Linked by Thom Holwerda on Mon 6th Feb 2006 22:21 UTC, submitted by Anonymous Goatherd
Thread beginning with comment 93419
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[2]: Other Reasons for the Switch
by rayiner on Tue 7th Feb 2006 04:10
in reply to "RE: Other Reasons for the Switch"
Hmm, good point. Should have read the article more carefully! In that case, POWER6's integer performance might be quite better than expected, since the PA-Semi chip he gets the estimates from is quite a bit simpler than POWER6 is projected to be (3 issue versus 4 issue, etc). If the 604 reference is accurate, POWER6 will be wider than the PA Semi chip, though narrower than the G5.





Member since:
2006-02-07
"POWER6 will be a very different design, taking a cue from Cell and the PPE and moving to a narrower core that can be clocked very high (4-5GHz). The performance estimates IBM has posted are mixed: 2200-2600 SPECint and 4600-5200 SPECfp."
It seems very unlikely that IBM would quote numbers this early. Could you provide a link? Btw, the real world tech article was speculation by David Kanter, not an IBM quote.