I agree, it is just that the current word is that Juniper is the codename for RV830/840.
Can we get the codenames right? Now many sites are posting news with different codenames pointing to different things.
Fill in the missing
- Cypress = 40 nm DX11 wafer?
- RV870 = 40 nm DX11 GPU?
- Evergreen = Graphics card based on 40 nm DX11 GPU?
Anandtech:
VR-Zone:Quote:
Evergreen is the codename for AMD's 40nm DX11 based product, although with AMD's codename shift this time around it is still a little ambiguous as to whether Evergreen is a GPU class or a specific configuration.
Semiaccurate:Quote:
Expected launch of RV870 codenamed "Cypress" (old) or "Evergreen" (new) is in September this year.
Huh?Quote:
showing off not only an Cypress (commonly mis-known as R870) wafer, but also working silicon
this is gonna get interesting...
I like the last slide saying "Sooner than you think" :D... hope that means summer sometime for this to be in my PC!
EDIT:
http://www.semiaccurate.com/2009/06/...ware-tomorrow/
Hmmm... what time is that in US Eastern Standard Time?Quote:
June 2, 2009
JUST BEFORE NOON Taipei time on Wendesday, ATI is going to to drop a bombshell on the graphics world, showing off working 40nm DX11 silicon. Nvidia has yet to tape out a DX11 part or put a 40nm chip on the market.
The cards that will be showed off are Evergreen/Cypress, what is commonly referred to as R870. That code name however is not used internally, so if you ever see any leaked specs or charts with R870 on them, they are flat out faked. There have been several of these bandied around, but they are all laughably off.
AMD promised that they would have DX11 cards out on the market by the Windows 7 launch, and barring a massive and as of yet undetected problem, it looks like they will make that with ease. In the mean time, at the company's Computex press conferences, Nvidia distinctly avoided any mention of DX11 or parts that run it..
Mid-day tomorrow, the winners and losers for the next 2-3 quarters will become very very clear.
EDIT2: Nevermind, it's 5:41am EST now, and it is therefore 5:41pm Wednesday in Taipei ;), already happened.
While you guys were busy away chatting about code names and what not, I was watching some vids used in the presentation.
http://www.youtube.com/user/Hardwarehorst
Courtesy Mau1wurf1977 from OCAU :)
Are 2000 SPs possible with such a small die size?
IMO unlikely.If 180mm2 is the figure for RV870,I think we will see 1000-1200 SP on that part.It's still a good bump over RV770 and coupled with higher clocks,lower power draw and DX11 support it could be best perf./watt part out there. Oh and 2xRV870 for high end part as usual :).
Its very nice to see the diesize. It could also explain why they want 5870X2 out. Since 5870 might not have so much raw power over 4890. But DX11...yumyum.
im betting this was done as a rush job, get a dx11 great chip out asap to get some market share. then i think once nvidia gets their monster chip out, ati will be ready with an even bigger one after the 40nm process is spot on. i think were either looking at the 5830 equivalent, or it will actually be the 5870, and we will see a 6870 very early 2010.
lets all hope this one can clock above 1ghz stock, and OC 20+% more, or there may be no need to jump after it if you already have a 4890. but the pricepoint seems to be well within the range of anyones budget
I absolutely agree. Judging from the disaster that is the HD 4770 production, I just can't see TSMC's current 40nm process being up to the challenge of producing a high-end chip for the next little while. What AMD will probably do is release a mid-range card to grab market share and then release their higher-end offering some time next year. From a yield and binning standpoint, an approach like that only makes sense.
don't believe NVIDA FUD!
AMD will have full DX11 lineup for Win 7 launch time-frame... so all the talks about problematic TSMC 40nm tech are completely off... yeah they've struggled in the first half of this year, but all the pieces are coming together, and all be fine in time for RV870 mass production, and that's sooner than expected ;)
did ati know if the 40nm process was going to be perfect in time for win7? probably not, so why spend a huge effort making such a killer chip, when there is no need to waste the R&D when they knew they would be the first dx11 card. make one just strong enough to replace the current generation, and be happy with it.
So is DX11 going to be worth it? I'm planning a build in August and I want to know whether I should hold out on the graphics card like I did with DX10. My 8800gt is still getting me through WoW, TF2, and L4D, and if DX11 is worth it I'll be holding out for a new card for another 4 months (edit: after august).
Good to see they have something at least somewhat ready.
im waiting till next spring before i buy a dx11 card, i want to see how things pan out. i rushed in to get dx10 and got screwed, i should have waited for 10.1
as long as your not buying any new games that kill the 8800, i dont see why you should buy anything yet.
http://www.fudzilla.com/images/stori...n_die_shot.jpg
How big do you guys think the die might be ?
its 180mm^2 like every review of it has said. the RV740 is 120mm^ and the RV770 is 280mm^2 (someone double check my numbers pls)
EDIT: Anandtech numbers
i dont know if i like the fact that a 4770x2 was suppose to be close to a 4890, and this is going to be no where near double the size of the 4770Quote:
We can ascertain from the wafer shot AMD provided us that there are about 19.5 dies vertically and 25.5 dies horizontally. As this is a 300mm wafer, we can sort of "guess" the dimensions of the chip at roughly 15.38mm x 11.76mm resulting in a die area of a little over 180mm^2. This is much smaller than the 55nm RV770 which is a 260mm^2 part, so if we expect a similar price target for the first Evergreen die that we saw with RV770, we could see a significant cost savings (and hopefully this tiny die will deliver a good general performance improvement as well).
Compare this die size to RV740 that weighs in at 137mm^2 and 826 million transistors, and we can estimate very loosely that Evergreen could come in at something over 1 billion transistors. Certainly the process has been tweaked since RV740 and Evergreen is a different architecture and layout, so scaling isn't exact or direct, but proportionally AMD should be able to fit that many transistors into only 180mm^2 on the 40nm process.
RV740 is 136mm2, RV770 was supposedly ~256mm2, really 276mm2? RV790 is ~282mm2.
Not confirmed.
The sources that are hearing these codenames have nothing other than the names, no pricing/performance brackets, no RV8x0 designations, no specifications.
CJ heard Cypress back before the end of '08. We guessed that is was the new codename for RV870 but he never actually confirmed it 100%.
Then when other news sites started picking up the new codenames for ATi just a week or so ago, he dropped another name, Juniper and speculated that it was RV830.
So basically we just have the codename of Cypress, which could be anything, a specific chip or maybe the codename for the entire family of chips and Juniper, which seems to be RV830.
We now have a couple sites reporting that this 180mm2 die is Evergreen, with no confirmations that I have gathered.
So there are quite a few different scenarios for the codenames but I think it would be best to just wait a couple of weeks until more info is leaked.
who is able to make a picture or two?;)
i have jet to see a launch where pictures arent out a few weeks after the production of first samples...
evergreen could be the successor of rv770pro, seems plausible...