RXXX Series Roadmap from AnandTech

The RV models have always been the mid-range chips in the past, eg. X700 is the RV410 and also has a 128-bit bus.
 
4 should be vertex shaders, I don't know what the 1 is (1 triangle per cycle at setup? I doubt it), 3 is three fragment shader units (working on quads) and 2 is two ROPs (working again on quads).
 
Dave Baumann said:
12 pipelines...
4-1-3-2...
Code:
16-1-1-1 R520 
16-1-3-1 R580
 4-1-3-2 RV530
 4-1-1-1 RV515

I'm kinda :oops: as to where this leeds for R580 ...

Anyways, 1.4ns memory on the mid-range cards should indicate a plenty supply, lets hope that also implies some faster bins available for the high-end soon. Oh yeah, and I'm getting tired of 128-bit memory busses.
 
RoOoBo said:
4 should be vertex shaders, I don't know what the 1 is (1 triangle per cycle at setup? I doubt it), 3 is three fragment shader units (working on quads) and 2 is two ROPs (working again on quads).
My current running theory is:

ROPs - tex/clock/pixel - pipes/ROP - Z/clock

But that would lead to a truly monster R580. :-?
 
RoOoBo said:
4 should be vertex shaders, I don't know what the 1 is (1 triangle per cycle at setup? I doubt it), 3 is three fragment shader units (working on quads) and 2 is two ROPs (working again on quads).

If 3 is three fragment shader quads then R420 would have 1 quad/ 4 pipes. not very likely...
 
so they are a year late from potentially Christmas 2004 to Christmas 2005... something really went wrong there at ATI :-?
 
incurable said:
My current running theory is:

ROPs - tex/clock/pixel - pipes/ROP - Z/clock

But that would lead to a truly monster R580. :-?

lets start this cronologicaly

1 ATI decoupled texture units from fragment quads to make them available vertex units ->R520

2 why should they limit one texture unit per pipe? once texture units are decoupled make one fragment able to make 3 requests instead of one. -> R580 & RV530

so it should be ROPs - ? - tex/clock/pixel - Z/clock
 
I think it's high time Dave dropped us just a few more hints about those numbers and what they're referring to. After all, my grandparents fought in the War for this! :cool:
 
RV530 is a beast
icon_exclaim.gif
icon_mrgreen.gif


Fragment shader rate:

RV530 - 7200M fragments/s
X800XL - 6400M fragments/s
X850XT - 8320M fragments/s

RV515 is a beast too:

RV515 - 3200M fragment/s
X550 - 1600M fragments/s
X300 - 1300M fragments/s
X700XT - 3800M fragments/s

Blimey.

Jawed
 
We have talked a little bit about R520 in the past, specifically concerning H.264 acceleration on the card. We have also hinted at performance rumors of the R520, but since the silicon has been overhauled so many times since then, R520 could be much different than what was originally speculated. (In fact, we know it is.
sooooooo

who wanna bet R520 was cancelled and ATi is going straight to R580 ?
 
Neeyik said:
I think it's high time Dave dropped us just a few more hints about those numbers and what they're referring to. After all, my grandparents fought in the War for this! :cool:

I would say that you need to consider what are the important elements (other people are considering "pipelines" to be and then a case of "do the math".
 
chavvdarrr said:
sooooooo

who wanna bet R520 was cancelled and ATi is going straight to R580 ?

I would say that comment was more of a justification why the BS specs everyone and their dog has peddled over the past 6 months don't bear much relation to reality. The article mentioned "silicon changes" and while some changes could be made in a bringup cycle, they are going to be mere tweaks and not large scale changes to the design or configuration of a chip.
 
Dave Baumann said:
I would say that you need to consider what are the important elements (other people are considering "pipelines" to be and then a case of "do the math".
That's just so not the answer I was looking for...:-? ;-)
 
lopri said:
128-bit memory?

The 700Mhz GDDR3 will help a bit to feed all the 12 pipes.

Core speed and mem speed of the RV530XT are supposedly also the same for the R520XT while the R520XL will rock at 500/500.
 
My theory makes a lot more sense

Dave Baumann said:
12 pipelines...
4-1-3-2...
Number of tape-outs/ nb of market failure/ nb of years needed for Ati to recover from this/ nb of Canadian graphical IHVs doomed (Matrox: Never Forget).
 
Back
Top