Geil Ultra Platinum PC4200 rules

Joined
Apr 2, 2003
Messages
747
I've been running my 2.4C at 3.0 (250fsb/5:4 divider) since I bought it; using Kingston HyperX PC3500. I recently swapped it out with the Geil PC4200 and am able to run at 3.3ghz (275fsb/1:1) so far. I tried 3.4ghz but it crashed during prime95. Hopefully I'll be able to go further after the ram burns in.
 
I think I'm processor limited. I'm running prime now at 300 fsb 5:4. It crashed at 1.6 vcore so I'm trying 1.675. What is the lowest vcore documented in a case of SNDS?
 
i think (note THINK) that it doesn't happen intil 1.75 or so on air cooling, and 1.8 or 1.85 with water. what ram voltage are you running at? o yeah, what timings?
 
i think anything under 1.7volts is safe for a p4, as long as ur mobo doesnt overvolt (ie: asus mobos overvolt when idle, undervolt when under load)
 
DRAM is 2.8v (MAX on IC7-G) and the timings are 3-4-4-8.

Unfortunately my PC crapped out after a couple minutes at 3.6 even with the extra voltage. I may take it up more to see if it will stick though.
 
Originally posted by skinlab
i think anything under 1.7volts is safe for a p4, as long as ur mobo doesnt overvolt (ie: asus mobos overvolt when idle, undervolt when under load)

I agree!!!


INteresting side thought here cause I think I know the answer!!!


With 1:1 you have to run the ram at cas 3??? With 5:4 have you tried it at 3.3ghz, and can you run it at cas 2??? If you have not tried it, go ahead and do it. If you can run cas 2 you will be surprised that the performance is not likely that much better....

IN fact see if with a 5:4 or even a 3:2 if you can get it stable at higher fsb...If you can get it stable with a 50-100mhz lead of clock speed I bet the lower ratio, lower cas timing, higher clocked system is overall superior....


I have tested many combinations on my system and found this to be the case and therefore I choose not to waste my money on bloated priced ram above the pc3500-pc3700 level...
 
This ram was only $296 for the gig and is extremely fast. My kingston could only do 216 at 2-2-3-6 no matter the ratio. The kingston was $240 after tax.

Your post doesn't make a lot of sense. I assume you're saying tighter timings > higher fsb, which is not always the case. If you'd read my post you'd see that a lower ratio was not helping my overclock at all. I'll try increasing my voltage to get to 300fsb, but I'm a little scared of going over 1.7v.
 
300 fsb at 5:4 at 1.7v looks like it's going alright. I'm running prime now, and it's already lasted longer than before. Wish me luck.
 
Cool!!!.....If you need 1.7v you are reaching chips limit...When I get that high .025v often cannot garner more then 1fsb....


I can get my system st 1.6v actual to be stable at 3.5ghz but at 3.552ghz it needs 1.67v actual and at 3.564ghz it needs 1.7v actual....

For me all the extra volatge and heat which is more of an increase going 50-60mhz with .07v then my total load heat increase going from 2.4 to 3.4ghz at stock voltage!!!
 
300 fsb crapped out again. I guess I'll be happy with 3.3ghz at 275 1:1. It's more than most people have. It's definately my cpu that is limiting me, though I think my ram craps out somewhere around ~285 fsb 1:1.
 
OK now try this....

Can you get the 5:4 stable at 292fsb with 5:4 and cas2 or cas 2.5??? If so try it and bench it and I guaranee you it will be better then the 3.3ghz even with 550ddr cause ddr speed doesn't mean as much as raw cpu speed and cas 3 timings really blow!!! A 200mhz clock, heck even running a 100mhz clock with 5:4 will likely be better....

At 5:4 with 292fsb we would be looking at 433ddr still but faster timings no doubt....
 
This is frustrating. Everything I have tested over 3ghz crashes during prime or anything video card related. Before changing the ram, the only instability I had was that every few weeks the computer would hang with a small square of noise at the top left corner of the screen or where the mouse was. With the computer at 3.3/3.4/3.6 the computer works fine until I stress it, then it either just freezes, or reboots. I tried 3.6 at 5:4 with 1.75v and it still rebooted.

:x
 
Originally posted by rogue_jedi
i think (note THINK) that it doesn't happen intil 1.75 or so on air cooling, and 1.8 or 1.85 with water. what ram voltage are you running at? o yeah, what timings?

regardless of the cooling used itll still happon. its electron migration that causes sudden death not heat.
 
Back
Top