anyone feel the same? 3dmark... newer gets higher score

[HiddenStupid]

anyone feel the same? 3dmark... newer gets higher score

go to tomshardware chart and shows that 3dmark.... the newer card gets higher score. say like the 8600gts vs. 7800gtx.... when look at indivdual particular game bench.... shows the 7800gtx higher score.... but when look at 3dmark.... 8600gts gets so higher score than 7800gtx.

make sme dont believe in 3dmark or any overall bench.... and ive been not believing in bench.... well only in actual in-game bench but not any machine generated number bench.

numbers and scores is nothing.... only real-time in-game bench. ff11 bench are good. fear bench are good. 3dmark bench i no trust.

 

[AphexDreamer]

I think you can dude. Its just like running FEAR's bench, except a different game. The benches use different ways to stress hardware. Show me where the 7000 Series gets more FPS in games then the 8000 Series.

 

[candle_86]

that actully makes sense, it why the G84 could out preform the G71 in Rainbow 6 Vegas also. Shaders, and shader speed, any game shader dependate over bandwith or rops will always run faster on the these cards. 3dmark06 is very shader dependant.

 

[HiddenStupid]

I think you can dude. Its just like running FEAR's bench, except a different game. The benches use different ways to stress hardware. Show me where the 7000 Series gets more FPS in games then the 8000 Series.

here is when the 7 is higher than the 8
http://www.tomshardware.com/charts/desktop-vga-charts/doom-3,560.html?p=1571,1587,1647,,

here is when the 8 is higher than the 7
http://www.tomshardware.com/charts/...1-0-2-sm2-0-score,545.html?p=1571,1587,1647,,

 

[HiddenStupid]

that actully makes sense, it why the G84 could out preform the G71 in Rainbow 6 Vegas also. Shaders, and shader speed, any game shader dependate over bandwith or rops will always run faster on the these cards. 3dmark06 is very shader dependant.

so the 8600gts has higher shader than the 7800gts? and rainbow six vegas uses alot of shader power? and yeah vegas does look like a shader game....

overall the 7800gtx is a faster card but inferior shader? and 8600gts is superior shader but slower card? 3dmark bench tech is in lots of shader? and thats why 8600gts shows greater score? confirmation would be great.

 

[candle_86]

yes the 7900GTX has 24shaders the 8600gts has 32shaders

 

[HiddenStupid]

yes the 7900GTX has 24shaders the 8600gts has 32shaders

no i mean the 7800gtx not the 7900gtx

 

[candle_86]

yes same shader count on both 24 shaders on 7800GTX and 7900GTX

 

[AphexDreamer]

hmmm. Another explanation might be that the games were programed to be more optimized on the generation of cards during the time period the games were released. Im sure if you were to compare the two cards on a much more modern game, the results would be quite the opposite.

 

[philbrown23]

yeah, and the clock differemce, and so much goes into synthetic benches, also those gamebanches were on max fps and sure the 7800gtx has a higher max fps but I guarantee that it does not hold that fps long, also thats kinda an unfair comparsion to base belif in benches on because the 8600gt would smoke the 7600gt and the 8800gtx would smoke the 7800gtx, basically those cards are about the same with different cores and a few more shaders here or there.

 

[Kursah]

Synthetic benches should be taken with a grain of salt anyways, unless that's what you compete in or build your rig specificly for (the you're competeting most likely)...the gaming results should matter. Most know that newer cards may have less specs than the card they replace, thus game worse, or have similar specs and better optimizations and game better.

Shaders have become more and more important "en masse", but not to the ATI extent in games, how the process is and how fast are also extremely improtant, so, more complicated shaders with higher speeds are still winning the battle on many gaming fronts in the NV vs. ATI "wars". Both companies have excellent releases that are pretty damn competetive in price and performance, and yet both still take different directions which is great in my mind. In the end, I could care less what 3DMark says, and would rather pay attention to how games I play are impacted or how they benefit from what I'm planning on purchasing, as both aren't necessarily linked.

 

[Palit_Guy]

@Kursah

I have been saying that since 3DMark01 and I couldn't agree with you more. Unless you play 3DMarkXX what does anyone care what score you get?

If you look at 3DMark scores and rank cards in order of their performance and then take those same cards and bench them on real games you will see that the order in which they perform is different.

In fact, I have yet to see any single game that aligns perfectly with performance levels on any game. Sometimes the difference in relative performance is really large but often it's just the difference of just one or two models.

 

[Focus182]

most 8600GTS is 512mb well mine is anyway and they are showing u a 256mb .. oh well xD it seems that the 8600GTS can perform better when its stresed more .. you run oblivion on 1024x768 and the 7800GTX will probly beat it .. but then u look at 1920x1600 and the 8600GTS will probly beat it

~Focus

 

[tkpenalty]

I honestly think tom's hardware can be taken with a grain of salt as well these days... Since the original owner sold it.