• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

6800 series slides

True I sell my current card to fund purchase of the new one, it's not rocket science. As for employment, I don't feel obliged to disclose that to anyone.
 
Rroff said:
Just be careful where you host images from in future then ;)
Click to expand...




Image shack.:confused:


GJ02 said:
works in a top secret lab, thats why his name never ever shows up when he is on the forum !

no but seriously, what do you do? you always seem to be on here, are you gibbo ?
Click to expand...

lol my post count is relatively small for the amount of time I have been registered. Rroff has nearly 3x the amount of posts I have but only been registered a year more than me.
 
Last edited:
RavenXXX2 said:
poor tessellation.
Click to expand...
It made me chuckle at AMD's claim of "tessellating the right way". I have heard it said amd only really paid lip-service to tessellation?. It's hardly suprising though when you think about amd's cypress chips only have a single tessellation unit and then compare that to nvidia's 16 tessellation units in fermi. Apparently, because of the so called buzzword surrounding microsoft’s latest graphics API was tessellation and it’s this feature that nvidia claims was responsible for fermi's late arrival. Nvidia will be king of tessellation for the foreseeable future no doubt, but it's pointless unless more game developers utilize the tech in more of their games. The upcoming Metro2034 will cripple any future graphics card though, lol.
 
Phoenixdancer said:
Blah blach blah.

300 Watts? Not bad considering a 480GTX uses more.

There is no doupt it will kick 480GTX bum
Click to expand...


Latest info suggests 1120 shaders, 32 rops at 255mm2 and better than GF104 performance quite easily, considering the GF104 seems to be around 360-370mm2(only really seen it quoted as 10% bigger than a 336mm2 5870 core), AMD are just destroying Nvidia on efficiency, yields will be higher, prices much lower, get more cores off a wafer and it performs better. Using 30W less.

This is the simple thing, AMD can produce a LOT more performance at the same power level Nvidia can, if the 6970 turns out to be 300W, you can count on it being 35-40% faster.

I don't honestly believe that though, and probably through it up to a 2gb card so 10-15W for the extra gig of mem.

I think people are simply basis 300W off the double a 6870 everything like the 5770 to 5870. Thats unlikely though, they'd be pushing 2200 shaders but also 470-480mm2 which is suicide.

1700-1900 looks about right, a few more shaders than the 5870 + 35% more efficiency.

My prediction on the SA forums btw was a 255mm2 chip with 1120 shaders, though I'll be honest and say thats what I guessed for the Barts Pro, I guessed 1280 for the Barts XT, ruddy close though :p
 
RavenXXX2 said:
So that's their excuse for having poor tessellation, looks like it's carried on through to the 6*** series. Where there is a decent level of tessellation it performs at previous generation level, hardly any tessellation it performs better than previous gen, that's progress for you....:D

Does the new AA not look way to blurry...
Click to expand...

I don't really see how you can say it's poor, at Tessellation factors up to 10 you're getting double the performance of the 5870 and considering this is Tessellation factor 9 how much smaller do you want the polygons to be?

f9a9ck.jpg


Sure Fermi might be faster at really extreme levels of Tessellation, but it won't actually make a difference to model quality.
 
AMD have their stuff locked down good. Were 2 days away from 6870 release and not benchmarks yet,
Info for 6970 wont be out for at least 3weeks.
I expect performance 6850 x2.
 
RavenXXX2 said:
So that's their excuse for having poor tessellation, looks like it's carried on through to the 6*** series. Where there is a decent level of tessellation it performs at previous generation level, hardly any tessellation it performs better than previous gen, that's progress for you....:D

Does the new AA not look way to blurry...
Click to expand...

You're getting confused, slightly, and not because its obvious either. Tesselation factor(cba'd to look up the actual defination) isn't the level of tesselation in a scene altogether, its the level of polygons per, whatever, square inch, whatever. At some stage more polygons in the same small area just kills performance while IQ isn't increasing anymore. Not every surface and every area being tesselated will improve IQ with a higher factor, as broken hope showed, a fairly small "factor" can have incredible detail, I doubt you'll see anything close to the "9" in that picture in any game ever, its simply too much detail.

But its not exactly explained in those slides, hopefully there will be a slide somewhere or reviewers explain what it means.

THe AA looks, I can't really decide, it looks like everything, including the area's between the lines is AA'd, which is washing it out slightly. I thought it looked washed out and blurry, but the more I stare at it the less obvious it seems :p

Honestly, most gens will bring new methods of AA, most of them are crap, get around minor issues in some games and will probably be improved upon, mostly I end up using the same basic/safe AA options

Assuming they are starting to implement full scene AA that can bypass limitations of other methods, in the future it will be improved, has to start somewhere really.
 
You must log in or register to reply here.
Back
Top Bottom