Cayman PRO/XT in nov:
"Touch me, I'm reality"
GTX580 in nov/dec:
"Wait for me, I'm a dream"
Printable View
Cayman PRO/XT in nov:
"Touch me, I'm reality"
GTX580 in nov/dec:
"Wait for me, I'm a dream"
Humm I have to call you on that one. There is the planning and execution of the 6870 on amd's side. That after only rumors up to the very last day. This time there are official slides open and motivated claiming Cayman is out in November. I think its a lot more than PR.
Same time it means Nvidia's rumors need to be taken into considerarion as well. The difference is Nvidia remains rumours but amd is as near fact as can be.
This is true.
Remeber, we already have Cayman's little brother, Barts, on the market. On Barts XT, you get near Cypress XT performance at 50-watts less energy consumption. That shows that AMD has room to stretch their legs on Cayman. This information, combined with the information on AMD's briefing slides should tell anyone with half a brain that AMD means business and that Cayman is not just a myth.
I guess we'll see who's right in a few weeks. ;)
I'm sure they both exist, I have not reason to think otherwise. It's just the semantics that one is more real than the other, its a bit silly.
We've gone from arguing over PR slides to arguing who's PR is more believable but we really don't really have any tangible info about the real deals yet.
Its funny that the less information there is available the more trivial and nit picky the arguments get...
It's not really semantics
One is an official slide with a roadmap, and AMD stating Cayman is out late november, go ahead and publish that on your review
The other is from Fudzilla...
Big difference
so gtx580 is real ??? what proof to you have???
a black gpu cooler ??? how's that for proof??? is it written on it gtx580 at least???
model name in drivers and nvidia's page ... so it must be real ... but they did remove the name shortly afterward right??? so why would they remove the name of something that exist ... ahh yes ... the good ol surprise we got a mid life rebranded kicker product technique .. and all of that info came days before barts launch ... coincidental or not you decide ...
so cayman is more real because we saw lots of slides .. and bart is as real as people on the forum who got one on launch day ...
I have no doubts gtx 580 will exist at some point. And what performance both it and Cayman will have is irrelevant to my argument. What I was stating is that AMD has been executing its announcements with very little deviation. Nvidia has not made any official annoucements, there have only been rumors of annoucements so far. And previous history does not help its cause when it comes to that.
To be fair though, the only thing late has been fermi cards as of late and earlier generations they have been on time. If we look at the past, g200, g80, g70 and g60 were on time and in the cases of g80 generation, AMD was 6 months compared to NV. And the product delivered from hype (r600) was pretty bad(I think worse than fermi even as it used more power and was alot slower) and AMD still made a pretty good recovery after.
Which is why I do not discard a 580 at all. Especially if its 384 bit 512 cuda. I did think 512 bit 768 cuda and 128 tmu like some rumors claimed was completely impossible but those rumors seem to have died out.
Its quite feasible for 512 cuda gtx580 max 300 watts to pop up and 15-20% faster than 480 is quite expected after 6 months in the workshop. Lets hope it happens soon!
PS- Nvidias target should be 5970. If they can match or even beat it with one chip at 40nm, then cudos for them really. Fermi is very held back at 40nm.
That can also be said as why would they even bother to add a product id if it didn't exist, glass half full or glass half empty mentality.
Its strangely difficult to go between ATI and Nvidia threads and not get drawn into semantics. We'll just have to wait and see what happens, as far as I can tell its pretty much a sure thing cayman will launch(according to the pr) the question now is Nvidia going to ambush the launch(alluded to by the pr).
Meh. Nvidia is attempting to do what it failed in 480? And that is "being the fastest single GPU around?" As far as I can recall 480 WAS the fastest single GPU around. It didn't fail on performance. Hell, compared to the AMD offerings it didn't fail on price either.
It failed on power consumption, and it failed hugely. It also failed to be on schedule, and it failed on living up to the hype. The latter two are obviously connected.
Fermi cards' timing has failed so horribly that a great mainstream card they released could keep its spot only for 3 months - why? Because it was overtaken by AMD's NEXT GEN offerings. There is no excuse for this, really.
If this time Nvidia's timing with 580 more or less matches with AMD's timing with Cayman; the card is on par with Cayman on price, performance and power, THEN they will have succeeded.
Also, this is about enthusiast / high end only. Right now Nvidia may not be the way to go at mainstream, but they provide very good and competitive alternatives to AMD, something they achieved with massive price cuts.
omg, nothin new?
No everyone is joying over the GTX580 the world's first 2d 3d-accelerator card.
Not much, although I just caught this in the Xbitslab Review. Emphasis mine
Quote:
A next-generation tessellator from AMD will be implemented in the Cayman GPU (Radeon HD 6900). AMD claims it to be three times as fast as the Cypress’ tessellator. We guess AMD’s engineers will focus on increasing the performance of the tessellator proper in the next implementations, possibly on optimizing the hull shaders. We can expect the future architectures, Southern Islands, Hecatonchires, etc., to bring some innovations into the very design of the tessellation pipeline, like what Nvidia offers in its Fermi architecture where each large array of stream processors has a dedicated tessellator for optimal data threading.
my prediction on the GTX 580 is late march early april
no need to get 580. cayman will blow it away
edit: that Xbit lab review on the 6800 series is good, nice find
http://www.xbitlabs.com/articles/vid...70-hd6850.html