Andy - Re: Deschutes, NoteBook PCs and Heat.
First, the fact that Deschutes chips are out in the field must tell you something about the status of the chip! It exists and apparently is being sampled.
I don't have a verifiable power dissipation number for the Deschutes, so let's derive one from the Pentium II - it's direct predecessor.
The Pentium II (0.35 microns) dissipates a worst case 36 watts (266 to 300 MHz) at 2.8 volts. The "standard" voltage for Intel's 0.25 micron technology, which includes the Deschutes & Tillamook, is 1.8 volts.
Power dissipation is proportional to the square of the operating voltage. Thus, the RATIO of power between the Deschutes and the Pentium II should be (1.8^2)/(2.8^2) = 0.41.
This should result in .41 x 36 = 14.9 watts for a Deschutes chip running at 266 to 300 MHz at 1.8 volts. This IS a lot of power! For reference, the Tillamook dissipates about 4 watts at 233 MHz.
Now, the Deschutes is capable of running at 450 Mhz at 1.4 to 2.2 volts. Since Intel has historically run its mobile CPUS at a lower supply voltage (and lower speed) than the DESKTOP versions, it would be very likely that the Deschutes could run at 300 MHz with supply voltages well below 1.8 volts., such as 1.4 volts as mentioned in Intel's upcoming ISSCC paper.
Using this scenario, the Deschutes should dissipate (1.4^2)/(2.8^2)*36 = 9 watts.
This 9 watts is comparable to what the Pentium MMX (0.35 micron) dissipates at 166 MHz, and this is used ubiquitously in notebooks (and IT RUNS HOT - I HAVE ONE!).
The problem, however, with Deschutes in notebooks is that these high end notebooks will be stuffed with LARGE displays (13" and 14") and other goodies, thereby INCREASING overall power dissipation of the system. Battery life then suffers proportionately.
Intel will probably be offering a 266 MHz Pentium MMX Tillamook (0.25 micron) as a more realistic approach for high end notebooks, until the overall power dissipation/battery life issues with the Deschutes are resolved.
Paul |
