AMD: R8xx Speculation

How soon will Nvidia respond with GT300 to upcoming ATI-RV870 lineup GPUs

  • Within 1 or 2 weeks

    Votes: 1 0.6%
  • Within a month

    Votes: 5 3.2%
  • Within couple months

    Votes: 28 18.1%
  • Very late this year

    Votes: 52 33.5%
  • Not until next year

    Votes: 69 44.5%

  • Total voters
    155
  • Poll closed .
We don't have any full die shot, it could very well be a 80/160SP part (2 or 4 SIMDs with 8 or 16 Vec5 units each).

The die shots seen on AMD's papers don't tell us anything as they could be photoshopped and the only high res "full die shot" I've seen (from a "reputable" fakes source) showed some inconsistencies (missing blocks).

The cut back shot wasn't particularly telling; but I've seen a full die shot (don't recall where) that showed 6 clusters.

***edit: here you go http://img208.imageshack.us/img208/4966/apuu.jpg

Intel netbooks all use IMG IP. Intel based laptops are definitely a market, but the desktop market definitely appears toast.

80 alu's -> 400 alu's => 5x jump in one shrink? I don't think so. Especially when you look at the trends in the past. So for cedar, no.

Redwood is more interesting. May be not immediately, but in 2-3 generations, I am expecting amd fusion chips to cannibalize this market as well.

<$100 market probably accounts for ~80% (anyone got better numbers?) of the unit sales. AMD has a real opportunity here to gobble up this market. Let's just hope LIano is not delayed anymore.

Besides Llano there's also Ontario which sounds like netbook (and possible lower) material to me.
 
I would suspect that the fusion chips will have their own ram for the gpus. My 4330 on my laptop has 512 megs of ddr 3 built in. I wouldn't be surprised if the fusion chips had similar
 
I would suspect that the fusion chips will have their own ram for the gpus. My 4330 on my laptop has 512 megs of ddr 3 built in. I wouldn't be surprised if the fusion chips had similar

Where would this RAM be located? It's not feasible to place enough RAM as to be useful to an IMR GPU on the same package as said GPU.
 
It will be interesting how much memory will be aviable to the gpu. Maybe with 64bit OS and 8GB u can have 4GB+ video memory ;)
Probably the Llano gpu will use hyperTransport to comunicate with cpu :?:.
 
It will be interesting how much memory will be aviable to the gpu. Maybe with 64bit OS and 8GB u can have 4GB+ video memory ;)
Probably the Llano gpu will use hyperTransport to comunicate with cpu :?:.

You could have up to ~192GB video mem on 64bit Win7 Professional or higher
But having that extra memory won't help anything really.
 
ATI delays 5830 again?

http://www.fudzilla.com/content/view/17546/1/

We were told that these GPUs are going to be delayed to factories and that it won’t make it to China before February 10. At this time most factories are closed for the Chinese New Year and won't open their doors for anyone. This might delay the shipping until February 20th.
Too bad, the likely performance/possible price might have made it a really sensible option for me and my trigger finger has been straining while browsing Newegg's 5770 and 5850 offerings.

A possible upside is that the 5830 might come with the rumored new bios I've been reading about.

I have three questions.

Is the talk of a new bios for 5xxx cards based on fact?

If so, would the delayed 5830's be likely candidates or is a current variant likely to be already in the chip?

Is it the gpu that gets the bios or is a bios flashed onto the board somewhere?

Tia!
 
I read that there was going to be some voltage increases. Just rumors, hence my question.

As to how an unannounced card can be delayed, I guess it can't, strictly speaking. :)
That's partly why I included a question mark in my OP.

But news of this upcoming part got leaked from someone or several people.

As to release dates being delayed when there was never an official mention of one, the same thing applies, people had sources as to that and then later newer information came to them.

No criticism was intended.

This is a part I'm interested in and have been tracking.

My searching about it provides new information.

I guess if speculation isn't desired by ATI/AMD then they need to keep things more in the dark.
 
I guess if speculation isn't desired by ATI/AMD then they need to keep things more in the dark.

Impossible where anyone with a blog can make something up from thin air, and the rest of the blogosphere picks it up as a valid rumor.

Unless you can think up a way to prevent people on the net from just coming up with stuff out of thin air. :)

Regards,
SB
 
Impossible where anyone with a blog can make something up from thin air, and the rest of the blogosphere picks it up as a valid rumor.

Unless you can think up a way to prevent people on the net from just coming up with stuff out of thin air. :)

Regards,
SB

Oh absolutely. I was thinking of leaked information from valid sources. Naturally people extrapolate from even just crumbs of information. I concede that that is when things can start to get silly. ;)
 
That would entirely defeat the point of integration!

Then how else will they give the GPU component the bandwidth it needs? IGPs are fine with side-port memory to increase the capabilities of the part, so why not a side port for the CPU as well?
 
Back
Top