Tech-Report comments on driver optimization saga

kemosabe

Veteran
"There is another important issue here, of course, and that's trust. Yes, the API provides a little wiggle room, but NVIDIA is flaunting the situation in multiple and evasive ways in order to make its products look better than they are. All indications are that NVIDIA deceived reviewers—and by proxy consumers, investors, and PC makers—when it told them the 44.03 Quality setting would do away with the funky "lightweight" trilinear method, then enabled that method on a per-application basis anyway. They dug the hole deeper when they claimed a 30% performance increase in UT2003 on the GFFX 5600. And this all happened at the same time as they were concocting multiple methods of inflating scores in 3DMark03 while waging an all-out PR assault on 3DMark's creators. I would think twice before putting my hard-earned cash down for an NVIDIA product in the midst of all of this deception."

http://www.tech-report.com/onearticle.x/5418
 
Isn't it interesting how many sites are being approached by nVidia and invited to meetings... Anyone would think they had a new product launching later this year... :devilish:
 
The are launching at least 3 more desktop ASICs plus two IGP/core logic solutions later this year, IIRC.

MuFu.
 
Must admit tech reports article is the best i have seen on the issue.

One thing interesting to note though...Brutally honest reviews without direct vendor influence and formalities is one of the main reasons web based HW review journalism took over the print based one in late 90's. Now, we see all top websites pretty much settle down to the same level as print media...Only a bit verbose and with too many charts because space is not an issue.

Sites like techreport,beyond3d etc in my opinion are the new wave of the future...We can tag THG,AT,HOCP and ET along with PCMag,PCGamer etc etc
 
Mighty frisky of 'em. I'm guessing we're going to see an Opteron MB with an IGP? As for the three desktop ASICs: 5700, 5900 Value, and NV40?
 
MuFu said:
The are launching at least 3 more desktop ASICs plus two IGP/core logic solutions later this year, IIRC.

MuFu.

And what about ATi? :D

wouldn't you like to tell us?

please, oh please!

Will ATi unveil their R360 and RV360 with the Radeon 9200 SE (end of July, right)?
 
Pete said:
Mighty frisky of 'em. I'm guessing we're going to see an Opteron MB with an IGP?

Yep - "nForceFX", I think.

As for the three desktop ASICs: 5700, 5900 Value, and NV40?

5700, 6000/6900 and NV40 (unless it gets pushed back more). There's NV34B and NV18C as well, but I have no idea which sectors they are targetting - probably the POS and mega-POS markets.

MuFu.
 
Does UT2003's texture filtering return to normal if it is named, say, GoofyFun.exe? I'm curious as to whether it's a cheat or a severely busted optimization (since it seems that the 44.03 drivers were quite rushed for the NV35 launch).

Of course, it's nothing short of fucking amazingly irresponsible to use UT2003 in the reviewer's guide and not adequately test it if it is a bug.

But if it was a cheat, it doesn't seem hidden enough. If it's exposed through filename alone, unlike other "optimizations" (hence the need for Antidetect)... it's like NVIDIA got lazy.
 
DaveBaumann said:

Hehe, over time, I've really began to like your answer type, Dave :)

In relation to what MuFu said: Also note that those are ASICs - nV will release other products this year based on current ASICs ( and one or two ASICs are also pontentially for this year, potentially for next and, well, maybe will never appear on the market, eh )


Uttar
 
MuFu said:
Pete said:
Mighty frisky of 'em. I'm guessing we're going to see an Opteron MB with an IGP?

Yep - "nForceFX", I think.

As for the three desktop ASICs: 5700, 5900 Value, and NV40?

5700, 6000/6900 and NV40 (unless it gets pushed back more). There's NV34B and NV18C as well, but I have no idea which sectors they are targetting - probably the POS and mega-POS markets.

MuFu.
GeForce 5 6000 :LOL:

And the 5700 can be the GeForce 5 5500. :D
 
Is there any chance it's a bug introduced by an optimization/bugfix that detects the .exe name?

Is this like the clip planes and buffer clearing (or not clearing or whatever) in 3dmark03 that almost had to have been done by hand?

Just wondering how likely that it's a bug, although right now it's pretty much guilty until proven innocent for nvidia with regard to any IQ loss/major performance increase, lol.
 
Interesting how RV360 has been discussed for months, and yet we've heard less reliable speculation about its specs than what's been floated about NV36 in the last few days alone. Seems NVIDIA enjoys the roadmap leak approach a bit more. :LOL:
 
Interesting how RV360 has been discussed for months, and yet we've heard less reliable speculation about its specs than what's been floated about NV36 in the last few days alone. Seems NVIDIA enjoys the roadmap leak approach a bit more.
Thats becuase there is little to discuss with the 360's. Anyone with a tad bit of controlled imagination can figure out what it is.

What is interesting now is to see what happens when R360/nv38 face off in the biggest game since well.. was there ever such a big game??? :eek:
 
kemosabe said:
Interesting how RV360 has been discussed for months, and yet we've heard less reliable speculation about its specs than what's been floated about NV36 in the last few days alone. Seems NVIDIA enjoys the roadmap leak approach a bit more. :LOL:

Heh...is any "speculation" reliable?....;)
 
Thats becuase there is little to discuss with the 360's. Anyone with a tad bit of controlled imagination can figure out what it is.

Controlled imagination? Wassat? :D

I tend to agree about R360, but is everyone fairly convinced that RV360 will offer nothing beyond a speed bump? No HierZ or other hardware optimizations? I am presuming ATI will want the huge performance gap between RV350 and R350 to be bridged somewhat with RV360/R360.

LOL, Walt, I suppose "speculation from a reliable source" sounds more à propos than "reliable speculation".
 
kemosabe said:
Interesting how RV360 has been discussed for months, and yet we've heard less reliable speculation about its specs than what's been floated about NV36 in the last few days alone. Seems NVIDIA enjoys the roadmap leak approach a bit more. :LOL:

Standard Nvidia spoiler - spread some FUD via leaks, and hope that people hang onto their cash to see what the new Nvidia cards are like, rather than spend money on a competitors product.

We saw this right from the first hints of NV30, all the way through its delayed production, delayed introduction, late ship, and final cancellation. It's still going on now.
 
We saw this right from the first hints of NV30, all the way through its delayed production, delayed introduction, late ship, and final cancellation. It's still going on now.

Certainly that's the label they're getting plastered with, but I don't think ATI would be wise to expect a repeat performance with future hardware.
 
Back
Top