PDA

View Full Version : Beyond 3D's GeForce FX preview


Pages : [1] 2 3

swanlee
03-03-03, 07:45 AM
While the review seemed concise what was the point of it since it did not compare against non Nvidia cards? Was this meant to be a strictly Nvidia fanboy review or something, or did Nvidia give them a talking to about comparing it to non Nvidia cards? Why do a review at all if your only going to compare it against older cards from the same company, of course it's gonna be faster.

panzaman
03-03-03, 08:04 AM
I think was nVidia BS PR department that obliged them to not compare against a competition card. If I was Beyond 3D I would now publish a revisited review of the ATI 9700 with benchmarks done on the same system specs they have used to benchmark the FX.

DaveBaumann
03-03-03, 08:05 AM
Hint: Take a look at any of our (p)reviews...

creedamd
03-03-03, 08:13 AM
The card didn't really show a great improvement over the ti4600, that should tell you how it is against the r300.

swanlee
03-03-03, 08:19 AM
Ok nevermind nvnews.net screwed up and called this a review even though on beyond3d's site it is just a preview. Hello mcfly, nvnews.net needs to fix the homepage and link to make sure people know it is just a preview.

I hope the offical review will compare against other cards, if not Nvidia needs to be questioned on their PR policies again if they put any pressure to not compare against other cards.

creedamd
03-03-03, 08:26 AM
Originally posted by DaveBaumann
Hint: Take a look at any of our (p)reviews...

Why put so much into a preview, most people want the straight facts r300vsgffx? Your preview was very well written. I don't think nvidia should have put the hold on you guys. I think you should have waited and got a retail card. Then put out what you wanted and call it a review.

DaveBaumann
03-03-03, 08:42 AM
Nobody is stopping us from doing comparisons - we just don't do it as a matter of practice. B3D is a 3D technology site and we like to explore the nuances of a particular architecture when we get something new in like this - this follows through will all of our reviews.

The chances are that this will be followed up with a separate comparison article sooner or later, but it will be worthwhile holding on a little longer for that...

DSC
03-03-03, 08:46 AM
Something's truly broken in the FX, in RtCW and SS:SE, slower than a Ti4600 when max AA+AF is enabled?!?!?! :eek: :mad:

The margin of improvements in other areas aren't so great either. Makes you wonder wtf is Nvidia doing in their labs..... :rolleyes:

creedamd
03-03-03, 08:46 AM
Originally posted by DaveBaumann
Nobody is stopping us from doing comparisons - we just don't do it as a matter of practice. B3D is a 3D technology site and we like to explore the nuances of a particular architecture when we get something new in like this - this follows through will all of our reviews.

The chances are that this will be followed up with a separate comparison article sooner or later, but it will be worthwhile holding on a little longer for that...

buy you did do a comparison, just not against a competitor, which is most sought after.

CurtMan
03-03-03, 09:12 AM
I think Dave has a good point in waiting to compare the FX and FX Ultra to the competition since the R350 is coming out very soon. Why compare the FX to the 9700 Pro when it can be compared to the 9800 Pro and 9600 Pro (where the 9600 Pro should be close to the 9700 in performance).

swanlee
03-03-03, 09:15 AM
As long as they do comparisons in the "actual review" I have no problems with this preview. Previews are just that previews. Although from the sounds of it even an actual review may not have comparisons which is pretty pointless in my opinion, people want to know how it stacks up against card from other makers.

Nuances of an architecture are fine but there is nothing stopping you guys from throwing in a 9700 pro score just for a basic comparison.

nutball
03-03-03, 09:30 AM
I don't understand what the fuss is about. If a guy who runs his own website wants is more interested in the subtleties of the architecture than the simple "which is faster?", why do people think they have a right to have a problem with that? Especially where do they get the right to start questioning whether he's been told what to do by NVIDIA?

IMO the B3D preview has done more to reveal what NV30 is and what NV30 isn't than pretty much any other article in the subject. IMO "throwing in a 9700 pro score just for a basic comparison" adds no information whatsoever, as it doesn't tackle what R300 is and what R300 isn't.

creedamd
03-03-03, 09:45 AM
Originally posted by nutball
I don't understand what the fuss is about. If a guy who runs his own website wants is more interested in the subtleties of the architecture than the simple "which is faster?", why do people think they have a right to have a problem with that? Especially where do they get the right to start questioning whether he's been told what to do by NVIDIA?

IMO the B3D preview has done more to reveal what NV30 is and what NV30 isn't than pretty much any other article in the subject. IMO "throwing in a 9700 pro score just for a basic comparison" adds no information whatsoever, as it doesn't tackle what R300 is and what R300 isn't.

we are making suggestions, not flaming the guy, don't turn this into something it's not.

People ask for suggestions all of the time, give yours then let it be, don't knock someone for their opinion because it's not the same as yours.

swanlee
03-03-03, 09:51 AM
I don't think it's a big deal since it's a preview, let them comapre it against a voodoo 1 for all I care. In a real review though the value of that review would be pretty much nothing. Problem was Nvnew annoucned this as an actual REVIEW not preview.

Sazar
03-03-03, 10:05 AM
IMO dave did a good job... if you have visited b3d and had a decko on all the conversations that are ongoing concerning various aspects of the video card in question... you will see that teh [p]review did in fact touch on many of the points raised...

I was quite happy with the [p]review seeing as we have enough r300 v/s nv30 fodder flying around the web... this is a good comparison tool since nvidia had been touting numbers themselves between the gf4ti 4600 and the gf FX 5800ultra in their PR releases...

good job dave... maybe one day soon I will get off my lazy arse and subscribe to your forum instead of just ghosting like the last several months lol...

kudos..

druga runda
03-03-03, 10:07 AM
Well for me it was a review and a great one at that... what's the point... this (p)review shows the strenghts and weaknesses of NV30 vs NV25... do your own comparisons between that and R300... check some of their radeon reviews and you will get the approximate picture where the things lay between the two... but I can bet many other sites will be throwing out such R300 vs NV30 reviews soon.

kyleb
03-03-03, 10:24 AM
Originally posted by swanlee
I don't think it's a big deal since it's a preview, let them comapre it against a voodoo 1 for all I care. In a real review though the value of that review would be pretty much nothing. Problem was Nvnew annoucned this as an actual REVIEW not preview.

well you must admit it is rather confuseing, i mean Dave even used the term "review" when reffering to it up there. ;)

swanlee
03-03-03, 10:28 AM
Eve ghough I'm ok with it as a preview, you have to compare against other cards if you want the review to be of any real value. i can pull R300 numbers ferom other sites but realisticall that's no way to compare as these other numbers could be from different systems, different memory, different MB's etc. If you want a review to be valued by a diverse audience you simply must compare with other cards available and not just of the same manf. Only the wors fanboys would only consider cards from Nvidia.

Typedef Enum
03-03-03, 10:30 AM
The more I see of it, the more it becomes obvious that nVidia was simply caught with their pants down by ATI.

As has been pointed out before...If there was no ATI/9700, the results would have been in-line with previous releases.

The shocking thing to me was the relatively small increase over the Ti4600...and when you then consider the normal FX (400/400), it really becomes that much worse IMHO.

This preview also reinforced the notion that nVidia's AA modes need to be completely overhauled. For all intents and purposes, anything beyond 4x is almost useless...the performance is not very good, and the actual IQ increase looks almost nonexistant. OF course, we have the 9700 to thank for painting their AA modes in such poor light.

tazdevl
03-03-03, 10:34 AM
Dave did a solid job as always.

I don't think it makes sense to preview a non-retail card against a mature card that's been out on the market for several months. If Dave get's a BFG in or the like in, makes sense... and I'm sure it will make a great read.

kyleb
03-03-03, 11:04 AM
Originally posted by Typedef Enum
OF course, we have the 9700 to thank for painting their AA modes in such poor light.

na, the voodoo5 made that stuff look like crap years ago. :D

Hellbinder
03-03-03, 11:25 AM
Eve ghough I'm ok with it as a preview, you have to compare against other cards if you want the review to be of any real value. i can pull R300 numbers ferom other sites but realisticall that's no way to compare as these other numbers could be from different systems, different memory, different MB's etc. If you want a review to be valued by a diverse audience you simply must compare with other cards available and not just of the same manf. Only the wors fanboys would only consider cards from Nvidia.

No, i dont agree.

There are a number of Different types of articles. Not one of the other "previews" on the internet that DID compare ATi to Nvidia Did it correctly, they mixed modes between the two, changed settings etc. They did not have nearly as complete and detailed a breakdown of the information. IMO It is simply FOOLISH for you to suggest that every single article should be the same.

If you take the Time to read and learn from this article you will be much better Equipped when you read all the *Fluf Stuffed* unbalanced, incomplete *reviews* that come out later.

John Reynolds
03-03-03, 12:01 PM
What I'm not seeing people comment on is that the driver default AF mode is Balanced. Which means all the scores are not with Application AF mode enabled.

Another interesting tidbit is Dave's speculation that while PS1.1 is running on integer operators, 1.4 runs on the new floating point hardware via PS2.0 support. If true, this really explains why Nvidia is so hardset against 3DMark03.

kyleb
03-03-03, 12:20 PM
good points, but i just assumed that it was handeld in such a way because nvidia insisted upon it. granted that would be downright shady on nvidia's part. but asusmeing that is the case; i think Dave handled himself respetably under those restrictions by at least clearly pointing these things out.

Cotita
03-03-03, 01:41 PM
Originally posted by DaveBaumann
Nobody is stopping us from doing comparisons - we just don't do it as a matter of practice. B3D is a 3D technology site and we like to explore the nuances of a particular architecture when we get something new in like this - this follows through will all of our reviews.

The chances are that this will be followed up with a separate comparison article sooner or later, but it will be worthwhile holding on a little longer for that...

I've alway had a great respect for Beyond3d and their reviews, but...

PTRRRRRRRRRRRRRRRRRRRR!!!!!!!

who are you kidding?

In the past it made sense to compare new nvidia boards with previous generation, because they were already the fastest thing out there. It took a Geforce 2 to defeat the geforce 256 and so on.

But now, comparing the nv30 only with the ti4600 is ridiculous. Not only that, your conclusions sound like the r300 doesn't even exist, since it's not mentioned at all during the preview.

On second toughts if nvidia gave me a nv30 for making it look good, well then it absolutely rocks.

Now, when can I expect to get mine?