• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

ncch-dl not hitting 200 mhz fsb - any ideas?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

mbentley

Gloriously Lead, Overclockix Chief Architect
Joined
Sep 26, 2002
Location
Indianapolis, IN
okay, here is my setup:

asus ncch-dl
2 x intel xeon 2.4 c1's
2 x corsair pc3200 @ 2-2-2-5
2 x mcx603v's cooling the xeons

i run fine when i am at 18 x 166, but when i run my multi at 15 and try to boost the fsb via clockgen to 200, i start to get prime95 errors around 180 - 185. i have my memory set to 266 so it should run at full speed when at 200 mhz fsb... any ideas?
 
well, i actually set the jumper to 200 mhz this time and it worked... i guess clockgen will only take you so far... hehe

now when i boot every couple of times i get a bios rom checksum error... any thoughts?
 
Last edited:
doesn't the ncch-dl boot up at max multiplier X FSB until it reads the BIOS? (so if you got 533 chips it's booting at 3.6ghz, and 400mhz chips would be 4.8ghz)
If so you might need to do the volt mod.

Also make sure RAM voltage is maxed (2.8 on your board i think) and try loose timings as well.
 
i went ahead and flashed the bios to the latest 1003 bios from the asus website and cleared the cmos before and after... that seemed to make it so the system would boot fine every time... who knows...
 
mbentley said:

i run fine when i am at 18 x 166, but when i run my multi at 15 and try to boost the fsb via clockgen to 200, i start to get prime95 errors around 180 - 185.


i think i read over at 2cpu.com that if the fsb jumper is set to 166 then the pci bus lock doesn't work...but if the fsb jumper is set to 165, then the pci lock works...

therefore, if you set the fsb jumper to 166 and then increase the fsb with clockgen in windows, your pci bus could be causing problems...

also, you might have to increase the vcore using the socket pin mod method for 1.6v to be stable at 3+ghz (especially with C1 chips)....

particularly to boot at 200mhz fsb...since, intially that will be 3.6ghz for your 2.4ghz 533mhz chips with 18x bootstrap multiplier...then as long as you've changed the multiplier to 15x previously, once it gets past the intial post it will boot at 3ghz instead of 3.6ghz...in any case, you will likely need 1.6v vcore
 
dustybyrd said:
i think i read over at 2cpu.com that if the fsb jumper is set to 166 then the pci bus lock doesn't work...but if the fsb jumper is set to 165, then the pci lock works...

therefore, if you set the fsb jumper to 166 and then increase the fsb with clockgen in windows, your pci bus could be causing problems...

also, you might have to increase the vcore using the socket pin mod method for 1.6v to be stable at 3+ghz (especially with C1 chips)....

particularly to boot at 200mhz fsb...since, intially that will be 3.6ghz for your 2.4ghz 533mhz chips with 18x bootstrap multiplier...then as long as you've changed the multiplier to 15x previously, once it gets past the intial post it will boot at 3ghz instead of 3.6ghz...in any case, you will likely need 1.6v vcore
thank you for your informative post. that makes quite a bit of sense. i am still not sure why clockgen couldn't get me to 200 mhz... i was actually using the 165 in the bios and then using clockgen to change it to 166 initially on booting up... oh well... i am at 200 mhz and haven't had any crashes or anything like that...
 
well i've done a 1.6v mod with the wires (confirmed to work in the bios) and now i am having some stability issues. i figured it would cause the opposite effect.

right now i am running the chipset at 1.5v (stock) and the memory at 2.6v (stock). i don't know if this might be a northbridge cooling issue or what. i know the chips can run ~3.0 at 18*165 rock stable so that is one reason i might think it could be the northbridge with the sink not making good contact or something...

what should i try now?
 
well take off the northbridge heatsink and apply some thermal epoxy or non conductive paste if it has a retention mechanism.

bump RAM voltage to 2.8 cuz that shouldn't be close to too much voltage....what were your temps at your max OC (appears 3.0) and what are your temps now at 3.0?
 
ajrettke said:
well take off the northbridge heatsink and apply some thermal epoxy or non conductive paste if it has a retention mechanism.

bump RAM voltage to 2.8 cuz that shouldn't be close to too much voltage....what were your temps at your max OC (appears 3.0) and what are your temps now at 3.0?

i just removed the northbridge, cleaned it and the heatsink and applied some as5. as for the temps with this board, i can't be certain as i haven't found any program (including mbm) that will find sensors on this thing... i bumped ram voltage up to 2.8.

right now i am having issues even getting into windows... gah
 
Last edited:
mbentley said:
fortron 550w eps model FSP550-60PLN


18Amps on the CPU rail

12*18=216 watts available

assuming 80% efficency, that's 172 watts available to the CPUs, assuming that no other power is drawn from the CPU rail

172/2=86 watts

I'm pretty sure a xeon at 1.6v will pull more than 86 watts. check the CPU rail with a multimeter. Make sure you're not checking the wrong rail.
 
mbentley said:
fortron 550w eps model FSP550-60PLN


I knoiw another member had an OCZ that was 18a on the 12v rial and 18a on the second rail.IT WOULD NOT POWER THE RIG...I have no clue what these processors chew up for amperage but I am convinced you need as close to 27a available to the processors from the 12v rail..
 
greenman100 said:
18Amps on the CPU rail

12*18=216 watts available

assuming 80% efficency, that's 172 watts available to the CPUs, assuming that no other power is drawn from the CPU rail

172/2=86 watts

I'm pretty sure a xeon at 1.6v will pull more than 86 watts. check the CPU rail with a multimeter. Make sure you're not checking the wrong rail.
I'm running my Vcore modded 1.6LVs at 3.12 GHz in a PC-DL with a 6800 128MB AGP, 2 x 512MB PC3200 DDR, a SATA 80GB hard drive, DVD-ROM, and DVD Burner from an AMS Mercury EPS12V PS. The AMS has split +12V rails with 18A dedicated to the motherboard and CPUs. It runs rock stable, no problems. The BIOS reads the +12V potential at +12.20V. The other +12V rail has 15A available for drives.
 
mbentley said:
well i've done a 1.6v mod with the wires (confirmed to work in the bios) and now i am having some stability issues. i figured it would cause the opposite effect.
Wires? The 1.5V to 1.6V mod requires a single u-wire in each socket bridging D2-D3. By "wires" did you mean one in each socket? If you used more than one per socket, the mod wasn't done correctly.
 
DaveB said:
I'm running my Vcore modded 1.6LVs at 3.12 GHz in a PC-DL with a 6800 128MB AGP, 2 x 512MB PC3200 DDR, a SATA 80GB hard drive, DVD-ROM, and DVD Burner from an AMS Mercury EPS12V PS. The AMS has split +12V rails with 18A dedicated to the motherboard and CPUs. It runs rock stable, no problems. The BIOS reads the +12V potential at +12.20V. The other +12V rail has 15A available for drives.


the 6800, HDs, and opticals dont matter

all I was asking him to do was check the CPU line with a multimeter
 
ajrettke said:
well take off the northbridge heatsink and apply some thermal epoxy or non conductive paste if it has a retention mechanism.

mbentley said:
i just removed the northbridge, cleaned it and the heatsink and applied some as5.

I was just wondering if AS5 was non conductive
 
Back