We can only hope that Apple truly rethinks its hardware.
Quote:
Originally Posted by Tallest Skil
Agreed.
What about Thunderbolt? Give 'em six of those.
TB is very important and I have to agree that a Pro needs those ports. I do wonder what that means for the GPU though. Right now I see that they have two options. One is to place the GPU on the motherboard the other is a specialized card slot. It is either that or TB becomes data only on the Mac Pro.
A Mac Pro would be a high performance Professional workstation. That is a machine with lots of cores and state of the art performance.
So an addition to the Mac Pro, not a replacement. But still a Xeon.
A midrange Mac is exactly that, a desktop machine running a desktop processor.
Okay, not a Xeon. They have the iMac for this. That's it.
It is no more a consumer machine than the many PCs that end up in businesses and professional offices.
So too much power for not enough use? A machine designed for that?
Really tallest I don't know why you or anybody else for that matter, have such a hard time grasping this.
What I don't understand is why such a loud minority has such a hard time grasping that Apple couldn't care less about a consumer tower.
The Mac Mini is their barebones "switcher" Mac. You don't have to replace everything you already have to get one.
The iMac is their "here's everything you need to have a modern computer" Mac. It's also the "hey, you want to do some high-level stuff, that's just fine look how well it works on just this thing" Mac.
The Mac Pro is their "you know exactly what you're doing and what you need to do it, so here it is" Mac.
Where does a consumer tower fit into that? To me it sounds like a "you don't have a clue what you need, so here's something you can screw around with until you get it right" Mac. Is that "Apple"?
…however not all workstations are high performance machines.
Haswell Mac Mini, then. Something actually cheap and still far more powerful than "the guy what works the spreadsheets" needs, but suitable for a per-cubicle basis in an office.
First off $1500 for an XMac isn't cheap so I'm not sure why you keep using the word cheap.
But you want these machines under the desks of every Bob Business and Angie Accounting like the crap Dell/HP towers we see today?
A $1500 computer with a current processor would be expensive even in PC land.
It's in "providing professional power for a specific need" price-land, though.
As far as working for 15 years that is total BS. The desktop line has been in decline for years with only the iMac holding onto sales. The fact is they have tried anything truly new outside of the Mini since Steve cam back to Apple.
G4 Cube; failure; too expensive. Sort of like the computer you're trying to pitch…
Apple actually has considerable respect in the industry now for Mac OS but sadly doesn't have the hardware most users need.
It's the same hardware as in any other PC. Better, in many cases. "Most users" don't need a tower. They don't need to get into their computer. They don't need to have a clue what to do with this little rectangle jobby here.
Apple is actually in a very strong position now with Mac OS and really should be trying to leverage that by innovating on the desktop with all the effort they out into the laptop line up. Done right Apple could increase Mac sales significantly.
Oh, I see them doing that! By getting rid of the mouse.
Handles are always nice to have. One idea that I have had is to mold them into the cube design. The "Handles" could also function as ports for airflow. A die cast housing would give them lots of options here. Done right the "Handles" wouldn't even look like handles but would rather be just another design element.
Quote:
Originally Posted by SolipsismX
I don't think it will lose it's handles and feet. I think it's likely they'll be very different but keeping it raised up for floor use and being able to carry/move the machine is important. Even if they can knock off 10 lbs that's still 30 lbs to lift, which can be even more clumsy to hold in a cube arrangement.
I love the idea of a cube but I asked previously how this can happen and still be accessible to all components without adding any complexity to the mix. I used the HDDs as an example. Imagine if you had to remove one HDD to get to one behind it or remove all of them, even if in some unifying special mount to get to the RAM risers. The only way I can see this being feasible is if Apple allows two sides to come apart. For instance, each side with the mobo running down the center. It saves money to put the chips and connectors on side of the board but we've seen with the iPhone they have plenty of skill in maximizing board space. This could reduce the mobo's footprint and by having two smaller fans on the sides for front and back they could potentially reduce fan noise and vibration, and save the user some power if the heat zones are better isolated.
Still, as cool as I can imagine I would surprised to see any cube design for a professional machine. Plus, at 20.1" tall which just fits in a rack sideways they'd be limited to just over 11" for the cube width so you can get 2 next to each otherwise you're losing space in a 23" wide rack. At 11"×11"×11" you get a whole bunch of other issues.
PS: I really do hope I'm wrong and we do see the rebirth of the NeXTcube.
Your concerns about accessibility are real and it is one of the reasons I hate the iMac so much. However I'm not sure why you believe that a cube implies bad design. It could of course but it doesn't have too. Remember we have come a very long way since the original next cube. While there are many approaches Apple could easily design a machine with a lift off shell with the majority of the components that would need tending to easily accessible from the exposed sides.
The other thing is that the cube wouldn't need a lot of the discreet crap that machines from 20 or so years ago needed. You won't need the following: a floppy, a optical, a sound card, a network card, you might not need a video car and probably a number of other things. These days all of those are on the mother board often compressed into multifunction chips. One need only look at Apples laptop to realize Apple has the engineer chops to do a high performance cube like machine.
Sadly most rack systems are 19" wide. That forces the width of the machine to less than ideal size if you want them two wide in a rack. If you want two wide per 19" rack you are looking at around 8-9 inches. Not bad for an XMAC but maybe just a little thin for a Mac Pro. Interestingly we do run into a problem with a rack convertible machine and that is proper cooling where in the front out the back is the norm.
I have the preference for a cube for practical reasons as I see them as easier to shelve or otherwise clear the clutter from my work area. Of course that implies a cube that is small enough to fit on a shelf in the first place. I can't help but to think that if the Mini was just a bit bigger it could very well give me 90% of what I'm looking for in an XMAC. That being easy access to drives, a discrete GPU and good performance. Such a Mini would certainly still be small enough to rack mount two wide for those that need it and for those that want it on a shelf someplace the Mini is no problem at all. Sometimes I think they should just kill the Mini in its current size and add an inch or so to the chassis to make it more than an entry level model especially at the $999 end.
Now that I've jumped from Mac Pro to XMac to Mini I've probably confused you even more. I hope however that I've illuminated the thread with my ideas as far as a cube based Mac Pro goes. I think it is a very possible direction for Apple but it will give up some of the features current Pro user like about the machine. The big one will be no internal disk arrays. The other thing Apple needs to do is to spur sales with unique performance features.
Handles are always nice to have. One idea that I have had is to mold them into the cube design. The "Handles" could also function as ports for airflow. A die cast housing would give them lots of options here. Done right the "Handles" wouldn't even look like handles but would rather be just another design element.
I love this idea!
The other thing is that the cube wouldn't need a lot of the discreet crap that machines from 20 or so years ago needed. You won't need the following: a floppy, a optical, a sound card, a network card, you might not need a video [card] and probably a number of other things.
It may not need them to be seperate cards but there will be slots ports that will work for all of those. I assume the primary use is for video cards. Do you think they'd get rid of PCIe slots? If so, then why even have this machine? Having two TB ports where the old and busted FW800 ports were would allow for a nice drop in price if they don't include a discrete GPU as standard. If it is for a server room then HD4000 is sufficient for most needs.
Sadly most rack systems are 19" wide. That forces the width of the machine to less than ideal size if you want them two wide in a rack. If you want two wide per 19" rack you are looking at around 8-9 inches. Not bad for an XMAC but maybe just a little thin for a Mac Pro. Interestingly we do run into a problem with a rack convertible machine and that is proper cooling where in the front out the back is the norm.
I didn't recall with more common but choose the largest for the sake of the argument as it still didn't paint a pretty picture. At 19" my shortened mockup seems more plausible.
Of course that implies a cube that is small enough to fit on a shelf in the first place.
How big is the average shelf? How much weight can it support? These are questions that would need to be addressed if they were to build a cube and I don't think the answer will be pleasant.
Sometimes I think they should just kill the Mini in its current size and add an inch or so to the chassis to make it more than an entry level model especially at the $999 end.
But they haven't done that and it seems to be pretty popular since they keep refreshing it so I think we need to understand that product better if we want to understand how the Mac Pro might change.
I think it is a very possible direction for Apple but it will give up some of the features current Pro user like about the machine.
Possible, sure, but in no way do I think it's the most probable solution to expect.
The big one will be no internal disk arrays.
I think that's away off. They still sell the MBP, Mac mini, and iMac with HDDS so I doubt that will come for some time. Even after those have all gone SSD I think the Mac Pro will still have the HDDs for a good time after (but part of that is because they change the design so infrequently).
You really don't know what I'm talking about and frankly I'm not sure what I can say to get you to grasp the fact that what I'm talking about has nothing to do with the consumer market!!! I'm talking very much about a machine for professional use where a workstation class machine isn't required. A machine with slots is often a requirement though.
I have to wonder if you have ever been in a very large complex that has offices, R&D, engineering, manufacturing and other areas where computers are used. I know of instances where several hindered PCs are used and not a one of them is used in the capacity of an office computer. In these sorts of installations standardized hardware makes support far easier and as such the hardware has to have PCIe slots. Those slots might employ any number of specialized cards or they may have no cards at all in them. The important thing here is the cost of the hardware and as such the Mac Pro isn't even in the ball game. A suitable compact XMac might be if the cost was reasonable. The point is Apple isn't even considered for such uses because they don't have professional class hardware priced right for the use. The rest of the building isn't likely to use an iMac either because built in screens aren't acceptable for professional usage. Basically Apple through its bull headed approach to desktop hardware hasn't had a chance in hell of getting placed into these sorts of business settings.
Now it is pretty well know that Steve didn't give a damn about the needs of business. It wasn't his goal to think different for the business world. However as a company Apple can change to at least offer something suitable. It might actually take a while for business to catch on but considering the quality problems that exist for the mainstream manufactures I really think they could start to win a few installations with the right hardware.
Quote:
Originally Posted by Tallest Skil
So an addition to the Mac Pro, not a replacement. But still a Xeon.
Okay, not a Xeon. They have the iMac for this. That's it.
So too much power for not enough use? A machine designed for that?
What I don't understand is why such a loud minority has such a hard time grasping that Apple couldn't care less about a consumer tower.
The Mac Mini is their barebones "switcher" Mac. You don't have to replace everything you already have to get one.
The iMac is their "here's everything you need to have a modern computer" Mac. It's also the "hey, you want to do some high-level stuff, that's just fine look how well it works on just this thing" Mac.
The Mac Pro is their "you know exactly what you're doing and what you need to do it, so here it is" Mac.
Where does a consumer tower fit into that? To me it sounds like a "you don't have a clue what you need, so here's something you can screw around with until you get it right" Mac. Is that "Apple"?
Haswell Mac Mini, then. Something actually cheap and still far more powerful than "the guy what works the spreadsheets" needs, but suitable for a per-cubicle basis in an office.
But you want these machines under the desks of every Bob Business and Angie Accounting like the crap Dell/HP towers we see today?
It's in "providing professional power for a specific need" price-land, though.
G4 Cube; failure; too expensive. Sort of like the computer you're trying to pitch…
It's the same hardware as in any other PC. Better, in many cases. "Most users" don't need a tower. They don't need to get into their computer. They don't need to have a clue what to do with this little rectangle jobby here.
Oh, I see them doing that! By getting rid of the mouse.
You talk a lot about consumer but frankly I think that is all you are familiar with. Like it or not Apple doesn't have a midrange business solution. You can point all you want at the iMac but nobody with any sort of business sense would buy that for use outside of an office environment. Frankly I see more PC outside the office than I do inside them.
As to that Haswell Mini it might turn my head for personal usage. But putting it into a diverse business setting would be difficult at best.
If I was an artist I would draw up a mock up but sadly it would look pretty pathetic.
Quote:
Originally Posted by SolipsismX
I love this idea!
It may not need them to be seperate cards but there will be slots ports that will work for all of those. I assume the primary use is for video cards. Do you think they'd get rid of PCIe slots? If so, then why even have this machine? Having two TB ports where the old and busted FW800 ports were would allow for a nice drop in price if they don't include a discrete GPU as standard. If it is for a server room then HD4000 is sufficient for most needs.
In a nut shell I see the need for slots for the foreseeable future. FireWire should go, that I agree with. A Pro machine should have at least 4 TB ports though. I'm not sure if the technology is there yet for 4 TB ports.
Quote:
I didn't recall with more common but choose the largest for the sake of the argument as it still didn't paint a pretty picture. At 19" my shortened mockup seems more plausible.
It really doesn't matter as this isn't Apples primary focus. I just see designs that easily convert to Rack Mount as solving issues for a certain number of Apples customers. Done right it doesn't cost any extra for the mainstream user so why not.
Quote:
How big is the average shelf? How much weight can it support? These are questions that would need to be addressed if they were to build a cube and I don't think the answer will be pleasant.
Well there are such elves then there are shelves. Book shelves, if properly built, shouldn't be a problem. Think of a stack of books end to end. This is one reason why library shelving is so expensive you end up with considerable weight on those racks.
Quote:
But they haven't done that and it seems to be pretty popular since they keep refreshing it so I think we need to understand that product better if we want to understand how the Mac Pro might change.
I use to think the Mini was a joke but it is growing on me. The fact is advancements in chip technology have turned the Mini into something that is at least respectable. Unfortunately it just doesn't have the graphics chops right now for all of my needs. I'm hopeful that Apple does right by Haswell in the Mini and slots in a GT3 supporting chip
Quote:
Possible, sure, but in no way do I think it's the most probable solution to expect.
I think that's away off. They still sell the MBP, Mac mini, and iMac with HDDS so I doubt that will come for some time. Even after those have all gone SSD I think the Mac Pro will still have the HDDs for a good time after (but part of that is because they change the design so infrequently).
The new Mac Pro will likely have space for at most two magnetic drives. External disk arrays will be connected by TB for those that need them. Primary storage will be SSD hopefully sitting on the PCI bus. Getting rid of the capacity to hold a disk array internally will allow for a significant shrinkage of the chassis and significantly reduce power supply size. In fact I wouldn't be surprised to see Apple put laptop sized drives in the new Mac Pro to save even more power. It has worked out well in the server market so it certainly can be done in the workstation market.
In any event I know some of my wishes for the Mac Pro seem like long shots to many in this thread but really Apple needs a workstation design that can carry forward for another 8 years or so. That means things like SSD for main storage is a must have. Support for TB is also a must have which brings into question how they will handle the GPU. Further things like 3D ram modules could dramatically change the way a machine is built allowing for size compact action once again. We probably won't get everything that I salivate over but if we get nothing then you might as well call Apple dead on the desktop.
As for what Apple is competing with the answer is simple they aren't competing with anything as they are a single source supplier for Macs. The goal here isn't competeing but meeting customer needs. The primary need here is a machine that is headless but yet a respectable performer. Further that machine should have a bit of modern technology at least as interesting as Apple laptops and some expansion capability. It isn't a lot to ask for and it would generat a lot of new sales.
It's often suggested that Apple can generate a lot more sales by having more options in the lineups but you have to demonstrate there is a significant market for the suggested options and the figures don't back it up for a tower in the $1000-2000 range. Apple already has the premium PC market locked down, just as they have with the tablet and smartphone markets.
I love the idea of a cube but I asked previously how this can happen and still be accessible to all components without adding any complexity to the mix.
He seems to describe a half cube for rack-mounting, which isn't technically a cube e.g 13x13x7. This is essentially a smaller Mac Pro on its side.
If they were going with an actual cube, I don't think it would be good going above 8". The 12" NeXT Cube was far too big:
It would have the same footprint as the Mac Pro does now but I wouldn't like to see it take up more horizontal space. If it has to sit on the floor, it'll get in the way of your feet and it certainly won't sit on the desk. I'd like to see a machine people can sit on their desk quite easily.
Based on the designs some were PPC-based. I was told they need to be rebooted once a week.
My, how times have changed. I once disabled Software Update on my Mini and it had an uptime of 260 days. Like TS, I wonder why they needed the weekly reboot.
I think that's away off. They still sell the MBP, Mac mini, and iMac with HDDS so I doubt that will come for some time. Even after those have all gone SSD I think the Mac Pro will still have the HDDs for a good time after (but part of that is because they change the design so infrequently).
I understand your point here sol, but do not through out the idea of a HDD-less Mac Pro. Apple can easily determine how the MP is being used, and if the percentage of uses with very large datasets far outweigh the number of uses with a single HDD in the MP it might be a good design. One or two SSD's for the OS and Apps and all HDD's external; a rebirth of the Xserve RAID if you will. It'll allow them to shrink the PSU and create a small Form Factor. Though personally I don't see any benefit to a redesign chassis. The current on will be 10 years old this June, so it's not without reason they haven't changed it. I do think they need to redesign the handles as they are annoyingly sharp if you need to carry one more than a foot or two.
It would have the same footprint as the Mac Pro does now but I wouldn't like to see it take up more horizontal space. If it has to sit on the floor, it'll get in the way of your feet and it certainly won't sit on the desk. I'd like to see a machine people can sit on their desk quite easily.
The old school desktop design, triple pizza box, was ok during the CRT area. Flatscreens gave the tower a favorable design. The iMac is the AIO and perfect as-is. They could go all external, simply put the MP in the keyboard with build in SSD, connect up to 14 HDD's over TB, create a new 30-40" ACD with a removable videocard slot. PCIe goes into an external box for those who need it.
Well, maybe not. I certainly don't see a reason for a redesigned chassis. And rack mountable? If they want to return to the enterprice/server market, they should start with a clean slate, not redesign a desktop computer with a 1GB of GDDR5 memory videocard.
Thunderbolt isn't a slot and its limited adoption means you won't see specialized cards moving to outside the box anytime soon. In simple terms there is no effective replacement for slots at this point. Frankly it doesn't appear that either Apple nor Intel are in a rush to promote the adoption of TB. Maybe that will change 5 years down the road but even then you still have legacy hardware to support.
Quote:
Originally Posted by Tallest Skil
And the iMac is invalidated because "slots", despite Thunderbolt providing that, correct?
… Not the glossy argument again. Is it?
Where do you get these ideas? Glossy has nothing to do with screen failures. The simple fact is screens go bad and you are at a huge disadvantage if that means replacing the entire PC. Beyond that screens are often specialized for the task at hand. For example a small screen for a crampe d work area while another might have a touch screen. Not to mention the iMacs are rather big machines and as such get in the way simply because the screen is large. If you have a bench full of lab equipment finding room for a 27" screen is an issue.
It makes a much bigger difference for some things than others. Admittedly I hate the way that most of these panels deal with glare reduction, but I hate glossy even more. I have used imacs in the past, and the reflections can be extremely annoying when dealing with colors. I'm not sure how much you know about displays, but anything gamma encoded works so that 1 + 1 != 2 either visually or physically. It's used to display images, text, etc. in a way that looks good at a lower contrast ratio. The appearance of parallel light rays reflecting off glass is really additive to these values, so it really decreases the perceived separation in them. If it was a complete non-issue, Apple wouldn't have attempted to mitigate the issue. If you completely blacken the room it's quite tolerable, but having to do that is annoying.
Quote:
Originally Posted by wizard69
Thunderbolt isn't a slot and its limited adoption means you won't see specialized cards moving to outside the box anytime soon. In simple terms there is no effective replacement for slots at this point. Frankly it doesn't appear that either Apple nor Intel are in a rush to promote the adoption of TB. Maybe that will change 5 years down the road but even then you still have legacy hardware to support.
Where do you get these ideas? Glossy has nothing to do with screen failures. The simple fact is screens go bad and you are at a huge disadvantage if that means replacing the entire PC. Beyond that screens are often specialized for the task at hand. For example a small screen for a crampe d work area while another might have a touch screen. Not to mention the iMacs are rather big machines and as such get in the way simply because the screen is large. If you have a bench full of lab equipment finding room for a 27" screen is an issue.
Actually thunderbolt has at least working products in a lot of the higher end specialty hardware, basically in the $500-5000 range. You don't have a lot of simple solutions without a relatively high minimum purchase though, which makes me somewhat skeptical of long term support. I don't think it's meant to be ubiquitous. It won't support the more extreme bandwidth requirements, and its ability to support multiple devices is extremely limited based on the device type. There are many things that don't work, and it's not clearly explained anywhere. For example a lot of people buy other displays then mistakenly try to use thunderbolt cables from Apple. The Apple display will support another thunderbolt display. For slightly atypical requirements, it's really possible to run into situations where the combination of desired hardware won't work with the available ports. I wouldn't personally make big investments in thunderbolt hardware. Marvin actually pointed out a while ago that some of these lists would be incredibly expensive for freelance individuals, yet at the freelance or smaller shop level, people don't always replace all their hardware simultaneously. Due to the cost, it's sensible to avoid the potential for orphaned peripheral devices unless they're paid for very quickly. I've personally gone the route of only purchasing computing equipment that can be paid for or justified within 12-18 months to avoid the issue of frankensteined setups.
In my experience, if Apple wants to drop something, support drops off very quickly. The express 34 slot remained on the 17" macbook pros longer than the others, yet reliable drivers and support for any of those cards dropped off a long time ago. Many of them aren't supported with Lion and on. A new thunderbolt revision could go the same way.
It makes a much bigger difference for some things than others. Admittedly I hate the way that most of these panels deal with glare reduction, but I hate glossy even more. I have used imacs in the past, and the reflections can be extremely annoying when dealing with colors. I'm not sure how much you know about displays, but anything gamma encoded works so that 1 + 1 != 2 either visually or physically. It's used to display images, text, etc. in a way that looks good at a lower contrast ratio. The appearance of parallel light rays reflecting off glass is really additive to these values, so it really decreases the perceived separation in them. If it was a complete non-issue, Apple wouldn't have attempted to mitigate the issue. If you completely blacken the room it's quite tolerable, but having to do that is annoying.
If you haven't seen the new iMacs, please look at them some time. I had another chance to look at them this weekend. It's very astonishing how strong the AR treatment is. It's the strongest AR treatment I remember seeing. Even in a brightly lit retail environment, the reflections are minimal, sometimes I had to look hard for reflections. And I'm annoyed with the 2011 iMac's reflections, in the 2012 models, I consider it to be a non-issue.
anything gamma encoded works so that 1 + 1 != 2 either visually or physically. The appearance of parallel light rays reflecting off glass is really additive to these values, so it really decreases the perceived separation in them.
You're going to perceive more of a reflection in the darker areas of an image but that's true of any display. The glare used to be bad but the new one has reduced it to the point that the glare isn't really a problem any more.
Obviously if you sit with your back to an open window, it's still a problem but it's not a mirror like it used to be and much easier to fix - a matte display would wash a reflection that bright over the whole display:
There are many things that don't work, and it's not clearly explained anywhere. For example a lot of people buy other displays then mistakenly try to use thunderbolt cables from Apple.
A lot (I assume you mean thousands) of people are buying non-Apple Thunderbolt displays and an Apple Thunderbolt cable? Why would they be doing that, what displays are they and why exactly don't they work?
I wouldn't personally make big investments in thunderbolt hardware.
Of course not because you want to see it fail for whatever reason. You'd presumably have been happy to buy a PCI-X device though only to see it made obsolete with PCIe. That standard only lasted 8 years. You probably also don't want to invest in DDR3 RAM as they'll change the slots with DDR4. Avoid Mini-displayport too, that'll probably fail.
In my experience, if Apple wants to drop something, support drops off very quickly. The express 34 slot remained on the 17" macbook pros longer than the others, yet reliable drivers and support for any of those cards dropped off a long time ago. Many of them aren't supported with Lion and on. A new thunderbolt revision could go the same way.
Yes, I'm sure a new Thunderbolt revision will obsolete all existing Thunderbolt hardware just like Firewire 800 made all Firewire 400 devices obsolete. No hang on, you could just plug one into the other with an updated cable. Maybe Apple and Intel can do that. We can only hope they know what they're doing but yeah, in the meantime don't buy any products from any manufacturer because it's not going to last more than 10 years, which is the minimum you should expect anything to last.
If you haven't seen the new iMacs, please look at them some time. I had another chance to look at them this weekend. It's very astonishing how strong the AR treatment is. It's the strongest AR treatment I remember seeing. Even in a brightly lit retail environment, the reflections are minimal, sometimes I had to look hard for reflections. And I'm annoyed with the 2011 iMac's reflections, in the 2012 models, I consider it to be a non-issue.
I will certainly take another look at the iMac when I need to replace my Mac Pro. I keep on reading posts like yours, and the possibility of switching to a cheaper solution seems all the more viable. The only 'heavy-duty stuff I do is mostly Aperture, and that doesn't require all the horsepower my Mac Pro is giving me. Still, the 4xHDD does have its advantages, and I even use the ODD. Or at least used to -- I've finished ripping all my optical discs except for the DVD's that are stacked up next to my DVD player. But those are all BR, and I simply cannot picture myself of buying a read-only BR drive to add another geez-knows-how-many-gigabytes to my current, hmm, I don't even know, but the internal storage on a MP is really great. Anyone know if there's a real storage bump looming on the horizon, like going from the current 4TB to, say, 200TB or something similarly crazy?
That's off topic, but the screen seems a likely candidate as a replacement for my 30" ACD. Thanks for your post JeffDM.
@Marvin: that is the most useful comparison picture I have ever seen, thank you very much indeed for that!
but how far out they are is never clear. I wouldn't expect them to jump very far because they'd sell fewer of them. Most likely they'll go to 6TB next year. They can probably go up 2TB every year on the 3.5" models:
If you haven't seen the new iMacs, please look at them some time. I had another chance to look at them this weekend. It's very astonishing how strong the AR treatment is. It's the strongest AR treatment I remember seeing. Even in a brightly lit retail environment, the reflections are minimal, sometimes I had to look hard for reflections. And I'm annoyed with the 2011 iMac's reflections, in the 2012 models, I consider it to be a non-issue.
Agreed. I'm using the 2010 iMac now, and it's generally workable in a pro setting. The glossy screen took some getting used to, but it's usable.
The main issue with the iMac is the lack of a user-replaceable hard drive.
Having to lose six working days for Apple's hard drive recall is ridiculous and downright inexcusable for pro users.
If the iMac is what Apple wants more pros to use, they should sell a version of the 27" without a hard drive at all, so that all confidential files can be kept safely at home/work if the machine has to be serviced. OWC could sell a version of the MiniStack made for iMac users with a RAID drive and removable Time Machine backup, and many pro users would, I think, be content with that.
I wouldn't expect them to jump very far because they'd sell fewer of them.
This is the first time I've thought of this regarding hard drives. And it makes me livid, because I wouldn't put it past them.
That's not how I'd operate, at least. If I had the ability to create a product ten times better than anything on the market, I'd do it. *shrug* That's all that needs said. I wouldn't make something twice as good to sell "more" of them, I'd make something ten times as good and watch the entire market buy my product because of the immensity of its improvement.
If the improvement is big enough, you'll start to pull in not only the people who wouldn't have bought it at a 2x improvement, but also the people who weren't planning to buy a new one at all. That's what they should be going for. That kind of one-upmanship is what we need in hard drives these days, in both camps.
You're going to perceive more of a reflection in the darker areas of an image but that's true of any display. The glare used to be bad but the new one has reduced it to the point that the glare isn't really a problem any more.
Obviously if you sit with your back to an open window, it's still a problem but it's not a mirror like it used to be and much easier to fix - a matte display would wash a reflection that bright over the whole display:
I have never seen a side by side. Thanks for that. For anything really critical, I suggest subdued lighting anyway, as there aren't any imac display hoods. I would even recommend it with matte displays if possible if you really need consistency. I also suggest allowing it to warm up for 30 minutes or so prior to judging color from a previous session due to shifts as the display stabilizes. Some are better than others though. I was trying to explain that reflections don't really follow the same behavior as a color managed device that emits light.
Quote:
What are some examples of Thunderbolt solutions that are vastly more expensive than equivalent PCIe solutions?
You're really misinterpreting my posts today. Wizard mentioned that you won't see specialized cards migrate. I mentioned that the expensive, specialized ones were the first to release thunderbolt versions. I've mentioned some of the tools made by Black Magic before, as they were some of the first to ship thunderbolt products. My claim was that it has working products from companies that have a sizable market on the Mac and margins that can absorb the extra development cost. I haven't seen as much consumer grade stuff at similar price points to what you would pay for usb3.
Quote:
Why won't it support more extreme bandwidths? That's in their roadmap.
I'm skeptical as to what machines will get this, the state of daisy chaining if you need to run a number of peripherals from different companies with different firmware, and intel's claims compared to what is actually delivered. They projected 10 years out with some uncertainty. It's also dependent on them adding that many PCI lanes, so that thunderbolt may derive such bandwidth and completely ignores the fact that display bandwidth may grow immensely by the end of the decade given the emergence of higher dpi displays. I'm still skeptical as to when we'll have shipping desktop products of that type. It started with phones simply because of the huge growth and small screen size. Desktop displays have been relatively stagnant for years though. The biggest difference is that prices have come down.
Quote:
A lot (I assume you mean thousands) of people are buying non-Apple Thunderbolt displays and an Apple Thunderbolt cable? Why would they be doing that, what displays are they and why exactly don't they work?
Again misinterpretation! I meant people buying a non-Apple display to use in conjunction with their mini or to add a large screen to their notebook. I wouldn't make that mistake on cables, but some people seem to do so. I actually
Quote:
Of course not because you want to see it fail for whatever reason. You'd presumably have been happy to buy a PCI-X device though only to see it made obsolete with PCIe. That standard only lasted 8 years. You probably also don't want to invest in DDR3 RAM as they'll change the slots with DDR4. Avoid Mini-displayport too, that'll probably fail.
None of that is really true. It was weird how the last G5s got PCIe with no PCI-X slots so it took half a cycle to see anything for them. In that sense it was an awkward transition to PCIe. You also ignored some of my previous points that I know you have read. Ram isn't really something you pass between machines, so the argument makes little sense there. Mini displayport works the same as displayport. The only thing I've complained about is lack of full displayport 1.2 and 10 bit out support from Apple.
Quote:
Yes, I'm sure a new Thunderbolt revision will obsolete all existing Thunderbolt hardware just like Firewire 800 made all Firewire 400 devices obsolete. No hang on, you could just plug one into the other with an updated cable. Maybe Apple and Intel can do that. We can only hope they know what they're doing but yeah, in the meantime don't buy any products from any manufacturer because it's not going to last more than 10 years, which is the minimum you should expect anything to last.
You misinterpreted me on a number of these points. I was saying whenever Apple changes direction on something, support drops off quickly whether it's from Apple, third parties, or a combination. Sometimes it can be 1-2 years from discontinuation. If you look at what I mentioned regarding staggering purchases of peripherals and primary computing devices, sometimes it's possible to just buy at a bad time (like late cycle or prior to a transition) because you happened to need it then. Right now I'm waiting for some stability there prior to investing in any storage solutions based around thunderbolt. The other problem is that it's coupled to a display, yet not every peripheral allows for daisy chaining.
Having to lose six working days for Apple's hard drive recall is ridiculous and downright inexcusable for pro users.
I wanted them to make 2.5" drives accessible from the base but they obviously realised that people would just end up buying the entry models and put Samsung or Crucial SSDs in them. It also complicates their manufacturing having accessible panels on the outside but I think it would have helped them in some ways because they'd have happier customers and faster service times. A technician could replace a broken drive in 5 minutes while someone was standing in the store instead of prizing the display off and gluing it back on again.
Accessible slots don't work so well for the Fusion drive setup though because if someone removes one drive, it would break it. Also, they obviously wanted 3.5" drives for the extra capacity and they are better being tightly secured to minimize noise. I'd be quite happy with an option where they put a blade SSD screwed in beside the RAM slots on the 27" and offered a 256GB and 512GB option on its own. There aren't really many competitively priced blade SSDs. There's a chance the SSD could get too hot that way but they can probably cool it down quite easily and it's a setup I'd prefer.
I wouldn't make something twice as good to sell "more" of them, I'd make something ten times as good and watch the entire market buy my product because of the immensity of its improvement.
The problem there though is sustaining a large company. If you sell a product that will mean you saturate the market very quickly, the sales will start to tail off and you might have to downsize the company. There are two sides to the ethics of this because while technological progress means that consumers get better products at lower prices, it's harder for companies to sustain growth and jobs will be lost all the way from manufacuring to retail.
I think the way they do things now is the better way simply because the whole reason anybody really does anything is to try and improve the quality of life and having a sustainable economic model is more important to this than having technology that isn't really essential. There's a balance that has to be made so that we don't just have to put up with junk products as that affects growth too but it wouldn't be wise of them to give us everything all at once. I really wonder what's going to happen to the whole tech industry in another 20-30 years given how far it has come in that time. It seems very much like computers will go the way of televisions and it all just becomes about the services. No doubt that's why companies are making moves to develop their own OS as that's the only guaranteed point of leverage anyone will eventually have over profitable services.
Right now, HDDs are around $0.05/GB and SSD is $0.60/GB. I can see all manufacturers just ditching HDDs and moving to faster, more profitable SSDs. There's not much point in having 200TB drives if it can still only read/write at 150MB/s and more importantly has really low random read/write speeds. SSD just needs to get down to $0.10/GB to replace any size of HDD because you can just stick more chips in. There's not the same weight and noise issues like with mechanical platters. Imagine in 2 years, they half SSD prices to get to $0.30/GB and another 2 years, it's $0.15/GB, you'd potentially get a 2TB 2.5" SSD for $300. Even if you can get a single 20TB HDD for the same price, people will still need RAID systems for reliability and if you have files that take up that much space, you need high performance.
I doubt we'll ever see consumer HDDs that are 60TB+ in size because the market appeal for them would be so small. I can see them focusing on speed more than capacity to the point that RAM and storage are one and the same - in effect 2TB of storage becomes 2TB of RAM and video memory or at the very least is tiered storage with small amounts of very fast memory and of course all soldered to the motherboard. While it seems crazy now to solder storage to the motherboard, when computers are so inexpensive, it will be a better model to just replace the whole board. The iPad Mini is $329, which means the parts cost less than $200 and that includes the display. For any board or chip failure, they'd charge $100 for the whole part.
I haven't seen as much consumer grade stuff at similar price points to what you would pay for usb3.
There are a few but you wouldn't say that there aren't as many consumer grade expressCard or PCIe cards as USB 3 products. The problem is that people class Thunderbolt as just a competitor to USB 3 and it's really missing the point of what Thunderbolt is. It's essentially a faster expressCard merged with the mini-displayport output.
There's a premium for the faster ports and active cables. Lacie's drives come with a $50 premium. G-tech is a bit more with 4TB USB 3 at $399 and with Thunderbolt, $538. Thunderbolt needs more volume, cheaper parts and faster licensing steps but it takes time for this to happen. The problem as always lies with cheap product manufacturers that side with USB 3 because it's good enough but it will never negate the need for multi-protocol ports.
I meant people buying a non-Apple display to use in conjunction with their mini or to add a large screen to their notebook. I wouldn't make that mistake on cables, but some people seem to do so.
A Thunderbolt cable wouldn't even plug into the display though so it's not even remotely a problem with Thunderbolt and I doubt it's a widespread issue. The laptops and Minis have HDMI for consumer displays.
Right now I'm waiting for some stability there prior to investing in any storage solutions based around thunderbolt.
Would you have hesitated to buy an eSATA drive? That standard is going to totally die out in favour of USB 3. At the end of the day, consumer electronics is like cars.You don't expect to have the same car for more than 10 years because you'd be able to afford a better one after that period of time. If a new storage standard comes along, you copy the data over to it and move on with your life.
You make out like Apple will get bored with Thunderbolt and just phase it out within 1-2 years. People said the same with displayport vs the much more widely adopted HDMI. They co-exist now just like TB and USB 3.
Comments
We can only hope that Apple truly rethinks its hardware.
Quote:
Originally Posted by Tallest Skil
Agreed.
What about Thunderbolt? Give 'em six of those.
TB is very important and I have to agree that a Pro needs those ports. I do wonder what that means for the GPU though. Right now I see that they have two options. One is to place the GPU on the motherboard the other is a specialized card slot. It is either that or TB becomes data only on the Mac Pro.
Originally Posted by wizard69
A Mac Pro would be a high performance Professional workstation. That is a machine with lots of cores and state of the art performance.
So an addition to the Mac Pro, not a replacement. But still a Xeon.
A midrange Mac is exactly that, a desktop machine running a desktop processor.
Okay, not a Xeon. They have the iMac for this. That's it.
It is no more a consumer machine than the many PCs that end up in businesses and professional offices.
So too much power for not enough use? A machine designed for that?
Really tallest I don't know why you or anybody else for that matter, have such a hard time grasping this.
What I don't understand is why such a loud minority has such a hard time grasping that Apple couldn't care less about a consumer tower.
The Mac Mini is their barebones "switcher" Mac. You don't have to replace everything you already have to get one.
The iMac is their "here's everything you need to have a modern computer" Mac. It's also the "hey, you want to do some high-level stuff, that's just fine look how well it works on just this thing" Mac.
The Mac Pro is their "you know exactly what you're doing and what you need to do it, so here it is" Mac.
Where does a consumer tower fit into that? To me it sounds like a "you don't have a clue what you need, so here's something you can screw around with until you get it right" Mac. Is that "Apple"?
…however not all workstations are high performance machines.
Haswell Mac Mini, then. Something actually cheap and still far more powerful than "the guy what works the spreadsheets" needs, but suitable for a per-cubicle basis in an office.
First off $1500 for an XMac isn't cheap so I'm not sure why you keep using the word cheap.
But you want these machines under the desks of every Bob Business and Angie Accounting like the crap Dell/HP towers we see today?
A $1500 computer with a current processor would be expensive even in PC land.
It's in "providing professional power for a specific need" price-land, though.
As far as working for 15 years that is total BS. The desktop line has been in decline for years with only the iMac holding onto sales. The fact is they have tried anything truly new outside of the Mini since Steve cam back to Apple.
G4 Cube; failure; too expensive. Sort of like the computer you're trying to pitch…
Apple actually has considerable respect in the industry now for Mac OS but sadly doesn't have the hardware most users need.
It's the same hardware as in any other PC. Better, in many cases. "Most users" don't need a tower. They don't need to get into their computer. They don't need to have a clue what to do with this little rectangle jobby here.
Apple is actually in a very strong position now with Mac OS and really should be trying to leverage that by innovating on the desktop with all the effort they out into the laptop line up. Done right Apple could increase Mac sales significantly.
Oh, I see them doing that! By getting rid of the mouse.
Handles are always nice to have. One idea that I have had is to mold them into the cube design. The "Handles" could also function as ports for airflow. A die cast housing would give them lots of options here. Done right the "Handles" wouldn't even look like handles but would rather be just another design element.
Quote:
Originally Posted by SolipsismX
I don't think it will lose it's handles and feet. I think it's likely they'll be very different but keeping it raised up for floor use and being able to carry/move the machine is important. Even if they can knock off 10 lbs that's still 30 lbs to lift, which can be even more clumsy to hold in a cube arrangement.
I love the idea of a cube but I asked previously how this can happen and still be accessible to all components without adding any complexity to the mix. I used the HDDs as an example. Imagine if you had to remove one HDD to get to one behind it or remove all of them, even if in some unifying special mount to get to the RAM risers. The only way I can see this being feasible is if Apple allows two sides to come apart. For instance, each side with the mobo running down the center. It saves money to put the chips and connectors on side of the board but we've seen with the iPhone they have plenty of skill in maximizing board space. This could reduce the mobo's footprint and by having two smaller fans on the sides for front and back they could potentially reduce fan noise and vibration, and save the user some power if the heat zones are better isolated.
Still, as cool as I can imagine I would surprised to see any cube design for a professional machine. Plus, at 20.1" tall which just fits in a rack sideways they'd be limited to just over 11" for the cube width so you can get 2 next to each otherwise you're losing space in a 23" wide rack. At 11"×11"×11" you get a whole bunch of other issues.
PS: I really do hope I'm wrong and we do see the rebirth of the NeXTcube.
Your concerns about accessibility are real and it is one of the reasons I hate the iMac so much. However I'm not sure why you believe that a cube implies bad design. It could of course but it doesn't have too. Remember we have come a very long way since the original next cube. While there are many approaches Apple could easily design a machine with a lift off shell with the majority of the components that would need tending to easily accessible from the exposed sides.
The other thing is that the cube wouldn't need a lot of the discreet crap that machines from 20 or so years ago needed. You won't need the following: a floppy, a optical, a sound card, a network card, you might not need a video car and probably a number of other things. These days all of those are on the mother board often compressed into multifunction chips. One need only look at Apples laptop to realize Apple has the engineer chops to do a high performance cube like machine.
Sadly most rack systems are 19" wide. That forces the width of the machine to less than ideal size if you want them two wide in a rack. If you want two wide per 19" rack you are looking at around 8-9 inches. Not bad for an XMAC but maybe just a little thin for a Mac Pro. Interestingly we do run into a problem with a rack convertible machine and that is proper cooling where in the front out the back is the norm.
I have the preference for a cube for practical reasons as I see them as easier to shelve or otherwise clear the clutter from my work area. Of course that implies a cube that is small enough to fit on a shelf in the first place. I can't help but to think that if the Mini was just a bit bigger it could very well give me 90% of what I'm looking for in an XMAC. That being easy access to drives, a discrete GPU and good performance. Such a Mini would certainly still be small enough to rack mount two wide for those that need it and for those that want it on a shelf someplace the Mini is no problem at all. Sometimes I think they should just kill the Mini in its current size and add an inch or so to the chassis to make it more than an entry level model especially at the $999 end.
Now that I've jumped from Mac Pro to XMac to Mini I've probably confused you even more. I hope however that I've illuminated the thread with my ideas as far as a cube based Mac Pro goes. I think it is a very possible direction for Apple but it will give up some of the features current Pro user like about the machine. The big one will be no internal disk arrays. The other thing Apple needs to do is to spur sales with unique performance features.
I love this idea!
It may not need them to be seperate cards but there will be slots ports that will work for all of those. I assume the primary use is for video cards. Do you think they'd get rid of PCIe slots? If so, then why even have this machine? Having two TB ports where the old and busted FW800 ports were would allow for a nice drop in price if they don't include a discrete GPU as standard. If it is for a server room then HD4000 is sufficient for most needs.
I didn't recall with more common but choose the largest for the sake of the argument as it still didn't paint a pretty picture. At 19" my shortened mockup seems more plausible.
How big is the average shelf? How much weight can it support? These are questions that would need to be addressed if they were to build a cube and I don't think the answer will be pleasant.
But they haven't done that and it seems to be pretty popular since they keep refreshing it so I think we need to understand that product better if we want to understand how the Mac Pro might change.
Possible, sure, but in no way do I think it's the most probable solution to expect.
I think that's away off. They still sell the MBP, Mac mini, and iMac with HDDS so I doubt that will come for some time. Even after those have all gone SSD I think the Mac Pro will still have the HDDs for a good time after (but part of that is because they change the design so infrequently).
You really don't know what I'm talking about and frankly I'm not sure what I can say to get you to grasp the fact that what I'm talking about has nothing to do with the consumer market!!! I'm talking very much about a machine for professional use where a workstation class machine isn't required. A machine with slots is often a requirement though.
I have to wonder if you have ever been in a very large complex that has offices, R&D, engineering, manufacturing and other areas where computers are used. I know of instances where several hindered PCs are used and not a one of them is used in the capacity of an office computer. In these sorts of installations standardized hardware makes support far easier and as such the hardware has to have PCIe slots. Those slots might employ any number of specialized cards or they may have no cards at all in them. The important thing here is the cost of the hardware and as such the Mac Pro isn't even in the ball game. A suitable compact XMac might be if the cost was reasonable. The point is Apple isn't even considered for such uses because they don't have professional class hardware priced right for the use. The rest of the building isn't likely to use an iMac either because built in screens aren't acceptable for professional usage. Basically Apple through its bull headed approach to desktop hardware hasn't had a chance in hell of getting placed into these sorts of business settings.
Now it is pretty well know that Steve didn't give a damn about the needs of business. It wasn't his goal to think different for the business world. However as a company Apple can change to at least offer something suitable. It might actually take a while for business to catch on but considering the quality problems that exist for the mainstream manufactures I really think they could start to win a few installations with the right hardware.
Quote:
Originally Posted by Tallest Skil
So an addition to the Mac Pro, not a replacement. But still a Xeon.
Okay, not a Xeon. They have the iMac for this. That's it.
So too much power for not enough use? A machine designed for that?
What I don't understand is why such a loud minority has such a hard time grasping that Apple couldn't care less about a consumer tower.
The Mac Mini is their barebones "switcher" Mac. You don't have to replace everything you already have to get one.
The iMac is their "here's everything you need to have a modern computer" Mac. It's also the "hey, you want to do some high-level stuff, that's just fine look how well it works on just this thing" Mac.
The Mac Pro is their "you know exactly what you're doing and what you need to do it, so here it is" Mac.
Where does a consumer tower fit into that? To me it sounds like a "you don't have a clue what you need, so here's something you can screw around with until you get it right" Mac. Is that "Apple"?
Haswell Mac Mini, then. Something actually cheap and still far more powerful than "the guy what works the spreadsheets" needs, but suitable for a per-cubicle basis in an office.
But you want these machines under the desks of every Bob Business and Angie Accounting like the crap Dell/HP towers we see today?
It's in "providing professional power for a specific need" price-land, though.
G4 Cube; failure; too expensive. Sort of like the computer you're trying to pitch…
It's the same hardware as in any other PC. Better, in many cases. "Most users" don't need a tower. They don't need to get into their computer. They don't need to have a clue what to do with this little rectangle jobby here.
Oh, I see them doing that! By getting rid of the mouse.
You talk a lot about consumer but frankly I think that is all you are familiar with. Like it or not Apple doesn't have a midrange business solution. You can point all you want at the iMac but nobody with any sort of business sense would buy that for use outside of an office environment. Frankly I see more PC outside the office than I do inside them.
As to that Haswell Mini it might turn my head for personal usage. But putting it into a diverse business setting would be difficult at best.
If I was an artist I would draw up a mock up but sadly it would look pretty pathetic.
Quote:
Originally Posted by SolipsismX
I love this idea!
It may not need them to be seperate cards but there will be slots ports that will work for all of those. I assume the primary use is for video cards. Do you think they'd get rid of PCIe slots? If so, then why even have this machine? Having two TB ports where the old and busted FW800 ports were would allow for a nice drop in price if they don't include a discrete GPU as standard. If it is for a server room then HD4000 is sufficient for most needs.
In a nut shell I see the need for slots for the foreseeable future. FireWire should go, that I agree with. A Pro machine should have at least 4 TB ports though. I'm not sure if the technology is there yet for 4 TB ports.
Quote:
I didn't recall with more common but choose the largest for the sake of the argument as it still didn't paint a pretty picture. At 19" my shortened mockup seems more plausible.
It really doesn't matter as this isn't Apples primary focus. I just see designs that easily convert to Rack Mount as solving issues for a certain number of Apples customers. Done right it doesn't cost any extra for the mainstream user so why not.
Quote:
How big is the average shelf? How much weight can it support? These are questions that would need to be addressed if they were to build a cube and I don't think the answer will be pleasant.
Well there are such elves then there are shelves. Book shelves, if properly built, shouldn't be a problem. Think of a stack of books end to end. This is one reason why library shelving is so expensive you end up with considerable weight on those racks.
Quote:
But they haven't done that and it seems to be pretty popular since they keep refreshing it so I think we need to understand that product better if we want to understand how the Mac Pro might change.
I use to think the Mini was a joke but it is growing on me. The fact is advancements in chip technology have turned the Mini into something that is at least respectable. Unfortunately it just doesn't have the graphics chops right now for all of my needs. I'm hopeful that Apple does right by Haswell in the Mini and slots in a GT3 supporting chip
Quote:
Possible, sure, but in no way do I think it's the most probable solution to expect.
I think that's away off. They still sell the MBP, Mac mini, and iMac with HDDS so I doubt that will come for some time. Even after those have all gone SSD I think the Mac Pro will still have the HDDs for a good time after (but part of that is because they change the design so infrequently).
The new Mac Pro will likely have space for at most two magnetic drives. External disk arrays will be connected by TB for those that need them. Primary storage will be SSD hopefully sitting on the PCI bus. Getting rid of the capacity to hold a disk array internally will allow for a significant shrinkage of the chassis and significantly reduce power supply size. In fact I wouldn't be surprised to see Apple put laptop sized drives in the new Mac Pro to save even more power. It has worked out well in the server market so it certainly can be done in the workstation market.In any event I know some of my wishes for the Mac Pro seem like long shots to many in this thread but really Apple needs a workstation design that can carry forward for another 8 years or so. That means things like SSD for main storage is a must have. Support for TB is also a must have which brings into question how they will handle the GPU. Further things like 3D ram modules could dramatically change the way a machine is built allowing for size compact action once again. We probably won't get everything that I salivate over but if we get nothing then you might as well call Apple dead on the desktop.
It's often suggested that Apple can generate a lot more sales by having more options in the lineups but you have to demonstrate there is a significant market for the suggested options and the figures don't back it up for a tower in the $1000-2000 range. Apple already has the premium PC market locked down, just as they have with the tablet and smartphone markets.
He seems to describe a half cube for rack-mounting, which isn't technically a cube e.g 13x13x7. This is essentially a smaller Mac Pro on its side.
If they were going with an actual cube, I don't think it would be good going above 8". The 12" NeXT Cube was far too big:
It would have the same footprint as the Mac Pro does now but I wouldn't like to see it take up more horizontal space. If it has to sit on the floor, it'll get in the way of your feet and it certainly won't sit on the desk. I'd like to see a machine people can sit on their desk quite easily.
Originally Posted by wizard69
The important thing here is the cost of the hardware and as such the Mac Pro isn't even in the ball game.
And the iMac is invalidated because "slots", despite Thunderbolt providing that, correct?
The rest of the building isn't likely to use an iMac either because built in screens aren't acceptable for professional usage.
… Not the glossy argument again. Is it?
My, how times have changed. I once disabled Software Update on my Mini and it had an uptime of 260 days. Like TS, I wonder why they needed the weekly reboot.
I understand your point here sol, but do not through out the idea of a HDD-less Mac Pro. Apple can easily determine how the MP is being used, and if the percentage of uses with very large datasets far outweigh the number of uses with a single HDD in the MP it might be a good design. One or two SSD's for the OS and Apps and all HDD's external; a rebirth of the Xserve RAID if you will. It'll allow them to shrink the PSU and create a small Form Factor. Though personally I don't see any benefit to a redesign chassis. The current on will be 10 years old this June, so it's not without reason they haven't changed it. I do think they need to redesign the handles as they are annoyingly sharp if you need to carry one more than a foot or two.
The old school desktop design, triple pizza box, was ok during the CRT area. Flatscreens gave the tower a favorable design. The iMac is the AIO and perfect as-is. They could go all external, simply put the MP in the keyboard with build in SSD, connect up to 14 HDD's over TB, create a new 30-40" ACD with a removable videocard slot. PCIe goes into an external box for those who need it.
Well, maybe not. I certainly don't see a reason for a redesigned chassis. And rack mountable? If they want to return to the enterprice/server market, they should start with a clean slate, not redesign a desktop computer with a 1GB of GDDR5 memory videocard.
Thunderbolt isn't a slot and its limited adoption means you won't see specialized cards moving to outside the box anytime soon. In simple terms there is no effective replacement for slots at this point. Frankly it doesn't appear that either Apple nor Intel are in a rush to promote the adoption of TB. Maybe that will change 5 years down the road but even then you still have legacy hardware to support.
Quote:
Originally Posted by Tallest Skil
And the iMac is invalidated because "slots", despite Thunderbolt providing that, correct?
… Not the glossy argument again. Is it?
Where do you get these ideas? Glossy has nothing to do with screen failures. The simple fact is screens go bad and you are at a huge disadvantage if that means replacing the entire PC. Beyond that screens are often specialized for the task at hand. For example a small screen for a crampe d work area while another might have a touch screen. Not to mention the iMacs are rather big machines and as such get in the way simply because the screen is large. If you have a bench full of lab equipment finding room for a 27" screen is an issue.
Quote:
Originally Posted by Tallest Skil
… Not the glossy argument again. Is it?
It makes a much bigger difference for some things than others. Admittedly I hate the way that most of these panels deal with glare reduction, but I hate glossy even more. I have used imacs in the past, and the reflections can be extremely annoying when dealing with colors. I'm not sure how much you know about displays, but anything gamma encoded works so that 1 + 1 != 2 either visually or physically. It's used to display images, text, etc. in a way that looks good at a lower contrast ratio. The appearance of parallel light rays reflecting off glass is really additive to these values, so it really decreases the perceived separation in them. If it was a complete non-issue, Apple wouldn't have attempted to mitigate the issue. If you completely blacken the room it's quite tolerable, but having to do that is annoying.
Quote:
Originally Posted by wizard69
Thunderbolt isn't a slot and its limited adoption means you won't see specialized cards moving to outside the box anytime soon. In simple terms there is no effective replacement for slots at this point. Frankly it doesn't appear that either Apple nor Intel are in a rush to promote the adoption of TB. Maybe that will change 5 years down the road but even then you still have legacy hardware to support.
Where do you get these ideas? Glossy has nothing to do with screen failures. The simple fact is screens go bad and you are at a huge disadvantage if that means replacing the entire PC. Beyond that screens are often specialized for the task at hand. For example a small screen for a crampe d work area while another might have a touch screen. Not to mention the iMacs are rather big machines and as such get in the way simply because the screen is large. If you have a bench full of lab equipment finding room for a 27" screen is an issue.
Actually thunderbolt has at least working products in a lot of the higher end specialty hardware, basically in the $500-5000 range. You don't have a lot of simple solutions without a relatively high minimum purchase though, which makes me somewhat skeptical of long term support. I don't think it's meant to be ubiquitous. It won't support the more extreme bandwidth requirements, and its ability to support multiple devices is extremely limited based on the device type. There are many things that don't work, and it's not clearly explained anywhere. For example a lot of people buy other displays then mistakenly try to use thunderbolt cables from Apple. The Apple display will support another thunderbolt display. For slightly atypical requirements, it's really possible to run into situations where the combination of desired hardware won't work with the available ports. I wouldn't personally make big investments in thunderbolt hardware. Marvin actually pointed out a while ago that some of these lists would be incredibly expensive for freelance individuals, yet at the freelance or smaller shop level, people don't always replace all their hardware simultaneously. Due to the cost, it's sensible to avoid the potential for orphaned peripheral devices unless they're paid for very quickly. I've personally gone the route of only purchasing computing equipment that can be paid for or justified within 12-18 months to avoid the issue of frankensteined setups.
In my experience, if Apple wants to drop something, support drops off very quickly. The express 34 slot remained on the 17" macbook pros longer than the others, yet reliable drivers and support for any of those cards dropped off a long time ago. Many of them aren't supported with Lion and on. A new thunderbolt revision could go the same way.
If you haven't seen the new iMacs, please look at them some time. I had another chance to look at them this weekend. It's very astonishing how strong the AR treatment is. It's the strongest AR treatment I remember seeing. Even in a brightly lit retail environment, the reflections are minimal, sometimes I had to look hard for reflections. And I'm annoyed with the 2011 iMac's reflections, in the 2012 models, I consider it to be a non-issue.
You're going to perceive more of a reflection in the darker areas of an image but that's true of any display. The glare used to be bad but the new one has reduced it to the point that the glare isn't really a problem any more.
Obviously if you sit with your back to an open window, it's still a problem but it's not a mirror like it used to be and much easier to fix - a matte display would wash a reflection that bright over the whole display:
What are some examples of Thunderbolt solutions that are vastly more expensive than equivalent PCIe solutions?
Why won't it support more extreme bandwidths? That's in their roadmap.
A lot (I assume you mean thousands) of people are buying non-Apple Thunderbolt displays and an Apple Thunderbolt cable? Why would they be doing that, what displays are they and why exactly don't they work?
Of course not because you want to see it fail for whatever reason. You'd presumably have been happy to buy a PCI-X device though only to see it made obsolete with PCIe. That standard only lasted 8 years. You probably also don't want to invest in DDR3 RAM as they'll change the slots with DDR4. Avoid Mini-displayport too, that'll probably fail.
Yes, I'm sure a new Thunderbolt revision will obsolete all existing Thunderbolt hardware just like Firewire 800 made all Firewire 400 devices obsolete. No hang on, you could just plug one into the other with an updated cable. Maybe Apple and Intel can do that. We can only hope they know what they're doing but yeah, in the meantime don't buy any products from any manufacturer because it's not going to last more than 10 years, which is the minimum you should expect anything to last.
I will certainly take another look at the iMac when I need to replace my Mac Pro. I keep on reading posts like yours, and the possibility of switching to a cheaper solution seems all the more viable. The only 'heavy-duty stuff I do is mostly Aperture, and that doesn't require all the horsepower my Mac Pro is giving me. Still, the 4xHDD does have its advantages, and I even use the ODD. Or at least used to -- I've finished ripping all my optical discs except for the DVD's that are stacked up next to my DVD player. But those are all BR, and I simply cannot picture myself of buying a read-only BR drive to add another geez-knows-how-many-gigabytes to my current, hmm, I don't even know, but the internal storage on a MP is really great. Anyone know if there's a real storage bump looming on the horizon, like going from the current 4TB to, say, 200TB or something similarly crazy?
That's off topic, but the screen seems a likely candidate as a replacement for my 30" ACD. Thanks for your post JeffDM.
@Marvin: that is the most useful comparison picture I have ever seen, thank you very much indeed for that!
There's something on the horizon:
http://arstechnica.com/gadgets/2011/10/researchers-increase-hard-drive-density-sixfold-with-salt/
but how far out they are is never clear. I wouldn't expect them to jump very far because they'd sell fewer of them. Most likely they'll go to 6TB next year. They can probably go up 2TB every year on the 3.5" models:
http://www.extremetech.com/computing/122921-seagate-hits-1-terabit-per-square-inch-60tb-drives-on-their-way
Quote:
Originally Posted by JeffDM
If you haven't seen the new iMacs, please look at them some time. I had another chance to look at them this weekend. It's very astonishing how strong the AR treatment is. It's the strongest AR treatment I remember seeing. Even in a brightly lit retail environment, the reflections are minimal, sometimes I had to look hard for reflections. And I'm annoyed with the 2011 iMac's reflections, in the 2012 models, I consider it to be a non-issue.
Agreed. I'm using the 2010 iMac now, and it's generally workable in a pro setting. The glossy screen took some getting used to, but it's usable.
The main issue with the iMac is the lack of a user-replaceable hard drive.
Having to lose six working days for Apple's hard drive recall is ridiculous and downright inexcusable for pro users.
If the iMac is what Apple wants more pros to use, they should sell a version of the 27" without a hard drive at all, so that all confidential files can be kept safely at home/work if the machine has to be serviced. OWC could sell a version of the MiniStack made for iMac users with a RAID drive and removable Time Machine backup, and many pro users would, I think, be content with that.
Originally Posted by Marvin
I wouldn't expect them to jump very far because they'd sell fewer of them.
This is the first time I've thought of this regarding hard drives. And it makes me livid, because I wouldn't put it past them.
That's not how I'd operate, at least. If I had the ability to create a product ten times better than anything on the market, I'd do it. *shrug* That's all that needs said. I wouldn't make something twice as good to sell "more" of them, I'd make something ten times as good and watch the entire market buy my product because of the immensity of its improvement.
If the improvement is big enough, you'll start to pull in not only the people who wouldn't have bought it at a 2x improvement, but also the people who weren't planning to buy a new one at all. That's what they should be going for. That kind of one-upmanship is what we need in hard drives these days, in both camps.
Quote:
Originally Posted by Marvin
You're going to perceive more of a reflection in the darker areas of an image but that's true of any display. The glare used to be bad but the new one has reduced it to the point that the glare isn't really a problem any more.
Obviously if you sit with your back to an open window, it's still a problem but it's not a mirror like it used to be and much easier to fix - a matte display would wash a reflection that bright over the whole display:
I have never seen a side by side. Thanks for that. For anything really critical, I suggest subdued lighting anyway, as there aren't any imac display hoods. I would even recommend it with matte displays if possible if you really need consistency. I also suggest allowing it to warm up for 30 minutes or so prior to judging color from a previous session due to shifts as the display stabilizes. Some are better than others though. I was trying to explain that reflections don't really follow the same behavior as a color managed device that emits light.
Quote:
What are some examples of Thunderbolt solutions that are vastly more expensive than equivalent PCIe solutions?
You're really misinterpreting my posts today. Wizard mentioned that you won't see specialized cards migrate. I mentioned that the expensive, specialized ones were the first to release thunderbolt versions. I've mentioned some of the tools made by Black Magic before, as they were some of the first to ship thunderbolt products. My claim was that it has working products from companies that have a sizable market on the Mac and margins that can absorb the extra development cost. I haven't seen as much consumer grade stuff at similar price points to what you would pay for usb3.
Quote:
Why won't it support more extreme bandwidths? That's in their roadmap.
I'm skeptical as to what machines will get this, the state of daisy chaining if you need to run a number of peripherals from different companies with different firmware, and intel's claims compared to what is actually delivered. They projected 10 years out with some uncertainty. It's also dependent on them adding that many PCI lanes, so that thunderbolt may derive such bandwidth and completely ignores the fact that display bandwidth may grow immensely by the end of the decade given the emergence of higher dpi displays. I'm still skeptical as to when we'll have shipping desktop products of that type. It started with phones simply because of the huge growth and small screen size. Desktop displays have been relatively stagnant for years though. The biggest difference is that prices have come down.
Quote:
A lot (I assume you mean thousands) of people are buying non-Apple Thunderbolt displays and an Apple Thunderbolt cable? Why would they be doing that, what displays are they and why exactly don't they work?
Again misinterpretation! I meant people buying a non-Apple display to use in conjunction with their mini or to add a large screen to their notebook. I wouldn't make that mistake on cables, but some people seem to do so. I actually
Quote:
Of course not because you want to see it fail for whatever reason. You'd presumably have been happy to buy a PCI-X device though only to see it made obsolete with PCIe. That standard only lasted 8 years. You probably also don't want to invest in DDR3 RAM as they'll change the slots with DDR4. Avoid Mini-displayport too, that'll probably fail.
None of that is really true. It was weird how the last G5s got PCIe with no PCI-X slots so it took half a cycle to see anything for them. In that sense it was an awkward transition to PCIe. You also ignored some of my previous points that I know you have read. Ram isn't really something you pass between machines, so the argument makes little sense there. Mini displayport works the same as displayport. The only thing I've complained about is lack of full displayport 1.2 and 10 bit out support from Apple.
Quote:
Yes, I'm sure a new Thunderbolt revision will obsolete all existing Thunderbolt hardware just like Firewire 800 made all Firewire 400 devices obsolete. No hang on, you could just plug one into the other with an updated cable. Maybe Apple and Intel can do that. We can only hope they know what they're doing but yeah, in the meantime don't buy any products from any manufacturer because it's not going to last more than 10 years, which is the minimum you should expect anything to last.
You misinterpreted me on a number of these points. I was saying whenever Apple changes direction on something, support drops off quickly whether it's from Apple, third parties, or a combination. Sometimes it can be 1-2 years from discontinuation. If you look at what I mentioned regarding staggering purchases of peripherals and primary computing devices, sometimes it's possible to just buy at a bad time (like late cycle or prior to a transition) because you happened to need it then. Right now I'm waiting for some stability there prior to investing in any storage solutions based around thunderbolt. The other problem is that it's coupled to a display, yet not every peripheral allows for daisy chaining.
Well I do as that's mostly how I find things.
I wanted them to make 2.5" drives accessible from the base but they obviously realised that people would just end up buying the entry models and put Samsung or Crucial SSDs in them. It also complicates their manufacturing having accessible panels on the outside but I think it would have helped them in some ways because they'd have happier customers and faster service times. A technician could replace a broken drive in 5 minutes while someone was standing in the store instead of prizing the display off and gluing it back on again.
Accessible slots don't work so well for the Fusion drive setup though because if someone removes one drive, it would break it. Also, they obviously wanted 3.5" drives for the extra capacity and they are better being tightly secured to minimize noise. I'd be quite happy with an option where they put a blade SSD screwed in beside the RAM slots on the 27" and offered a 256GB and 512GB option on its own. There aren't really many competitively priced blade SSDs. There's a chance the SSD could get too hot that way but they can probably cool it down quite easily and it's a setup I'd prefer.
The problem there though is sustaining a large company. If you sell a product that will mean you saturate the market very quickly, the sales will start to tail off and you might have to downsize the company. There are two sides to the ethics of this because while technological progress means that consumers get better products at lower prices, it's harder for companies to sustain growth and jobs will be lost all the way from manufacuring to retail.
I think the way they do things now is the better way simply because the whole reason anybody really does anything is to try and improve the quality of life and having a sustainable economic model is more important to this than having technology that isn't really essential. There's a balance that has to be made so that we don't just have to put up with junk products as that affects growth too but it wouldn't be wise of them to give us everything all at once. I really wonder what's going to happen to the whole tech industry in another 20-30 years given how far it has come in that time. It seems very much like computers will go the way of televisions and it all just becomes about the services. No doubt that's why companies are making moves to develop their own OS as that's the only guaranteed point of leverage anyone will eventually have over profitable services.
Right now, HDDs are around $0.05/GB and SSD is $0.60/GB. I can see all manufacturers just ditching HDDs and moving to faster, more profitable SSDs. There's not much point in having 200TB drives if it can still only read/write at 150MB/s and more importantly has really low random read/write speeds. SSD just needs to get down to $0.10/GB to replace any size of HDD because you can just stick more chips in. There's not the same weight and noise issues like with mechanical platters. Imagine in 2 years, they half SSD prices to get to $0.30/GB and another 2 years, it's $0.15/GB, you'd potentially get a 2TB 2.5" SSD for $300. Even if you can get a single 20TB HDD for the same price, people will still need RAID systems for reliability and if you have files that take up that much space, you need high performance.
I doubt we'll ever see consumer HDDs that are 60TB+ in size because the market appeal for them would be so small. I can see them focusing on speed more than capacity to the point that RAM and storage are one and the same - in effect 2TB of storage becomes 2TB of RAM and video memory or at the very least is tiered storage with small amounts of very fast memory and of course all soldered to the motherboard. While it seems crazy now to solder storage to the motherboard, when computers are so inexpensive, it will be a better model to just replace the whole board. The iPad Mini is $329, which means the parts cost less than $200 and that includes the display. For any board or chip failure, they'd charge $100 for the whole part.
There are a few but you wouldn't say that there aren't as many consumer grade expressCard or PCIe cards as USB 3 products. The problem is that people class Thunderbolt as just a competitor to USB 3 and it's really missing the point of what Thunderbolt is. It's essentially a faster expressCard merged with the mini-displayport output.
There's a premium for the faster ports and active cables. Lacie's drives come with a $50 premium. G-tech is a bit more with 4TB USB 3 at $399 and with Thunderbolt, $538. Thunderbolt needs more volume, cheaper parts and faster licensing steps but it takes time for this to happen. The problem as always lies with cheap product manufacturers that side with USB 3 because it's good enough but it will never negate the need for multi-protocol ports.
It goes to 20Gbps per channel next year as they will use PCIe 3 lanes. Displayport 1.2 will be supported this year and is enough for 4k resolution.
A Thunderbolt cable wouldn't even plug into the display though so it's not even remotely a problem with Thunderbolt and I doubt it's a widespread issue. The laptops and Minis have HDMI for consumer displays.
Would you have hesitated to buy an eSATA drive? That standard is going to totally die out in favour of USB 3. At the end of the day, consumer electronics is like cars.You don't expect to have the same car for more than 10 years because you'd be able to afford a better one after that period of time. If a new storage standard comes along, you copy the data over to it and move on with your life.
You make out like Apple will get bored with Thunderbolt and just phase it out within 1-2 years. People said the same with displayport vs the much more widely adopted HDMI. They co-exist now just like TB and USB 3.