Initial 955XE PCMark05 results

Printable View

Show 100 post(s) from this thread on one page

12-23-2005, 10:58 PM
mike

Quote:

Originally Posted by eva2000

i posted a pcmark05 excel calculator here http://i4memory.com/showthread.php?t=787 you can change each component's results for the suite score and see how it effects the final score... me thinks the hdd scores helped alot for 955XE :)

changing just hdd start up, file decryption, hdd general and audio compression on my excel spreadsheet for my opteron 170@2800mhz to your 955XE values put me at 10,200+ pcmark05 without taking into account virus scan which is a big boost in scores :D

LOL - paperbenching doesn't work too well crossplatform - specially not like that;)

Here's a comparo with a only 4.89GHZ on the 955XE with a slow single drive that would score 9mb/s in XP startup compared to the current #1 ( which is on a X2 http://service.futuremark.com/compare?pcm05=199017 )

http://img388.imageshack.us/img388/2...dino8iv.th.jpg

As you can clearly see only in test #1 does the X2 @ 3.265GHZ pulls a tiny winy bit on the 955XE - in all other cpu tests the 955XE has the upper hand. and this was ONLY @ 4.9GHZ - I can run PCMark up to 5.8GHZ so far....which would COMPLETELY change the comparo...

As far as 3DMark05 I did a comparo with a FX57 on air @ 3GHZ and the 955XE on air @ 4.5GHZ - the FX57 was 300points faster. But I will look inot it some more when I get more time. Think here the X2 and FX57 will keep the upper hand - while 03 and 01 could bring yet a whole other picture where the Intel could get really strong under LN2 or triple cascade

Quote:

Originally Posted by neoman

Very nice scores idd :O Whit what are u cooling ur CPU ? Vapo or Casecade ?

Was running dry ice - but not nearly to it's full potential... only ~ -20C on the cores

Quote:

Originally Posted by mrlobber

Mike, is that some special ram for you there which would be soon released into retail or are they only exceptional sticks picked out of the existing DDR2 around?

Actually it was some ram I got for evaluation - I just got another set thru retail channels which didn't do :banana::banana::banana::banana: - don't want badmouth the company - but bascially it's the same fatty D9 Micron chips that are in Corsair 5400UL and 8000UL DDR2 ram. So yes they were just exceptionally pciked sticks
12-25-2005, 08:59 PM
eva2000

Quote:

Originally Posted by mikeguava

LOL - paperbenching doesn't work too well crossplatform - specially not like that;)

and here i thought it did

Opteron 170 @2800mhz at 1.536v
2GB OCZ PC4000EB @254mhz 3-3-2-5 7-14-2223 at 2.64v
512MB HIS X1800XT @742/881
4X 80GB Hitachi 7K80 SATAII NCQ raid 0

http://www.fileshosts.com/DFI/NF4_SL...5_10962_tn.jpg

Compare against my 7710 score

http://www.fileshosts.com/DFI/NF4_SL...2vs7710_tn.jpg
12-25-2005, 09:01 PM
mike

Lol - software ramdrive changes everything yet again - doesn't it? GJ! I also see virus scan, file decyption etc. also got a nice boost! Awesome.

I think the more interesting comparison is X2/955XE on how the score in indivudal tests that are cpu related.
12-25-2005, 09:08 PM
eva2000

yes it does and in the real world now outside of pcmark05, i could use 4GB i-ram drive for windows OS i think and then use my 4x 80GB Hitachi 7K80 SATAII NCQ raid 0 array for data drives for hdd general usage and multi-task'd hdd virus scan (which the raid 0 array gets 101MB/s) like tasks = best of both worlds and higher performance theorectically than what i could get with PCmark05 :D
12-25-2005, 09:17 PM
mike

Quote:

Originally Posted by eva2000

yes it does and in the real world now outside of pcmark05, i could use 4GB i-ram drive for windows OS i think and then use my 4x 80GB Hitachi 7K80 SATAII NCQ raid 0 array for data drives for hdd general usage and multi-task'd hdd virus scan (which the raid 0 array gets 101MB/s) like tasks = best of both worlds and higher performance theorectically than what i could get with PCmark05 :D

:confused:
12-25-2005, 09:24 PM
eva2000

i mean Futuremark disapproves of software ramdrives officially for pcmark05 but for real world use they're nice :)

my latest

Opteron 170 @2850mhz at 1.6v
2GB OCZ PC4000EB @259mhz 3-3-2-5 7-14-2223 at 2.64v
512MB HIS X1800XT @742/881
4X 80GB Hitachi 7K80 SATAII NCQ raid 0

http://www.fileshosts.com/DFI/NF4_SL...5_11074_tn.jpg

11,074 vs 10,962 and 11,074 vs 7,710
http://www.fileshosts.com/DFI/NF4_SL...vs10962_tn.jpg http://www.fileshosts.com/DFI/NF4_SL...4vs7710_tn.jpg

3 way multi compare of 11,074 vs 10,962 vs 7,710
http://www.fileshosts.com/DFI/NF4_SL.../pcmark05.html

.
12-25-2005, 09:25 PM
mike

what point are you trying to make here?
12-25-2005, 09:28 PM
eva2000

1. benching is fun :D
2. for pcmark05 drive speed/performance is heavily weighted some what :)

3d /2d components would be higher if i used my 2x 256MB 7800GTX SLI so well >11,074
12-25-2005, 09:41 PM
mike

so still slower than the 955XE - right? And only if you use software ram drive - not iram...lol
12-25-2005, 09:45 PM
s7e9h3n

Quote:

Originally Posted by mikeguava

so still slower than the 955XE - right? And only if you use software ram drive - not iram...lol

Sorry mike, don't look up, but I think Eva beat your score ;)
12-25-2005, 09:47 PM
eva2000

hard to say you're using 2x 7800GTX SLI clocked at ? as well right ?

that's a hefty component too i loose 800pts from changing from 2x 7800GTX SLI stock speed to 1x HIS X1800XT in pcmark05 as well
12-25-2005, 09:49 PM
s7e9h3n

Quote:

Originally Posted by eva2000

hard to say you're using 2x 7800GTX SLI clocked at ? as well right ?

that's a hefty component too i loose 800pts from changing from 2x 7800GTX SLI stock speed to 1x HIS X1800XT in pcmark05 as well

Well, to be fair, you're not using DI either eva, and if you are, it's time to get yourself a better container cuz +36C won't cut it :p:
12-25-2005, 09:54 PM
eva2000

Corsair Hydrocool 200 1/4" tubed water cooling
http://www.fileshosts.com/watercooli...4_SLI_DR/html/

in 35+ C weather but got portable air con to drop room temp to 28-30C heh

i can certainly see that 955XE has improved alot compared to other intel cpus though
12-25-2005, 10:03 PM
mike

Well - sorry if I stepped on your toes - didn't mean to. I was referring to your initial post that using my HDD startup and HDD useage on your system would never score you 10700points - espcially not giving you ANY gains in any of the other test you were referring to.

I never looked at your claculator - I only assumed that this was what your claculator gave you as a result - which again in real world would have been wrong.

Espcially using the score of a Intel system can be applied to a AMD system. I have run very extensive test on both platforms and this was simply a wrong statement.

RE dual cards:
No, I am running a single GTX, and not nearly at what it can do. My single 256mb card easioly scores over 12k and my 512mb card scores easily over 13k. If you want to compare each others scores on by one - let's do that. Sorry I don't have a X1800XT at the moment. But if that's what you are intending to do - let's play
12-25-2005, 10:07 PM
eva2000

no no.. no toes stepped on... all is fine :D

Just responding to kunaak's and others comments on the boost the hdd start up and general hdd usage does to your Pcmark05 score not any other benchmark :) For 3dmark05 you easily beat me my X1800XT stock hsf air in the system @2800mhz only does 11,058 3dmark05 heh
12-25-2005, 10:10 PM
s7e9h3n

Quote:

Originally Posted by mikeguava

Well - sorry if I stepped on your toes - didn't mean to. I was referring to your initial post that using my HDD startup and HDD useage on your system would never score you 10700points - espcially not giving you ANY gains in any of the other test you were referring to.

I never looked at your claculator - I only assumed that this was what your claculator gave you as a result - which again in real world would have been wrong.

Espcially using the score of a Intel system can be applied to a AMD system. I have run very extensive test on both platforms and this was simply a wrong statement.

RE dual cards:
No, I am running a single GTX, and not nearly at what it can do. My single 256mb card easioly scores over 12k and my 512mb card scores easily over 13k. If you want to compare each others scores on by one - let's do that. Sorry I don't have a X1800XT at the moment. But if that's what you are intending to do - let's play

Uh oh....Mike's about to take out the big guns ;)
12-25-2005, 10:24 PM
Charles Wirth

We know who is gonna win at PCMark. ;)
12-25-2005, 10:26 PM
eva2000

well not me on my current cpu cooling LOL
12-25-2005, 10:27 PM
mike

Quote:

Originally Posted by eva2000

i can certainly see that 955XE has improved alot compared to other intel cpus though

actually the score differnce clock for clock is alomst non existent - similar like the difference between a 4600+ and a 4800+ - the huge imporvement is that the 955XE can bench @ around 6ghz while the 8x0 line was only able to bench around 4.5GHZ
12-25-2005, 10:35 PM
eva2000

Quote:

Originally Posted by mikeguava

the huge imporvement is that the 955XE can bench @ around 6ghz while the 8x0 line was only able to bench around 4.5GHZ

that's as well, but what i was talking about was the 955XE air/water results posted on this forum too much better than previous intel cpus on the same cooling and running cooler apparently

because i only have air/water cpu cooling i rarely look at phase change results to determine what cpu is best for me heh
12-25-2005, 11:02 PM
mike

Quote:

Originally Posted by eva2000

that's as well, but what i was talking about was the 955XE air/water results posted on this forum too much better than previous intel cpus on the same cooling and running cooler apparently

because i only have air/water cpu cooling i rarely look at phase change results to determine what cpu is best for me heh

gotcha - and I can assure you that the 955XE on air is faster than the current X2 on water - on the all the cpu relevant scores . Let's see what AMD will come out with soon. Tide can always turn quickly

Show 100 post(s) from this thread on one page