<i>Full update on thread:
x86-secret has published <A HREF="http://babelfish.altavista.com/babelfish/trurl_pagecontent?lp=fr_en&trurl=http%3a%2f%2fwww.x86-secret.com%2f%3foption%3dnewsd%26nid%3d808" target="_new">extensive benchmarks</A> for a 2.55Ghz Dothan. If their results are to be believed, it is quite simply astonishing! The 2.55Ghz Dothan is sometimes even faster than an A64 4000+, and definitely faster than P4 solutions. If taken to 2.8Ghz, dothan makes all current other processors look bad! </i>
Also, AOpen seems to be so proud as to put the i855-based <A HREF="http://global.aopen.com.tw/products/mb/" target="_new">motherboard</A> right in the middle of their products main page. A quick look at the <A HREF="http://global.aopen.com.tw/products/mb/i855GMEm-LFS.htm" target="_new">specs</A> for that motherboard shows that it's quite underpowered in terms of memory - it still uses DDR333 and only AGP 4x slot, not 8x... Imagine if Dothan were to sit on a more heavy-duty motherboard instead of this cute tiny mobo.
Also, too bad we don't quite know how much wattage a 2.6 needs (and therefore, how much heat it generates). All we know is that a 2.1Ghz dothan puts out 21W. If a 2.6Ghz didn't put out much more than that... well, you all know what I'm going to say. It's fairly obvious that Toledo will initially clock at less than 2.6Ghz.
In any case, it is fairly obvious that Intel is holding the gun on dothan. I'm thinking that the only reason for that is that dothan could become stronger than P4s! I mean, a 2.6Ghz dothan on air with dual channel memory and enhanced FSB (533, 800, whatever) would have performance enough to make a serious stand against 3.2, 3.4 and maybe even 3.6Ghz prescotts!
Also, the whole thing suggests that yonah might manage clock speeds in excess of 3Ghz with air cooling. (taped out several weeks ago, ready to roll in 2005... but no, not for desktops... :frown: ) Heck, current dothans can do 2.6Ghz on air and yonah <b>is 65nm-based</b>, which should improve clock speeds further! As far as we know, it's not completely impossible that smithfield is actually yonah-based and clocks at 2.8, 3.0 and 3.2Ghz... If that were the case, I'd gladly have them delay smithfield for another 3 months to the very late 2005 than get an impossible netburst-dual-core processor.<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 11/02/04 11:44 PM.</EM></FONT></P>
x86-secret has published <A HREF="http://babelfish.altavista.com/babelfish/trurl_pagecontent?lp=fr_en&trurl=http%3a%2f%2fwww.x86-secret.com%2f%3foption%3dnewsd%26nid%3d808" target="_new">extensive benchmarks</A> for a 2.55Ghz Dothan. If their results are to be believed, it is quite simply astonishing! The 2.55Ghz Dothan is sometimes even faster than an A64 4000+, and definitely faster than P4 solutions. If taken to 2.8Ghz, dothan makes all current other processors look bad! </i>
Also, AOpen seems to be so proud as to put the i855-based <A HREF="http://global.aopen.com.tw/products/mb/" target="_new">motherboard</A> right in the middle of their products main page. A quick look at the <A HREF="http://global.aopen.com.tw/products/mb/i855GMEm-LFS.htm" target="_new">specs</A> for that motherboard shows that it's quite underpowered in terms of memory - it still uses DDR333 and only AGP 4x slot, not 8x... Imagine if Dothan were to sit on a more heavy-duty motherboard instead of this cute tiny mobo.
Also, too bad we don't quite know how much wattage a 2.6 needs (and therefore, how much heat it generates). All we know is that a 2.1Ghz dothan puts out 21W. If a 2.6Ghz didn't put out much more than that... well, you all know what I'm going to say. It's fairly obvious that Toledo will initially clock at less than 2.6Ghz.
In any case, it is fairly obvious that Intel is holding the gun on dothan. I'm thinking that the only reason for that is that dothan could become stronger than P4s! I mean, a 2.6Ghz dothan on air with dual channel memory and enhanced FSB (533, 800, whatever) would have performance enough to make a serious stand against 3.2, 3.4 and maybe even 3.6Ghz prescotts!
Also, the whole thing suggests that yonah might manage clock speeds in excess of 3Ghz with air cooling. (taped out several weeks ago, ready to roll in 2005... but no, not for desktops... :frown: ) Heck, current dothans can do 2.6Ghz on air and yonah <b>is 65nm-based</b>, which should improve clock speeds further! As far as we know, it's not completely impossible that smithfield is actually yonah-based and clocks at 2.8, 3.0 and 3.2Ghz... If that were the case, I'd gladly have them delay smithfield for another 3 months to the very late 2005 than get an impossible netburst-dual-core processor.<P ID="edit"><FONT SIZE=-1><EM>Edited by Mephistopheles on 11/02/04 11:44 PM.</EM></FONT></P>