PDP Systems Patriot XBLK PC3200 Database

[komodo]

A64 0.60 beta

 

Bandwidth

 

Bandwidth

[Super Nade]

deleted

[magg]

For you guys running >2.8V on your PDP...active cooling?

 

If so, how are you doing it? I'm tempted to try one more time to push some power through these guys to see if I can get em stable higher, especially with what a few of you have shown for timings and speeds @ 2.9V.

 

From what I've tried so far, mine still don't like anything higher than 2.7V :

[chennhui]

magg, I has active cooling on both DRAM and PWMIC. I mounted a 12v fan blowing downward on the mosfet and DRAM area. These fan ventilating fresh air into the casing and also redured the CPU temperature. This has significantly increase the stability.

[lmmaudio]

Up to 2.9 I don't think you need active cooling on them. I recently got an 80mm Revoltec Dark Blue on it, just in case... I'll try to push to 3.0 some time in the near future, but I don't think it will help my winnie mem controller lol.

[chennhui]

Some interesting finding on PDP XBL PC3200's voltage and timings:-

 

Memory Frequency, Timings, Voltage, Remarks

279, 3.0-4-4, 2.6v, Not stable

279, 3.0-4-4, 2.9v, Prime95 stable

279, 2.5-4-3, 2.9v, Not stable

279, 2.5-4-3, 2.6v, Prime95 stable

 

Results speak itself. Aggressive timings required lower voltage.

[Super Nade]

deleted

[WheresWaldo]

I am working on my backup system and finally have a good handle on these Patriot PC3200 XBLK (2x512). I am running a 3500+ Clawhammer 280x10 at 1.625v with RAM 2.5,4,3,7 1T @ 280 HTT @ 2.80v. It is memtest stable overnight. I do not have any HDD attached to the system and only have a DVD/RW connected as Primary on IDE1. All the memory settings are on auto or the default using 331-3 BIOS. I'm sure if I lower the CPU multiplier I could get more speed out of the RAM, but what for, since I already have the CPU at 2.8 Ghz.

[Super Nade]

deleted

[Big B]

Okay, here's my XBL PC3200 settings:

 

2x256MB

 

HTT: 260MHz

Multi: 9x

LDT: 3x

 

DRAM stuff:

1:1

1T

 

2.5-3-3-7 (have to check the rest)

2.8V

 

Now, I know this stuff will go up to 285MHz with some slight timing adjustments. The bad news is that my CPU doesn't want to go past it's current speed. It seems that anything past a 260MHz HTT doesn't fly, even if I drop the multi. I still need to look around. Can't gripe about a 30% OC without any sweat, though.

[WheresWaldo]

...

Now, I know this stuff will go up to 285MHz with some slight timing adjustments. The bad news is that my CPU doesn't want to go past it's current speed. It seems that anything past a 260MHz HTT doesn't fly, even if I drop the multi. I still need to look around. Can't gripe about a 30% OC without any sweat, though.

BB

I was in the same situation until I remembered that I was only at 1.550v CPU. I was stable at 270, memtest at least, but at 275 I had 96 errors in eight hours. I upped the DRAM voltage to 2.9v and had over 8000 errors in just a few hours. So I backed down the DRAM to 2.8v and slowly upped the CPU*VID so now I'm at 1.65v and everything is good to 280. I am testing 282 now. Since I still have no drives for this PC I have more patience to try to take the FSB up in very small increments

 

Just to be clear this is not the system in my sig

[WheresWaldo]

282 was a no-go in memtest, a lot of errors on the 2nd pass. Tried 281 and same errors on 2nd pass. Saved settings in CMOS Reloaded and went back to stock settings, flashed with 414-3 and reloaded CMOS settings and lowered the errors at 281 to 2 errors in pass 4 on test #5. Will do a little more testing. Will start tweaking settings. If I can get 2850 stable I will usually drop the 50Mhz and run 24x7 at 2.8Ghz. That's what I did with the rig in sig. It will run stable at 2650Mhz (memtest, prime95, superPi, OCCT all pass) so I drop the 50Mhz (50Mhz is really insignificant at these speeds, but it makes me feel better) and run 2.6Ghz 24x7.