There's some new news floating about on AMD's latest Athlons (based on Palomino core).
Look at
http://www.theregister.co.uk/content/3/21938.html and
http://www.theregister.co.uk/content/3/21825.html There's some good and not so good information there. The good news is that they will be released on October 7th. The not so good is that max clockspeed has only been increased to 1.533 Ghz. This isn't quite as bad as it sounds, since the Palomino contains a number of core improvements to reduce heat and increase performance per clock. In general a Palomino at the same clockspeed as a Thunderbird will perform between 5 and 10% faster overall. Pricing will also be VERY competitive with Intel.
Now for what I don't like (from my point of view):
The processors will be named "Athlon XP" and called:
Model 1500 -> runs at 1333 Mhz true and = to about 1400 Tbird
Model 1600 -> 1400 Mhz
Model 1700 -> 1466 Mhz
Model 1800 -> 1533 Mhz
The reason for the "model" designation can only be to attempt to convince the general (and uneducated for the most part) public that AMD's Athlon is faster clock per clock than the Intel P4. This is 100% true. The deceiving part is that some people are going to assume that the model 1800 runs at 1.8Ghz and if they find out it
only runs at 1.533 Ghz they are going to be angry. In AMD's defense a Palomino core at 1.533 Ghz will pretty easily perform better than a 1.8 (or 2 Ghz for that matter) P4 in nearly every potential benchmark. What AMD has done is to even print "Model 1800" on the CPU and has told MB manufacturers to release bios updates to show on boot-up MODEL 1800. Only by using Sandra or CPUID will you be able to find out the true clockspeed. This I really don't like.
From a marketing standpoint AMD is playing their cards right IMO and following Intel's lead in naming CPU archetecture components with goofy, meaningless, names. AMD's new name for the Palomino's added features is "Quantispeed Technology" and I presume it refers to the hardware prefetch logic built into Palomino, along with full SSE SIMD support. (Very similar to Intel's "Netburst" archetecture, though admittedly MUCH less deceivingly named Intel's use of "Netburst" and related press releases amount to false advertising IMO. )
AMD is also capitalizing on the release of Windows XP (xp means eXPerience btw
) by naming their new Athlon design similarly. This is smart, and helps to differentiate that the new Athlon XP is in fact improved in other than just clockspeed.
On the other hand AMD is playing it safe in only releasing up to 1.533 Ghz "Model 1800". They
probably won't be able to go much higher than 1.8 Ghz on their current manufacturing process until early next year. The last thing they need to do is to threaten Intel into releasing Northwood at 2.2Ghz or 2.4 Ghz now.
Other news from AMD is that the specs for the new design "Sledgehammer" and "Clawhammer" cores will be released on around the 10th of October. There is also good news in that AMDs SOI .13 Micron process is currently sampling products. This will enable significantly higher clockspeeds than the current .18 micron process. AMD looks to be in line for a 6 month wait for initial production on this process. By this time Intel will have long transistioned production to a .13 micron process themselves, though I'm unaware of SOI production capabilities from Intel at this point. Maybe AKDejaVu knows something here that I don't...
I'm really curious to hear some feedback from him on the confirmation that the "model" designations will in fact be used.
