Since I was having lots of success running MP3s all day (and Prime95 at night) I thought I'd bring down the voltage down to 2.5v. And it works flawlessly while the chip is running at 266Mhz! I plan to hook up a GPS receiver to the unit and an LCD screen somehow so it will be an MP3/GPS receiver (and hopefully DVD if I have room in my car)
If you are wondering I put a big socket370 heatsink/fan (with enough room to spare so I can close it with the case) and a P5 heatsink on the silver chip. Works great!
05-03-2000 15:37:51
RE:2.5v *stable* (modified 0 times)
Brian
How did you attach the heatsink to the silver chip? thermal epoxy? thermal glue?
have a bunch of sinks around and was just wondering what the best way to attach one. did you have any stability problems without a sink on the silver chip? mine feels cool to the touch during operation, not sure if a sink on it would be necessary or not. thanks for the input.
05-03-2000 16:03:40
RE:2.5v *stable* (modified 0 times)
Dr. Evil
I used double sided thermal tape, it looks blue on one side and white on the other, they can be had for $.50 or so at computers shows and other stores. Gonna try 2.4 and 2.3 tonight!
05-03-2000 16:09:34
RE:2.5v *stable* (modified 0 times)
Linuxguru
OK, I can confirm that 33x8 = 266 is *very* stable at 2.5 v. Both the CPU and the Cx5530 run cool as cucumbers for hours. How does one get to higher multipliers and/or clocks and is it stable?
05-04-2000 13:04:08
RE:2.5v *stable* (modified 0 times)
Linuxguru
BTW, mine is a stock websurfer with no mods to the CPU heatsink and no heatsink for the Cx5530. My guess is that the lower voltage makes a huge difference because the power consumption scales as V^2 * f for CMOS. So assuming that we don't hit the process capacitance limitations, for every 10% drop in voltage, we can crank the frequency up about 20% and still maintain identical thermal stability.
05-04-2000 13:11:30
RE:2.5v *stable* (modified 0 times)
Linuxguru
Actually, there's more: Since we're reducing the voltage swings, for a given slew rate at each node (limited by gate capacitance), there's less delay before the voltage gets to the switching threshold and therefore it will tolerate higher clocks at lower voltages.
05-04-2000 13:18:31
RE:2.5v *stable* (modified 0 times)
Dr. Evil
I will try 2.3v in a couple minutes and see if it's stable. Will post back to see if it's stable.
05-04-2000 14:45:36
RE:2.5v *stable* (modified 0 times)
Dr. Evil
Success! 2.3 volts working fine and dandy! Been running for 15min without any lockups. I let winamp play a couple MP3s in a loop and it's still goin!
05-04-2000 14:58:08
RE:2.5v *stable* (modified 0 times)
Dr. Evil
My goal is to get 2.0v working before I stick it into my car w/ my GPS!
I tried getting an IO, SOLD OUT EVERYWHERE! I was going to use the LCD screen =\, gonna buy a touchscreen instead. =)
i-opener opener laptop notebook computer help
drivers dll free windows dos repair fix linux mac macintosh 2000 95 98 nt pc
configure hardware software sound video netscape explorer network networking
lan wan software cmos fat bios printer card mouse modem ide scsi cd rom
controllers scanner tape hard drive cgi scripts source code mp3
