• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

X58 overclocking club

Eehhh, I want to, really I do.... But they'd have to sponsor me. I won't give them more money, I've killed too many of their boards through the years.

In order of DEATH at a quickness.

Biostar - look at these boards wrong, they die. I don't even understand why!!!
AsRock - Partially because well, asRock. BUT I love their off the wall ingenuity!
MSI - Every single board I ever overclocked on any type of cooling died. Not right away though!!
Gigabyte, I've killed a couple, but in general very robust.
Asus, I've killed ONE single board through the last 15 years of extreme overclocking. One. That's it.
Model was M3A32MVP wifi deluxe app solo. I probably benched at least 2 dozen or more chips on that board.
Including extreme temps. That's what it took to finally kill it. Over a span of a few years, not in a month though.
I built a system for my little brothers (board was MSI Z370-A Pro) and it was fine ^_^
 
Uhh.. <---- See avatar to the left? Not exactly into air cooling my stuff, I bench competitively.
Using air with my main PC (with a GByte board & GPU) and no problems. The X58 system has watercooling though the 290X is air-cooled.
 
Using air with my main PC (with a GByte board & GPU) and no problems. The X58 system has watercooling though the 290X is air-cooled.
I was just trying to portray what others don't really get to experience. Sure any system can last and last, but I really beat on my hardware lol.
Should have included DFI, I haven't killed any of those yet though. I still have 2 left from socket 939.

I have one remaining MSI board that was RMA'ed 2 times. The 3rd board stayed working and I still have it. Some AM2 job. It's delicate, I bet if I where to pull it out, it would just die before I hooked the cables up to it.

Nope, free MSI board, I'll beat the shit out of it and that might make me a believer again. But I won't purchase one.
 
Still have it?
Unfortunately no, I sold it last summer with my R3F to help fund our trip to Alberta. I really shouldn't have and every time I talk about it it hurts :cry:

They say it gets easier with time.. it does not :(

Still have all my awesome ram though :D

Edit:

I do like my Asus boards, they have been awesome for me.
 
Agreed on ASUS X58 boards. I almost exclusively have their boards for X58 although I do want a GB or better EVGA board (only non-ASUS is X58 SLI LE).

Got a P6X58D-E, P6X58-E WS, P6X58D Premium, and Rampage II Extreme, all fantastic boards.

Looking back MSI boards were hot garbage for the most part... Biostar had their moments though, the TZ77XE4 is an amazing motherboard. MSI nowadays makes killer budget boards (B450 Tomahawk/Max/II), and across their lineup has DDR4 OC thoroughly figured out.
 
Agreed on ASUS X58 boards. I almost exclusively have their boards for X58 although I do want a GB or better EVGA board (only non-ASUS is X58 SLI LE).

Got a P6X58D-E, P6X58-E WS, P6X58D Premium, and Rampage II Extreme, all fantastic boards.

Looking back MSI boards were hot garbage for the most part... Biostar had their moments though, the TZ77XE4 is an amazing motherboard. MSI nowadays makes killer budget boards (B450 Tomahawk/Max/II), and across their lineup has DDR4 OC thoroughly figured out.
MSI B450 M-ATX?? Nah, my ROG B450-I would thoroughly spank any of em. But I'm not sure if it would be considered a budget board, though it's not an X series chipset.

Bet I could kill that TZ77 with a single CPU some LN2 and a hit of v-core nitrous. But then again, I don't think I've clocked anything on 1155, so I should just keep my mouth shut haha.

Unfortunately no, I sold it last summer with my R3F to help fund our trip to Alberta. I really shouldn't have and every time I talk about it it hurts :cry:

They say it gets easier with time.. it does not :(

Still have all my awesome ram though :D

Edit:

I do like my Asus boards, they have been awesome for me.

Aw bummer. When my W11 decided to update kill it's self, I lost all my Xeon submissions. Well they might still be there in Windows.old file, but I don't have the system up and running.

Family trips are more important. And that's why you sold it. You gave your family memories that'll last forever. I'd sell ALL of my gear for that any second. Just don't let my wife read that.... :fear:
 
How you can have so bad luck that an update kills the OS? :(
 
How you can have so bad luck that an update kills the OS? :(
Well I don't really know. If that's the worst luck for my PC adventures this year, I'll take it.

At our team site we have a specific thread called "What have you killed today" -

I try and make it a serious and strict point to stay as far away from that thread as possible.

In fact, I read it and shiver at some of the bad luck other's have had.
 
So question to ya'll 'enthusiasts' squeezing every performance drop out of x58, with Spider-Man out I was curious if any of you have ran the game successfully with RT enabled w/ x58?

My game insta-crashes on the 3070 with RT on, but runs flawlessly 60FPS+ maxed 1440p+DLSS w/ RT off. Game is patched to latest 1.817 'supposed RT fix' patch. I have dialed the OC on the GPU and CPU thinking instability but no avail, I've read exactly one user on reddit in-touch with a nixxies (who handled the port) dev directly saying the RT implementation uses AVX/AVX2 instructions which we all know is no-bueno for us if that's the case till patched - if ever. Just wanting to stop chasing the rabbit if that's the case and enjoy the game.
 
I have a p6x58d premium and went from an i7 920 to a X5675 this year and have been trying to get it stable for the past 7 months, my goal when buying the X5675 was to reach 4.5Ghz stable but temps were good only when gaming, when stress testing with AIDA64 temps hit over 80C frequently, I'm on air with a Hyper 212 LED Turbo so I decided to bring clock down to 4.4GHz to finally try and get it stable. I still get BSODs from time to time, after so many tweeks and time working on it to get it stable makes me think the problem is with my RAM sticks not the overclock itself (the BSODs is always memory related, DRIVER_IRQL_NOT_LESS_OR_EQUAL D1), even though they were fine with the i7 920 overclocked to 3.6GHz and never gave me any BSODs I think at least one of the sticks might be dying, I recently used memtest86 to check for problems and each one passed flawlessly for over 20+ hours running though.

Before I bought the Hyper 212 LED Turbo I was running the i7 920 with stock cooler and stock speed, I had 6x4 sticks giving me 24GBs, after installing the new cooler I lost triple channel and had to downgrade to 4x4GB for 16GBs (the first 2 DIMMs stopped working), the stock cooler was dusty so I either damaged the DIMMs slots while installing the Hyper 212 or I just need to clean it up to get tripple channel back. The RAM sticks are very old ones that I've been using since 2011/2012, it's the corsair XMS3 with XMP for 1600MHz at 1.65V, soon I will be buying 3 HyperX 8GB sticks rated for 1600MHz at 1.5V to see if I can finally get rid of the BSODs once and for all.

Currently the CPU is at 4.4Ghz with 1.38V (23x192) and 1.35V QPI, I will be posting some screenshots of bios settings and other stuff while I try to get tripple channel back before buying new RAM.

CPU-Z:
1661548665568.png
1661548707296.png
1661548854704.png


Overclock and bios settings on AIDA64:
1661549041685.png

1661548993078.png


Temps while browsing:
1661549129365.png


All the BSODs past months and last one earlier today:
1661549315526.png

1661549370239.png
 
I have a p6x58d premium and went from an i7 920 to a X5675 this year and have been trying to get it stable for the past 7 months, my goal when buying the X5675 was to reach 4.5Ghz stable but temps were good only when gaming, when stress testing with AIDA64 temps hit over 80C frequently, I'm on air with a Hyper 212 LED Turbo so I decided to bring clock down to 4.4GHz to finally try and get it stable. I still get BSODs from time to time, after so many tweeks and time working on it to get it stable makes me think the problem is with my RAM sticks not the overclock itself (the BSODs is always memory related, DRIVER_IRQL_NOT_LESS_OR_EQUAL D1), even though they were fine with the i7 920 overclocked to 3.6GHz and never gave me any BSODs I think at least one of the sticks might be dying, I recently used memtest86 to check for problems and each one passed flawlessly for over 20+ hours running though.

Before I bought the Hyper 212 LED Turbo I was running the i7 920 with stock cooler and stock speed, I had 6x4 sticks giving me 24GBs, after installing the new cooler I lost triple channel and had to downgrade to 4x4GB for 16GBs (the first 2 DIMMs stopped working), the stock cooler was dusty so I either damaged the DIMMs slots while installing the Hyper 212 or I just need to clean it up to get tripple channel back. The RAM sticks are very old ones that I've been using since 2011/2012, it's the corsair XMS3 with XMP for 1600MHz at 1.65V, soon I will be buying 3 HyperX 8GB sticks rated for 1600MHz at 1.5V to see if I can finally get rid of the BSODs once and for all.

Currently the CPU is at 4.4Ghz with 1.38V (23x192) and 1.35V QPI, I will be posting some screenshots of bios settings and other stuff while I try to get tripple channel back before buying new RAM.

CPU-Z:
View attachment 259496View attachment 259497View attachment 259498

Overclock and bios settings on AIDA64:
View attachment 259500
View attachment 259499

Temps while browsing:
View attachment 259501

All the BSODs past months and last one earlier today:
View attachment 259505
View attachment 259507

My experience with 5670s is that thermals hit the knee of the hockey stick when pushing stable voltage past about 4.2. Good RAM is a must, but cooling might be your limiting factor with that 212.
 
4.4 is pretty heavy for a 212, I have the Evo. It barely cooled my old X5690 at 4000MHz with a Linpack load and a butt load of caseflow.
 
4.4 is pretty heavy for a 212, I have the Evo. It barely cooled my old X5690 at 4000MHz with a Linpack load and a butt load of caseflow.
The problem there was unlikely the heatsink and more the fan. I have the stock heatsink with my Dell T3500 and a 120mm fan with a custom shroud. See below..
IMG_20210707_075718_1 - Copy.jpgIMG_20210707_075702 - Copy.jpg
The fan runs at 2100ish RPM. It's not loud and allows running the W3680 at 4.1GHZ. Never goes above 70C.

If your Hyper212 was struggling, you needed a better fan, or higher fan power profile in the PWM section of your BIOS.
 
The problem there was unlikely the heatsink and more the fan. I have the stock heatsink with my Dell T3500 and a 120mm fan with a custom shroud. See below..
View attachment 259536View attachment 259535
The fan runs at 2100ish RPM. It's not loud and allows running the W3680 at 4.1GHZ. Never goes above 70C.

If your Hyper212 was struggling, you needed a better fan, or higher fan power profile in the PWM section of your BIOS.
Is yours boosting? I was running a static OC at 200x20, 1.275v . could never get the boost to work right :D

Was using R3F, plenty of power there, 1.6v no problem for 4800 with the windows open in the winter :D

Edit:

Was using 120x38 strapped to it :laugh:
 
Is yours boosting? I was running a static OC at 200x20, 1.275v . could never get the boost to work right :D
It's a Dell and running it soft OC'd with ThrottleStop required hitting the BIOS to turn off the boost options. However, it does run 100% stable with 20mins of Prime95 to verify stability.

EDIT: Just booted into it's BIOS and verified that SpeedStep is also disabled.
 
Last edited:
My experience with 5670s is that thermals hit the knee of the hockey stick when pushing stable voltage past about 4.2. Good RAM is a must, but cooling might be your limiting factor with that 212.
Yes, I agree. With the current voltages I'm using cooling isnt good enough, this is because on desperation to get it stable I raised QPI from 1.3 to 1.35, I noticed a huge change on temps after this, before it used to stay below 45C while idle and below 60C while gaming. I'm not an expert when it comes to overclock, but I think I will be able to get QPI down to 1.3~ again with the new RAM sticks which are rated as 1.5V instead of the ones I'm using now (1.65V). Which temps should be ideal while idle and gaming?
4.4 is pretty heavy for a 212, I have the Evo. It barely cooled my old X5690 at 4000MHz with a Linpack load and a butt load of casef
The turbo version is a bit better than the Evo, which temps should be ideal while idle and gaming?
 
Yes, I agree. With the current voltages I'm using cooling isnt good enough, this is because on desperation to get it stable I raised QPI from 1.3 to 1.35, I noticed a huge change on temps after this, before it used to stay below 45C while idle and below 60C while gaming. I'm not an expert when it comes to overclock, but I think I will be able to get QPI down to 1.3~ again with the new RAM sticks which are rated as 1.5V instead of the ones I'm using now (1.65V). Which temps should be ideal while idle and gaming?

The turbo version is a bit better than the Evo, which temps should be ideal while idle and gaming?

Temp targets are to a certain degree a matter of preference. Broadly speaking, keep it under Tjmax and you're fine. Lots of folks seem to like a 70C load limit, which is a really tough target to hit when chasing a decent OC. I personally like my distance to Tjmax to be at least 5C under heaviest typical workload, which is WCG for my X58 box. Temps in other scenarios is what they is.
 
Broadly speaking, keep it under Tjmax and you're fine. Lots of folks seem to like a 70C load limit, which is a really tough target to hit when chasing a decent OC.
This is because many of the 1366 Xeons have a TJMax below 70C.

For example, my W3680 has a 67.9C limit.

My X5680 is 78.5C.

My X5675 and X5650 are both 81.3C.

The thermal limits vary chip to chip on the 1366 platform so most people want to keep there CPU's well under 70C to be safe, even when OCing.
 
I just cleaned the cooler and case which were both a bit dusty and brought clock down to 4.2GHz to check how thermals would be with low voltage compared to 4.4 I had before and thermals chocked me, now I know if I get any BSODs is not gonna be because of cooling.

1661940243046.png
 
@İsmailTPC_Hastanesi has created modded X58 BIOSes for some motherboards, which now support PCIe NVMe add-in cards.

Check it out here:

Pretty cool stuff :)
 
Back
Top