Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
I keep hearing the same tired tropes from people who are believing the Apple marketing hype on the new Mac Pro and want to make some points here that are absolutely clear.

First of all, Apple is clearly on a marketing blitz to present the new Mac Pro as some super computer. The black polished case, the "FirePro" graphics, and Phil Shiller rattled off the high end specs in a way that Jobs would not have done. Why is Apple making the case the new Mac Pro is a computing monster? Obviously because of the new, non upgradable diminutive size. They have to convince everybody this is not just as powerful as the old Mac Pro, but more so. Secondarily, the knew the specs on the new Xeons was not a big improvement from previous gen (the power comes from multiple GPUs) so there too they have to convince us since single proc is the only option.

Yet another expert weighs in. . .

So, the myths ...

Bla bla bla, yet another expert weighs in. . .
 
Being probably the only non-pro user here I sincerely hope Apple will at least have one option for a base-spec on the new MP that is cheaper than the previous entry-level MPs.

I do a lot of video work (editing, encoding etc..) and I've outgrown my iMac in every way. I've pushed the thing to its limits, gone through 2 display panels (constant heat generation caused 2 panel failures, first one thankfully was within AC); as much as I love my iMac and has served me so far I think the all-in-one concept just isn't for me anymore.

The problem is that while I can easily afford a MP I've just had a tough time justifying its price. It seems that I fall in that niche in between the mac mini and the traditional MP, and where an all-in-one is no longer a preferred option. Apple has ignored this segment, maybe rightly so as it's probably very small, but I'm selfishly hoping that with the new direction taken with the MP there will be a consumer-priced model.

Depending on which application that you run, you may find that moving to a PC is a better deal. You can get a better machine without going the expensive xeon route for less than the price of an imac.
 
If they do make a model for under $1500, I hope to discover that one can buy that lowest spec model and swap in the highest spec CPU/GPUs for massive savings. It would also be nice to make a y-splitter for the PCIe SSD connector, so one could put two larger SSDs inside that would share that 1250MB/second speed. (Not sure why people are bumping that internal flash storage up to 1.5GB/second.)
 
Given an unlimited budget to pull from the HP or Dell workstation catalog and drop two W9000s in the system I don't think it would have been any less impressive. The Mari system requirements list the W7000 , W8000, W9000 line up as certified cards ( http://www.thefoundry.co.uk/products/mari/system-requirements/#tested-graphics-cards )

Mari runs on Linux ( Unix). OS X is Unix also. Mari needs OpenGL 3.2. OS X has OpenGL 3.2. If not pressed about putting a Mac menu bar up on the screen it shouldn't take very long at all to get something running (unless there are some quirky Linux GUI bindings they incorporated into the app. ). A polished Mac specific/optimized app will take time.

The unoptimized is offset by just throwing highest end BTO option at the problem and brute forcing around it.

It would be far more impressive if pulled out an entry level Mac Pro with the targeted entry level GPU and had turned in an slightly less quick results. Then the upcoming optimizations would not matter as much. They probably do. The open question is whether the stack can deliver. That uncertainty is what many folks are really grumbling about. If it looked like what they expected they'd be more confident that it would eventually work. It doesn't.

If the Mac Pro delivers at an acceptable price to many, then they will sell more than a modest number of these. It isn't going to be the exact same group of folks, but a sizable group.

I use Mari at work on a Xeon Dell T7500 system and Windows 8, ati 7870. Mari is no where near as responsive as in that demo by Jack. Sure I could build a system with W9000's and SSD's and probably see close to similar performance, but FireGl and SSD standard out of the box in the nMP is impressive and was also pointed out by the presenters. Even the low-end nMP would likely outperform what I see currently in native windows.
 
I use Mari at work on a Xeon Dell T7500 system and Windows 8, ati 7870. Mari is no where near as responsive as in that demo by Jack. Sure I could build a system with W9000's and SSD's and probably see close to similar performance, but FireGl and SSD standard out of the box in the nMP is impressive and was also pointed out by the presenters. Even the low-end nMP would likely outperform what I see currently in native windows.

I noticed its responsiveness. I've never used Mari, but I have used Mudbox. I'm also aware of the differences. It's just the closest parallel I can think of at the moment. I wonder how your machine would perform on Linux.
 
PUTTANG NEW MAC PRO MITHS TOO REST:

MITH - IT"S A TRASHCAN
FACT - no u idiots its a personal sized cumptuer

MITH - it will run windows 98
FACT - UMMMM it"s called mac Ossex u morans

MITH - U CAN"T run a NVIDIA graphic PROCESSOR in it
FACT - ok, that's true but why do u need 2? RU DUMB? LOLL idiots

MITHS = PUTT TOO REST.

I have seen this before. It was in a high school grammar textbook in the chapter called, "Don't Let this Happen To You."
 
Wow really? "Putting to rest the myths"? Sounds more like you just made some stuff up based on assumption.

I've said this in other threads and I'll say it again, the fact that Pixar, Black Magic and The Foundry are all praising this machine means its a fantastic, high end machine. I'm not sure if you saw the Pixar demo but the speed at which they were painting channels with an un-optimized Mari was astounding.

Indeed. I hope this thread gets deleted before I see some people repeating the tripe posted by op. If I do, then I know whom to blame.
 
I think his assertion that the contrary are widely held beliefs (myths) is where this post is flawed.

Yeah, I guess there's that too. Kind of a miss.



-------------------

I think it's kinda strange everyone is flaming this guy and no is in time-out yet. I know if some of these posts were from me the mods would have timed me out already! Maybe they just like me too much. :D
 
I think it's kinda strange everyone is flaming this guy and no is in time-out yet. I know if some of these posts were from me the mods would have timed me out already! Maybe they just like me too much. :D

Haha... based on the OP's absence from this thread, maybe he got the time out! :eek:
 
I do a lot of video work (editing, encoding etc..) and I've outgrown my iMac in every way. I've pushed the thing to its limits, gone through 2 display panels (constant heat generation caused 2 panel failures, first one thankfully was within AC); as much as I love my iMac and has served me so far I think the all-in-one concept just isn't for me anymore.

There is a difference between all-in-one as a class of machines and a design failure. The panel overheating is largely a design failure, not a class wide systemic problem. The newest iMacs collect heat from the major producers and blow the hot air in the opposite direction of the panel. The old design "flowed" the heat up and out through the middle underneath the panel. 20/20 hindsight but that isn't a great idea.


It seems that I fall in that niche in between the mac mini and the traditional MP, and where an all-in-one is no longer a preferred option. Apple has ignored this segment, maybe rightly so as it's probably very small, but I'm selfishly hoping that with the new direction taken with the MP there will be a consumer-priced model.

Since the hang-up is primary price, not performance, probably not. Longer term there are 2 primary drivers that Apple uses to address that in between market.

a. Mac mini gets better. For users who have workloads growing slower than tech growth, the Mac Mini will catch them over time.


b. Used/Refurb... cheaper Mac Pros. Price is the issue. Cheaper solves it. While Apple doesn't get much money this way for hardware it suppresses the size of the gap. For those who sell to partially finance a trade-up it actually does indirectly drive sales of new.

2-3 years from now there will probably be 2013 Mac Pros going for around $1,700-1,800 ( if start at $2500 now that is a 32% discount at lowest price. ). Same is true now. There is a $1,800 Mac Pro 2010 refurb on Apple store right now in the USA.


The other issue is just to keep fixing major problems with the iMac. No high speed I/O outside the box... Thunderbolt. limited graphics performance .... entry desktop class graphics ( 680MX just a significantly underclocked desktop GPU of the current generation.). Overheating LCD panels ... fix thermal management.

If completely optimize the iMac and Mini and there is a huge gap... fill it. Until then why throw a product in there that will try very hard to wipe out?



None of those are going to completely close the gap, but they do render it relatively quite small.

Frankly, is any xMac they wanted to sell would have to fight uphill against those exact same forces.
 
do you just copy/paste this reply in all the threads you're in?
here's the question- why haven't you bought one of the 16 core dell or hp?

Because I am waiting for all the facts. Just like everyone else here, we have a lot of supposition. I don't actually have to make a decision until we see the prices.

If the cost of the nMP & externals & the cost of a Dell or HP is a wash, then the decision process becomes harder.

If the nMP keeps the same price points as the current one - then moving to a Dell or HP is a no-brainer.

Besides what else do we have to do while we are waiting.
 
I think you confuse GB and Gb. 8 bits in a byte, this "cheap" SSD does 1.25 GigaBytes per second. Cubemmal, your post shows how little you know about technology.


I keep hearing the same tired tropes from people who are believing the Apple marketing hype on the new Mac Pro and want to make some points here that are absolutely clear.

First of all, Apple is clearly on a marketing blitz to present the new Mac Pro as some super computer. The black polished case, the "FirePro" graphics, and Phil Shiller rattled off the high end specs in a way that Jobs would not have done. Why is Apple making the case the new Mac Pro is a computing monster? Obviously because of the new, non upgradable diminutive size. They have to convince everybody this is not just as powerful as the old Mac Pro, but more so. Secondarily, the knew the specs on the new Xeons was not a big improvement from previous gen (the power comes from multiple GPUs) so there too they have to convince us since single proc is the only option.

So, the myths ...

  • The machine is using FirePro cards. This is pure marketing hype, the machine isn't using any cards at all. It's using AMD consumer chips on a custom board, just as the FirePro graphics cards do. All that makes a FirePro a FirePro is the driver, and Apple has all it's own drivers. All Apple is buying from AMD is consumer chips and maybe some support silicon.
  • The machine is using expensive parts. Other than the Xeon chip the machine is using cheap parts. Cheap AMD consumer graphics chips, cheap SSD assembly which they use in Macbook Air consumer computers, fewer parts overall as they've ditched everything possible.
  • It has to be more expensive. This small cylinder is cheaper to manufacture due to a smaller BOM (Bill Of Materials), guaranteed. With a usual 10x markup the BOM has an enormous effect on final price.
  • Apple doesn't care about pricing. This one is really stupid. Apple is a consumer device company now, do you really think they'll price this in the stratosphere, especially considering what was said above? If anything they want to bring the price down so as to sell on volume, which is how they operate on every other product line. Price drops are more likely than any price increase.
  • Apple cares about the "pros". This one is a bit of a stretch, but I don't think Apple really cares about pros anymore. Again, it is pure marketing hype. From a physical standpoint I see a machine that is a step back from the old Mac Pro (less internally configurable) with a lot of marketing to convince us otherwise. Their behavior on the old Mac Pro (unbelievable waits between upgrades) and FCP demonstrates how much they care about the tiny professional market. Apple cares about consumers and this looks like a high end consumer (prosumer) computer.

These are the facts. Anything further is speculation, but as I've written elsewhere I expect a machine that is priced at or below current prices due to all these factors.
 
Depending on which application that you run, you may find that moving to a PC is a better deal. You can get a better machine without going the expensive xeon route for less than the price of an imac.

I've been considering it for a while now. And as tempting as it is, ever since switching from windows to mac 5 years ago I really do enjoy using OSX a lot more than Windows (I still use a W7 machine daily at work). But you're right, cost-wise I can get a decked out windows workstation for as much as an iMac. I'll have to wait and see how the nMP is priced and decide then.

There is a difference between all-in-one as a class of machines and a design failure. The panel overheating is largely a design failure, not a class wide systemic problem. The newest iMacs collect heat from the major producers and blow the hot air in the opposite direction of the panel. The old design "flowed" the heat up and out through the middle underneath the panel. 20/20 hindsight but that isn't a great idea.




Since the hang-up is primary price, not performance, probably not. Longer term there are 2 primary drivers that Apple uses to address that in between market.

a. Mac mini gets better. For users who have workloads growing slower than tech growth, the Mac Mini will catch them over time.


b. Used/Refurb... cheaper Mac Pros. Price is the issue. Cheaper solves it. While Apple doesn't get much money this way for hardware it suppresses the size of the gap. For those who sell to partially finance a trade-up it actually does indirectly drive sales of new.

2-3 years from now there will probably be 2013 Mac Pros going for around $1,700-1,800 ( if start at $2500 now that is a 32% discount at lowest price. ). Same is true now. There is a $1,800 Mac Pro 2010 refurb on Apple store right now in the USA.


The other issue is just to keep fixing major problems with the iMac. No high speed I/O outside the box... Thunderbolt. limited graphics performance .... entry desktop class graphics ( 680MX just a significantly underclocked desktop GPU of the current generation.). Overheating LCD panels ... fix thermal management.

If completely optimize the iMac and Mini and there is a huge gap... fill it. Until then why throw a product in there that will try very hard to wipe out?



None of those are going to completely close the gap, but they do render it relatively quite small.

Frankly, is any xMac they wanted to sell would have to fight uphill against those exact same forces.

That's interesting info, thanks. I didn't know the extent of the heat management changes made to the new iMacs. If the LCD panels in the new iMacs are truly shielded now and don't suffer the heat exposure like before then maybe I should look at them again. I still like the idea of a separate CPU and display which is why I'm really excited about the nMP, just waiting on pricing now to decide.
 
I keep hearing the same tired tropes from people who are believing the Apple marketing hype on the new Mac Pro and want to make some points here that are absolutely clear.

First of all, Apple is clearly on a marketing blitz to present the new Mac Pro as some super computer. The black polished case, the "FirePro" graphics, and Phil Shiller rattled off the high end specs in a way that Jobs would not have done. Why is Apple making the case the new Mac Pro is a computing monster? Obviously because of the new, non upgradable diminutive size. They have to convince everybody this is not just as powerful as the old Mac Pro, but more so. Secondarily, the knew the specs on the new Xeons was not a big improvement from previous gen (the power comes from multiple GPUs) so there too they have to convince us since single proc is the only option.

So, the myths ...

  • The machine is using FirePro cards. This is pure marketing hype, the machine isn't using any cards at all. It's using AMD consumer chips on a custom board, just as the FirePro graphics cards do. All that makes a FirePro a FirePro is the driver, and Apple has all it's own drivers. All Apple is buying from AMD is consumer chips and maybe some support silicon.
  • The machine is using expensive parts. Other than the Xeon chip the machine is using cheap parts. Cheap AMD consumer graphics chips, cheap SSD assembly which they use in Macbook Air consumer computers, fewer parts overall as they've ditched everything possible.
  • It has to be more expensive. This small cylinder is cheaper to manufacture due to a smaller BOM (Bill Of Materials), guaranteed. With a usual 10x markup the BOM has an enormous effect on final price.
  • Apple doesn't care about pricing. This one is really stupid. Apple is a consumer device company now, do you really think they'll price this in the stratosphere, especially considering what was said above? If anything they want to bring the price down so as to sell on volume, which is how they operate on every other product line. Price drops are more likely than any price increase.
  • Apple cares about the "pros". This one is a bit of a stretch, but I don't think Apple really cares about pros anymore. Again, it is pure marketing hype. From a physical standpoint I see a machine that is a step back from the old Mac Pro (less internally configurable) with a lot of marketing to convince us otherwise. Their behavior on the old Mac Pro (unbelievable waits between upgrades) and FCP demonstrates how much they care about the tiny professional market. Apple cares about consumers and this looks like a high end consumer (prosumer) computer.

These are the facts. Anything further is speculation, but as I've written elsewhere I expect a machine that is priced at or below current prices due to all these factors.

Thank You.

Your tone will likely cause conniptions in the Kool Aid/apologist crowd but facts are facts.

You will have a hard time convincing folks that "Professional" GPUs aren't really any different than consumer grade ones. It's a Naked Emperor that they don't want to confront.

But I will add what I can to this idea.

For some time it has been possible to read the quality of Fermi cards GPU core and spit this out as a number between 1 and 100. Most GTX480s came in around 65-85 since these had to run the fastest clocks.

Want to guess where the Quadro 4000, 5000, and 6000 boards came in? I only handled a few of each but they were always graded as inferior to the GTX480 cores. They ranged from 47-78. So, the chips were binned as needing lower clocks to run stable and...they ended up in the $2K-5K cards. I have no direct knowledge of AMDs policies but there is no reason to believe any differently.

As far as BOM...more materials = more money. Heavy old one = MORE MASS = MORE MATERIALS. Unless the new one is made of Gold and Platinum, there is no denying that there is a whole lot less "stuff" inside which translates into cheaper to build.

I think Apple is going to be able to shut a bunch of us complainers up the old fashioned way, pricing this at what it is worth. $1,500 for an entry level and who can complain?
 
Besides what else do we have to do while we are waiting.

yeah.. i know. youre right

one thing on a more serious note though.. is people using apps that may lean towards gpu acceleration abilities (as in, you see similar or somewhat similar apps doing it and benefiting from it).. you might want to contact the developers and see where there at on the whole matter.. if they say no way then of course you should be looking towards more cpu based solutions.. if they say something like - "oh, wait til you see what we have cooking up.. give us a few more months" -- then you might want hold off a while.
 
Thank You.

Your tone will likely cause conniptions in the Kool Aid/apologist crowd but facts are facts.

You will have a hard time convincing folks that "Professional" GPUs aren't really any different than consumer grade ones. It's a Naked Emperor that they don't want to confront.

But I will add what I can to this idea.

For some time it has been possible to read the quality of Fermi cards GPU core and spit this out as a number between 1 and 100. Most GTX480s came in around 65-85 since these had to run the fastest clocks.

Want to guess where the Quadro 4000, 5000, and 6000 boards came in? I only handled a few of each but they were always graded as inferior to the GTX480 cores. They ranged from 47-78. So, the chips were binned as needing lower clocks to run stable and...they ended up in the $2K-5K cards. I have no direct knowledge of AMDs policies but there is no reason to believe any differently.

so what you're saying is that the more expensive cards are higher quality and run faster?
or am i missing something?
 
yeah.. i know. youre right

one thing on a more serious note though.. is people using apps that may lean towards gpu acceleration abilities (as in, you see similar or somewhat similar apps doing it and benefiting from it).. you might want to contact the developers and see where there at on the whole matter.. if they say no way then of course you should be looking towards more cpu based solutions.. if they say something like - "oh, wait til you see what we have cooking up.. give us a few more months" -- then you might want hold off a while.

Oh, I agree, I am keeping a close eye on development for all of my programs. The issue right now is the fact that most of the customer base for the programs I use are not clamoring for GPU acceleration. There are other things that they want, and are still wrapping their heads around the new features of the current programs.
 
Oh, I agree, I am keeping a close eye on development for all of my programs. The issue right now is the fact that most of the customer base for the programs I use are not clamoring for GPU acceleration. There are other things that they want, and are still wrapping their heads around the new features of the current programs.

it's almost as if it's a "not if, but when" type of situation. (i say almost because i'm basing this off of one rendering app so i don't want to sound as if i'm some sort of openCL expert or anything like that)

with a 5770, i'm seeing a 30%+ speed increase when enabling gpu acceleration.. that alone is more than i'd see in 16cores vs 12.. i make the guess that with 2 mid gpus in the new mac, i'll see the same type of performance gains i'd get out of having a 32core machine vs 12...

if that turns out to be true, or even close to being true, the developers are going to have to bring it in (or some other methods for performance other than a 'throw more cpus at the problem' mindset)..or they're going to get out competed..
 
I think you confuse GB and Gb. 8 bits in a byte, this "cheap" SSD does 1.25 GigaBytes per second. Cubemmal, your post shows how little you know about technology.

This one is the best response, somehow he's figured me out! However it seems you haven't read the thread, including the retraction on the SSD part.

IRL I'm a software architect with 25 years experience and you're running my software every time you take a plane flight or use your cell phone. Feel safe now :cool:
 
This one is the best response, somehow he's figured me out! However it seems you haven't read the thread, including the retraction on the SSD part.

IRL I'm a software architect with 25 years experience and you're running my software every time you take a plane flight or use your cell phone. Feel safe now :cool:

Like your list of myths, I suspect any software you wrote has been thoroughly debugged by others. :)
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.