PDA

View Full Version : No mention of 'it' from Anand


Pages : [1] 2 3 4

Typedef Enum
05-16-03, 09:48 AM
Interesting.

To this point in time, Anandtech hasn't mentioned the story that pretty much made the rounds everywhere else.

Why?

What's more interesting is the fact that he was just getting ready to unleash that "DetonatorFX" article @ the end of the week when this whole thing exploded.

I find that rather odd. So, they don't mention it at all...they were supposed to release an article detailing this spiffy new driver. If there's nothing to this whole story in the first place, why didn't he release his Detonator article?

Me thinks the guy took a step back, and began to actually take a freakin' _look_ at the screen for once while doing these articles, and began to question his findings...Ya think?

surfhurleydude
05-16-03, 09:51 AM
No, I just think that Anand is always late with his articles. Some of the round-ups Anand promotes in previews never even come out. I can't really say I'm real surprised!

Typedef Enum
05-16-03, 10:32 AM
Not according to his 5900 preview.

Later this week we will bring you a comparative review detailing the performance of all of the anisotropic filtering modes as well as a better idea of exactly how much performance the new Detonator FX drivers will bring you.

Well, it's definitely "later this week' and this preview is nowehere to be found.

I'm positive that this article would have gone out the door in/around the official day it was released...but very likely, it was held back due to the issue that was uncovered by ExtremeTech.

As I said in that other monster thread, you put your credibility on the line when you either defend what appears to be going on, or deflect it entirely.

Hanners
05-16-03, 10:38 AM
I may have missed it, but did Digit Life fail to mention anything about the 3DMark2003/nVidia issues in their 5900 Ultra review? I notice they used 3DMark2003 as a benchmark, and saw no comments regarding the results.

Joe DeFuria
05-16-03, 11:06 AM
Originally posted by Typedef Enum
Me thinks the guy took a step back, and began to actually take a freakin' _look_ at the screen for once while doing these articles, and began to question his findings...Ya think? [/B]

I agree mostly...but I think instead of questioning the results, he's just waiting for nvidia to give him a "plausible" explanation for the cheating issue, so that he can pass it on to the public in his revew.

This way, he "acknowledges" the issue, but he can claim he has some grounds to not actually investigate it himself.

The question is...can nvidia actually come up with something that is "plausible?" (Or stated another way, what will Anand accept as plausible?)

Sazar
05-16-03, 11:08 AM
Originally posted by Joe DeFuria
I agree mostly...but I think instead of questioning the results, he's just waiting for nvidia to give him a "plausible" explanation for the cheating issue, so that he can pass it on to the public in his revew.

This way, he "acknowledges" the issue, but he can claim he has some grounds to not actually investigate it himself.

The question is...can nvidia actually come up with something that is "plausible?" (Or stated another way, what will Anand accept as plausible?)

I contend that the plausibility depends entirely on the reviewer to an extent and their understanding of what is happening...

personally I would have loved to see the explanation being given to b3d and have the peeps there disect it for validity... :)

Dazz
05-16-03, 11:34 AM
They may also be using their own methords to find such problems/cheating rather then pass on old news.

DadGT
05-16-03, 12:11 PM
I find it very interesting that there are no 3DMark2003 or 2001 results in the Anandtech review of the 5900U. If the 44.03 drivers "opitmizations" are truly only for 3DMark2k3, then the article should be relatively free of taint. So far, I have heard no mention of the same "clipping plane" issues from using other games.

Sazar
05-16-03, 12:13 PM
Originally posted by DadGT
I find it very interesting that there are no 3DMark2003 or 2001 results in the Anandtech review of the 5900U. If the 44.03 drivers "opitmizations" are truly only for 3DMark2k3, then the article should be relatively free of taint. So far, I have heard no mention of the same "clipping plane" issues from using other games.

apparently... per some posts @ b3d... serious sam is exhibiting similar issues... but they are related to benches conducted using the first version... not SS: SE

Hanners
05-16-03, 12:47 PM
Originally posted by nam ng
Digit-life had plenty of guys extremely competent, very unlike most others pretending to be experts. They ain't no fools...

I don't know about that, I've seen some pretty stupid comments come from that site over the years...

Hanners
05-16-03, 12:52 PM
Originally posted by CoWBoY
If you can't trust your own eyes, who's eyes can you trust?

Do the testing yourself, it will provide you with far less disagreement. :D

If I could afford (or even get my hands on) a 5900 Ultra, I'd do it without a second thought. Sadly, like most people I'm not in such a priveledged position, so we have to rely on others to do the testing for us....

Hanners
05-16-03, 12:57 PM
Originally posted by CoWBoY
I suppose their is no point in disgreeing with what someone says then until you see for yourself correct? :p

This forum would get mighty boring if that happened.....

digitalwanderer
05-16-03, 02:06 PM
Anand is right now in closeted talks with nVidia & that Doomed3 guy about how they're going to screw the public now with some falsified/optimized benchmark results!!!!!! :mad:


(Well, you wanted it livelied up a bit...didn't you?) :p

Lezmaka
05-16-03, 04:59 PM
Maybe he thinks it's not worth mentioning. Did his site post anything about Quack? Sure, some things like this and quack are big to some people, but surely not all. I don't see anything on the major news sites like news.com or wired.

muzz
05-16-03, 05:13 PM
He's probably waiting to hear a response from NV ( MY guess, and thats ALL it is)........ at which point he will tell the public what they say, which will be " They said it's a bug, and they are working FEVERISHLY[ :rolleyes: ] to find out what it is, I will report back when we know........

At least ET had the NADDS to speak up ( although they MAY [doubtful, but possible] find out they were not ENTIRELY correct).

Hellbinder
05-16-03, 05:59 PM
At least ET had the NADDS to speak up ( although they MAY [doubtful, but possible] find out they were not ENTIRELY correct).

Im sorry, but it is what it is. In this case there is no way that it can be attributed to a driver bug. Driver bugs do not add Clip Planes just beyond the movement Rails of specific benchmarks...

digitalwanderer
05-16-03, 06:18 PM
Originally posted by Hellbinder
Im sorry, but it is what it is. In this case there is no way that it can be attributed to a driver bug. Driver bugs do not add Clip Planes just beyond the movement Rails of specific benchmarks...

Now, now Hellbinder...let's not be too hasty.

I'm sure that nVidia is working VERY hard right now at coming up with a very good reason for that....or buying some people who will for them. :)

muzz
05-16-03, 06:20 PM
The word is Feverishly Dig....

Hard doesn't do it justice.....:D

digitalwanderer
05-16-03, 06:41 PM
Originally posted by muzz
The word is Feverishly Dig....

Hard doesn't do it justice.....:D

I'm sorry, you are indeed correct.

:lol:

muzz
05-16-03, 06:45 PM
TY sir.........;)

StealthHawk
05-16-03, 08:50 PM
Originally posted by Hellbinder
Im sorry, but it is what it is. In this case there is no way that it can be attributed to a driver bug. Driver bugs do not add Clip Planes just beyond the movement Rails of specific benchmarks...

Come on now, Hellbinder.

They were both cheats :rolleyes:

Why is it that you believe ATI's explanation that it is a bug but immediately say that nvidia is lying.

The nature of the two cheats may be different. ATI was just substituting some "optimizations" while they got their driver fully functional. In other words, they were showing the performance that would be expected with a driver coming out very soon. The same thing can't really be said about nvidia. ie, there is no evidence at this point that some essential features have not been implemented into nvidia's driver.

Rogozhin
05-16-03, 08:54 PM
Most of the problem comes with nvidia doing this before with another set of dets that decreased the IQ in 3dmark to yield a higher score, so they figured they would have to do it without using IQ cheats.

Rogo

Darth Rancid
05-17-03, 01:10 AM
ATi has a glorious history of cheating and being caught, beginning with the Rage Pro's "Winbench 3D" drivers that made it beat up a Voodoo2 in Winbench, but failing to improve anything else... and seemingly ending with Quack. Maybe they've learnt their lesson now.

Typedef Enum
05-17-03, 01:16 AM
Wow...So you're saying that over the course of the last 5-6 years, there have been 2 instances? I'm not so sure I would call that "glorious," though I hate _any_ level of cheating...

The flip side is that, despite all the accolades given to the driver writers @ nVidia, they have certainly walked a very tight line between either all-out cheating (this case, or possibly the earlier 3DM2003 IQ issues), or just very suspect (IE huge claims of driver performance increases, all the while, sacrificing quality) practices.

Sazar
05-17-03, 01:20 AM
Originally posted by StealthHawk
Come on now, Hellbinder.

They were both cheats :rolleyes:

Why is it that you believe ATI's explanation that it is a bug but immediately say that nvidia is lying.

The nature of the two cheats may be different. ATI was just substituting some "optimizations" while they got their driver fully functional. In other words, they were showing the performance that would be expected with a driver coming out very soon. The same thing can't really be said about nvidia. ie, there is no evidence at this point that some essential features have not been implemented into nvidia's driver.

considering the facts around the matter :) and considering the history of the drivers ati had released as well as the particular source of the information as well as the exe 'exposing' the so-called 'cheats' in quake... the nature of the 'cheats' becomes different...

the so-called optimizations had been in the driver releases for a while before the 8500 cards dropped...

what nvidia is doing is a different scenario... and especially considering they propagated the 'expose' of the quake/quack issue it does bring the whole issue under another light :)

I am sure all IHV's optimise... but in THIS day and age of accountability it is unfortunate this has happened... especially considering the potential of the cards themselves...