1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

winchester, new castle, manchester ect....

Discussion in 'General Hardware' started by HiddenStupid, Aug 24, 2007.

  1. HiddenStupid New Member

    Joined:
    Jul 17, 2007
    Messages:
    741 (0.27/day)
    Thanks Received:
    8
    What is the name of all AMD prcessors? I mean why did they name it?

    does each name superior to another? is it by performance ranking? does it matter? I have atlon64 3200+ winchester.... is the ahlon64 3200+ new castle better?

    can someone list of all the names in order by performance ranking?
     
  2. von kain New Member

    Joined:
    Sep 13, 2006
    Messages:
    782 (0.26/day)
    Thanks Received:
    49
    Location:
    athens/greece
    take a look @ the cpu database the main difference is the stepping the ht ant the transistors number





    p.s we have the same cpu
     
  3. Tatty_One

    Tatty_One Super Moderator Staff Member

    Joined:
    Jan 18, 2006
    Messages:
    16,941 (5.19/day)
    Thanks Received:
    2,745
    Location:
    Worcestershire, UK
    Well like anything, it's not quite as simple as that, processor "names" (effectively core names) is not the only ingredient, therefore NEW processors are introduced either because there is a new socket......Socket A.....S754.....S939.....SAM2 etc, often but not always with a socket upgrade you get a feature or fabrication improvement........for example........130nm > 90nm (Winchester to Venice.....Clawhammer to San Diego as an example).....or SSE2 > SSE3 (Winchester > Venice again as an example).

    New cores does not necessarily mean a new socket though, could just mean a new fabrication process or feature set which as I have said could be fabrication process, instruction set or FSB rating etc, for example, Socket A at the end of it's life saw Barton core;s rated at 333FSB upgraded to 400FSB that remained socket A. On top of that, usually, but not always the improvement does bring enhanced performance but there are exceptions, many would say that clock for clock, some of the S754 processors were a match for their equivilent Sockt 939 ones, despite socket 939's have what was then the new "on die" dual memory controller.

    Here is a list showing Core/CPU name, L2 cache and fabrication process, there are some cores before Socket A but this will take you back far enuff:

    Socket A

    Thoroughbred Athlon XP/Sempron 256K 130nm
    Duron 256K 130nm
    Thorton Athlon XP/Sempron 256K 130nm
    Barton Athlon XP 512K 130nm

    Socket 754

    Clawhammer Athlon 64 1024K 130nm
    Newcastle Athlon 64 512K 130nm
    Paris Sempron 256K 130nm
    Palermo Sempron 128K/256K 90nm SSE3 enabled

    Socket 939

    Newcastle Athlon 64 512K 130nm
    Sledgehammer Athlon 64/FX 1024K 130nm
    Winchester Athlon 64 512K 90nm
    Venice Athlon 64 512K 90nm SSE3 enabled
    San Diego Athlon 64/FX 1024K 90nm SSE3 enabled
    Manchester Athlon 64 X2 2 x 512K 90nm SSE3 + Dual Core
    Toledo Athlon 64 X2 2 x 1024K 90nm SSE3 + Dual Core
    Opterons had a Toledo followed by a Denmark core I think.

    Socket AM2....goes something like this in more detail:

    Windsor (dual)......2x 512 or 2x 1mb L2 cache....some "EE" as well as standard 90nm

    Orleans (single)....512mb L2 cache......90nm

    Manilla (single).....256mb or 2x 256mb L2 cache....90nm.....THESE WERE AM2 SEMPERONS

    Barcelona (dual)....65nm 2x 512mb L2


    I may have missed the odd one there ot got something wrong, not sure, just quickly done this from memory! I am sure someone will tell me, thewre is a lot more I could say on these processors but as you can see, there is enuff here to bore the wart of a pigs arse!
     
    Last edited: Aug 24, 2007

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page