Elite Bastards comment on the upcoming catalyst 3.5 set, cheating, & the beta 3.5s

[digitalwanderer]

Yeah, it's a shameless self-pimp...but I finally got a half-way decent article up on the frontpage of Elite Bastards (http://www.elitebastards.com) and I was feeling all proud and full of a sense of accomplishment. :D (And relieved as hell that I could still write something that would pass me peer review over there! ;) :lol: )

It's not much, but it doesn't bash nVidia so I thought y'all wouldn't mind too much. :)

It's just about the upcoming catalyst 3.5 set, the beta pre-3.5 leak this weekend, and ATi's "optimizations" in some benchmarks.

If nothing else, it should get an interesting discussion going. :cool:

[Sazar]

Originally posted by digitalwanderer
Yeah, it's a shameless self-pimp...but I finally got a half-way decent article up on the frontpage of Elite Bastards (http://www.elitebastards.com) and I was feeling all proud and full of a sense of accomplishment. :D (And relieved as hell that I could still write something that would pass me peer review over there! ;) :lol: )

It's not much, but it doesn't bash nVidia so I thought y'all wouldn't mind too much. :)

It's just about the upcoming catalyst 3.5 set, the beta pre-3.5 leak this weekend, and ATi's "optimizations" in some benchmarks.

If nothing else, it should get an interesting discussion going. :cool:

your peer review needs to check your spelling though digi :)

last line down

and I just wanted to see if'n anyone else thought so too

:)

[Sazar]

Originally posted by digitalwanderer
No, "if'n" is trademark Digispeak; sort of like "methinks" and "pagh". ;)

it could well be... lol... but it kills the flow of the sentence...

its like a singer singing opera and catching a frog in the throat half way through a note...

:D

-edit-

hmmm... consdering all the stuff bout nvidia and yourself and what not... your site does not appear to be that big... 200 odd registered users ?

whats the daily uniques average ?

[Sazar]

Originally posted by digitalwanderer
Well if it takes away from the flow that is an entirely different story! Consider it changed, and my thanks to you on the tip/critique. :cool:



np m8... :)

keep posting on further developments :D

[Hellbinder]

I am actually a little upset at this release.... more like upset in general..

As with most CATALYST™ releases, the performance has increased in numerous situations. Here are some examples observed in CATALYST™ version 3.5.
Improvements in the ATI OpenGL drivers have experienced performance gains such as:
• GL Excess performance improves 12-14% on the RADEON™ 9700 PRO and 9800 PRO. Modest benefits are noticed on our other graphic products as well
• Unigraphics performance (ViewPerf Ugs-03) improves 27% on the RADEON™ 9800 PRO, 52% on the RADEON™ 9700 PRO, and 79% on the RADEON™ 9500 PRO
• Data Explorer performance (ViewPerf DV-08) improves 9% on RADEON™ 9800 PRO, and the RADEON™9700 PRO. The RADEON™ 9500 PRO improves by 14.5%

How can this not be *cheating* ????

Only improvemetns in these specific *BENCHMARKS*. Means they are application detecting and tweaking for a higher Benchmark score. With little effect on real games.

I am honestly a little Irritated right now about the whole thing. I am sick to the core of all companies and And review sites emphasizing Prefabed Benchmarks. I am begining to find myself in some agreement with Kyle except for different motivations.

Screw all this 3dmark, Shadermark, Glexcess, Villagemark, Specview etc etc etc junk. Its a giant Meaningless pissing contest that has ZERO effect on the real world of gaming.

[The Baron]

Screw all this 3dmark, Shadermark, Glexcess, Villagemark, Specview etc etc etc junk. Its a giant Meaningless pissing contest that has ZERO effect on the real world of gaming.
In other news, Hell has frozen over. :D

I agree with you, HB. I was also really annoyed at ATI's hubris in the release notes posted at EB:The 3DMark2003 shader optimizations found in previousCATALYST™ releases have been removed
Excuse me? The shader optimizations "found" in previous releases?

Like they were complete accidents? OK, thanks PR People! :rolleyes:

[The Baron]

I think I need to send some email to Unwinder and see when he's going to release his Anti-Application-Detector script...

[The Baron]

Hell... the 26th needs to hurry up and arrive so I can get my finished UT2k3 GPU Killer Map from a bud and then maybe finish my article on how bots are the only accurate way to benchmark anymore.

Heh. Yeah. That's it. Bots. Only way to do it (unless Unwinder hacks the drivers)...

[The Baron]

Won't he have to re-write it for the new Cat3.5 set? (I do not think he's a beta tester, but I don't know.)
I would imagine so...

[Hellbinder]

i posted about this subject at B3D and was told to *Calm down* by Saem :) ...

To which my response is this...

ok...

So tweaking for benchmark improvements is *ok* then. fixes that only affect Glexcess and Specview. Yet have no affect on actual games..

Tell me.. Given that they are 45-80% increases in these specific benchmarks yet dont affect anything else.. how is it possible they are not application specific?

Look, I just want to know. Either this kind of thing is *Ok* or its not. If its *OK* then unwinder needs to back the truck up with his utility that negates all application specific tweaks. Because he is calling it Cheating.

[Nutty]

Is it not possible that the new drivers are just making better use of the hardware, and they just happened to state what this improvement in efficiency does to those specific benchmarks?

Thats the kind of optimizing I like. But I doubt thats the case. Especially the huge increases they state. Its gotta be something iffy.

[saturnotaku]

Originally posted by Nutty
Especially the huge increases they state. Its gotta be something iffy.

I was gonna say I would be very suspicious because of those huge increases. Someone, though should do a gaming comparison because the new drivers could very well show some improvements in gaming. But as long as gaming isn't hurt in any way I guess it's all right. Man, I'm so confused as to what to make of all this I really am. :|

[deejaya]

Originally posted by digitalwanderer
What scared me about the bit you quoted was how they SPECIFICALLY stated that it was the "found" optimizations that were removed, sort of implying that there might just be some unfound ones they left in. :(

Like the 3dMark2001 ones? Or were they proven to be "legitimate"? I'm with all of you on this, the whole thing is enough to give you a big migraine. :p

And the way they worded the release notes, it sounds to me like they are general rendering improvements, and quote the %'s for things which these general rendering improvements improve. But who can really tell? I doubt ATi would make the same mistake twice, especially in the same release as fixing one, but stranger things must surely have happened (and will happen).

[schuey74]

From what I've read....
"Improvements in the ATI OpenGL drivers have experienced performance gains such as:"
I don't see the problem. I have not yet tested but several people at the Rage3D forums are saying that their OpenGL performance is up, especially with AA/AF. If actual OpenGL games get 7% - 10% improvement then specific benchmarks can probably gain 25% - 30% because they don't have to calculate the physics, AI, controls, & sound.

I can't imagine why ATI would do anything to harm their image after seeing the way everyone reacted towards Nvidia.

The alarms will go off after some cat 3.5 reviews are released and we can see how OpenGL performance is in real games. Of course, Unwinder's utility would be the best.

[DaveW]

Originally posted by Hellbinder
I am actually a little upset at this release.... more like upset in general..

How can this not be *cheating* ????

Only improvemetns in these specific *BENCHMARKS*. Means they are application detecting and tweaking for a higher Benchmark score. With little effect on real games.

I am honestly a little Irritated right now about the whole thing. I am sick to the core of all companies and And review sites emphasizing Prefabed Benchmarks. I am begining to find myself in some agreement with Kyle except for different motivations.

Screw all this 3dmark, Shadermark, Glexcess, Villagemark, Specview etc etc etc junk. Its a giant Meaningless pissing contest that has ZERO effect on the real world of gaming.

I know how you feel. I get excited seeing the words "performance increases", then I read further and see "20% increase in XXX mark", "25% increase in YYY mark"... WTF? Those aren't "performance increases" they are just benchmark optimizations. The catalyst team should spend less time arseing around with benchmark hacks and more time fixing existing issues.

The Counterstrike bug still isn't fixed. How hard can it be to fix a bug in a 5 year old game engine? It should be child's play to a professional driver team. If they can't fix an old simple game engine like Halflife, how can we trust them to properly run newer game engines like Doom 3 and Halflife 2 which at 10 times more complex?

[Quinn1981]

Originally posted by Sazar
whats the daily uniques average ?
Most people are scared of the name. Stats can be found here (http://w.extreme-dm.com/?login=elitebas).

The most uniques we've had in a day were 5000 something. The public stats haven't been up that long, but I think they are way more accurate than the server log stats that only I can see.

We would have more members I imagine if we would have had an equally nerdy name, but... we can't deny our true selves. We're bastards. :rolleyes:

[Skuzzy]

I could/can be a bastard,..just dunt know if I got what it takes to be an 'elite' bastard. :D

[Skuzzy]

Uhmmm,..I do have a problem plugging the pie hole up. It just sort of falls out of there. Sometimes it is connected to a cognizant thought though. :D

Besides,.I always post knowing I am probably going to get hammered. Masochist? Nah,...just goofy that way. It's the price ya pay when you express an opinion or try to help sometimes.

[Sazar]

Originally posted by Quinn1981
Most people are scared of the name. Stats can be found here (http://w.extreme-dm.com/?login=elitebas).

The most uniques we've had in a day were 5000 something. The public stats haven't been up that long, but I think they are way more accurate than the server log stats that only I can see.

We would have more members I imagine if we would have had an equally nerdy name, but... we can't deny our true selves. We're bastards. :rolleyes:

k... I was just wondering... listening to digi the last few weeks I had the assumption that EB was a huge website with a massive following so I was a bit confused looking @ the member base... is all :)

[Sazar]

Originally posted by digitalwanderer
No, no.....the site is still all small and cozy the way I like it; it's just my ego that keeps making it look so big to ya. ;)

:)

/me pimps his site a bit now...

over 12k uniques today and counting :)

[tertsi]

Well look at these C3.4 release notes, old but still ....

C3.4 release notes (http://pdownload.mii.instacontent.net/ati/drivers/Catalyst_34_Release_Notes.html)

The following is a list of DirectX9 Performance Improvements noticed in the latest release of CATALYST™.

* 3DMark03 scores are improved as much as 8% across our entire DirectX 9 (dx9) product line
* The difficult Mother Nature (dx9) test now runs as much as 20% faster than previous drivers ... blaah blaah

Sounds similar like C3.5 release notes.

[StealthHawk]

Originally posted by DaveW
The Counterstrike bug still isn't fixed. How hard can it be to fix a bug in a 5 year old game engine? It should be child's play to a professional driver team. If they can't fix an old simple game engine like Halflife, how can we trust them to properly run newer game engines like Doom 3 and Halflife 2 which at 10 times more complex?

What do you mean "still?" It was "fixed" in the Catalyst3.4s. At least, my system would no longer crash when I pressed esc. It would just take a loooong time to switch back with the 3.4s.

[Sazar]

Originally posted by StealthHawk
What do you mean "still?" It was "fixed" in the Catalyst3.4s. At least, my system would no longer crash when I pressed esc. It would just take a loooong time to switch back with the 3.4s.

only driver set I had a problem with this was cat 3.2's...

none of the other cats have given me a problem with CS... there is a slight delay once in a while... but other than that its fine...

are people actually experiencing a problem with the 3.5's like in the 3.2's or are they acting on hearsay ?

[Quinn1981]

I wasn't able to use the 3.4s, so I'm thinking the same will happen with the 3.5s. The 3.2s have been great for me thus far. I need to at least try the latest I guess. The Diggy 8500 isn't the best card to take advantage of newer drivers though. :)

[StealthHawk]

Originally posted by Sazar
only driver set I had a problem with this was cat 3.2's...

none of the other cats have given me a problem with CS... there is a slight delay once in a while... but other than that its fine...

are people actually experiencing a problem with the 3.5's like in the 3.2's or are they acting on hearsay ?

No problem with CS and the Cat3.5's here. Still the same long pauses when pressing esc that existed with Cat3.4, but it doesn't freeze like it used to with Cat3.2.

I'd say most of the "problems" are simply hearsay.