5900Ultra review at [H]ard OCP

[Doomtrooper]

This site was for all of the above Stealthawk..B3D was 1st to show what 'application' was and used in their reviews, then hardware.fr.

[H] was not the 1st, in fact they didn't know about until reading it here.

http://www.beyond3d.com/forum/viewtopic.php?t=4772&highlight=application

 

[Dave Baumann]

This site was for all of the above Stealthawk..B3D was 1st to show what 'application' was and used in their reviews, then hardware.fr.

We missed the trick in the 5800 review by putting the benchmarks number up in 'Balanced' (the driver default) but pointing out that this wasn't real Trilinear. I then talked to Brent just before starting this thread which coincided with Sean doing the 5600/5200 review and I believe Sean included those settings in the review after Brent talked to him.

 

[K.I.L.E.R]

Is the word 'fraudulent' too strong to describe what is happening?

fraudulent

\Fraud"u*lent\, a. [L. fraudulentus, fr. fraus, fraudis, frand: cf. F. fraudulent.] 1. Using fraud; trickly; deceitful; dishonest.

I don't think so.

 

[PSarge]

Is the word 'fraudulent' too strong to describe what is happening?

'twas a dark night when the men in suits came a-knocking upon K.I.L.E.R's door. A night he wouldn't forget......

 

[K.I.L.E.R]

Is the word 'fraudulent' too strong to describe what is happening?

'twas a dark night when the men in suits came a-knocking upon K.I.L.E.R's door. A night he wouldn't forget......

They can come and knock on my door. I will just back up my claim.

 

[StealthHawk]

This site was for all of the above Stealthawk..B3D was 1st to show what 'application' was and used in their reviews, then hardware.fr.

[H] was not the 1st, in fact they didn't know about until reading it here.

http://www.beyond3d.com/forum/viewtopic.php?t=4772&highlight=application

Nobody said that B3D didn't discover the settings...just that AFAIK [H] actually used it in their review first.

March 13@[H]: http://www.hardocp.com/article.html?art=NDQ0
April 7@B3D: http://www.beyond3d.com/previews/nvidia/nv31/

Don't know about hardware.fr as I don't go to foreign sites for my reviewing needs.

 

[Pete]

You should, considering the scarcity of satisfactory reviews in English. HW.fr has done some good reviews of DX9 cards. You don't need French to interpret the graphs, but Google does a decent job of translation if you must read the commentary.

 

[WaltC]

Since Brent didnt come here, Dave went over there.
http://www.hardforum.com/showthread.php?s=&threadid=640474&perpage=15&pagenumber=3
Sorry if this post pisses you off Doomtrooper....

I perused the thread from Dave's post forward, including Brent's responses, and although I thought I had become jaded to this kind of rhetoric it still astonished me. Basically, Brent's rationale for ignoring the trilinear filtering issue was that he was just too busy to do a thorough review but that he'd "check into it" and maybe say something about it later, time and schedule permitting, of course. The fact that he hadn't "personally verified" the trilinear issue prior to publishing the review, although he did admit prior knowledge of the problem, didn't seem to bother him in the least.

Wow. Brent says he's "busy" doing a "9600 review" and so can't immediately address the issue. How does the fact that nVidia's removed trilinear filtering support from its drivers for detail textures not rate high on the scale of issues important to people thinking of buying the products he's reviewing?

Brent's position seems to be that bilinear filtering is "good enough." I've said it before: nVidia and its partners in crime seem to be working overtime to set back the general state of the 3D chip industry to what it was several years ago. It boggles the mind to hear a hardware reviewer state that the difference between trilinear and bilinear filtering is such an unimportant issue relative to IQ and performance that its consideration in a review of a $500US 3D card isn't even worth his time. This is a bridge we crossed years ago, and I thought burned behind us. Heh...I guess not. Amazing stuff these days.

IMO, if a hardware reviewer is "too busy" to properly cover issues like these he's too busy to be in the business of writing hardware reviews. Why there should be any question at all of the relevancy of issues like this in a 3D-card review is completely beyond me.

 

[Doomtrooper]

I read that thread too, the logic used by him as it being a 'ok' optimization is ..well I'm speechless. The idiocy from that comment is pathetic.
Here we have a IHV disabling a setting that you have selected in your control panel 'on purpose' to gain frames. Nvidia has disabled what you the end user selected without your permission.

I'm sorry the more I read this Drivel, it just keeps adding more and more to the evidence that the entire internet review circle is a big pay-off.

How about:

http://www.motherboards.org/articlesd/hardware-reviews/1257_3.html

A Ti4200 scoring 4361 in 3Dmark 03

 

[digitalwanderer]

Mr. Baumann, me heartfelt gratitude to you.

I've been reading the thread over at [H] about Brent's review and found to me less-than-surprise that I am again phantom-banned over there. (It doesn't say I'm banned, but I can't post. ) It drove me a little nuts since there were SO many things I wanted to point out over there to Brent and I couldn't!

But you did, and you did it well.

Thanks for going over there and posting up what you did; your questions were fair, rational, and damning as hell. The posts by you, creedAMD, and hardBBQ kept me blood-pressure down while trying to read that.

I gotta go e-mail Brent, he's always been reasonable and open in the past with me about stuff thru e-mails....how come I think that is about to change too?

 

[Brent]

I just wanted to say thanks to Dave for explaining and summerizing the situation for me. I do understand it better now.

I still want to do my own testing first of course, but I will see about putting an amend in the review to alert users of this issue and that we are looking into it.

 

[Reverend]

Oh man... now that I read Brent's signature, it becomes even more incomprehensible what he wrote in that HardForum thread and to Dave.

 

[Doomtrooper]

I just wanted to say thanks to Dave for explaining and summerizing the situation for me. I do understand it better now.

I still want to do my own testing first of course, but I will see about putting an amend in the review to alert users of this issue and that we are looking into it.

I would just like to say maybe next time be a little more careful about coming to your conclusions, especially working for a sight famous for editorials such as 'benchmarking right' and as Reverend said 'your sig'.

 

[digitalwanderer]

Re: Mr. Baumann, me heartfelt gratitude to you.

I gotta go e-mail Brent, he's always been reasonable and open in the past with me about stuff thru e-mails....how come I think that is about to change too?

The heck with e-mail, I left a message for you over here Brent.
(Sorry for the crossboard post, but it ain't exactly conducive to the civil atmosphere around here...I tend to get ranty when I get emotional, a nice advantage to not being a professional. )

 

[Althornin]

i do wish we could cease the personal attacks on kyle, et. al. at the [H].
It serves as nothing but a vent, and just lowers the S/N ratio of the forums.

 

[Reverend]

i do wish we could cease the personal attacks on kyle, et. al. at the [H].
It serves as nothing but a vent, and just lowers the S/N ratio of the forums.

I encourage deserved criticisms of websites/website-personnels (including our own), not out-and-out ventings/flames.

Believe it or not, it helps (and not just within the site).

 

[Bambers]

At 1600x1200 we encountered an odd problem with the 9800 Pro. It pulls in 54.2FPS with NoAA and NoAF and it also pulls this same result in 2XAA/8XAF, 4XAA/8XAF and 6XAA/8XAF. Indicating that Anti-Aliasing doesn’t seem to be initializing at this resolution.

I see lots of reviews with comments like this. Use your eyes, tell us if AA really is on or not (unless you actually meant that actually no AA was being applied)

On my 15" monitor it was pretty hard to see the difference between 4x and 6x AA at 1152x864 but I recently purchased a new 19" diamondtron tube based monitor and at 1600x1200@85Hz I can easily pick out the edge graduations at 4xAA and the jaggies at 2048x1536@80Hz 2xAA are quite noticable. Its easy enough to see when AA is off or on.

 

[5150 Joker]

I don't see why everyone puts so much credibility in HardOCP as it is. They've shown time and again that they will sell-out for the highest dollar available, it's capitalism at it's finest afterall. Digi--I'm glad you make people aware of this stuff on other boardsf but does it need to be put so dramatically? I mean come on, "OMG DAVE CUSSED..".

 

[digitalwanderer]

Digi--I'm glad you make people aware of this stuff on other boardsf but does it need to be put so dramatically? I mean come on, "OMG DAVE CUSSED..".

Sorry, melodrama is me stock and trade. I'm a color commentator, not a technical expert.

I will go and re-phrase that one out of respect for Mr. Baumman, me apologies for any offense.

 

[Quinn1981]

I feel as if a lot of people are getting the run-around. If the results aren't truthful to what the card is able to do globally, then it isn't worth posting. That was the point Kyle has tried to make with inferences to 3DMark2003. If the results can't be trusted, then why use the given app or results from a certain website?

For [H] to have consistent results with their claims, this problem must be solved. And a point of reason, shouldn't a journalist be too busy satisfying their readers rather than satisfying their review unit suppliers?

As for the specific incident in UT2003, while I agree with Brent that IQ is not all it's cracked up to be if you can't tell the difference right away (which is what I think he's getting at), shouldn't the power to control IQ and performance be in the hands of the user, or better put, the person who purchased the graphics card?

I also don't know what to think about a website that paid careful attention to ATI's application specific "tweaking" in Quake 3, but instead finds no immediate time to handle this situation and very rarely comments on current accounts of cheating by a certain company we have all grown to love.

My dad told me this one time, "It's hard to build up a good reputation, but easy to have it torn down." I suggest [H] works hard to uphold it's long standing grandeur. I'm sure there are people willing to bring much more attention to this matter.