Feedback on GeForce 6800U Preview

[cornelious0_0]

Originally posted by MemoryInAGarden
I agree that there should be more raw speed comparisons, but benchmarking at a lot of different resolutions and different AA/AF settings would be incredibly time consuming and frustrating for Kyle and Brent. I'm sure they have better things to do than benchmark at a plethora of different settings.

Then explain why it isn't done the same as it always was.....I was MUCH more pleased with the old method becuase it was MUCH more like the "every res, every lvl" approach, but not enough to annoy the reader. There is a happy medium.....and we HAD it.

 

[neisius]

Originally posted by cornelious0_0
Come on [H].....stop trying to feed me what YOU think is the "best playable settings" in these reviews.....cus it means nothing to so many of us

yeh, what he said

 

[Flipside]

Originally posted by STL
The new type of "review" sucks. Others in this thread have already pinpointed out why. Trying to equalize framerates and comparing the resulting image quality settings is morally bankrupt, and hides the fact that the GeForce 6800 Ultra is blisteringly faster than current cards.

You're right, but yet so very wrong. I think the new review method was effective, because with today's blistering fast hardware 150FPS over 100FPS isn't going to add anything to the gaming experience. What you are going to notice is if you can run the resolution higher and ramp up the visual goodies.

The truth is that you can't compare apples to apples, because the whole point of introducing new technology, is that it'll bring something new to the market that current products cannot provide. So what if the 6800U scores 50 FPS more than the 9800XT's already impressive speed - if it can match the 9800XT, but add new visual features and higher resolutions, then it really proves that this new technology is worth the upgrade.

 

[cornelious0_0]

if it can match the 9800XT, but add new visual features and higher resolutions, then it really proves that this new technology is worth the upgrade.

I like that point you made, but my previous statments still stand. It's obviously nice to see that it can perform the same with higer IQ.....but it isn't completey out of the question to have the rest of the information there as well. Some of us buy new video cards for different reasons and I personally WANT to know that it is 50fps faster in a certain game.....whether I can notice that 50fps or not.....it's there.

Hate me for it but that's still my opinion, and I still hate the system.

 

[MrHappyGoLucky]

Originally posted by evilchris
He is a massive fanboy. That's what he meant. If 420 came out first, and you showed it owning the 5950, he'd say it was a fair comparison.

Not really...... The R3xx core already owns the 5950

 

[MrHappyGoLucky]

Originally posted by cornelious0_0
It's obviously nice to see that it can perform the same with higer IQ

I think you are getting IQ confused. The 6800 series still is lacking in the IQ department compared to the R3xx series. Take a look at most screen shots and you will see they are still shortcutting quality to gain performance. Yes the 6800 can run at higher resolution with AA and AF, but didn't look too good.

Originally posted by cornelious0_0

Hate me for it but that's still my opinion, and I still hate the system.

You can never hate someone for their opinion.

 

[evilchris]

Originally posted by MrHappyGoLucky
I think you are getting IQ confused. The 6800 series still is lacking in the IQ department compared to the R3xx series. Take a look at most screen shots and you will see they are still shortcutting quality to gain performance. Yes the 6800 can run at higher resolution with AA and AF, but didn't look too good.

Actually you can't tell the difference between them anymore. Mike over at NVNews did what no reviewer is doing. He had a poll, and you had to pick 1 of 2 screenshots which looked better to you. He didn't tell which was the XT and which was the 6800. most people had picked the 6800 as looking slightly better before he revealed the source. The NV40 is not shortcutting ANYTHING. The image quality is stellar and on par with anything else out there. You are just so obsessed with ATi I think your little brain just TRIES to find faults in the 6800 image, while ignoring the XT image. The *only* current issue is Farcry, which is not the "NV40 shortcutting". That blockiness problem will be fixed. The GAME is running NV3x code extensions on the NV40.

 

[conscript]

I don't know what all the hub bub is about. i like the different review analysis. Would I base a h/w purchase on one review? Absolutely not, no matter how much I trusted the reviewer. But let's face it guys. When products like the 6800 come out, there's dozens of reviews up the same day, and if you're going to think about shelling out 400-500 bucks for a card, I have to assume you'll read most if not all of those reviews. I think it is only a benefit that we have reviewers look at things different ways. When I will have a problem with it is if everyone moved to using Kyle and Brent's method. At the end of the day I want to see b&w apple to apple comparisons too. But last i checked my url wasn't stuck on the [h], and I can get that data somewhere else. Guys, I think you should embrace this, and I truly hope a few other sites come up with different standards for testing. I want as much different data as possible so I can make my OWN decision... until ati & nvidia start sending me sample to test myself

p.s. thank god ati is releasing their 420 soon too, because once i can see a comarison between it and the 6800, i will most certainly begetting one of the two... my only hesitation will be whether I should wait a little longer for pci-e...

 

[Sandman]

Personally, I kind of like the new review format. Adding some apples to apples comparisons to it would make it even better. And I don't think that you'd have to add too many of them, so it won't be too much extra work.

The only concern I have is that there may be 2+ different settings that get approximately the same framerates. For example, let's say that some video card gets about 60 fps at 4xAA and 4xAF, and also gets about 65 fps at 2xAA and 16xAF. Which do you choose?


Finally, as a bit of an aside, is there any chance that we can get a short comparison of this card vs. some of nvidia's older cards? I'd love to see how much faster this card is than a GF4, GF3 and maybe even down to a TNT2 or so (or lower if you really want to get ancient, riva128 anyone? ). Just put them all at 1600x1200 with whatever IQ settings they can all do equally and show a graph. So we're talking about a 1 page article with a graph and some historical tidbits. But it'd be fun to see how far we've come. Also, people like me with pretty old cards (GF3) will see how much improvement there is in an upgrade to the 6800u.

 

[conscript]

Originally posted by MrHappyGoLucky


You can never hate someone for their opinion.

actually, that's about the only reason you can hate someone. otherwise, we'd all be the same and hate everyone including ourselves

 

[zoltan]

Here's my thought.

This would have been an excellant review if we actually could get this product. But before hand we all want to know just how much better it is than the past generation by an IQ and not differn't settings. People only want to know this type of "review" after they get the card so "we" can set accordingly and know how well "we" should be doing. Good job [H] but bad timing.

 

[MemoryInAGarden]

Originally posted by zoltan
Here's my thought.

This would have been an excellant review if we actually could get this product. But before hand we all want to know just how much better it is than the past generation by an IQ and not differn't settings. People only want to know this type of "review" after they get the card so "we" can set accordingly and know how well "we" should be doing. Good job [H] but bad timing.

Everyone else would have had a review out anyway. If they didn't put out a review people would be complaining about that. It's a lose-lose situation for them.

 

[conscript]

Originally posted by zoltan
Here's my thought.

This would have been an excellant review if we actually could get this product. But before hand we all want to know just how much better it is than the past generation by an IQ and not differn't settings. People only want to know this type of "review" after they get the card so "we" can set accordingly and know how well "we" should be doing. Good job [H] but bad timing.

umm, i don't know about you, but I don't like spending half a grand on a video card and roll the dice if it can handle 1600x1200 with 8x/4x after the fact...

 

[gb25]

Originally posted by Brent
Um, but you are seeing that. That is the whole point, to see how much better the card can play the games compares to another card.

For example FS 2004, the 6800U plays it smoothly at 1280x1024 4XAA/16XAF vs. the 9800XT at 1280x1024 2XAA/8XAF, AND with higher performance as is shown.

Ok, so where do we see how the 6800u's image quality at 1280x1024 4XAA/16XAF compares to the image quality of the 9800xt at 1280x1024 2XAA/8XAF?

 

[cornelious0_0]

I think you are getting IQ confused. The 6800 series still is lacking in the IQ department compared to the R3xx series. Take a look at most screen shots and you will see they are still shortcutting quality to gain performance. Yes the 6800 can run at higher resolution with AA and AF, but didn't look too good.

First off, you're confusing what I was saying, don't assume you knew what I "meant" and decide to undermind me like that.....it's irritating.

I didn't say that the 6800 had BETTER image quality, I just said that it performs better at higher IQ settings.....which it does. I agree that the 9800XT still has SLIGHTLY (to the point where it's hard to tell looking at most still shots, so you would NOT notice the difference in a game) better AA, but the AF on the 6800 does look better to me.

Next time I (or anyone else) says something you don't understand, please, PM them about it. Do not assume to much around here, because the internet is a horrible medium for getting across meaning and emphasis.

 

[FrgMstr]

Originally posted by Nexu
I'd have to say I agree with these two. I prefer the apples to apples resolution and quailty settings then tell me why the two aren't the same (i.e. IQ differences/rendering and so on). The other reviews I read (firingsquad, anand and tom's) all took the 6800U to 1600-1200 with AA and AF cranked up and that is usually when the 6800U started to separate from the others. Running tests at 12x10 (which were shown to run fine on the older hardware and be CPU limited in many cases) IMO took away from this GPU review.

edit: sorry about the improper quoting :/

Then I suggest you stick with other sources for your information. No hard feelings.

 

[FrgMstr]

Originally posted by onetwo
None of benchmarks made it clear that:

1) The NV40 is greatly held back by the CPU in most games. The one benchmark that clearly hinted at this wasn't supported by the text or supplementary benchmarks necessary to come to any sort of conclusion.

2) The NV40 is 2 - 3 times more playable at 1600x1200 and up resolutions than exisiting hardware -- many games can have every bell and whistle enabled and still run ~60fps.

3) The new 8xAA implementation results in a massive performance hit and is not to be considered usable.

These are major insights that your 'new' reviewing system failed to impart upon the reader.

IMO, try differentiating yourselves with great text, not unorthodox tests.

1. Maybe in benchmarks, but show me that in realworld gameplay please.

2. We did not find this to be the case in REAL WORLD GAMEPLAY.

3. That is why we did not use it.

 

[FrgMstr]

Originally posted by cornelious0_0
Come on [H].....stop trying to feed me what YOU think is the "best playable settings" in these reviews.....cus it means nothing to so many of us, and you've lost a number of loyal readers (including myself) over it.

If how the game really plays means nothing to you , then I suggest you do not read our articles.

Bottom line is that it is about WHAT WE THINK. If you do not like our logic, then I appreciate that, but I am not going to do it "wrong" as I see it.

Bottom line is that we have NEVER run HardOCP like a business, and have always done what we thought is right, no matter the impact to readership. I am not going to stop now either.

 

[gb25]

Originally posted by FrgMstr
Another thing that is interesting about this journey is that we have been doing video card reviews this way for months and getting excellent feedback. It is only at this junction that we got such a hostile reply. It seems to me that everyday folks that visit our site in order to try to figure out what to buy that week have been very happy with our new format.

Personally, the only reason I'm speaking out now about the way you are doing video card reviews is that this is the first review I've read using the new format. Had I read an earlier review using this new format, I probably wouldn't have said anything anyways because it doesn't really matter that much to me. I don't have any problems with going to another site to find a review. Anyways, I'm just adding my 2 cents now because the issue has been raised.

Keep the opinions coming. I want to hear the good, the bad, and the ugly. I have yet to read anything that makes me want to change formats.

I have a few complaints with the new review format. Most of which are pretty insignificant to me. But my main complaint is the issue that fugu brought up. You say that it's something that "can be done", but in my opinion, it's something that absolutely must be done.

In my opinion, for a video card comparison, image quality and performance go hand in hand. Without a screenshot, fps are just numbers. Without benchmarks, screenshots are just pretty pictures. In this preview, you just give us a lot of numbers and pretty pictures that have no correlation to the other. Which makes them pretty useless to me.

Take the Battlefield Vietnam review for example.

http://hardocp.com/article.html?art=NjA2LDc=

In the chart/benchmark I see that the 6800u at 1280x960 4xaa/8xaf performs comparably to the 5950u at 1024x768 noaa/noaf and the 9800xt at 1024x768 noaa/8xaf. The min, max, and average framerates are all pretty close, and the line graph is also similar for all three. Fine, I understand that. But how are we supposed to know how much better the image quality for the 6800u is than the 9800xt?


Benchmarks:
6800u - 1280x960 4xaa/8xaf
9800xt - 1024x768 0xaa/8xaf

Screenshots:
6800u - 1024x768 4xaa/8xaf
9800xt - 1024x768 4xaa/8xaf

Taking the benchmark/chart in to consideration, I know that the 6800u performs well at the settings used in the benchmarks. I also know that the 9800xt performs well at the settings used in the benchmark. But I can't tell what the image quality for either card looks like at the settings used for the benchmarks. Instead, I get to see what they look like with completely different settings.

So, tell me, how am I supposed to know how much better the 6800u at 1280x960 4xaa/8xaf looks than the 9800xt at 1024x768 0xaa/8xaf without screenshots? What good is a screenshot of the 9800xt with 4xaa when the only benchmark is not using any aa at all? What good is a screenshot of the 6800u running at 1024x768 while the benchmark is done at 1280x960?

 

[FrgMstr]

Originally posted by cornelious0_0
Exactly, people need to know which card is FASTER,

Actually, we don't care which is "Faster". Faster is relative. I want to know what card will give me the greatest level of IQ at a playable framerate and that is what we are going to base our reviews on. It is just that simple.

 

[FrgMstr]

Originally posted by Sandman
Personally, I kind of like the new review format. Adding some apples to apples comparisons to it would make it even better. And I don't think that you'd have to add too many of them, so it won't be too much extra work.

As mentioned earlier in this thread, this is already decided upon being a normal addition to the current format.

 

[davidj]

I THINK that I get it. The [H] team feels that their procedures produce the most informative results possible. They have given this chance for community feedback and have not seen any hard data that compells them to make LARGE changes.

So do not expect the [H] to compromise it's ethics just because some disagree with it's way of doing things. When I really believe in something I will stick by my guns.

Running the [H] like a business and being a buisness are distinctive. I'm looking at all of the advertisers as I type so, TEH [H] IS A BUISNESS.

None of these threads have to be emotional. We are free to read the reviews here, elswhere or both.

Looking forward to the next one Brent

May the Geforce be with you

 

[MrHappyGoLucky]

Originally posted by conscript
actually, that's about the only reason you can hate someone. otherwise, we'd all be the same and hate everyone including ourselves

I tend to researve my hatred towards people based on thier actions rather than their opinion.

 

[cornelious0_0]

2. We did not find this to be the case in REAL WORLD GAMEPLAY.

Oh really? It might not be 3x as fast (that's kind of an exageration) but then how do you respond to these numbers that cleary show it EASILY outdoing the 5950 and 9800XT at high res/AA+AF.....often by about 100%, and never any less then about 50%.....which is still a hell of a lot more then OUR review showed:

Call of Duty @ 1600x1200 /w 4xAA 8xAF

IL-2 Sturmovik @ 1600x1200 res /w 4xAA 8xAF

Quake III @ 1600 res /w 4xAA 8xAF

Splinter Cell @ 1600 res /w 8xAF

Tomb Raider AOD @ 1600 res

Lock On: Modern Air Combat @ 1600 res

Unreal2k4 @ 1600res /w 4xAA 8xAF

Halo @ 1600res

Far Cry @ 1600res /w 4xAA

I spose you're gonna come up with something on how these aren't credible results.....'cus they (honestly, not pesonally) kick the shit out of [H]'s results.....and really show the cards true performance. It is true that some games at some resolutions still show all teh cards neck and neck check out Unreal2k4 @ 800x600 with 4xAA/8xAF.....it's barely beating out the previous gen cards.....by a mere 1.7fps. Kick the res up to 1600x1200 and the gap grows to 21.8fps.....leaving the 6800 the only card thats half playable at that setting. This is shown time and time again with many games, I'm no just making this up. If [H] is trying to say that it ISN'T that much faster at high detail /AA AF settings they better look again, 'cus their preview is the "odd man out" in this case and looks very strange.

If how the game really plays means nothing to you , then I suggest you do not read our articles

I dont' know how the hell you managed to turn that around but you really need to stop playing with my words here, that's not at all what I was talking about. It's important for a game to be playable at high quality settings, but it's the fact that you're TELLING me what setting is best, and REALLY limiting the hell out of the results you've giving me is what pisses me off.

Actually, we don't care which is "Faster". Faster is relative. I want to know what card will give me the greatest level of IQ at a playable framerate and that is what we are going to base our reviews on. It is just that simple.

Again, you're playing with my words. If you would just show us each card at each resolution, at each AA/AF level (which really wouldn't take that long at all) then we could figure out for ourselves which card is faster AND gives the best play experience at high details. Its almost like you're assuming that we're not able to do this and so you have decided to tell us what has been deemed "acceptable". I don't understand how you can say the term "faster" is relative.....it isn't. If a card is faster, it's faster.....plain and simple. I know that sounds quite simplistic, but it really isnt' that complicated. If you would give us ALL the numbers we coudl see which card is faster, or to phrase it differently, which card is fastest at high IQ settings......which is EXACTLY what you say you're going for with these reviews of yours.

I know that you're not really a dumb guy but these reponses you've been coming up with are a little pathetic FrgMstr, you actually expect me to buy any of this. I'm not going to take you (or the site) seriously if you keep turning my words around, underminding me, and talking like you know what's best for us. That last one isn't directed at you personally, moreso at [H].....because its just teh feeling I'm left with after reading that preview.

Come back at this with whatever kind of reponse you want, but I dont see how you're gonna find any way to get around the fact that those numbers above speak for themselves.....and THAT'S reality, whether you guys like it or not. I'm sorry, but I just don't like the fact that you're making nvidia's gem (the fastest thing out there) look like your everyday card.....while everyone else on the net is giving us the truth. That preview gives a person no reason to be excited about the card. bumping the res up one more step and attaining the same fps is nothing impressive, and [H] really messed this one up buy not showing ANY results from high resolution/high AA+AF settings, which is where the card shines.

It's nice that you say you're trying to make the card look good.....but you're also stealing away a lot of it's thunder with a POS preview like that.....I'd be wary to purchase one if all I read was your preview. Thankfully I've been elsewhere (before coming here to boot) so I've been able to see the truth, and I'm still excited about the card.

 

[MrHappyGoLucky]

It's nice that you say you're trying to make the card look good.....but you're also stealing away a lot of it's thunder with a POS preview like that.....I'd be wary to purchase one if all I read was your preview. Thankfully I've been elsewhere (before coming here to boot) so I've been able to see the truth, and I'm still excited about the card.

Actually I would place more value on understanding how well the game played instead of reading all the drone reviews that run the same exact benchmarks everytime. It makes me more inclined to read a review that wasn't a copy/paste of the last review and simply did a search and replace for the model number.

I think you can't get past the fact a video card running 80fps can look better and provide a much more enjoyable gaming experience than a different video card running 160fps even though they have the "same settings". Based on the screenshots, it's for real.

Besides the whole Far Cry graphical errors, here are two screenshots to look at. Now tell me the 6800 looks better. I'm sure if I went through all the screenshots I would find the 9800ProXT looks better 3 out of 5 times. That is way more important to me than raw FPS.

Take a look at the tail rutters.

9800ProXT

6800Ultra

 

[Brent_Justice]

Originally posted by cornelious0_0
It's nice that you say you're trying to make the card look good.....but you're also stealing away a lot of it's thunder with a POS preview like that.....I'd be wary to purchase one if all I read was your preview. Thankfully I've been elsewhere (before coming here to boot) so I've been able to see the truth, and I'm still excited about the card.

We aren't stealing any thunder away from the card. If you read the preview you will see that our results showed that it was able to play every game faster and with more iq than the 5950U and the 9800XT, it won every game test.

How is that stealing thunder by showing it to be the best card for gaming?

Perhaps you missed this paragraph in the conclusion as well?

The GeForce 6800Ultra delivered the best realworld gameplay in every game we tested. NVIDIA has vastly improved performance over the GeForceFX series with the GeForce 6 series GPU. In most of today's popular games, the GeForce 6800Ultra noticeably outpaces the ATI Radeon 9800XT.

 

[sKiTz0]

small idea to throw in.. GIFs that switch between identical screenies on different cards so theyre easy to compare would be cool.

 

[evilchris]

Originally posted by MrHappyGoLucky


Take a look at the tail rutters.

9800ProXT

6800Ultra

Heh, I don't see a difference at all. Sorry Happy, NVIDIA's AA is just as good now.

 

[LegendKiller]

Originally posted by evilchris
Heh, I don't see a difference at all. Sorry Happy, NVIDIA's AA is just as good now.

You=blind?


Kyle and Brent,

When you first came out with this method I applauded it, nobody in the benchmarking and review area has even considered performing reviews in this fashion. The work that it takes to create a review like this must be staggering. The reviews I did for GotApex look like childs play compared to what you have produced.

Numbers are just that, numbers. They tell you nothing other than something is 10% faster. But what does that 10% get you? What good is that 10% to me? Does it equal 10% better resolution? How do you quantify that?

You have done so in your reviews and I find that truly insightful. While there are people who will always have to measure themselved by a yard stick, you have shown some a way to measure not quantity but quality.

For those of you who cannot get this fact through your thick skulls, then perhaps you do need a mix of benchies that will provide it, that is up to Kyle and Brent. However, when I first saw the 6800u and its associated graphs, I immediately knew how hot it was. Too bad many people do not think like some do.


Again, keep up the great work, you guys are rare among a flock of "status quo" review sites.


LegendKiller
www.gotapex.com reviewer.

 

[evilchris]

Originally posted by LegendKiller
You=blind?

20/15

If you see a difference, it's your fanboy brain. Go check out nvnews where Mike posted comparison shots but did NOT tell which was which. You'll be suprised. Next time try being more objective as in " Hey, I see a difference in ( name a part of the plane here ), you don't see that? " Not "You=blind?". That doesn't support your belief really well. I alt tabbed those two about 50 times and wouldn't be able to guess which was which. NVIDIA's AA is as good as r3xx now, and sometimes better depending on the game. ( judged from SS's on the various reviews )

 

[dderidex]

Originally posted by LegendKiller
Numbers are just that, numbers. They tell you nothing other than something is 10% faster. But what does that 10% get you? What good is that 10% to me? Does it equal 10% better resolution? How do you quantify that?

C'mon, DUH.

'Just numbers' tells you how much bigger your intarw3b p3n0s will be if you buy this new card.

Seeing that it can run 6xAA at the same FPS another card can only run 4xAA doesn't tell you how big your intarw3b p3n0s will be, does it?

But, running all tests at 1024x768 with no FSAA and no Aniso and showing a 25% differnce in score allows you sleep well knowing your intarw3b p3n0s will be growing by a full quarter!

Who cares how much more QUALITY it can put out - that's irrelevant! The only thing that matters is the length of the bar graph! Not how you use it!

 

[lzsbkl]

its all about the Benjamins
PS did you have a nice vacation in Canada lads

 

[Vagrant Zero]

Only if they stayed away from drunk Canadian men with really poor vision.

*shudders*

 

[@trapine]

[START RANT HERE]

its all about the Benjamins
PS did you have a nice vacation in Canada lads

So now because kyle and brent deside to show people The performance the card's are REALLY capable of in GAMES the very thing you BUY a CARD FOR?
They are suddenly being somehow payed or influenced by ATI
Just as a side note its going to be most intresting to see if the same outrage is diplayed when they do the same kind of review on the R420?
TO me this just all reaks of people with a strong prefrence for nvidia that really dont give a shit about anything nvidia actually have brought to the table an is new Apart from the fact that it can do 12000+3Dmarks and gets 5000FPS=Pathetic!

Some of you people need to Be reminded why you buy the damed card for and thats to play games.Not so you can get a moral hardon about why your card is so much better than the compatiton at 3DWANK
[/END RANT HERE]

 

[lzsbkl]

Originally posted by @trapine
[START RANT HERE]


So now because kyle and brent deside to show people The performance the card's are REALLY capable of in GAMES the very thing you BUY a CARD FOR?
They are suddenly being somehow payed or influenced by ATI
Just as a side note its going to be most intresting to see if the same outrage is diplayed when they do the same kind of review on the R420?
TO me this just all reaks of people with a strong prefrence for nvidia that really dont give a shit about anything nvidia actually have brought to the table an is new Apart from the fact that it can do 12000+3Dmarks and gets 5000FPS=Pathetic!

Some of you people need to Be reminded why you buy the damed card for and thats to play games.Not so you can get a moral hardon about why your card is so much better than the compatiton at 3DWANK
[/END RANT HERE]

I am no nvidia fanboy (i have a 9600pro in my rig) but the whole article smacked of bias and doesn't show what a huge leap forward the 6800 actually is. My first post was meant tongue in cheek by the way.

 

[Themacabre]

Originally posted by evilchris
20/15

If you see a difference, it's your fanboy brain. Go check out nvnews where Mike posted comparison shots but did NOT tell which was which. You'll be suprised. Next time try being more objective as in " Hey, I see a difference in ( name a part of the plane here ), you don't see that? " Not "You=blind?". That doesn't support your belief really well. I alt tabbed those two about 50 times and wouldn't be able to guess which was which. NVIDIA's AA is as good as r3xx now, and sometimes better depending on the game. ( judged from SS's on the various reviews )

No he's right...if you don't see a difference in the tail then...

A) You didn't look at the screenshots

or

B) You're blind

This isn't about being a "fanboy" and maybe it's just a problem with that particular game (driver problem?)...but saying you don't see a difference at all makes you look foolish.

 

[PopCorn]

Originally posted by LegendKiller
You=blind?


Kyle and Brent,

When you first came out with this method I applauded it, nobody in the benchmarking and review area has even considered performing reviews in this fashion. The work that it takes to create a review like this must be staggering. The reviews I did for GotApex look like childs play compared to what you have produced.

Numbers are just that, numbers. They tell you nothing other than something is 10% faster. But what does that 10% get you? What good is that 10% to me? Does it equal 10% better resolution? How do you quantify that?

You have done so in your reviews and I find that truly insightful. While there are people who will always have to measure themselved by a yard stick, you have shown some a way to measure not quantity but quality.

For those of you who cannot get this fact through your thick skulls, then perhaps you do need a mix of benchies that will provide it, that is up to Kyle and Brent. However, when I first saw the 6800u and its associated graphs, I immediately knew how hot it was. Too bad many people do not think like some do.


Again, keep up the great work, you guys are rare among a flock of "status quo" review sites.


LegendKiller
www.gotapex.com reviewer.

These thoughts mirror my own from earlier in the thread, but I thought this post was good enough to be quoted again.

Ditto.

Once again, if I want the cookie cutter review, I'll go elsewhere. After reading |H|'s preview (remember this wasn't a review), and then reading other 'reviews', I feel that |H| balanced things very well.

Esoteric max FPS numbers and 3dmark numbers with Beta Drivers mean Jack shit. It's all about how the card plays and handles games. If all you care about is benchmark numbers and bragging rights to increase your e-penis size, then whatever.

Nvidia has made great strides with this Card. But all the facts aren't in yet. A real Driver set from nvidia and ATI's card coming up to bat will complete the puzzle. Pretty simple really.

Until then, I'm glad to have a review site here that doesn't simply follow the pack.

 

[Anderu]

My take is, that unless ATI has something up their sleeve that takes the world by complete surprise and puts NVidia solidly in the backseat again, what we have here is solid competition on a more or less even playing field.

We have not had that in over a year. I used NVidia cards from the Riva 128 all the way up to the Ti4600. But when the 5800 line came along, I had to go with the 9700pro because it was CLEARLY the better "bang for the buck" than anything the 5800 line had to offer.

However, if this race is as close as it is shaping up to be, my next card purchase will be a much more difficult decision.....but that is a GOOD thing, and about damn time!

 

[cornelious0_0]

Originally posted by Brent
We aren't stealing any thunder away from the card. If you read the preview you will see that our results showed that it was able to play every game faster and with more iq than the 5950U and the 9800XT, it won every game test.

How is that stealing thunder by showing it to be the best card for gaming?

Perhaps you missed this paragraph in the conclusion as well?

I didn't miss anything, the one thing I'm still getting at is not once did I see any benchmarks at 1600x1200 with AA and AF cranked up, which shows the cards true power. BTW, at those ultra high settings it is STILL the best performer, and makes fools out of everything else on the market.

There is another thing I don't understand, and that's how your benchmarks don't match up with a number of the others out there. Take for instance Unreal2k4 in your preview. You guys showed the 6800 running at 1280x1024 res with 4xAA and 16xAF.....with an average fps of 46.5.....whereas the review over at the firingsquad has it running at 1280x1024 with 4xAA and 8xAF.....but at 65.6fps. I know for certain that not even my 9800XT takes a 20fps hit from going up to 16xAF. Either you guys are using different methods of testing or something is up.

I'm basically through arguing about this, because you still arent getting what I'm saying.....but that's alright, I'll just continue to read elsewhere and forget about taking another look at any of your video card "reviews".

 

[MemoryInAGarden]

Originally posted by cornelious0_0
I didn't miss anything, the one thing I'm still getting at is not once did I see any benchmarks at 1600x1200 with AA and AF cranked up, which shows the cards true power. BTW, at those ultra high settings it is STILL the best performer, and makes fools out of everything else on the market.

There is another thing I don't understand, and that's how your benchmarks don't match up with a number of the others out there. Take for instance Unreal2k4 in your preview. You guys showed the 6800 running at 1280x1024 res with 4xAA and 16xAF.....with an average fps of 46.5.....whereas the review over at the firingsquad has it running at 1280x1024 with 4xAA and 8xAF.....but at 65.6fps. I know for certain that not even my 9800XT takes a 20fps hit from going up to 16xAF. Either you guys are using different methods of testing or something is up.

I'm basically through arguing about this, because you still arent getting what I'm saying.....but that's alright, I'll just continue to read elsewhere and forget about taking another look at any of your video card "reviews".

They're not reviewing cards in the traditional sense. What they're trying to do is show what resolutions and AA/AF levels are playable in various current games.

Also, why are you hinting around that the results were skewed or or falsified in some reason? Do they have a real motive for making the card look worse than it really is? After all, they got this card before any of the general public can. I don't understand why they would shit on the card, especially when they're getting the benefit of getting it before it's widely available.