Something wrong with 9700?

[Doomtrooper]

Yep thanks for posting gkar1, never meant to steal your thunder..I was just showing off some of your shots. I should have mentioned you before using them .. I apologize.

Mighty envious of certain individuals...oh well I'll wait it out a bit longer

 

[rubank]

It doesn´t seem to be much of a point in taking this further, but I´ll make these last remarks.

Some of you say, in essance: â€where´s your proof, show us the pics on which you ground your comparisonâ€. This is however a moot point since I haven´t said that 9700 looks bad in comparison to this or that (o.k. I mentioned my Radeon32, I´ll get back to that). I´m saying 9700 looks bad period. And i base that statement on what can be seen in the screenshots published here and on iXBT.
I´m not claiming any other comparable card to be superior, that´s not my point at all. My point is simply that 9700 should look better than it appears to be. Now, have I said this clearly enough for anyone to understand?

I said my old Radeon looks good, and it does. But I can´t offer anything meaningful for comparison, since I don´t have all the games to produce it. I have the Nascar demo, but I´m not able to run in higher res. than 1280x960 and I can´t force AA, and editing rend_d3d.ini for bigger textures doesn´t seem to make that much difference but I can tell you that textures seem sharper than the 9700 screenshots and shadows, while still ugly, is every bit as â€good†as on the 9700.

I have 3DMark of course, but I can´t get any screenshots out of it. But please, have a look at iXBT:s screenshots from Dragothic. Look at the house in the center of the picture compared to the houses farther away at the left of the picture. Look at the ground. Comparing aniso 0 to aniso 16Q gives you what?
My old Radeon does considerably better than this, even if I can´t prove it to you. Look at the little animated comparison between 9700 and 8500; it´s too small to really tell, but it seems the 8500 has the edge here.

I have games like CMR2, STCC2, MBTR. They´re a year or more old. But they look damned good. Take CMR2: nice crisp textures and shadows that look like shadows. Take MBTR, have you seen better scenery? With my aging hardware it still looks very good compared to the 9700 screenshots.

So, the newest, highest end hardware playing current software (I assume) have nothing to offer in terms of IQ compared to my old setup? AA of course, but personally I´d take sharp textures and good aniso over AA if that would have to be the choice.

I´m disappointed, I vent my frustration and some of you get pissed. Big deal.

Finally: Foodman, you´re truly lost. Your remarks are so stupid I won´t comment.

 

[Doomtrooper]

Don't even bother comparing Ixbit ?? They didn't even have the settings correct in the 6X FSAA shots, they still say the 8500 uses Rip Mapping and the Forums are a real joy to read..although difficult with a translator.

There is plenty of shots posted by me and other members that look amazing...I don't understand your skepticism at all.

 

[DemoCoder]

They only look amazing from a resolution/aliasing point of view. None of them look amazing from a 3d graphics point of view. I mean, given the capabilities of even today's 8500/Gf3 cards, these titles stink. Look at the Doom3, Everquest 2, Halo 2, etc screenshots to see what's possible to do, even on older DX8 hardware. Or just look at ATI's SIGGRAPH demos.


That's what rubank is complaining about. Buying a $500 card to play blockly looking, no real shadows, no real lighting, games that were designed for TNT/Voodoo3 hardware, where the only benefit is smoothly out existing artifacts.

His standards for good graphics are more attuned towards lighting, shadows, and geometric detail -- artistic content, and not just deartifacting games that still look like are based on 1995 technology.

Is this ATI's fault? Nope. Is it a problem with the R300? Nope. But don't think that those screenshots are "impressive" from a graphics point of view. I find them ho-hum myself. Artistic and technological failures compared to *what's possible* on the even 3 year old hardware.

 

[Doomtrooper]

I lean strongly towards the UT shots myself, since UT was my favorite game

I think UT looks sweet in those shots.

 

[demalion]

...

Is this ATI's fault? Nope. Is it a problem with the R300? Nope. But don't think that those screenshots are "impressive" from a graphics point of view. I find them ho-hum myself. Artistic and technological failures compared to *what's possible* on the even 3 year old hardware.

But look at the thread title. If this were "Games don't take advantage of the R300" or even "Why get an R300 when games look like this?", there wouldn't have been the same type of response. There was indeed the claim that there was something wrong the R300 or drivers and that is what the replies allude to. The point you are making has already been made in these forums, and I don't recall anyone disagreeing with those particular examples in that light (though they might disagree with some other examples, like I would for NwN once all the features its offers work properly). Replies disagreeing with you (atleast mine, were there any others?) are not on those points, but on your defense of rubank when that is not what he communicated.

 

[OpenGL guy]

This is however a moot point since I haven´t said that 9700 looks bad in comparison to this or that (o.k. I mentioned my Radeon32, I´ll get back to that). I´m saying 9700 looks bad period.

If you aren't comparing it to anything, then you have no basis for "good" or "bad".

And i base that statement on what can be seen in the screenshots published here and on iXBT.

If you look at the SS:SE screenshots, you can definitely tell that the 9700 has better anisotropic filtering than the 8500.

Secondly, the guys on iXBT missed the fact that 6x AA does look superior to 4x. If you look at their shots, you can see more gradiations with 6x AA than 4x, as it should be. While it doesn't look like much magnified so much (how can you appreciate AA when it all pixelated?) it looks fantastic at normal res.

I´m not claiming any other comparable card to be superior, that´s not my point at all. My point is simply that 9700 should look better than it appears to be.

How do you know it should look better? Without comparing it to anything, you have no basis for this statement. See the logic flaw here?

Now, have I said this clearly enough for anyone to understand?

Maybe it's clear to you, but it doesn't make much sense.

I said my old Radeon looks good, and it does. But I can´t offer anything meaningful for comparison, since I don´t have all the games to produce it. I have the Nascar demo, but I´m not able to run in higher res. than 1280x960 and I can´t force AA, and editing rend_d3d.ini for bigger textures doesn´t seem to make that much difference but I can tell you that textures seem sharper than the 9700 screenshots and shadows, while still ugly, is every bit as "good" as on the 9700.

You can use "Print screen" on your keyboard to take screenshots, then paste them into Paint.

I have 3DMark of course, but I can´t get any screenshots out of it.

"Print screen" works here too.

But please, have a look at iXBT:s screenshots from Dragothic. Look at the house in the center of the picture compared to the houses farther away at the left of the picture. Look at the ground. Comparing aniso 0 to aniso 16Q gives you what?

I don't think the settings got loaded. There's an issue with the release driver where your "Control Panel" settings don't take place unless you do a mode switch. If the desktop was at 1024x768, then there won't be a mode switch when you launch 3D Mark at the same resolution.

My old Radeon does considerably better than this, even if I can´t prove it to you. Look at the little animated comparison between 9700 and 8500; it´s too small to really tell, but it seems the 8500 has the edge here.

Actually, the Radeon 8500 is suffering from a lack of Trilinear filtering in the Dragothic shot. Also, did you notice that there is no "45 degree problem" with anisotropic filtering?

I have games like CMR2, STCC2, MBTR. They´re a year or more old. But they look damned good. Take CMR2: nice crisp textures and shadows that look like shadows. Take MBTR, have you seen better scenery? With my aging hardware it still looks very good compared to the 9700 screenshots.

The 9700 can't make bad textures look better. No video card can do this. What it can do is offer more features (vertex and pixel shaders, DX9, etc.), offer enough performance to run higher res and give you great AA and anisotropic filtering.

If you're expecting new video cards to make old games look better, then aside from AA and anisotropic filtering, you're likely to be disappointed with any card. New cards can't make high-res textures out of low-res ones. New cards can't add more geometry to low poly models (unless they support Truform ). New cards can't add more color to games that use a limited palette (Quake anyone?). Maybe your expectations were not realistic?

So, the newest, highest end hardware playing current software (I assume) have nothing to offer in terms of IQ compared to my old setup? AA of course, but personally I´d take sharp textures and good aniso over AA if that would have to be the choice.

The anisotropic filtering is better on the 9700 than on previous Radeon products, so I don't see your point here. Plus AA is much faster. Performance is dramatically improved, which means higher resolutions are playable. Doesn't this all improve image quality?

I´m disappointed, I vent my frustration and some of you get pissed. Big deal.

People have tried to discuss this but you don't seem willing. Did you read my previous post? I didn't see a reply.

 

[demalion]

It doesn´t seem to be much of a point in taking this further, but I´ll make these last remarks.

Some of you say, in essance: â€where´s your proof, show us the pics on which you ground your comparisonâ€. This is however a moot point since I haven´t said that 9700 looks bad in comparison to this or that (o.k. I mentioned my Radeon32, I´ll get back to that). I´m saying 9700 looks bad period. And i base that statement on what can be seen in the screenshots published here and on iXBT.
I´m not claiming any other comparable card to be superior, that´s not my point at all. My point is simply that 9700 should look better than it appears to be. Now, have I said this clearly enough for anyone to understand?

"that 9700 should look better than it appears to be." Dave addressed this pretty clearly I think. Democoder in a way also, if you read carefully. The rest of this post is my last stab.

Your expectations are arbitrary and unspecified...it isn't conducive to much discussion except people asking for examples and comparison, which you got, and subsequent criticism if you state you don't have any.

I said my old Radeon looks good, and it does. But I can´t offer anything meaningful for comparison, since I don´t have all the games to produce it. I have the Nascar demo, but I´m not able to run in higher res. than 1280x960 and I can´t force AA, and editing rend_d3d.ini for bigger textures doesn´t seem to make that much difference but I can tell you that textures seem sharper than the 9700 screenshots and shadows, while still ugly, is every bit as â€good†as on the 9700.

? Yeah, my shots on my 8500 look way better than jvd's for NwN...but I'm not going out and stating his 9700 is flawed. If I had an 9700, my shots would look better still, because of the way I'd configure it...jvd doesn't own the game, so he hasn't tweaked it like I have. Wouldn't be very noticeable though, because NwN doesn't take advantage of any features unique to the 9700 compared to the 8500...the card did just come out, and NwN is a few months old.
In both of these cases, it DOES look a LOT better than it would on your Radeon 32 DDR simply because the software takes advantage of features your current card doesn't have. Is this a surprise to you?

I have 3DMark of course, but I can´t get any screenshots out of it. But please, have a look at iXBT:s screenshots from Dragothic. Look at the house in the center of the picture compared to the houses farther away at the left of the picture. Look at the ground. Comparing aniso 0 to aniso 16Q gives you what?

Well, I doubt you zoom in on that section and that frame like they did when you look at it on your machine, that is why we need a frame of reference for your expectations.

My old Radeon does considerably better than this, even if I can´t prove it to you. Look at the little animated comparison between 9700 and 8500; it´s too small to really tell, but it seems the 8500 has the edge here.

Hmm...well, I can't disagree with something that is not even specified, I guess.

I have games like CMR2, STCC2, MBTR. They´re a year or more old. But they look damned good. Take CMR2: nice crisp textures and shadows that look like shadows. Take MBTR, have you seen better scenery? With my aging hardware it still looks very good compared to the 9700 screenshots.

Where are the screen shots of the 9700 you are compring to for these games? This brings to mind when I said earlier: "...the way it reads it is like rubank is comparing some arbitrary sharp texture rendering in some game he has in mind to some arbitrary impression of the screenshots carried over in what seems a bit of hyperbole." You seem to be bearing this out. I hope you can take away from this that this type of comparison achieves nothing at all? Or, if you can't agree with that, atleast the difference between DemoCoder's points and your own as you've presented them?

So, the newest, highest end hardware playing current software (I assume) have nothing to offer in terms of IQ compared to my old setup? AA of course, but personally I´d take sharp textures and good aniso over AA if that would have to be the choice.

I think you should read some of DemoCoder's posts a bit more carefully. I don't agree with the defense of you in them, but if you fully understand what is being said there maybe things will be a bit clearer to you.

I´m disappointed, I vent my frustration and some of you get pissed. Big deal.

? Well, I wasn't pissed, the most I was was frustrated I believe. Criticism doesn't necessitate anger. <-see?

 

[gkar1]

As an owner of a 9700, which you obviously are not. I can say that image quality is improved by a factor of 1.5 times the radeon 8500. Aniso looks fantastic, specially in older games like Half-life. It looks even better than my old voodoo5 with 4xaa. Add 6xAA and my eyes can't believe a directx 7 game could look so good. The same goes with the giants shot i posted earlier.

Your unfounded rants make me wish more and more of that ignore button in this forum. Hint hint

 

[DemoCoder]

demalion,
yes, rubank doesn't understand that the newer cards can't improve the lighting, geometry, or textures of old games automagically.

For a brief second, I entertained the idea that the NASCAR shots were of low-poly cars "tru-formed" into smoother cars, but whose shadow projections weren't. That's the only scenario where a card could conceivably yield "downgraded" looking geometry.

 

[Dave Baumann]

For a brief second, I entertained the idea that the NASCAR shots were of low-poly cars "tru-formed" into smoother cars, but whose shadow projections weren't. That's the only scenario where a card could conceivably yield "downgraded" looking geometry.

The title doesn't support it.

 

[rubank]

@ OpenGL guy

I disagree that "good" or "bad" has to be related to a comparison. Of course this kind of statement is made in the context of my personal reference framework, but that doesn´t mean I have to compare to a specific item. I think I´m entitled to state my opinion on, let´s say, the design of a new product without reference to a comparable object, e.g. I might think a certain new car is ugly and should be able to say so without making a comparison to, say, Ferrari XX. I think the same applies here; I point to what I think are shortcomings in picture quality, I don´t compare (other than to my old h/w), and since shortcomings isn´t a good thing this is how I know 9700 should do better. Sorry if this seems illogical and unclear to you.

I´m not debating AA. This is obviously a very good thing, as far as it doesn´t deteriorate texture clarity (which has been known to happen, but I´m not saying it does so with 9700).

Very nice of you to point me to the Prnt Scrn key, not to mention Paint
Rest assured, I have some experience with a computer and I do have software that even surpasses Paint
When I say I can´t get any screenshots out of 3dMark it means just that; Prnt Scrn doesn´t work, and neither does Corel Capture (in 3dMark).

Dragothic: driver issues with aniso, you say. That´s ok. they can be fixed, but what about the blurry textures? I don´t see this kind of blurryness on my vintage card.

Yes I´ve noticed there´s no 45 degree problem. No, I don´t expect new hardware to perform miracles with bad software. Regardless of what you might think I am not an idiot.

@ DemoCoder

Of course I understand that new h/w can´t improve geometry, lighting, textures and whatnot in older games (so, you take me for a moron too). But are these screenshots from "older games"? Yes, 3dMark is a bit long i the tooth, but the rest? And as I have already said, there are oldies out there looking better, so why not use those for screenshots? I fail to see the point in taking screenshots from software so limited that you can´t show even the slightest advancement from the new hardware capabilities, software that in fact puts the new stuff at a disadvantage.

-----
From all that´s been said here, I draw the conclusion that 9700 is "perfect" but the games stink Interesting.
Some people have a problem with others stating their opinion. Not so interesting.

 

[DemoCoder]

Those games started in development probably 18 months ago, so they are targeted at hardware API generations behind. You won't see any games that really show off DX9 for atleast 2 years. Doom3, EQ2, and others are taking advantage of hardware capable of only per pixel dot product. That is, Doom3 isn't a DirectX9 level game, it was designed for the GF3 level.


Just like on game consoles, the first directx9 enabled games will take atleast 1.5 years to come out, and those will be "first generation" titles that just barely use the new features correctly. To really see what's possible, it will probably be 3-4 years.

 

[pascal]

Yeah, It is all a question of subjective viewpoint:

- You think it looks bad
- I think it looks very good

 

[demalion]

@ OpenGL guy

I disagree that "good" or "bad" has to be related to a comparison. Of course this kind of statement is made in the context of my personal reference framework, but that doesn´t mean I have to compare to a specific item.

To quote you:

I have games like CMR2, STCC2, MBTR. They´re a year or more old. But they look damned good. Take CMR2: nice crisp textures and shadows that look like shadows. Take MBTR, have you seen better scenery? With my aging hardware it still looks very good compared to the 9700 screenshots.

Your attempting to redefine what you are doing. You are saying you don't need a framework (and in fact don't have one), we are telling you your framework is completely screwed up. What you are doing is comparing the look of 1960s Porsche (these games you list here) to a 1980s VW beetle (the review shots) instead of a 1980s Porsche, and concluding that nothing new is offered by car designers after 20 years.

I think you are being dishonest because I don't think a response to my post would allow you to mask the flaw in your approach, and I note that you haven't provided one as of yet. Maybe you just haven't gotten around to it, but it doesn't look that way to me right now. To be clear, a reply to my last post to you would be in order, though I'd expect only selective replies to parts of this one if you really are committed to trying to redefine your stance.

I think I´m entitled to state my opinion on, let´s say, the design of a new product without reference to a comparable object, e.g. I might think a certain new car is ugly and should be able to say so without making a comparison to, say, Ferrari XX. I think the same applies here; I point to what I think are shortcomings in picture quality, I don´t compare (other than to my old h/w), and since shortcomings isn´t a good thing this is how I know 9700 should do better. Sorry if this seems illogical and unclear to you.

But you did compare. The statement "I don't compare (other than to my old h/w)" is really saying "I compared to my old hardware"...are you trying to trick us? And, as stated, the way you do so is flawed. Why this is so has been covered a few times already in detail, and you've never addressed it at all (see my prior post to you for an opportunity to address it), only said it is ok, and now try to redefine what you are doing.

...
From all that´s been said here, I draw the conclusion that 9700 is "perfect" but the games stink Interesting.
Some people have a problem with others stating their opinion. Not so interesting.

Have a problem with others stating their opinion? I told you exactly what the problem was, and you ensured that you avoided responding to anything related to that. That is what is "Not so interesting." in this thread.

Let me see if I start a thread on how in my opinion the GeForce 4 Ti 4600 is half as fast as the 8500 and pick one benchmark with aniso and ignore any comments by others as to how my benchmarking is flawed, what kind of reaction I'd get. Worse, it is like if I compared counterstrike framerates for a Radeon 7500 to UT 2003 framerates for the GF 4.

 

[spx]

Nvidia marketing machine gearing up?

For crises sake, what's the freakin connection?

The connection is an unsunbstansiated rumour of "bad image quality"? of the new radeons on the eve of their launch... I could be wrong

 

[rubank]

@ Demalion

But look at the thread title. If this were "Games don't take advantage of the R300" or even "Why get an R300 when games look like this?", there wouldn't have been the same type of response. There was indeed the claim that there was something wrong the R300 or drivers ...

Yes, look at the title. There´s a question mark at the end. I don´t claim that there is a flaw in the 9700, I just put forth the question whether or not the lacking IQ is h/w related, and yes - I do question the output presented.

...your defense of rubank when that is not what he communicated.

That is entirely possible since english is not my native language.

Your expectations are arbitrary and unspecified...

I haven´t said anything direct about my expectations, have I? I comment on what I see in the pictures I refer to, and I have specified several items that I put in question.

...people asking for examples and comparison, which you got...

Don´t understand.

...If I had an 9700, my shots would look better still, because of the way I'd configure it...jvd doesn't own the game, so he hasn't tweaked it like I have....

Empty words without substance.

...In both of these cases, it DOES look a LOT better than it would on your Radeon 32 DDR simply because the software takes advantage of features your current card doesn't have. Is this a surprise to you?

My point is, that apart from AA it doesn´t, again referring to the pictures I have mentioned. It wouldn´t surprise me if it did, I´m surprised it doesn´t.

Well, I doubt you zoom in on that section and that frame like they did when you look at it on your machine, that is why we need a frame of reference for your expectations.

What zoom?. Did you try clicking the pics? I have said elsewhere that I´m not able to provide screenshots from 3dMark. Does it mean I have to shutup?

Where are the screen shots of the 9700 you are compring to for these games? ...

I admit it´s not a valid comparison as such, but I think of it more in terms of reference framework. These games are widely spread, and demos are readily available. I assumed screenshots weren´t taken from inferior software, but rather more current software showing the new h/w from its sunny side. I shouldn´t have done that.

Your attempting to redefine what you are doing. You are saying you don't need a framework (and in fact don't have one)..

This makes no sense to me

..we are telling you your framework is completely screwed up..

Define who you represent, or limit yourself to first person statements.

...What you are doing is comparing the look of 1960s Porsche (these games you list here) to a 1980s VW beetle (the review shots)...

So now you confirm that the reveiw software is inferior

I think you are being dishonest...

Careful now..

..To be clear, a reply to my last post to you would be in order, though I'd expect only selective replies to parts of this one if you really are committed to trying to redefine your stance...

Your expectations are arbitrary and unspecified.

But you did compare. The statement "I don't compare (other than to my old h/w)" is really saying "I compared to my old hardware"...are you trying to trick us? And, as stated, the way you do so is flawed...

How would this be a trick? I think I´m perfectly clear. The alleged flaw is in your mind.

Let me see if I start a thread on how in my opinion the GeForce 4 Ti 4600 is half as fast as the 8500 and pick one benchmark with aniso and ignore any comments by others as to how my benchmarking is flawed, what kind of reaction I'd get. Worse, it is like if I compared counterstrike framerates for a Radeon 7500 to UT 2003 framerates for the GF 4.

And you accuse me of redefining? When did this thread turn into a question of framerates? This has nothing to do with the question at hand. IQ is a matter of opinion and subject to discussion; I haven´t seen anyone state that 80 FPS is in fact 43 FPS.

 

[jvd]

demalion - give me your config file or settings for neverwinter nights tell me what to do and i will tweak it and see if it helps.

 

[Sharkfood]

When I say I can´t get any screenshots out of 3dMark it means just that; Prnt Scrn doesn´t work, and neither does Corel Capture (in 3dMark).

Use F12. 3DMark locks out all keystrokes during a run, except F12 which is it's own built-in screencapture utility. Check the 3dmark directory for the file it dumps.

 

[demalion]

@ Demalion

But look at the thread title. If this were "Games don't take advantage of the R300" or even "Why get an R300 when games look like this?", there wouldn't have been the same type of response. There was indeed the claim that there was something wrong the R300 or drivers ...

Yes, look at the title. There´s a question mark at the end. I don´t claim that there is a flaw in the 9700, I just put forth the question whether or not the lacking IQ is h/w related, and yes - I do question the output presented.

Are you an nVidia employee playing troll and spreading FUD as the 9700 launches? Your statements persist in an unfounded and unspecified claim of image quality problems, and your defense consists of comparisons of completely different games you own to screenshots of other games you do not. You seem to be trying to avoid the issue of whether the one point of comparison you say you observe directly can be substantiated by saying you can't take a screenshot, and ignore that what you are comparing to is an isolated static image that by its nature will not look like what you see going by at > 30 frames per second on your screen.

Would this be fair? You can't say that I'm claiming you are an "nVidia troll" without being a hypocrite, so where does that leave us when it comes to attacking the stance I present in that paragraph? And, unlike you, I have a chain of specific evidence in the form of your posts in this thread to back it up, so I'm actually in a much better position than you are.

...your defense of rubank when that is not what he communicated.

That is entirely possible since english is not my native language.

Your expectations are arbitrary and unspecified...

I haven´t said anything direct about my expectations, have I? I comment on what I see in the pictures I refer to, and I have specified several items that I put in question.

No, your text quite clearly states expectation. FYI, you can make strong assertions even by putting a "?" at the end. As I hope I illustrated above.

...people asking for examples and comparison, which you got...

Don´t understand.

Well, you only provide a partial quote here. The sentence was saying that your question, without any proof or evidence, made the case that there was something wrong with the 9700 based on screenshots from a review, compared to completely different images from completely different games. As a result, people asked you for useful comparisons, and explained how your comparisons were not useful, and criticized you when it became evident that you don't have any.

...If I had an 9700, my shots would look better still, because of the way I'd configure it...jvd doesn't own the game, so he hasn't tweaked it like I have....

Empty words without substance.

Are you trying to misrepresent me, or do you really not understand the meaning of my words?

...In both of these cases, it DOES look a LOT better than it would on your Radeon 32 DDR simply because the software takes advantage of features your current card doesn't have. Is this a surprise to you?

My point is, that apart from AA it doesn´t, again referring to the pictures I have mentioned. It wouldn´t surprise me if it did, I´m surprised it doesn´t.

Well, I was talking about NwN. I can assure you, that it most definitely would, for the reasons I state. I also explained clearly how comparing Dungeon Siege or 3dmark on one card to CMR 2 and whatever else you didon another is about as useless a comparison to determine what a video card offers over another as possible.

Well, I doubt you zoom in on that section and that frame like they did when you look at it on your machine, that is why we need a frame of reference for your expectations.

What zoom?. Did you try clicking the pics? I have said elsewhere that I´m not able to provide screenshots from 3dMark. Does it mean I have to shutup?

Oh, you are comparing the full screen shots...I thought your description was referring to the isolated section of the image (but you are right that it isn't zoomed).

OK, I'll take a screenshot of 3dmark and compare it to their shot of the 8500, as that will be a direct comparison with a nice solid frame of reference...

Heh, OK, I've done that side by side and this clarifies some of your reticence in the criticism of your comparisons. Namely, one of your comparisons is quite valid (though unsupported by you) and from your view that factor is a focus rather than the extreme problems with the rest of your comparisons. Let me clarify:

You are quite correct about the relatives fuzziness of 9700 shots, quite likely in comparison to what you see on your card as well. The thing is the 8500 shots are just as fuzzy compared to my 8500 shot. Pretty much the exact same spot (I 'm not sure how they got that image without a streak of flame in it, I can't duplicate that), and if this is in dispute I can provide the screenshot that were what we asked you for for this particular point.

So this seems to show that within the "frame of reference" of xbit's testing, the 9700 doesn't have problems, but within the larger framework, xbit's testing does have problems. Please read with care...I'm sure wording this complex in your first language would be above my ability to understand even after a few years of study, so I'm not trying to look down on you when I ask you to make sure you understand that sentence fully.

Your confusion stems from when we criticize you for your 3dmark comparison, we are asking for a screenshot, and when we are criticizing your other comparisons (and I continue to do) we are saying your rationale behind them are completey invalid.

Where are the screen shots of the 9700 you are compring to for these games? ...

I admit it´s not a valid comparison as such, but I think of it more in terms of reference framework.

Well, if you'd done that earlier you'd have been met with less criticisms. In fact, if you'd only focused on 3dmark the thread would be a few less pages in length...with a screenshot, we'd likely still be on the first page by the time we established what we know so far. Most of my criticism is because "it's not a valid comparison as such", but atleast you've admitted that now.

These games are widely spread, and demos are readily available. I assumed screenshots weren´t taken from inferior software, but rather more current software showing the new h/w from its sunny side. I shouldn´t have done that.

I can't say it any more clearly than this: your set of assumptions here exhibit major flaws, and though you construe it to be other people taking you for a "moron" (your word I believe), it is just that the conclusions they reached (which did not include the word "moron") are pretty logical based on how flawed those assumptions are. The details to this are provided in adequate detail elsewhere.

Your attempting to redefine what you are doing. You are saying you don't need a framework (and in fact don't have one)..

This makes no sense to me

I can understand not having English as a first language and how subtle phrasing can be unclear. Read the whole text together at once, because you break it up here...reading it together at once is required to understand my point. Or, read from "But you did compare" later in my text for a restatement of the same general principle.

..we are telling you your framework is completely screwed up..

Define who you represent, or limit yourself to first person statements.

"screwed up" is my personal label for the various comments several people, who you've responded to so you know who they are, have made about your support for your comments. Since the words "screwed up" appears in my text and not in theirs, the phrase is attributable to me. There is no claim of representation intended to be included in that statement, though the implication is there that others do find the same faults in your statements (and my understanding of their text is that they have expressed that they do).

...What you are doing is comparing the look of 1960s Porsche (these games you list here) to a 1980s VW beetle (the review shots)...

So now you confirm that the reveiw software is inferior

Well, some people like the way a VW bug looks more than a Porsche, or don't care about the difference...but you were going on about specific styling points you attributed to be positive, so in that context yes the "review software is inferior".

I think you are being dishonest...

Careful now..

..To be clear, a reply to my last post to you would be in order, though I'd expect only selective replies to parts of this one if you really are committed to trying to redefine your stance...

Your expectations are arbitrary and unspecified.

Hmm...that was a particularly meaningless way to respond to a partial quote, hence my impression of dishonesty on your part. Read that bold part again. I hope you can understand where my impression of dishonesty is coming from.

But you did compare. The statement "I don't compare (other than to my old h/w)" is really saying "I compared to my old hardware"...are you trying to trick us? And, as stated, the way you do so is flawed...

How would this be a trick? I think I´m perfectly clear. The alleged flaw is in your mind.

To...further illustrate:

You begin a statement "I don't compare", and then carry on with the premise that you don't compare. But...the full statement is "I don't compare (other than to my old h/w)", and I point out that full statement is "I compare to my old hardware" phrased in such a way as to excuse carrying on with the premise that you don't compare (when in fact you do). Hence the term "trick", and my impression of dishonesty.

Let me see if I start a thread on how in my opinion the GeForce 4 Ti 4600 is half as fast as the 8500 and pick one benchmark with aniso and ignore any comments by others as to how my benchmarking is flawed, what kind of reaction I'd get. Worse, it is like if I compared counterstrike framerates for a Radeon 7500 to UT 2003 framerates for the GF 4.

And you accuse me of redefining? When did this thread turn into a question of framerates? This has nothing to do with the question at hand.

Hmm...well, it is a matter now for me of whether my impression of your dishonesty or your admission of some English unfamiliarity is the reason you attempt to describe my analogy as some sort of actual argument. To be fair to you, I'll explain regardless of my own doubt: This example is an analogy (an example of something different to show a parallel to a subject being discussed) that is purposefully flawed in the same lack of objective data and relevant context as your argument. To be a more accurate analogy, it would have to be phrased in the form of a question.

IQ is a matter of opinion and subject to discussion; I haven´t seen anyone state that 80 FPS is in fact 43 FPS.

You didn't read the the analogy very closely at the end. I'm pretty sure your Radeon card (or the faster 7500 I mention) gets higher framerates in counterstrike than a GF 4 does in UT 2003, and using that data someone could make a much better case than you have that there is something wrong with the GF 4's speed (namely, there is direct and objective data, namely fps, with a similar lack of a useful frame of reference as your argument...but your supposition lacks even the direct and objective data).