Sign in with
Sign up | Sign in
Your question

3dmark06 results are in...

Last response: in Systems
Share
November 6, 2006 1:59:01 AM

MOBO: ASUS M2N32-SLI
CPU: Athlon XP 4200+ AM2
RAM: 2048MB DDR2 800MHz
HDD: 960GB SATA II
GPU: EVGA 7900GTO

SM2.0: 2756 points
HDR/SM3.0: 2677 points
CPU: 1594
Total Score 6142
Ranking out of 283 similar systems: 43

I got my ass handed to me by a Intel C2D X6800 with a 7950GTX, the top ranked system scored over 16,000 points

Red Valley brought my system to it's knees....Ugh

I don't guess i did bad for a box-stock custom build with no tweaking or overclocking, but i definately see a G80 graphics card and a much (very much) faster CPU in my future.......I'm hoping the 5000+ or maybe Opteron will give me the boost i need so i can stay with this motherboard, if not.........then I'm changing my name to CONROE C2D

More about : 3dmark06 results

November 6, 2006 12:22:58 PM

Sorry to hear that. Download and try 3DMark05 and see what you get. It could very well be due to the fact that your mother board has a Southbridge chipset as well as a Northbridge. Also, just a question, did you download the AMD cpu drivers for your processor? Windows only has AMD cpu support for K6's and earlier. Installing the CPU drivers could drastically improve perfomance and scores. Also, even though you saved a ton by going with the 7900GTO, a 7900 GTX will have double the scores and performance of a 7900GTO. They literally squeeze every ounce of perfomance out of a 7900GTX. Also, invest in a Western Digital Raptor HDD. They're VERY FAST, with 8ms or less access times and they have transfer speeds of up to 122 mb/s. You're set for storage for life almost :tongue: . Otherwise it looks like a great setup, just needs tweaks is all.
November 6, 2006 1:25:26 PM

Quote:
Sorry to hear that. Download and try 3DMark05 and see what you get. It could very well be due to the fact that your mother board has a Southbridge chipset as well as a Northbridge. Also, just a question, did you download the AMD cpu drivers for your processor? Windows only has AMD cpu support for K6's and earlier. Installing the CPU drivers could drastically improve perfomance and scores. Also, even though you saved a ton by going with the 7900GTO, a 7900 GTX will have double the scores and performance of a 7900GTO. They literally squeeze every ounce of perfomance out of a 7900GTX. Also, invest in a Western Digital Raptor HDD. They're VERY FAST, with 8ms or less access times and they have transfer speeds of up to 122 mb/s. You're set for storage for life almost :tongue: . Otherwise it looks like a great setup, just needs tweaks is all.


Your comment about Windows support for AMD cpu's. Windows XP has provided support for K8 since last April. It is available as a patch on the Microsoft Update site. Its full name is 'Windows Driver Package - Advanced Micro Devices (AMDK8) Processor (04/28/2006 1.3.1.0)'.

-Bob
Related resources
a c 82 à CPUs
November 6, 2006 2:03:18 PM

Quote:
MOBO: ASUS M2N32-SLI
CPU: Athlon XP 4200+ AM2
RAM: 2048MB DDR2 800MHz
HDD: 960GB SATA II
GPU: EVGA 7900GTO

SM2.0: 2756 points
HDR/SM3.0: 2677 points
CPU: 1594
Total Score 6142
Ranking out of 283 similar systems: 43

I got my ass handed to me by a Intel C2D X6800 with a 7950GTX, the top ranked system scored over 16,000 points

Red Valley brought my system to it's knees....Ugh

I don't guess i did bad for a box-stock custom build with no tweaking or overclocking, but i definately see a G80 graphics card and a much (very much) faster CPU in my future.......I'm hoping the 5000+ or maybe Opteron will give me the boost i need so i can stay with this motherboard, if not.........then I'm changing my name to CONROE C2D


My GTO on PCI-Ex4 with a E6600 on 2Gb of DDR400 is hitting about 5,500 on '06. The big hitting systems you are seeing there will be SLI in all probability. Don't feel too bad, its a good score.
November 6, 2006 2:23:54 PM

Quote:
Red Valley brought my system to it's knees....Ugh


It's supposed to - virtually every machine crawls through there - isn't that the one rendering the graphics calculations directly from the CPU instead of the GPU?
November 6, 2006 2:40:15 PM

SPARTAN-117 makes little sense. What are you talking about that he gets a low score because his mobo has a NB and SB?

7900GTX is not double the performance of a 7900GTO.
November 6, 2006 6:46:35 PM

Yes, that is one of the CPU tests, my 4200 was hitting 0 to 1 frames per second.
I'd be curious to see how AMD's new CPU/GPU combined chip does on red Valley.......it may just be the way to go for me.
The graphics rendering power of a dedicated graphics card combined with a dual-purpose CPU/GPU Quad-core chip.
The tables may turn once again, at least for a while.
November 6, 2006 7:59:49 PM

AMD has dual core drivers that Windows doesn't. mad-dog's CPU is a dual core 4200+ and AMD makes a specific drivers package for them, when I built my system I got the version 3 drivers and already they're on 3.1 something, so the AMD dual core specific drivers will help alot. Also you're right about the package, but a % of us who use Windows won't go to the website at all for fear of the infamous WGA bug. If it don't come in the auto update chances are I don't have it.
November 6, 2006 8:12:55 PM

I'm not saying that having both a North and South bridge is lowering his score, I'm saying its a possibility. You would know that if you had actually read what I wrote. The reason DDR2 boards have both chipsets, according to Tom's Hardware, is that the Southbridge works together with the CPU memory controller to operate the DDR2 memory. Its a little less efficient and has been noted to cause loss of performance, but not by a significant amount. Also if you look at the Tom's Hardware VGA charts you'll see that the 7900GTX scores almost a 1000 pts (2000 vs 3000) more than a 7900GTO, not quite double but almost.
November 6, 2006 8:47:37 PM

50% = 100% ???
a c 82 à CPUs
November 6, 2006 9:01:36 PM

Quote:
Also if you look at the Tom's Hardware VGA charts you'll see that the 7900GTX scores almost a 1000 pts (2000 vs 3000) more than a 7900GTO, not quite double but almost.


Odd maths, but precisely which toms VGA charts is the GTO in, I can't see it in the std VGA charts, or is it some other tom?
November 6, 2006 9:52:21 PM

I second this what charts are you referring to. I have looked up and down on the VGA charts which are located at http://www23.tomshardware.com/graphics.html and i can not seem to find the GTO anywhere. I have 2-GTO's in SLI as my sig shows and with 3dMark06 i hit 9396 with NO OC'ing at all. I am about 70% sure that if i pushed a little OC on the cards i could run right with the GTX cards. Now 9396 is not an off the wall score but it is what i scored i hope to soon start OC'ing and see how high i can get this baby to go.
November 6, 2006 9:54:24 PM

True, the VGA charts listed don't have my card listed.
The 7900GTO has 512Mb of DDR3 and a 24 pixel pipeline vs the 7900GT having 256MB DDR3 and a 12 pixel pipeline.
I'm shocked to see how much better the 7900GTX performs than the 7900GTO, the way i understand it is the 7900GTX is just a overclocked 7900GTO......go figure that the GTX was 2x as expensive as the GTO.
I'm gonna RMA the GTO and try to make a decision on the 7900GTX or the X1950XTX.
i thought just adding another 7900GTO and run it in SLI but it still wouldn't come close to the GTX or X1950 except in 3dmark06......in real world gaming the GTO wont cut the mustard, even in SLI mode.
November 6, 2006 10:04:07 PM

The only difference between the GTO and GTX is memory speed NOTHING else. GTO memory clocked at 1320MHz GTX - 1600MHz. That is all that is different. And the GTX is not and over clocked GTO. The GTO is an under clocked GTX the GTO was released way after the GTX. But do as you will my friend has 2 BFG 7900 GTX in SLI and we run right beside each other in all benchmarks we have ran. But i spend 1/2 of what he did in video cards price wise. Here is a nice review of the card http://www.firingsquad.com/hardware/evga_e-geforce_7900...
November 6, 2006 10:14:16 PM

A GTX simply does not double the GTO's scores, end of story. No matter how you slice it, it's not even close. Nevermind 2000 to 3000....where does that come from? How do you say you are "shocked at how much better the GTX performs than the GTO"? Where is this information you are quoting? Sorry, but it doesn't exist......your info is incorrect. I know, I've benched both cards here at this very keyboard.

How does 2 7900GTO's not match a 7900GTX?!?!?!?! Dude, seriously, where is your info coming from? Wow, that is simply the biggest piece of misinformation I've seen here EVER. 7900GTOs in SLI will spank any single card on the market, bar none (excluding the 8800, which isn't released yet). A 1950XTX won't even come close, not even maybe. Even a 7900GX2 won't touch GTO's in SLI. Wow, I'm astonished at your misinformation, it's simply wrong. Sorry to tell you, but if you RMA your GTO and get a GTX, you'll be sorely disappointed, not to mention far more broke in the pocketbook.

A stock GTO can score somewhere around 5900 - 6000, depending on your system specs. When overclocked to GTX speeds, it will score about 6400-6500, plus or minus several hundred points.

So, assuming even a 1,000 variance in the total score (which is more than generous), you're going from 5,800 to 6,800 (and that is generous). That is a 15% increase. Not 100%. Not 50%. 15%. That's it, and that's being generous. It's actually more like 10% performance difference. And overclocking a GTO will bring you to within 3-4% difference. That's nothing, a few frames in any game.

So, to the Original Poster.......a GTO is an amazing performer for the price, simply awesome compared to the GTX. The GTX is a few % higher, nothing more than that. Everyone knows this, and it's been proven. Someone claiming 50%, 100% or whatever difference is unfortunately very mistaken.

Regardless, your system is good. But an 8800GTX will spank the 7900GTO ;) 


I hope this clears it up and this misinformation does not continue.
November 6, 2006 10:14:53 PM

Well, if you want a faster CPU. The AM2 Opterons look nice.

http://www.newegg.com/Product/ProductList.asp?Category=...

If you can afford one, the 1216 (2.4Ghz) or the 1218 (2.6Ghz) would be a great processor. They also have more cache than the Athlon X2 you have. That would be better when multitasking. I think I'm going to buy myself a 1212 and overclock it.
November 6, 2006 10:18:57 PM

I'm already running SATA II's in RAID1 for redundancy, i understand that may slow me down a little but from what i understand configuring RAID0 doesn't provide that much of a speed boost.
The transfer speed is 3.0Gig /sec, IDE can't touch that...
November 6, 2006 10:22:37 PM

The HD's in RAID will rarely hit a full 3.0.....saturation to that point is actually quite rare. So they are faster than IDE but we're not talking minutes or seconds....we're talking minor in regular tasking, but somewhat more in longer data transfers.

Sorry man, but first the GTO/GTX bogus claim and now this? Where DO you get your misinformation????

Seriously bud, you should check your facts. Misleading people (even if unintended) here causes alot of headaches.
November 6, 2006 10:29:46 PM

which came first the chicken or the egg, who cares.....whether the GTO is underclocked or the GTX is overclocked makes no difference to me because one came out before the other......I'm sorry you're confused.

i should've stated that as saying the rated transfer speed for a SATAII drive is 3.0Gig/sec.......i didn't mean to imply that my system is operating at that transfer speed.
November 6, 2006 10:33:07 PM

Don't worry about your scores, very acceptable and who needs more benchmark points except hardcore overclockers and world record attempts.

Hard drive upgrade will do very little for your scores and if you are happy with the gaming performance, you should be happy with your system!
November 6, 2006 10:36:18 PM

Quote:
True, the VGA charts listed don't have my card listed.
The 7900GTO has 512Mb of DDR3 and a 24 pixel pipeline vs the 7900GT having 256MB DDR3 and a 12 pixel pipeline.
I'm shocked to see how much better the 7900GTX performs than the 7900GTO, the way i understand it is the 7900GTX is just a overclocked 7900GTO......go figure that the GTX was 2x as expensive as the GTO.
I'm gonna RMA the GTO and try to make a decision on the 7900GTX or the X1950XTX.
i thought just adding another 7900GTO and run it in SLI but it still wouldn't come close to the GTX or X1950 except in 3dmark06......in real world gaming the GTO wont cut the mustard, even in SLI mode.
That's incorrect; the 7900GT has 24 pixel pipelines, and the 7900GTO is barely slower than the 7900GTX. Chances are the scores you're comparing are of an SLI system. SLI is not recognized by the ORB, however the performance increases are still there.
November 6, 2006 10:39:32 PM

Excellent point Heyyou27.....it could be that mad-dog is accidentally looking at the SLI scores. It is proven fact that a GTO is extremely close in performance to a GTX, both in benchmarks and in real world apps. However, comparing a GTO to GTX's in SLI......well, that's another story altogether....that would be a world of difference.

Mad-dog, did you perhaps confuse the scores between single and SLI? That could explain your understanding of the difference..........?
November 6, 2006 10:39:33 PM

Well it does make some difference considering that the GTO is an under clocked GTX you can simply OC the card and preform just the same as a GTX. And it is NOT like the chicken and the egg we know which one came first here and that is the GTX look it up if you would like. The complete build of the GTO was built off of the GTX from design of the card to almost exact specs. From all i can find some of the GTO's might have a different kind of memory on them which would explain for the lower clocked memory.
November 6, 2006 10:43:19 PM

I guess my OP did sound like i was comparing my system to the other one, but that is not the case.
i only meant to use that score to showcase what can be done with a system, It tis a huge difference of almost 10,000 points.
the Op did state that i ranked 43 out of 283 similar systems or something like that, so i don't believe it's me that's blowing things out of proportion..
i only wanted to show what could be done............lol
November 7, 2006 12:10:51 AM

I know, I was confused too. But after some looking at other benchmarks on the web I determined that the 7900GTO is listed on Toms Hardware's VGA chart as a 7900 GT Top (odd I know).
November 7, 2006 12:13:12 AM

I can't speak for all of us here, but I guarantee me and oh say 50% of the guys on this forum don't ever visit Microsoft for update packages outside of the Auto update.
November 7, 2006 12:22:40 AM

Look in the Toms Hardware VGA charts for the 7900GT Top and the 7900GTX and compare the 2. Now I know the specs for the 7900GT Top aren't listed correctly for the 7900GTO, but after comparing benchmarks from several other sites on the web I've determined that the 7900GTO is listed on Toms Hardware VGA charts as the 7900GT Top. Also, I'm not saying that the 7900GTO sucks in SLI, I'm saying that as a standalone card its scoring about a 1000pts less than a 7900 GTX. I guarantee that 2 7900GTO's in SLI will whip a single 7900GTX anyday. You wanna know why, and this one is sraight off of several VGA manufactuerers websites, when you go to SLI even though your physical hardware capacity has doubled, your actually graphics processing power has quadrupled.
November 7, 2006 12:57:08 AM

The only difference in the GTX and GTO is the downclocked speeds, AND a change in Voltage.

The GTO is missing one small transistor, which will not let you get the vcore up to GTX power levels.

My GTO is running within 20 ticks of the GTX speed, but will not hit it even after flashing the bios.

http://www.mvktech.net/component/option,com_joomlaboard...

Halfway down is the picture. More than you wanted to know.
November 7, 2006 1:05:34 AM

You want physical, tangible data, well thats what you're going to get. Accorrding to the specs listed at newegg a eVGA GeForce 7900 GTO has a core clock of 650mhz, memory at 660mhz (1320mhz for DDR), 256-bit memory interface and 512mb of RAM. Don't believe me, here's the link to newegg http://www.newegg.com/Product/Product.asp?Item=N82E1681...

Also accorrding to newegg, a eVGA 7900GTX has a core clock of 650mhz, memory at 800mhz (1600mhz for DDR), 256-bit memory interface and 512mb of RAM. Again if you don't believe me see for yourself http://www.newegg.com/Product/Product.asp?Item=N82E1681...

As for the where I'm getting my benchmark info I checked the VGA charts at Toms Hardware and determined, after comparing several other benchmarks on the web, that the 7900GTO is listed as a 7900GT Top. Don't believe me check the link http://www23.tomshardware.com/graphics.html?modelx=33&m....

Quriteria are: 7900GTX w/G71, 650mhz core, 800mhz (1600mhz for DDR) memory, 512mb RAM vs. 7900GT Top w/G71, 520mhz core, 720mhz (1440mhz for DDR) memory, 256mb RAM. Test was in 3DMark06 (v. 1.0.2), 1024x768, no AA, Trilinear Filtering.

Also, I have my own math to prove what I'm saying. The numbers I'm going to reference come from 3DMark05 Pro (v. 1.2.0). Now, because my video card came from BFG, my 7900GTX was already OC'ed to 670mhz core and 820mhz (1640mhz for DDR) on the memory. Now before I oced the VGA from stock I scored a 9350. After I OC'ed the VGA from stock to a 3% overclock using nVidia's nTune app. (v. 1.1) my VGA came to rest at 688mhz core and 841mhz (1682mhz for DDR) memory. This time I scored a 9527.

So according to my math an increase of 18mhz on core speed an increase of 21mhz (42mhz for DDR) memory will result in an score that is 177pts higher than a non OC'ed VGA. So quite conceivably, you can OC a 7900GTO upto a 7900GTX, but that's unlikley and highly un-recommended.

So, I'm pretty sure that the almost 150mhz (300mhz for DDR) memory clock speed explains why a 7900GTO scores almost a 1000pts less in 3DMark06 Pro than a 7900GTX.

I hope this explains what I've been talking about and clears up any confusion. :wink:

--------------------

My System:

AMD 64 X2 3800+ OC'ed to 2.2 Ghz, BFG 7900 GTX 512 mb OC'ed, ASUS A8N-SLi Deluxe, 2 Gb Kingston Value RAM CAS Latency 2.5-3-3-6, Creative Labs Sound Blaster Audigy 4, Westen Digital 74 Gb Ratop ADFD, 2x Western Digital 250 Gb Caviar SE, 3x Thermaltake Hardcano 14 HDD coolers, Thermaltake Blue Orb II CPU cooler, Zalman VF-Cu 900 Blue LED VGA cooler, Thermaltake Extreme Spirit Northbridge cooler, and a PC Power & Cooling 510w PSU.
a c 82 à CPUs
November 7, 2006 7:10:01 AM

Quote:
eVGA GeForce 7900 GTO has a core clock of 650mhz, memory at 660mhz (1320mhz for DDR), 256-bit memory interface and 512mb of RAM.

7900GT Top w/G71, 520mhz core, 720mhz (1440mhz for DDR) memory, 256mb RAM.


Unfortunately you posted information to prove yourself wrong and so you said it yourself, they are not equal, different amounts of RAM, different core speeds and different memory speeds.
November 7, 2006 10:23:23 PM

Did you not read what I wrote? From what I've seen on the web the 7900GTO is listed in the Tom's Hardware VGA charts as a 7900GT Top. The only reason I listed the specs was in case the link failed and that others could see what I saw. Now I know everyone on this site thinks Tom's Hardware is always right, but your all forgeting that the humans who run the Toms Hardware website are just that, human and thus they're falliable. So you can take your "I'm right you're wrong" attitude and shove it up your ass!!!! :!:

--------------

My System:

AMD 64 X2 3800+ OC'ed to 2.2 Ghz, BFG 7900 GTX 512 mb OC'ed, ASUS A8N-SLi Deluxe, 2 Gb Kingston Value RAM CAS Latency 2.5-3-3-6, Creative Labs Sound Blaster Audigy 4, Westen Digital 74 Gb Ratop ADFD, 2x Western Digital 250 Gb Caviar SE, 3x Thermaltake Hardcano 14 HDD coolers, Thermaltake Blue Orb II CPU cooler, Zalman VF-Cu 900 Blue LED VGA cooler, Thermaltake Extreme Spirit Northbridge cooler, and a PC Power & Cooling 510w PSU.
a c 82 à CPUs
November 7, 2006 10:47:26 PM

Yes I did read what you wrote and I quoted the piece I had particular issue with, but to spell it out for you.

Look at the clock speeds and the RAM speeds and the RAM amounts, they are different cards, you can't really compare results from different sites anyway, and certainly not to determine that one card is in fact another. By that measure we'll take my stock GTO scores on '06 of 5500, therefore it is twice that of a GTX.

You can hopefully see it doesn't work, also the processor weighs heavily on '06, the rest of the system will change the results, so you can't take an anandtech or an xbitlabs result and use them to compare against a toms result. None of the results are any better or more accurate than any other, but they are only comparable within themselves.

I could understand that toms might have named it wrong, but to also name all of the attributes wrong is unlikely, especially as there is such a thing as an Asus 7900 GT TOP. However going on past history it is taking toms a significant amount of time to post new cards onto the DB so I'm not surprised its not there.

As for your last point, grow up, you've been very defensive from post one on this thread, and not really seemed to have much knowledge, for instance there are very few CPU's that hit >1 FPS in red valley, so its not a 'driver issue'.

The OP could have benefitted from knowing that their score is actually quite normal, but instead you spread misinformation, probably through misinterpretation rather than anything malicious I grant you. The effect of this is that someone might now be buying a new card when they did not really need one.
November 7, 2006 11:15:08 PM

I'm sorry :cry:  . I've been under stress lately and quiet often it seems lately, I've lashed out without having every fact at hand. Being a full time college student and having a full time job will do this to you :x . You're right about ASUS having a 7900GT Top, I'd never found it because it's listed as a 7900 GT on newegg and I thought a 7900GT Top was a breed unto its own. I wasn't saying his score was bad, and I certaintly wasn't trying to make him buy a new VGA, but I'm confused and maybe you can help me? What's so drastically different in 3DMark06 that the scores appear lower than those that are in 3DMark05? As I've said, I scored a 9350 stock in 3DMark05 on my system, so would this translate into roughly a 5000-6000 for a score in 3DMark06? And concerning the "driver issue", this one actually came from personal experience. I was having problems of all sorts with anything until I installed the AMD CPU drivers for my 64 X2. It was like a parting of the clouds so to speak, therfore I thought it might help him.

And to anyone whom I may have offended, I'm sorry :cry:  .

----------------------------

My System:

AMD 64 X2 3800+ OC'ed to 2.2 Ghz, BFG 7900 GTX 512 mb OC'ed, ASUS A8N-SLi Deluxe, 2 Gb Kingston Value RAM CAS Latency 2.5-3-3-6, Creative Labs Sound Blaster Audigy 4, Westen Digital 74 Gb Ratop ADFD, 2x Western Digital 250 Gb Caviar SE, 3x Thermaltake Hardcano 14 HDD coolers, Thermaltake Blue Orb II CPU cooler, Zalman VF-Cu 900 Blue LED VGA cooler, Thermaltake Extreme Spirit Northbridge cooler, and a PC Power & Cooling 510w PSU.
November 8, 2006 5:00:19 AM

I apologize for making it seem like i thought my scores were low when indeed they weren't, I originally scored 43/283 similar systems.

I only posted that high score of 16,000 something to show what could be done with the right gear, settings and control environment.

From what i can tell the 7900 GTO "IS NOT LISTED", only the 7900 GS which is a totally different animal.

Before i decided on the 7900 GTO i read review after review and everyone said it's a steal for $250 because it's basically a underclocked 7900 GTX.

Too bad the latest VGA charts don't list the 7900 GTO so we can see how it does stack up against the competition, i wanna believe that the GTX is much faster than the GTO but i don't believe it is.

5-10% maybe 15% with a strong backwind.......i can't justify upgrading to the GTX unless someone can show me that the GTX is "ALL THAT", then I'll whip out the plastic and lay it down for the balance.

On a side note:
As i stated before, the 8800's are the future of gaming but without VISTA to support the DirectX/10 i believe they are a waste of money with a Windows OS.

Microsoft has already stated they are not releasing DirectX/10 drivers for Win XP so if you wanna reap the full benefits of the new cards you gotta upgrade to VISTA............blackmail i tell ya, pure and simple blackmail.
a c 82 à CPUs
November 8, 2006 7:21:38 AM

I'm also running a GTO, any tests that I can do to give you a little piece of mind.

I'm thinking that if someone with a GTX could post their '06 scores broken down into the 3 categories and maybe the frame rates too it might help to see what kind of boost that extra memory speed will give.

Those 16,000 scores seems to me to me SLI GTX although that puts a single GTX at maybe 10,000 given inefficencies, so it may even be quad, and the GX2's are being treated as 2xgtx's. with a slower memory speed.



Like I said my system is hobbled due to DDR400 and PCI-E x 4.
a c 82 à CPUs
November 8, 2006 7:31:32 AM

Don't worry about it, in fact you might be able to help.

If you can do a similar screenie of '06 like I have done below we can try and isolate out the processor, and then let mad dog see what the difference really is.

In my experience a 10,500 '05 dropped to a 5,500 '06, it has a heavier CPU bias from what I have read on the forum and so my E6600 is lifting me a little I think. so your 9,350 I'd expect to drop to maybe 4,500 with no processor bias, maybe a little lower. It'd still be nice to the Sm2.0 and Sm3.0 scores with the framerates.

A word of advice, if you are trying to do two full time things the forumz probably are not the right place for you right now, they are addictive to some poeple (possibly myself included, but I'm between contracts right now), take that how it is meant, becareful how you spend your college time, you ony get to do it once (unless you are very lucky or rich :wink: ). And there are better things than PC's at college.

You might be right about the drivers, but very little below highly overclocked FX's and C2D's get anything much about 1 in that specific test, so I doubt it would give him much.
November 8, 2006 12:38:16 PM

I do have a 7900GTX from BFG, so the stock speeds are faster than a 7900GTX from oh say eVGA. I've also got one question in regards to CPU generated stuff in 3DMark05, is it common for a 64 X2 3800+ oc'ed to 2.2ghz to do 5-7fps by itself? I'm pretty sure by what you said that its hard to get higher than 1fps in red valley on 3DMark06 with what I've got. As to the thing about being careful about 2 FT's at once, you ain't kiddin! But going to a community college with transfer programs to major 4 year schools helps alot. My 7900GTX has SM 3.0 and CineFX 4.0 with HDR and a max res of 2560x1600. The problem with 3DMark06 is that I'm at school or work all day so I don't want to leave it running all day w/o me being there, and I live with my parents (for now anyway) so my PC is in my room and this is when it would be preferable to have a PC w/o a side window :tongue: , so that when I do have a chance to leave it run to download 3DMark06 (about 3hrs ETA on my high speed cable) I usually turn it off and go to sleep :lol:  . Hope to get 3DMark06 soon though and give you all my scores so we can all better interpret 3DMark06.

----------------------------

My System:

AMD 64 X2 3800+ OC'ed to 2.2 Ghz, BFG 7900 GTX 512 mb OC'ed, ASUS A8N-SLi Deluxe, 2 Gb Kingston Value RAM CAS Latency 2.5-3-3-6, Creative Labs Sound Blaster Audigy 4, Westen Digital 74 Gb Ratop ADFD, 2x Western Digital 250 Gb Caviar SE, 3x Thermaltake Hardcano 14 HDD coolers, Thermaltake Blue Orb II CPU cooler, Zalman VF-Cu 900 Blue LED VGA cooler, Thermaltake Extreme Spirit Northbridge cooler, and a PC Power & Cooling 510w PSU.
a c 82 à CPUs
November 8, 2006 1:19:37 PM

the '05 cpu test i not as harsh as the '06 test, effectively think of '06 as having increased image quality created by the CPU, hence it is a much harder test.

Hopefully someone else might have the gtx details.
November 8, 2006 1:55:57 PM

I'm currently download 3DMark06 and hopefully by tonight, tommorow at the latest, I'll have the results.
November 8, 2006 2:21:01 PM

And i, in return will download 3dmark05 so we can compare apples to apples and oranges to oranges.....other than the differences in CPU of course.

3 hours to D/L '06.......maybe at 56K, i think it took around 45min with my DSL, you've gotta go back and try different mirror sites till you find the one that's kickin out around 350Mb/sec
November 8, 2006 2:26:42 PM

Ok there seems to be some confusion here.

6100 is fine for a 7900GTO.

3Dmark06 cant tell the difference between a single card and SLi, that 16000 score was more than likely a highly overclocked pair of water cooled or LN2 cooled cards and CPU. Thats basically 8000 per card, not toooo much higher than yours.

There is no difference in pipes/shaders etc between the 7900GT, the 7900GTO, the 7950GT, and the 7900GTX. The only G71 (7900) based card without 24 pipes is the 7900GS, which has 20. The difference between them is just core and memory speeds.

The 7600GT has 12 pipes and must be the one you are thinking of.
November 8, 2006 2:36:36 PM

ok, when you break it down like that it all makes sense....
it was clear as mud before, THANK-YOU
November 8, 2006 3:04:34 PM

Hey mad-dog, I'm currently restarting my 3DMark06 download using the 2nd TX mirror and its 65% done after 15min, so if your 3DMark05 download fails 1/2 way through like my 06 did, retry using the 2nd mirror.
a c 82 à CPUs
November 8, 2006 3:24:42 PM

Quote:
ok, when you break it down like that it all makes sense....
it was clear as mud before, THANK-YOU


I thought I'd tried to make it clear(ish) earlier, never mind I often get accused of over complicating things.

I'm actually looking forward to this data coming through.
November 8, 2006 3:45:03 PM

My 3DMark06 scores are in now, I used the free edition's standard setup. I was ranked 54th out of 119 systems similar to mine. Scores are as follows:

Overall: 5942

SM 2.0: 2635

HDR/SM 3.0: 2550

CPU: 1615

I would venture to guess that the only reason my scores are little lower than mad-dog's is because I'm using a DDR setup that has no Southbridge chipset dedicated to working the RAM :? .

----------------------

My System:

AMD 64 X2 3800+ OC'ed to 2.2 Ghz, BFG 7900 GTX 512 mb OC'ed, ASUS A8N-SLi Deluxe, 2 Gb Kingston Value RAM CAS Latency 2.5-3-3-6, Creative Labs Sound Blaster Audigy 4, Westen Digital 74 Gb Ratop ADFD, 2x Western Digital 250 Gb Caviar SE, 3x Thermaltake Hardcano 14 HDD coolers, Thermaltake Blue Orb II CPU cooler, Zalman VF-Cu 900 Blue LED VGA cooler, Thermaltake Extreme Spirit Northbridge cooler, and a PC Power & Cooling 510w PSU.
a c 82 à CPUs
November 8, 2006 3:55:55 PM

OK, so for comparison.

---------Spartan -13thMon-- Mad Dog
---------GTX x16 -GTO x4--GTO x16
06
SM2.0----2635 ---2411-----2756
SM3.0----2550 ---2143-----2677
CPU------1615 ---2035 ----1594
Overall---5942 ---5593----6142

05
-----------9350-----10,500---8146

Have stopped initial thoughts, need to have a think about how to interpret these. This could be saying that PCI-Ex4 gives about a 10% performance drop, or that spartans GTX is underperforming, or that Mad-dogs is overperforming. need to go away and think for a bit. Especially given the '05 scores.

Also need to figure out how to do lists/tables better. :?

It may not be possible to get much from this.

I'll just edit this for now when I get another result or two in.
November 8, 2006 4:07:11 PM

alrighty then, 8146 on '05
please keep in mind that my system is stock with no overclocking, the entire system is using default settings for comparative purposes.
November 8, 2006 6:17:25 PM

You need to add a zero to my 3DMark05 score to prevent others from being confused, we all know that there's been enough of that for a year :tongue: . I'm not for sure, and please read the article at Toms Hardware to check me (its 62 pages so take your time)( http://www.tomshardware.com/2006/05/23/amd_reinvents_it... ). The article is about the AM2 socket and how it compares to Intel in memory bandwidth usage. Now from what I read AMD's on-dye memory controller works with the Southbridge chipset to control the memory, Intel's version of this is just a larger L2 Cache. Now as I understand it the AMD on-dye controller works in tandem with the Southbridge to operate the memory, and at clock speeds higher than 2.4ghz AMD begins to beat Intel chips in the area of better memory perfomance. This could explain the discrepencies in the scores we have.

Now ASUS could have very well updated its architeture on its boards so that Hyper-Transport and DDR2 get along better, but I highly doubt this. One thing does come to mind though: on my ASUS board, if you insert 4 sticks of DDR-400 memory, the BIOS automatically under clocks the memory to DDR-333 and sets the Hyper-Transport bus to 2200mhz. This could very well be interfering with my VGA and causing the lower scores. Also, if anyone knows what normal is supposed to be :roll: , my PCI-x16 bus has a speed of 2500mhz. And another thing comes to mind also, my ASUS board has an on-board SLi selector (read: physical hardware) in addition to the requiered bridge for all SLi compatible VGA's.

I'll try to post my 3DMark05 scores as well for SM, HDR and CPU.

-------------------------

My System:

AMD 64 X2 3800+ OC'ed to 2.2 Ghz, BFG 7900 GTX 512 mb OC'ed, ASUS A8N-SLi Deluxe, 2 Gb Kingston Value RAM CAS Latency 2.5-3-3-6, Creative Labs Sound Blaster Audigy 4, Westen Digital 74 Gb Ratop ADFD, 2x Western Digital 250 Gb Caviar SE, 3x Thermaltake Hardcano 14 HDD coolers, Thermaltake Blue Orb II CPU cooler, Zalman VF-Cu 900 Blue LED VGA cooler, Thermaltake Extreme Spirit Northbridge cooler, and a PC Power & Cooling 510w PSU.
a c 82 à CPUs
November 8, 2006 6:33:06 PM

I don't understand the switch in ordering in '05 vs '06.

Perhaps they have increased the bandwidth CPU to GPU in the '06 tests, although from what I heard '06 was more processor intensive, and therefore the C2D shoud have had the biger advantage and PCI_ex4 the lesser disadvantage here.

If anyone else wants to add some scores, GTO's and GTX's only please.
November 8, 2006 6:43:47 PM

Could you expand on this? Do you mean ordering as in the way the results are listed?, or do you mean the order in which the tests are performed?

Also, I noted that there were advertisements for AGEIA in the load screens. Could this mean that 3DMark06 also tests your system for in game physics calcs?
!