R420 and R423 details from HardOCP

[AlphaWolf]

Well if nvidia can run a .13 micron part with 222million transistors @ 400 mhz ATi ought to be able to get a 160-180 million transistor part to run at 500mhz on the .13 micron process. Surely ATi was able to manage 380mhz with a 110 million transistor chip on the .15 micron process. I think though that the 16 pipeline variety of the R420 will be over 200 million transistors. ATi has a history of making cooler running processors compared to nVidia at the same process and relatively same transistor count but in this case they might be running into some limitations and implement similar cooling a la nVidia, maybe.

ATI managed 432 mhz on a 110 million transistors with the 9800XT.

The 9800xt isnt on low k

It wasn't on .13 micron either, but that wasn't really his point was it.

 

[991060]

Looking at the memory bandwidth, you just have to choose where you want your bottleneck

Shader driven applications ain't that sensitive to bandiwdth, a 20% clock boost( 500MHz->600MHz) should give R420 a margin in future games.

Now the biggest concern is the yield and the power consumption( linear related to the clock speed? ) IMHO.

 

[Evildeus]

Is this information new?
RV380 = X600
RV370 = X300

 

[AlphaWolf]

Is this information new?
RV380 = X600
RV370 = X300

I've seen it before (week or so ago), dunno if it was posted here or not.

 

[ERK]

[quote:HardOCP]Hard|OCP reports that various versions of the R420 will be branded as RADEON X800 PRO, RADEON X800 XT, RADEON X880 XT and RADEON X800 SE citing a “classified ATI documentâ€￾. The solutions will enable different number of rendering pipelines: 12, 16, 16, 8 and will function at clock-speeds of 500MHz, 600MHz, 600MHz and 450MHz respectively. Higher-end graphics cards will have 256 or 512MB of GDDR3 memory functioning at 1000 and 1200MHz for “PROâ€￾ and “XTâ€￾ parts. The performance-mainstream offering RADEON X800 SE will boast with 800MHz DDR memory with 128-bit bus.

So if this is correct and the R420 does comeout with 16 pipes and 600 Mhz Clock with 1200 Mhz GDDR3, how much memory bandwidth are we looking at??

I know the NV40 with 16 pipes and 400Mhz Clock with 550 Mhz GDDR3 gives a memory Bandwidth of 35.2GB/sec(as per Dave's preview).

News Source: XBit Labs

Maybe I missed something, but what is the difference between X800XT and X880XT?
Thanks

 

[Joe DeFuria]

Maybe I missed something, but what is the difference between X800XT and X880XT?
Thanks

X800XT= AGP 8X
X880XT = PCI Express

 

[plat]

pci-express i think

 

[AlphaWolf]

Well I was clearly beaten to the punch x3.

x880 = peg

 

[ERK]

Thanks, folks!
ERK

 

[Ardrid]

I don't know, 600MHz seems like a stretch. If we look at RV360, which uses the same process and has less than half the transistors of R420, ATI was able to clock it to 500MHz. Now, more than doubling the transistors and using the same process, it just doesn't seem like they're going to be able to hit 600MHz on R420. That's just me though

 

[Stryyder]

I don't know, 600MHz seems like a stretch. If we look at RV360, which uses the same process and has less than half the transistors of R420, ATI was able to clock it to 500MHz. Now, more than doubling the transistors and using the same process, it just doesn't seem like they're going to be able to hit 600MHz on R420. That's just me though

R420 was easily overclocked to 580+Mhz without artifacts and that was there first attempt at the technology Personally I wouldn't be suprised if their highes yield boards hit 600 easily.

 

[Ardrid]

I don't know, 600MHz seems like a stretch. If we look at RV360, which uses the same process and has less than half the transistors of R420, ATI was able to clock it to 500MHz. Now, more than doubling the transistors and using the same process, it just doesn't seem like they're going to be able to hit 600MHz on R420. That's just me though

R420 was easily overclocked to 580+Mhz without artifacts and that was there first attempt at the technology Personally I wouldn't be suprised if their highes yield boards hit 600 easily.

When did this happen? To my knowledge, no one has their hands on R420. Did you mean RV360?

 

[Joe DeFuria]

I don't know, 600MHz seems like a stretch. If we look at RV360, which uses the same process and has less than half the transistors of R420, ATI was able to clock it to 500MHz. Now, more than doubling the transistors and using the same process, it just doesn't seem like they're going to be able to hit 600MHz on R420. That's just me though

You can't use RV360 as a basis for predicting R420 in this way.

RV360 is designe for much higher volume (and higher yields), and lower power than the R420 will be. RV360 has a much different set of goals than the R420.

For example, what's the "fastest" that the Radeon 9000 is clocked at? 275 Mhz...that's approx 35 million transistors on TSMC's 0.15u process. Now look at the R360: 110 million transistors on the same process....and 400+ Mhz.

 

[Stryyder]

I don't know, 600MHz seems like a stretch. If we look at RV360, which uses the same process and has less than half the transistors of R420, ATI was able to clock it to 500MHz. Now, more than doubling the transistors and using the same process, it just doesn't seem like they're going to be able to hit 600MHz on R420. That's just me though

R420 was easily overclocked to 580+Mhz without artifacts and that was there first attempt at the technology Personally I wouldn't be suprised if their highes yield boards hit 600 easily.

When did this happen? To my knowledge, no one has their hands on R420. Did you mean RV360?

Yes I did sorry for the oops brain not working well today

 

[kemosabe]

I don't know, 600MHz seems like a stretch. If we look at RV360, which uses the same process and has less than half the transistors of R420, ATI was able to clock it to 500MHz. Now, more than doubling the transistors and using the same process, it just doesn't seem like they're going to be able to hit 600MHz on R420. That's just me though

R420 was easily overclocked to 580+Mhz without artifacts and that was there first attempt at the technology Personally I wouldn't be suprised if their highes yield boards hit 600 easily.

When did this happen? To my knowledge, no one has their hands on R420. Did you mean RV360?

Ditto. Not that I find it that hard to believe, but where have there been any reports, or even rumours, about R420 clocking that high? :?

Edit: Never mind, just saw your correction.

 

[Dave Baumann]

To my knowledge, no one has their hands on R420.

Huh. Really?

 

[demalion]

Bang! Zoom! Right in the kisser...

To the moon, Wavey, to the moon!

 

[jimmyjames123]

ATI's final clock speeds may be somewhat dependent on how the R420 compares to the NV40. If they can accomplish their goals by using a 500Mhz core instead of 600Mhz, then why not? This is definitely going to be an interesting race.

I also think that R420 reviews/previews may be just a few weeks away. I'm sure there are several people who have already gotten their hands on, such as top developers and possibly even some reviewers

 

[Ardrid]

To my knowledge, no one has their hands on R420.

Huh. Really?

Well, no one that's actually confirmed anything. I'm pretty sure reviewers have their hands on them.

 

[Pete]

To my knowledge, no one has their hands on R420.

Huh. Really?

Yes, well, remember that this is a forum for people on the outside trying to look in. There was an implied "that's shared that info with us" in Adrid's post, just as there was an implied, "Nyah, nyah!" in Dave's.

Seriously, where's the R420's Vegetto-Ex? We need the inside scoop so we can publicly mock it and then sheepishly admit our inner cynics!

So, Dave, does all the rush reviewing and partisan sniping lower your enjoyment of new cards, or are you still thrilled when you see advances like NV40? I'll admit that some reviews didn't show NV40 in the best light, but that card seems to flat out fly compared to NV3x--a really impressive turnaround by nVidia.