Any case redesign would likely do away with internal hard drives all together or at least get rid of the array support. Supporting disk array internally is a step in the wrong direction, the internals need to be simplified so that maybe (hopefully) Apple can update the machine at regular intervals.
Don't knock it before you see it. Also rethink how you handle bulk storage. Why? Because I don't think there is a chance in hell that Apple will go in that direction. To put it simply the market neither needs nor wants such machines.
What!!! Are you kidding? A workstation without internal hard drives. That's not a workstation at all. That would almost certainly kill the product and drive away the customers who used to purchase it.
You thinking is from the dark ages of computing, there is no rational for magnetic drives in a modern computer. To put it simply they are to damn slow. Secondary storage in a modern workstation should be solid state sitting on PCI-Express buses. Bulk storage can sit out side the box on any number of alternative interfaces.
As for customers that use to purchase the machine, they can either get with the program or break out their VIC 20's. I mean really guys what is with people that want to regress to the dark ages? I've been using various computers since that VIC 20 mentioned above, the last thing I want to do is to see technology stagnate because a couple of fools can't open their minds to new ways of doing things.
Frankly we have not even gotten a clue as to what Apples new pro will look like. When it does come it might be a good idea to concentrate on what is good about the new machine instead of what is missing. Frankly this point of view is as silly as the people complaining about the optical missing in the new iMac.
"Simplifying the internals" has absolutely nothing to do with Apple's lack of commitment to the Mac Pro lineup or the long interval between updates.
You missed the point, Apple needs a Pro type machine that can grab some decent sales. A simplified machine means that Apple can offer a platform at a much low cost than today's machine.
On the flip side a complex machine is far harder to update. As such it is a factor in update cycles.
Apple have said in the past, as I think was probably pointed out somewhere in this thread, that they were not going to add USB 3 until it was natively supported, which, supposedly, will be the case with the Ivy Bridge Xeons. There really is no excuse for Apple's laggardly approach to this product in recent years.
The "excuse" is the lack of sales. The lack of sales is the only reason why I expect to see a heavily refactored Mac Pro. Lets face it there wasa huge demand for the Mac Pro Apple would be paying more attention to the platform. So really I don't think Apple has a choice, either they make a Mac Pro that appeals to a wider audience or they scrap the machine. There is virtually no demand for the machine you are asking for.
I have no idea where you get the idea that the market neither needs nor wants "such machines". That plainly is not the case.
Apples sales? Seriously guy take a good long look at Apples sales, 80% of Apples Mac sales are laptops, the iMac takes up the vast majority of the desktop sales and then the Mini follows that. So realistically what do you think Mac Pro sales are?
Lets say it is 3% of Mac sales, of that 3% how many of those users do you think make use of internal disk arrays? I'm going to take a shot here and say it is les than 10% of that 3%. Internal disk arrays are a thing of the past, the demand isn't there to justify passing the expense on to every Mac user.
As for Blu-ray, it is unlikely that Apple will add that any time soon, if ever. On the other hand, there are third party solutions.
There are more people than Apple is willing to admit who are sitting on the fence, waiting to see what Apple does next. They will go elsewhere if necessary, but, if they do, it will be because Apple drove them off.
No rational business person is going to commit to a product they haven't seen. However good business man making their tools work for them. Anybody that rushes off in a fit because Apple didn't do it their way isn't the type of customer Apple needs. Frankly I'm tired of hearing from people that want to drag Apple back into the last century technology wise. At the very least make a thoughtful and rational evaluation of where Apple takes the Mac Pro.
For the OP, a hackintosh will work for a number of solutions, but there are not that many Xeon hackintoshes out there and, if Apple drops the Mac Pro, there simply may not be adequate OS support for the hardware.
Bingo. As for hackintoshes, lots of luck structuring a business around that sort of crap. All that wasted effort would be better spent on simply configuring the new Mac Pro to work for you.
As a 12 core user, you certainly are in need of more than an i7 quad core which is limited to 32 GB RAM (which is not error correcting).
Do realize that very shortly we will have chips with 12 cores and accelerators with +50 intel compatible cores. Intel is also working on other members of the Phi family with built in super computing networking and other features. Frankly we have all sorts of new technologies on the hip origin that could go into the New Mac Pro, some of these technologies could dramatically alter how the Mac Pro is seen as a pro computer. It might not be the image you have burnt into our mind but that would not make it less of a pro machine. The whole point I'm trying to make in this thread is that Apple can't build a machine for past user needs and expect it to sell or for that matter to even get people to note it. They need to build a machine that is built for the future.
No rational business person is going to commit to a product they haven't seen. However good business man making their tools work for them. Anybody that rushes off in a fit because Apple didn't do it their way isn't the type of customer Apple needs. Frankly I'm tired of hearing from people that want to drag Apple back into the last century technology wise. At the very least make a thoughtful and rational evaluation of where Apple takes the Mac Pro.
Bingo. As for hackintoshes, lots of luck structuring a business around that sort of crap. All that wasted effort would be better spent on simply configuring the new Mac Pro to work for you.
I don't agree with everything in that post, but I'm not feeling like breaking it down regarding storage and things. 64 bit computing did alleviate the use of scratch disks in a lot of cases. This was a big use of internal bays. I still think they're useful for storage, but 64GB of ram really kills the need for scratch data in most use cases. When these applications could only access 3GB directly, they had to dump it somewhere, so an internal RAID 0 was often the best solution.
Regarding hackintoshes, no matter how many people claim to own stable ones, I'd never go that route for a business. I've never suggested one to anyone I know that uses Macs within their businesses. A hacked solution is just a bad idea there. I think if the mac pro and Apple's other offerings are a bad solution, you should look at Windows rather than attempt an unsupported configuration. They're really better for hobbyists. If you look at insanely mac, many of them own macbook pros and hackintoshes. DIY computing can be cheap enough to take up as a very nerdy hobby.
No rational business person is going to commit to a product they haven't seen. However good business man making their tools work for them. Anybody that rushes off in a fit because Apple didn't do it their way isn't the type of customer Apple needs. Frankly I'm tired of hearing from people that want to drag Apple back into the last century technology wise. At the very least make a thoughtful and rational evaluation of where Apple takes the Mac Pro.
Bingo. As for hackintoshes, lots of luck structuring a business around that sort of crap. All that wasted effort would be better spent on simply configuring the new Mac Pro to work for you.
I don't agree with everything in that post, but I'm not feeling like breaking it down regarding storage and things. 64 bit computing did alleviate the use of scratch disks in a lot of cases. This was a big use of internal bays. I still think they're useful for storage, but 64GB of ram really kills the need for scratch data in most use cases. When these applications could only access 3GB directly, they had to dump it somewhere, so an internal RAID 0 was often the best solution.
The big problem with building a Mac Pro big enough to support a disk array of rotating media in my mind is that the vast majority of Mac Pro users don't need a built in solution. As such it makes the Mac Pro more expensive than it needs to be and slows development cycles.
Note that I'm not saying that some users don't need an array just that these days it makes more sense to put the array in an external box. An external box provides the user with certain advantages one important one being that it decouples the Mac Pro and array storage technologies.
On the flip side I realize that internal storage, that is secondary storage inside the Mac Pro is important. However looking forward this storage has to move off the SATA bus to maximize transfer rates. So Apple would be far better off basically saying the same thing about magnetic drives as the have about optical, the technology has run its course in the case of a high performance work station! Needs do vary but there is nothing to stop Apple from introducing two or more storage dedicated PCI-Express slots.
Regarding hackintoshes, no matter how many people claim to own stable ones, I'd never go that route for a business. I've never suggested one to anyone I know that uses Macs within their businesses. A hacked solution is just a bad idea there.
I agree 100%! There is just too much that can go wrong for most businesses. Even if the guy doing the Hackintosh is competent it is still a significant drain on business resources.
I think if the mac pro and Apple's other offerings are a bad solution, you should look at Windows rather than attempt an unsupported configuration. They're really better for hobbyists. If you look at insanely mac, many of them own macbook pros and hackintoshes. DIY computing can be cheap enough to take up as a very nerdy hobby.
Well I'm not sure about Windows????.
I like to look at it this way, if you run a little tool and die business you may have a couple of CNC machines that you pay for over a long number of years. Over those years the machines are bolted and grouted to the floor as such you can't change them out every time a new project comes in. Instead you learn to make the machine do what you need done.
Buying a computer and using a computer isn't much different. It isn't normally bolted to the floor but you can't expect to change it out with every new project that comes along. In effect you need to make money off that hardware until it makes business sense to upgrade. So you learn to make the platform you got do the job at hand. In the end I expect next years Mac Pro to be a faster machine that may require adaptation on established user configuration. This isn't a bad thing at all but it does make some people feel uncomfortable and in some case down right angry. In some cases you will never get an individual to see the light, Apple just has to hope that the new machine has enough allure that it causes people to reconsider their preconception of what a pro computer is.
The big problem with building a Mac Pro big enough to support a disk array of rotating media in my mind is that the vast majority of Mac Pro users don't need a built in solution. As such it makes the Mac Pro more expensive than it needs to be and slows development cycles.
Note that I'm not saying that some users don't need an array just that these days it makes more sense to put the array in an external box. An external box provides the user with certain advantages one important one being that it decouples the Mac Pro and array storage technologies.
On the flip side I realize that internal storage, that is secondary storage inside the Mac Pro is important. However looking forward this storage has to move off the SATA bus to maximize transfer rates. So Apple would be far better off basically saying the same thing about magnetic drives as the have about optical, the technology has run its course in the case of a high performance work station! Needs do vary but there is nothing to stop Apple from introducing two or more storage dedicated PCI-Express slots.
Well things consume PCI lanes even if they're run as SATA. I think the stick type ssds Apple uses still show up as SATA. They may have their own controllers.
Quote:
I agree 100%! There is just too much that can go wrong for most businesses. Even if the guy doing the Hackintosh is competent it is still a significant drain on business resources.
I'm not a fan of hacked solutions for business. Windows actually has certain advantages in software. Their gpu drivers also have some features that Apple is unlikely to support in the next couple years, such as full support for displayport 1.2.
Quote:
Well I'm not sure about Windows????.
I like to look at it this way, if you run a little tool and die business you may have a couple of CNC machines that you pay for over a long number of years. Over those years the machines are bolted and grouted to the floor as such you can't change them out every time a new project comes in. Instead you learn to make the machine do what you need done.
Buying a computer and using a computer isn't much different. It isn't normally bolted to the floor but you can't expect to change it out with every new project that comes along. In effect you need to make money off that hardware until it makes business sense to upgrade. So you learn to make the platform you got do the job at hand. In the end I expect next years Mac Pro to be a faster machine that may require adaptation on established user configuration. This isn't a bad thing at all but it does make some people feel uncomfortable and in some case down right angry. In some cases you will never get an individual to see the light, Apple just has to hope that the new machine has enough allure that it causes people to reconsider their preconception of what a pro computer is.
Some workflows are much more portable than others. Going back a few years, things were much more segregated. Certain things either wouldn't run on Macs, or they got terrible ports. It had certain niche markets that were supported really well, so you could have certain Mac specific plugins. It was also a lot of work dealing with the different file systems in a mixed environment. Today this is less of an issue. My point was merely that if they're not going to buy a Mac, they shouldn't try to prolong the use of OSX with a hacked solution. It is typically a long term migration if you switch your OS. Anyway what I've suggested for the mac pro is that they need some kind of growth leverage outside of Macs. The workstation market isn't really growing, so they can't rely on inertia. The health of the line would likely rely on making something attractive enough to leverage more users away from Windows boxes or occasionally Linux (popular in larger film studios).
My current desktop is a 3.46 GHz Hexa-core Mac Pro. It has two SSDs and four HDDs (two 2 TB, two 3 TB). Both SSDs are on a 6G SATA PCIe card, and I've added eSATA for external backups And yes, it has a Blu-ray drive, which I use all the time to "consume" movies. BD offers superior quality over streaming compressed 1080p content, just like CDs offer higher quality than iTunes downloads.
For my needs this is a dream system, and no it is NOT overkill. I use it for lightroom, photoshop, and HDR processing, along with the usual email, word processing, and internet. When I added all those HDDs to it, I could buy bare HDDs, no need for expensive TB enclosures or a rat's nest of cables. Upgrading to 6G SATA was simple, but honestly the speed difference is negligible for my needs. This computer has more power than I need at the moment, but I plan to use it for another 5-7 years, and it may be good to go for another ten.
Could I do the same with an iMac? Yes, but it would be one ugly, space consuming setup. I'd have either a bunch drive enclosures, or one TB 4 bay enclosure that cost a crazy amount of money. Also, an external BD drive. The Mac Pro takes up far less space, and it is much cheaper to upgrade than an iMac. If a drive goes bad, I just swap it out. On an iMac, if a drive goes bad I have to lug it into an Apple servicer for replacement. Best of all, I can choose what monitors I want to use.
If Apple wants to sell more towers, I'd advise them to offer a non-Xeon tower with internals comparable to the iMac in performance. Use desktop components: they're cheaper and faster. And for the love of God, lose the thin fetish! What does thin get you with an iMac? Overheating, less expandability, less capability. You can't even see how thick it is when you're using it, unless it's to bend over and around to reach the card reader that is inexplicably on the back of the iMac. Design a tower in which form follows function.
If Apple wants to ditch their most loyal customer base - the content creation industry, that is their right, but it would be like Chevy EOLing the Corvette. Sure, the Corvette doesn't make Chevy much return on the R&D investment. They may even come close to breaking even. But Chevy knows that the Corvette is part of what defines their brand, and it showcases their engineering chops. Same with the Mac Pro. I've heard countless times from people who comment that their company uses Macs for all video and graphic arts, and many of these people buy a Mac precisely for this reason. The Mac Pro, or more accurately, Apple's presence in the creative arts community, helps define Apple as a serious player. Without content creation, Apple is a gadget maker who dabbles in desktop computing - not a computer someone buys for serious work.
Of course Apple can discontinue all their desktops and laptops that aren't as profitable as their iBaubles, and they will remain profitable. But one day, their dominance in the gadget markets will ebb, and then what? Apple will just be another phone maker among dozens. Nothing special, and certainly not a company serious users turn to for computing solutions.
For my needs this is a dream system, and no it is NOT overkill. I use it for lightroom, photoshop, and HDR processing, along with the usual email, word processing, and internet. When I added all those HDDs to it, I could buy bare HDDs, no need for expensive TB enclosures or a rat's nest of cables. Upgrading to 6G SATA was simple, but honestly the speed difference is negligible for my needs. This computer has more power than I need at the moment, but I plan to use it for another 5-7 years, and it may be good to go for another ten.
That was my point on growth. The system would receive more attention if it had the power to encourage further rounds of upgrades and leverage new customers from Windows equivalents. You stick with what you have because it works. A new one wouldn't significantly improve the way you're able to work. If LR raw processing was in real time or the machine could interpret raws in real time while retaining more information and applying adjustments at a lower level, it would influence many people like you to invest in new hardware, as it would change the way you work. If you compare Lightroom today to the early Sinar digital back (which had to be tethered via firewire) software, you'd see what I mean. People don't just upgrade based on perceived speed. There are cyclical upgrades, but the time when you see many people upgrade simultaneously is when it changes the way they can work. Otherwise you're already looking at sunken costs and a workflow that works.
That's certainly true about encouraging new rounds of upgrades, and it could explain why Apple hasn't upgraded the Mac Pro recently. The new Xeon processors don't offer the sorts of performance increases you describe, and at least in my experience, SATA III isn't enough of an improvement over SATA II to make me want a whole new system. To hear people whining about switching to windows over the lack of SATA III and USB 3.0 seems silly IMO.
There is something new that will be shipping in volume next year that could enable the sort of performance leap you describe, hmmm. Could be that the new Mac Pro is designed around it. It would explain the delay, and Apple's confidence in waiting. It could differentiate Mac Pros from other Xeon workstations if cleverly implemented. And it would require Apple and Intel to work closely together to bring a product to fruition, something Intel has been doing with other companies on this new innovation. Knight's Corner. :P
In any event, even without me buying a new Mac Pro, Apple still gets plenty of business out of me. I buy every OS X release, and my share of iBaubles. Perhaps even more importantly, I prosyletize OS X and help Windows users switch to Macs. They see my Mac Pro, see what I do with it, are amazed that Apple does more than make phones and ipods. Then they go out and buy iMacs. Did my Mac Pro directly lead to the iMac sales? Not really, but I can't say it did not either. It's part of the Apple ecosystem, and people respond to a broad computing ecosystem. Call it the bandwagon effect. The 17" MacBook Pro has the same effect - "whoa, that's a sweet laptop! I don't need anything that big, but I want a MacBook now!"
That's certainly true about encouraging new rounds of upgrades, and it could explain why Apple hasn't upgraded the Mac Pro recently. The new Xeon processors don't offer the sorts of performance increases you describe, and at least in my experience, SATA III isn't enough of an improvement over SATA II to make me want a whole new system. To hear people whining about switching to windows over the lack of SATA III and USB 3.0 seems silly IMO.
You really need to do your reading. The available chipsets will still be the same damn thing, and you're talking about 225-300W specialized cards. The $400 ones seem to be testing samples. I don't expect them to see volume sales at that. While it would be awesome to see if that thing can run OpenCL, this doesn't seem like something Apple would do. The current case also lacks enough cooling to dissipate that off a single card, and that is the lower power version. The mac pro actually has enough problems with cards approaching 200W, so you'd definitely require redesigned airflow. Even then I don't see it. They ignored Teslas. They never made CUDA any kind of priority even though early development exceeded OpenCL, and NVidia cards Fermi and on support OpenCL. What would cause the change in behavior now?
If you want to weave Xeon Phis into your supercomputers for number-crunching offload, then you probably will want the passively cooled Xeon Phi 5110P PCIe card. This one has more cores fired up and a slightly slower clock speed, and can deliver its 1.01 teraflops within a 225-watt power envelope – the same thermal limit that other GPU coprocessors for servers need to stay within. The 5110P card has the Xeon Phi chip with 60 cores, 30MB of cache memory on the die, plus 8GB of GDDR5 memory and a peak of 320GB/sec of memory bandwidth.
Obviously it would have to be a custom solution, not simple support for a PCIe card. That's why I mentioned that Apple and Intel would have to work together on it.
Since the OP kicked this off with an ARM-based Mac Pro, I figured this wasn't a reality-based thread.
The big problem with building a Mac Pro big enough to support a disk array of rotating media in my mind is that the vast majority of Mac Pro users don't need a built in solution. As such it makes the Mac Pro more expensive than it needs to be and slows development cycles.
With SSDs there's not much use of internal RAID, if that's what you mean. There is still a need for big, cheap storage space, and internal HDDs fit the bill. If anything there is more need for internal HDDs now than before SSDs. Previously, even photoshop users needed fast RAID volumes which it could be argued are preferably put in an external SAS enclosure. For many pros, it's now possible to keep everything in the tower except for external backups. Video editing is of course another matter, but even there I know of a few amateurs using the Mac Pro's internal bays until they absolutely need (can afford) a faster external solution.
I fail to see how four internal HDD bays slows Mac Pro development cycles. Apple's used the same case design for countless cycles. They add some to the cost but I'd bet it's not much. The case is 1.5" taller or so, and there's the SATA controller and ports on the logic board. That's such a common addition to chipsets it's probably only a few cents more.
On the other hand, a smaller Mac Pro does have appeal Apple could axe the SATA controller entirely, losing the ODDs and HDDs. The SSDs go on the PCIe bus, and the Mac Pro gets a whole lot smaller. Limit it to a single socket Xeon with up to 10-12 cores. Thunderbolt can handle any PCIe cards, and it's tricky to include TB with a standard PCIe video card, so just go with integrated graphics. Intel's integrated graphics are far better than they used to be, so no problem there. May as well stick with triple channel memory since it's faster, so it only needs three DIMM slots. With only three DIMMs, it would be cheaper and thinner to just solder them to the logic board, offer either 12 GB or 24 GB and that's more than enough RAM for 95% of Mac Pro users. Now this Mac Pro is small enough to give it a reach around, so ditch the front ports which only add unecessary cost anyways. The PSU can go in a brick, no need for it to be internal taking up all that space. With so little to upgrade inside, there's no reason for user access, so just seal it up in aluminum. Now we're cooking with gas, this is one thin Mac Pro! The only problem is the giant Xeon heatsink. We can do without it, an Ivy Bridge quad-core i7 is fast enough for 90% of Mac users. Now the power requirements are low enough to put the PSU back in it.
OMG, Apple already released the new Mac Pro! Here it is!
The big problem with building a Mac Pro big enough to support a disk array of rotating media in my mind is that the vast majority of Mac Pro users don't need a built in solution. As such it makes the Mac Pro more expensive than it needs to be and slows development cycles.
With SSDs there's not much use of internal RAID, if that's what you mean. There is still a need for big, cheap storage space, and internal HDDs fit the bill. If anything there is more need for internal HDDs now than before SSDs. Previously, even photoshop users needed fast RAID volumes which it could be argued are preferably put in an external SAS enclosure. For many pros, it's now possible to keep everything in the tower except for external backups. Video editing is of course another matter, but even there I know of a few amateurs using the Mac Pro's internal bays until they absolutely need (can afford) a faster external solution.
Nope, not at all what I mean. Many Mac Pro users only see the machine in the configurations they use thus this matra that the Mac Pro must support internal disk arrays. My point is that, that is the desire of a small minority iPod Mac Peo users. Most Mac Pro users benefit from fast secondary storage, there is no doubt there. However the need for vast quantities of such storage doesn't sit with every Mac Pro user.
I fail to see how four internal HDD bays slows Mac Pro development cycles. Apple's used the same case design for countless cycles. They add some to the cost but I'd bet it's not much. The case is 1.5" taller or so, and there's the SATA controller and ports on the logic board. That's such a common addition to chipsets it's probably only a few cents more.
It slows development time because they need to validate that all of those SATA ports and the mechanical design of the board. Think about how many true new motherboards we have seen for the Mac Pro lately. The nature of the Mac Pro case stagnate design.
On the other hand, a smaller Mac Pro does have appeal :) Apple could axe the SATA controller entirely, losing the ODDs and HDDs. The SSDs go on the PCIe bus, and the Mac Pro gets a whole lot smaller. Limit it to a single socket Xeon with up to 10-12 cores.
Sounds nice doesn't it! As long as it has a free PCI Express slot or two it would make for a very nice midrange machine.
Thunderbolt can handle any PCIe cards, and it's tricky to include TB with a standard PCIe video card, so just go with integrated graphics.
Actually TB is not in any way a replacement for internal slots.
Intel's integrated graphics are far better than they used to be, so no problem there.
Intel isn't there yet so an discrete GPU is advisable. That will no doubt change in a year or two.
May as well stick with triple channel memory since it's faster, so it only needs three DIMM slots. With only three DIMMs, it would be cheaper and thinner to just solder them to the logic board, offer either 12 GB or 24 GB and that's more than enough RAM for 95% of Mac Pro users. Now this Mac Pro is small enough to give it a reach around, so ditch the front ports which only add unecessary cost anyways. The PSU can go in a brick, no need for it to be internal taking up all that space. With so little to upgrade inside, there's no reason for user access, so just seal it up in aluminum. Now we're cooking with gas, this is one thin Mac Pro! The only problem is the giant Xeon heatsink. We can do without it, an Ivy Bridge quad-core i7 is fast enough for 90% of Mac users. Now the power requirements are low enough to put the PSU back in it.
I think that is maybe a bit much. I still want access to internal PCI Express slots.
OMG, Apple already released the new Mac Pro! Here it is!
Not really, Silly benchmarks aside it still doesn't perform anything like a Mac Pro. That being said I may buy one next year if nothing better comes from Apple.
I was thinking about something while reading this post:
I wonder how Apple will do Multiple Thunderbolt ports on the New Mac Pro for 2013.
I mean check these specs out and then I'll finish.
USB 3.0
SATA III, 6.0 Gbs
128 GB RAM Max
1600 MHz RAM
2 processors 8 cores each
3.2 GHz (probably around 2.9-3.2) GHz, Turboboost to 4.0-4.3 GHz
16 cores, 32 virtual cores
PCIExpress 3.0
10/100/1000 Ethernet
Now here is the trick part how would Apple do the Video card?
Would they put a 1 GB Video card in the PCI slot with 2 DVI/1 MiniDisplay or 1 DVI/2 MiniDisplay and then ...
... have 2 additional Thunderbolt ports? or even 3 or 4 if they expect users to use adapters for Firewire 800 or even Ethernet (10GB) in the Future?
If this is the case you could theoretically have 4 monitors (I would love 3) easily hooked up to your Mac Pro using Thunderbolt ports? and the Video card? I don't actually know but does the E5 Ivy Bridge have a GPU like the Sandy Bridge and Ivy Bridge Mobile CPUs?
Oh well just wondering what other may think.
Video card plus 2 Thunderbolt? So you could hook up 5 monitors? NICE!!
Now here is the trick part how would Apple do the Video card?
Would they put a 1 GB Video card in the PCI slot with 2 DVI/1 MiniDisplay or 1 DVI/2 MiniDisplay and then ...
... have 2 additional Thunderbolt ports? or even 3 or 4 if they expect users to use adapters for Firewire 800 or even Ethernet (10GB) in the Future?
If this is the case you could theoretically have 4 monitors (I would love 3) easily hooked up to your Mac Pro using Thunderbolt ports? and the Video card? I don't actually know but does the E5 Ivy Bridge have a GPU like the Sandy Bridge and Ivy Bridge Mobile CPUs?
This is somewhat of a nitpick, but most desktop cards are much higher on video memory at this point. The reason the mac pro has 1GB versions is because they're using cards that came out around the very beginning of 2010. They were both announced in late 2009. It should be 2-3GB today. With Apple 2 is likely. DVI is also an extremely unlikely choice. New displays are somewhat split between hdmi and displayport. DVI isn't even common on Windows at this point, as it doesn't need to be. Things like VGA held on longer due to its use in projectors. E5s aren't scheduled to gain integrated graphics with Ivy Bridge. I didn't look at what changes with Haswell, and I have no idea if they could squeeze more on there without really limiting the clock rates of higher core count chips. It would be more likely in the single package models. The E3 Xeons already have integrated graphics. They're single cpu package only.
Quote:
Originally Posted by wizard69
It slows development time because they need to validate that all of those SATA ports and the mechanical design of the board. Think about how many true new motherboards we have seen for the Mac Pro lately. The nature of the Mac Pro case stagnate design.
I believe the stick ssds are still using sata, even with the weird plug. I'd have to check if any PCI card type ssds are usable under OSX.
Now here is the trick part how would Apple do the Video card?
Would they put a 1 GB Video card in the PCI slot with 2 DVI/1 MiniDisplay or 1 DVI/2 MiniDisplay and then ...
... have 2 additional Thunderbolt ports? or even 3 or 4 if they expect users to use adapters for Firewire 800 or even Ethernet (10GB) in the Future?
If this is the case you could theoretically have 4 monitors (I would love 3) easily hooked up to your Mac Pro using Thunderbolt ports? and the Video card? I don't actually know but does the E5 Ivy Bridge have a GPU like the Sandy Bridge and Ivy Bridge Mobile CPUs?
This is somewhat of a nitpick, but most desktop cards are much higher on video memory at this point. The reason the mac pro has 1GB versions is because they're using cards that came out around the very beginning of 2010. They were both announced in late 2009. It should be 2-3GB today. With Apple 2 is likely. DVI is also an extremely unlikely choice. New displays are somewhat split between hdmi and displayport. DVI isn't even common on Windows at this point, as it doesn't need to be. Things like VGA held on longer due to its use in projectors. E5s aren't scheduled to gain integrated graphics with Ivy Bridge. I didn't look at what changes with Haswell, and I have no idea if they could squeeze more on there without really limiting the clock rates of higher core count chips. It would be more likely in the single package models. The E3 Xeons already have integrated graphics. They're single cpu package only.
Quote:
Originally Posted by wizard69
It slows development time because they need to validate that all of those SATA ports and the mechanical design of the board. Think about how many true new motherboards we have seen for the Mac Pro lately. The nature of the Mac Pro case stagnate design.
I believe the stick ssds are still using sata, even with the weird plug. I'd have to check if any PCI card type ssds are usable under OSX.
Yes they are still using SATA which will become a problem in the very near future. AS to PCI cards usable under OS/X, what difference does that make. Apple would have to develop the hardware standards and the software to do this.
Not really, Silly benchmarks aside it still doesn't perform anything like a Mac Pro. That being said I may buy one next year if nothing better comes from Apple.
I was attempting snark. However I do like the idea of a Mac Pro with several blade SSDs for storage, with all the HDDs in a separate (optional) enclosure connected via Apple's proprietary version of a sick fast SAS connection. Of course I'd prefer a standard SAS connection that enabled Mac Pros to be used with any SAS enclosure, but this is Apple we're talking about. They could even make the HDD enclosure stackable with the Mac Pro. This would enable design posibilities for the Mac Pro, and a even a couple versions of the HDD enclosure such as a 4-6 HDD version and an 8-12 HDD version.
Note that Thunderbolt enclosures would not cut it for this idea, the seperate enclosure would require speeds equivalent to having the HDDs connected to internal SATA III ports on the logic board.
By separating the HDD into a separate enclosure, the new Mac Pro wouldn't need another case redesign when SSDs entirely supersede HDDs. Just make the HDD enclosure optional for a few design cycles and discontinue it when SSDs drop enough in price. Many users wouldn't even bother with the HDD enclosure, a Mac Pro with a couple SSDs would be fine and they could use a USB 3.0 or Thunderbolt enclosure for backups and/or data.
Quote:
Originally Posted by hmm
I believe the stick ssds are still using sata, even with the weird plug. I'd have to check if any PCI card type ssds are usable under OSX.
The Solos are nice because they can be used with a variety of SSDs. With the Accelsior, you're stuck with SandForce, although OWC's firmware is reputed to be more stable than others (not my experience).
If Apple wants to sell more towers, I'd advise them to offer a non-Xeon tower with internals comparable to the iMac in performance. Use desktop components: they're cheaper and faster. And for the love of God, lose the thin fetish! What does thin get you with an iMac? Overheating, less expandability, less capability. You can't even see how thick it is when you're using it, unless it's to bend over and around to reach the card reader that is inexplicably on the back of the iMac. Design a tower in which form follows function.
You have me drooling here Junkyard. Offer iMac performance in a tower and I will gladly pay $1500 for it. Being able to replace components easily, by myself when they go bad, my choice of monitor with enough internal space for 2 hard drives and an optical drive.
$1500 is a bargain for not being stuck with the uselessness of thin in a desktop computer.
I hate to disagree because of friction. But I have 2 Mac Pro 1,1 and I don't want to wait for the Mac Pro (2013) but I have to. I see my machines as they are setup and they are so keen. They each have 4 TBs of storage. One is an LDAP 2TB RAID 10 server for the family, along with other accounts for managing iTunes etc. The other is my development design studio, setup in a 1 TB RAID 1, and 1 TB for Windows 7 for gaming, and 1 TB for Scratch for Video / Photo / Audio projects. The machines run great still (16 GB of RAM a must) but they have jobs as media servers for video and audio to the whole house, when not in use for working.
I think the whole argument that storage via classic platter media is dead, is dead wrong. You may have a minor point with the fact that some people don't need media servers cause they can watch Netflix, Hulu, Amazon and etc for the Television content, but all on demand kinda sucks. Let's say you download something that is 1 GB, and you are on a laptop browsing, watching Netflix on your TV, browsing the internet slows down and lag sucks WhEnEvEr it hits.
I see the Mac Pro easily with options for, if you want a Server for your House, for Family, with photos and video and music, and then kids doing fun projects in the future with iPad and iPhone camera, hey who knows, but you easily need 4-8 TB of storage, easily. And now FINALLY, with Thunderbolt, you can back up restore that much data, to an external enclosure 4-8 TB.
But let say you have a company and a solo operation, where you want to edit 1 video with 5-10 video sources, you definitely want at least (for 720p/1080p 24/60 content) 4x512 SSDs RAID 0 or 5 on 6 Gb/s, and then when you are done you would back up your project to an external Thunderbolt device, clean up and go on again from there.
But in both cases, both setups are perfectly viable and reasonable. I just think about setting up my younger brother's and my brother in law's families with sweet home servers for all of there media needs. One is getting into Photography, but loves Windows, the other likes gaming, and taking photos like 300 GB of kids photos, and then videos for the kids iPads. But they both need servers with RAID 10 storage or at least RAID 1 with external thunderbolt external backup for in case of fire or cold storage.
In addition they can do all of their editing of videos and photos and have it all backed up. And play video games if they want to, while having it served up properly.
Seriously, these are system's that families need in general that last for up to the next 8-10 years. Yes you have to spend $3000, to $5000 to get this setup but in the long run, it's so much worth it to have your family taken care of.
oh and btw off topic, Cluster servers, will probably be done in the future by Apple at an ExTrEmElY more reduced price via Thunderbolt adapters of some sort if not just plain ethernet with MacMini, I know they already do, but this is getting really powerful and cheap. Think about if you could buy a MacMini for $599 every month for 8 months. For $4,800 you could have a MacMini cluster that you added on to month by month, at the end with 8x4x2 64 cores. (This of course is when the quads come down), pretty interesting. Cause honestly, blade servers and xserves are a joke if you can stack 32 mac mini's in the same space a Mac Pro (maybe more eek) takes, what's that 32x8 = 256 cores,
Since the OP kicked this off with an ARM-based Mac Pro, I figured this wasn't a reality-based thread.
You can ask one Olympian to lift 500 pounds or you can ask 10 normal people to help lift 50 pounds. Which is reality based? Both are. Is one solution better than the other? It depends. For many applications, many hands make light work. An ARM processor may be like the normal people and the latest x86 like the Olympic weight-lifter, but do I really care what is inside the machine if they can both accomplish the same work? No I don't.
One Olympic weight lifter cannot scale above its roughly 500 pound limit. But average people are cheap to come by. 100 average people can lift way more than one Olympic athlete. If each average person can lift 50 pounds, than 100 people can lift 5,000 pounds - way more than the weight lifter. Sometimes quantity has a quality all its own.
There is no question audio, video, and photo work are massively parallel problems. Are these not the main source of taxing the CPUs? Then you would be better served by a parallel solution not a serial solution. Therefore ARM chips are not a stupid solution.
Of course, one could come back and say what about having 10 Olympic weight lifters, not just one? This too is a valid solution, and in fact, the one we currently have. Which is better? I really do not care which solution I have if the work outcome is the same. Therefore, it all comes down to cost. How much does cost to hire 10 Olympic weight lifters verses how much does it cost to hire 100 average people? Realistically, how much do x86 chips cost verses a boatload of equivalent ARM chips? If they are the same cost than I really do not see why Apple should do anything different. But if one solution is much cheaper, then that is what Apple should do.
Actually, cost is not the only factor. Heat is another big deal. I am not a CPU guru so I do not know the answers to these questions. Maybe somebody else here does. I am sure Apple has this data. The fact that Apple seems to be dragging their feet on the Pro side indicates to me that a significant shift is coming soon.
This theory fits the data better than proposing Apple just drop all its pro level machines and software. What is pro level today becomes the consumer machine of tomorrow. That is why Apple needs to stay in the Pro market and why they will. Therefore I see the Xeon Phi or a massively parallel ARM new MacPro being the way forward in the near future.
Do realize that very shortly we will have chips with 12 cores and accelerators with +50 intel compatible cores. Intel is also working on other members of the Phi family with built in super computing networking and other features. Frankly we have all sorts of new technologies on the hip origin that could go into the New Mac Pro, some of these technologies could dramatically alter how the Mac Pro is seen as a pro computer. It might not be the image you have burnt into our mind but that would not make it less of a pro machine. The whole point I'm trying to make in this thread is that Apple can't build a machine for past user needs and expect it to sell or for that matter to even get people to note it. They need to build a machine that is built for the future.
Personally, I think Apple will keep on going with the slim thing.
You have 'capable' boxes. iPads. iPhones. Very powerful for their size. You have the new iMac (the most powerful iMac yet...in an even thinner enclosure...) You have the i7 Mac Mini which, if it had the 680 MX in it...would fly off the shelves, no doubt in my mind about that. (Blame Apple's artificial 'upsell' model for the lack of discrete gpu...which says it all.)
...then you have the Mac Pro. It's a big box. An out of date big box. People no longer need computers the size of a big room to do certain tasks. That's why smaller beasts do the jobs that can be done easier with lighter 'mammal' computers. eg the iPad can do 9/10 things most casual users do with their mamputer. :P
Now. The pro is wayyyyyy out of date. Over due for a facelift and a comprehensive redesign. The Wizard has been calling for a smaller but 'upgradable' beast. Does a Pro still need to be that big to include a couple of drives and dual processors and a decent gpu?
The current Mac Mini can 'context' an entry Mac Pro depending on which 'silly' benchmark you use. That's Apple's fault for not upgrading the Pro? Or is it because desktops vs laptop cpu development is flattening out. The exponential development of laptop and in particular the iPad's power is nothing short of jaw dropping. The recent 'iToys' humble the last generation of PPC Mac kit. It seems the mammals are chasing down the dinosaurs.
So what does a Mac Pro have to include? The same case? Room for 2? 3? 4? Hard drives? 2 internals? And an external port to hook up to an array? Gpus? 1 top of range? Or room for SLI? Given Apple's (and even Intel's...) recent focus on performance for watt, what does a re-imagined Pro look like given that most of the 10% (let's be generous) of really demanding tasks...for those demanding users.
I'm plumping for software to catch up to hardware. I'd offer a broader range of Pro. To increase volume? A consumer quad core with 680MX with SSD. For the pro? A 8 core cube with GPU that can fit into that enclosure. A couple of SSDs HDs. Want more power? Buy another and cluster. X-Grid. About time Apple used software to redefine power and to drive sales of their 'Pro.' For the 'Prosumer' and 'Pro.' It would perhaps be just a little bigger than x2 Mini's one on top of the other. Access via a similar mechanism to the Cube or a screw top biscuit lid aka like the Mini. Maybe with some go faster air vents on their somewhere. I want it to be as stunning as the original Cube...but have the understated brilliance of the iPhone. (Give me a slate Pro...)
Want more power? Stick an array on it. Want more power? External GPUs are around the corner. Want more displays? Use your Thunderbolt ports. etc. It's not like Towers have ever been all in ones. You have bits sticking in them and out of them. Or want more power? Just buy another 'Cube' and pass a render over to the 'grid.'
It's got to the stage where the iMac has all the power 'I' need. (I couldn't have said that when the Bondi iMac was around. I didn't think I'd ever see a top of the range iMac kicking around an entry Mac Pro. Yet here we are.)
Price a prosumer 'Pro' at £1100 (by the time you add Apple's 27 inch display, you're already at £1900!) Quad core. Decent GPU. No display. Entry 'Pro' six core at £1395 SSD/Fusion as standard. (Yes Apple...nickle and dimers that you are...) 8 core at £1795. With top end gpu included. Dual Model? Starting at 2k upward.
Comments
As for customers that use to purchase the machine, they can either get with the program or break out their VIC 20's. I mean really guys what is with people that want to regress to the dark ages? I've been using various computers since that VIC 20 mentioned above, the last thing I want to do is to see technology stagnate because a couple of fools can't open their minds to new ways of doing things.
Frankly we have not even gotten a clue as to what Apples new pro will look like. When it does come it might be a good idea to concentrate on what is good about the new machine instead of what is missing. Frankly this point of view is as silly as the people complaining about the optical missing in the new iMac. You missed the point, Apple needs a Pro type machine that can grab some decent sales. A simplified machine means that Apple can offer a platform at a much low cost than today's machine.
On the flip side a complex machine is far harder to update. As such it is a factor in update cycles. The "excuse" is the lack of sales. The lack of sales is the only reason why I expect to see a heavily refactored Mac Pro. Lets face it there wasa huge demand for the Mac Pro Apple would be paying more attention to the platform. So really I don't think Apple has a choice, either they make a Mac Pro that appeals to a wider audience or they scrap the machine. There is virtually no demand for the machine you are asking for. Apples sales? Seriously guy take a good long look at Apples sales, 80% of Apples Mac sales are laptops, the iMac takes up the vast majority of the desktop sales and then the Mini follows that. So realistically what do you think Mac Pro sales are?
Lets say it is 3% of Mac sales, of that 3% how many of those users do you think make use of internal disk arrays? I'm going to take a shot here and say it is les than 10% of that 3%. Internal disk arrays are a thing of the past, the demand isn't there to justify passing the expense on to every Mac user. No rational business person is going to commit to a product they haven't seen. However good business man making their tools work for them. Anybody that rushes off in a fit because Apple didn't do it their way isn't the type of customer Apple needs. Frankly I'm tired of hearing from people that want to drag Apple back into the last century technology wise. At the very least make a thoughtful and rational evaluation of where Apple takes the Mac Pro. Bingo. As for hackintoshes, lots of luck structuring a business around that sort of crap. All that wasted effort would be better spent on simply configuring the new Mac Pro to work for you.
Do realize that very shortly we will have chips with 12 cores and accelerators with +50 intel compatible cores. Intel is also working on other members of the Phi family with built in super computing networking and other features. Frankly we have all sorts of new technologies on the hip origin that could go into the New Mac Pro, some of these technologies could dramatically alter how the Mac Pro is seen as a pro computer. It might not be the image you have burnt into our mind but that would not make it less of a pro machine. The whole point I'm trying to make in this thread is that Apple can't build a machine for past user needs and expect it to sell or for that matter to even get people to note it. They need to build a machine that is built for the future.
Quote:
Originally Posted by wizard69
No rational business person is going to commit to a product they haven't seen. However good business man making their tools work for them. Anybody that rushes off in a fit because Apple didn't do it their way isn't the type of customer Apple needs. Frankly I'm tired of hearing from people that want to drag Apple back into the last century technology wise. At the very least make a thoughtful and rational evaluation of where Apple takes the Mac Pro.
Bingo. As for hackintoshes, lots of luck structuring a business around that sort of crap. All that wasted effort would be better spent on simply configuring the new Mac Pro to work for you.
I don't agree with everything in that post, but I'm not feeling like breaking it down regarding storage and things. 64 bit computing did alleviate the use of scratch disks in a lot of cases. This was a big use of internal bays. I still think they're useful for storage, but 64GB of ram really kills the need for scratch data in most use cases. When these applications could only access 3GB directly, they had to dump it somewhere, so an internal RAID 0 was often the best solution.
Regarding hackintoshes, no matter how many people claim to own stable ones, I'd never go that route for a business. I've never suggested one to anyone I know that uses Macs within their businesses. A hacked solution is just a bad idea there. I think if the mac pro and Apple's other offerings are a bad solution, you should look at Windows rather than attempt an unsupported configuration. They're really better for hobbyists. If you look at insanely mac, many of them own macbook pros and hackintoshes. DIY computing can be cheap enough to take up as a very nerdy hobby.
Note that I'm not saying that some users don't need an array just that these days it makes more sense to put the array in an external box. An external box provides the user with certain advantages one important one being that it decouples the Mac Pro and array storage technologies.
On the flip side I realize that internal storage, that is secondary storage inside the Mac Pro is important. However looking forward this storage has to move off the SATA bus to maximize transfer rates. So Apple would be far better off basically saying the same thing about magnetic drives as the have about optical, the technology has run its course in the case of a high performance work station! Needs do vary but there is nothing to stop Apple from introducing two or more storage dedicated PCI-Express slots. I agree 100%! There is just too much that can go wrong for most businesses. Even if the guy doing the Hackintosh is competent it is still a significant drain on business resources. Well I'm not sure about Windows????.
I like to look at it this way, if you run a little tool and die business you may have a couple of CNC machines that you pay for over a long number of years. Over those years the machines are bolted and grouted to the floor as such you can't change them out every time a new project comes in. Instead you learn to make the machine do what you need done.
Buying a computer and using a computer isn't much different. It isn't normally bolted to the floor but you can't expect to change it out with every new project that comes along. In effect you need to make money off that hardware until it makes business sense to upgrade. So you learn to make the platform you got do the job at hand. In the end I expect next years Mac Pro to be a faster machine that may require adaptation on established user configuration. This isn't a bad thing at all but it does make some people feel uncomfortable and in some case down right angry. In some cases you will never get an individual to see the light, Apple just has to hope that the new machine has enough allure that it causes people to reconsider their preconception of what a pro computer is.
Quote:
Originally Posted by wizard69
The big problem with building a Mac Pro big enough to support a disk array of rotating media in my mind is that the vast majority of Mac Pro users don't need a built in solution. As such it makes the Mac Pro more expensive than it needs to be and slows development cycles.
Note that I'm not saying that some users don't need an array just that these days it makes more sense to put the array in an external box. An external box provides the user with certain advantages one important one being that it decouples the Mac Pro and array storage technologies.
On the flip side I realize that internal storage, that is secondary storage inside the Mac Pro is important. However looking forward this storage has to move off the SATA bus to maximize transfer rates. So Apple would be far better off basically saying the same thing about magnetic drives as the have about optical, the technology has run its course in the case of a high performance work station! Needs do vary but there is nothing to stop Apple from introducing two or more storage dedicated PCI-Express slots.
Well things consume PCI lanes even if they're run as SATA. I think the stick type ssds Apple uses still show up as SATA. They may have their own controllers.
Quote:
I agree 100%! There is just too much that can go wrong for most businesses. Even if the guy doing the Hackintosh is competent it is still a significant drain on business resources.
I'm not a fan of hacked solutions for business. Windows actually has certain advantages in software. Their gpu drivers also have some features that Apple is unlikely to support in the next couple years, such as full support for displayport 1.2.
Quote:
Well I'm not sure about Windows????.
I like to look at it this way, if you run a little tool and die business you may have a couple of CNC machines that you pay for over a long number of years. Over those years the machines are bolted and grouted to the floor as such you can't change them out every time a new project comes in. Instead you learn to make the machine do what you need done.
Buying a computer and using a computer isn't much different. It isn't normally bolted to the floor but you can't expect to change it out with every new project that comes along. In effect you need to make money off that hardware until it makes business sense to upgrade. So you learn to make the platform you got do the job at hand. In the end I expect next years Mac Pro to be a faster machine that may require adaptation on established user configuration. This isn't a bad thing at all but it does make some people feel uncomfortable and in some case down right angry. In some cases you will never get an individual to see the light, Apple just has to hope that the new machine has enough allure that it causes people to reconsider their preconception of what a pro computer is.
Some workflows are much more portable than others. Going back a few years, things were much more segregated. Certain things either wouldn't run on Macs, or they got terrible ports. It had certain niche markets that were supported really well, so you could have certain Mac specific plugins. It was also a lot of work dealing with the different file systems in a mixed environment. Today this is less of an issue. My point was merely that if they're not going to buy a Mac, they shouldn't try to prolong the use of OSX with a hacked solution. It is typically a long term migration if you switch your OS. Anyway what I've suggested for the mac pro is that they need some kind of growth leverage outside of Macs. The workstation market isn't really growing, so they can't rely on inertia. The health of the line would likely rely on making something attractive enough to leverage more users away from Windows boxes or occasionally Linux (popular in larger film studios).
I see no need at all to EOL the Mac Pro.
My current desktop is a 3.46 GHz Hexa-core Mac Pro. It has two SSDs and four HDDs (two 2 TB, two 3 TB). Both SSDs are on a 6G SATA PCIe card, and I've added eSATA for external backups And yes, it has a Blu-ray drive, which I use all the time to "consume" movies. BD offers superior quality over streaming compressed 1080p content, just like CDs offer higher quality than iTunes downloads.
For my needs this is a dream system, and no it is NOT overkill. I use it for lightroom, photoshop, and HDR processing, along with the usual email, word processing, and internet. When I added all those HDDs to it, I could buy bare HDDs, no need for expensive TB enclosures or a rat's nest of cables. Upgrading to 6G SATA was simple, but honestly the speed difference is negligible for my needs. This computer has more power than I need at the moment, but I plan to use it for another 5-7 years, and it may be good to go for another ten.
Could I do the same with an iMac? Yes, but it would be one ugly, space consuming setup. I'd have either a bunch drive enclosures, or one TB 4 bay enclosure that cost a crazy amount of money. Also, an external BD drive. The Mac Pro takes up far less space, and it is much cheaper to upgrade than an iMac. If a drive goes bad, I just swap it out. On an iMac, if a drive goes bad I have to lug it into an Apple servicer for replacement. Best of all, I can choose what monitors I want to use.
If Apple wants to sell more towers, I'd advise them to offer a non-Xeon tower with internals comparable to the iMac in performance. Use desktop components: they're cheaper and faster. And for the love of God, lose the thin fetish! What does thin get you with an iMac? Overheating, less expandability, less capability. You can't even see how thick it is when you're using it, unless it's to bend over and around to reach the card reader that is inexplicably on the back of the iMac. Design a tower in which form follows function.
If Apple wants to ditch their most loyal customer base - the content creation industry, that is their right, but it would be like Chevy EOLing the Corvette. Sure, the Corvette doesn't make Chevy much return on the R&D investment. They may even come close to breaking even. But Chevy knows that the Corvette is part of what defines their brand, and it showcases their engineering chops. Same with the Mac Pro. I've heard countless times from people who comment that their company uses Macs for all video and graphic arts, and many of these people buy a Mac precisely for this reason. The Mac Pro, or more accurately, Apple's presence in the creative arts community, helps define Apple as a serious player. Without content creation, Apple is a gadget maker who dabbles in desktop computing - not a computer someone buys for serious work.
Of course Apple can discontinue all their desktops and laptops that aren't as profitable as their iBaubles, and they will remain profitable. But one day, their dominance in the gadget markets will ebb, and then what? Apple will just be another phone maker among dozens. Nothing special, and certainly not a company serious users turn to for computing solutions.
Quote:
Originally Posted by Junkyard Dawg
For my needs this is a dream system, and no it is NOT overkill. I use it for lightroom, photoshop, and HDR processing, along with the usual email, word processing, and internet. When I added all those HDDs to it, I could buy bare HDDs, no need for expensive TB enclosures or a rat's nest of cables. Upgrading to 6G SATA was simple, but honestly the speed difference is negligible for my needs. This computer has more power than I need at the moment, but I plan to use it for another 5-7 years, and it may be good to go for another ten.
That was my point on growth. The system would receive more attention if it had the power to encourage further rounds of upgrades and leverage new customers from Windows equivalents. You stick with what you have because it works. A new one wouldn't significantly improve the way you're able to work. If LR raw processing was in real time or the machine could interpret raws in real time while retaining more information and applying adjustments at a lower level, it would influence many people like you to invest in new hardware, as it would change the way you work. If you compare Lightroom today to the early Sinar digital back (which had to be tethered via firewire) software, you'd see what I mean. People don't just upgrade based on perceived speed. There are cyclical upgrades, but the time when you see many people upgrade simultaneously is when it changes the way they can work. Otherwise you're already looking at sunken costs and a workflow that works.
That's certainly true about encouraging new rounds of upgrades, and it could explain why Apple hasn't upgraded the Mac Pro recently. The new Xeon processors don't offer the sorts of performance increases you describe, and at least in my experience, SATA III isn't enough of an improvement over SATA II to make me want a whole new system. To hear people whining about switching to windows over the lack of SATA III and USB 3.0 seems silly IMO.
There is something new that will be shipping in volume next year that could enable the sort of performance leap you describe, hmmm. Could be that the new Mac Pro is designed around it. It would explain the delay, and Apple's confidence in waiting. It could differentiate Mac Pros from other Xeon workstations if cleverly implemented. And it would require Apple and Intel to work closely together to bring a product to fruition, something Intel has been doing with other companies on this new innovation. Knight's Corner. :P
In any event, even without me buying a new Mac Pro, Apple still gets plenty of business out of me. I buy every OS X release, and my share of iBaubles. Perhaps even more importantly, I prosyletize OS X and help Windows users switch to Macs. They see my Mac Pro, see what I do with it, are amazed that Apple does more than make phones and ipods. Then they go out and buy iMacs. Did my Mac Pro directly lead to the iMac sales? Not really, but I can't say it did not either. It's part of the Apple ecosystem, and people respond to a broad computing ecosystem. Call it the bandwagon effect. The 17" MacBook Pro has the same effect - "whoa, that's a sweet laptop! I don't need anything that big, but I want a MacBook now!"
Quote:
Originally Posted by Junkyard Dawg
That's certainly true about encouraging new rounds of upgrades, and it could explain why Apple hasn't upgraded the Mac Pro recently. The new Xeon processors don't offer the sorts of performance increases you describe, and at least in my experience, SATA III isn't enough of an improvement over SATA II to make me want a whole new system. To hear people whining about switching to windows over the lack of SATA III and USB 3.0 seems silly IMO.
You really need to do your reading. The available chipsets will still be the same damn thing, and you're talking about 225-300W specialized cards. The $400 ones seem to be testing samples. I don't expect them to see volume sales at that. While it would be awesome to see if that thing can run OpenCL, this doesn't seem like something Apple would do. The current case also lacks enough cooling to dissipate that off a single card, and that is the lower power version. The mac pro actually has enough problems with cards approaching 200W, so you'd definitely require redesigned airflow. Even then I don't see it. They ignored Teslas. They never made CUDA any kind of priority even though early development exceeded OpenCL, and NVidia cards Fermi and on support OpenCL. What would cause the change in behavior now?
http://www.theregister.co.uk/2012/11/12/intel_xeon_phi_coprocessor_launch/
Quote:
If you want to weave Xeon Phis into your supercomputers for number-crunching offload, then you probably will want the passively cooled Xeon Phi 5110P PCIe card. This one has more cores fired up and a slightly slower clock speed, and can deliver its 1.01 teraflops within a 225-watt power envelope – the same thermal limit that other GPU coprocessors for servers need to stay within. The 5110P card has the Xeon Phi chip with 60 cores, 30MB of cache memory on the die, plus 8GB of GDDR5 memory and a peak of 320GB/sec of memory bandwidth.
Obviously it would have to be a custom solution, not simple support for a PCIe card. That's why I mentioned that Apple and Intel would have to work together on it.
Since the OP kicked this off with an ARM-based Mac Pro, I figured this wasn't a reality-based thread.
Quote:
The big problem with building a Mac Pro big enough to support a disk array of rotating media in my mind is that the vast majority of Mac Pro users don't need a built in solution. As such it makes the Mac Pro more expensive than it needs to be and slows development cycles.
With SSDs there's not much use of internal RAID, if that's what you mean. There is still a need for big, cheap storage space, and internal HDDs fit the bill. If anything there is more need for internal HDDs now than before SSDs. Previously, even photoshop users needed fast RAID volumes which it could be argued are preferably put in an external SAS enclosure. For many pros, it's now possible to keep everything in the tower except for external backups. Video editing is of course another matter, but even there I know of a few amateurs using the Mac Pro's internal bays until they absolutely need (can afford) a faster external solution.
I fail to see how four internal HDD bays slows Mac Pro development cycles. Apple's used the same case design for countless cycles. They add some to the cost but I'd bet it's not much. The case is 1.5" taller or so, and there's the SATA controller and ports on the logic board. That's such a common addition to chipsets it's probably only a few cents more.
On the other hand, a smaller Mac Pro does have appeal
OMG, Apple already released the new Mac Pro! Here it is!
I was thinking about something while reading this post:
I wonder how Apple will do Multiple Thunderbolt ports on the New Mac Pro for 2013.
I mean check these specs out and then I'll finish.
USB 3.0
SATA III, 6.0 Gbs
128 GB RAM Max
1600 MHz RAM
2 processors 8 cores each
3.2 GHz (probably around 2.9-3.2) GHz, Turboboost to 4.0-4.3 GHz
16 cores, 32 virtual cores
PCIExpress 3.0
10/100/1000 Ethernet
Now here is the trick part how would Apple do the Video card?
Would they put a 1 GB Video card in the PCI slot with 2 DVI/1 MiniDisplay or 1 DVI/2 MiniDisplay and then ...
... have 2 additional Thunderbolt ports? or even 3 or 4 if they expect users to use adapters for Firewire 800 or even Ethernet (10GB) in the Future?
If this is the case you could theoretically have 4 monitors (I would love 3) easily hooked up to your Mac Pro using Thunderbolt ports? and the Video card? I don't actually know but does the E5 Ivy Bridge have a GPU like the Sandy Bridge and Ivy Bridge Mobile CPUs?
Oh well just wondering what other may think.
Video card plus 2 Thunderbolt? So you could hook up 5 monitors? NICE!!
Laters...
Quote:
Originally Posted by rezwits
Now here is the trick part how would Apple do the Video card?
Would they put a 1 GB Video card in the PCI slot with 2 DVI/1 MiniDisplay or 1 DVI/2 MiniDisplay and then ...
... have 2 additional Thunderbolt ports? or even 3 or 4 if they expect users to use adapters for Firewire 800 or even Ethernet (10GB) in the Future?
If this is the case you could theoretically have 4 monitors (I would love 3) easily hooked up to your Mac Pro using Thunderbolt ports? and the Video card? I don't actually know but does the E5 Ivy Bridge have a GPU like the Sandy Bridge and Ivy Bridge Mobile CPUs?
This is somewhat of a nitpick, but most desktop cards are much higher on video memory at this point. The reason the mac pro has 1GB versions is because they're using cards that came out around the very beginning of 2010. They were both announced in late 2009. It should be 2-3GB today. With Apple 2 is likely. DVI is also an extremely unlikely choice. New displays are somewhat split between hdmi and displayport. DVI isn't even common on Windows at this point, as it doesn't need to be. Things like VGA held on longer due to its use in projectors. E5s aren't scheduled to gain integrated graphics with Ivy Bridge. I didn't look at what changes with Haswell, and I have no idea if they could squeeze more on there without really limiting the clock rates of higher core count chips. It would be more likely in the single package models. The E3 Xeons already have integrated graphics. They're single cpu package only.
Quote:
Originally Posted by wizard69
It slows development time because they need to validate that all of those SATA ports and the mechanical design of the board. Think about how many true new motherboards we have seen for the Mac Pro lately. The nature of the Mac Pro case stagnate design.
I believe the stick ssds are still using sata, even with the weird plug. I'd have to check if any PCI card type ssds are usable under OSX.
Quote:
Originally Posted by hmm
Quote:
Originally Posted by rezwits
Now here is the trick part how would Apple do the Video card?
Would they put a 1 GB Video card in the PCI slot with 2 DVI/1 MiniDisplay or 1 DVI/2 MiniDisplay and then ...
... have 2 additional Thunderbolt ports? or even 3 or 4 if they expect users to use adapters for Firewire 800 or even Ethernet (10GB) in the Future?
If this is the case you could theoretically have 4 monitors (I would love 3) easily hooked up to your Mac Pro using Thunderbolt ports? and the Video card? I don't actually know but does the E5 Ivy Bridge have a GPU like the Sandy Bridge and Ivy Bridge Mobile CPUs?
This is somewhat of a nitpick, but most desktop cards are much higher on video memory at this point. The reason the mac pro has 1GB versions is because they're using cards that came out around the very beginning of 2010. They were both announced in late 2009. It should be 2-3GB today. With Apple 2 is likely. DVI is also an extremely unlikely choice. New displays are somewhat split between hdmi and displayport. DVI isn't even common on Windows at this point, as it doesn't need to be. Things like VGA held on longer due to its use in projectors. E5s aren't scheduled to gain integrated graphics with Ivy Bridge. I didn't look at what changes with Haswell, and I have no idea if they could squeeze more on there without really limiting the clock rates of higher core count chips. It would be more likely in the single package models. The E3 Xeons already have integrated graphics. They're single cpu package only.
Quote:
Originally Posted by wizard69
It slows development time because they need to validate that all of those SATA ports and the mechanical design of the board. Think about how many true new motherboards we have seen for the Mac Pro lately. The nature of the Mac Pro case stagnate design.
I believe the stick ssds are still using sata, even with the weird plug. I'd have to check if any PCI card type ssds are usable under OSX.
Yes they are still using SATA which will become a problem in the very near future. AS to PCI cards usable under OS/X, what difference does that make. Apple would have to develop the hardware standards and the software to do this.
Quote:
Not really, Silly benchmarks aside it still doesn't perform anything like a Mac Pro. That being said I may buy one next year if nothing better comes from Apple.
I was attempting snark. However I do like the idea of a Mac Pro with several blade SSDs for storage, with all the HDDs in a separate (optional) enclosure connected via Apple's proprietary version of a sick fast SAS connection. Of course I'd prefer a standard SAS connection that enabled Mac Pros to be used with any SAS enclosure, but this is Apple we're talking about. They could even make the HDD enclosure stackable with the Mac Pro. This would enable design posibilities for the Mac Pro, and a even a couple versions of the HDD enclosure such as a 4-6 HDD version and an 8-12 HDD version.
Note that Thunderbolt enclosures would not cut it for this idea, the seperate enclosure would require speeds equivalent to having the HDDs connected to internal SATA III ports on the logic board.
By separating the HDD into a separate enclosure, the new Mac Pro wouldn't need another case redesign when SSDs entirely supersede HDDs. Just make the HDD enclosure optional for a few design cycles and discontinue it when SSDs drop enough in price. Many users wouldn't even bother with the HDD enclosure, a Mac Pro with a couple SSDs would be fine and they could use a USB 3.0 or Thunderbolt enclosure for backups and/or data.
Quote:
Originally Posted by hmm
I believe the stick ssds are still using sata, even with the weird plug. I'd have to check if any PCI card type ssds are usable under OSX.
OWC Accelsior
Velocity Solo x2
Anglebird Wings (moving to 6G SATA next year)
The Solos are nice because they can be used with a variety of SSDs. With the Accelsior, you're stuck with SandForce, although OWC's firmware is reputed to be more stable than others (not my experience).
Quote:
Originally Posted by Junkyard Dawg
I see no need at all to EOL the Mac Pro.
If Apple wants to sell more towers, I'd advise them to offer a non-Xeon tower with internals comparable to the iMac in performance. Use desktop components: they're cheaper and faster. And for the love of God, lose the thin fetish! What does thin get you with an iMac? Overheating, less expandability, less capability. You can't even see how thick it is when you're using it, unless it's to bend over and around to reach the card reader that is inexplicably on the back of the iMac. Design a tower in which form follows function.
You have me drooling here Junkyard. Offer iMac performance in a tower and I will gladly pay $1500 for it. Being able to replace components easily, by myself when they go bad, my choice of monitor with enough internal space for 2 hard drives and an optical drive.
$1500 is a bargain for not being stuck with the uselessness of thin in a desktop computer.
I hate to disagree because of friction. But I have 2 Mac Pro 1,1 and I don't want to wait for the Mac Pro (2013) but I have to. I see my machines as they are setup and they are so keen. They each have 4 TBs of storage. One is an LDAP 2TB RAID 10 server for the family, along with other accounts for managing iTunes etc. The other is my development design studio, setup in a 1 TB RAID 1, and 1 TB for Windows 7 for gaming, and 1 TB for Scratch for Video / Photo / Audio projects. The machines run great still (16 GB of RAM a must) but they have jobs as media servers for video and audio to the whole house, when not in use for working.
I think the whole argument that storage via classic platter media is dead, is dead wrong. You may have a minor point with the fact that some people don't need media servers cause they can watch Netflix, Hulu, Amazon and etc for the Television content, but all on demand kinda sucks. Let's say you download something that is 1 GB, and you are on a laptop browsing, watching Netflix on your TV, browsing the internet slows down and lag sucks WhEnEvEr it hits.
I see the Mac Pro easily with options for, if you want a Server for your House, for Family, with photos and video and music, and then kids doing fun projects in the future with iPad and iPhone camera, hey who knows, but you easily need 4-8 TB of storage, easily. And now FINALLY, with Thunderbolt, you can back up restore that much data, to an external enclosure 4-8 TB.
But let say you have a company and a solo operation, where you want to edit 1 video with 5-10 video sources, you definitely want at least (for 720p/1080p 24/60 content) 4x512 SSDs RAID 0 or 5 on 6 Gb/s, and then when you are done you would back up your project to an external Thunderbolt device, clean up and go on again from there.
But in both cases, both setups are perfectly viable and reasonable. I just think about setting up my younger brother's and my brother in law's families with sweet home servers for all of there media needs. One is getting into Photography, but loves Windows, the other likes gaming, and taking photos like 300 GB of kids photos, and then videos for the kids iPads. But they both need servers with RAID 10 storage or at least RAID 1 with external thunderbolt external backup for in case of fire or cold storage.
In addition they can do all of their editing of videos and photos and have it all backed up. And play video games if they want to, while having it served up properly.
Seriously, these are system's that families need in general that last for up to the next 8-10 years. Yes you have to spend $3000, to $5000 to get this setup but in the long run, it's so much worth it to have your family taken care of.
oh and btw off topic, Cluster servers, will probably be done in the future by Apple at an ExTrEmElY more reduced price via Thunderbolt adapters of some sort if not just plain ethernet with MacMini, I know they already do, but this is getting really powerful and cheap. Think about if you could buy a MacMini for $599 every month for 8 months. For $4,800 you could have a MacMini cluster that you added on to month by month, at the end with 8x4x2 64 cores. (This of course is when the quads come down), pretty interesting. Cause honestly, blade servers and xserves are a joke if you can stack 32 mac mini's in the same space a Mac Pro (maybe more eek) takes, what's that 32x8 = 256 cores,
Laters...
Quote:
Since the OP kicked this off with an ARM-based Mac Pro, I figured this wasn't a reality-based thread.
You can ask one Olympian to lift 500 pounds or you can ask 10 normal people to help lift 50 pounds. Which is reality based? Both are. Is one solution better than the other? It depends. For many applications, many hands make light work. An ARM processor may be like the normal people and the latest x86 like the Olympic weight-lifter, but do I really care what is inside the machine if they can both accomplish the same work? No I don't.
One Olympic weight lifter cannot scale above its roughly 500 pound limit. But average people are cheap to come by. 100 average people can lift way more than one Olympic athlete. If each average person can lift 50 pounds, than 100 people can lift 5,000 pounds - way more than the weight lifter. Sometimes quantity has a quality all its own.
There is no question audio, video, and photo work are massively parallel problems. Are these not the main source of taxing the CPUs? Then you would be better served by a parallel solution not a serial solution. Therefore ARM chips are not a stupid solution.
Of course, one could come back and say what about having 10 Olympic weight lifters, not just one? This too is a valid solution, and in fact, the one we currently have. Which is better? I really do not care which solution I have if the work outcome is the same. Therefore, it all comes down to cost. How much does cost to hire 10 Olympic weight lifters verses how much does it cost to hire 100 average people? Realistically, how much do x86 chips cost verses a boatload of equivalent ARM chips? If they are the same cost than I really do not see why Apple should do anything different. But if one solution is much cheaper, then that is what Apple should do.
Actually, cost is not the only factor. Heat is another big deal. I am not a CPU guru so I do not know the answers to these questions. Maybe somebody else here does. I am sure Apple has this data. The fact that Apple seems to be dragging their feet on the Pro side indicates to me that a significant shift is coming soon.
This theory fits the data better than proposing Apple just drop all its pro level machines and software. What is pro level today becomes the consumer machine of tomorrow. That is why Apple needs to stay in the Pro market and why they will. Therefore I see the Xeon Phi or a massively parallel ARM new MacPro being the way forward in the near future.
Quote:
Do realize that very shortly we will have chips with 12 cores and accelerators with +50 intel compatible cores. Intel is also working on other members of the Phi family with built in super computing networking and other features. Frankly we have all sorts of new technologies on the hip origin that could go into the New Mac Pro, some of these technologies could dramatically alter how the Mac Pro is seen as a pro computer. It might not be the image you have burnt into our mind but that would not make it less of a pro machine. The whole point I'm trying to make in this thread is that Apple can't build a machine for past user needs and expect it to sell or for that matter to even get people to note it. They need to build a machine that is built for the future.
Good post from Mr. Wizard here.
Lemon Bon Bon.
Personally, I think Apple will keep on going with the slim thing.
You have 'capable' boxes. iPads. iPhones. Very powerful for their size. You have the new iMac (the most powerful iMac yet...in an even thinner enclosure...) You have the i7 Mac Mini which, if it had the 680 MX in it...would fly off the shelves, no doubt in my mind about that. (Blame Apple's artificial 'upsell' model for the lack of discrete gpu...which says it all.)
...then you have the Mac Pro. It's a big box. An out of date big box. People no longer need computers the size of a big room to do certain tasks. That's why smaller beasts do the jobs that can be done easier with lighter 'mammal' computers. eg the iPad can do 9/10 things most casual users do with their mamputer. :P
Now. The pro is wayyyyyy out of date. Over due for a facelift and a comprehensive redesign. The Wizard has been calling for a smaller but 'upgradable' beast. Does a Pro still need to be that big to include a couple of drives and dual processors and a decent gpu?
The current Mac Mini can 'context' an entry Mac Pro depending on which 'silly' benchmark you use. That's Apple's fault for not upgrading the Pro? Or is it because desktops vs laptop cpu development is flattening out. The exponential development of laptop and in particular the iPad's power is nothing short of jaw dropping. The recent 'iToys' humble the last generation of PPC Mac kit. It seems the mammals are chasing down the dinosaurs.
So what does a Mac Pro have to include? The same case? Room for 2? 3? 4? Hard drives? 2 internals? And an external port to hook up to an array? Gpus? 1 top of range? Or room for SLI? Given Apple's (and even Intel's...) recent focus on performance for watt, what does a re-imagined Pro look like given that most of the 10% (let's be generous) of really demanding tasks...for those demanding users.
I'm plumping for software to catch up to hardware. I'd offer a broader range of Pro. To increase volume? A consumer quad core with 680MX with SSD. For the pro? A 8 core cube with GPU that can fit into that enclosure. A couple of SSDs HDs. Want more power? Buy another and cluster. X-Grid. About time Apple used software to redefine power and to drive sales of their 'Pro.' For the 'Prosumer' and 'Pro.' It would perhaps be just a little bigger than x2 Mini's one on top of the other. Access via a similar mechanism to the Cube or a screw top biscuit lid aka like the Mini. Maybe with some go faster air vents on their somewhere. I want it to be as stunning as the original Cube...but have the understated brilliance of the iPhone.
Want more power? Stick an array on it. Want more power? External GPUs are around the corner. Want more displays? Use your Thunderbolt ports. etc. It's not like Towers have ever been all in ones. You have bits sticking in them and out of them. Or want more power? Just buy another 'Cube' and pass a render over to the 'grid.'
It's got to the stage where the iMac has all the power 'I' need. (I couldn't have said that when the Bondi iMac was around. I didn't think I'd ever see a top of the range iMac kicking around an entry Mac Pro. Yet here we are.)
Price a prosumer 'Pro' at £1100 (by the time you add Apple's 27 inch display, you're already at £1900!) Quad core. Decent GPU. No display. Entry 'Pro' six core at £1395 SSD/Fusion as standard. (Yes Apple...nickle and dimers that you are...) 8 core at £1795. With top end gpu included. Dual Model? Starting at 2k upward.
That's my solution. What's yours?
Lemon Bon Bon.