To: Scumbria who wrote (61357 ) 6/11/1999 3:22:00 AM From: Tenchusatsu Read Replies (1) | Respond to of 1571802
Thanks, Scumbria, for your notes. I just got back from playing some b-ball, which is why I still have to catch up with all these posts. My initial feeling is that Dirk is exaggerating some of the advantages of the K7, but then I remind myself that the K7 probably doesn't need any exaggeration. Those 3D Winbench scores, however, aren't too reliable since they seem to change with the wind (and the system configuration, and the drivers, etc.). As for SpecInt and SpecFP scores, well, at least now we finally have some ballpark figures regarding K7 performance, so we no longer have to resort to benchmarks run by a high school kid. ;-) Here's a few other points in the presentation that I'll comment on: <The point-to-point connections on the bus allow for much higher speeds than Intel CPU's. 400 MHz is easily obtainable.> This statement is very interesting, since Intel is sticking to the multiprocessor bus with Merced. I don't know about the 400 MHz speed being "easily attainable," or else current Alpha 21264 systems would be running their processor port faster than just 333 MHz. <No third party chipsets until the end of the year. Lots of support after then (VIA, SIS, etc.)> This is perhaps the biggest concern. AMD really doesn't have the fab space to dedicate toward chipsets. The lack of available third-party chipsets means the K7 infrastructure is still in question. Like I said before, it's hard for me to believe that Dirk Meyer has any reason to stretch the truth here. K7 really does sound like serious competition to Intel. (Manufacturing concerns aside, of course.) Tenchusatsu