Wife's vid card crapped out the other night, need to replace it with something that'll play her beloved Sims 3 at decent resolution on her 22" monitor. Torn between these two current choices at newegg:
Thanks for the feedback all, ended up going for the 9600 GSO because of the price, hard to beat $40 after rebate, and it leaves a bit more funds in the budget for my next vid card upgrade for MY OWN machine - you know, the one that counts!
Sorry to dig up an old thread, wanted to add something. I found this thread because the egg is sold out of my beloved ASUS 9600 GSO (I have 2) and I needed another video card. I wasn't sure about getting the EVGA 9600 GSO 1GB DDR2 compared to the ASUS (DDR3 512MB), and I searched for comparisons.... I had already ordered the EVGA... wasn't sure if I did right.
Google ranked this thread pretty high... so here I am.
Longer story short... I did a comparison of the cards on the same box, identical drivers, using 3DMark06. For the price, the EVGA or the ASUS will run any games with decent resolutions for the average gamer.... yes, if you want to spend hundreds more, a different card will bench way better.
My benchmarks with both cards at stock clocks were virtually identical at 7666points for the ASUS and 7639 for the EVGA.
Mildly OC'd is where it really changed, the DDR3 was able to easily push harder to 8503 points and the EVGA started going sour at 8328 points (still, virtually identical scores....)
ASUS OC'd 627/1598/1568 (Still had headroom) 8503 points
EVGA OC'd 701/1752/2004 (Pretty much topped out) 8328 points
Conclusion: They are the same card, and the memory makes very little difference.
So, a DDR3 card outperforms a much higher clocked DDR2 card (seriously, the RAM is clocked 1/3 higher on the DDR2 card) while still having overclock headroom and they are pretty much the same Sorry, I think your conclusions are very flawed.
------------------------------If the automobile had followed the same development cycle as the computer, a Rolls-Royce today would cost $100, get a million miles to the gallon, and explode once a year, killing everyone inside.
PSA
Reply to outlw6669
I don't get it. Did you purposely not OC the ASUS card any higher so the scores would work out to be similar?
I gave them both a rough guesstimate of what worked. The EVGA puked. The ASUS (DDR3) still perhaps had headroom... They were what I felt was a rough OC ratio over stock. (doing the math, the EVGA was 25% over the memory, while the ASUS was 12%) The engine was ~15% for both. So not quite the same OC... my bad.
As far as your comment... the ASUS got a better score at a lower clock and a lower % OC.... for the win.....
I'd say they are very similar..... YMMV depending on OC. They are identical STOCK..... YMMV
I purposely did not OC the ASUS any higher because I was not after the limit of the card.... and to make people offer wild assumptions.
Message edited by JonnyQ on 11-05-2009 at 09:37:18 PM
Well it seems like the proper conclusion is that they perform similarly at stock settings(stock in this case being that one of the cards has a factory OC) but the DDR3 card has a definite advantage once you OC. And they engines aren't on the same OC, they both use the same chip so the same OC would be the same speed. Just because EVGA factory OCed the card doesn't mean the ASUS isn't just capable of the exact same OC.
It also matters what resolution is used. Even at stock the DDR3 card should have a much larger memory bandwidth. At low resolutions that may not come into play but at higher ones it's likely pretty important.
So, a DDR3 card outperforms a much higher clocked DDR2 card (seriously, the RAM is clocked 1/3 higher on the DDR2 card) while still having overclock headroom and they are pretty much the same Sorry, I think your conclusions are very flawed.
Ummm, yeah.... That extra 27 points will make a huge difference gaming..... STOCK.
The difference between the two cards is huge, as far as memory clocks... sure. The scores are still similar. The DDR2 was higher clocked STOCK. How is that conclusion FLAWED?
My point was that people here made assumptions that there was a HUGE difference between these two cards.... I would assume STOCK, as the guy was using it in his wifes computer.....
If I get time, I'll throw the ASUS back in and run it higher, my guess it'll score 9000.... but I doubt it'll hit mem speed of 1800 for the same 1/3 over... but who really can make a comparison of OC'd GPU's and expect that to be the norm in that line of cards?
YMMV, but it'll be nearly the same STOCK.
Perhaps no one would be upset and claim my testing faulty had I not OC'd.... (Remember... they scored virtually identical STOCK)
Well it seems like the proper conclusion is that they perform similarly at stock settings(stock in this case being that one of the cards has a factory OC) but the DDR3 card has a definite advantage once you OC. And they engines aren't on the same OC, they both use the same chip so the same OC would be the same speed. Just because EVGA factory OCed the card doesn't mean the ASUS isn't just capable of the exact same OC. It also matters what resolution is used. Even at stock the DDR3 card should have a much larger memory bandwidth. At low resolutions that may not come into play but at higher ones it's likely pretty important.
Good point, I never considered using the same engine clock as a baseline...... Resolutions were the same for testing, stock 3DMark06.... which is low by todays standards.
My original point was intended to compare stock cards...... (1 GB DDR2 Vs 512MB DDR3) I sometimes stray from the intended path.
Message edited by JonnyQ on 11-05-2009 at 10:06:12 PM
A stock clocked card runs at nVidia's Specified Speeds (also Directly from nVidia).
For a g92 based 9600GSO (96Sp variant) that would be 550Mhz core, 1375Mhz shader and 800Mhz (1600Mhz DDR, DDR3 cards) or 500Mhz (1000Mhz DDR, DDR2 cards) for the RAM.
A factory overclocked card is still overclocked.
If you want to do a real comparison, clock both cards exactly the same and see what happens.
Saying both cards perform nearly the same because the weaker one is overclocked by EVGA = flawed thinking.
------------------------------If the automobile had followed the same development cycle as the computer, a Rolls-Royce today would cost $100, get a million miles to the gallon, and explode once a year, killing everyone inside.
PSA
Reply to outlw6669
A stock clocked card runs at nVidia's Specified Speeds (also Directly from nVidia).
For a g92 based 9600GSO (96Sp variant) that would be 550Mhz core, 1375Mhz shader and 800Mhz (1600Mhz DDR, DDR3 cards) or 500Mhz (1000Mhz DDR, DDR2 cards) for the RAM.
A factory overclocked card is still overclocked.
If you want to do a real comparison, clock both cards exactly the same and see what happens.
Saying both cards perform nearly the same because the weaker one is overclocked by EVGA = flawed thinking.
Screwed thinking is buying two cards, UNDERclocking one of them and making a comparison.....
I am not here to please you.
I did a real comparison, they are virtually identical... STOCK... as in "out of the box"... like a real person would obtain them, as in STOCK form... clear enough?
------------------------------If the automobile had followed the same development cycle as the computer, a Rolls-Royce today would cost $100, get a million miles to the gallon, and explode once a year, killing everyone inside.
PSA
Reply to outlw6669