Most users don't need a high end NEC or Eizo instead of the perfectly good IPS panel in the iMac either.
I think you mis my point completely, not everybody is an AV artist, some people have 3D printers on their desks, or a desk full of instrumentation, or pictures of the wife, maybe a PCB router, a disk array or any number of other things. Some of us run into real problems with the size of the iMac.
Most users don't need a high end NEC or Eizo instead of the perfectly good IPS panel in the iMac either.
The reason I mention NEC at times is that you can find their best 27" for roughly the same price as a thunderbolt display through some of the online retailers, at least in the US.
I think you mis my point completely, not everybody is an AV artist, some people have 3D printers on their desks, or a desk full of instrumentation, or pictures of the wife, maybe a PCB router, a disk array or any number of other things. Some of us run into real problems with the size of the iMac.
They have problems with even the 21" iMac? If you're that space constrained you can't fit a Mac Pro either.
Your size argument is bullshit since most (meaning 99.99%) will need some kind of monitor and it doesn't get much smaller than 21" these days.
You just dislike AIOs which is fine. What is not fine is BS about how the iMac sucks and can't do the same job as a Mac Pro or your mythical xMac.
This confirms the older scores. The multi-threaded render benchmark was 38% faster than the E5-2687W. The latter scores 25.3 in Cinebench with dual CPUs so single is about 12.65 x 1.38 = 17.5 vs 16 in the current 12-core (9% faster). The 64-bit Geekbench score was noted as over 30,000 and the 64-bit of the 2012 12-core MP is ~25800 (16% faster).
The CPUs in the old one cost $1440x2 = $2880 x 1.3 markup = $3744
The single CPU likely won't cost over $2000 x1.3 = $2600
This price difference means that there will be an opportunity for Apple to sell a machine faster than the top 2012 12-core over $1000 cheaper.
The performance benchmarks will disappoint spec whores just like the RAM limit and the bandwidth on the TB ports but it ensures higher growth rates. A dual 12-core would cost over $7k and a lot of people don't spend that much regularly. By improving performance-per-dollar, it makes it more accessible to a higher volume of buyers while leaving some in need of upgrades.
Some people might find a dual E5-2697v2 workstation tower for over $7k more appealing than a single CPU Mac Pro for $5k and migrate away but I think a higher volume of people will prefer the lower prices on the 12-core CPUs than people who want the highest possible dual-CPUs.
Intel's improvements have been a bit disappointing. Although they doubled the core count in 3 years, they dropped the clock speed and they delayed the CPU architecture so now the MP is on Ivy Bridge rather than Haswell. If that wasn't the case, the new MP would have a much more significant upgrade.
Still, when you think about the roadmap ahead, consider that Intel might be able to double core count again in 2-3 years so that means a 24-core/48-thread MP. Obviously a dual-CPU provides that now but how many people does a 48 CPU core machine really appeal to? The Mac Pro is a workstation and is designed for real-time work. The dual GPUs will crunch through OpenCL processes much better than the Xeon does and workflows involving Da Vinci and Adobe CS will benefit hugely from that. For people who absolutely need the CPUs for bulk computing, they can buy more than one machine. CPU compute tasks typically don't require real-time feedback.
I like the direction they've taken with it because although they are taking away some control from the user, they are making great decisions like forcing every user to have a PCIe SSD. Not every Mac Pro owner is going to be technically minded enough to know how big a difference that will make. Even people who put in their own SSDs have been stuck on 3Gbps. Dual GPUs are regarded as redundant additions to some people but they'll manage to drive 4K/Retina displays better than typical workstations where a second GPU might not even be a consideration.
Marvin, what you wrote is interesting, I guess, but IMO what the majority of users are most interested in will be the configuration of the base model and its price. Based on past history Apple should offer more than just a high end model. So I guess in order to find out we just wait and wait and wait... .
Marvin, what you wrote is interesting, I guess, but IMO what the majority of users are most interested in will be the configuration of the base model and its price. Based on past history Apple should offer more than just a high end model. So I guess in order to find out we just wait and wait and wait... .
I have to concur though I'm not sure Apple realizes this yet! The market for uber workstations just isn't that huge these days. Actually it never has been huge.
My hope is that Apple took the approach they did, with the new Mac Pro, to in part control cost to actually be able to market the machine at a far more reasonable price. If they haven't recognized the Mac Pros previous failure in the market place was partly due to an abusive price structure they will have learned nothing. In simplest terms the old Mac Pro wasn't a good value in terms of the base machine for several years now so hopefully the New Mac Pro corrects this.
Marvin, what you wrote is interesting, I guess, but IMO what the majority of users are most interested in will be the configuration of the base model and its price. Based on past history Apple should offer more than just a high end model. So I guess in order to find out we just wait and wait and wait...
The current low-end uses a $294 processor. There is an Ivy Bridge quad at this price:
There are entry FirePro GPUs too. SSDs can scale from 256GB. They can hit anywhere in the region $1999-2499.
I don't know where people are getting the idea they'll only have a high-end model. They have no reason to jump from a quad-core iMac to a 12-core MP and leave nothing in between and they've never done this.
It's only 3 weeks until the next event. If that's just for iOS, then it'll be another month but TB2 should be introduced in 3 weeks as well as all the CPUs.
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
How did they get a product to test when there is no final released product? Remember, the only MacPros that exist are in Apple's research labs. They still have to get Intel to ship the processors and chip sets so they can begin production.
Even 10.9 which is required hasn't been sent to Golden Master yet, so any tests that you see with regards to these new systems is 100% complete BULLSQUAT.
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
Oh, Cinebench only tests Open GL, but doesn't seem to test Open CL. so, i would wait until the actual product is released and then look at all of the different tests and then see how well it runs a specific application that you want to run vs another OS system running a similarly priced box that supposed to be as close as possible.
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
You might also check AFTER 10.9 gets released because I'm sure Cinebench will probably have to go through an update process to make it run for 10.9. The current version of Cinebench is for 10.8, 10.7, but makes no mention of 10.9, which hasn't been Golden Master, yet. Patience grasshopper.
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
They did the same thing with the 2012 Mac Pro where they compared a 12-core to a previous 8-core. Price has to factor into it because if they sell a new architecture 12-core at the same price as a previous architecture 8-core then they'd be right to say you get up to double the performance at the same price point.
What they say on the site is up to double the floating point performance. If you look at a Geekbench score of the 2.4GHz 2012 12-core (this one happens to be running Windows so the scores will differ a little - the OS X score is listed as 17431 but the score sheet isn't showing up):
I don't expect they'll be selling the top 12-core CPU at $3799 so I think they'd struggle to justify 2x faster but they would have last year too with the 12-core to 8-core comparison they used as I don't think they matched the price points either. These numbers are used for marketing. If buyers don't like the performance-per-dollar of the new Mac Pro then they don't have to buy it but it will use the same CPUs as anybody else.
How did they get a product to test when there is no final released product? Remember, the only MacPros that exist are in Apple's research labs. They still have to get Intel to ship the processors and chip sets so they can begin production.
Even 10.9 which is required hasn't been sent to Golden Master yet, so any tests that you see with regards to these new systems is 100% complete BULLSQUAT.
They had working models on stage at WWDC used for a software demo. They were running a beta version of Mavericks. They don't just get the CPUs when Intel releases them as they have to build machines to ship when the processors launch. The CPU launch is when Intel goes into mass production.
Within a week of setting up my iMac and TBD, that height difference began to really annoy me, so I ordered a SlimKey Stand for the display that puts the TBD approximately .100" or so above the top of the iMac's display - something only I would notice. An added benefit was that I also picked up 4 USB ports and some extra storage space for the Magic Trackpad and the remote (which I seldom use).
I'm now searching for some sort of connection bar or spacer to prevent the displays from accidentally bumping each other - and yeah, it would be more aesthetically pleasing. I haven't found anything yet, so I may have to go the DIY route.
Sorry for going off topic, but this did catch my eye.
They did the same thing with the 2012 Mac Pro where they compared a 12-core to a previous 8-core. Price has to factor into it because if they sell a new architecture 12-core at the same price as a previous architecture 8-core then they'd be right to say you get up to double the performance at the same price point.
What they say on the site is up to double the floating point performance. If you look at a Geekbench score of the 2.4GHz 2012 12-core (this one happens to be running Windows so the scores will differ a little - the OS X score is listed as 17431 but the score sheet isn't showing up):
I don't expect they'll be selling the top 12-core CPU at $3799 so I think they'd struggle to justify 2x faster but they would have last year too with the 12-core to 8-core comparison they used as I don't think they matched the price points either. These numbers are used for marketing. If buyers don't like the performance-per-dollar of the new Mac Pro then they don't have to buy it but it will use the same CPUs as anybody else.
They had working models on stage at WWDC used for a software demo. They were running a beta version of Mavericks. They don't just get the CPUs when Intel releases them as they have to build machines to ship when the processors launch. The CPU launch is when Intel goes into mass production.
Cinebench tests CPU rendering as well as OpenGL.
Running Geekbench tests on Beta software thinking that that is reliable? Really?
I know cinebench goes CPU, I thought that would be a given, but it doesn't test for Open CL.
My suggestion, don't be shackled to these figures.
FIrst, wait until the FINAL version of Mavericks comes out, the REAL MacPros are released, and then make sure Cinebench and Geekbench have OFFICIALLY said that the version you are running has been updated for Mavericks and THEN test. Many times, these tests need to be be updated when a major OS release comes out. But you should know that by now.
I would also see what happens with certain apps after they've been updated as well. Some will be improving their speeds so, I would wait until updates on apps before running speed tests.
This is only of my pet peeves, discussing and pre-release information on speed tests when NOTHING is officially released and the testing software isn't updated.
It can be VERY misleading if they aren't actually accurate.
As far as pricing.
I don't know which processor(s) they will offer for the entry level, etc. I did read that it's supposed to go UP to 6GB of GPU memory. Is that per card? So are there choices to get the product with less memory per GPU? I know they are supposed to SHIP standard with two GPU, but the text for that section on Apple's site suggests that the amount of GPU memory maybe another option.
I don't know what the pricing is for the CPUs/GPUs they are using, since they aren't specific.
Running Geekbench tests on Beta software thinking that that is reliable? Really?
Beta hardware most likely too! Often Intel will go through a series of engineering samples as various issues are corrected or the build process tweaked. Beyond the fact that GeekBench isn't all that great of a way to tet a system, you need a stable system free of debug code and other slow ups that don't ship with released code.
I know cinebench goes CPU, I thought that would be a given, but it doesn't test for Open CL.
In the end the only real benchmarks are those that test the software that is your primary money maker. How well the Mac Pro will perform will vary widely with the type of App being used and its revision level.
My suggestion, don't be shackled to these figures.
Exactly! One should concern themselves with apps and work loads they actually run.
FIrst, wait until the FINAL version of Mavericks comes out, the REAL MacPros are released, and then make sure Cinebench and Geekbench have OFFICIALLY said that the version you are running has been updated for Mavericks and THEN test. Many times, these tests need to be be updated when a major OS release comes out. But you should know that by now.
Beyond that new instructions can often benefit many pro apps. Without a recompile targeting the new hardware you won't see the complete picture.
I would also see what happens with certain apps after they've been updated as well. Some will be improving their speeds so, I would wait until updates on apps before running speed tests.
This is only of my pet peeves, discussing and pre-release information on speed tests when NOTHING is officially released and the testing software isn't updated.
It can be VERY misleading if they aren't actually accurate.
You are completely justified with respect to your peeves.
As far as pricing.
I don't know which processor(s) they will offer for the entry level, etc. I did read that it's supposed to go UP to 6GB of GPU memory. Is that per card? So are there choices to get the product with less memory per GPU? I know they are supposed to SHIP standard with two GPU, but the text for that section on Apple's site suggests that the amount of GPU memory maybe another option.
I don't know what the pricing is for the CPUs/GPUs they are using, since they aren't specific.
I'm really hoping Apple is taking the same tact that they have with the AIRs and other notebooks. That is to use advanced technology to gain an advantage in the marketplace that allows for fairly agressive pricing. In the end if the new Mac Pro has not addressed base price issues of the old MacPro it will tank in the market place.
Beta hardware most likely too! Often Intel will go through a series of engineering samples as various issues are corrected or the build process tweaked. Beyond the fact that GeekBench isn't all that great of a way to tet a system, you need a stable system free of debug code and other slow ups that don't ship with released code.
In the end the only real benchmarks are those that test the software that is your primary money maker. How well the Mac Pro will perform will vary widely with the type of App being used and its revision level.
Exactly! One should concern themselves with apps and work loads they actually run.
Beyond that new instructions can often benefit many pro apps. Without a recompile targeting the new hardware you won't see the complete picture.
You are completely justified with respect to your peeves.
I'm really hoping Apple is taking the same tact that they have with the AIRs and other notebooks. That is to use advanced technology to gain an advantage in the marketplace that allows for fairly agressive pricing. In the end if the new Mac Pro has not addressed base price issues of the old MacPro it will tank in the market place.
Here's what we DON"T know. All of the final specs and pricing. I can wait another month or two. I think they will release the final numbers/specs in Oct. (Just a hunch).
but it's not going to be half the speed of the production version. The sample has to be close to the production model in clock speed, features and power consumption otherwise there's little point in having the sample.
There may be performance improvements with final software and production hardware but the important detail as far as it applies to the objection to Apple marketing up to 2x the CPU performance is that the new top-end 12-core will not be double the performance of the old top-end 12-core nor close to it. That would require Intel's performance-per-watt to increase 4x in 2 architecture steps, which doesn't happen. There should have been 3 steps since the old Mac Pro but there weren't. It's fair for people to object to their marketing but marketing material is usually not that accurate and they typically detail the comparison they use in their marketing pages. They did the same thing last year:
"in other promotional copy, Apple did reveal that the fastest custom configuration "Mid-2012" Mac Pro -- the Mac Pro "Twelve Core" 3.06 (2012/Westmere) -- is between 1.2 and 1.5 times faster than the fastest custom configuration "Early 2009" Mac Pro -- the Mac Pro "Eight Core" 2.93 (2009/Nehalem).
First, it is worth noting that this official comparison is a synthetic performance test using the "STREAM" 5.8 benchmark, and is a comparison with the much earlier "Early 2009" line rather than the previous, and effectively identical, "Mid-2010" Mac Pro line."
Some people will be disappointed they can't buy two 12-core processors but a Mac Pro like that would cost over $7500 so it affects very few people and Apple has never used the highest end processors in the past. What Apple would have used if they'd gone with two CPUs is a dual 8-core and it would have offered up to 50% more CPU performance for about $1300 more.
That extra performance option isn't essential because real-time feedback isn't required from CPU tasks. People who need more CPU performance can buy more machines e.g get a 6-core slave Mac Pro in addition to the 12-core. Less convenient in some cases but media software is using OpenCL more so the GPUs will provide good value there.
Here's what we DON"T know. All of the final specs and pricing. I can wait another month or two. I think they will release the final numbers/specs in Oct. (Just a hunch).
The final specs will be interesting. Apple provides plenty of wiggle room with the use of the "up to" phrase constantly. I'm just hoping that the design affords them a wider more competitive pricing range.
but it's not going to be half the speed of the production version. The sample has to be close to the production model in clock speed, features and power consumption otherwise there's little point in having the sample.
There may be performance improvements with final software and production hardware but the important detail as far as it applies to the objection to Apple marketing up to 2x the CPU performance is that the new top-end 12-core will not be double the performance of the old top-end 12-core nor close to it. That would require Intel's performance-per-watt to increase 4x in 2 architecture steps, which doesn't happen. There should have been 3 steps since the old Mac Pro but there weren't. It's fair for people to object to their marketing but marketing material is usually not that accurate and they typically detail the comparison they use in their marketing pages. They did the same thing last year:
"in other promotional copy, Apple did reveal that the fastest custom configuration "Mid-2012" Mac Pro -- the Mac Pro "Twelve Core" 3.06 (2012/Westmere) -- is between 1.2 and 1.5 times faster than the fastest custom configuration "Early 2009" Mac Pro -- the Mac Pro "Eight Core" 2.93 (2009/Nehalem).
First, it is worth noting that this official comparison is a synthetic performance test using the "STREAM" 5.8 benchmark, and is a comparison with the much earlier "Early 2009" line rather than the previous, and effectively identical, "Mid-2010" Mac Pro line."
Some people will be disappointed they can't buy two 12-core processors but a Mac Pro like that would cost over $7500 so it affects very few people and Apple has never used the highest end processors in the past. What Apple would have used if they'd gone with two CPUs is a dual 8-core and it would have offered up to 50% more CPU performance for about $1300 more.
That extra performance option isn't essential because real-time feedback isn't required from CPU tasks. People who need more CPU performance can buy more machines e.g get a 6-core slave Mac Pro in addition to the 12-core. Less convenient in some cases but media software is using OpenCL more so the GPUs will provide good value there.
I went to Intel's site and it doesn't list a 12 core E5 series processor. here's the link that I went to. If anyone knows where they have information on a 12 Core E5 series processor, let me know.
Comments
I think you mis my point completely, not everybody is an AV artist, some people have 3D printers on their desks, or a desk full of instrumentation, or pictures of the wife, maybe a PCB router, a disk array or any number of other things. Some of us run into real problems with the size of the iMac.
Quote:
Originally Posted by nht
Most users don't need a high end NEC or Eizo instead of the perfectly good IPS panel in the iMac either.
The reason I mention NEC at times is that you can find their best 27" for roughly the same price as a thunderbolt display through some of the online retailers, at least in the US.
They have problems with even the 21" iMac? If you're that space constrained you can't fit a Mac Pro either.
Your size argument is bullshit since most (meaning 99.99%) will need some kind of monitor and it doesn't get much smaller than 21" these days.
You just dislike AIOs which is fine. What is not fine is BS about how the iMac sucks and can't do the same job as a Mac Pro or your mythical xMac.
From MacRumors - Upcoming 12-Core Xeon CPU Destined for Apple's New Mac Pro Posts Impressive Benchmark Scores
From Tom's Hardware - Intel's 12-Core Xeon With 30MB of L3: The New Mac Pro's CPU?
Read & Discuss…!
This confirms the older scores. The multi-threaded render benchmark was 38% faster than the E5-2687W. The latter scores 25.3 in Cinebench with dual CPUs so single is about 12.65 x 1.38 = 17.5 vs 16 in the current 12-core (9% faster). The 64-bit Geekbench score was noted as over 30,000 and the 64-bit of the 2012 12-core MP is ~25800 (16% faster).
The CPUs in the old one cost $1440x2 = $2880 x 1.3 markup = $3744
The single CPU likely won't cost over $2000 x1.3 = $2600
This price difference means that there will be an opportunity for Apple to sell a machine faster than the top 2012 12-core over $1000 cheaper.
The performance benchmarks will disappoint spec whores just like the RAM limit and the bandwidth on the TB ports but it ensures higher growth rates. A dual 12-core would cost over $7k and a lot of people don't spend that much regularly. By improving performance-per-dollar, it makes it more accessible to a higher volume of buyers while leaving some in need of upgrades.
Some people might find a dual E5-2697v2 workstation tower for over $7k more appealing than a single CPU Mac Pro for $5k and migrate away but I think a higher volume of people will prefer the lower prices on the 12-core CPUs than people who want the highest possible dual-CPUs.
Intel's improvements have been a bit disappointing. Although they doubled the core count in 3 years, they dropped the clock speed and they delayed the CPU architecture so now the MP is on Ivy Bridge rather than Haswell. If that wasn't the case, the new MP would have a much more significant upgrade.
Still, when you think about the roadmap ahead, consider that Intel might be able to double core count again in 2-3 years so that means a 24-core/48-thread MP. Obviously a dual-CPU provides that now but how many people does a 48 CPU core machine really appeal to? The Mac Pro is a workstation and is designed for real-time work. The dual GPUs will crunch through OpenCL processes much better than the Xeon does and workflows involving Da Vinci and Adobe CS will benefit hugely from that. For people who absolutely need the CPUs for bulk computing, they can buy more than one machine. CPU compute tasks typically don't require real-time feedback.
I like the direction they've taken with it because although they are taking away some control from the user, they are making great decisions like forcing every user to have a PCIe SSD. Not every Mac Pro owner is going to be technically minded enough to know how big a difference that will make. Even people who put in their own SSDs have been stuck on 3Gbps. Dual GPUs are regarded as redundant additions to some people but they'll manage to drive 4K/Retina displays better than typical workstations where a second GPU might not even be a consideration.
I have to concur though I'm not sure Apple realizes this yet! The market for uber workstations just isn't that huge these days. Actually it never has been huge.
My hope is that Apple took the approach they did, with the new Mac Pro, to in part control cost to actually be able to market the machine at a far more reasonable price. If they haven't recognized the Mac Pros previous failure in the market place was partly due to an abusive price structure they will have learned nothing. In simplest terms the old Mac Pro wasn't a good value in terms of the base machine for several years now so hopefully the New Mac Pro corrects this.
The current low-end uses a $294 processor. There is an Ivy Bridge quad at this price:
http://www.techpowerup.com/cpudb/1672/xeon-e5-1620-v2.html
There are entry FirePro GPUs too. SSDs can scale from 256GB. They can hit anywhere in the region $1999-2499.
I don't know where people are getting the idea they'll only have a high-end model. They have no reason to jump from a quad-core iMac to a 12-core MP and leave nothing in between and they've never done this.
It's only 3 weeks until the next event. If that's just for iOS, then it'll be another month but TB2 should be introduced in 3 weeks as well as all the CPUs.
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
Quote:
Originally Posted by Mike Fix
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
How did they get a product to test when there is no final released product? Remember, the only MacPros that exist are in Apple's research labs. They still have to get Intel to ship the processors and chip sets so they can begin production.
Even 10.9 which is required hasn't been sent to Golden Master yet, so any tests that you see with regards to these new systems is 100% complete BULLSQUAT.
Quote:
Originally Posted by Mike Fix
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
Oh, Cinebench only tests Open GL, but doesn't seem to test Open CL. so, i would wait until the actual product is released and then look at all of the different tests and then see how well it runs a specific application that you want to run vs another OS system running a similarly priced box that supposed to be as close as possible.
Quote:
Originally Posted by Mike Fix
The current Mac Pro 2.93 gets a 15.21 on cinebench... The new mac pro, that we've been waiting years for...scores 17.63. Apple should remove their 2x processing power from their product page for the new mac pro.
You might also check AFTER 10.9 gets released because I'm sure Cinebench will probably have to go through an update process to make it run for 10.9. The current version of Cinebench is for 10.8, 10.7, but makes no mention of 10.9, which hasn't been Golden Master, yet. Patience grasshopper.
They did the same thing with the 2012 Mac Pro where they compared a 12-core to a previous 8-core. Price has to factor into it because if they sell a new architecture 12-core at the same price as a previous architecture 8-core then they'd be right to say you get up to double the performance at the same price point.
What they say on the site is up to double the floating point performance. If you look at a Geekbench score of the 2.4GHz 2012 12-core (this one happens to be running Windows so the scores will differ a little - the OS X score is listed as 17431 but the score sheet isn't showing up):
http://browser.primatelabs.com/geekbench2/2046462
you can see the floating point score is 20892. The leaked 12-core Mac Pro score is here:
http://browser.primatelabs.com/geekbench2/2064275
That shows a floating point score of 40753.
I don't expect they'll be selling the top 12-core CPU at $3799 so I think they'd struggle to justify 2x faster but they would have last year too with the 12-core to 8-core comparison they used as I don't think they matched the price points either. These numbers are used for marketing. If buyers don't like the performance-per-dollar of the new Mac Pro then they don't have to buy it but it will use the same CPUs as anybody else.
They had working models on stage at WWDC used for a software demo. They were running a beta version of Mavericks. They don't just get the CPUs when Intel releases them as they have to build machines to ship when the processors launch. The CPU launch is when Intel goes into mass production.
Cinebench tests CPU rendering as well as OpenGL.
Quote:
Originally Posted by PhilBoogie
The iMac next to a TB Display looks stupid:
http://wpmu.org/thunderbolt-display-imac-height/
Within a week of setting up my iMac and TBD, that height difference began to really annoy me, so I ordered a SlimKey Stand for the display that puts the TBD approximately .100" or so above the top of the iMac's display - something only I would notice. An added benefit was that I also picked up 4 USB ports and some extra storage space for the Magic Trackpad and the remote (which I seldom use).
I'm now searching for some sort of connection bar or spacer to prevent the displays from accidentally bumping each other - and yeah, it would be more aesthetically pleasing. I haven't found anything yet, so I may have to go the DIY route.
Sorry for going off topic, but this did catch my eye.
Quote:
Originally Posted by Marvin
They did the same thing with the 2012 Mac Pro where they compared a 12-core to a previous 8-core. Price has to factor into it because if they sell a new architecture 12-core at the same price as a previous architecture 8-core then they'd be right to say you get up to double the performance at the same price point.
What they say on the site is up to double the floating point performance. If you look at a Geekbench score of the 2.4GHz 2012 12-core (this one happens to be running Windows so the scores will differ a little - the OS X score is listed as 17431 but the score sheet isn't showing up):
http://browser.primatelabs.com/geekbench2/2046462
you can see the floating point score is 20892. The leaked 12-core Mac Pro score is here:
http://browser.primatelabs.com/geekbench2/2064275
That shows a floating point score of 40753.
I don't expect they'll be selling the top 12-core CPU at $3799 so I think they'd struggle to justify 2x faster but they would have last year too with the 12-core to 8-core comparison they used as I don't think they matched the price points either. These numbers are used for marketing. If buyers don't like the performance-per-dollar of the new Mac Pro then they don't have to buy it but it will use the same CPUs as anybody else.
They had working models on stage at WWDC used for a software demo. They were running a beta version of Mavericks. They don't just get the CPUs when Intel releases them as they have to build machines to ship when the processors launch. The CPU launch is when Intel goes into mass production.
Cinebench tests CPU rendering as well as OpenGL.
Running Geekbench tests on Beta software thinking that that is reliable? Really?
I know cinebench goes CPU, I thought that would be a given, but it doesn't test for Open CL.
My suggestion, don't be shackled to these figures.
FIrst, wait until the FINAL version of Mavericks comes out, the REAL MacPros are released, and then make sure Cinebench and Geekbench have OFFICIALLY said that the version you are running has been updated for Mavericks and THEN test. Many times, these tests need to be be updated when a major OS release comes out. But you should know that by now.
I would also see what happens with certain apps after they've been updated as well. Some will be improving their speeds so, I would wait until updates on apps before running speed tests.
This is only of my pet peeves, discussing and pre-release information on speed tests when NOTHING is officially released and the testing software isn't updated.
It can be VERY misleading if they aren't actually accurate.
As far as pricing.
I don't know which processor(s) they will offer for the entry level, etc. I did read that it's supposed to go UP to 6GB of GPU memory. Is that per card? So are there choices to get the product with less memory per GPU? I know they are supposed to SHIP standard with two GPU, but the text for that section on Apple's site suggests that the amount of GPU memory maybe another option.
I don't know what the pricing is for the CPUs/GPUs they are using, since they aren't specific.
Quote:
Originally Posted by wizard69
Beta hardware most likely too! Often Intel will go through a series of engineering samples as various issues are corrected or the build process tweaked. Beyond the fact that GeekBench isn't all that great of a way to tet a system, you need a stable system free of debug code and other slow ups that don't ship with released code.
In the end the only real benchmarks are those that test the software that is your primary money maker. How well the Mac Pro will perform will vary widely with the type of App being used and its revision level.
Exactly! One should concern themselves with apps and work loads they actually run.
Beyond that new instructions can often benefit many pro apps. Without a recompile targeting the new hardware you won't see the complete picture.
You are completely justified with respect to your peeves.
I'm really hoping Apple is taking the same tact that they have with the AIRs and other notebooks. That is to use advanced technology to gain an advantage in the marketplace that allows for fairly agressive pricing. In the end if the new Mac Pro has not addressed base price issues of the old MacPro it will tank in the market place.
Here's what we DON"T know. All of the final specs and pricing. I can wait another month or two. I think they will release the final numbers/specs in Oct. (Just a hunch).
It was an engineering sample they tested here:
http://www.tomshardware.co.uk/ivy-bridge-ep-xeon-e5-2697-v2-benchmarks,review-32756.html
but it's not going to be half the speed of the production version. The sample has to be close to the production model in clock speed, features and power consumption otherwise there's little point in having the sample.
There may be performance improvements with final software and production hardware but the important detail as far as it applies to the objection to Apple marketing up to 2x the CPU performance is that the new top-end 12-core will not be double the performance of the old top-end 12-core nor close to it. That would require Intel's performance-per-watt to increase 4x in 2 architecture steps, which doesn't happen. There should have been 3 steps since the old Mac Pro but there weren't. It's fair for people to object to their marketing but marketing material is usually not that accurate and they typically detail the comparison they use in their marketing pages. They did the same thing last year:
http://www.everymac.com/systems/apple/mac_pro/faq/mac-pro-mid-2012-performance-benchmarks.html
"in other promotional copy, Apple did reveal that the fastest custom configuration "Mid-2012" Mac Pro -- the Mac Pro "Twelve Core" 3.06 (2012/Westmere) -- is between 1.2 and 1.5 times faster than the fastest custom configuration "Early 2009" Mac Pro -- the Mac Pro "Eight Core" 2.93 (2009/Nehalem).
First, it is worth noting that this official comparison is a synthetic performance test using the "STREAM" 5.8 benchmark, and is a comparison with the much earlier "Early 2009" line rather than the previous, and effectively identical, "Mid-2010" Mac Pro line."
Some people will be disappointed they can't buy two 12-core processors but a Mac Pro like that would cost over $7500 so it affects very few people and Apple has never used the highest end processors in the past. What Apple would have used if they'd gone with two CPUs is a dual 8-core and it would have offered up to 50% more CPU performance for about $1300 more.
That extra performance option isn't essential because real-time feedback isn't required from CPU tasks. People who need more CPU performance can buy more machines e.g get a 6-core slave Mac Pro in addition to the 12-core. Less convenient in some cases but media software is using OpenCL more so the GPUs will provide good value there.
The final specs will be interesting. Apple provides plenty of wiggle room with the use of the "up to" phrase constantly. I'm just hoping that the design affords them a wider more competitive pricing range.
Quote:
Originally Posted by Marvin
It was an engineering sample they tested here:
http://www.tomshardware.co.uk/ivy-bridge-ep-xeon-e5-2697-v2-benchmarks,review-32756.html
but it's not going to be half the speed of the production version. The sample has to be close to the production model in clock speed, features and power consumption otherwise there's little point in having the sample.
There may be performance improvements with final software and production hardware but the important detail as far as it applies to the objection to Apple marketing up to 2x the CPU performance is that the new top-end 12-core will not be double the performance of the old top-end 12-core nor close to it. That would require Intel's performance-per-watt to increase 4x in 2 architecture steps, which doesn't happen. There should have been 3 steps since the old Mac Pro but there weren't. It's fair for people to object to their marketing but marketing material is usually not that accurate and they typically detail the comparison they use in their marketing pages. They did the same thing last year:
http://www.everymac.com/systems/apple/mac_pro/faq/mac-pro-mid-2012-performance-benchmarks.html
"in other promotional copy, Apple did reveal that the fastest custom configuration "Mid-2012" Mac Pro -- the Mac Pro "Twelve Core" 3.06 (2012/Westmere) -- is between 1.2 and 1.5 times faster than the fastest custom configuration "Early 2009" Mac Pro -- the Mac Pro "Eight Core" 2.93 (2009/Nehalem).
First, it is worth noting that this official comparison is a synthetic performance test using the "STREAM" 5.8 benchmark, and is a comparison with the much earlier "Early 2009" line rather than the previous, and effectively identical, "Mid-2010" Mac Pro line."
Some people will be disappointed they can't buy two 12-core processors but a Mac Pro like that would cost over $7500 so it affects very few people and Apple has never used the highest end processors in the past. What Apple would have used if they'd gone with two CPUs is a dual 8-core and it would have offered up to 50% more CPU performance for about $1300 more.
That extra performance option isn't essential because real-time feedback isn't required from CPU tasks. People who need more CPU performance can buy more machines e.g get a 6-core slave Mac Pro in addition to the 12-core. Less convenient in some cases but media software is using OpenCL more so the GPUs will provide good value there.
I went to Intel's site and it doesn't list a 12 core E5 series processor. here's the link that I went to. If anyone knows where they have information on a 12 Core E5 series processor, let me know.
http://www.intel.com/content/www/us/en/processor-comparison/compare-intel-processors.html?select=server