rainbrodash666

Picked this up a bit ago and starting to get it ready to be revived to be used in a crysis crushing 2007/8 build. Hopefully I can use the original pcb so I can use my backup 3870x2 for quad crossfire instead of spare parts.

my friend just wants something that has an upgrade path, im thinking this isn't the right way for now. thanks, ive been on amd forever so the intel socket upgrade path is kinda foreign to me.

im looking on marketplace for a pc for a friend, this pc has a 7600k so its LGA 1151 and i was trying to see if the cpu could be upgraded but i cant id the motherboard and the seller isn't responding yet.

im building my "ultimate" crysis machine. and i need cpu coolers. this board uses AMD socket F. so the cooler rings mount with 2 bolts each (like socket 939 but im not 100% sure it is the same) so i have to find a cooler that i can modify or has those mounts, or one that uses the stock AMD clip mounting, I am open to doing air or custom water cooling if you caan find blocks that will work for under $100 for the pair. also im hoping to keep the build period correct to the late 2000's the build: asus l1n64-sli ws,2x AMD Athlon 64 x2 FX74, 4x2gb ocz reaper, 2x wd velociraptor 150gb in raid, ati HD3870x2 (2 of them if i can fix the other one) and windows vista ultimate or windows 7 ( haven't decided yet) attached a pic that shows mounting hardware.

yeah I would say to rma the cooler. when i had a pump failure it displayed similar symptoms with the one tube being significantly warmer due to little to no fluid flow. my cpu would go to 90c+ idle though

did you change jsut the usb boot option back or did you reset the bios to original options, might have switched to the cpu's igpu, or are you plugging into your graphics card or the motherboard.

also the difference between teh 3 and 4 pin fan headers is the 3 pin are meant to be controlled with voltage or not at all, 4 pin connections always run at 12v and the 4th pin tells the fan how fast to turn the 12v on and off to achieve a certain speed. this is why why when the pump header was set to pwm it produced the same results as dc@100%

A9 is the asus Q code for entering bios i believe. those codes let you know what the motherboard is doing, not a temperature readout. I would check with OCCT or some other temperature monitor in OS to see cpu temps. also in bios under system health/system monitor or something like that. as for the aio header it should be pushing a steady 12v no matter what so thats why the rpm did not change.

ok so I decided to oc my 1800x since im heavily considering a new cpu soon (5800X3d here i come!) i have the cpu core voltage set to auto 1.4v for 3950mhz. and it is stable heres the issue get conflicting voltage readings with occt and my corsair link software for my h100i gtx. in occt the sensors i track are core 0 VID: IDLE 1.35 LOAD: 1.294v MAX: 1.35v cpu core voltage (svi2 tfn): IDLE: 1.388-1.4V LOAD: 1.33-1.35v MAX 1.40v cpu core vid (effective): IDLE: 1.35V LOAD: 1.35v MAX: 1.35v in corsair link my voltage reads corsair vcpu: IDLE: 1.41V LOAD: 1.44-1.45V im not super worried becasue the cpu is staying well under 80c but has anyone else seen corsair link misread cpu voltage? or is it that occt is wrong?

i was just noting that the timing of the post lines up suspiciously well, not the whole content. and billet labs reasons for not wanting a prototype back make sense to me. in my opinion LMG should try and get the original back to billet labs AND pay for the replacement cost as a good gesture EDITED TO MAKE A LITTLE MORE SENSE.

this seems to line up with a (at time of posting)14 hour old billet labs post on reddit saying that lmg now knows who bought the prototype cooler and offering to try and get it back. and confirming that it isnt in the possession of a competitor company.

previous mount and paste spread was good, i noticed that my version of occt was quite outdated and updated it. now they dont show the "cputin" sensor the cpu and gpu on die/package temps are within range of normal 70-80c max on load. but if i load up the older occt the cputin sensor is still in the 94c to 98c range with a cpu+gpu or gpu only load. (makes me think even more this is a vrm temp sensor) i turned off overtemp protection so i could watch the sensors longer. manually setting my cpu package voltage to 1.275v instead of auto(saw up to 1.44v) seems to cap the cputin reading at 91/92c under load. changing the voltage does not change the cpu/gpu die temperatures significantly, just the cputin sensor value. and it seems stable so i will do this for now.

good point, i did notice the backplate could move a bit when the cooler is off.

you have to remove the ring around the top of the fan that says AMD or the stock cooler is too tall to fit in this case. it just uses the stock mount with the 4 screws

my moms asrock a300 pc has been shutting down randomly on her lately. SYSTEM SPEC: ryzen 3200g stock settings, stock cooler modded to fit, 16gb 3000mhz ddr4, 1tb m.2 i can recreate the crashes with the power test on occt, the 'CPUTIN" reading quickly goes to 95c while the actual cpu core reading is much cooler. if i run a cpu only load on occt temps are as expected. if i run a 3d load on occt the cputin reading also quickly increases. yesterday i dusted the pc and replaced the thermal paste as this pc is pushing 3 years old now but those had no effect. im going to replace teh thermal paste again to see if i just did a poor job. i have a suspicion that the "cputin" reading is actually somewhere on the motherboard vrm as it is so much higher than the other on chip readings but have no way to figure that out now.