CInebench 10 and new Mac Pro 8 core Nehalem


#1

Hi to everybody. I’m in the process of updating my computer and I was waiting for the new Mac Pro. Finally they are out, but in my opinion the cinebench performance aren’t so impressive. I want to know if this can be an optimization problem of cinebench that is still based on Cinema 10 engine or is a real hardware limit.
If it is the last I think that at this moment an overclocked I7 920 would be the deal for me.
Thanks!

(as usual no PC/MAC war please, I’m interested only in understanding real performance gain of new machines)


#2

where did you see Cinebench results for it?


#3

Here is the result, check the cinebench tab under 3D section…

it says 1.3x faster =)


#4

On the apple website is stated only the 2,93 GHz performance gain in respect of “old” 3,2 GHz link
Than I have found several benchmarks in Macrumors forum that confirm more or less the apple one. You can find the 2,93 GHz at middle page at this link and also the 2,26 GHz in the following post.

For easy reading
Mac Pro 8 Core Nehalem 2,93 GHz Multiple CPU 25644
Mac Pro 8 Core Nehalem 2,26 GHz Multiple CPU 18088

Seem trustable


#5

Yeah it seems like the 8-core machines get almost 8 times the single thread score, for the first time (on a mac, anyway). Due, no doubt, to the hyperthreading, which means that there are 16 virtual cores at work. When put in that perspective, the per thread incremental benefit falls to less than 0.5…

Nick


#6

You cant complain that an 8 core machine is only 8 times faster than a single core machine, there arent 16 real cores so diving the score up by x16 simply isnt a fair comparison.


#7

Going from a crummy dual 2.3 G5, I think I’d be happy if one of you guys bought me one! :wink:


#8

Oh, I wasn’t complaining. In fact, I’m surprised that there is so much benefit from the hyperthreading…

Nick


#9

man…i wish i had money…i still have a single 1.8 ghz amd chip…i would kill for just a i7920.


#10

Hm I am on a 8x3.0 ghz Xeon and it scores around 20.000. Does that mean it has absolutely no use for me to upgrade? :hmm:

Actually kinda good :slight_smile: , 2 more years!


#11

these numbers are off!.. I think…

the I7 nahelem is scoring way more than this…

http://www.overclock3d.net/news.php?/cpu_mainboard/core_i7_3_2ghz_processor_blitzes_cinebench_10/1

I dunno what to make of it.


#12

Back in the day I was blown away at the difference my new 3.0 GHz Quad Mac Pro made to my renders out of C4D in comparison to my old Dual 2 GHz G5. It was like night and day even though the G5 was in it’s day a revelation going from my old machine. So it’s that time again, but this time I think I am going to have to wait. Because of the current financial climate the price of the top of the line machine that I have specced out is A$12,500.00 and that doesn’t include a monitor. Ouch. It’s pretty outrageous to be honest. And I want a 30 inch screen because I have started doing HD work pretty much all the time now.

The trouble is I don’t want to go to the 2.66 GHz version as it probably is only slightly faster than the current Mac Pros, which will no doubt be offered refurbished at reduced to clear prices.

I was so looking forward to this update but it’s not earth shattering enough for me, yet

all the best

rich


#13

Am I the only one noticing that the “new, awesome” 8 core 2.26ghz Mac Pro that is $3299 is actually SLOWER in clock speed and performance than the “old” 8 core 2.8ghz Mac Pro that can be had for like $2400 refurbished?

That’s almost $1000 more for… little bit better graphics? Cleaner innards?

I love my Mac Pro like a brother, but what is up with that? New machine significantly more expensive underperforms compared to a year ago’s model?

Am I missing something?

-shoqman


#14

It says “soon to be released” and the date was 20/08/2008
So i’m sure they did some more tweaking before the final release, but d00d 45000+?!?!?!

Come on who could complain about those scores. That’s incredible.

I’m extremly happy with my 14,000 scores, so I’d probably blow a gasket to see something like that.


#15

Well is arguable that it has 16 virtual cores, apple did rush this one out. Now you have to make your RAM work in sets of 3 and the dudes(Apple) instead of placing 3GB * 2 they went and place 6GB in a 1GB slot, that is very nasty business.


#16

The new MacPro is using server cpus and chipset. This solution has a much higher memory bandwidth (and can utilize it) compared to a single core i7 motherboard. Therefore it makes sense to use a combination of 2 x 3 RAM chips.

But the unquestionable downside of server cpus/chipset is the price. My personal opion is, that both (Apple & Intel) are skimming their customers with this machines (Intel’s price for the 2.66 GHz server cpu is $1000, while you get the regular 2.66 GHz Core i7 version for $290).

Best regards,

wbj


#17

They aren’t far off form majority of Benchmarks. And knowing Intel’s knowledge and access to Cinebench I’d wager that wasn’t an accurate presentation with a version of Cinebench that is a little beyond Cinebench 10 and not really comparable. They were probably running a Cinebench that was using the R11 rendering core instead of the R10 one which would throw off the numbers, or perhaps they tweaked the benchmark file. just by using R11 and turning off AA I can get my Q6600 to render the scene in 27 seconds instead of the almost 2 minutes it usually takes, that really changes the benchmark. I know when We did the benchmarks using the 24 core we had to use a modified version of cinebench that used the R11 core simply because the R10 core was limited to 16 processors, so it wouldn’t represent 24 very well.

Here is an example showing how several people with older CPu’s got comparable scores by using the R11 renderengine with cinebench. The problem there is the bench marking standards are based of an certain engine and when that engine is dramatically improved the original benchmark stats are no longer accurate so values will always be extreme. Gives you a good idea how much better R11 is compared to R10 for rendering though.

http://www.xtremesystems.org/forums/showthread.php?t=212575

to date I think the best cinebench I’ve seen from a third party source for an I7 was in the mid 20k range, 24-26k on an overclocked system, but then thats for a single physical cpu, while the macs are dual. I wish apple would give the actual benchmark instead of saying 1.3x faster than before.


#18

Yes wbj but why are they shipping them 6 gb on a one gig per slot? I did buy a MacPro and the RAM from apple was not fully buffered so i had to eat my 2 gb of RAM doesnt matter now but it came on 2 * 1GB sticks.


#19

You get the best performance out of a single socket i7 board by using 3 ram sticks - and you get the best performance out of a dual socket i7 board by using 6 ram sticks…

Best regards,

wbj


#20

If finances matter, $1440 for 0.27GHz isn’t worth it IMO.

A 2.66 with 16GB of RAM and the upgraded vid card ends up being about half the amount you quoted (with Applecare and in CDN funds). That’s plenty of machine for most applications isn’t it? Doesn’t seem like you should have to spend $12,5000 just for the machine. Besides, you can always get the box now and whatever additions later on can’t ya? What are you adding that jacking the price up so high?