NVIDIA 60.72 Fiasco

[Kombatant]

Complain if it's there when the boards are released.

So it's OK if it's there during the time that sites benchmark the card and formulate people's opinions, but it's removed when the card hits retail?

 

[Nick Spolec]

Complain if it's there when the boards are released.

Yeah, by the time the 6800 goes to retail, Nvidia should have hex-detection ready to go So renaming will be useless.

 

[Snarfy]

Complain if it's there when the boards are released.

i just think its fishy that all of a sudden, all these new problems are cropping up in farcry with the 61.11 drivers...

i mean, nvidia is supposed to have AWESOME drivers, right? so why is it that all of a sudden, all of these issues pop up:

117597 [TWIMTBP]NV40,WinXP: Corruption when trying to take a screenshot with AA enabled 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 [TWIMTBP]NV38/40-WinXP: Lighting/shadow problem in FarCry.

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

For a company who's biggest selling point is its allegedly 'great drivers', either they are slouching majorly or they are trying to stack things in there favor? Why else would they release 61.11 at the time they did, with MORE problems than 60.72? It boggles my mind!

 

[KimB]

So it's OK if it's there during the time that sites benchmark the card and formulate people's opinions, but it's removed when the card hits retail?

As long as the optimizations used have become general by that time, giving the same or better performance.

 

[Snarfy]

So it's OK if it's there during the time that sites benchmark the card and formulate people's opinions, but it's removed when the card hits retail?

As long as the optimizations used have become general by that time, giving the same or better performance.

Yeah, some optimizations. They gut the image quality...

 

[Snarfy]

double post, sorry

 

[KimB]

i mean, nvidia is supposed to have AWESOME drivers, right? so why is it that all of a sudden, all of these issues pop up:

It's a new card. Drivers are always relatively raw with a brand-new architecture.

Oh, and the R420 is much, much more similar to the previous generation than the NV40, so its drivers are bound to be more mature. I'm more interested in comparing the NV40's drivers to the R300's closer to release.

 

[KimB]

Yeah, some optimizations. They gut the image quality...

Um. Where have you been? The image quality problems on specular highlights appeared on all nVidia cards with the 1.1 patch. It's a game issue.

 

[Nick Spolec]

Oh, and the R420 is much, much more similar to the previous generation than the NV40, so its drivers are bound to be more mature. I'm more interested in comparing the NV40's drivers to the R300's closer to release.

Gee.. I don't know. How long do you think Nvidia has been working on the NV40 and it's drivers? Do you not think they thoroughly tested it? And FarCry is a TWIMTBP game..

I find it very suspicious that the beta "review" drivers contain bugs that just so happen to raise performance on what could be summed up as the biggest graphical water-mark game currently out.

 

[Kombatant]

i mean, nvidia is supposed to have AWESOME drivers, right? so why is it that all of a sudden, all of these issues pop up:

It's a new card. Drivers are always relatively raw with a brand-new architecture.

Oh, and the R420 is much, much more similar to the previous generation than the NV40, so its drivers are bound to be more mature. I'm more interested in comparing the NV40's drivers to the R300's closer to release.

The architecture is not THAT new to warrant immature drivers. And besides, I believe our topic of discussion here was questionable "optimizations" and not driver immaturity.

 

[Snarfy]

i mean, nvidia is supposed to have AWESOME drivers, right? so why is it that all of a sudden, all of these issues pop up:

It's a new card. Drivers are always relatively raw with a brand-new architecture.

Oh, and the R420 is much, much more similar to the previous generation than the NV40, so its drivers are bound to be more mature. I'm more interested in comparing the NV40's drivers to the R300's closer to release.

The architecture is not THAT new to warrant immature drivers. And besides, I believe our topic of discussion here was questionable "optimizations" and not driver immaturity.

Yah, Kombatant is a catalyst beta tester, so he knows a thing or two about drivers 8)

 

[KimB]

Gee.. I don't know. How long do you think Nvidia has been working on the NV40 and it's drivers? Do you not think they thoroughly tested it? And FarCry is a TWIMTBP game..

Where have you been? I've seen drivers on first-release cards. I've owned a few of those cards. Drivers are always raw. They are always buggy. They are always slow.

Really good drivers don't come about until ~6 months or so after initial release.

I find it very suspicious that the beta "review" drivers contain bugs that just so happen to raise performance on what could be summed up as the biggest graphical water-mark game currently out.

I'm not saying this is the case. I'm saying that nVidia used a "quick fix" on the bugs for these first drivers, something that should become more robust with future releases.

 

[stevem]

Actually, I wonder how optimized either IHVs' drivers are...? For Nvidia it's a given, but although ATI have experience with a 2 quad design, how optimal is the load balancing for a 4 quad design or (possibly worse) a 3 quad, non power of two design...?

 

[Joe DeFuria]

Oh, and the R420 is much, much more similar to the previous generation than the NV40, so its drivers are bound to be more mature. I'm more interested in comparing the NV40's drivers to the R300's closer to release.

So what you're saying is, nVidia's driver suck, and ATI's are good. OK.

 

[KimB]

Actually, I wonder how optimized either IHVs' drivers are...? For Nvidia it's a given, but although ATI have experience with a 2 quad design, how optimal is the load balancing for a 4 quad design or (possibly worse) a 3 quad, non power of two design...?

It wouldn't be any different at all to simply add pipelines. They're completely parallel.

The difference in the drivers will come down to shader execution.

 

[KimB]

So what you're saying is, nVidia's driver suck, and ATI's are good. OK.

I'm saying that ATI's had a long time to work on the R3xx drivers. nVidia hasn't had much time to work on the NV4x drivers.

The R420 really would be more accurately called the R380 or something similar.

 

[KimB]

The architecture is not THAT new to warrant immature drivers. And besides, I believe our topic of discussion here was questionable "optimizations" and not driver immaturity.

Not yet available isn't new? Since when?

 

[Snarfy]

this isnt about pipes or clocks or transistors, fellas...

its about how nvidia sent out the 61.11 drivers to testers, knowing full well that they had MORE bugs than the 60.72 version in order to steal some of Ati's thunder by lowering the image quality in order to up the performance of their card.

Hey, look here:

That was the 60.72 drivers running far cry. you can SEE the blockiness in the screenies.

then, they release the 61.11 drivers adding on some MORE iq degredation!

117597 [TWIMTBP]NV40,WinXP: Corruption when trying to take a screenshot with AA enabled 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 [TWIMTBP]NV38/40-WinXP: Lighting/shadow problem in FarCry.

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

THEN, adding insult to injury, that german site tries something interesting, and hey, guess what!

when they rename farcry.exe to fartcry.exe, SHAZAM! performance with the 61.11 drivers drops dramatically!

all of this, PLUS nvidias past record of doing the SAME things when the 5800u sucked...

Why, Nvidia, why

 

[stevem]

It wouldn't be any different at all to simply add pipelines. They're completely parallel.

The difference in the drivers will come down to shader execution.

I'm not convinced it's that simple. The complexity of keeping all the pipelines as busy as possible & maximizing IPC seems challenging. Scheduling will be a key task. The driver/compiler teams are going to be very busy...

 

[Joe DeFuria]

I'm saying that ATI's had a long time to work on the R3xx drivers. nVidia hasn't had much time to work on the NV4x drivers.

...And as a consequence, ATI's drivers are good, and nVidia's suck. Right?

The R420 really would be more accurately called the R380 or something similar.

Whatever makes you happy. The NV40 should actually be called the NV30...becuase it would be best for everyone concerned if the NV3x never existed at all and was erased from memory.