[R300King!]

http://www.forum-3dcenter.org/vbulle...58#post5449558

What's this?

R600:
64ALUs
16TMUs

RV630:
24ALUs
8TMUs

RV610:
8ALUs
4TMUs

[Ailuros]

Originally Posted by nAo

New 'advanced' custom AA resolve passes are nice and everything but it seems to me that nothing could stop competition to implement the same filters, in fact I'm willing to bet R6xx uses its shaders code to implement those filters.

Just about when I was happy that Quincunx was gone....heh

[nAo]

Originally Posted by Ailuros

Just about when I was happy that Quincunx was gone....heh

It's going to be much better than quincunx, I experimented a few months ago wiith custom AA resolve filters and even a simple gaussian filter on a 4X multisampled image (just a bit wider than a single pixel) was given extremely good results + a slight blur..

[Jawed]

Blimey that AA is going to be fun to play with But only 2x, 4x and 8xMSAA? Hmm, I was hoping for 4x, 8x, 12x. I really want to see movies of these CFAA modes now...

I see I just about got there in time to post my speculation on the ALU operation

LOL, I moaned about the asymmetry of the ring-bus, now it's symmetric. I even convinced myself it was inevitably asymmetric, hahahaha.

I've asked before, I'm gonna ask again, what use are fp32-per-component bilinearly filtered texels?

Jawed

[BRiT]

More ATI Slides, these showing performance comparison to Radeon 1950XTX, for Stencil Shadow, HDR, and Crossfire [ http://www.forum-3dcenter.org/vbulle...&postcount=414 ]

[dnavas]

Originally Posted by nAo

New 'advanced' custom AA resolve passes are nice and everything but it seems to me that nothing could stop competition to implement the same filters, in fact I'm willing to bet R6xx uses its shaders code to implement those filters.

In theory, the R600 has slightly more math/shader power to throw at something like this, though.

Originally Posted by Jawed

R5xx introduced a dedicated branch pipeline. The diagram appears to show a little blue runty thing, that's prolly it.

Thanks, I missed that.

-Dave

[Bouncing Zabaglione Bros.]

Originally Posted by Ailuros

Just about when I was happy that Quincunx was gone....heh

Hasn't there been general talk for quite a while of moving away from hardcoded AA methods applied via control panel to the whole screen, to AA applied at the discretion of the developer directly where it is needed using shaders as the mechanism?

[tEd]

Originally Posted by nAo

It's going to be much better than quincunx, I experimented a few months ago wiith custom AA resolve filters and even a simple gaussian filter on a 4X multisampled image (just a bit wider than a single pixel) was given extremely good results + a slight blur..

Can you share the results? (Pictures)

[Rangers]

Well it's nice to see this thread got exciting again with at least rumors that R600 can beat GTX.

I still have a hard time believing it though, because why $400 if that's the case?

So glad they moved NDA up to May 2..

Someone back there mentioned supposedly Crysis running faster on R600..again that wouldn't surprise me if the TMU limits are true. Crysis is probably going to stress shaders like no game ever before by a factor of several.

[INKster]

Originally Posted by BRiT

More ATI Slides, these showing performance comparison to Radeon 1950XTX, for Stencil Shadow, HDR, and Crossfire [ http://www.forum-3dcenter.org/vbulle...&postcount=414 ]

That scale on the slides comparing it to the R580...
Shady, very shady.

[nAo]

Originally Posted by dnavas

In theory, the R600 has slightly more math/shader power to throw at something like this, though.

That's true but you need also to sample your sub samples..

[Bouncing Zabaglione Bros.]

Originally Posted by INKster

That scale on the slides comparing it to the R580...
Shady, very shady.

I noticed that, but it is somewhat mitigated by the R580 baseline being clearly marked, and the percentage increase clearly up the side, as well as just the frames on the bars.

[BRiT]

Originally Posted by INKster

That scale on the slides comparing it to the R580...
Shady, very shady.

And what would you have them compare it to? Or do you mean how they only show 1 block of "80 to 100%" and then leave 5 more blocks above that to go upto 200% ? Yeah, it's very marketting-like, but at least they provide you with raw numbers too.

[INKster]

Originally Posted by Bouncing Zabaglione Bros.

I noticed that, but it is somewhat mitigated by the R580 baseline being clearly marked, and the percentage increase clearly up the side, as well as just the frames on the bars.

I have no problem with that.
It's just that, this way, whoever sees those graphs and doesn't know any better will see a R600 with a bar that looks as if it was several times faster than a R580, when in fact it's closer to 0.8x times faster, at best.

If the graphs started at zero with 10% intervals, the impact would be much less pronounced.
I don't like it when there's this marketing effort trying to obscure clarity. Not everyone notices these subtleties.
And frankly, AMD is starting to "usurp" ATI's clear way of doing marketing (they took over this department directly, didn't they ?).

[Ailuros]

Originally Posted by nAo

It's going to be much better than quincunx, I experimented a few months ago wiith custom AA resolve filters and even a simple gaussian filter on a 4X multisampled image (just a bit wider than a single pixel) was given extremely good results + a slight blur..

No objection there; one objection is that filter samples get counted as AA sample densities and a close second that if we'd be talking about 16x or even 24xCFAA how "slight" will the blur be there?

[Ailuros]

Originally Posted by Bouncing Zabaglione Bros.

Hasn't there been general talk for quite a while of moving away from hardcoded AA methods applied via control panel to the whole screen, to AA applied at the discretion of the developer directly where it is needed using shaders as the mechanism?

I don't think the custom AA filter thingy of R600 has anything to do with shader antialiasing.

[Jawed]

Originally Posted by BRiT

And what would you have them compare it to? Or do you mean how they only show 1 block of "80 to 100%" and then leave 5 more blocks above that to go upto 200% ? Yeah, it's very marketting-like, but at least they provide you with raw numbers too.

That's the problem with this HD era, 16:9 aspect ratio robs your bars of their height

Bring back 4:3 I say.

Jawed

[Bouncing Zabaglione Bros.]

Originally Posted by INKster

I have no problem with that.
It's just that, this way, whoever looks at those graphs and doesn't know any better will see a R600 with a bar that looks as if it was several times faster than a R580, when in fact it's closer to 0.8x times faster, at best.

Reading a graph like that isn't rocket science. If you're the sort of person that is only going to see the pretty coloured bars and not look to see what the scale means (especially when it is so clearly marked), you've really no business looking at anything as complex as a bar graph!

[nAo]

Originally Posted by Ailuros

No objection there; one objection is that filter samples get counted as AA sample densities and a close second that if we'd be talking about 16x or even 24xCFAA how "slight" will the blur be there?

I agree on the first objection, it's just marketing but we shouldn't get suprised by that, every company acts this way.
Dunno about the second one but I'd like to find that out as soon as possible, even though the fact that they have narrow and wide filters is already a hint for us, imho

[INKster]

Originally Posted by Bouncing Zabaglione Bros.

Reading a graph like that isn't rocket science. If you're the sort of person that is only going to see the pretty coloured bars and not look to see what the scale means (especially when it is so clearly marked), you've really no business looking at anything as complex as a bar graph!

Refer to the second part of my previous post.
Like i said, i have no problem with that, but it's still "unusual" in ATI's ways of doing marketing.
I would expect that from Nvidia or Intel, as their marketing departments have a very aggressive reputation.

[tEd]

Originally Posted by INKster

Refer to the second part of my previous post.
Like i said, i have no problem with that, but it's still "unusual" in ATI's ways of doing marketing.
I would expect that from Nvidia or Intel, as their marketing departments have a very aggressive reputation.

It's not ati anymore.

[Rangers]

Originally Posted by INKster

Refer to the second part of my previous post.
Like i said, i have no problem with that, but it's still "unusual" in ATI's ways of doing marketing.
I would expect that from Nvidia or Intel, as their marketing departments have a very aggressive reputation.

ATI did the same thing on R520 and people bitched then too.

It's not new even from ATI.

[Bouncing Zabaglione Bros.]

Originally Posted by INKster

Refer to the second part of my previous post.
Like i said, i have no problem with that, but it's still "unusual" in ATI's ways of doing marketing.
I would expect that from Nvidia or Intel, as their marketing departments have a very aggressive reputation.

And it's been flagged for a long time as being an area where the former ATI was deficient compared to the marketing beast that is Nvidia. I don't think it would be a bad thing if AMD gets a bit more aggressive on the marketing front. It's not like these graphs are misleading if you actually take the trouble to read them. They are pretty mild compared to other graphs I've seen where you don't even get a scale on the side.

[INKster]

Originally Posted by tEd

It's not ati anymore.

Semantics.
R600 is still an ATI child.

[Kaotik]

Originally Posted by INKster

Refer to the second part of my previous post.
Like i said, i have no problem with that, but it's still "unusual" in ATI's ways of doing marketing.
I would expect that from Nvidia or Intel, as their marketing departments have a very aggressive reputation.

Maybe they put old AMD's marketing team on the job?