Re[T]ardocp BS again

[Unit01]

This post represents my personal opinion about it only. First i haven't read the whole editorial at hardocp. But instead i've gathered info from quotes on anandtech and beyond3d, and people whom have read the editorial and voiced their opinion on it.

Remember the 3dmark issue that NV so happily screwed with? Sure we wont optimize for benchmarks. Then it was revealed they optimized specifically for benchmarks(and games). Hardocp finished off their BS editorial about the issue saying that it wasn't such a big deal since it's just a synthetic benchmark.

Well the current UT2003 filtrering situation wont end any different from that one. In a new editorial written by hardocp(i wont link it as i refuse to give them accidental click-hits). They say that it's ok if nvidia performs bilinear filtrering in ut2003 vs ati's trilinear filtrering. They say it's completely fair and apples to apples(why do people use that phrase ) comparison between FX5900ULTRA and Radeon9800Pro. How the hell can they make a conclusion like that? It's ok that nvidia performs less work while ati performs more work and thereby gets a lower score?

Maybe i'm just exagerating and [H]ardocp is finishing of their transformation into Re[T]ardocp.

Wooba Wooba

 

[TheGreatGrapeApe]

Told you, all benchmarks are coming to an end. Both cheating and only one coming back, means everything is up for questioning. I wonder if that same test method explains some of the scoring in the VGA buyer's guide.

Personally I saw this this morning right at the end of work and didn't post a link because it still seemed vaguey IMO. The conclusion is, it's not the same, it's not as good, but 'I can't tell'.
Well, I'll need to see a little more to truely be sure. It definitely doesn't make me feel good about what WILL become important when chosing features for the next cards. Will we return to BIG FPS at the price of IQ, even if it's a small hit? 95% IQ of the last card but 2% increase in frames. After a few gens out IQ is down alot and only for a small increase in FPS, but we dind't noticed because compared to the last card it wasn't a noticeable difference. I knw that's simplistic, but it's kinda the feeling I'm getting. 'Well only the purists' will notice', is what I hear from the bored/boardroom, yet we'll be the one's paying the premium FOR these features we wish we had.
So what's in a Benchmark? Not much. Hope Aquamark makes it out soon. And I hope they have some good lawyers!

Oh well, once again; All this kinda stuff Sux!
Freakin' gonna have nightmares now before going to bed. Aaaahh Textures, MipMaps, Polygons!

Freakin' Graphics Cards/Mfrs!

EDIT: Just thought I'd leave you with the Editor's note;
<font color=blue>Do let us say this. If you are spending your hard earned dollars on video cards now days and doing so based on framerate numbers alone, you are cheating yourself</font color=blue>

That goes for all you who saying playing at 16x12 is better than 1280x1024 4XAA/8XAF. :tongue:

Help, my dreams are droping frames! Gotta replace the grey-matter.

- You need a licence to buy a gun, but they'll sell anyone a stamp <i>(or internet account)</i> ! <A HREF="http://www.redgreen.com" target="_new"><font color=green>RED</font color=green> <font color=red>GREEN</font color=red></A> GA to SK <P ID="edit"><FONT SIZE=-1><EM>Edited by TheGreatGrapeApe on 07/21/03 05:09 PM.</EM></FONT></P>

 

[c0d1f1ed]

They say that it's ok if nvidia performs bilinear filtrering in ut2003 vs ati's trilinear filtrering.

I don't see any bilinear filtering, it's all trilinear...

 

[Unit01]

Of course it is
http://www.beyond3d.com/forum/viewtopic.php?t=6719

Re[T]ardocp selected shots and positions where the transitions are hard to spot, thereby justifying their "doesn't matter" BS.

Wooba Wooba

 

[phial]

standard bilinear texture filtering on my GF4 sucks ass. if you are moving steadily in one direction , you can see the textures "jerk" to the different mipmap ranges

its sick. drivers dont affect this. same with all versions.. only alternative is AF, which gives a hefty 20% performance hit at 4x

-------

<A HREF="http://www.quake3world.com/ubb/Forum1/HTML/001355.html" target="_new">*I hate thug gangstas*</A>

 

[c0d1f1ed]

That -is- trilinear. Bilinear filtering shows -abrupt- mipmap level changes from one pixel to the other. Try running it on a TNT2. Sure, Nvidia is using a different method than ATI but any optimization technique is a good one if the difference is not noticable while playing, which is the case. Trust me, bilinear is noticable while playing. Trilinear isn't, with ATI cards as well as with Nvidia cards.