More nVidia cheating?

Quote:

---

To not care is to live in ignorance... sorry, but it is the truth. I would want to know if ATI or nVidia is cheating... EVERY TIME they do it. This is important information and for those of us whom are intelligently speaking of a certain gfx card, we need to know what each company is up to to make/give a truly informed decision/recommendation.

---

I agree that we should care, and voice our disapproval of their practice, even though our voices may be drowned out by those who say they don't. If nVidia cheats, I wanna know.:mad: If ATI cheats, I wanna know too:mad: . That way, if we decide to buy a card from either company, we would, at the very least, make an informed decision. ;) Even then, I grow tired of all this flaming whenever a thread like is posted.:( I like ATI but if they cheat I'd wanna know 'bout it, that way if I choose to go with ATI or nVidia, then, I've made a informed choice based on personal preferences. After all these flamewars, people can get jaded and become cynical 'bout these vid card companies (Honestly, who can blame them?) Please no flaming, OK? :) Just airing my $0.02 worth.;):)

Quote:

---

Originally posted by Ashpool


Are you even aware of what we're talking about?

---

Yea, I am, I just worded it awkwardly. I meant to say nVidia is doing the exact same thing in Unreal Tournament 2003 with AF that ATI did two years ago in Quake III. They're lowering IQ without the user's consent.

You pay for a Porsche 911 Turbo and you get a normal 911 with a CD who play the turbo sound , the strategy is not bad :).
The bottom line is that they pulled out 2 generations videocards to compete with the 9700/9800 series and when they see the final disappointing result they desperately cheat , cut quality , overclock , cut FSAA , and etc only to barely be the winner .
They sould concentrate in good hardware like they did before and not in cheats and stuff like this , they never need cheats to make the GF2 , GF3 or GF4 a stellar videocard , cards had raw power in that time and the king was undisputed , now they are pathetic.

Quote:

---

Originally posted by AMD_Man


Yea, I am, I just worded it awkwardly. I meant to say nVidia is doing the exact same thing in Unreal Tournament 2003 with AF that ATI did two years ago in Quake III. They're lowering IQ without the user's consent.

---

What's our "standard" for AF quality? There is none, so it is hardly a cheat. If you want high image quality buy ATI, want the fastest FPS possible, buy Nvidia.Compared to the last generation of cards, the image quality is still better than it's been. Sort of like arguing about how fast your toaster, makes toast. You don't like it? Buy a different one.

Quote:

---

Originally posted by rhettro


What's our "standard" for AF quality? There is none, so it is hardly a cheat. If you want high image quality buy ATI, want the fastest FPS possible, buy Nvidia.Compared to the last generation of cards, the image quality is still better than it's been. Sort of like arguing about how fast your toaster, makes toast. You don't like it? Buy a different one.

---

Using the "AntiDetect" script, it was proven that the nVidia hardware is capable of better image quality, but they're compromising image quality for speed even though they don't make it explicitly clear in their driver settings. If you're going to consider the whole Quack issue a cheat, then nVidia is definitely also cheating in UT2003.

When it comes to games, if people don't notice IQ differences, they don't care. For most people, it is the final image that matters. (Synthetic benchmarks, they may care depending on what exactly the benchmarks are testing, but that's another story/thread).

The colorized thing, just highlights the exact filtering technique, in respect to being trilinear, nothing more. (As shown, in UT2K3 anisotropic filtering isn't highlighted by the colored bands for either card, so using the color band thing to judge overall IQ is flawed.)

The "quack" thing was really noticiable.

For this, on the beyond3d forum, they have to explicitly circle "differences" (I wouldn't be able to see the difference if it weren't circled). Thus, people don't care.

The real issue is the availability of "application" mode.

Quote:

---

Originally posted by rhettro
What's our "standard" for AF quality? There is none, so it is hardly a cheat.

---

Not exactly. Do you guys remember the original Nvidia FSAA cheat? You could select "8xFSAA" in the display controls and without telling you, it would drop back to 4x or 2x depending on the resolution you were running. So when the first benchmarks went up, they found no difference in framerate between the fsaa levels - reason: they weren't running at the specified levels. The difference isn't just the final output quality, the difference is what you THINK the card is doing compared to what it is actually doing. That can be considered nothing but a cheat.

In any case, this is off the topic of whether or not we should care. Palydr said it quite well:

Quote:

---

To not care is to live in ignorance... sorry, but it is the truth. I would want to know if ATI or nVidia is cheating... EVERY TIME they do it. This is important information and for those of us whom are intelligently speaking of a certain gfx card, we need to know what each company is up to to make/give a truly informed decision/recommendation.

---

Some people really DO prefer ignorance, hence the phrase "ignorance is bliss" (and its use in The Matrix) but I prefer knowing what is going on.

The kinds of cheats we have seen from both ATI and Nvidia in the past rise almost to the level of fraud. They are false advertising.

To borrow the car analogy, imagine the deciding factor in a car purchase decision for you was the 0-60 time. You check the company's websites and read the times, then buy the one with the faster 0-60 time. Then you try it for yourself and find that you don't get anywhere near the performance you were expecting. You do a little investigating and find that the car company did its 0-60 run downhill.

I think this analogy fits pretty well, but the problem is there are very few people outside of boards like this one who understand the parallels to the computer industry. Most people can spot and understand false advertising on a car, but very few people even understand video card specs and performance info.

Not to be too arrogant, but its acutally up to people like *US* and the writers of review sites like Sharky's to keep these companies honest. If we don't demand it, no one will even know what is going on.

Quote:

---

Originally posted by russ_watters

Not to be too arrogant, but its acutally up to people like *US* and the writers of review sites like Sharky's to keep these companies honest. If we don't demand it, no one will even know what is going on.

---

I agree. However, I also agree that this is getting old - simply because reports about Nvidia cheating don't seem to stop coming in. It really bothers me that they do things this way.

Timisorakill hit the nail on the head:

Quote:

---

The bottom line is that they pulled out 2 generations videocards to compete with the 9700/9800 series and when they see the final disappointing result they desperately cheat , cut quality , overclock , cut FSAA , and etc only to barely be the winner .

---

And there's more than enough evidence to suggest that this is the case. There have been several new reviews that show this clearly.

Nvidia should focus more on making their hardware faster, insteasd of focusing on creating the illusion of faster hardware.

-Ryan

Why do some of you insist on posting just to say this thread will be a flamewar, this thread is old, who cares, etc.?? Stop trolling the forums... I'm sick and tired of scrolling through posts who's sole purpose is to put down the topic or act all 133t... get over it. There have always been and always will be topics that you don't like or don't care about.

Quote:

---

Originally posted by russ_watters


To borrow the car analogy, imagine the deciding factor in a car purchase decision for you was the 0-60 time. You check the company's websites and read the times, then buy the one with the faster 0-60 time. Then you try it for yourself and find that you don't get anywhere near the performance you were expecting. You do a little investigating and find that the car company did its 0-60 run downhill.

---

Then this would be indicitive of poor research on your part. If you buy a product based on marketing fluff from the website, then you are deserving of the title of ignorant. I agree with the other posters, this topic is getting old. We can assume there is a standard when none exist, and then labast companies for their non-complience. I think that's a waste of time. It's the end result that I'm interested in. Remember, perception is the providence of the mind. What is more perceptable choppy framerate or slight texture blurring in game. Nvidia have made their choice. Nvidia can only undermine it's IQ so far until it becomes uncompetitive with what's available. In most cases the changes to IQ in game are inperceptable, but we will notice the smooth framerate. Again, if you do not like Nvidia's image quality, buy something else, Nvidia do not have a monopoly on the video card business.

Quote:

---

Originally posted by rhettro


Then this would be indicitive of poor research on your part.

---

Whoa there rhettro...
If a company lies in their advertizing, it's fraud. It's not "your fault" for not doing proper research. Because what they told you (0-60 speed) is based on a standard.

Quote:

---

Nvidia have made their choice. Nvidia can only undermine it's IQ so far until it becomes uncompetitive with what's available. In most cases the changes to IQ in game are inperceptable, but we will notice the smooth framerate. Again, if you do not like Nvidia's image quality, buy something else, Nvidia do not have a monopoly on the video card business.

---

No...Nvidia is not just undermining it's IQ. It's manipulating standard timedemos (most likely in much the same way as 3dmark) to give the impression of faster performance. Don't you think it's a bit odd that Nvidia's card loses it's apparant speed advantage when custom timedemos are used instead?
Since most people go by these timedeomo results as a deciding factor in their purchase, this is obviosuly an attempt to fool the customers into thinking their product performs better than the competition - when in most cases, this is the exact opposite of what's really happening (see some of the recent reviews out there).

If this were just image quality that they were fudging with, then I think this wouldn't be as big of a deal. But when scores are artificially manipulated and inflated, it tends to piss people off.

-Ryan

Quote:

---

Originally posted by ryandinan


And there's more than enough evidence to suggest that this is the case. There have been several new reviews that show this clearly.

Nvidia should focus more on making their hardware faster, insteasd of focusing on creating the illusion of faster hardware.

-Ryan

---

It's true, Nvidia haven't caught up with ATI yet. And they have tweaked their drivers to get more performance to make their card more marketable. But to say that all of Nvidia's tweaks are tie to image quality is jumping the gun in my opionion, otherwise ATI's performance mode would be competive with Nvidia's offerings. Last I checked Nvidia was planning on making faster video cards, but it's a little harder to fab new hardware than to write software.

Quote:

---

Originally posted by ryandinan
[B]

Whoa there rhettro...
If a company lies in their advertizing, it's fraud. It's not "your fault" for not doing proper research. Because what they told you (0-60 speed) is based on a standard.


No...Nvidia is not just undermining it's IQ. It's manipulating standard timedemos (most likely in much the same way as 3dmark) to give the impression of faster performance. Don't you think it's a bit odd that Nvidia's card loses it's apparant speed advantage when custom timedemos are used instead?
Since most people go by these timedeomo results as a deciding factor in their purchase, this is obviosuly an attempt to fool the customers into thinking their product performs better than the competition - when in most cases, this is the exact opposite of what's really happening (see some of the recent reviews out there).

If this were just image quality that they were fudging with, then I think this wouldn't be as big of a deal. But when scores are artificially manipulated and inflated, it tends to piss people off.

-Ryan

---

If Nvidia is guilty of fraud then I think you should sue them. I'll continue to get my information from multiple INDEPENDENT sources before I make a purchase. I'm not surprised at all about the lack luster performance of the custom demos, because they have obviously been tweaked. That's not the point, the point is would you rather have a ti4600 or a 5900 Ultra. The 5900 Ultra has better IQ than the ti4600 and more speed to boot. When a company's latest product is better than it's former then it is generally consider a success. I haven't read one review of the 5900 Ultra that says it is slow or ugly.

Quote:

---

Originally posted by rhettro


It's true, Nvidia haven't caught up with ATI yet. And they have tweaked their drivers to get more performance to make their card more marketable. But to say that all of Nvidia's tweaks are tie to image quality is jumping the gun in my opionion, otherwise ATI's performance mode would be competive with Nvidia's offerings. Last I checked Nvidia was planning on making faster video cards, but it's a little harder to fab new hardware than to write software.

---

I think we're on different pages here..
It's one thing to tweak your drivers to squeeze out that last bit of performance - as long as that performance is real and not just trickery of numbers (as is the case with the timedemos).

As far as Nvidia's image quality - I think it should be noted, that way people have the information to help them decide what to buy. If they want extra performance at the cost of IQ, then that's their decision - but that extra performance better be there, and not be some inflated number.
Unfortunately, this appears to not be the case with Nvidia. Not only do they offer sub-par IQ when compared to ATI, the performance is at best equal, if not lower than ATI in the majority of the acclaimed game tests.

If consumers are given the straight dope, the decision is much more black and white. Here's the senario:

Card A offers performance equal to or greater than Card B, offers better IQ than card B, and costs less than card B.

Card B offers performance equal to or less than card A, offers inferior IQ than card A, and costs more than card A.

From a price/perfoamane standpoint, the decision is clear. However, due to Nvidia's recent practices, the above aspects have been clouded, and even falsified. If these things weren't discovered, Nvidia would appear to have a card that performs better. This is quite a bit different than 'tweaking' the drivers.

-Ryan

While no one has flamed anyone (thanks for that), I think this thread has outlived its useful life and now everything is just a rehash of what has already been said.