ATI products now PCI-SIG compliant

[overclocked_enthusiasm]

http://www.pcisig.com/developers/compliance_program/integrators_list/pcie/#components

http://www.theinquirer.net/?article=19475

 

[Smurfie]

Well, Charlie was man enough to admit he's wrong. Which is more than I can say about this forum.

Yes, this is what I really feel about this forum these days.

 

[Phantom69]

http://www.beyond3d.com/forum/viewtopic.php?t=17830

 

[whql]

Well, Charlie was man enough to admit he's wrong. Which is more than I can say about this forum.

Yes, this is what I really feel about this forum these days.

Bit of a pointless comment considering that a forum is a collection of people with differing opinions - how can a forum admit its wrong when there are going to be opposing view points? I seem to remember a discussion on this forum about this particular issue and there were some that were saying that lack of compliance doesn't result mean that are not actually compliant, whilst others were clearly indicating that ati weren't compliant - how can the "forum" admit its wrong there since some of the forum was correct and some of it wasn't? Would it please you if those that believed ati weren't compliant admit they were wrong?

 

[Mariner]

Well, Charlie was man enough to admit he's wrong. Which is more than I can say about this forum.

Yes, this is what I really feel about this forum these days.

Well, to be fair it wasn't "this forum" who were claiming ATI chips had failed PCI-E compatibility. It was one of the NV-centric wackos who visit - jimmyjames, radar or Ruined. I forget which one.

 

[gordon]

Well he was wrong the day he released the article since there was at least one ATI pci express product on the sig list that very same day.

 

[Pete]

So R480's already out and about, eh?

 

[kemosabe]

So R480's already out and about, eh?

Yup, and Mr. Baumann hasn't even dropped the usual cryptic clues. Is ATI tightening the screws even further, or is a fall refresh just not sexy enough?

 

[Mariner]

We're not expecting too much from R480, are we? Perhaps 575MHz core with faster memory, too? That's the kind of thing I'd expect but I would be be pleased if I was happily surprised!

 

[Dave Baumann]

My description of R480/430 so far would be an "availability refresh" (plus, there is another as well).

 

[karlotta]

My description of R480/430 so far would be an "availability refresh" (plus, there is another as well).

Does this mean what ever GDDR3 speed and size is out?

 

[Pete]

(plus, there is another as well).

Another Skywalker?!

 

[Dave Baumann]

Yes, it is a twin.

 

[Dave Baumann]

My description of R480/430 so far would be an "availability refresh" (plus, there is another as well).

Does this mean what ever GDDR3 speed and size is out?

I'm saying that the onus on this refresh may not be quite so much about readdressing speed issues, but availability (although I suspect there will be a speed hump in there somewhere).

 

[tEd]

My description of R480/430 so far would be an "availability refresh" (plus, there is another as well).

Does this mean what ever GDDR3 speed and size is out?

I'm saying that the onus on this refresh may not be quite so much about readdressing speed issues, but availability (although I suspect there will be a speed hump in there somewhere).

Do we know the exact reason for the shortage? I heard dfferent things(lack of high speed memory,Lack of .13lowk ressources at TMSC,Flip-chip substrate shortage). What kind of issues would the r480 address to improve availability? R480 also supposed to use .13 lowk isn't it?

 

[kemosabe]

Anyone else think January sounds a tad late? Or would that just be the Euro launch (not that you monarchists are unimportant, of course )? Also, 600MHz memory clock sounds weird unless they're still stuck with 1.6ns GDDR3 and clocking them to the hilt or severely underclocking 700MHz modules.

 

[Pete]

Late? For a X800 refresh that's less than 10% faster than the XTPE? B/c it ain't the R520, I guess it does sound a bit late.

 

[kemosabe]

Late? For a X800 refresh that's less than 10% faster than the XTPE? B/c it ain't the R520, I guess it does sound a bit late.

Dave has mentioned in another thread that this refresh is more about improving availability than a major performance boost. I don't think we should expect more than a 10% increase in core clock, though I was quite confident about the 700MHz GDDR3.

 

[Dave Baumann]

Anyone else think January sounds a tad late?

Thats probably for "the other".

 

[Pete]

I'm a bit lost, here. Are the "twins" AGP and PCIe sides of the same coin? Is R520 out of the question for 1H05, and probably scheduled for Q3 (exactly two 18-month "generations" after R300?

And, uh, WTH is R430? Is that the availability respin of R420, and R480/483(?) the AGP/PCIe refreshes?

Guh.