Yes, it is that simple. For single threaded applications the 3.33Ghz 6 core will turbo up to 3.6Ghz and the 2.93Ghz 12 core model will only go to 3.46Ghz. The 3.33Ghz 6 core will be faster if the application is using only one core.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
3.33 6core vs 2.66 12 core
- Thread starter allupons
- Start date
- Sort by reaction score
Yes, it is that simple. For single threaded applications the 3.33Ghz 6 core will turbo up to 3.6Ghz and the 2.93Ghz 12 core model will only go to 3.46Ghz. The 3.33Ghz 6 core will be faster if the application is using only one core.
Thanks a lot! Well that's good news for my wallet, haha
it's time to switch to a Mac it's been 16 years I had one, the good old 100mhz Mac performa 6300Taking my thread back for a second. So I did order the 6 core with a 5870 gpu upgrade last sunday night and it is still marked as not yet shipped. There were some annoying issues with my card's security limits, but we added a second card and the authorizations seemed to go well at that point. However, I believe that most people that ordered 6 core machines AFTER me have had theirs start shipping by the end of this week. I gave apple a call to see if anything weird was preventing shipping and the guy I talked to said everything looked fine on his end. I am basically curious how many that ordered earlier this week have had theirs ship or not.
Where can you verify this "turbo" information? If it is true, then when in turbo mode, the difference in speed between them actually reduces significantly from .4 to .14
Still, while this is interesting stuff, most people are not going to realize any noticeable difference during use of software so most of this banter is pointless academics at best.
Just look up the intel page on each processor for example http://ark.intel.com/Product.aspx?id=47920 . Btw the 2.93 will only go up to 3.33 not 3.46
in keeping with the thread topics, will dual multi core be better than single multi core for running more than one application at the same time?
I understand that photoshop does not benefit from the 12 core, and that Final Cut Pro doesn't so much, yet, either but should I assume that running a raw processor such as Capture One Pro on the same machine, going back and forth between it and cs4 or cs5 would make use of the split core banks?
I like using the mbp for raw processing, on the side, but am curious about the options.
If all you do is photoshop and video, how is the 12 core really even useful, let alone "necessary".
I understand that photoshop does not benefit from the 12 core, and that Final Cut Pro doesn't so much, yet, either but should I assume that running a raw processor such as Capture One Pro on the same machine, going back and forth between it and cs4 or cs5 would make use of the split core banks?
I like using the mbp for raw processing, on the side, but am curious about the options.
If all you do is photoshop and video, how is the 12 core really even useful, let alone "necessary".
If you are a heavy After Effects user that would be one reason to go all out as it can take full advantage of as many cores as you have. 3D Rendering, heavy virtualization, and Xcode compilation of large programs would be other reasons.
Photoshop isn't a good reason for a 12 core.
Photoshop isn't a good reason for a 12 core.
It is not. There are plenty around here that will say it is but it isn't. If you looking at overall system preformance the other elements of the system play a role.
Implicit in those that say it does are the assumptions that you have enough RAM and disk I/O speed allocated to the system where they aren't playing a major factor. That is typically easily done if run some older software on relatively small sized problems.
If you turn into a problem that can load up into the respective L3 caches and run a single core at then yeah you will see a marginal difference. The problem is that doesn't happen in real world when folks run multiple applications at the same time and have bigger than L3 cache problems.
I just posted Cinebench 11.5 and Geekbench 64 bit results for various Mac Pros including the 2.8 4-core, 3.33 6-core, 2.4 8-core, and 2.66 12-core:
http://barefeats.com/wst10.html
I'll add more permutations to the page as I receive them from "remote mad scientists."
I also recommend Lloyd Chambers page comparing the 3.33 6-core to three other Macs running Photoshop, After Effects, and several other apps. He also has memory bandwidth tests.
http://macperformanceguide.com/Reviews-MacProWestmere.html
Meanwhile, I'm compiling comparative results for Compressor, After Effects, iMovie, X-Plane, Portal, etc. Stay tuned.
http://barefeats.com/wst10.html
I'll add more permutations to the page as I receive them from "remote mad scientists."
I also recommend Lloyd Chambers page comparing the 3.33 6-core to three other Macs running Photoshop, After Effects, and several other apps. He also has memory bandwidth tests.
http://macperformanceguide.com/Reviews-MacProWestmere.html
Meanwhile, I'm compiling comparative results for Compressor, After Effects, iMovie, X-Plane, Portal, etc. Stay tuned.
brilliant job as always bare! thanks very much. is it possible for handbrake benchmarks?
cant wait for the compressor updates
And tons of other companies have workstation boards with 6-12 or more ram slots. Sticking with a limitation of intel's reference board instead of doing what the other companies are doing and coming up with a design that has more? Yeah, that's lazy. Not to mention that apple has eight slots on the 8/12 mobo so it's not like it's that outlandish of an idea (but even that should be 9 or 12 since it's triple channel memory).
What kind of benchmark is used for After Effects? I don't use AE (have it though) and I'd like to try one on my 8-core.
I am in exactly the same boat as you...I had originally ordered 8-core, dual cpu on 9 Aug. Then changed my mind and cancelled. I had to wait a few days for apple to release the hold on my cc which they finally did on sun (15 Aug). I promptly re-odered the six-core, and as of today STILL have the "Not yet shipped" on the status page. I also called, and was treated like, "why are you calling us, its on the website?" and was also told that the order looks ok, but since it is BTO (I ordered standard config with the 5870), that it takes awhile. Yeah right. Its says delivery date from 24-30 Aug so right now I'm pretty much thinking its going to be closer to 30 Aug.
brilliant job as always bare! thanks very much. is it possible for handbrake benchmarks?
cant wait for the compressor updates
HandBrake and Compressor are definitely on the "to do" list.
Send me a forum private message and I'll send you the test project and test instructions. BTW, even if you don't have After Effects, you can download a 30 day trial from Adobe.
Here's a Portal 'appetizer' for the 4-core 2010 Mac Pro with Radeon HD 5870 versus other Macs:
http://barefeats.com/wst10g.html
I'm going to be adding other game results including X-Plane, WoW, ETQW, CoD4, Team Fortress 2, Starcraft, Bioshock, etc. And I'll be posting results for the 5870 on previous generation Mac Pros.
http://barefeats.com/wst10g.html
I'm going to be adding other game results including X-Plane, WoW, ETQW, CoD4, Team Fortress 2, Starcraft, Bioshock, etc. And I'll be posting results for the 5870 on previous generation Mac Pros.
Barefoots,I am very,very happy that you have the time and patience to post us all this info.
Appreciated.
A lot!
Radeon HD 5870 in 2008 Mac Pro
I plugged the HD 5870 into the 2008 Mac Pro 8-core 3.2GHz. Works like a charm.
Here's an appetizer. I'm benchmarking Portal First Slice (which all my remote testers are using).
At 1920x1200 with these settings:
Model Detail = High, Texture Detail = High, Shader Detail = High
Water Detail = Reflect All, Shadow Detail = High, Color Correction = Enabled
Antialiasing Mode = None, Filtering Mode = Anisotropic 4X
Vertical Sync = Disabled, Motion Blur = Enabled
The 6-core Westmere 3.33 (2010) with Radeon HD 5870 = 233 FPS avg
The 8-core Harpertown 3.2 (2008) with Radeon HD 5870 = 250 FPS avg
The 2008 model is faster!!
I plugged the HD 5870 into the 2008 Mac Pro 8-core 3.2GHz. Works like a charm.
Here's an appetizer. I'm benchmarking Portal First Slice (which all my remote testers are using).
At 1920x1200 with these settings:
Model Detail = High, Texture Detail = High, Shader Detail = High
Water Detail = Reflect All, Shadow Detail = High, Color Correction = Enabled
Antialiasing Mode = None, Filtering Mode = Anisotropic 4X
Vertical Sync = Disabled, Motion Blur = Enabled
The 6-core Westmere 3.33 (2010) with Radeon HD 5870 = 233 FPS avg
The 8-core Harpertown 3.2 (2008) with Radeon HD 5870 = 250 FPS avg
The 2008 model is faster!!
More Radeon HD 5870 results in Mac Pro 2008 vs 2010
More Radeon HD 5870 results in Mac Pro 2008 vs 2010:
World of Warcraft "Narache Totem to Tree Run"
1920x1200 Ultra with 4X Multisampling
2010 Hexacore = 140 FPS avg
2008 Octocore = 153 FPS avg
2560x1600 Ultra with 4X Multisampling
2010 Hexacore = 110 FPS avg
2008 Octocore = 114 FPS avg
In fairness, not all 3D Games are faster on the 2008.
X-Plane 9.6 at 2560x1600 with High Quality and 2X AA
2010 Hexacore = 173 FPS avg
2008 Octocore = 105 FPS avg
That's all the appetizers for now. Back to testing.
More Radeon HD 5870 results in Mac Pro 2008 vs 2010:
World of Warcraft "Narache Totem to Tree Run"
1920x1200 Ultra with 4X Multisampling
2010 Hexacore = 140 FPS avg
2008 Octocore = 153 FPS avg
2560x1600 Ultra with 4X Multisampling
2010 Hexacore = 110 FPS avg
2008 Octocore = 114 FPS avg
In fairness, not all 3D Games are faster on the 2008.
X-Plane 9.6 at 2560x1600 with High Quality and 2X AA
2010 Hexacore = 173 FPS avg
2008 Octocore = 105 FPS avg
That's all the appetizers for now. Back to testing.
There is a caveat when using the Radeon HD 5870 on the 2006-2008 Mac Pro. The two Mini DisplayPorts are "negative function" no matter what adapter you have. Only the dual-link DVI port can be used on the older Mac Pros.
The 2009 and 2010 Mac Pros do support the Mini DisplayPort with or without a Mini DisplayPort to DVI adapter (for displays with DVI connectors). That's probably why Apple officially supports the 5870 and 5770 only on the 2009 and 2010.
The 2009 and 2010 Mac Pros do support the Mini DisplayPort with or without a Mini DisplayPort to DVI adapter (for displays with DVI connectors). That's probably why Apple officially supports the 5870 and 5770 only on the 2009 and 2010.
these are very interesting results - such massive differences! goes to show exactly how much the coding of the game/software comes into play, and how much 2 cores can make such big differences
thanks very much for these results.
So, along this video card tangent, is anyone familiar with this? :
NVIDIA Quadro FX 4800 for Mac by PNY / Part number: TW386LL/A $1800
1.5GB GDDR3 GPU memory, 192 cuda
http://store.apple.com/us/product/TW386LL/A?mco=MTQxMTc4NDg
I found that the fx 4800 above has, besides 50% more ram:
# 384-bit memory interface (higher than the 256 of the 5870)
# 76.8 GB/sec memory bandwidth (1/2 the 5870)
Is this any better than the "new" 5870, or not as good? It's certainly more expensive.
I noticed on some software, like After Effects, that "open gl 2.0" is a requirement - how does one be sure which cards have that feature?
Further about cards, and actual FCP type video work, would the 6 core 3.33 be "enough for now" or is the 12 core "substantially" faster? I am not too concerned about a couple seconds here or there but rather real appreciable speed difference.
From the sounds of this thread info, photoshop work is "faster" on the fastest ghz not so much number of cores, but is that even so much different to worry about? My idea is if the 12 core is better for some things, and will be better for more rewritten software in the years to come, what's the point of going 6 core except for cost?
If video work will be "plenty fast enough" on a 6 core, then it seems that is a fine route to go. I suppose I'd rather invest $2500 to become more money in a few years, than spend it unnecessarily on a 12 core if it's not really all that much better.
Lastly, what the heck was the 12 core really designed for is the 6 is so close?
-more lastly
is the 12 core inherently "better" because long term, for ram intensive uses, you can get a lot more inside which outweighs the ghz aspect?I have to read up, does cs5 finally "use" more ram thus functioning faster and not needing scratch disk? On my old g5 the functions I see super slow are saving the layered file as a psd (1gb+ most often) but also complex lasso with max feather and huge brush size mask painting. I realize some of these may be the old dual 2.0 and it's agp video card ? nvidia gforce 6800 w/ 256mb ram. machine has 6.5 gb running cs3.
retain ssd integrity via periodic reformatting ???
Can't the ssd simply be "hard erased" or reformatted back to original integrity/performance?
Don't do this.
Boot drive? Sure. Scratch disk? Bad idea. Especially not a tiny SSD. The scratch disk is constantly being written to, over and over. That will quickly degrade the performance of the drive and waste your money.
Can't the ssd simply be "hard erased" or reformatted back to original integrity/performance?