PDA

View Full Version : Leaked benchmarks - R420 vs. NV40


Pages : [1] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

radunovicL
05-03-04, 05:11 PM
http://www.beyond3d.com/forum/viewtopic.php?t=12059

It's interesting. r420 pro and XT beats r420. Tom used the new 60.xx drivers and

Known 61.11 driver bugs:

117597 [TWIMTBP]NV40,WinXP: Corruption when trying to take a screenshot with AA enabeld in Far Cry.

117561 [TWIMTBP]NV38/40,WinXP: Corruption of some weapons in Far Cry.

117474 [TWIMTBP]NV40,WinXP: Fog broke in Far Cry.

113476 NV38/40-WinXP: Lighting/shadow problem in FarCry.

117575 [TWIMTBP] NV40/38: Banding visible from wall in FarCry

If this all true, but i guess it is, NVIDIA is nothing but bunch of morons and losers...

radunovicL
05-03-04, 05:12 PM
I meant r420 pro and XT beats NV40 (Far Cry)...

radunovicL
05-03-04, 05:16 PM
This is normal quality of Far Cry. I guess the biggest moron is Tom...

Casper
05-03-04, 05:21 PM
Yes Tom prooves time after time he don´t no **** about hardware. Like when he tested the A64 with crap memory :rolleyes:

Someone was kind and link this from that thread also quite hilarious in retroperspective :)

http://www17.tomshardware.com/graphic/20030127/geforce_fx-29.html

radunovicL
05-03-04, 05:24 PM
This is all unofficial, but seems that 61.11 is the biggest cheating NVIDIA did so far. I was saying here for long time, that FP24 will be the key in the feature, and thatn NV40 still has poor shading performance. Why the hell they don't use FP32 in FAR CRY?

Edge
05-03-04, 05:27 PM
If this all true, but i guess it is, NVIDIA is nothing but bunch of morons and losers...

What the hell? That's a pretty idiotic statement. Why are they morons and losers? Because an UNOFFICIAL, UNRELEASED driver has graphic corruption issues with one game (most of which have NOTHING to do with performance, I highly doubt corrupt screenshots with AA enabled is somehow magically speeding the game up)? I'd hate to see what you think of ATI if you're judging the entire company with issues in one game using unofficial drivers.

Also those benchmarks are pretty useless since they're almost certainly CPU limited. I'd like to see some REAL benchmarks that use AA/Aniso and higher resolution. When the 9800xt comes within 10% of the x800 XT, you know something's up. And umm...why does the FX5950 have a higher min. framerate than the 9800xt?

Casper
05-03-04, 05:28 PM
Because they use the NV3X optimizations and will continue to do so in default mode. Hopefully they will allow users to disable them in the feature though unless they gather it will hurt sales of course.

radunovicL
05-03-04, 05:28 PM
Anand should be better, and I'm going to skip over Tom's BS...

lIqUID
05-03-04, 05:29 PM
whats even less interesting is the bench is obviously CPU limited and tells us NOTHING about one card beating another, now go troll along :lame:

radunovicL
05-03-04, 05:32 PM
What the hell? That's a pretty idiotic statement. Why are they morons and losers? Because an UNOFFICIAL, UNRELEASED driver has graphic corruption issues with one game (most of which have NOTHING to do with performance, I highly doubt corrupt screenshots with AA enabled is somehow magically speeding the game up)? I'd hate to see what you think of ATI if you're judging the entire company with issues in one game using unofficial drivers.

Also those benchmarks are pretty useless since they're almost certainly CPU limited. I'd like to see some REAL benchmarks that use AA/Aniso and higher resolution. When the 9800xt comes within 10% of the x800 XT, you know something's up. And umm...why does the FX5950 have a higher min. framerate than the 9800xt?

They released 61.11 with the purpose. They are morons, because they know for bugs (actually it's not bug, they keep nv30 with the purpose) and they can fix it, but guess what? Having FP32 in Far Cry would make 6800Ultra to look like joke...

Dude I would not compare NV35 with r360 at all, because NV35 crap does PS1.1 in Far Cry mixed with PS2.0 FP16, but r360 does PS2.0 all the time having FP24...

radunovicL
05-03-04, 05:33 PM
whats even less interesting is the bench is obviously CPU limited and tells us NOTHING about one card beating another, now go troll along :lame:

Wait for tomorrow :lol2:
I'm going to laugh...

CaiNaM
05-03-04, 05:39 PM
http://www.beyond3d.com/forum/viewtopic.php?t=12059

wow.. the xt scores sure suck, don't they? higher clocks * 4 more pipes = no difference? it even loses in "breed"... rather disappointing.

lIqUID
05-03-04, 05:40 PM
If my life revolved around a videocard I might laugh as well, but maybe Ill just laugh at you for caring so much. :screwy:

Casper
05-03-04, 05:42 PM
wow.. the xt scores sure suck, don't they? higher clocks * 4 more pipes = no difference? it even loses in "breed"... rather disappointing.

Look at the settings. Normal quality 1024x768 no aa and anisotrophic = very nice CPU benchmark...

It don´t tell anything really and of course the x800 with it´s higher clock ought to having a higher max fps. What is more interesting is the lowest.

But then you can´t really compare it fairly anyway thanks to the drivers they used for the NV40.

That is a 100 % useless benchmark really. It don´t tell you anything.

SH64
05-03-04, 05:43 PM
If this all true, but i guess it is, NVIDIA is nothing but bunch of morons and losers...

easy there cowboy .. its just a leaked benchmark .. remember the leaked NV40 benchies from [H] ?? & how we draw million of conclusions around it & finally it proved to be not correct in many cases .

CaiNaM
05-03-04, 05:46 PM
Look at the settings. Normal quality 1024x768 no aa and anisotrophic = very nice CPU benchmark...

It don´t tell anything really and of course the x800 with it´s higher clock ought to having a higher max fps. What is more interesting is the lowest.

But then you can´t really compare it fairly anyway thanks to the drivers they used for the NV40.

That is a 100 % useless benchmark really. It don´t tell you anything.

heh, cas... you miss my point.

photophreak314
05-03-04, 05:46 PM
How did Tom's get an X800 XT, if they are going to be released a month after the X800 pro? I'd say it's all fabricated and :lame:

Lets all just wait and see.

CaiNaM
05-03-04, 05:46 PM
easy there cowboy .. its just a leaked benchmark .. remember the leaked NV40 benchies from [H] ?? & how we draw million of conclusions around it & finally it proved to be not correct in many cases .

logic is irrelevant when trolling ;)

CaiNaM
05-03-04, 05:47 PM
How did Tom's get an X800 XT, if they are going to be released a month after the X800 pro? I'd say it's all fabricated and :lame:

Lets all just wait and see.

no.. frankly it seems the xt is easier to get then pro as far as pre-production eval hardware...

photophreak314
05-03-04, 05:48 PM
Dude I would not compare NV35 with r360 at all, because NV35 crap does PS1.1 in Far Cry mixed with PS2.0 FP16, but r360 does PS2.0 all the time having FP24...

Seeing as most of the shaders are 1.1 anyways on both cards, I'd say "R360 does PS2.0 all the time having FP24" is wrong, as it's doing mostly 1.1, with some 2.0 FP24, while nvidia is doing mostly 1.1 and some 2.0 FP16

lIqUID
05-03-04, 05:48 PM
How did Tom's get an X800 XT, if they are going to be released a month after the X800 pro? I'd say it's all fabricated and :lame:

Lets all just wait and see.

OMG! I didnt even think about that LOL.... FAKE!

jAkUp
05-03-04, 05:52 PM
XT will be released in 2 weeks. Those scores are CPU limited and say 0. Trust me guys, I will have all my info at 12 tonight... Including some rough FarCry benchmarks at 1600x1200..... 8x/8x. Trust me guys, you dont wanna take this graph realistically...

nForceMan
05-03-04, 06:01 PM
It's interesting. r420 pro and XT beats r420.

If this all true, but i guess it is, NVIDIA is nothing but bunch of morons and losers...

I was saying here for long time, that FP24 will be the key in the feature, and thatn NV40 still has poor shading performance.

Dude I would not compare NV35 with r360 at all, because NV35 crap does PS1.1 in Far Cry mixed with PS2.0 FP16, but r360 does PS2.0 all the time having FP24...
You are definitely one of the biggest ignorant morons I've ever met. All your statements are just pure :bs:

jAkUp
05-03-04, 06:05 PM
Just wait till you see the 1600x1200 scoress. ;)

SH64
05-03-04, 06:12 PM
Just wait till you see the 1600x1200 scoress. ;)

You sound that you know something we dont know :wtf: ??

do tell .. do tell :D