InvestorsHub Logo

mas

Followers 13
Posts 14434
Boards Moderated 0
Alias Born 01/08/2004

mas

Re: Grimes post# 41143

Sunday, 08/01/2004 4:38:41 PM

Sunday, August 01, 2004 4:38:41 PM

Post# of 97585
well sgolds pretty nailed it but to add to that.

1. Yes

2. Until 65nm Prescott or redesigned Dothan hits in 2006. A 2+Ghz single-core Montecito is the darkhorse here. Who knows what could happen if Intel could harness that FP power on common desktop applications.

3. Well Netburst is eol at 65nm, a marketing driven processor which has hit power rocks, so it will have to be Dothan derivatives. The 2 Ghz Dothan reviews have looked good to me, when they double the fsb and release the power limits it will be very close to K8 in say 32bit games performance. 64 bit will probably open up a new advantage for K8 when WinXP64 is released as EM64T is probably a quick kludge. The Dothan with EM64T will be designed from the start with that and should be on a par with K8.

4. They probably started in early 2003 when Prescott early steppings started sucking so should be ready out of the oven in 2005-6. By then AMD should be around 3 GHz with K8 and readying K9 for release. The K9 is a long-pipeline SMT design according to rumors with entry speeds close to 4 Ghz. If AMD get it right it will have ipc close to K8 with a 1Ghz advantage wink. If they get it wrong then they will have to go back and redesign K8 (longer pipeline) to get it to 45nm at 4+GHz. AMD should at least have performance parity or more likely a speed grade advantage as they have had since Jan 2004.


Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent AMD News