Ok I shoulda phrased that differently. AMD will use the steady pace for the following months, all the while continuing with Tbreds and possibly Barton. It won't be the most competing product, but it's normal as the Athlon is dying, and that spending any more R&D for more performance on a dying platform is really useless and loses money.
Next, Hammer is definitly gonna be a success, I just feel it. You may not beleive me, but trust me, this is AMD, not Intel with P4. They will always stay with the MHZ ISN'T EVERYTHING moto, and they gotta stick by it by providing us with extreme performance. I am more than sure the first ClawHammer will have a 20% increase of overall performance over the top of the line P4. Again I just think that way, but I also find it logical if AMD stated all this. Do not forget, it's not just 64 bit that drives it, it's the incredibly 32-bit it has. They've added a galore of improvements, a mem controller on-die, 512K L2 cache, Hyper Transport, possible SSE2 (if so, Intel is more screwed), 12 stage pipeline for better clock speeds, improved branch prediction, and tons more little tidbits. Overall it is estimated to have a 25-30% boost over current Athlons. Rests to see if that holds true, but even 20% is nice, since it's still better than P4s, and the ability to ramp up in clockspeed thanks to a 20% increase in pipeline and 0.13m/0.09m process, it should be able to compete P4s easily. Prescott is the real deal.
I also analyzed the PR a bit. Since 66MHZ Palomino equals 100MHZ Tbird by AMD's assumptions, then a 66MHZ Hammer should be about a 300MHZ Tbird. If that is true, then ramping up clock speed translates into very high PR ratings easily, and thus explaining why PR4400 comes right in the next quarter after ClawHammer's release. PR4400 is quite competitive, and I have yet to see a 4.4GHZ P4 NW B by 1rst Q of 2003. In that context, Prescott is what will make this competition more interesting.
There, was that enough ranting?
--
For the first time, Hookers are hooked on Phonics!!