Got bored

[PhilDoc]

Was getting a little bored waiting for the "old" DFI to arrive so I decided to have a little fun. I had a couple of Delta 190cfm fans that I used on a Baker custom Cpu cooler and decided to strap them on the Ultra-120 X. A little supprised, but one worked better than two in a push/pull. Brought my idle temps down to 34C. I was using 1.65v despite what cpu-z says. could have had a little higher but the snipping tool for vista takes so long to save. 59 db was getting to me anyway.

[Gothic]

Incredible chip you got there!

3.3GHz on air! It should be one of those 0704 or higher demons!

Whats the max for SPI 1M?

PS.: Now I noticed... 1.375v for 3.3GHz on air??????

[PhilDoc]

Incredible chip you got there!

3.3GHz on air! It should be one of those 0704 or higher demons!

Whats the max for SPI 1M?

PS.: Now I noticed... 1.375v for 3.3GHz on air??????


Thanks bud, It was really 1.65v. Dont know why cpu-z reports it wrong. Don't know SPI 1M, I'll download it a give it try later this evening. Should be pretty close to the 3.3G though.

[Gothic]

Any info on the steppings, batch, etc etc?

Thx!

[Boite de P4]

Whoww Impresive !
You must try with a DOD :D

GG PhilDoc :up:

[knopflerbruce]

C'mon give us some stepping info:D

[PhilDoc]

Any info on the steppings, batch, etc etc?

Thx!

Stepping is LCB9E 0630BPMW. Dont know where the find the batch. Just let me know what info you want and I'll get it for you when I break it down.

[PhilDoc]

Whoww Impresive !
You must try with a DOD :D

GG PhilDoc :up:
Thanks, sorry, don't know what DOD is, lol, but I'll probably break out the LS or Baker and give phase change a try. I'll need a new mount kit for the LS, maybe in a couple of weeks. The baker probably runs a little colder, but the block was too large to run without the IHS in the past. Maybe it'll fit better on the DFI.

[Gothic]

You can find the revision after de stepping... like OSA185AA6CD[XX]

Where XX is the revision... as she's naked, it easy to find it out... :D

[PhilDoc]

You can find the revision after de stepping... like OSA185AA6CD[XX]

Where XX is the revision... as she's naked, it easy to find it out... :D

I'll get the info for ya later this evening. I haven't removed the IHS yet, but if I put her on Phase Change, I'll probably take it off.

[Trunks]

That's a nice cpu you got there :)

[lee34]

good cpu.

[PhilDoc]

Thanks guys. Anyway, decided to have more fun so I turn the AC on in my computer room and dropped the ambiet to 64F which dropped my idle temps to 31C. Thought it would give me more out the 1M SPI, but this was the best at 1.65v. Almost had 3.296G but always crapped out just at the end. The time is poor because the memory and HTT was set low to isolate the cpu

[PhilDoc]

I was able to get this screen however at 1.68V. Besides the better temps the other main reason for the improvement was loading up Corel Capture, which takes a very quick screen shot. The vdimm was 2.6V, but I doubt that made any difference.

[Gothic]

Impressive!!! Keep pushing it!

Do you think a 3.3GHz 1M PI is possible, PhilDoc?

Wouldn't these PI times get some seconds lower under XP? Just my opinion...

[PhilDoc]

Impressive!!! Keep pushing it!

Do you think a 3.3GHz 1M PI is possible, PhilDoc?

Wouldn't these PI times get some seconds lower under XP? Just my opinion...

Thanks my friend, Don't know about 3.3G 1M SPI. Maybe when I get the new board and I can give it a little more vcore or it gets colder around here.

You're probably right about XP, don't know anything about SPI times and Vista. I think with the right settings 2003 use to be the best for SPI, but I dont know anymore.

Did have a question about the memory I have. I know that the bh-5 and the Micron I have will take some volts, had them both above 3.3v, but not sure how high I can go with the infineon. Its the same chips that Mushkin puts in their pc4000 redline. 2.9v is what I've been able to find but don't know what the max is. This is the best so far doing SPI 1M. Nothing but heat spreder cooling them, but don't really know if that makes any difference.

[Gothic]

BH-5 is that old story... toast it till it doesn't get benefit from it... I've seen ppl runing them up to 4.2v.... lol

Micron is a different story... It depends on wich revision you runing... I remeber Micron -5B D (Crucial Ballistix Z503) liked no more than 2.7v 24/7, and still gave some nice clocks... like 280~305MHz @ 3-3-3-7-1T... but there's stability and MHz gains even @ 3.5v... but they die with such a high voltage...

Infineon chips BE/CE-5/6 would hit their 24/7 stability @ 2.6~2.7v.. and would hardly benefit from anything over 2.9v....

Well... hope I didn't confuse you... lol

[PhilDoc]

Nope, not at all. The micron I have in my old pair of OCZ are the one's you mentioned. I think I got over 250 2225 at 3.3v, but never ran them long at that voltage.

I was one of the ones you ran my bh-5 at over 4 volts, lol. Not very long. Just benching. Usually, 3.3v 24/7. Had a pair that did 274 2225, but lent them out and never got them back.

I believe I have the CE-5 Infineon. So I'll wont take them higher than 2.9v probably leave them at 2.8v. OCZ still waranties them for life at 2.8.

It all sounds good, I can live with what I've got. Thanks again for your help :toast:

[Gothic]

Any news? Has the DFI already arrived?

Looking foward for results from this 939 beast!

[PhilDoc]

Actually, kinda stuck atm. I had the same problem with dividers that you had. I didn't like 166. Strange I could do them fine before. Anyway, after some frustration with that. I ran it on the stock cooler and was able to run SPI at 3.207G, best with a 86 cfm fan and the Ultra-120 was 3.187G. I'm thinking, I'm golden.

So I stick on the Ultra and the best I can do SPI is 3.233G. Even with a 190 cfm Delta that's all I can run. Changing the voltage does very little. Its strange. Tried changing ram, using multipliers. Nothing past that. I can do a screen at over 3.3G put no PI.

Maybe its something in the bios or the version I'm running. Anyway, have to run for a bit. I'll try a few bios setting and see what happens. Sorry I haven't looked at your threads, I'll take a look when I get back.

Too frustrated for words atm, lol

[PhilDoc]

Sorry for taking so long. Any, I guess I should read a little before I get all frustrated. Seem you have to boot in around 250 fsb to get over 250 fsb with clockgen. Even with that it wasn't much more than the MSI. It would only let me get up to 1.62v and I could only get the ambient down to 68F instead of 64F like I had the other night. Well at least, I now have 1T and dual channel. Anyway, here's your SPI at 3.3G. Tried to get a little more but it was a no go. Maybe when it gets a little cooler here. It was over 100 degrees outside

Sorry no cpu-z and I hid the fsb. I was using a multi of 12 and a divider of 180. The vcore was 1.62v

BTW, I changed bioses just to see if I could do better. No better OC, but the divider that was unstable changed. You may want to try a different bios with your situation.

[Gothic]

Now that was A NICE improvement! Didn't realize you were runing 2T.... that would explain the high times.... but I still think that Vista is stealing you some seconds... lol

I surelly have to try that 329 BIOS.... did the 166 divider worked for you?

ps.: Nice OC and time!

[Gothic]

Oh!

BTW if you need some MHz goals, the actual WR is 3807.86MHz for Opty 185....

http://valid.x86-secret.com/show_oc.php?id=186341

Good luck!

[PhilDoc]

Now that was A NICE improvement! Didn't realize you were runing 2T.... that would explain the high times.... but I still think that Vista is stealing you some seconds... lol

I surelly have to try that 329 BIOS.... did the 166 divider worked for you?

ps.: Nice OC and time!


Thanks, the 166 divider works for me in the 329 bios, but couldn't use it in the 406 bios. Had to use the 150 divider. The 150 divider wont run in the 329 bios, lol. I suspect they're tuned for different memory types.

3807, wow, that's up there, but doable if I could get this thing cold. Well, there's an answer to every problem. Hopefully in a couple of weeks that will fall, lol

[PhilDoc]

Still bored lol, so I decided to see how the memory oc. The best I did was 282, but here's a pi running 1:1 on the memory 275 X 12. 3/10 of a second quicker. Thought I would have gotten higher but by the time I got around to running pi it was 10F warmer outside. While it was cooler I managed to gain 60 mhz on the suicide shot. 1.68 vcore using cool air. I sure hope I can get the phase change working on this.

[Gothic]

Looking good!

BTW just got 2x 120mm nasty 160CFM FANS! Will be testing those tonight!

Ps.: You should really get some P-ATA IDE and install w2k3 on it!

[PhilDoc]

Looking good!

BTW just got 2x 120mm nasty 160CFM FANS! Will be testing those tonight!

Ps.: You should really get some P-ATA IDE and install w2k3 on it!

There you go, that should help. Let me say, they're loud, lol. I put my delta 150's in for that run. I would have used the 190s, but I can run the 150s on my fan controller. That way I don't have to remount the fans, I can just turn them down, lol. 53db X2 is still plenty loud.

Still use the ice if you can and if you can get some cold air blowing on your 160s it'll help.

I've got both XP and 2003 and a 160G sata on the system along with my raid setup. That was one of the problems I had before. It seems I had the raptor raid on channel 3 and 4, with the 160G on channel 1. It didn't like that. Once I put the raptor on channels 1 and 2, with the 160G on channel 4, everything ran fine.

Anyway, I'll try to load either XP or 2003 on the 160G and see what happens.

Good luck tonight. :up:

[PhilDoc]

I guess you're right about needing a PATA. Tried to load both XP and 2003 and they both crapped out, lol. Must be something to do with having both on the same controller. Anyway, I'll pic up a cheap PATA next week and give it a try.