Pascal: What do we know? Discussion, Latest News & Updates: 1000M Series GPU's

Discussion in 'Gaming (Software and Graphics Cards)' started by J.Dre, Oct 11, 2014.

Thread Status:
Not open for further replies.
  1. HTWingNut

    HTWingNut Potato

    Reputations:
    21,578
    Messages:
    35,409
    Likes Received:
    9,862
    Trophy Points:
    931
    It sounds unbelievable, but likely true...
     
  2. Deks

    Deks Notebook Prophet

    Reputations:
    1,088
    Messages:
    4,583
    Likes Received:
    1,795
    Trophy Points:
    231
    The likelyhood of Pascal being 80% faster than 980M is a bit out there though.
    I wouldn't expect more than 50% increase since that seems more in line with historical progression.
     
  3. HTWingNut

    HTWingNut Potato

    Reputations:
    21,578
    Messages:
    35,409
    Likes Received:
    9,862
    Trophy Points:
    931
    Historical Progression:

    285M Tesla --> 485M Fermi: Vantage P GPU: 5631 to 12191 = +116%
    485M Fermi --> 680M Kepler: 3DM11 P GPU: 2709 to 5897 = +118%
    680M Kepler --> 980M Maxwell: 3DM11 P GPU: 5897 to 12531 = +112% / Fire Strike Graphics: 4049 to 9682 = +139%

    980M Maxwell --> 1080M Pascal ==> ??? at least +100%!?
     
  4. J.Dre

    J.Dre Notebook Nobel Laureate

    Reputations:
    3,700
    Messages:
    8,316
    Likes Received:
    3,819
    Trophy Points:
    431
    That's my hope and was my original prediction. Pascal should be 100%+ faster. However, I'm anticipating them to hold back Pascal, intentionally, at least for the first release. There will probably be at least two (if not more) iterations of Pascal, 2017 offering another 50%+ over this year's release.

    Pascal's shelf-life will not end this year. That we can all agree on. Re-branding is game within the game.
     
    HTWingNut likes this.
  5. HTWingNut

    HTWingNut Potato

    Reputations:
    21,578
    Messages:
    35,409
    Likes Received:
    9,862
    Trophy Points:
    931
    Yep. Just that Maxwell never officially got a refresh of the top end GPU's like other architectures did. Just more of the same, really. 860M and lower got refreshes to the 9xxM level, but 970m and 980m still trucking along after 18 months without even a refresh in sight.
     
  6. PrimeTimeAction

    PrimeTimeAction Notebook Evangelist

    Reputations:
    250
    Messages:
    541
    Likes Received:
    1,135
    Trophy Points:
    156
    The jump between Kepler and Maxwell should have been even bigger. Remember originally Maxwell was supposed to be die shrink. But they had to skip 20nm due to issues with TSMC and finfet.
    And I think 780m was worthwhile re-brand of 680m. I would have been happy to see a similar re-brand of 980m.
    Edit: Remember this?
    [​IMG]
     
    HTWingNut and i_pk_pjers_i like this.
  7. Prema

    Prema Your Freedom, Your Choice

    Reputations:
    8,239
    Messages:
    6,093
    Likes Received:
    15,537
    Trophy Points:
    681
    Ethrem, tgipier, jaybee83 and 2 others like this.
  8. HTWingNut

    HTWingNut Potato

    Reputations:
    21,578
    Messages:
    35,409
    Likes Received:
    9,862
    Trophy Points:
    931
    Right. Just an architecture change improved performance dramatically on the same node size.
     
  9. Cloudfire

    Cloudfire (Really odd person)

    Reputations:
    7,279
    Messages:
    10,310
    Likes Received:
    2,878
    Trophy Points:
    581
    Shouldnt we start to see 1080M leaks soon if its available in products in mid June? Only 3 months away
     
    moviemarketing likes this.
  10. J.Dre

    J.Dre Notebook Nobel Laureate

    Reputations:
    3,700
    Messages:
    8,316
    Likes Received:
    3,819
    Trophy Points:
    431
    Yeah, should be seeing leaks around GTC in April.

    If NVIDIA doesn't announce Pascal and present a working Pascal GPU at GTC, then they probably won't launch until late summer.
     
Loading...
Thread Status:
Not open for further replies.

Share This Page