Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

2009 Mac Pro in hand - unboxing etc.

CrisMatthews said:
Not a benchmark I am familiar with, but here you go...

http://db.xbench.com/merge.xhtml?doc1=353678&doc2=1&setCookie=true

It's from my 2009 Quad 2.66 with 6Gb RAM.

Cheers

Cris.
Click to expand...

Thanks a lot! Here's a comparison with my Mac Book Pro 2,5 ghz:
http://db.xbench.com/merge.xhtml?doc2=353788

To me xbench gives valuable information of how the machine performs in daily life with apps, as it includes a lot of standard procedures that's common in most apps..

I think the Nehalem quad 2,66 really is lightning fast, - make a comparison with the 2008 2x2,8 ghz machine. Test by test, the nehalem beats it in important procedures. To get really high scores, much ram and a fast disk is necessary.
Thanks a lot. Hope some more will drop in with a test score later.
I'm gonna buy a new machine for my musicstudio, and was thinking of a quad 2,93, but not really sure yet. (Have already sold my powermac 2x2,5 which was a stunning powerful machine 4-5-years ago1:)

Cheers!
 
janht said:
Thanks a lot! Here's a comparison with my Mac Book Pro 2,5 ghz:
http://db.xbench.com/merge.xhtml?doc2=353788

To me xbench gives valuable information of how the machine performs in daily life with apps, as it includes a lot of standard procedures that's common in most apps..

I think the Nehalem quad 2,66 really is lightning fast, - make a comparison with the 2008 2x2,8 ghz machine. Test by test, the nehalem beats it in important procedures. To get really high scores, much ram and a fast disk is necessary.
Thanks a lot. Hope some more will drop in with a test score later.
I'm gonna buy a new machine for my musicstudio, and was thinking of a quad 2,93, but not really sure yet. (Have already sold my powermac 2x2,5 which was a stunning powerful machine 4-5-years ago1:)

Cheers!
Click to expand...
huh, id say get an octo instead :S
 
Surprised by same size!

I was surprised that after 4 years, the outside case is still the same size!

Left: 2009 MP 2x2.26, Radeon 4870
Right: 2004 G5 1x1.8
 

Attachments

  • SSCN0539.JPG
    SSCN0539.JPG
    73.9 KB · Views: 303
And lets continue the benchmarking,it it helps anyone to decide wether to go the MP09 way...


Cinebench10 :

MP 4x3.00ghz woodcrest (august2006/1.1)

1CPU : 3162
xCPU : 10527
OpenGL: 2591
 
Abidubi said:
Also supporting diminishing returns...

Multi-threaded speedup:

2.66 quad - 413%
2.93 quad - 376%

2.26 octo - 641%
2.66 octo - 640%
2.93 octo - 629%

Faster you go, less of a benefit.

Cinebench10_Numbers.jpg
Click to expand...
Sorry, noob question. I've done a quick search and not come up with anything. What is the percentage value of the "multcore" (multi-core??) black box set against? What is it actually comparing and is it of any great importance?
 
ncc1701d said:
Sorry, noob question. I've done a quick search and not come up with anything. What is the percentage value of the "multcore" (multi-core??) black box set against? What is it actually comparing and is it of any great importance?
Click to expand...

It is how fast it is when using all cores compared to using 1. If it is 4X as fast, thats 400%. If there is no change, it is 100%.
 
Abidubi said:
It is how fast it is when using all cores compared to using 1. If it is 4X as fast, thats 400%. If there is no change, it is 100%.
Click to expand...
Thanks Abidubi! So it's comparing it's speed with itself. I guess then the green and yellow boxes are more of a comparison against each other.
 
I was wondering....have most of you guys who have purchased a Nehalem Mac Pro configured it with a second SuperDrive?
 
Abidubi said:
It is how fast it is when using all cores compared to using 1. If it is 4X as fast, thats 400%. If there is no change, it is 100%.
Click to expand...

Yup! And it's important in this case because the new chips have a different architecture that takes better advantage of multiple cores. So for example in the case of the 2.66 quad 413% can be realized which is exceptional and quite a difference from the 315% realized from the previous architecture.

It also shows an overhead in this particular test too! Notice that the best octad scores only 641% which is nowhere close to the 800%+ that it would be getting if it scaled like the quad. :)
 
summitscout said:
I was surprised that after 4 years, the outside case is still the same size!

Left: 2009 MP 2x2.26, Radeon 4870
Right: 2004 G5 1x1.8
Click to expand...

Wow, seeing these two machines side-by-side really demonstrates how much the internal design as progressed since the G5

The 2009 Mac Pro is gorgeous

Who cares about external looks...
 
Tesselator said:
Yup! And it's important in this case because the new chips have a different architecture that takes better advantage of multiple cores. So for example in the case of the 2.66 quad 413% can be realized which is exceptional and quite a difference from the 315% realized from the previous architecture.

It also shows an overhead in this particular test too! Notice that the best octad scores only 641% which is nowhere close to the 800%+ that it would be getting if it scaled like the quad. :)
Click to expand...
Thanks Tesselator! However... I think I'm going to go a fully kitted out 2009 quad 2.93, but when I see only 376% versus much higher readings from the others (plus you suggest that the previous architecture getting 315% is not very good, which is very close to the 376% comparatively speaking), I get worried. But, since I don't fully understand what I'm looking at, it's probably best I remain stoooopid! ;)
 
Hey guys. I haven't read this entire thread yet so sorry if this is a duplicate question. How high does the 2.26 GHz processor step to when in Turbo Mode?
 
ncc1701d said:
Thanks Tesselator! However... I think I'm going to go a fully kitted out 2009 quad 2.93, but when I see only 376% versus much higher readings from the others (plus you suggest that the previous architecture getting 315% is not very good, which is very close to the 376% comparatively speaking), I get worried. But, since I don't fully understand what I'm looking at, it's probably best I remain stoooopid! ;)
Click to expand...

The quad 2.93 is nearly the same than the I7 core 940. And there is a lot of review that did cinebench 10 bench and show a score over 400%.

On barefeats the multicore result is also higher than the one on tesselator's graph. Maybe the on show here is a little low. We need more bench to have a more accurate result.
 
ncc1701d said:
Thanks Tesselator! However... I think I'm going to go a fully kitted out 2009 quad 2.93, but when I see only 376% versus much higher readings from the others (plus you suggest that the previous architecture getting 315% is not very good, which is very close to the 376% comparatively speaking), I get worried. But, since I don't fully understand what I'm looking at, it's probably best I remain stoooopid! ;)
Click to expand...

It just means that the new machines are speedier when apps use all of the cores.

If the machine is showing a 300% increase it means it's the same actual speed as 3 cores total. If it's getting 400% it's like getting another full core working for you.

But each core or percentage point remains respective to the core speed. If you get 380% at 2.93GHz that's still faster than 410% on a 2.26GHz box. (though less respective efficiency) You can think of it (in this example) as 3.8 cores of 2.93GHz vs. 4.1 cores of 2.26GHz. Multiply those relationships and you will get the Yellow bar for overall speed.

It's not too complicated. ;)

.
 
can you add the second processor later if you want to a new MAC Pro?

I bought the single CPU version
 
jketron said:
can you add the second processor later if you want to a new MAC Pro?

I bought the single CPU version
Click to expand...

In a way you may be able to. If speculations (and common sense) hold true then you can replace the riser card (daughter board) with one that has 2 CPU sockets and the extra DIMM slots. The riser board in single processor machines have no sockets for those tho - so you can't just buy the 2nd proc and plop it in.
 
Tesselator said:
In a way you may be able to. If speculations (and common sense) hold true then you can replace the riser card (daughter board) with one that has 2 CPU sockets and the extra DIMM slots. The riser board in single processor machines have no sockets for those tho - so you can't just buy the 2nd proc and plop it in.
Click to expand...

Which will probably be $500-$700 so would be unlikely to ever be worth doing unless a repalacement was needed due to repair.
 
I run man windows open at the same time but in most cases the content is static. I don't do much in the way of graphic editing. The most intensive application I run World Of Warcraft and I love to turn up all the graphics.

I have the GT120 card and I'm thinking about getting another to run a second monitor.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back