-
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.
GTX680 to arrive at the end of February!
- Thread starter CinderzFTW
- Start date
Oh it _should_ beat it but it won't be by any spectacular amount. That said there is a whole lot of rumor atm and very little solid information even from somewhat reliable sources. (From which I was told the top tier part for the first Kepler release (not flagship) was penciled for +60% shaders +20% clockspeed on the GTX480 - which would have put it at 768SP (or equivalent to a 768SP Fermi not sure which but rumours support 768SP) and 840MHz).
Last edited:
Have you any evidence of what memory is actually used. I mean a definate answer that shows it without any sort of pre caching(Unless we are talking above 1080p)
Soldato
- Joined
- 29 Sep 2010
- Posts
- 5,702
So much speculation. 
No idea atm what the specs are on their flagship card but everything seems to indicate this will be the top model for their mid-range lineup in the general Kepler series.
600 series tho seems to be designated for Fermi optical shrinks so if it actually is a "680" then it may be just a 28nm Fermi
Soldato
Ultra settings here with 2560x1600 don't top 1500mb.
No idea atm what the specs are on their flagship card but everything seems to indicate this will be the top model for their mid-range lineup in the general Kepler series.
600 series tho seems to be designated for Fermi optical shrinks so if it actually is a "680" then it may be just a 28nm Fermi
So when would keplar roll out
Soldato
2GB?!
Two of these in SLI will be a downgrade for me then!!!
Ok, not a downgrade, but more of a sidestep
Two of these in SLI will be a downgrade for me then!!!
Ok, not a downgrade, but more of a sidestep
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 33,188
can people not use a little common sense, a 680gtx, a 780gtx, a mustang 28, or a keplersuperdupafasterthan 7970.
The NAME is irrelevant, when people say 780gtx, or 680gtx, or BIG kepler, they are all referring to the same damn thing. The faster single core Nvidia gpu from the next gen.
It also wouldn't matter if a midrange is a fermi shrink, a fermi on 28nm with a lot more shaders, or a kepler on 28nm with more shaders.
We don't know the specs, or the performance, its guesswork, the names are for want of a better word(or maybe its spot on, I dunno) a place holder for the card we are all talking about.
Some generations AMD/nvidia rebrand old gen stuff with new gen names, sometimes they shrink old gen stuff on a new process and give it a new gen name then decide its unfair to call their "real" new gen that name and jump forward another step. AMD and Nvidia have both done every combination of every type of name change there is so predicting the final names based of almost certain low end 28nm parts being called 600 series is daft.
You could have 28nm low end 600 parts, and still have the high end new gen be 680gtx, you could have fermi 40nm low end 600 parts, and still have the high end Keplers be 680gtx. Guessing won't change the fact that the fastest biggest core will be the fastest biggest core, no matter how fast or what architecture it is, and same goes for the middle core.
Meh, I'm high(not fun, pain medication
) and I've seen probably 50 posts on "but but, I thought it was a 780/680/999999gtx/potatocakefishcard, it's peeing me off.
The NAME is irrelevant, when people say 780gtx, or 680gtx, or BIG kepler, they are all referring to the same damn thing. The faster single core Nvidia gpu from the next gen.
It also wouldn't matter if a midrange is a fermi shrink, a fermi on 28nm with a lot more shaders, or a kepler on 28nm with more shaders.
We don't know the specs, or the performance, its guesswork, the names are for want of a better word(or maybe its spot on, I dunno) a place holder for the card we are all talking about.
Some generations AMD/nvidia rebrand old gen stuff with new gen names, sometimes they shrink old gen stuff on a new process and give it a new gen name then decide its unfair to call their "real" new gen that name and jump forward another step. AMD and Nvidia have both done every combination of every type of name change there is so predicting the final names based of almost certain low end 28nm parts being called 600 series is daft.
You could have 28nm low end 600 parts, and still have the high end new gen be 680gtx, you could have fermi 40nm low end 600 parts, and still have the high end Keplers be 680gtx. Guessing won't change the fact that the fastest biggest core will be the fastest biggest core, no matter how fast or what architecture it is, and same goes for the middle core.
Meh, I'm high(not fun, pain medication
) and I've seen probably 50 posts on "but but, I thought it was a 780/680/999999gtx/potatocakefishcard, it's peeing me off.
Soldato
- Joined
- 11 Oct 2007
- Posts
- 2,597
It could just be Nvidea is rushing out a 28nm variant of existing technology while they finish off the Keplar range. Would allow them to compete with the 79x0 cards, thus retain market share etc. Could be quite likely if it really is an April eta.
Personally whatever it is, I would love to see a card dropped on us that will lower the price of competing tech and last gen cards.
Personally whatever it is, I would love to see a card dropped on us that will lower the price of competing tech and last gen cards.
can people not use a little common sense, a 680gtx, a 780gtx, a mustang 28, or a keplersuperdupafasterthan 7970.
The NAME is irrelevant, when people say 780gtx, or 680gtx, or BIG kepler, they are all referring to the same damn thing. The faster single core Nvidia gpu from the next gen.
It also wouldn't matter if a midrange is a fermi shrink, a fermi on 28nm with a lot more shaders, or a kepler on 28nm with more shaders.
We don't know the specs, or the performance, its guesswork, the names are for want of a better word(or maybe its spot on, I dunno) a place holder for the card we are all talking about.
Some generations AMD/nvidia rebrand old gen stuff with new gen names, sometimes they shrink old gen stuff on a new process and give it a new gen name then decide its unfair to call their "real" new gen that name and jump forward another step. AMD and Nvidia have both done every combination of every type of name change there is so predicting the final names based of almost certain low end 28nm parts being called 600 series is daft.
You could have 28nm low end 600 parts, and still have the high end new gen be 680gtx, you could have fermi 40nm low end 600 parts, and still have the high end Keplers be 680gtx. Guessing won't change the fact that the fastest biggest core will be the fastest biggest core, no matter how fast or what architecture it is, and same goes for the middle core.
Meh, I'm high(not fun, pain medication
wtb potatocakefishcard plz
HMMMMMM sounds tasteyyyy
This is one of those rare instances where I agree with Drunkenmaster. Take 'GTX 680' or 'GTX 780' as a metaphor for the 'Kepler GTX 580'. Arguing over the name is the very definition of arguing over semantics.
Its just we have news about the 680 which by specs is not the full 'bread and butter' which has been confirmed as the 7XX series,
Which leaves us to think the 6XX series will be die shurnk fermi and the 7XX will be the new keplar architechture
Today I've read one article claiming a 680GTX will arrive in February, another that new parts from NVIDIA won't be here until April.
My conclusion: no-one has a cogent clue when NVIDIA will be releasing either Kepler or a Fermi die-shrink and the dates are just smoke on the wind. And whilst everyone debates, AMD has the fastest single-GPU card *now*.
My conclusion: no-one has a cogent clue when NVIDIA will be releasing either Kepler or a Fermi die-shrink and the dates are just smoke on the wind. And whilst everyone debates, AMD has the fastest single-GPU card *now*.