Edit: Decided to add something to this. There are many valid uses that start to become strained easier than you'd think on lighter hardware, and thunderbolt as it is today can't really fill those voids no matter how much you kick and scream about it. You can hook up a couple displays, storage, backup storage (or keep primary storage internal if speed is less of an issue), faster gpus (not talking about gaming benchmarks either), and whatever else. One of my problems with thunderbolt right now is the desire to push everything onto a single port per computer then say to daisy chain when it hasn't ben proven effective in such setups. The clustering idea doesn't actually work. I've noted on macrumors a few people use imacs for after effects work. The common complaint is that they get blazingly hot. Being within a few degrees of throttling or shutdown isn't really a good idea if it's going to run for hours. The problem is that lately that's what you get with Apple. They don't really design it with these uses in mind. You could play games over a macbook pro, but those fans will max out the entire time. Essentially they're not designed to be run at their full potential for longer periods of time. This is mostly hypothetical, but Apple's solutions break whenever your usage patterns don't line up with their design goals.
By the way, you can't actually use anything beyond 96GB of ram in OSX. It's an OS limitation.
One more link...
http://www.chaosgroup.com/en/2/envyspot.html?i=15
Somewhere down the page they reference a 16 core workstation. They've also mentioned rendering smaller projects on a single workstation in a couple of their videos.
Quote:
In the short term, this matters. In the long term, it doesn't. I go back to this often but 10 years ago, I would have said the same thing to you. Today I have a phone that's more powerful than the machine I would have said couldn't be replaced. So now, I'm saying in less than 10 years time, none of this will matter.
That doesn't mean the Mac Pro can be discontinued tomorrow without casualties. What it means is that it's going away very soon and Apple could do it tomorrow with little to no meaningful consequence.
The group of Mac Pro users is not exclusively professional. There are professional digital painters, professional graphic designers, professional developers, professional barristers, professional architects, professional doctors and so on who rely on computers just as much as the high-resource users.
The high resource users tend to need more than a Mac Pro. 3D animation needs a render farm, same with certain scientific fields. The Mac Pro holds the middle ground where it's not enough to render movie quality visual effects on its own but offers better performance than consumer machines albeit at great expense.
These same professionals have gotten by with far worse over the years.
We've all seen or read about it. Mini computers will never replace mainframes. Laptops will never replace desktops. Blah blah. Okay you're missing something here about why people dislike seeing the old thing going away. They're often left with another moving target that's designed for a different market segment. You can make it work, but it's not always ideal. The imacs and macbook pros can run close to throttling temperatures when functioning normally when they're fully utilized. They aren't within spec for use exceeding 8 hours a day in such a manner. Basically they're not built for hours of heavy tasks from a design standpoint, although they might tolerate it (I haven't tried). Imacs are much more prone to downtime. You don't have any kind of LUT based system for display matching or hardware level calibration. It uses a glass panel yet there are no treated glass options. The point is if you designed something as a successor aimed at a similar market, it wouldn't be this. You can argue in favor of abandoning them, but not wanting to take on a leveraged solution isn't really the same thing as holding back progress. Beyond that, who cares if you can pick up the computer and take it with you if you're tied to a lot of other larger stationary hardware already. Further editing 4k on one beyond slip edits doesn't sound like fun, although that's not really my area. Creative Cow had a discussion on this a while back, but I can't find it right now.
Now I've noticed some animators that are independent of big shops do render on single workstations. It's probably also true with motion graphics guys. As to having used far worse in the past, that's pretty meaningless as computing has always been a moving target.
http://www.linuxjournal.com/article/9653
Note the increase in render time hours. Computer hardware keeps getting faster, yet it doesn't always outpace requirements. Better hardware means less animation would need to be done via proxy models with lower poly counts, meaning you'd have better feedback. I don't really care what the computer looks like. I care about design priorities. I don't care if they make it 10 mm thinner if this means I lose half the ports. When you're looking at leveraged solutions, you need to remember that the point of focus of these other solutions is also changing. If they were running in place on their target market, it would probably make the transition easier, but overall no one should buy a computer that doesn't suit their needs.
All I hear from Mac Pro users is 'don't take away my PCI slots, don't take away my desktop GPU, don't take away my giant metal box, I must be able to use my hypothetical 100GB/s PCI cards and my hypothetical SLI capability and my affordable PCI SSD drives and the 128GB of RAM that I totally bought and almost max out'. And yet at the same time saying it's ok to stick an i7 in there as long as it's cheaper.
I still don't think you get this. While those users exist, the point of a machine is to solve problems. Personally I don't care what it looks like. The form factor difference between mini and mac pro means very little to me. The imac has terrible ergonomics and too many flaws to list. If I ever considered one, it would be about price rather than function, which doesn't say much for the machine. It might also be influenced by some of the terrible scaling at times under OSX. I don't know if I ever mentioned that OpenGL has been terrible since Snow Leopard. Anyway getting off track here, but the reason people don't want to see it go most likely has more to do with the lack of a near functionally seamless replacement. Also please don't reference the base eyesore mac pro again. It's a cheap cop out configuration with an inflated price tag.
- ah but can the Macbook Pro import a 4k film off a Kona capture card while outputting 128 channels of audio over a DSP card from a PCI SSD and running Metro 2033 in the background connected up to 3 monitors. Didn't think so.
The scenarios are going to get more adventurous as time goes on and all it does is create noise. All people are talking about is raw spec and there's a disconnect between that and practical applications.
Would Apple gain from discontinuing the Pro? No, the same volume of users migrate to lower margin hardware or alternatives.
Would Apple gain from a minor update to the Pro? No, desktop sales have no growth and that update isn't compelling.
Would Apple gain from redesigning the Pro? In a small way, yes as they could increase their market volume in higher margin hardware by offering good value.
Whatever they do, the Mac Pro contributes something like 2% of their profits at most. Sure, it's not about the money but what is it about? Their volume isn't high so they aren't making an impact on any community.
I have yet to see it run something like that well along with an appropriate display. You are pretty lacking on ports with a macbook pro if you ignore the stupid television commercials and actually set out to build a setup around a macbook pro. This was my point. They didn't design it for that, so you're going to run into more gotchas that should be addressed on Apple's end. I still think you just like new shiny gadgets. I don't spite you for that. It's just that the science fiction comment supports it.
Edited by hmm - 6/1/12 at 11:02pm
