Asrock 990FX Extreme3 - Issues

[DigitalUK]

im having a fair few problems with this board, has anyone else got this board as im wondering if the chipset is faulty.
CPU = 1090T
Cooler = Corsair H100
Memory = Corsair Vegenance (4x 4GB 1866)
PSU = Thermaltake Toughpower 700w

at first (bios 1.0) the cpu would throttle down at max load, no problem a bios was released (v1.20) which seems to have fixed the throttle, but even at stock the chipset heatsink is red hot (and i do mean hot almost burn your fingers hot). any overclock of anything at or above 3.8ghz will result in a BSOD or hang (not down to voltages tried from 1.4750 to 1.55v) exact same BSOD/Hangs (please note i didnt noticed the chipset HS temp until after trying to overclock as the board was brand new)
i removed the chipset heatsink cleaned and replaced with MX-2 but still chipset getting very hot (cant find a way to read the chipset temp at all) being a 990FX chipset i expected a little more out of it.

CPU Temps Idle 22c Max load 33c (Intel burn Test & Prime) 4ghz load temp before BSOD/Hang 42c
also memory was run at stated 1.5v at 1600,1333 & 1066 and also tested 1 stick at a time

 

[PaulieG]

im having a fair few problems with this board, has anyone else got this board as im wondering if the chipset is faulty.
CPU = 1090T
Cooler = Corsair H100
Memory = Corsair Vegenance (4x 4GB 1866)
PSU = Thermaltake Toughpower 700w

at first (bios 1.0) the cpu would throttle down at max load, no problem a bios was released (v1.20) which seems to have fixed the throttle, but even at stock the chipset heatsink is red hot (and i do mean hot almost burn your fingers hot). any overclock of anything at or above 3.8ghz will result in a BSOD or hang (not down to voltages tried from 1.4750 to 1.55v) exact same BSOD/Hangs (please note i didnt noticed the chipset HS temp until after trying to overclock as the board was brand new)
i removed the chipset heatsink cleaned and replaced with MX-2 but still chipset getting very hot (cant find a way to read the chipset temp at all) being a 990FX chipset i expected a little more out of it.

CPU Temps Idle 22c Max load 33c (Intel burn Test & Prime) 4ghz load temp before BSOD/Hang 42c
also memory was run at stated 1.5v at 1600,1333 & 1066 and also tested 1 stick at a time

I have a Extreme4 on the way, so I'll let you know if it has the same issue. There is a fairly easy solution (if hangs are a heat issue), regardless of whether or not it should get that hot. Get a Antec spot cool, or just tape a 40mm fan on top of the chipset heatsink.

 

[de.das.dude]

didnt they include a fan for the mosfet?

 

[PaulieG]

didnt they include a fan for the mosfet?

The EX4 does. Not sure about the EX3.

 

[de.das.dude]

yeah, i wish i had moolah. i really liek that board.

 

[DigitalUK]

no fan at all, im almost thinking not fit for purpose as ive never felt heat like it on a chipset heatsink even the nvidia 780i never felt anywhere near this hot to touch.

i did try blasting air at the chipset to see if that was it, didnt seem to change anything

 

[PaulieG]

no fan at all, im almost thinking not fit for purpose as ive never felt heat like it on a chipset heatsink even the nvidia 780i never felt anywhere near this hot to touch.

i did try blasting air at the chipset to see if that was it, didnt seem to change anything

You might just have a bad board. The reviews overall are pretty good, from what I've seen.

 

[Exeodus]

I have never seen a AMD NB run that hot. I wonder if the board is overvolting the northbridge? If it is, it would cause the BSOD issue the OP has been experiencing.

 

[DigitalUK]

just checked the NB Vid using AOD bios 1.20 - auto setting = v1.2000 chipset hot to touch, downgraded bios to v1.00 NB Vid is set to 1.1750 and is much cooler under load.

 

[TheoneandonlyMrK]

a o.025 volt increase should not on its own acount for a big temp difference imho, im no expert on AMD rigs but imho something else is going on or being enabled by the newer bios which is putting undue stress on the NB,

Id consider returning it to the shop for a swap if it were me or wait on a newer bios to hopefully fix it ,but in that case do contact asrock and complain to ensure they know of the issue ,afaik two different boards still have the same type of chipset and my boards NB runs at 30 permanent ish with 1.3volts on it so yours sounds a bit too hot

 

[de.das.dude]

yep your board is poop. the last RMA'd mobo i had was like this. the NB hsf would burn my fingers :O
finally the lan port on that mobo derped and i got a new rma.

my NB was >60C with 30C ambient.
:/ i burnt my finger so many times >.> .....

 

[DigitalUK]

think ill return it and go for a sabretooth or something. i cant even get a temp reading on this board and ive tried everything. thanks for the input guys

 

[eidairaman1]

1 for 1 swap with the Maker of the board is best bet.

 

[de.das.dude]

yeah. keep the board. i was at the ASUS Rma center here and saw a couple of 99FX sabretooths for rma.

ASUS is going down man.

 

[Exeodus]

yeah. keep the board. i was at the ASUS Rma center here and saw a couple of 99FX sabretooths for rma.

ASUS is going down man.

I think you may be right. The very first true DOA board I ever purchased was a Sabertooth P67 board. I couldn't believe it!

 

[eidairaman1]

Ever since the first Asus board ive purchased back in 2002 ive been warry and not very trustful/satisfied with them, they are definitely overrated, Over the Years ive heard Asus Problems out the wazoo compared to Gigabyte, MSI, Biostar, ECS. Ive replaced several Customers machines that had burnt out Asus boards with ECS and never a problem.

 

[DigitalUK]

i was looking at the Gigabyte GA-990FXA-UD3 before i read that the UD5 & 7 suffer from vdroop) but i got all hung up on the UEFI, ive always used alot of MSI boards for myself (but the msi 990fx boards seemed abit boring) and ive used loads of ASROCK boards in loads of builds and have always been pretty good.

as an update, ive turned everything off auto and the chipset is now almost cool to the touch now plus pointed a fan at the VRM heatsink which was getting pretty hot. im starting to wonder if this is a bios fault, as im running at 4ghz at the moment (Bios 1.20) and passes 1024 wprime but cant seem to do 10 passes of intel burn test (high).

 

[eidairaman1]

Tell u truth Id send a Report of your finding to ASRock actually because 1.20 Bios just says Update CPU Code n nothing more, Perhaps another bios update is needed since the board is so friggin new. (this is the reason i dont buy boards till like 6 months or a year down the line because of quirks)

I wouldnt Grab a Gigabyte board right now because of the troubles with the x79 series (who knows how many other boards are affected with faulty bios and caps/mosfets).

tbh Im not looking at the system innards when im using the computer, just gaming, facebook, Camera IM, movies downloads. the one component that gets about 99.99% attention on a desktop computer is the monitor...

 

[micropage7]

have you checked the hsf pressure to the processor? does it pretty tight?
if everything is ok the err really from the board

 

[DigitalUK]

ok i have been playing with this board for the last few days and have come to the conclusion that the board is fine it is the bios that is the problem, specificly the voltages. i have fitted a noctua fan above the VRM for good measure (thats just common sence..lol) but the BSOD were all based around the 000124 error which is usually not enough voltage, but according to bios i had it set to say 1.5000v which should get this 1090t upto at least 3.8-3.9ghz but would blue screen. as i was adjusting i thought what the hell and did a suicide run with 1.5500v but still BSOD 000124 (cant be right). i had CPU-Z and Core temp open and noticed that core temp was showing a VID of 1.4500v. it looks like the bios is setting the wrong voltages (undervolting by .1000)

screenshots below.
from Bios CPU Core set to 1.6000 (suicide)

from Windows.

 

[eidairaman1]

Look at AsRock Forums if any exist and see if anyone else is experiencing the same issue with the same exact board. You may want to up chain this Issue to AsRock. Push Comes to shove 1 for 1 swap the board