The Conroe865PE thread... for AGP users... stock and mod

[b1gt1m]

Hey guys! How yer all doin! Sorry it's been a while, 10 months? :O Well we all gotta work so yer know how it goes huh? Yep so it's nice to be back...

 

[Deleted member 24505]

Hey guys! How yer all doin! Sorry it's been a while, 10 months? :O Well we all gotta work so yer know how it goes huh? Yep so it's nice to be back...

WB

 

[b1gt1m]

Attempting alternative route to overclocking...

Yep sooo, I'm back on here and I'm still rockin the spec - please see the sceenshot - anyway, I got talking to a friend of mine a few nights ago at a party, we we're discussing the dark art of overclocking - and I'm talking back in the days of 486's! He gave me an idea... considering my CPU is unlocked by standard, I can just drop the FSB to some thing low, then raise the multiplier and so on, while having the pci bus locked down. this is doable thanx to the 5/4 and 1/1 bios mods I posted on here about 8 months ago ish... well here we go!!!

 

[b1gt1m]

Oh, I'm running Windows 7 Ultimate 32bit edition.

 

[rtwjunkie]

Welcome back, bigt1m!

 

[Jstn7477]

The new brain for my 4CoreDual-SATAII arrived today. We'll see how much it overclocks and if I can easily conductive pen mod it like my X6800. It's an SLACP QX6800 (G0, not the old SL9UK B3) so hopefully it overclocks well, though the board isn't exactly beefy. It also has "leet" VID apparently (1.337v) and the real voltage bounces around everywhere in the 1.2v range.

 

[TRWOV]

My Conroe865PE isn't 100% stable with the QX6800(G0) @ 3.19Ghz. It passes Geekbench, Cinebench and a bunch others but crashes on IBT. I can attest that's not an IBT bug as it crashed twice when I crunched on it but then again two crashes over a 2.5 month period of 24/7 crunching is very good IMO.

Of course, not two boards are the same (I recall Captain Harlock OCing a QX6700 to 3.24Ghz) but 3.2Ghz is very close to the VRMs limits. The highest IBT stable OC I get is 3.1Ghz (281x11) and I prefer it that way for the increased memory bandwidth.

I don't know why but the 5:4 bios is very unstable on my board.

 

[Jstn7477]

My Conroe865PE isn't 100% stable with the QX6800(G0) @ 3.19Ghz. It passes Geekbench, Cinebench and a bunch others but crashes on IBT. I can attest that's not an IBT bug as it crashed twice when I crunched on it but then again two crashes over a 2.5 month period of 24/7 crunching is very good IMO.

Of course, not two boards are the same (I recall Captain Harlock OCing a QX6700 to 3.24Ghz) but 3.2Ghz is very close to the VRMs limits. The highest IBT stable OC I get is 3.1Ghz (281x11) and I prefer it that way for the increased memory bandwidth.

I don't know why but the 5:4 bios is very unstable on my board.

You're right, the Kentsfield completely sacks the VRM on these boards. I had to back down to 260*12 because the machine got a 124 BSOD overnight. The VID is 1.337 and under load the measured voltage drops from 1.29v to 1.126-1.232v which is very low. I can vmod to 1.412v but I bet the VRM section is already overwhelmed, especially without heatsinks. I am using an Zalman Fatal1ty CPU cooler that should be running air over the VRM but probably not enough.

I wish there were some FSB 1066 Yorkfields but I really shouldn't put any more money into such an old system. The QX6800 set me back $100 but I can use it in two other boards after this one fails.

 

[b1gt1m]

Well, see the picture for my results, I just can't seem to get past x11 on the multiplier... it's rock-steady at x10!? Weird... yer thoughts please... Allo TRWOV how yer doin!

 

[b1gt1m]

Oh Sh*t!

OK, I seem to have a problem here, I've just been running LinX to test my processor, and after 2 mins 32 secs it stopped and reported an error and I'm running at 266*11 -RAM 3/2 @ 177Mhz!!? That should not be happening!!! Seriously worried here...

 

[b1gt1m]

This is the current set up...

 

[b1gt1m]

Also, this seems to show that my CPU is running a bit too warm? My HSF is a TUNIQ Tower 120!!!

 

[TRWOV]

QX6700 is a B3 chip with higher leakage (130w TDP). If your VRMs aren't sinked I guess it would overwhelm them easily.

Does it crash at stock settings? Temps seem to be about the same I had with air cooling (using a Sanyo AIO WC ATM).

 

[Jstn7477]

I'd love to get one of these and replace the stock VRM on my board for fun.

http://www.techpowerup.com/159347/evga-offers-overclockers-the-100-epower-board.html

 

[b1gt1m]

Hi guys, good question TRWOV - I'll just test it now. :wink:

 

[b1gt1m]

Well guys, this is most curious! My settings are 266*10 which is stock settings - RAM is at 3/2 Divider 177Mhz @ 2.5 3 2 5 - Stock settings, with Vdimm set to AUTO!!! I'm really con fused, see pic! Help? :shadedshu

 

[TRWOV]

I'd love to get one of these and replace the stock VRM on my board for fun.

http://tpucdn.com/img/12-01-25/evga_epower_board_02_thm.jpg

http://www.techpowerup.com/159347/evga-offers-overclockers-the-100-epower-board.html

That actually looks like a fun project. I'm using a Silverstone SFX 450w PSU because of the size of the Sanyo AIO cooler so I think I could tuck it under the PSU easily. Now I must get another board to play with

If I can't find one for sale I guess could replace the VRMs, couldn't I?

Well guys, this is most curious! My settings are 266*10 which is stock settings - RAM is at 3/2 Divider 177Mhz @ 2.5 3 2 5 - Stock settings, with Vdimm set to AUTO!!! I'm really con fused, see pic! Help? :shadedshu

mmm, seems to do fine at stock settings. I suppose your board can't handle an overclocked B3. Try getting a QX6800 G0 (SLACP 110w).

 

[b1gt1m]

Omfg!!!

TRWOV!!! I took yer advice, I've just found myself a QX6800, but yer not gonna believe this... it's a *genuine* Engineering Sample!!!! It's designated as 'QXWX' for... wait for it... $25!!!!! I can't believe my luck!! I've heard stories that if I email intel, they'll send out a better cpu... I should take delivery over the next 3 days... will keep you posted... check this out!

http://www.intel.com/support/processors/sb/CS-030747.htm

 

[TRWOV]

The ES chips are also B3 but it should at least run at 2.92Ghz without issue. The G0 chips were released around the same time as the QX6850 (not compatible with the Conroe865PE so don't bother getting one).

 

[b1gt1m]

Hmmmmmmm maybe, maybe not... I know the QX6850 is incompatible with the Conroe865PE as it runs at 1333 where as the QX6800 runs at 1066. Still how great is this - a QX6800 ES for next to nothing!!!! Awesome!!!!!

 

[b1gt1m]

OMFG!!? Can't believe my luck!!!

TRWOV!!!! Just installed the CPU an booted up the system, take a close look at the screen shot and you're thoughts would be appreciated, I wish you could see the size of the smile on my face!!!!

 

[TRWOV]

WTFITF? A G0 ES? and for $25? You hit the jackpot with that one. You could sell your QX6700 and even get a little profit.

Most I've got with my board is x12 and not 100% stable. How much can you raise it?

 

[Jstn7477]

I had to back down my QX6800 to 3.06GHz as the system locked up, had to be power cycled and somehow lost the date/time but retained every other setting. This processor just obliterates the weak VRM on this board and it drops down to 1.200v under load when the VID is 1.337v.

 

[TRWOV]

Pretty much where mine is. No wonder it doesn't even show up on the CPU support list for the Conroe865PE.

 

[b1gt1m]

Yea I know, awesome!! Soooo anyway, I've been leaving the CPU alone, and trying the BIOS's with the 5/4 and 1/1 dividers, I couldn't for the life of me get the system to boot!! So after a bit of thinking and a coffee, I figured I'll use the 1.72 BIOS, drop the multiplier to x10 and raise the FSB to 300 - that way the CPU should run at 2.99 instead of 2.93 and the RAM at 200 instead of 177 at 3/2 and all at low voltages. Well guess what, it worked like a charm! The thing that's baking my noodle though is: I use OCZ Platinum DDR RAM, so if it boots at DDR427 - it shouldn't be a problem...