Go Back   nV News Forums > Hardware Forums > Benchmarking And Overclocking

Newegg Daily Deals

Reply
 
Thread Tools
Old 11-11-03, 03:03 PM   #73
Morrow
Atari STE 4-bit color
 
Morrow's Avatar
 
Join Date: May 2003
Posts: 798
Default

Quote:
Originally posted by Sazar

Quote:
Originally posted by AnteP
well ATi usually performs slower (not this much though) in tests with massive CPU dependency
in my experience that is
really ?

didnt know about that...
Yes, this is true and the reason for this is because HW T&L of the R3x0 and prior cores is not working as advertised... in fact it's broken.

T&L on ATI cards is mainly done by the main CPU and not by the graphiccard (VPU) as it should be, therefore when using games or benches with require heavy CPU workload, the performance of Radeons drops proportionally more dramatically than on GeForce cards.

This explains also why high-end GeForce cards run great on slower systems where high-end Radeons need more CPU horsepower to unleash their real performance.
Morrow is offline   Reply With Quote
Old 11-11-03, 03:10 PM   #74
bkswaney
Mr. Extreme!
 
bkswaney's Avatar
 
Join Date: Aug 2002
Location: SC
Posts: 3,421
Send a message via Yahoo to bkswaney
Default

Quote:
Originally posted by Morrow
really ?

didnt know about that...


Yes, this is true and the reason for this is because HW T&L of the R3x0 and prior cores is not working as advertised... in fact it's broken.

T&L on ATI cards is mainly done by the main CPU and not by the graphiccard (VPU) as it should be, therefore when using games or benches with require heavy CPU workload, the performance of Radeons drops proportionally more dramatically than on GeForce cards.

This explains also why high-end GeForce cards run great on slower systems where high-end Radeons need more CPU horsepower to unleash their real performance.
[/quote]

If this is true how is ati selling the card on the fact it has t&l?


BTW: my rig lost about 900 points. I'm now scoring a little over 6000.
bkswaney is offline   Reply With Quote
Old 11-11-03, 03:11 PM   #75
sebazve
 
Join Date: Jul 2002
Location: Montevideo, Uruguay
Posts: 421
Default

Quote:
Originally posted by Morrow
really ?

didnt know about that...


Yes, this is true and the reason for this is because HW T&L of the R3x0 and prior cores is not working as advertised... in fact it's broken.

T&L on ATI cards is mainly done by the main CPU and not by the graphiccard (VPU) as it should be, therefore when using games or benches with require heavy CPU workload, the performance of Radeons drops proportionally more dramatically than on GeForce cards.

This explains also why high-end GeForce cards run great on slower systems where high-end Radeons need more CPU horsepower to unleash their real performance.
[/quote]


what the **** are you smoking??
__________________
Signatures are a waste of bandwidth!
thanks rwolf!!!!! :-P
sebazve is offline   Reply With Quote
Old 11-11-03, 03:14 PM   #76
Nutty
Sittin in the Sun
 
Nutty's Avatar
 
Join Date: Jul 2002
Location: United Kingdom
Posts: 1,835
Send a message via MSN to Nutty
Default

I dont think it can be a large proportion of T&L done on the CPU. If it was all, that would mean all vertices get transformed on the cpu. It just wouldn't be fast enough.

It may be they do a few more matrix setups, or have to do more DMA setting up on the cpu which uses slightly more cpu time. But the T&L does work. You would get the speed of a TNT2 otherwise, which was the last card that did everything on the cpu.
Nutty is offline   Reply With Quote
Old 11-11-03, 03:16 PM   #77
The Baron
Guest
 
Posts: n/a
Default

I thought that was the Savage2000.
  Reply With Quote
Old 11-11-03, 03:17 PM   #78
saturnotaku
Apple user. Deal with it.
 
Join Date: Jul 2001
Location: The 'burbs, IL USA
Posts: 12,502
Default

Quote:
Originally posted by Nutty
The point is, they wouldn't. Unless IQ is being sacrificed. If its not, then wheres the problem?

I mean I dont care that NV is cheating if the IQ stays the same. I really dont.

I dont see the difference with having a compiler in a driver making shaders work better, or a driver engineer making shaders work better. Ideally the former is preferred, but until they get a miracle compiler built, it aint gonna happen.

The whole issue of benchmarks is just totally flawed. I thought everyone agree'd ages ago to stop paying attention to PenisMarks.. sorry 3dmarks?

You should judge a card on what you're gonna use it for. If you're gonna play games, bench it with games. If you're gonna be a 3dmark bragger, then yeah use 3dmark. But frankly, I couldn't care less if NV said, yeah we cheat our ass off in 3dmarks.. I mean I just dont care. As long as my games run fine, and my programming works, thats all I care about.

Whats also annoying is that Futuremark are being paid by IHV money, which comes from you ppl that buy cards, to just sit there and churn out more anti-cheat patches that in the grand scheme of things just dont mean squat.

Why is it okay for JC to code a path for NV3X, but Benchmarks wont?

I'm starting to sound like a right nvidiot now, (which I prolly am) but I'm just soo bored of this 3dmark thing.

I can see the point of view of having a program that is supposed to run identical on hardware, to test performance. But that _isn't_ how things work!
This is precisely how I feel. I'm still waiting for someone to show me how much (if at all) image quality is lowered with using the old patch as opposed to the new one. If we're talking a couple out of place pixels when the image is zoomed in 300%, I think that's being just a tad ridiculous.

This is just like the Quack thing. ATI fixed image quality problems while getting just as good, if not better, performance. Do we know that this is going on here? What exactly is this new patch detecting and removing and what are the visual differences?
saturnotaku is offline   Reply With Quote
Old 11-11-03, 03:19 PM   #79
andypski
Registered User
 
Join Date: Jun 2003
Posts: 34
Default

Quote:
Yes, this is true and the reason for this is because HW T&L of the R3x0 and prior cores is not working as advertised... in fact it's broken.

T&L on ATI cards is mainly done by the main CPU and not by the graphiccard (VPU) as it should be, therefore when using games or benches with require heavy CPU workload, the performance of Radeons drops proportionally more dramatically than on GeForce cards.
Total unmitigated rubbish.
andypski is offline   Reply With Quote
Old 11-11-03, 03:23 PM   #80
Razor04
Registered User
 
Join Date: Jan 2003
Posts: 205
Default

Quote:
Originally posted by Nutty
Why is it okay for JC to code a path for NV3X, but Benchmarks wont?
I can't believe you just said that...that has to be the stupidest thing I have ever read. A benchmark is supposed to be fair and impartial...an apples to apples comparison or as close to one as is possible. What you are suggesting is an apples to oranges comparison where oranges are the preferred fruit. I hope I made this simple enough for you to see your error.

And about those compilers...I have no problem with them at all so long as they don't decrease image quality or do stuff that wasn't intended by the developer. If a developer asks for output A using X instructions the compiler should optimize X instructions but still output A. Anytime that output A is different because of a compiler should raise red flags.

Tests like the one here with 3DMark show that it sure as hell wasn't the compiler providing the optimization but hand picked "optimizations" instead. A compiler would not suddenly lose performance in a program because that program changes itself so it could no longer be detected by cheating drivers.

NVIDIA has definately lost me as a customer till at least the NV50...and that is provided they remove all their old cheats and can show they have a IQ and perfomance lead on ATI.
Razor04 is offline   Reply With Quote

Old 11-11-03, 03:25 PM   #81
Richteralan
The Enigma
 
Richteralan's Avatar
 
Join Date: Jan 2003
Location: State College, PA
Posts: 758
Send a message via ICQ to Richteralan Send a message via MSN to Richteralan
Talking

wow, NV cheating is like a fashion thing, if you don't say it, you're out, you're not "trendy"....

haha...duh~~
Richteralan is offline   Reply With Quote
Old 11-11-03, 03:27 PM   #82
subbo
Registered User
 
subbo's Avatar
 
Join Date: May 2003
Posts: 452
Default

I'd just like to say, that this thing makes the general public more pissed at 3dmark for ruining their scores than nvidia supposedly cheating with their drivers. I still wont buy ati though, been there and it was not all that peachy. Though, without ATI we'd just now be celebrating the launch of the GF2.
subbo is offline   Reply With Quote
Old 11-11-03, 03:33 PM   #83
fallguy
 
fallguy's Avatar
 
Join Date: Oct 2003
Location: ^ Next to her
Posts: 155
Default

AthlonXP1800, I have never seen anyone more in denial than you.
fallguy is offline   Reply With Quote
Old 11-11-03, 03:34 PM   #84
bkswaney
Mr. Extreme!
 
bkswaney's Avatar
 
Join Date: Aug 2002
Location: SC
Posts: 3,421
Send a message via Yahoo to bkswaney
Default

I'm sure not going to loose any sleep over this.
I'm scoring 6000 on this non cheating bench.

I can play every game I have at way above the FPS
needed for smooth game play.
The IQ is as good as my 9800 was to my eyes.

So that's all "I" need.
I think the performance will still get better as newer drivers r released.
It seems nvidia has had a hard time figuring out the drivers for the NV3x
compared to there older hardware.

Most of the game companies don't seem to mind coding for NV
so why worry about it.
bkswaney is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


Similar Threads
Thread Thread Starter Forum Replies Last Post
NVIDIA Driver 295.53 installs into kernel 3.4 without a patch! jdmcdaniel3 NVIDIA Linux 3 06-08-12 09:41 AM
Need Help Installing NVIDIA Tesla M2070Q in Linux RHEL5 Ferianto85 NVIDIA Linux 0 05-18-12 08:35 PM
Rumor regarding lack of 680 availability ViN86 Rumor Mill 6 05-09-12 04:48 PM

All times are GMT -5. The time now is 01:33 PM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 1998 - 2014, nV News.