To: peter_luc who wrote (27480 ) 2/5/2001 8:14:28 PM From: Dan3 Read Replies (2) | Respond to of 275872 Re: Scumbria is also one of the very big heavywights Yes, if not THE heavyweight. Scumbria didn't fold even after Paul DeMone started making veiled hints that he had inside info from real results and implied Scumbria was making a fool of himself for doubting P4's performance. The exchanges, many of them fascinating and quite educational went on and on (as you well know - but others on this thread may not visit Ace's message board). When the real numbers came out DeMone never admitted that he'd miscalled it - nor would he admit that Scumbria had been absolutely spot on all along. I think that Paul (DeMone, not you-know-who) still has hopes that some new compiler will come out and someone will release software using it that vindicates his conclusions. But so far, Scumbria and Hans DeVries have been the shining lights of correctly estimating processor performance from processor architectural diagrams. Hans' site at chip-architect.com is a great one, it would be great if he could find time to add to it. Paul Hsieh should probably be listed in there too - he was skeptical of P4 as well and has posted great analyses atazillionmonkeys.com Paul DeMone is a great analyst too, but P4's wonderfully convoluted architecture, with its compellingly beautiful complexity, seems to have misled him a bit. I'm sure I've left out many others equally deserving, but these 4 have been the ones who seem to have put the most effort into showing the rest of us what was up with the new architectures. Regards, Dan PS - my English major wife for some reason delights in pronouncing misled as miiz'uld, I don't quite know why :-)