ATI Radeon HD 4000 Series discussion

[cegras]

First a question for you:
How reliable is Wikipedia?

I find wiki to be quite reliable, especially if they draw from first-order sources, which they frequently do for almost any science topic - I can almost depend on them for a quick first read of any topic I am not interested in.

They have a very good article generation and upkeep structure, but it's fouled by the fact that anyone can edit. If they close THAT off, it might suddenly become a reliable resource.

[Jethro]

http://www.nordichardware.com/news,7841.html


?

[purecain]

wooooow, if that isnt right nordichardware are going to get slated.... on the flip side, if its real... woooooooooooow... i dont think its real unfortunatly....

[LordEC911]

http://www.nordichardware.com/news,7841.html


?

wooooow, if that isnt right nordichardware are going to get slated.... on the flip side, if its real... woooooooooooow... i dont think its real unfortunatly....

Wanna read through the thread?
Or you could just read page 61...

[Jethro]

Have been following the thread but been busy all weekend with my daughter and it seems ive missed a few pages o.0. Either way groovy news!

[purecain]

Have been following the thread but been busy all weekend with my daughter and it seems ive missed a few pages o.0. Either way groovy news!

snap!!

[AliG]

http://www.nordichardware.com/news,7841.html


?

ugh, don't you guys actually read this thread before you repost things?


Man I'm getting sick of this, its not you personally, but people need to really learn to read threads before the post things

[zerazax]

Someone should really update the first post though w/ the latest infos

[generics_user]

LoL. You mis-understand, really. TMU's are TEXTURE MEMORY UNITS...cache...

The real problem with R670 is AA implementation. If they fix that, they can get away with 24 TMUs, although this is truly dependant on how many shaders there are...the number of shaders will dictate the number of TMUs.

Helmore, while I partially agree with you, it does not see mto make much sense that AF was a big problem. 16xAF @ 2560x1600 is not a show-stopper. AA WILL make an app unplayable, so while texture filtering and z-ops were a bit anemic, they weren't the real problem.

AA in shaders is essentailly the future, yes. When we get 4xMSAA standard...but that's still a couple of gens off yet...more likely to appear the same time as raytracing...

So ATI had a foray into it, didn't work well for them, but i think they understand the problems they are facing. Doesn't mean they have to beat the dead horse...

TMUs are texture mapping units, not memorx units

[LowRun]

ugh, don't you guys actually read this thread before you repost things?


Man I'm getting sick of this, its not you personally, but people need to really learn to read threads before the post things

Talking about repost, you've posted that same kind of comment like a dozen time already. This is bound to happen in a 70 pages thread so chill out

[AliG]

TMUs are texture mapping units, not memorx units

that's what I thought

[Lightman]

Quick copy&paste

This score in Crysis are authentical ... i got very similar FPS on my GTX 280 ...

GTX 280 is not brutal powerfull card! performance is good, but i expected many more ....

According testing of Radeon HD4850 in CF i can say - radeons are better then GTX 280!

by ORB here..


As we all know he is not an AMD/ATi fan
But his information are legit for sure....

[Miwo]

sweet
2 more weeks.......

hoping to hear some more about 1GB cards soon

[w0mbat]

If i knew how fast gtx280 is i maybe could confirm ORB´s info.

[Lightman]

If i knew how fast gtx280 is i maybe could confirm ORB´s info.

Lets say 38FPS avg. for QC 2.66GHz at 1920x1080 DX10 VH (this is the slower score leaked today)

[Hornet331]

If i knew how fast gtx280 is i maybe could confirm ORB´s info.

singel 280gtx in crysis @1900x1200 (no aa/af) VH -> 36,8fps

[Extelleron]

If i knew how fast gtx280 is i maybe could confirm ORB´s info.

Supposedly ~37 FPS in Crysis @ 1920x1200 Very High / No AA

And supposedly ~X4800 in Vantage.

That's all we have right now.

[003]

Supposedly ~37 FPS in Crysis @ 1920x1200 Very High / No AA

Wow, that sucks. If that's true, looks like we may have another FX series on our hands, or more specifically, the FX 5800 Ultra.

[Hornet331]

Wow, that sucks. If that's true, looks like we may have another FX series on our hands, or more specifically, the FX 5800 Ultra.

do you even know what SLI 8800gtx and 9800gtx scores at this res in crysis?


wooping 25-27fps.

[Extelleron]

Wow, that sucks. If that's true, looks like we may have another FX series on our hands, or more specifically, the FX 5800 Ultra.

It's not that bad.

http://www.tomshardware.com/reviews/...w,1792-10.html

According to that, @ 1920x1200 DX10 VH w/ reduced shadow setting, 9800GX2 gets 32.1 FPS and 8800 Ultra gets 21.3 FPS.

So GTX 280 is roughly ~2x 8800 Ultra.

[Hu1kamania]

Looks like it is still the fastest single chip solution, which lately seems pretty important to a lot of people. Unlike the 5800 Ultra when it came out.

[kemist]

LoL. You mis-understand, really. TMU's are TEXTURE MEMORY UNITS...cache...

This post and like the next 3 posts after it are all wrong. First off TMU's are not Texture memory units, they are texture mapping units, though they do have large amounts of cache associated to them.

Um, TMU's do the texture processing? NO!!! The shaders do! So yes, that IS the case! More specifically, the inconnection speed between the TMUs and the shader core is much faster in nV's solution, hence the difference.

This too is wrong, why would you have TMUs if they are not doing the texture processing? Maybe you are confusing offloading AA work from RBE's into the shaders, but this is not occurring with TMUs yet.

Its been shown that there is nothing wrong with Ati's AA in and of itself, in fact the hit from just AA is often less than nvidias at the same setting. Its AF that tanks performance. The source below shows this, especially with HQAF which is default for Ati now. If you want old AF you need to use ati tray tools or other progs like that.

Source: http://www.computerbase.de/artikel/h...d_3870_rv670/5

[Helmore]

Thanks for clearing that up kemist, I was thinking the same thing and although I wasn't sure of the specifics, I did know cadaveca was wrong on this.
About RV770, we still don't know anything about it really, there are still some whom are doubting the 800 SP part (I just hope it's true) and nobody has any idea on the rest of the architecture (IE. TMUs and stuff like that). Although I'm pretty sure there won't be any separately clocked shader domains, that's just to much of a diversion from their current RV670 architecture.

[madcho]

Thanks for clearing that up kemist, I was thinking the same thing and although I wasn't sure of the specifics, I did know cadaveca was wrong on this.
About RV770, we still don't know anything about it really, there are still some whom are doubting the 800 SP part (I just hope it's true) and nobody has any idea on the rest of the architecture (IE. TMUs and stuff like that). Although I'm pretty sure there won't be any separately clocked shader domains, that's just to much of a diversion from their current RV670 architecture.

serparate clock shader is easy to do. GPU are in order, use a fifo between the other side of the gpu, and rate your clock shader as you want.

EASY TO DO

The real question; ATI found it usefull ?

[Macadamia]

serparate clock shader is easy to do. GPU are in order, use a fifo between the other side of the gpu, and rate your clock shader as you want.

EASY TO DO

The real question; ATI found it usefull ?

Yeah right, what about latencies and synchronization?
You make it sound like a FIFO doesn't induce big latencies. And their shader units aren't engineered to run clocks that high too.


It's 800SP and flat clock, easy engineering and easy on the die size too.