AMD says "Prove it!"

[Lanek]

a hint... you don't see how it is with the fin.... the power of the image i believe ..

What do you want to say ? AMD don't use vapor chambers or AMD don't know how make vapor chambers when they have use it on some models since the 2900XT ? im not sure i follow you on what this thread is about .

[STEvil]

I cannot see a seal/ gap or bolts to hold the two pieces of copper together to attain vapor...?
(nice grammar)

No need for petty grammar fights. Not everybody's first language is english, and not all institutions of learning do a worthwhile job of teaching it these days.

[NEOAethyr]

nothing really ? are you sure you have not seen anything who will not let you clock it as you want ? as something who make some white smoke when it happend ..

Yeah I know about the vmods burning up the card's .


In my mind, ati/amd just cannot compete at all without tex/mip adjustment settings.
I want 100% of the detail, not 5%...
I might sound ignorant or something but honestly it's how I feel about it.

Textures are not supposed to be blurry 2 ft from the player.
Well, they are normally, but I don't like it that way, on my nvidia card's I have a choice if I regedit it or mod the driver inf.
On ati/amd, I have the settings for it in the registry, but they don't do anything .
Last time I knew it was still this way, the 9x00 gen, dx9, but that was along time ago and I'de love to be proven wrong on that one.
It's the main thing keeping me from "looking" at an ati card.
Yeah that's right, I won't even look at them...
No offence.

I have no probs recommending a buddy an ati card.
But it's not for me.

[ZL1Killa]

someone needs to make some *** **** better drivers, I don't care whose is bigger, I just want some better drivers.

[halfwaythere]

One of the reasons 6990 is so good in games is actually the latest drivers: 11.4. Someone mentioned they changed the drivers writing team and it kind of shows already. So it appears amd is at least 2 up on nvidia: better drivers and a better built card.

[saaya]

One of the reasons 6990 is so good in games is actually the latest drivers: 11.4. Someone mentioned they changed the drivers writing team and it kind of shows already. So it appears amd is at least 2 up on nvidia: better drivers and a better built card.

they have 2 teams working back to back I thought?

And yes, sad to see all this horsepower go to waste with image quality tweaks... on both cards :/

There should be options to disable ALL optimization...

[SKYMTL]

This thread is quite interesting, as is the post by Dave. Basically, I take this as a toss up between the two in overall performance and we all know that 3DMark11 is far from "industry standard".

I don't think anyone has anything to prove here other than their ability to deliver stable performance to the consumer with a minimum amount of BS.

[killerxp100]

"industry standard"... More like AMD's standard & Nvidia's standard.

[BeepBeep2]

I cannot see a seal/ gap or bolts to hold the two pieces of copper together to attain vapor...?
(nice grammar)

Vapor chamber is essentially a flat heatpipe like I said before.
If you look at the bottom side of the heatsink you see a little square region sticks out. In there is the vapor chamber, actually that whole copper peice is the base/vapor chamber. The Aluminum fins on top are the rest of the heatsink.

Sapphire uses bolts in their vapor chamber design to secure the fan they usually stick in that area

[Mungri]

I cannot see a seal/ gap or bolts to hold the two pieces of copper together to attain vapor...?
(nice grammar)

Vapor chambers were used on the 5770, 5970, and 6900 range reference designs.

They look like a flat copper heatpipe which is filled with coolant fluid, and when it heats up it vaporises and spreads throughout the chamber. The aluminum fins conduct the heat from the vapor chamber and cool the vapor back down.

It is potentially a more effective design than heatpipe coolers, but the ATI reference cooler fans are still completely terrible, and MSI's custom heatpipe based coolers are still a lot better and cooling the GPU down.

[insurgent]

Erskine (arguably not AMD itself) is just asking for an "industry standard" test, no matter how useless it may be since they provided their own (mostly) irrelevant but accurate proof that their card is the fastest while Nvidia just claimed without proof. It's all marketing fluff, in a blog no less.

[tbone8ty]

waiting for Nvidia's rebuttal........or should i say "Annoucement" hahahahahahharhrahrahrahrha

[saaya]

In the end, I actually side with AMD on this one...
No1 they launched first and marked the spot
No2 after NVIDIA launched the 590 AMD said they are faster in xyz

NVIDIA doesn't specify how they ate faster... seems their marketing department is too lazy

Traditionally they would produce a slideshow highlighting where and how they are faster, and would disagree, and fanboys would argue for a few weeks.
But this time NVIDIA didn't specify anything...

[tiro_uspsss]

wow... so many upset nv fanbois.. evident by the: 'fix the drivers' posts.. I have a 78GTX with VIVO, that never got x64 VIVO drivers - can I whinge & whine like a lil boi now? - oh let me guess - 'its too old' re: the v.card.. yeah, so? all the more time to release drivers right? what about when it was released? & what about getting the drivers out in a reasonable time frame?

go ahead & be the fanbois u are.. for me & my experience: nv drivers have sucked harder & more frequently than ATi

[DarthShader]

[Bail_w]

Team red is #winning here. Charlie Sheen style, lol.

[NEOAethyr]

wow... so many upset nv fanbois.. evident by the: 'fix the drivers' posts.. I have a 78GTX with VIVO, that never got x64 VIVO drivers - can I whinge & whine like a lil boi now? - oh let me guess - 'its too old' re: the v.card.. yeah, so? all the more time to release drivers right? what about when it was released? & what about getting the drivers out in a reasonable time frame?

go ahead & be the fanbois u are.. for me & my experience: nv drivers have sucked harder & more frequently than ATi

You're right, nvidia don't have x64 wdm vivo drivers.
They haven't updated the x86 ones in years...

I since switched to my easycap usb dongle, it has a phillips chip as well.
Because non of the newer card's have vivo (sucks).
I already modded and combined them.., not sure if there is x64 for it though now that I think about it.

I think the majority of recent drivers for nvidia have been alright.
Much better releases then the years past where you'de go 100.x ver's before another decent driver foundation... (nvidia)

The 45.xx drivers.
The 84.26 driver.
And now pretty much all the latest drivers are all what's decent.
There were 3 huge gaps of complete crap.

It's all good though, I don't have a complaint about that .

There should be options to disable ALL optimization...

I totally agree.

If both ati and nvidia did that there would be real competition.
And ati could compete in the pro market as well (firegl is a joke if you ask me).

There's no reason why pc games have to look like console games in terms of quality of there textures.
I vote for no blurring what so ever, and no need for AF, that should be gotten rid of.

Some of you might think I'm talking bs, seriously though, all AF does is blur the living crap out of your mip maps and reduces performance.
It just looks better when you have blurred textures to begin with.
Trilinear is cleaner then AF...

There's a performance penalty when you don't blur the textures with lossy mip mapping methods.
But that penalty is almost long gone since the days of the geforce 4.
It still reduces your fps, but the small diff it does make in loss of performance is worth the large increase of quality, it's a whole world of difference.

All a quadro is, is slightly lower mip bais by default, some lossy optimizations disabled, and some speed restrictions removed.
And that's not even full quality... (not that it matters, you can fix that)

[STEvil]

Some of you might think I'm talking bs, seriously though, all AF does is blur the living crap out of your mip maps and reduces performance.
It just looks better when you have blurred textures to begin with.
Trilinear is cleaner then AF...

There's a performance penalty when you don't blur the textures with lossy mip mapping methods.
But that penalty is almost long gone since the says of the geforce 4.
It still reduces your fps, but the small diff it does make in loss of performance is worth the large increase of quality, it's a whole world of difference.

All a quadro is, is slightly lower mip bais by default, some lossy optimizations disabled, and some speed restrictions removed.
And that's not even full quality... (not that it matters, you can fix that)

About the trilinear vs anisotropic, yeah, you're talking BS.

[NEOAethyr]

Lol, I can prove it to you dude, I promise.
I have to find my ut2004 demo.
That is the best example I can think of to show you off the top of my head.

Give me until morning, even if I have to install the full game which I got somewhere.
It's noticeable in most games, but some more then others.

[generics_user]

Before claiming the fastest graphics card, fix your drivers, the fact that the latest official drivers still have the stuttering bug on the top right corner of the screen in windows 7 64 bit says alot, 4 months after the release of the 6950 and 6970, this bug still exists

OMG IT SUCKS; the cursor sometimes stutters when i try to play flashing windows buttons

seriously i never noticed this until someone pointed it out on this forum a month ago and i actually forgot about that bug until you noticed it again...

which sane windows user sits there; moving a cursor over these 3 buttons trying to decide whether to maximize; minimize or close a window


i bet that you lost more time of your life posting this massage than failing to hit the close button on a maximized window due to some mouselag only present when you actually hit the display edge and so miniscule that it's hardly noticeable?


i'm sure that amd has better things to do than fixing this bug which has no effect at all on the user experience if 99.9% of all users (this number is 100% correct)

[duploxxx]

If both ati and nvidia did that there would be real competition.
And ati could compete in the pro market as well (firegl is a joke if you ask me).

what is the joke here? your statement or the cards?

FireGL stopped in 2007 with R600, after that it was Firepro with R770, it seems like you need to upload some new roadmaps in your brains....
sure the Quadro has the performance lead, but you will also pay much much more for it... price/performance ratio is the key and not some old retared IT who always buy the brand names....
check OEM, they offer way more FirePro then they used to and that is the key to volume.

[zanzabar]

what is the joke here? your statement or the cards?

FireGL stopped in 2007 with R600, after that it was Firepro with R770, it seems like you need to upload some new roadmaps in your brains....
sure the Quadro has the performance lead, but you will also pay much much more for it... price/performance ratio is the key and not some old retared IT who always buy the brand names....
check OEM, they offer way more FirePro then they used to and that is the key to volume.

it actually depends on what u are doing, and in raw OGL power ati still has the better cards its just cuda is in allot of stuff now and NV is better at bulk renders but for a live workstation ati is better most of the time.

[cronicash]

+1

Still dont see the bolt/rivets

vapor
Its done on a much smaller scale compared to nvid imo...
If its so great then why is it nvid publicize it more than ati. I fail to see it in their features/ specs on many sites...

I know how it works... nubs fail if they did not see what i was getting at.

Have fun ranting...

Single float compute is much higher on ati, this is what raytracing is mostly about, other bloated ambient, occlusive logarithms (like in octane, unlike arion which is cpu where applicable) may bring this down yet ati still would be better in gpgpu imo considering the single float performance imo...

[insurgent]

What's this cursor lag bug? It didn't show in my 4870 and does not show in my 6950... Win7 x64 10.x to 11.4 drivers. Only bugs I've experienced recently is in Disciples 3 which, as far as anyone's concerned, is really buggy whatever system you use.

[generics_user]

What's this cursor lag bug? It didn't show in my 4870 and does not show in my 6950... Win7 x64 10.x to 11.4 drivers. Only bugs I've experienced recently is in Disciples 3 which, as far as anyone's concerned, is really buggy whatever system you use.

didn't notice that lag at all until someone pointed out how to:

maximize a window
move your cursor to the top right edge of your scrren (you have to touch the corner with your mouse)
now move it around on the minize/maximize/close buttons
there you have your lag

my take on this bug: not noticeable in real life

and i actually think that this is intended to be so as the mouse really sticks to the close button pretty much like autoaim on consoles