1.
Of the 4000+: "Moreover, its rating seems to us a bit too high, because its advantage over Athlon 64 3800+ is only 1-3%."
Bearing in mind that the only difference between the two is the extra L2 cache, can this be an indicator that we have already reached the point at which extra cache doesn't help Athlon64 performance? It seems to make sense, since AMD has been pushing for its transition from Clawhammer to Newcastle- I guess they realized from fairly early on that extra MHz would improve performance more significantly than extra cache. Along similar lines, compare the performance deltas between 3800/4000 (extra cache, same speed) & 4000/FX55 (same cache, extra speed); in the majority of benchmarks, the latter jump yields a higher gain than the former. Another commendation for AMD engineers!
Bearing in mind that the only difference between the two is the extra L2 cache, can this be an indicator that we have already reached the point at which extra cache doesn't help Athlon64 performance? It seems to make sense, since AMD has been pushing for its transition from Clawhammer to Newcastle- I guess they realized from fairly early on that extra MHz would improve performance more significantly than extra cache. Along similar lines, compare the performance deltas between 3800/4000 (extra cache, same speed) & 4000/FX55 (same cache, extra speed); in the majority of benchmarks, the latter jump yields a higher gain than the former. Another commendation for AMD engineers!
[Posted by: mavalpha
| Date: 10/19/04 12:29:24 AM]
| Date: 10/19/04 12:29:24 AM]


