"We’re now shooting the show on the RED EPIC cameras in 4K ... and we’ve moved to Final Cut Pro X. We’ve been able to do things on Leverage that no other cable show does simply because we can afford to do it using our all-digital workflow. It’s very rare to see a television show that averages 40 digital effects per episode. Or has four- or five-day sound mixing sessions. We’re able to do it and still produce a show for basically $1.8 million an episode. We think that Final Cut Pro X shows how simply and inexpensively a powerful file-based workflow can be implemented - not only does it change the price, but it actually changes creatively the way in which we work. We don’t have to wait to lock picture to start our digital effects shop.”
If someone rests a $1.8m budget on the ability of a $300 app to produce the final product, they either have to replace it when it fails or the software can handle it just fine.
The FCP to FCPX transition wasn't very good in that FCP had been languishing, and FCPX debuted in a typical 1.0 product state with the older one immediately EOL. The rest goes back to what I've already mentioned. You need some combination of hardware and software that meet the requirements of an entire working solution. The primary advantage of the desktop solution there is flexibility. You can usually find some custom solution to make it work. It doesn't mean other machines won't work if they align well with the required tasks. The link you posted is interesting in that it's the kind of place you'd expect to see newer workflows adopted first. Larger productions may have far more pipeline dependencies that make sudden shifts extremely difficult. They may have a lot of custom code to make things work, and ultimately what is important is that work is completed on time to the best quality possible.
[quote]At this stage, people should be grateful Apple is even bothering to make another one instead of condeming them for not doing it sooner because really, dropping it wouldn't affect them one bit and where would that leave potential buyers? They have no option but to buy a crummy PC and they're the ones that suffer for it.
[VIDEO]
^ Mac Pro buyer on the phone to the Apple Store wondering when the new one is coming out.[/quote]
It's possible to be a professional content creator with a Mac Mini. That doesn't make it preferable to use a Mini, does it?
It depends on the job I'd say. If your work is more storage limited than processor-limited like with video editing then do you spend $2500 on a quad-core Mac Pro or do you spend $799 on a quad-core Mini and $1099 on a Pegasus R4? You still need to load the MP with drives and a RAID card to match the storage performance.
You can get a faster MP but you're pretty much at $3800 before you see the benefit, which is already double the Mini setup and again you have to load it up.
If you take costs out of the equation, faster is better but you wouldn't always recommend the $6200 model fully loaded with SSDs and 96GB RAM and the highest GPU because it's not always needed.
When you look at people who are at the top of their game, time and time again they show it doesn't matter what you use:
"Steven Spielberg, surprisingly enough, for all the wondrous effects he has brought to the screen in films like Jurassic Park (1993), was one of them. Of digital editing he has complained: "It doesn't smell like film. It smells like an electronic lab." In the '90s, seeing that the end was nigh, he began buying up old Moviolas and their spare parts to guard against a future shortage.
By the second decade of the new millennium, though, even he had succumbed to the inexorable pressure of modernity. His recent films Tintin (2011), War Horse (2011), and Lincoln (2012) were all cut on an Avid, though he insists he's anxious to return to analog as soon as possible. Such devotion to the past is comforting in a way, revealing that genius need not depend solely on flashy new technology, though it is certainly unusual these days. As Walter Murch wrote over a decade ago: "The persistence of the Moviola into the last decade of the twentieth century is about as surprising as seeing an old manual Underwood typewriter loaded onto the Space Shuttle."
"“We all love the Moviola, that’s what got us were we are. Steve even loved the smell of it, the feel of it. He really enjoyed it,” reveals Kahn. “But when you sit down on an Avid and you want a trim right away, you don’t have to go to a box to get it, you just push a button and the trim is there. It is really quite an amazing machine and I wonder how they can build such a thing!” Kahn admits that he was fairly quick editing on film, but that Avid is ten times quicker for him. Film had given him the eye to quickly see matches, a skill which translated over to working on the computer."
They did amazing jobs manually cutting film in old mechanical machines and in some cases prefer it. You get a different mindset with limited technology. If you have to manually chop a piece of footage, it's a more important decision to make so you make it more carefully. When you work on a slower machine, you are similarly forced to find ways to optimize your work.
While having more resources to play with gives you a higher bound on creative freedom, it gives you a freedom to be wreckless too. It's like with large storage sizes and loading up your internal drives, at first it feels great to have 12TB internally but then you need a new machine, how do you move that 12TB of data to your new machine? What do you do when you run out of space or need to back it up, you can't put more inside so you have to go external after all. When you find one Mac Pro isn't fast enough, you have to start network rendering just like you would with any other machine.
Here's one where the Mac Pro buyer is explaining to Apple that they are the ones who are out there fighting for their cause but they just need to look less embarrassed doing it:
They did amazing jobs manually cutting film in old mechanical machines and in some cases prefer it. You get a different mindset with limited technology. If you have to manually chop a piece of footage, it's a more important decision to make so you make it more carefully. When you work on a slower machine, you are similarly forced to find ways to optimize your work.
This is probably the best argument I've yet heard for the Mini or iMac over a Mac Pro. Get sh!t done slower, so you have more time to think. Think carefully about optimizing your work.
It's also amusing that the Apple hardware defenders keep mentioning the "Pegasus". That just proves how limited fast external drive options are for Apple's disposable hardware - you have one brand a couple models to choose. Of course Thunderbolt storage has limited bandwidth compared to the professional SAS solutions for a Mac Pro, but you know that.
Here's one where the Mac Pro buyer is explaining to Apple that they are the ones who are out there fighting for their cause but they just need to look less embarrassed doing it:
This is probably the best argument I've yet heard for the Mini or iMac over a Mac Pro. Get sh!t done slower, so you have more time to think. Think carefully about optimizing your work.
You don't necessarily get things done noticeably slower. Productivity doesn't scale directly with raw hardware performance.
Having a machine that is 3x faster doesn't mean that you finish a job 3x quicker unless it's all raw processing. If you spend 20 hours editing a 60 minute video, it takes 20 hours on every machine. When you export, it might take 1.5 hours on the Mini and 0.5 hour on the top of the line Mac Pro (not likely as video encoding doesn't scale like that either). You've saved 1 hour out of 21.5 hours (5% more productive).
Even if you do effects and you render out a 15 minute render every hour on a low-end machine in an 8-hour day, the MP would save 8 x 10 minutes = 80 minutes.
HP tried to work this out and remember they are trying to sell people on the idea of buying higher powered machines:
Even though they are claiming 45-95% more productivity for adding a second CPU, they are quoting savings of 1.5-2 hours a day. That's nowhere near 45%, which is no doubt why they added their 'conservative' graph at the end. Sure an hour or two a day can save money over a period of time but a lot of things can save that time too and where does it end? Why stop at two processors, why not four? Does having two machines help? You can't just tack on power and assume it's going to make a big difference to a human-limited task.
It's also amusing that the Apple hardware defenders keep mentioning the "Pegasus". That just proves how limited fast external drive options are for Apple's disposable hardware - you have one brand a couple models to choose. Of course Thunderbolt storage has limited bandwidth compared to the professional SAS solutions for a Mac Pro, but you know that.
Yes, this has been covered in the past. There are more drives than the Pegasus (Lacie for example and USB 3 drives), it just happens to be one of the better performing ones. SAS setups don't really offer much more practical bandwidth than Thunderbolt because you hit the read/write speeds of the drive much sooner, but you know that.
While having more resources to play with gives you a higher bound on creative freedom, it gives you a freedom to be wreckless too. It's like with large storage sizes and loading up your internal drives, at first it feels great to have 12TB internally but then you need a new machine, how do you move that 12TB of data to your new machine? What do you do when you run out of space or need to back it up, you can't put more inside so you have to go external after all. When you find one Mac Pro isn't fast enough, you have to start network rendering just like you would with any other machine.
Here's one where the Mac Pro buyer is explaining to Apple that they are the ones who are out there fighting for their cause but they just need to look less embarrassed doing it:
You've drifted into hyperbole with the video links. I'm starting to wonder if Tallest gained control of your account. These problems are regularly addressed. If you're keeping a lot of storage internally, it's typical to leave the boot drive out of that. Single volumes can be transplanted later. I wouldn't personally use Raided internal drives for storage, which is what you seem to suggest here. With single volumes just move them. If you need real storage, you have a much wider range of available HBAs and RAID or JBOD boxes that can be used with a Mac Pro, although usb3 will probably cannibalize the lower end of this. The capabilities existed prior to thunderbolt. In case you are curious what annoys me, it comes down to the choices available when locked to a single company. With the imac, it would displace the use of superior displays. I can't fit both. The internals are probably enough. With the mini, the gpu would choke. I already choke it anyway. I've considered going to Windows simply because you can find a wider range of cards. If I did that, the machine I'd end up with would probably be something like an E5 hex core, 2 PCI cards with one being a mid range Quadro and the other being an HBA, as much ram as it would hold, and I'd stick with NEC or Eizo on displays going forward. Ironically I might consider a Fermi card. Kepler is terrible at floating point math.
You've drifted into hyperbole with the video links.
They illustrate two very important points. The first is that Apple won't be convinced about the MP by writing letters to Tim or running a Facebook campaign. Apple isn't a charity, it's a business and everything they do needs to make business sense. If people want to keep seeing a new Mac Pro, show them the money - that's the only way they can measure that people still want them and holding onto them for 6 years isn't the way to do that.
The second is that Mac Pro buyers need to realise that Apple won't fail without them. They are the ones hanging by a thread. Apple is the most valuable company in the world right now and I don't think the magnitude of that has sunk in. They used to be near bankruptcy and relied on people buying their beige desktops. They don't have that reliance any more and there isn't a debt of gratitude to be paid to tower fans because iMacs and laptops are far more powerful than the equivalent towers they offered back then.
These problems are regularly addressed. If you're keeping a lot of storage internally, it's typical to leave the boot drive out of that. Single volumes can be transplanted later. I wouldn't personally use Raided internal drives for storage, which is what you seem to suggest here.
Some people suggest internal RAID is an advantage of a larger machine - that you don't need a 'rats nest' with an external RAID box or don't need to pay for something like a Pegasus. As you say, it's not a good idea to RAID your system drive and you wouldn't want to use software RAID on the 3 other drives nor would you use Apple's hardware RAID so you have to just buy external storage connected via PCI card using up a slot and creating a rats nest on top of the already huge Mac Pro. You might get away with a 3rd party RAID card but then you compromise reliability and stability with the drivers.
In case you are curious what annoys me, it comes down to the choices available when locked to a single company. With the imac, it would displace the use of superior displays. I can't fit both.
So, you can fit a Mac Pro with an 8" depth and a 27" diagonal plus a separate display but you can't fit an iMac with an 8" stand depth (much less for the rest) and a 27" diagonal with a separate display?
It's pretty much to scale with all the components. There's plenty room in that shape for a powerful workstation. The side could be held on magnetically with pins and the base could have a gap large enough for fingers so it can be moved without trapping them while also allowing airflow underneath.
There might be enough room for more PCI slots if they wanted but they can also do dual 3-4TB Fusion drives. The power cable goes in at the bottom instead of the top so it doesn't get in the way of the ports. There's also no airflow blockage.
The top-end one could be about $4500 with a 12-core chip and they can offer whatever GPUs they like (with 6x Mini-DP outputs). The RAM would probably be in front of the CPU cooler too because they might not be able to make it a drawer and they'd want a guard round the cooler to prevent someone poking it while it was spinning.
You don't necessarily get things done noticeably slower. Productivity doesn't scale directly with raw hardware performance.
Having a machine that is 3x faster doesn't mean that you finish a job 3x quicker unless it's all raw processing. If you spend 20 hours editing a 60 minute video, it takes 20 hours on every machine. When you export, it might take 1.5 hours on the Mini and 0.5 hour on the top of the line Mac Pro (not likely as video encoding doesn't scale like that either). You've saved 1 hour out of 21.5 hours (5% more productive).
Even if you do effects and you render out a 15 minute render every hour on a low-end machine in an 8-hour day, the MP would save 8 x 10 minutes = 80 minutes.
HP tried to work this out and remember they are trying to sell people on the idea of buying higher powered machines:
Even though they are claiming 45-95% more productivity for adding a second CPU, they are quoting savings of 1.5-2 hours a day. That's nowhere near 45%, which is no doubt why they added their 'conservative' graph at the end. Sure an hour or two a day can save money over a period of time but a lot of things can save that time too and where does it end? Why stop at two processors, why not four? Does having two machines help? You can't just tack on power and assume it's going to make a big difference to a human-limited task.
Yes, this has been covered in the past. There are more drives than the Pegasus (Lacie for example and USB 3 drives), it just happens to be one of the better performing ones. SAS setups don't really offer much more practical bandwidth than Thunderbolt because you hit the read/write speeds of the drive much sooner, but you know that.
So what exactly is your point? That we don't need fast hardware because slower works well enough? You used a lot of words to state the obvious, although things are more complex than your wordy post suggests. Faster hardware not only gets sh!t done faster in the present, it remains viable longer into the future. Contrary to Apple's landfill fodder business strategy, most people who buy $6000 Xeon workstations expect the hardware to not only support today's software, but tomorrow's. In other words, build a faster system, and developers will find a way to exploit all the power and then some.
About SAS storage, it can scale considerably better than TB, and more cheaply. Arguing in favor of TB at this point is just Apple fanboy delusions. The next revision of TB should double the number of available PCIe lanes, so then it could be a good solution depending on how many vendors support it.
You don't necessarily get things done noticeably slower. Productivity doesn't scale directly with raw hardware performance.
Again that depends. If your machine is loaded to the point that the UI can't keep up with your typing then you need a faster machine.
Having a machine that is 3x faster doesn't mean that you finish a job 3x quicker unless it's all raw processing. If you spend 20 hours editing a 60 minute video, it takes 20 hours on every machine.
This isn't totally correct either. A faster machine can benefit the user simply by reducing lag.
When you export, it might take 1.5 hours on the Mini and 0.5 hour on the top of the line Mac Pro (not likely as video encoding doesn't scale like that either). You've saved 1 hour out of 21.5 hours (5% more productive).
Even if your logic above made sense, you seem to forget that people multitask. Saving time on rendering isn't as important as having a usable machine while rendering.
Even if you do effects and you render out a 15 minute render every hour on a low-end machine in an 8-hour day, the MP would save 8 x 10 minutes = 80 minutes.
If you continue working, on whatever, that is 80 minutes more productivity. Effectively that gets another job out the door on the same day.
HP tried to work this out and remember they are trying to sell people on the idea of buying higher powered machines:
Even though they are claiming 45-95% more productivity for adding a second CPU, they are quoting savings of 1.5-2 hours a day. That's nowhere near 45%, which is no doubt why they added their 'conservative' graph at the end. Sure an hour or two a day can save money over a period of time but a lot of things can save that time too and where does it end?
It ends at two processors because that is where machines remain economical. The reality is the extra cost of that processor is a bargain in most cases and would likely pay for itself in a couple of months. That is with the simplistic approach to productivity they are measuring by.
Why stop at two processors, why not four.
Because two socket machines are relatively economical.
Does having two machines help? You can't just tack on power and assume it's going to make a big difference to a human-limited task.
You make the assumption that the human is the limitation here. That is hard to swallow if the human is constantly waiting on the machine.
Yes, this has been covered in the past. There are more drives than the Pegasus (Lacie for example and USB 3 drives), it just happens to be one of the better performing ones. SAS setups don't really offer much more practical bandwidth than Thunderbolt because you hit the read/write speeds of the drive much sooner, but you know that.
Imagine a Mac Pro with 2TB of solid state storage in the boot / applications drive. Or maybe a spilt with 1TB allocated to the volume manage for tiered storage. It is nice to argue about the number of sockets in a machine but a total rethink of the Mac Pros architecture would go a very long way to making the machine viable over the long run. As much as we argue over the bottle necks in multi core or multi socket machines a very real and often overlooked choke point in modern systems is secondary storage. Without gains here significant gains in performance in the future may be harder to obtain no matter how many sockets or cores you have.
They illustrate two very important points. The first is that Apple won't be convinced about the MP by writing letters to Tim or running a Facebook campaign. Apple isn't a charity, it's a business and everything they do needs to make business sense. If people want to keep seeing a new Mac Pro, show them the money - that's the only way they can measure that people still want them and holding onto them for 6 years isn't the way to do that.
Marvin I can't believe you are still hung up on the idea that it is the users fault for the current sales situation with the Mac Pro. You say Apple isn't a charity but yet you say the way to save them is to donate more money to Apple as if they where a charity. That my good man it totally asinine.
Seriously it is Apples responsibility to come up with an economical and suitable modern computer for the Pro and advanced users. To say that these people need to buy more often is just shows a massive misunderstanding of the market that such machines sell into. I can just imagine going up to the CEO of any company, big or small, and telling them that hey buddy you need to buy Mac Pros more often so that Apple will keep making them. The concept is hilarious to say the least.
The second is that Mac Pro buyers need to realise that Apple won't fail without them. They are the ones hanging by a thread. Apple is the most valuable company in the world right now and I don't think the magnitude of that has sunk in. They used to be near bankruptcy and relied on people buying their beige desktops. They don't have that reliance any more and there isn't a debt of gratitude to be paid to tower fans because iMacs and laptops are far more powerful than the equivalent towers they offered back then.
The fact that laptops are more powerful has little to do with the fact that they have ignored the desktop user. All we really want is to see the same level of innovation on the desktop that we have seen in the laptop line up. It is not unreasonable to demand that from a company like Apple. Like it or not having a respectable desktop lineup just shows your customer base that you are serious about Mac OS.
Now I have repeatedly stated that I think their desktop line up is borked. You basically have three models that are highly compromised and thus serve a limited scope of customers. This has impacted their sales of desktops and frankly it is pretty hard for anybody to deny this. All the way back to 2008 I made a decision to buy a MBP not because I really needed it but because there wasn't a viable desktop solution. Nothing has really changed other than the iMac & Mac Pro have gotten worst as far as being a viable desktops. I was this " close to buying a Mini in 2012 but Apple decided to bork that machine instead of upgrading the GPU to something a rational person would want. So really if you don't understand why people feel forced into buying laptops from Apple I really don't know what to say. Apple can do a lot better on the desktop but that entails scraping the entire lone up, adjusting margins and looking to the future. In stead we have Apple sitting on the desktop hardware letting it stagnate.
Some people suggest internal RAID is an advantage of a larger machine - that you don't need a 'rats nest' with an external RAID box or don't need to pay for something like a Pegasus. As you say, it's not a good idea to RAID your system drive and you wouldn't want to use software RAID on the 3 other drives nor would you use Apple's hardware RAID so you have to just buy external storage connected via PCI card using up a slot and creating a rats nest on top of the already huge Mac Pro.
Most of the above is BS. Internal RAID has had and still has its place in the world. So does software RAID and even Apples controller, it really depends upon use cases. However with the advent of TB I no longer see a pressing need for lots of internal drives.
You might get away with a 3rd party RAID card but then you compromise reliability and stability with the drivers.
No more so than software issues can play havoc with external RAID boxes. It is certainly easier to unplug an external box and blame the vendor though.
So, you can fit a Mac Pro with an 8" depth and a 27" diagonal plus a separate display but you can't fit an iMac with an 8" stand depth (much less for the rest) and a 27" diagonal with a separate display?
They illustrate two very important points. The first is that Apple won't be convinced about the MP by writing letters to Tim or running a Facebook campaign. Apple isn't a charity, it's a business and everything they do needs to make business sense. If people want to keep seeing a new Mac Pro, show them the money - that's the only way they can measure that people still want them and holding onto them for 6 years isn't the way to do that.
Holding them 6 years isn't something that everyone does. You're examining a few forum posters, which is a silly way to extract data. The other thing you fail to account for is that the number of mac pros they sold in 2006 or the number of G5s they sold in 2004 may not satisfy them today, as their growth likely comes from other lines. There's a difference between slipping in units sold and simply being outpaced by the rest of their line. Outside of this last year, I expect the latter is more the case. That wouldn't get the mac pro off the hook if it's scheduled for cancellation. I'm just saying that you're looking at a limited market that has typically worked due to high margins. It's not so much of a dead market as you claim. It generates far better margins for oems than any of their lower products. Apple is in a unique situation. They've retained relatively high margins across their line, and their growth is driven elsewhere.
Quote:
The second is that Mac Pro buyers need to realise that Apple won't fail without them. They are the ones hanging by a thread. Apple is the most valuable company in the world right now and I don't think the magnitude of that has sunk in. They used to be near bankruptcy and relied on people buying their beige desktops. They don't have that reliance any more and there isn't a debt of gratitude to be paid to tower fans because iMacs and laptops are far more powerful than the equivalent towers they offered back then.
I'm not one of the people who buys into this idea. Some people are really weird in that they attempt to personify a corporation. Regarding the progression of lighter hardware, big points are growth in requirements and budget elasticity. The acceptance of lighter hardware has taken over somewhat on Windows too. The problem for some people is when you're dealing with a single company, if your requirements don't align well with what they make, you are out of luck. I've tested everything I use on Windows. While I'd have to make a couple hardware changes, it would work. I'm extremely annoyed by how ram hungry Lion and Mountain Lion have been. I max out memory constantly as some of the applications I deal with will use everything I throw at them. It annoys me when I feel that extra is tied up due to a bloated OS. Fedora runs a few things that I require, but some of them are problematic. I'd move to Linux if I could get everything working, just to have a nice light/stable OS. Ubuntu and Debian won't support certain things, or I would have switched already. I think I mentioned what I'd go with if I transitioned to Windows. I test a lot of this stuff to the configuration I require. I bug test everything, and download Windows trials of all software. I have a backup plan if OSX becomes unworkable, and I've avoided any software published by Apple for the past few years. This is far more trivial to me than it may be to others.
Quote:
Some people suggest internal RAID is an advantage of a larger machine - that you don't need a 'rats nest' with an external RAID box or don't need to pay for something like a Pegasus. As you say, it's not a good idea to RAID your system drive and you wouldn't want to use software RAID on the 3 other drives nor would you use Apple's hardware RAID so you have to just buy external storage connected via PCI card using up a slot and creating a rats nest on top of the already huge Mac Pro. You might get away with a 3rd party RAID card but then you compromise reliability and stability with the drivers.
There are better boxes out there, and if we're talking about a 12TB raid, I'd go with WD's enterprise grade SATA drives. It comes out slightly more expensive, but you can put together a much more robust solution for a fairly similar amount of money.
Quote:
So, you can fit a Mac Pro with an 8" depth and a 27" diagonal plus a separate display but you can't fit an iMac with an 8" stand depth (much less for the rest) and a 27" diagonal with a separate display?
Is it difficult to understand they wouldn't occupy the same space? Towers typically fit beneath a desk. My point with the imac was that it would have to fit in side by side with another display. At 27", I'd need to make it the primary display viewed dead on due to size. It's not sufficient for that. The differences are somewhat hedged by tradeoffs. Mac pro can take a slightly wider range of gpus. A few people have gotten Fermi to work on it, which is stronger at floating point math than any Kepler card currently on the market. Fermi was just heavily optimized for computation. It can take a couple extra drives internally (note I didn't say a RAID storage array). It's still a bit faster. The imac provides a display within its total cost. That really is appealing, but Apple displays in my own experience have been plagued with too many problems. I have something that is relatively problem free at the moment. I'd like to retain that.
Quote:
Originally Posted by Marvin
Some people suggest internal RAID is an advantage of a larger machine - that you don't need a 'rats nest' with an external RAID box or don't need to pay for something like a Pegasus. As you say, it's not a good idea to RAID your system drive and you wouldn't want to use software RAID on the 3 other drives nor would you use Apple's hardware RAID so you have to just buy external storage connected via PCI card using up a slot and creating a rats nest on top of the already huge Mac Pro. You might get away with a 3rd party RAID card but then you compromise reliability and stability with the drivers.
So, you can fit a Mac Pro with an 8" depth and a 27" diagonal plus a separate display but you can't fit an iMac with an 8" stand depth (much less for the rest) and a 27" diagonal with a separate display?
Actually some of the third party ones have been better supported than Apple's own solutions. Have you ever looked at reviews on their Mac Pro raid card or DOA drive complaints with the Pegasus raid? Apple is the last one you should trust there. You mentioned 12TB. At that level if I had to put something together in the form of a hardware raid, it would likely run $2-3k. Apple charges $2300 for that. My way might end up more expensive, but I could put together a more stable solution with one of Areca's desktop raids and Western Digital RE4s. Apple is usually the problem when it comes to driver stability. If we're talking about $40 eSATA cards, that's a different story. Those things are sold on razor thin margins.
Here's roughly how the internals could fit inside an 8" cube:
It's pretty much to scale with all the components. There's plenty room in that shape for a powerful workstation. The side could be held on magnetically with pins and the base could have a gap large enough for fingers so it can be moved without trapping them while also allowing airflow underneath.
While it would be a bit tight the one thing I do like about getting down to about 8" is that you could design the product with a viable rack mount solution. That is a bracket and bolting arrangement to put two in a standard rack width. It doesn't have to be pretty as this would go into installations where people don't care.
Honestly though I'd rather that Apple completely broke with the past and just deleted traditional magnetic drive bays in a machine this small. Put in a couple of dedicated PCI Express slots and implement solid state storage. Frankly I'm not sure why Apple hasn't done this on the Mini yet. If magnetic drives are needed, for whatever reason put in a couple of slots/bays for laptop drives.
There might be enough room for more PCI slots if they wanted but they can also do dual 3-4TB Fusion drives. The power cable goes in at the bottom instead of the top so it doesn't get in the way of the ports. There's also no airflow blockage.
I would rather have one accessible external PCI Express slot beyond the GPU slot. Further such a machine cries out for a base model with an integrated GPU. Intel would do well to offer up a socket that supports GPU enabled processors or more traditional multi core processors. Such a beast would allow Apple to support a wider array of customers on one platform. They might even be able to ditch the Mini.
The top-end one could be about $4500 with a 12-core chip and they can offer whatever GPUs they like (with 6x Mini-DP outputs).
I'm still wondering how Apple will address the TB issues with the new Pro Machine. I could see them soldering the GPU right on the motherboard. This is one thing about the TB initiative that is still a mystery.
The RAM would probably be in front of the CPU cooler too because they might not be able to make it a drawer and they'd want a guard round the cooler to prevent someone poking it while it was spinning.
Why bother with a drawer? Seriously build everything onto the motherboard or at least as much as possible putting slower parts on a daughter card.
I still think 8" is a little tight for a computer bound for the Pro market but you are on the right track trying to condense the machine down to the basic elements of a Pro machine. What will be interesting is the port selection, I'm just interested in what Apple sees as appropriate for the next 6 years or so.
Holding them 6 years isn't something that everyone does. You're examining a few forum posters, which is a silly way to extract data.
What was and still is silly, is the idea the he offered up that poor Mac Pro sales are our fault because we have not purchased Mac Pros frequently enough. That is absurd and frankly irrational. I know their is a certain element in society that has to go out and buy a new iPhone every time one comes out. That is not the Pro market at all. Now some business can justify fairly quick replacement schedules but it is not odd at all to see the same machine in use for five years or more in business.
The other thing you fail to account for is that the number of mac pros they sold in 2006 or the number of G5s they sold in 2004 may not satisfy them today, as their growth likely comes from other lines. There's a difference between slipping in units sold and simply being outpaced by the rest of their line. Outside of this last year, I expect the latter is more the case.
From what I've been able to determine it is slippage in sales that is the problem, more so a very significant slippage in sales. Of the desktops only the iMac has been positive in the USA.
That wouldn't get the mac pro off the hook if it's scheduled for cancellation. I'm just saying that you're looking at a limited market that has typically worked due to high margins. It's not so much of a dead market as you claim. It generates far better margins for oems than any of their lower products. Apple is in a unique situation. They've retained relatively high margins across their line, and their growth is driven elsewhere.
When looking at value, from the consumers point of view, the current Mac Pro is a big zero in the base models. This drives sales downward as the market for the high performance versions is more limited. I really don't see the concept of a high performance Mac Pro as so much of an issues, as is Apples big gap between the Mini and the tricked out Mac Pro. From my perspective what is killing the Mac Pro more than anything else is that apple doesn't have a midrange desktop machine. That machine could be derived from the Mac Pro or not but the lack of such a machine drive the perception that Apple has nothing worthwhile on the desktop.
I'm not one of the people who buys into this idea. Some people are really weird in that they attempt to personify a corporation. Regarding the progression of lighter hardware, big points are growth in requirements and budget elasticity. The acceptance of lighter hardware has taken over somewhat on Windows too.
The Windows world has basically crashed. People have literally jumped off the treadmill.
The problem for some people is when you're dealing with a single company, if your requirements don't align well with what they make, you are out of luck.
That is a two way street. Apple has basically ran the entire desktop line up into the gutter through mismanagement and not caring. In a nut shell they are out of luck as far as customers for that hardware. It is showing up in sales of desktop hardware.
I've tested everything I use on Windows. While I'd have to make a couple hardware changes, it would work. I'm extremely annoyed by how ram hungry Lion and Mountain Lion have been. I max out memory constantly as some of the applications I deal with will use everything I throw at them. It annoys me when I feel that extra is tied up due to a bloated OS. Fedora runs a few things that I require, but some of them are problematic. I'd move to Linux if I could get everything working, just to have a nice light/stable OS. Ubuntu and Debian won't support certain things, or I would have switched already:grumble: .
I ran Linux from about the time NT service pack 3 came out until 2008 when I went with a Mac. I can not understate just how frustrating Linux can be. You have a system built around a decent kernel that really could go somewhere if they could just address the UI issues that have been there for ever. Well that and the constant need to do complete installs every 6 months. Linux is a nice hobby, don't get me wrong as I still have a machine or two at the house, but it isn't the place to go if you want the hardware and software to work reliably.
I think I mentioned what I'd go with if I transitioned to Windows. I test a lot of this stuff to the configuration I require. I bug test everything, and download Windows trials of all software. I have a backup plan if OSX becomes unworkable, and I've avoided any software published by Apple for the past few years. This is far more trivial to me than it may be to others.
After my stint with Linux I've gone all in with Apples software. Interestingly when I first got the Mac I did try using some of the open source stuff (mostly desktop productivity) that ran on the Mac. Eventually I realized that this stuff suffered from the same issues as the software running on Linux. Eventually I installed the Apple software and frankly it is good enough with zero hassles.
There are better boxes out there, and if we're talking about a 12TB raid, I'd go with WD's enterprise grade SATA drives. It comes out slightly more expensive, but you can put together a much more robust solution for a fairly similar amount of money.
Is it difficult to understand they wouldn't occupy the same space? Towers typically fit beneath a desk. My point with the imac was that it would have to fit in side by side with another display. At 27", I'd need to make it the primary display viewed dead on due to size. It's not sufficient for that. The differences are somewhat hedged by tradeoffs. Mac pro can take a slightly wider range of gpus. A few people have gotten Fermi to work on it, which is stronger at floating point math than any Kepler card currently on the market. Fermi was just heavily optimized for computation. It can take a couple extra drives internally (note I didn't say a RAID storage array). It's still a bit faster. The imac provides a display within its total cost. That really is appealing, but Apple displays in my own experience have been plagued with too many problems. I have something that is relatively problem free at the moment. I'd like to retain that.
The above highlights possible the biggest issue with the Mac, once you get outside of the Mac mainstream as far as software is concerned your choice dwindle fast. The need for a hacker mentality to get things to work is almost as bad as Linux. Some companies just seem to have a serious inability to support their hardware on the Mac. A further software, especially in the engineering world has been slow to come to the Mac. So it does feel at times like the Linux world. However Max OS/X redeems itself via iOS integration and a bunch of apps that really are state of the art as far as usability goes.
So what exactly is your point? That we don't need fast hardware because slower works well enough?
Not quite, I'm not suggesting everyone should be happy with Macbook Airs. I'm saying that there are diminishing returns on investing in the fastest hardware the more that time goes on.
Say, for example, Apple put up a new computer tomorrow that had 4x E5-2687W CPUs, dual GTX 690s and then slapped a price tag of $11,000 on it. Is that something people would want? If raw performance is so important then why wouldn't people want it?
If we were back in 1995, those kind of prices meant the difference between getting a job done and not getting it done because some tasks couldn't be done with too little power or too little RAM. These days, the gains in productivity from faster hardware are getting lower because most of what we do is real-time.
Once the productivity gains from a new machine aren't significant enough, people hold onto machines longer. Then the company selling them doesn't see enough demand so they focus on hardware that does see the demand. Eventually an upgrade is required and there's no new model to be found.
Faster hardware not only gets sh!t done faster in the present, it remains viable longer into the future.
Yes, typically 3 years longer. Hardware goes up about 50% in performance every year (it doesn't do this every year but it tends to average out at this over a period of time) so in 3 years, we get 1.5^3 = ~3x faster. The top-end MP has a 3x faster CPU than the top-end iMac so in 3 years they are about the same. The GPUs make the gap even smaller because the GTX680M in the iMac is 70% of the performance of the GTX680 that would have gone into the Pro so the Pro would barely be a year ahead in terms of the GPU.
This trend may change if Intel decides to bump CPUs up 15% year on year and focus on GPUs but I think they'll focus more on CPUs again once the GPUs move up high enough.
Saving time on rendering isn't as important as having a usable machine while rendering.
Even a 12-core MP would lag if you max it out though. That's just a case of limiting the processes e.g give it 6 out of 8 threads and 12GB RAM out of 16GB or whatever.
It ends at two processors because that is where machines remain economical.
I think it always comes down to balancing needs and costs and it could be the case that there are enough people willing to pay for a $4000+ machine to justify 2P machines and not enough to justify a $10,000 4P machine but I think fewer and fewer people are willing to spend $4000+ on computer equipment now. Here's an example of someone who replaced XServes with Minis:
I can just imagine going up to the CEO of any company, big or small, and telling them that hey buddy you need to buy Mac Pros more often so that Apple will keep making them. The concept is hilarious to say the least.
It might seem hilarious but it's the way supply and demand works. Take away the demand and you take away the supply. You're right the supplier has a responsibility to create a demand for something but Apple has been at this a very long time and they know how big the markets are in each sector and their supply priorities reflect that.
Even if Apple brought out the most awesome Mac Pro anyone can imagine next year, it still won't sell in huge numbers with a starting price above $2000 and they won't drop the price lower than that. So making a great machine won't make a bit of difference unless that market grows substantially or they cut the price massively. As I showed earlier, lowering the margins doesn't give a direct increase in volume.
Holding them 6 years isn't something that everyone does. You're examining a few forum posters, which is a silly way to extract data.
It keeps swinging back and forth. Either the Mac Pro is great because it lasts, which means people are holding onto them or after 3-4 years, they are no faster than the lower end and people are upgrading. If the lower end matches them in performance after that shorter period, maybe these people switch to the lower end models but then people say that can't be happening because the lower end doesn't offer xyz benefits. It can't be all things at once.
I think a lot of people are migrating to lower machines as time goes on, this shows in the buying rates for laptops vs desktops and iMacs vs towers. I think tower buyers aren't upgrading often either. Both of these situations create a smaller demand for the Mac Pro and this shows in Apple's treatment of it.
There's a difference between slipping in units sold and simply being outpaced by the rest of their line. It's not so much of a dead market as you claim.
Before the iMac though, the towers used to be priced a lot lower. They displaced them in favour of the iMac in order to offer the best quality for the whole package in that range and make something compelling to buy. The $2500+ target audience is small and the market which would normally have had to go with a tower (DTP / CAD / video editing markets) can now get the work done on something much cheaper and even portable.
While the number of people in the high-resource audience continues to grow, the lower end hardware meets more needs and the higher end hardware encourages people to upgrade less.
My point with the imac was that it would have to fit in side by side with another display
I would say it's wasteful to stick it on the floor but you can do it if you want or even just sit the NEC/Eizo in front of it. It's still cheaper than the equivalent Mac Pro.
Maybe but every machine has a power target. The Mini is 85W, the iMac is 310W, the MP is currently 1000W. They'd just be making the Mini less attractive to people who like a tiny desktop machine and want to run it as a low power server.
What was and still is silly, is the idea the he offered up that poor Mac Pro sales are our fault because we have not purchased Mac Pros frequently enough. That is absurd and frankly irrational. I know their is a certain element in society that has to go out and buy a new iPhone every time one comes out. That is not the Pro market at all. Now some business can justify fairly quick replacement schedules but it is not odd at all to see the same machine in use for five years or more in business.
From what I've been able to determine it is slippage in sales that is the problem, more so a very significant slippage in sales. Of the desktops only the iMac has been positive in the USA.
That doesn't surprise me. A lot of the cliche Mac customers seem to be trending toward the use of imacs. My suggestion is that they purchase directly from Apple. If the display exhibits poor uniformity, return it. I've also explained to some of them what display calibration can and cannot fix and why they shouldn't cheap out on drives for their RAIDs.
Quote:
When looking at value, from the consumers point of view, the current Mac Pro is a big zero in the base models. This drives sales downward as the market for the high performance versions is more limited. I really don't see the concept of a high performance Mac Pro as so much of an issues, as is Apples big gap between the Mini and the tricked out Mac Pro. From my perspective what is killing the Mac Pro more than anything else is that apple doesn't have a midrange desktop machine. That machine could be derived from the Mac Pro or not but the lack of such a machine drive the perception that Apple has nothing worthwhile on the desktop.
The markets they're likely to approach with such a machine are somewhat limited. To maintain health with such a line, they'd need to attract customers that already buy into this hardware class from other vendors. The workstation market has shown growth in many quarters. 2010 showed some amount of growth. 2011 and 2012 were slow for obvious reasons. Budgets aren't necessarily elastic, and below $3000 + display, things have been moving very slowly.
Quote:
The Windows world has basically crashed. People have literally jumped off the treadmill.
I've noticed this to a degree, but it addresses certain things well enough, and you can configure decent hardware to the required spec quite easily. I tend to explore solutions due to a need to ensure I can make something work, and I've been a bit disappointed with OSX lately. I need the leanest OS possible. Windows isn't really that. It's just that it has a better range of gpus and bit more freedom in terms of customization. I don't think it's irrational to maintain contingency plans. I don't maintain a notion that I would somehow be holding Apple captive if they don't produce the machine I want to buy.
Quote:
I ran Linux from about the time NT service pack 3 came out until 2008 when I went with a Mac. I can not understate just how frustrating Linux can be. You have a system built around a decent kernel that really could go somewhere if they could just address the UI issues that have been there for ever. Well that and the constant need to do complete installs every 6 months. Linux is a nice hobby, don't get me wrong as I still have a machine or two at the house, but it isn't the place to go if you want the hardware and software to work reliably.
The vfx industry incorporates Linux in many areas, and I use a lot of the same software. Because of this there certified kernels and installation instructions for a lot of software. I've read them, so I know how much of a pain in the ass it is. I've been a bit disappointed in OSX with Lion onward. Snow Leopard was great. It's not possible to stay on older OS revisions forever, as you eventually miss too much in the way of added frameworks and software support. I usually update my OS mid cycle after reading through lists of known bugs for any critical software. It's quite boring.
Quote:
After my stint with Linux I've gone all in with Apples software. Interestingly when I first got the Mac I did try using some of the open source stuff (mostly desktop productivity) that ran on the Mac. Eventually I realized that this stuff suffered from the same issues as the software running on Linux. Eventually I installed the Apple software and frankly it is good enough with zero hassles.
There isn't a lot that I really need. I use a huge range of software. Some of it is used rather infrequently. FCPX isn't necessary as Creative Suite includes Premiere, even though I need it very rarely. If I didn't have Creative Suite, it would be different. The upgrade costs aren't that bad once you're past the upfront costs. The suite upgrade prices aren't much higher than single applications. This means it's not as big of a deal retaining software that I need less frequently. It used to be that you could upgrade up to 3 versions back with individual software products. Adobe reduced that to one to push their suites on everyone. It's annoying, but you get a lot of capability. I never personally liked Aperture. The rest isn't much different on Windows or OSX.
Quote:
The above highlights possible the biggest issue with the Mac, once you get outside of the Mac mainstream as far as software is concerned your choice dwindle fast. The need for a hacker mentality to get things to work is almost as bad as Linux. Some companies just seem to have a serious inability to support their hardware on the Mac. A further software, especially in the engineering world has been slow to come to the Mac. So it does feel at times like the Linux world. However Max OS/X redeems itself via iOS integration and a bunch of apps that really are state of the art as far as usability goes.
I've definitely mentioned that before. I do use some open source software, but I'm reasonably selective. I tend to avoid the "labor of love" projects supported by a single developer for anything important. I don't consider the RAID thing to be a hacker mentality. Marvin and I disagree on the quality of solutions provided by Apple at times. Apple Store reviews on the thunderbolt raid mention doa drives. This shouldn't really happen with a pre-configured raid solution, which should have been tested prior to shipping. It's not the same as just buying the drives as raw components on their own. I'd just stick with solutions that have been well supported in the past, and for a 12TB raid, you should be really picky on the hard drives used. The cheapest I'd go would be Caviar Blacks, as they use similar hardware to the RE4s. The RE4s have different firmware. iOS integration isn't that big of a deal to me. I have an iphone, and it's helpful, but not entirely crucial for me. The iphone was kind of an easy solution. It covers my requirements, but I will never wait outside the store to buy the latest one.
A lot of the cliche Mac customers seem to be trending toward the use of imacs.
The popularity of the iMacs isn't really a new thing. Even if you go back to the year 2000, Apple reported the split in numbers and the iMac outsold everything else:
"Sales of 1,377,000 systems including over 700,000 iMac™ consumer desktops and 235,000 iBook™ consumer portables"
This was back when the entry tower desktop was $1599 and the entry iMac was $799. It doesn't say how many Powerbooks vs Powermacs were in the remaining 442,000 units but if it was the same ratio as iMacs to iBooks, it would be 150k PBs and 290k PMs.
Knowing that Mac Pro shipments are under 100k / quarter now, it's clear that the tower desktop market has shrunk considerably or I guess a more accurate way to put it would be that people are happy to spend a $1000-2000 budget on an all-in-one instead of a tower and Apple makes better margins that way as they sell the display too and the way they make better margins from the towers is jack up the price. If they sold the towers at the same price as the iMacs again, it wouldn't be worth the trouble because you can bet not everyone would buy Apple's displays (like people who prefer NEC/Eizo).
To drive the growth that people here suggest, Apple would need to take the display out of the iMac and sell that at a lower price. But as I say, they just lose a display sale and have a machine without the iMac's unique selling points. If they use the same parts as the Mac Pro, they basically just have to cut their margins but they will still hit the upper end of the $1000-2000 range and they won't even double their volume.
The best route forward for them is to just play it out the same way they've been doing already. Scale down the operation, ship it back to the US so the machines can be assembled there (one factory can probably handle the 2,500 / day worldwide target). They can keep it running as long as the shipment volume stays above a certain level and the profit it makes is worth the effort. But regardless of whether people agree with it, people still need to buy them more often or their shipment volume reports will just slide and they'll think people don't want them any more. They might not drop the line for a few years but you can bet they'd stick with the long refresh cycles.
The popularity of the iMacs isn't really a new thing. Even if you go back to the year 2000, Apple reported the split in numbers and the iMac outsold everything else:
"Sales of 1,377,000 systems including over 700,000 iMac™ consumer desktops and 235,000 iBook™ consumer portables"
This was back when the entry tower desktop was $1599 and the entry iMac was $799. It doesn't say how many Powerbooks vs Powermacs were in the remaining 442,000 units but if it was the same ratio as iMacs to iBooks, it would be 150k PBs and 290k PMs.
I didn't realize it took over that early.
Quote:
Knowing that Mac Pro shipments are under 100k / quarter now, it's clear that the tower desktop market has shrunk considerably or I guess a more accurate way to put it would be that people are happy to spend a $1000-2000 budget on an all-in-one instead of a tower and Apple makes better margins that way as they sell the display too and the way they make better margins from the towers is jack up the price. If they sold the towers at the same price as the iMacs again, it wouldn't be worth the trouble because you can bet not everyone would buy Apple's displays (like people who prefer NEC/Eizo).
I get that. I tend to view things relative to my own needs. I'm not terribly biased against Apple displays. I just don't think they're as good, having compared on many occasions through different design generations. I'm confident that advancements in technology will bring these things further together. They all pretty much use the same panels at this point. A few years ago this wasn't the case. Some of the other vendors pulled out due to shrinking profits. As you know LG maintains an existence through extreme volume.
Quote:
To drive the growth that people here suggest, Apple would need to take the display out of the iMac and sell that at a lower price. But as I say, they just lose a display sale and have a machine without the iMac's unique selling points. If they use the same parts as the Mac Pro, they basically just have to cut their margins but they will still hit the upper end of the $1000-2000 range and they won't even double their volume.
That isn't necessarily what I'm saying they should map out as a current trajectory. That's basically something that would benefit me. I'm not arrogant enough to feel that would influence their business decisions.
Quote:
The best route forward for them is to just play it out the same way they've been doing already. Scale down the operation, ship it back to the US so the machines can be assembled there (one factory can probably handle the 2,500 / day worldwide target). They can keep it running as long as the shipment volume stays above a certain level and the profit it makes is worth the effort. But regardless of whether people agree with it, people still need to buy them more often or their shipment volume reports will just slide and they'll think people don't want them any more. They might not drop the line for a few years but you can bet they'd stick with the long refresh cycles.
Part of that is intel. 2009 to 2010 didn't provide much. Westmere was an incomplete lineup. Sandy Bridge E came late. That's a lot of flat growth in the lower tiers of the mac pro line, which likely carry a fair amount of its volume. They haven't maintained a lot of mac pro models overall. The 1,1 sold until the end of the first quarter in 2008. Given that many applications didn't release universal binary versions until at least 2007, it's reasonable to figure that many of those purchases were likely in 2007. In some of the edge cases, I can see how they would have been disappointed with the 2010 update and waited for Sandy in 2011. The lack of a predictable cycle and gains can make people hesitant, even if they don't want to see it go away. I'm personally less concerned. Looking at things going forward, I'm not tied to Mac specific software at this point. I can pretty much use whatever I want. If need be I'll go the custom build route for my own use (with the best parts of course), but I'm not typical. If the mini gets to the point of just barely good enough, I'll go that route and simply upgrade annually. In my opinion the mac pro won't survive at its current pricing structure. The quad mac pro has shown little growth. Sandy Bridge E shows little growth in that cpu variant (quad cpu at the $300 mark). They'd still need to implement a third party chipset for usb3 whether they go with sandy or ivy. Ivy Bridge E5s do not use a different chipset, therefore no usb3 is inclusive. SATA III and PCIe 3 do offer some help. Trying to push something at that price with the sludge like performance growth won't go anywhere. The more expensive models are likely worth retailing, but they need something to keep the volume above a minimum level.
Comments
Quote:
Originally Posted by Marvin
http://images.apple.com/finalcutpro/in-action/electric/
"We’re now shooting the show on the RED EPIC cameras in 4K ... and we’ve moved to Final Cut Pro X. We’ve been able to do things on Leverage that no other cable show does simply because we can afford to do it using our all-digital workflow. It’s very rare to see a television show that averages 40 digital effects per episode. Or has four- or five-day sound mixing sessions. We’re able to do it and still produce a show for basically $1.8 million an episode. We think that Final Cut Pro X shows how simply and inexpensively a powerful file-based workflow can be implemented - not only does it change the price, but it actually changes creatively the way in which we work. We don’t have to wait to lock picture to start our digital effects shop.”
If someone rests a $1.8m budget on the ability of a $300 app to produce the final product, they either have to replace it when it fails or the software can handle it just fine.
The FCP to FCPX transition wasn't very good in that FCP had been languishing, and FCPX debuted in a typical 1.0 product state with the older one immediately EOL. The rest goes back to what I've already mentioned. You need some combination of hardware and software that meet the requirements of an entire working solution. The primary advantage of the desktop solution there is flexibility. You can usually find some custom solution to make it work. It doesn't mean other machines won't work if they align well with the required tasks. The link you posted is interesting in that it's the kind of place you'd expect to see newer workflows adopted first. Larger productions may have far more pipeline dependencies that make sudden shifts extremely difficult. They may have a lot of custom code to make things work, and ultimately what is important is that work is completed on time to the best quality possible.
Quote:
Originally Posted by Marvin
Here's this posturing again that it's not possible to be a professional content creator with anything less than a Mac Pro....
It's possible to be a professional content creator with a Mac Mini. That doesn't make it preferable to use a Mini, does it?
[VIDEO]
^ Mac Pro buyer on the phone to the Apple Store wondering when the new one is coming out.[/quote]
Cute video
It depends on the job I'd say. If your work is more storage limited than processor-limited like with video editing then do you spend $2500 on a quad-core Mac Pro or do you spend $799 on a quad-core Mini and $1099 on a Pegasus R4? You still need to load the MP with drives and a RAID card to match the storage performance.
You can get a faster MP but you're pretty much at $3800 before you see the benefit, which is already double the Mini setup and again you have to load it up.
If you take costs out of the equation, faster is better but you wouldn't always recommend the $6200 model fully loaded with SSDs and 96GB RAM and the highest GPU because it's not always needed.
When you look at people who are at the top of their game, time and time again they show it doesn't matter what you use:
http://brightlightsfilm.com/78/78-cutters-way-the-mysterious-art-of-film-editing-walter-murch-margaret-booth_daseler.php
"Steven Spielberg, surprisingly enough, for all the wondrous effects he has brought to the screen in films like Jurassic Park (1993), was one of them. Of digital editing he has complained: "It doesn't smell like film. It smells like an electronic lab." In the '90s, seeing that the end was nigh, he began buying up old Moviolas and their spare parts to guard against a future shortage.
By the second decade of the new millennium, though, even he had succumbed to the inexorable pressure of modernity. His recent films Tintin (2011), War Horse (2011), and Lincoln (2012) were all cut on an Avid, though he insists he's anxious to return to analog as soon as possible. Such devotion to the past is comforting in a way, revealing that genius need not depend solely on flashy new technology, though it is certainly unusual these days. As Walter Murch wrote over a decade ago: "The persistence of the Moviola into the last decade of the twentieth century is about as surprising as seeing an old manual Underwood typewriter loaded onto the Space Shuttle."
http://www.btlnews.com/awards/contender-–-editor-michael-kahn-war-horse/
"“We all love the Moviola, that’s what got us were we are. Steve even loved the smell of it, the feel of it. He really enjoyed it,” reveals Kahn. “But when you sit down on an Avid and you want a trim right away, you don’t have to go to a box to get it, you just push a button and the trim is there. It is really quite an amazing machine and I wonder how they can build such a thing!” Kahn admits that he was fairly quick editing on film, but that Avid is ten times quicker for him. Film had given him the eye to quickly see matches, a skill which translated over to working on the computer."
They did amazing jobs manually cutting film in old mechanical machines and in some cases prefer it. You get a different mindset with limited technology. If you have to manually chop a piece of footage, it's a more important decision to make so you make it more carefully. When you work on a slower machine, you are similarly forced to find ways to optimize your work.
While having more resources to play with gives you a higher bound on creative freedom, it gives you a freedom to be wreckless too. It's like with large storage sizes and loading up your internal drives, at first it feels great to have 12TB internally but then you need a new machine, how do you move that 12TB of data to your new machine? What do you do when you run out of space or need to back it up, you can't put more inside so you have to go external after all. When you find one Mac Pro isn't fast enough, you have to start network rendering just like you would with any other machine.
Here's one where the Mac Pro buyer is explaining to Apple that they are the ones who are out there fighting for their cause but they just need to look less embarrassed doing it:
[VIDEO]
Quote:
Originally Posted by Marvin
They did amazing jobs manually cutting film in old mechanical machines and in some cases prefer it. You get a different mindset with limited technology. If you have to manually chop a piece of footage, it's a more important decision to make so you make it more carefully. When you work on a slower machine, you are similarly forced to find ways to optimize your work.
This is probably the best argument I've yet heard for the Mini or iMac over a Mac Pro. Get sh!t done slower, so you have more time to think. Think carefully about optimizing your work.
It's also amusing that the Apple hardware defenders keep mentioning the "Pegasus". That just proves how limited fast external drive options are for Apple's disposable hardware - you have one brand a couple models to choose. Of course Thunderbolt storage has limited bandwidth compared to the professional SAS solutions for a Mac Pro, but you know that.
You don't necessarily get things done noticeably slower. Productivity doesn't scale directly with raw hardware performance.
Having a machine that is 3x faster doesn't mean that you finish a job 3x quicker unless it's all raw processing. If you spend 20 hours editing a 60 minute video, it takes 20 hours on every machine. When you export, it might take 1.5 hours on the Mini and 0.5 hour on the top of the line Mac Pro (not likely as video encoding doesn't scale like that either). You've saved 1 hour out of 21.5 hours (5% more productive).
Even if you do effects and you render out a 15 minute render every hour on a low-end machine in an 8-hour day, the MP would save 8 x 10 minutes = 80 minutes.
HP tried to work this out and remember they are trying to sell people on the idea of buying higher powered machines:
http://h20331.www2.hp.com/Hpsub/downloads/Multi-Processor_WhitePaper_090611.pdf
Even though they are claiming 45-95% more productivity for adding a second CPU, they are quoting savings of 1.5-2 hours a day. That's nowhere near 45%, which is no doubt why they added their 'conservative' graph at the end. Sure an hour or two a day can save money over a period of time but a lot of things can save that time too and where does it end? Why stop at two processors, why not four? Does having two machines help? You can't just tack on power and assume it's going to make a big difference to a human-limited task.
Yes, this has been covered in the past. There are more drives than the Pegasus (Lacie for example and USB 3 drives), it just happens to be one of the better performing ones. SAS setups don't really offer much more practical bandwidth than Thunderbolt because you hit the read/write speeds of the drive much sooner, but you know that.
Quote:
Originally Posted by Marvin
While having more resources to play with gives you a higher bound on creative freedom, it gives you a freedom to be wreckless too. It's like with large storage sizes and loading up your internal drives, at first it feels great to have 12TB internally but then you need a new machine, how do you move that 12TB of data to your new machine? What do you do when you run out of space or need to back it up, you can't put more inside so you have to go external after all. When you find one Mac Pro isn't fast enough, you have to start network rendering just like you would with any other machine.
Here's one where the Mac Pro buyer is explaining to Apple that they are the ones who are out there fighting for their cause but they just need to look less embarrassed doing it:
You've drifted into hyperbole with the video links. I'm starting to wonder if Tallest gained control of your account. These problems are regularly addressed. If you're keeping a lot of storage internally, it's typical to leave the boot drive out of that. Single volumes can be transplanted later. I wouldn't personally use Raided internal drives for storage, which is what you seem to suggest here. With single volumes just move them. If you need real storage, you have a much wider range of available HBAs and RAID or JBOD boxes that can be used with a Mac Pro, although usb3 will probably cannibalize the lower end of this. The capabilities existed prior to thunderbolt. In case you are curious what annoys me, it comes down to the choices available when locked to a single company. With the imac, it would displace the use of superior displays. I can't fit both. The internals are probably enough. With the mini, the gpu would choke. I already choke it anyway. I've considered going to Windows simply because you can find a wider range of cards. If I did that, the machine I'd end up with would probably be something like an E5 hex core, 2 PCI cards with one being a mid range Quadro and the other being an HBA, as much ram as it would hold, and I'd stick with NEC or Eizo on displays going forward. Ironically I might consider a Fermi card. Kepler is terrible at floating point math.
They illustrate two very important points. The first is that Apple won't be convinced about the MP by writing letters to Tim or running a Facebook campaign. Apple isn't a charity, it's a business and everything they do needs to make business sense. If people want to keep seeing a new Mac Pro, show them the money - that's the only way they can measure that people still want them and holding onto them for 6 years isn't the way to do that.
The second is that Mac Pro buyers need to realise that Apple won't fail without them. They are the ones hanging by a thread. Apple is the most valuable company in the world right now and I don't think the magnitude of that has sunk in. They used to be near bankruptcy and relied on people buying their beige desktops. They don't have that reliance any more and there isn't a debt of gratitude to be paid to tower fans because iMacs and laptops are far more powerful than the equivalent towers they offered back then.
Some people suggest internal RAID is an advantage of a larger machine - that you don't need a 'rats nest' with an external RAID box or don't need to pay for something like a Pegasus. As you say, it's not a good idea to RAID your system drive and you wouldn't want to use software RAID on the 3 other drives nor would you use Apple's hardware RAID so you have to just buy external storage connected via PCI card using up a slot and creating a rats nest on top of the already huge Mac Pro. You might get away with a 3rd party RAID card but then you compromise reliability and stability with the drivers.
So, you can fit a Mac Pro with an 8" depth and a 27" diagonal plus a separate display but you can't fit an iMac with an 8" stand depth (much less for the rest) and a 27" diagonal with a separate display?
[IMG ALT=""]http://forums.appleinsider.com/content/type/61/id/19019/width/500/height/1000[/IMG]
It's pretty much to scale with all the components. There's plenty room in that shape for a powerful workstation. The side could be held on magnetically with pins and the base could have a gap large enough for fingers so it can be moved without trapping them while also allowing airflow underneath.
There might be enough room for more PCI slots if they wanted but they can also do dual 3-4TB Fusion drives. The power cable goes in at the bottom instead of the top so it doesn't get in the way of the ports. There's also no airflow blockage.
The top-end one could be about $4500 with a 12-core chip and they can offer whatever GPUs they like (with 6x Mini-DP outputs). The RAM would probably be in front of the CPU cooler too because they might not be able to make it a drawer and they'd want a guard round the cooler to prevent someone poking it while it was spinning.
Quote:
Originally Posted by Marvin
You don't necessarily get things done noticeably slower. Productivity doesn't scale directly with raw hardware performance.
Having a machine that is 3x faster doesn't mean that you finish a job 3x quicker unless it's all raw processing. If you spend 20 hours editing a 60 minute video, it takes 20 hours on every machine. When you export, it might take 1.5 hours on the Mini and 0.5 hour on the top of the line Mac Pro (not likely as video encoding doesn't scale like that either). You've saved 1 hour out of 21.5 hours (5% more productive).
Even if you do effects and you render out a 15 minute render every hour on a low-end machine in an 8-hour day, the MP would save 8 x 10 minutes = 80 minutes.
HP tried to work this out and remember they are trying to sell people on the idea of buying higher powered machines:
http://h20331.www2.hp.com/Hpsub/downloads/Multi-Processor_WhitePaper_090611.pdf
Even though they are claiming 45-95% more productivity for adding a second CPU, they are quoting savings of 1.5-2 hours a day. That's nowhere near 45%, which is no doubt why they added their 'conservative' graph at the end. Sure an hour or two a day can save money over a period of time but a lot of things can save that time too and where does it end? Why stop at two processors, why not four? Does having two machines help? You can't just tack on power and assume it's going to make a big difference to a human-limited task.
Yes, this has been covered in the past. There are more drives than the Pegasus (Lacie for example and USB 3 drives), it just happens to be one of the better performing ones. SAS setups don't really offer much more practical bandwidth than Thunderbolt because you hit the read/write speeds of the drive much sooner, but you know that.
So what exactly is your point? That we don't need fast hardware because slower works well enough? You used a lot of words to state the obvious, although things are more complex than your wordy post suggests. Faster hardware not only gets sh!t done faster in the present, it remains viable longer into the future. Contrary to Apple's landfill fodder business strategy, most people who buy $6000 Xeon workstations expect the hardware to not only support today's software, but tomorrow's. In other words, build a faster system, and developers will find a way to exploit all the power and then some.
About SAS storage, it can scale considerably better than TB, and more cheaply. Arguing in favor of TB at this point is just Apple fanboy delusions. The next revision of TB should double the number of available PCIe lanes, so then it could be a good solution depending on how many vendors support it.
Imagine a Mac Pro with 2TB of solid state storage in the boot / applications drive. Or maybe a spilt with 1TB allocated to the volume manage for tiered storage. It is nice to argue about the number of sockets in a machine but a total rethink of the Mac Pros architecture would go a very long way to making the machine viable over the long run. As much as we argue over the bottle necks in multi core or multi socket machines a very real and often overlooked choke point in modern systems is secondary storage. Without gains here significant gains in performance in the future may be harder to obtain no matter how many sockets or cores you have.
Seriously it is Apples responsibility to come up with an economical and suitable modern computer for the Pro and advanced users. To say that these people need to buy more often is just shows a massive misunderstanding of the market that such machines sell into. I can just imagine going up to the CEO of any company, big or small, and telling them that hey buddy you need to buy Mac Pros more often so that Apple will keep making them. The concept is hilarious to say the least. The fact that laptops are more powerful has little to do with the fact that they have ignored the desktop user. All we really want is to see the same level of innovation on the desktop that we have seen in the laptop line up. It is not unreasonable to demand that from a company like Apple. Like it or not having a respectable desktop lineup just shows your customer base that you are serious about Mac OS.
Now I have repeatedly stated that I think their desktop line up is borked. You basically have three models that are highly compromised and thus serve a limited scope of customers. This has impacted their sales of desktops and frankly it is pretty hard for anybody to deny this. All the way back to 2008 I made a decision to buy a MBP not because I really needed it but because there wasn't a viable desktop solution. Nothing has really changed other than the iMac & Mac Pro have gotten worst as far as being a viable desktops. I was this " close to buying a Mini in 2012 but Apple decided to bork that machine instead of upgrading the GPU to something a rational person would want. So really if you don't understand why people feel forced into buying laptops from Apple I really don't know what to say. Apple can do a lot better on the desktop but that entails scraping the entire lone up, adjusting margins and looking to the future. In stead we have Apple sitting on the desktop hardware letting it stagnate. Most of the above is BS. Internal RAID has had and still has its place in the world. So does software RAID and even Apples controller, it really depends upon use cases. However with the advent of TB I no longer see a pressing need for lots of internal drives. No more so than software issues can play havoc with external RAID boxes. It is certainly easier to unplug an external box and blame the vendor though. That wasn't what was said.
Quote:
Originally Posted by Marvin
They illustrate two very important points. The first is that Apple won't be convinced about the MP by writing letters to Tim or running a Facebook campaign. Apple isn't a charity, it's a business and everything they do needs to make business sense. If people want to keep seeing a new Mac Pro, show them the money - that's the only way they can measure that people still want them and holding onto them for 6 years isn't the way to do that.
Holding them 6 years isn't something that everyone does. You're examining a few forum posters, which is a silly way to extract data. The other thing you fail to account for is that the number of mac pros they sold in 2006 or the number of G5s they sold in 2004 may not satisfy them today, as their growth likely comes from other lines. There's a difference between slipping in units sold and simply being outpaced by the rest of their line. Outside of this last year, I expect the latter is more the case. That wouldn't get the mac pro off the hook if it's scheduled for cancellation. I'm just saying that you're looking at a limited market that has typically worked due to high margins. It's not so much of a dead market as you claim. It generates far better margins for oems than any of their lower products. Apple is in a unique situation. They've retained relatively high margins across their line, and their growth is driven elsewhere.
Quote:
The second is that Mac Pro buyers need to realise that Apple won't fail without them. They are the ones hanging by a thread. Apple is the most valuable company in the world right now and I don't think the magnitude of that has sunk in. They used to be near bankruptcy and relied on people buying their beige desktops. They don't have that reliance any more and there isn't a debt of gratitude to be paid to tower fans because iMacs and laptops are far more powerful than the equivalent towers they offered back then.
I'm not one of the people who buys into this idea. Some people are really weird in that they attempt to personify a corporation. Regarding the progression of lighter hardware, big points are growth in requirements and budget elasticity. The acceptance of lighter hardware has taken over somewhat on Windows too. The problem for some people is when you're dealing with a single company, if your requirements don't align well with what they make, you are out of luck. I've tested everything I use on Windows. While I'd have to make a couple hardware changes, it would work. I'm extremely annoyed by how ram hungry Lion and Mountain Lion have been. I max out memory constantly as some of the applications I deal with will use everything I throw at them. It annoys me when I feel that extra is tied up due to a bloated OS. Fedora runs a few things that I require, but some of them are problematic. I'd move to Linux if I could get everything working, just to have a nice light/stable OS. Ubuntu and Debian won't support certain things, or I would have switched already
Quote:
Some people suggest internal RAID is an advantage of a larger machine - that you don't need a 'rats nest' with an external RAID box or don't need to pay for something like a Pegasus. As you say, it's not a good idea to RAID your system drive and you wouldn't want to use software RAID on the 3 other drives nor would you use Apple's hardware RAID so you have to just buy external storage connected via PCI card using up a slot and creating a rats nest on top of the already huge Mac Pro. You might get away with a 3rd party RAID card but then you compromise reliability and stability with the drivers.
There are better boxes out there, and if we're talking about a 12TB raid, I'd go with WD's enterprise grade SATA drives. It comes out slightly more expensive, but you can put together a much more robust solution for a fairly similar amount of money.
Quote:
So, you can fit a Mac Pro with an 8" depth and a 27" diagonal plus a separate display but you can't fit an iMac with an 8" stand depth (much less for the rest) and a 27" diagonal with a separate display?
Is it difficult to understand they wouldn't occupy the same space? Towers typically fit beneath a desk. My point with the imac was that it would have to fit in side by side with another display. At 27", I'd need to make it the primary display viewed dead on due to size. It's not sufficient for that. The differences are somewhat hedged by tradeoffs. Mac pro can take a slightly wider range of gpus. A few people have gotten Fermi to work on it, which is stronger at floating point math than any Kepler card currently on the market. Fermi was just heavily optimized for computation. It can take a couple extra drives internally (note I didn't say a RAID storage array). It's still a bit faster. The imac provides a display within its total cost. That really is appealing, but Apple displays in my own experience have been plagued with too many problems. I have something that is relatively problem free at the moment. I'd like to retain that.
Quote:
Originally Posted by Marvin
Some people suggest internal RAID is an advantage of a larger machine - that you don't need a 'rats nest' with an external RAID box or don't need to pay for something like a Pegasus. As you say, it's not a good idea to RAID your system drive and you wouldn't want to use software RAID on the 3 other drives nor would you use Apple's hardware RAID so you have to just buy external storage connected via PCI card using up a slot and creating a rats nest on top of the already huge Mac Pro. You might get away with a 3rd party RAID card but then you compromise reliability and stability with the drivers.
So, you can fit a Mac Pro with an 8" depth and a 27" diagonal plus a separate display but you can't fit an iMac with an 8" stand depth (much less for the rest) and a 27" diagonal with a separate display?
Actually some of the third party ones have been better supported than Apple's own solutions. Have you ever looked at reviews on their Mac Pro raid card or DOA drive complaints with the Pegasus raid? Apple is the last one you should trust there. You mentioned 12TB. At that level if I had to put something together in the form of a hardware raid, it would likely run $2-3k. Apple charges $2300 for that. My way might end up more expensive, but I could put together a more stable solution with one of Areca's desktop raids and Western Digital RE4s. Apple is usually the problem when it comes to driver stability. If we're talking about $40 eSATA cards, that's a different story. Those things are sold on razor thin margins.
While it would be a bit tight the one thing I do like about getting down to about 8" is that you could design the product with a viable rack mount solution. That is a bracket and bolting arrangement to put two in a standard rack width. It doesn't have to be pretty as this would go into installations where people don't care.
Honestly though I'd rather that Apple completely broke with the past and just deleted traditional magnetic drive bays in a machine this small. Put in a couple of dedicated PCI Express slots and implement solid state storage. Frankly I'm not sure why Apple hasn't done this on the Mini yet. If magnetic drives are needed, for whatever reason put in a couple of slots/bays for laptop drives. I would rather have one accessible external PCI Express slot beyond the GPU slot. Further such a machine cries out for a base model with an integrated GPU. Intel would do well to offer up a socket that supports GPU enabled processors or more traditional multi core processors. Such a beast would allow Apple to support a wider array of customers on one platform. They might even be able to ditch the Mini. I'm still wondering how Apple will address the TB issues with the new Pro Machine. I could see them soldering the GPU right on the motherboard. This is one thing about the TB initiative that is still a mystery. Why bother with a drawer? Seriously build everything onto the motherboard or at least as much as possible putting slower parts on a daughter card.
I still think 8" is a little tight for a computer bound for the Pro market but you are on the right track trying to condense the machine down to the basic elements of a Pro machine. What will be interesting is the port selection, I'm just interested in what Apple sees as appropriate for the next 6 years or so.
Not quite, I'm not suggesting everyone should be happy with Macbook Airs. I'm saying that there are diminishing returns on investing in the fastest hardware the more that time goes on.
Say, for example, Apple put up a new computer tomorrow that had 4x E5-2687W CPUs, dual GTX 690s and then slapped a price tag of $11,000 on it. Is that something people would want? If raw performance is so important then why wouldn't people want it?
If we were back in 1995, those kind of prices meant the difference between getting a job done and not getting it done because some tasks couldn't be done with too little power or too little RAM. These days, the gains in productivity from faster hardware are getting lower because most of what we do is real-time.
Once the productivity gains from a new machine aren't significant enough, people hold onto machines longer. Then the company selling them doesn't see enough demand so they focus on hardware that does see the demand. Eventually an upgrade is required and there's no new model to be found.
Yes, typically 3 years longer. Hardware goes up about 50% in performance every year (it doesn't do this every year but it tends to average out at this over a period of time) so in 3 years, we get 1.5^3 = ~3x faster. The top-end MP has a 3x faster CPU than the top-end iMac so in 3 years they are about the same. The GPUs make the gap even smaller because the GTX680M in the iMac is 70% of the performance of the GTX680 that would have gone into the Pro so the Pro would barely be a year ahead in terms of the GPU.
This trend may change if Intel decides to bump CPUs up 15% year on year and focus on GPUs but I think they'll focus more on CPUs again once the GPUs move up high enough.
Even a 12-core MP would lag if you max it out though. That's just a case of limiting the processes e.g give it 6 out of 8 threads and 12GB RAM out of 16GB or whatever.
I think it always comes down to balancing needs and costs and it could be the case that there are enough people willing to pay for a $4000+ machine to justify 2P machines and not enough to justify a $10,000 4P machine but I think fewer and fewer people are willing to spend $4000+ on computer equipment now. Here's an example of someone who replaced XServes with Minis:
http://www.cultofmac.com/78914/why-did-apple-discontinue-xserve-because-its-a-dinosaur/
This sort of thing will happen the more that cheaper and lower power hardware can do the jobs of the more expensive machines.
It might seem hilarious but it's the way supply and demand works. Take away the demand and you take away the supply. You're right the supplier has a responsibility to create a demand for something but Apple has been at this a very long time and they know how big the markets are in each sector and their supply priorities reflect that.
Even if Apple brought out the most awesome Mac Pro anyone can imagine next year, it still won't sell in huge numbers with a starting price above $2000 and they won't drop the price lower than that. So making a great machine won't make a bit of difference unless that market grows substantially or they cut the price massively. As I showed earlier, lowering the margins doesn't give a direct increase in volume.
It keeps swinging back and forth. Either the Mac Pro is great because it lasts, which means people are holding onto them or after 3-4 years, they are no faster than the lower end and people are upgrading. If the lower end matches them in performance after that shorter period, maybe these people switch to the lower end models but then people say that can't be happening because the lower end doesn't offer xyz benefits. It can't be all things at once.
I think a lot of people are migrating to lower machines as time goes on, this shows in the buying rates for laptops vs desktops and iMacs vs towers. I think tower buyers aren't upgrading often either. Both of these situations create a smaller demand for the Mac Pro and this shows in Apple's treatment of it.
Before the iMac though, the towers used to be priced a lot lower. They displaced them in favour of the iMac in order to offer the best quality for the whole package in that range and make something compelling to buy. The $2500+ target audience is small and the market which would normally have had to go with a tower (DTP / CAD / video editing markets) can now get the work done on something much cheaper and even portable.
While the number of people in the high-resource audience continues to grow, the lower end hardware meets more needs and the higher end hardware encourages people to upgrade less.
I would say it's wasteful to stick it on the floor but you can do it if you want or even just sit the NEC/Eizo in front of it. It's still cheaper than the equivalent Mac Pro.
Maybe but every machine has a power target. The Mini is 85W, the iMac is 310W, the MP is currently 1000W. They'd just be making the Mini less attractive to people who like a tiny desktop machine and want to run it as a low power server.
Quote:
Originally Posted by wizard69
What was and still is silly, is the idea the he offered up that poor Mac Pro sales are our fault because we have not purchased Mac Pros frequently enough. That is absurd and frankly irrational. I know their is a certain element in society that has to go out and buy a new iPhone every time one comes out. That is not the Pro market at all. Now some business can justify fairly quick replacement schedules but it is not odd at all to see the same machine in use for five years or more in business.
From what I've been able to determine it is slippage in sales that is the problem, more so a very significant slippage in sales. Of the desktops only the iMac has been positive in the USA.
That doesn't surprise me. A lot of the cliche Mac customers seem to be trending toward the use of imacs. My suggestion is that they purchase directly from Apple. If the display exhibits poor uniformity, return it. I've also explained to some of them what display calibration can and cannot fix and why they shouldn't cheap out on drives for their RAIDs.
Quote:
When looking at value, from the consumers point of view, the current Mac Pro is a big zero in the base models. This drives sales downward as the market for the high performance versions is more limited. I really don't see the concept of a high performance Mac Pro as so much of an issues, as is Apples big gap between the Mini and the tricked out Mac Pro. From my perspective what is killing the Mac Pro more than anything else is that apple doesn't have a midrange desktop machine. That machine could be derived from the Mac Pro or not but the lack of such a machine drive the perception that Apple has nothing worthwhile on the desktop.
The markets they're likely to approach with such a machine are somewhat limited. To maintain health with such a line, they'd need to attract customers that already buy into this hardware class from other vendors. The workstation market has shown growth in many quarters. 2010 showed some amount of growth. 2011 and 2012 were slow for obvious reasons. Budgets aren't necessarily elastic, and below $3000 + display, things have been moving very slowly.
Quote:
The Windows world has basically crashed. People have literally jumped off the treadmill.
I've noticed this to a degree, but it addresses certain things well enough, and you can configure decent hardware to the required spec quite easily. I tend to explore solutions due to a need to ensure I can make something work, and I've been a bit disappointed with OSX lately. I need the leanest OS possible. Windows isn't really that. It's just that it has a better range of gpus and bit more freedom in terms of customization. I don't think it's irrational to maintain contingency plans. I don't maintain a notion that I would somehow be holding Apple captive if they don't produce the machine I want to buy.
Quote:
I ran Linux from about the time NT service pack 3 came out until 2008 when I went with a Mac. I can not understate just how frustrating Linux can be. You have a system built around a decent kernel that really could go somewhere if they could just address the UI issues that have been there for ever. Well that and the constant need to do complete installs every 6 months. Linux is a nice hobby, don't get me wrong as I still have a machine or two at the house, but it isn't the place to go if you want the hardware and software to work reliably.
The vfx industry incorporates Linux in many areas, and I use a lot of the same software. Because of this there certified kernels and installation instructions for a lot of software. I've read them, so I know how much of a pain in the ass it is. I've been a bit disappointed in OSX with Lion onward. Snow Leopard was great. It's not possible to stay on older OS revisions forever, as you eventually miss too much in the way of added frameworks and software support. I usually update my OS mid cycle after reading through lists of known bugs for any critical software. It's quite boring.
Quote:
After my stint with Linux I've gone all in with Apples software. Interestingly when I first got the Mac I did try using some of the open source stuff (mostly desktop productivity) that ran on the Mac. Eventually I realized that this stuff suffered from the same issues as the software running on Linux. Eventually I installed the Apple software and frankly it is good enough with zero hassles.
There isn't a lot that I really need. I use a huge range of software. Some of it is used rather infrequently. FCPX isn't necessary as Creative Suite includes Premiere, even though I need it very rarely. If I didn't have Creative Suite, it would be different. The upgrade costs aren't that bad once you're past the upfront costs. The suite upgrade prices aren't much higher than single applications. This means it's not as big of a deal retaining software that I need less frequently. It used to be that you could upgrade up to 3 versions back with individual software products. Adobe reduced that to one to push their suites on everyone. It's annoying, but you get a lot of capability. I never personally liked Aperture. The rest isn't much different on Windows or OSX.
Quote:
The above highlights possible the biggest issue with the Mac, once you get outside of the Mac mainstream as far as software is concerned your choice dwindle fast. The need for a hacker mentality to get things to work is almost as bad as Linux. Some companies just seem to have a serious inability to support their hardware on the Mac. A further software, especially in the engineering world has been slow to come to the Mac. So it does feel at times like the Linux world. However Max OS/X redeems itself via iOS integration and a bunch of apps that really are state of the art as far as usability goes.
I've definitely mentioned that before. I do use some open source software, but I'm reasonably selective. I tend to avoid the "labor of love" projects supported by a single developer for anything important. I don't consider the RAID thing to be a hacker mentality. Marvin and I disagree on the quality of solutions provided by Apple at times. Apple Store reviews on the thunderbolt raid mention doa drives. This shouldn't really happen with a pre-configured raid solution, which should have been tested prior to shipping. It's not the same as just buying the drives as raw components on their own. I'd just stick with solutions that have been well supported in the past, and for a 12TB raid, you should be really picky on the hard drives used. The cheapest I'd go would be Caviar Blacks, as they use similar hardware to the RE4s. The RE4s have different firmware. iOS integration isn't that big of a deal to me. I have an iphone, and it's helpful, but not entirely crucial for me. The iphone was kind of an easy solution. It covers my requirements, but I will never wait outside the store to buy the latest one.
The popularity of the iMacs isn't really a new thing. Even if you go back to the year 2000, Apple reported the split in numbers and the iMac outsold everything else:
http://www.apple.com/pr/library/2000/01/19Apple-Reports-First-Quarter-Profit-of-183-Million.html
"Sales of 1,377,000 systems including over 700,000 iMac™ consumer desktops and 235,000 iBook™ consumer portables"
This was back when the entry tower desktop was $1599 and the entry iMac was $799. It doesn't say how many Powerbooks vs Powermacs were in the remaining 442,000 units but if it was the same ratio as iMacs to iBooks, it would be 150k PBs and 290k PMs.
Knowing that Mac Pro shipments are under 100k / quarter now, it's clear that the tower desktop market has shrunk considerably or I guess a more accurate way to put it would be that people are happy to spend a $1000-2000 budget on an all-in-one instead of a tower and Apple makes better margins that way as they sell the display too and the way they make better margins from the towers is jack up the price. If they sold the towers at the same price as the iMacs again, it wouldn't be worth the trouble because you can bet not everyone would buy Apple's displays (like people who prefer NEC/Eizo).
To drive the growth that people here suggest, Apple would need to take the display out of the iMac and sell that at a lower price. But as I say, they just lose a display sale and have a machine without the iMac's unique selling points. If they use the same parts as the Mac Pro, they basically just have to cut their margins but they will still hit the upper end of the $1000-2000 range and they won't even double their volume.
The best route forward for them is to just play it out the same way they've been doing already. Scale down the operation, ship it back to the US so the machines can be assembled there (one factory can probably handle the 2,500 / day worldwide target). They can keep it running as long as the shipment volume stays above a certain level and the profit it makes is worth the effort. But regardless of whether people agree with it, people still need to buy them more often or their shipment volume reports will just slide and they'll think people don't want them any more. They might not drop the line for a few years but you can bet they'd stick with the long refresh cycles.
Quote:
Originally Posted by Marvin
The popularity of the iMacs isn't really a new thing. Even if you go back to the year 2000, Apple reported the split in numbers and the iMac outsold everything else:
http://www.apple.com/pr/library/2000/01/19Apple-Reports-First-Quarter-Profit-of-183-Million.html
"Sales of 1,377,000 systems including over 700,000 iMac™ consumer desktops and 235,000 iBook™ consumer portables"
This was back when the entry tower desktop was $1599 and the entry iMac was $799. It doesn't say how many Powerbooks vs Powermacs were in the remaining 442,000 units but if it was the same ratio as iMacs to iBooks, it would be 150k PBs and 290k PMs.
I didn't realize it took over that early.
Quote:
Knowing that Mac Pro shipments are under 100k / quarter now, it's clear that the tower desktop market has shrunk considerably or I guess a more accurate way to put it would be that people are happy to spend a $1000-2000 budget on an all-in-one instead of a tower and Apple makes better margins that way as they sell the display too and the way they make better margins from the towers is jack up the price. If they sold the towers at the same price as the iMacs again, it wouldn't be worth the trouble because you can bet not everyone would buy Apple's displays (like people who prefer NEC/Eizo).
I get that. I tend to view things relative to my own needs. I'm not terribly biased against Apple displays. I just don't think they're as good, having compared on many occasions through different design generations. I'm confident that advancements in technology will bring these things further together. They all pretty much use the same panels at this point. A few years ago this wasn't the case. Some of the other vendors pulled out due to shrinking profits. As you know LG maintains an existence through extreme volume.
Quote:
To drive the growth that people here suggest, Apple would need to take the display out of the iMac and sell that at a lower price. But as I say, they just lose a display sale and have a machine without the iMac's unique selling points. If they use the same parts as the Mac Pro, they basically just have to cut their margins but they will still hit the upper end of the $1000-2000 range and they won't even double their volume.
That isn't necessarily what I'm saying they should map out as a current trajectory. That's basically something that would benefit me. I'm not arrogant enough to feel that would influence their business decisions.
Quote:
The best route forward for them is to just play it out the same way they've been doing already. Scale down the operation, ship it back to the US so the machines can be assembled there (one factory can probably handle the 2,500 / day worldwide target). They can keep it running as long as the shipment volume stays above a certain level and the profit it makes is worth the effort. But regardless of whether people agree with it, people still need to buy them more often or their shipment volume reports will just slide and they'll think people don't want them any more. They might not drop the line for a few years but you can bet they'd stick with the long refresh cycles.
Part of that is intel. 2009 to 2010 didn't provide much. Westmere was an incomplete lineup. Sandy Bridge E came late. That's a lot of flat growth in the lower tiers of the mac pro line, which likely carry a fair amount of its volume. They haven't maintained a lot of mac pro models overall. The 1,1 sold until the end of the first quarter in 2008. Given that many applications didn't release universal binary versions until at least 2007, it's reasonable to figure that many of those purchases were likely in 2007. In some of the edge cases, I can see how they would have been disappointed with the 2010 update and waited for Sandy in 2011. The lack of a predictable cycle and gains can make people hesitant, even if they don't want to see it go away. I'm personally less concerned. Looking at things going forward, I'm not tied to Mac specific software at this point. I can pretty much use whatever I want. If need be I'll go the custom build route for my own use (with the best parts of course), but I'm not typical. If the mini gets to the point of just barely good enough, I'll go that route and simply upgrade annually. In my opinion the mac pro won't survive at its current pricing structure. The quad mac pro has shown little growth. Sandy Bridge E shows little growth in that cpu variant (quad cpu at the $300 mark). They'd still need to implement a third party chipset for usb3 whether they go with sandy or ivy. Ivy Bridge E5s do not use a different chipset, therefore no usb3 is inclusive. SATA III and PCIe 3 do offer some help. Trying to push something at that price with the sludge like performance growth won't go anywhere. The more expensive models are likely worth retailing, but they need something to keep the volume above a minimum level.