Hardwareanalysis bench are fake

[PsharkJF]

Ah, 3DFX...
[nostalgia][/nostalgia]

Ignoring the benchmarks for the moment, Sander's right about one thing. ATi's only option is to deny.

 

[Blastman]

Originally posted by: keysplayr2003

Your attempt at psychology is cute at best. I'll post as much as I want to whomever I wish. Where did I ever post that it is 100% real? Hmmm? Oh, that's right. I didn't.

(earlier) ?

? And ATI deserves a bit of ridicule for this poor excuse of a video card

Well, it does look like some people have already made their minds on the R520 based on the HA benches.

Considering Sassens?s anti-ATI rhetoric, those HA benches have to be treated with a great amount of suspicion.

About the R520 ?sucking? because it is only 16 pipes.

If the R520 has a pooled shader architecture those 16 pipes could be 25-40% more efficient (or more) than a conventional 16 pipe architecture. That means a 16-pipe card could match a 20 -22 pipe card clock for clock.

If it was 40%, a 16 pipe R520 would only need to be clocked 460mhz to match the 430 mhz 7800GTX in shader power. At 600mhz, the R520 would have considerably more shader power than a 7800. Because of its high clock rates the R520 is going to likely have a lot more Vertex power too.

Certainly ATI is going to need a better showing than what is in those HA benches if the R520 hits the stores at $599. Even ATI fans agree here. But ATI taped the R520 out a year ago. If the performance was no better than a X850 XTPE clocked 600/1400 (which is what those HA benches look like), they would have pushed forward the R580. So I expect the R520 to be better than what is in those HA benches.

 

[imported_g33k]

Those benches may be a tad low. With finalized drivers, you may see a small perfomance increase. But those benches will be close to the released product, imo.

 

[xtknight]

Originally posted by: sodcha0s
Wow, you people will argue over ANYTHING video card related. Get a freaking life! A video card isn't the be-all end-all of your existence.

Why do you think this is the Video forum?

 

[BFG10K]

I personally have some doubts about the validity of the benchmarks, especially after the ATi rep's comments. Ah well, I guess we'll find out on October 2.

 

[Biatche1488]

Originally posted by: ronnn

Originally posted by: Jeff7181

Originally posted by: Biatche1488
They are FAKE some site said it its just a x800xtpe overclock
They did that because they are not invite at the official lauch
Just look at the half life 2 result 512mb+600mz+new architecture+faster memory+16shader+512bitringbus+youre all blind those bench are fake
And they dont talk about shader model 3 in splinter cell
And where they got there drivers we dont now they build them
The only card availble at this time its the x1800xl ati big partners can only get x800xl at this time

Am I the only one who had to take 2 full minutes to read and understand that?

Likely the only one who read it.

lol i know im really bad for making post

 

[Keysplayr]

Originally posted by: Blastman

Originally posted by: keysplayr2003

Your attempt at psychology is cute at best. I'll post as much as I want to whomever I wish. Where did I ever post that it is 100% real? Hmmm? Oh, that's right. I didn't.

(earlier) ?

? And ATI deserves a bit of ridicule for this poor excuse of a video card

Well, it does look like some people have already made their minds on the R520 based on the HA benches.

Considering Sassens?s anti-ATI rhetoric, those HA benches have to be treated with a great amount of suspicion.

About the R520 ?sucking? because it is only 16 pipes.

If the R520 has a pooled shader architecture those 16 pipes could be 25-40% more efficient (or more) than a conventional 16 pipe architecture. That means a 16-pipe card could match a 20 -22 pipe card clock for clock.

If it was 40%, a 16 pipe R520 would only need to be clocked 460mhz to match the 430 mhz 7800GTX in shader power. At 600mhz, the R520 would have considerably more shader power than a 7800. Because of its high clock rates the R520 is going to likely have a lot more Vertex power too.

Certainly ATI is going to need a better showing than what is in those HA benches if the R520 hits the stores at $599. Even ATI fans agree here. But ATI taped the R520 out a year ago. If the performance was no better than a X850 XTPE clocked 600/1400 (which is what those HA benches look like), they would have pushed forward the R580. So I expect the R520 to be better than what is in those HA benches.

I'll say this. After all this time waiting and hoping for at least a better performing card from ATI, and all the BS we have been exposed to by a staggering amount of rumors, I am certainly dissappointed IF this bench holds any truth. A 16 pipe top offering is really not what I was hoping for as well as the rest of the ATI fans. I actually wanted the X1800XT to be quite a bit faster than the 7800GTX because this would force Nvidia to make their next move (which I hear they are ready to do but felt they didn't need to). I hope ATI can get at least a 24 pipe card out real soon.

 

[Turtle 1]

Key: I kinda sort of agree except for I already like my XT800XT.If it is any kinda real improvement. I will be happy If it brings a better feeling as how I interact with all the hardware and software features of my PC I well be happy with it . The G70 has brought the CPU'S to theirs kneese already. I suspect the R520 won't beable to do much better. From here on out its going to be about rendering. I want to see true HD in everthing I view on the screen I want to feel like I can reach out and touch it .

 

[stnicralisk]

Originally posted by: lavaheadache
Here is another link stating that the benches are fake. http://www.guru3d.com/newsitem.php?id=3069

Actually that link simply quotes CH as stating they are fake which is the same thing quoted in the previous link. Guru isnt actually saying theyre fake as they cannot know.

Here is what I want to know - how would CH decide they were done with an overclocked 850xt? Why wouldnt the numbers just be made up?

 

[stnicralisk]

Originally posted by: Rollo

Originally posted by: nitromullet

You are free to disregard me as you see fit Tanclearas. Perhaps those who read my posts and consider them see that I provide useful information from time to time, sometimes before anyone else? (e.g. I was the first on the web to post my experiences with nVidias shimmer fix)

You do post a lot of useful info, but i think that you should stop referring to the 78.03 drivers as a "shimmer fix", it did not fix the issue. I played about 8 hours of WoW over the past few days with the 78.03 drivers, 16xAF, LoD clamped, HQ, etc... and still the AF shimmer was present. I just wanted to point this out that to the rest of us that own a GTX, your claims of the shimmer being fixed (when I can see it's not with my own eyes) makes me question your credibility. You seem like a good guy overall, and I would think that you would rather be honest about the "fix" than let it tarnish your credibilty. Obvioulsy, I have just as much financially invested into my video card(s) as you do, so I don't want the shimmer to be there either, but no so much that I will choose to ignore it.

I'm not trying to mislead anyone Nitro. I didn't see the shimmer where I went back looking for it, but I am aware that it still exists to a lesser extent on some games.

I use the term "fix" for lack of a better one, perhaps I should have said "shimmer reducing".

You sure were quick to bash me when I said I doubted the new drivers would fix the issue though werent you. You deserve any flaming you get and also I told you so. You also stated that there was barely any performance loss and we saw that to be false too.

 

[Keysplayr]

Originally posted by: Turtle 1
Key: I kinda sort of agree except for I already like my XT800XT.If it is any kinda real improvement. I will be happy If it brings a better feeling as how I interact with all the hardware and software features of my PC I well be happy with it . The G70 has brought the CPU'S to theirs kneese already. I suspect the R520 won't beable to do much better. From here on out its going to be about rendering. I want to see true HD in everthing I view on the screen I want to feel like I can reach out and touch it .

Yup. And this is how I feel about what ATI is doing. Here.

 

[nRollo]

Originally posted by: stnicralisk
You sure were quick to bash me when I said I doubted the new drivers would fix the issue though werent you. You deserve any flaming you get and also I told you so. You also stated that there was barely any performance loss and we saw that to be false too.

I'm quick to bash you here too stnicralisk!

1. You're Off Topic

2. My posting info that was 90% right before anyone else on the net had a LOT more value than you whining "I doubt it"

3. The drivers eliminated the shimmer in some games, greatly reduced it in others, and didn't cause any greater performance hit than going from "Quality" to "High Quality" did before.

4. ATI optomizations shimmer as well

Meh, and begone.

 

[apoppin]

Originally posted by: stnicralisk

Originally posted by: lavaheadache
Here is another link stating that the benches are fake. http://www.guru3d.com/newsitem.php?id=3069

Actually that link simply quotes CH as stating they are fake which is the same thing quoted in the previous link. Guru isnt actually saying theyre fake as they cannot know.

Here is what I want to know - how would CH decide they were done with an overclocked 850xt? Why wouldnt the numbers just be made up?

it is LIKELY they are not real . . . Sassen has a vendetta against ATI and he covered his butt by saying the benchs are NOT HIS.
:roll:

using an o/c'd xtPE would give some validity to the benchs AS IF the r520 was an o/c'd x800 gpu.

i bet they are fabricated

 

[nRollo]

Originally posted by: apoppin

Originally posted by: stnicralisk

Originally posted by: lavaheadache
Here is another link stating that the benches are fake. http://www.guru3d.com/newsitem.php?id=3069

Actually that link simply quotes CH as stating they are fake which is the same thing quoted in the previous link. Guru isnt actually saying theyre fake as they cannot know.

Here is what I want to know - how would CH decide they were done with an overclocked 850xt? Why wouldnt the numbers just be made up?

it is LIKELY they are not real . . . Sassen has a vendetta against ATI and he covered his butt by saying the benchs are NOT HIS.
:roll:

using an o/c'd xtPE would give some validity to the benchs AS IF the r520 was an o/c'd x800 gpu.

i bet they are fabricated

Sassen having a vendetta doesn't mean it's "likely" he lied, it means it's possible he lied. Sassen not having a vendetta could possibly be lying.

We don't know. To me, those benches look in line with what a faster 16 pipe card with some memory and shader tweaks might do.

 

[apoppin]

Originally posted by: Rollo

Originally posted by: apoppin

Originally posted by: stnicralisk

Originally posted by: lavaheadache
Here is another link stating that the benches are fake. http://www.guru3d.com/newsitem.php?id=3069

Actually that link simply quotes CH as stating they are fake which is the same thing quoted in the previous link. Guru isnt actually saying theyre fake as they cannot know.

Here is what I want to know - how would CH decide they were done with an overclocked 850xt? Why wouldnt the numbers just be made up?

it is LIKELY they are not real . . . Sassen has a vendetta against ATI and he covered his butt by saying the benchs are NOT HIS.
:roll:

using an o/c'd xtPE would give some validity to the benchs AS IF the r520 was an o/c'd x800 gpu.

i bet they are fabricated

Sassen having a vendetta doesn't mean it's "likely" he lied, it means it's possible he lied. Sassen not having a vendetta could possibly be lying.

We don't know. To me, those benches look in line with what a faster 16 pipe card with some memory and shader tweaks might do.

you missed my point - "i" am saying the 'likely' - just as you are saying 'possibly unlikely'
[. . . and i base my 'bet' on Sander's motive and lack of responsibility/consequences for lying

but its true - we don't know for sure - yet.

 

[Duvie]

No need to attack Rollo ppl...At least he is going with the side that has empirical information unlike most of you...I say that since most of the arguments for these being fake are fanboy opinions and comments borderline on slanderous...if AT, Techreport come out and confrim the numbers are similar then I move to ban biatche and ackmed his little biatche..This is for rudely insulting ppl of this forum....

Calling ppl stupid cause they believe posted benches from some site when you have no credible evidence to prove this or other empirical data to refute it shows who the real stupid ones are.

 

[ronnn]

Originally posted by: Duvie
No need to attack Rollo ppl...At least he is going with the side that has empirical information unlike most of you...I say that since most of the arguments for these being fake are fanboy opinions and comments borderline on slanderous...if AT, Techreport come out and confrim the numbers are similar then I move to ban biatche and ackmed his little biatche..This is for rudely insulting ppl of this forum....

Calling ppl stupid cause they believe posted benches from some site when you have no credible evidence to prove this or other empirical data to refute it shows who the real stupid ones are.

Well have your movement.

 

[TerracideDK]

Originally posted by: otispunkmeyer
soloflys bench mark post is in dutch, but i was jus replying jokingly because he said "whacky french"

Exuse me?!
Hardinfo.dk is in danish.
I should now, I am a dane.

Terra...

 

[orangat]

Theres going to be a lot of red faces after the truth comes out (either way).

Personally I think they are fake after ATI officially called it an ouija board benchmark. But how much better can the real ones be, that is the question.

 

[Duvie]

Originally posted by: orangat
Theres going to be a lot of red faces after the truth comes out (either way).

Personally I think they are fake after ATI officially called it an ouija board benchmark. But how much better can the real ones be, that is the question.

I dont see where most ppl who see numbers will have a red face...At least we dont blindly believe what we cant see cause we dont like what the numbers have shown....If he numbers are false then by all means the author of the article should be held responsible...

 

[nRollo]

Originally posted by: Duvie

Originally posted by: orangat
Theres going to be a lot of red faces after the truth comes out (either way).

Personally I think they are fake after ATI officially called it an ouija board benchmark. But how much better can the real ones be, that is the question.

I dont see where most ppl who see numbers will have a red face...At least we dont blindly believe what we cant see cause we dont like what the numbers have shown....If he numbers are false then by all means the author of the article should be held responsible...

No doubt. Why would we be embarassed for thinking posted information that could be correct might be correct?

I'd be more embarassed if I closed my mind to the possibility they are correct.

 

[Blastman]

Originally posted by: Rollo

4. ATI optomizations shimmer as well

Err?. no. A few games may exhibit shimmering on ATI cards, but that?s because of the way the textures are set up in the game. Not because of any ATI optimizations. Running A.I. on/off makes no difference on ATI hardware as does running a performance (Bilinear) setting. ATI is not under sampling.

 

[orangat]

Originally posted by: Rollo

Originally posted by: Duvie

Originally posted by: orangat
Theres going to be a lot of red faces after the truth comes out (either way).

Personally I think they are fake after ATI officially called it an ouija board benchmark. But how much better can the real ones be, that is the question.

I dont see where most ppl who see numbers will have a red face...At least we dont blindly believe what we cant see cause we dont like what the numbers have shown....If he numbers are false then by all means the author of the article should be held responsible...

No doubt. Why would we be embarassed for thinking posted information that could be correct might be correct?

I'd be more embarassed if I closed my mind to the possibility they are correct.

You should be embarrassed because you keep insisting ATI is lying by dismissing ATIs official statement that the benches were from an ouija board.

Conversely if the benches turn out be be real then other people are going to be red-faced.

 

[apoppin]

Originally posted by: Duvie

Originally posted by: orangat
Theres going to be a lot of red faces after the truth comes out (either way).

Personally I think they are fake after ATI officially called it an ouija board benchmark. But how much better can the real ones be, that is the question.

I dont see where most ppl who see numbers will have a red face...At least we dont blindly believe what we cant see cause we dont like what the numbers have shown....If he numbers are false then by all means the author of the article should be held responsible...

you don't get it . . . still . . . the author set it up so he CANNOT be held responsible

i am quite certain now that the benchs are fake . . . by lying - ati has much to lose; sanders has nothing to lose

 

[nRollo]

Originally posted by: apoppin

Originally posted by: Duvie

Originally posted by: orangat
Theres going to be a lot of red faces after the truth comes out (either way).

Personally I think they are fake after ATI officially called it an ouija board benchmark. But how much better can the real ones be, that is the question.

I dont see where most ppl who see numbers will have a red face...At least we dont blindly believe what we cant see cause we dont like what the numbers have shown....If he numbers are false then by all means the author of the article should be held responsible...

you don't get it . . . still . . . the author set it up so he CANNOT be held responsible

i am quite certain now that the benchs are fake . . . by lying - ati has much to lose; sanders has nothing to lose

Well, if you're "quite certain" I guess we should believe you, not Sassen.

He's probably not "quite certain"....errrr......ummmmm........well.....................