I've got almost 25 years hardware experience and I don't think I would have thought of that for that error. I'd have been tearing my hair out on this one. I've reused parts in upgrade builds since the 90's and never thought of this. Thanks Jay.
Thats the first thing were i would look since reinstalling ios is a lot more work I begin with plugging and unplugging wires and then gpu memory and cpu ...
@@Ivantonkic2 I hear you. That would be very frustrating, not knowing what fixed the issue. In my career when I come across something like this I make a mental note and file it away for future use.
Remembering that it is a common error on some platforms, in the old days when there was still the extreme series from Intel, if you pressed too much or left too much slack, some memory channels of the usual 4 simply wouldn't boot, you could use the system but without those memory channels. In simpler systems that have fewer pins it is something much more difficult to happen.
Jay, I literally just finished adding a water-cooling loop to my system and was dealing with some weird issues and was worried that socket pressure could have been one of the issues. This video was released at the perfect time.
I had the same issue with my Asus H410M-, Corsair LPX ram with the Deepcool L360 ARGB with the tubes from the cpu block touching/pushing the ram when i mounter the tubes coming off the cpu block other way(without touching/pushing the ram) it works fine
Tolerance stackups can miss during the manufacture of a cooler/aio...something that jaybeebee1cent doesn’t get. Yes, you will have to tweak around with torques to compensate. Start with low uniform torques, wait for boot and work your way from there based on the temps you see.
hey could you come back to update this? I have a watercooled build as well and thinking this could be an issue I'm dealing with and never thought about
@@Z3RK3R2020 At this point, I'm leaning towards this problem more likely being cause by an aggressive undervolt setting rather than the socket pressure being too high but I'll need some time for stability issues to present themselves
This is LITERALLY me this past month. I’d never built a PC before and I’m currently on my first try. Brand new cooler, motherboard, ram, CPU, GPU, power supply, everything! I ended up having the error 55 issue and taking it in to a local tech shop to help me out. We’ve been trying to figure out exactly what the issue is for a couple of weeks now. I even replaced my CPU and Mobo after figuring out it wasn’t my RAM that was the issue. Incredibly frustrating and you cannot find any information about this anywhere online. Thank you for bringing this to the attention of people.
For a long time, friends have told me that you want to tighten your cooler JUST until it starts to resist with normal pressure, and I just thought it was to avoid threading. Now I know better. That's incredibly scary for someone who doesn't have dozens of Mo-Bos and CPUs, and now I'll be even more careful. Thanks, Jay! I love these little 'how to take care of your parts' videos from time to time, because longevity of GOOD components is incredibly important, and I'm all about this careful maintenance stuff.
Yooooo just last week my cousin was having this exact issue after building a pc that we could not track at all. We did reuse the aio cooler and we did our best not to fuck up the tightening but we're all human. We even checked out your videos then to see if you already talked about something like this but couldn't find anything exact until now. You're a life saver Jay, thank you so much!
I personally would've swapped the AIO back to the original to begin with as that was the only recent hardware change. If it were a brand new system build, I can see how this would be frustrating. Excellent tutorial as usual jay!
And this is why, if you ever have multiple upgrades planned for a build, you should install them _one at a time_ and then _test them thoroughly._ That way, if something goes wrong, you know which part caused the issue.
Sort of funny as I had an issue yesterday. Swapped a 120mm AIO for a Hyper 212 Black. Finished up and the PC wouldnt boot with both sticks of RAM. Ended up being the Hyper 212 bracket doesn't work with my motherboard, as tightening it adds pressure. Swapped the AIO back and boom it ran fine.
Great video Jay. I used to see this all the time when I was a SunMicrosystems hardware engineer back in 2001-2006. Granted, a different platform but the principle is the same. If you would over tighten the CPU module onto the system board of an E4500, E6500, or tighten the torx screws in the wrong order, the host would never pass POST. POST on these Enterprise beasts was quite a process. We were issued with a torque driver precisely calibrated for the correct torque to guard against this. Love the office refurb too. Great channel. Please keep it coming! Best, Jonny
I use to love them before i start to build PC as my job,i had so much weird things happen including exactly this what is happening in the video,i'm looking at this video and literally shaking my head.
Thanks for this video. I've never had this problem myself in spite of the fact that I've build well over a hundred gaming rigs over the past 30 years and have fitted air and liquid coolers of many different brands including some that needed to be modified in order to fit. I never tightened any of them down with a lot of force, though but always just enough so they wouldn't move. Now that I am aware of this being a possible problem when fitting an older cooler on a similar sized socket mount I'll be able to use this knowledge when troubleshooting problems because as you said, your first thought when you get errors like this is a problem with the RAM, once you eliminate that the motherboard, once you eliminate that the CPU, once you eliminate that the power supply. Your first thought is not going to be "Hey maybe it's the CPU cooler."
I experienced this very frustrating issue on my second build ever. Basically had to remove everything and then reseat everything twice to fix it. Crazy
I had similar symptoms with a i5 2500k and a z68 long time ago when installing a HYPER212EVO. Unscrewing the HYPER212, reseating the cpu and carefully centering the cooler while cross tighting the screws fixed my issue. It all depends on how the manufacturer decides the pressure amount their product needs to apply on the socket to be efective
This happened to me when I installed the CPU contact frame, two weeks ago. I followed the instructions from the kit and videos I watched. It appears that the 0.6Nm was too much for my setup. 0.15Nm was perfect and thermals are also great (no more crashes). It took me a couple hours to figure it out, but everything is perfect now. Thanks for your awesome videos, Jay!
We only live physically once so we shouldn’t be afraid to do anything bro, i smoke weed on my TH-cam channel and i ate Mcdonald’s inside Walmart 😈😈, screw anyones opinion 👌👌
That's exactly what happened with my AIO! I overtightened it and every so often it would just blue screen. After hours of nearly tearing my hair out as I troubleshot, I tried loosening the screws on the pump and voilà! No more random blue screens. After that, I just tightened the cooler down just until the cooler doesn't slide or move on the CPU when you try wiggling it around.
That was the method I used too and haven't had any problems. Haven't built a PC in a LONG time so was totally unsure of how torque would work with the springs in the mix. I settled on tightening the screws a little at a time in a criss-cross pattern until my heat sink quit sliding/moving with reasonable pressure from the sides and left it at that.
LGA would be fine if they designed them as removeable modules that have 10 micron tolerances, because then it wouldn't cost as much as a motherboard to replace a socket.
Your biggest hurdle with what you suggested is that you would have to make sure the socket is mounted as perfect as humanly possible, or else it may not make proper contact with the motherboard circuitry and nothing will work at all and could possibly brick the socket or the board or the cpu or anything plugged into it, but, if you could make a system where this wasn’t an issue, hypothetically, I’d be down for a replaceable socket, but it’s not like you could change generations though, cause the board layout for a 10th gen Intel is not the same as the layout for a 12th gen Intel and they would not be cross compatible even in the slightest
@@What_In_This_World Like I said the tolerances would have to be within 10 microns, anything much larger and the actual connections might not always line up, however if all LGA sockets were designed for this concept from the ground up we would be in a much more reasonable place because it would make cheaper boards $60-100 reasonable to repair with a 20-30$ socket, currently replacement socket prices are $5-10 but the labor is upwards of $100.
@@dagarath I was going to say that sockets already are replaceable modules. It's just that it requires soldering. But your last comment cleared that misunderstanding when you referred to it as currently being a $5 socket and $100 of labor. So that's fair enough, but: How would you securely mount this modular socket to the motherboard? Surely it would require a retention mechanism of it's own if you're not soldering the socket onto the motherboard. Adding an extra retention mechanism seems counterintuitive to me if the problem your trying to solve is the potential cost of replacement.
GUYS, Jay, you have no idea how much I have been racking my brain for the past 3 weeks after my mobo upgrade trying to figure out why my pc was unstable! I never considered that the pump was screwed on to tight! This video was a saving grace lol thanks man!
Thank you Jay! I had an 12600k build for a client that would boot but bluescreen' randomly after a few hours. Well I loosened up the cpu cooler per your suggestion and it appears to be stable and has passed multiple stress tests! Glad I watched your video because it definitely saved me hours of frustration.
it aint the overtightening per se its the bending of said overtightend cpu/MB/socketr etc :D if you for what ever reason would have a Motherboard that has a 2cm steelplate stabilizing it and your cooler plate is also a 2cm steelplate with a small insert in the middle out of copper and we asume you can tightening it in a way that those 2 plates are perfeclty plane to each other, you can pretty much squeeze to the point where the silicon is almost crumbling(or the ihs starting to get flatend out like in a hydraulicpress) and wont have any errors what so ever
I'm the same as a bunch of others here. I too wouldn't have thought to think about the cooler tention. Really interesting. Always something new to learn. Thanks Jay.
Fascinating ... I've never had such a problem but I'll add this to my checklist if I ever get a problem in the future. This is why I've been subscribed to this channel for years, it's interesting to get troubleshooting tips from people who build systems on a regular basis.
(To the viewers) You can also come across a situation where all your RAM sticks are individually working and each memory slot on your motherboard are individually working but as soon as you install a pair of memory stick where it activates the dual channel memory, now your system becomes unstable and crash. That's a thing and it's important to remember that the memory controller is in the CPU nowadays, among other things.
holy hell Jay, thank you so much for this video. My PC has been having that exact issue where it totally locks up and I have to hard reset it. I've been going batpoop crazy trying to figure out what has been causing this issue since April-ish and now thanks to this video I may have the fix. So again thank you sir because you might have just saved my sanity and wallet.
YES! Biggest problems I've seen people do with tech is WAY too much torque (this is not a car tire going down a bumpy road.) Than after that absolutely no focus on torque balance of things with multiple bolts. I had actually bought a micro torque driver not just to speed up the torque balance procedures but to show how far off the rails customers were with their DYI upgrade. I'd bet my last dollar that if you tested this, you would have found two corners relatively the same torque and than one corner the threads are barely grabbing while the opposite corner was suitable torque for a car part. Like a see-saw it will jack the CPU cockeyed into the socket.
This is immensely helpful. I built my first computer almost 20 years ago, and been regularly building since, and never once thought about this as a troubleshooting step.
Great info Jay. It never occurred to me that the Z-axis might be different from one socket to another that are otherwise compatible. I'd love to see you test a hardware fix such as adding washers to raise the height, which I would think would give more even pressure than simply loosening the screws.
Brings me back to the launch of Skylake, when Scythe had to update their coolers with thicker metal washers. The Z-axis was actually the same (or quite close), but the substrate was thinner, so high pressure mounts would bend it.
This is going to help a ton of people, especially with how the new 12th gen has changed the Z Axis height... I noticed that one of my Corsair AIO manuals explicitly says to not over tighten the screws now. I guess they are aware of this issue, but I honestly just heard about this issue from one of your earlier videos. Now when I install a cooler that has the spring loaded screws, I make sure to not bottom out the screws. Just let the spring tension do the work.
Thank you so much for including two codes today. That was very cool of you to follow through. I had the exact same issue with putting a Corsair H50 on a socket 2011 E5-2680 v1. The original heatsink was allowing the CPU to idle at 75°C so I decided to put an older AIO on it. I freaked myself out because after restarting the system it would not post. Turned out loosening the screws about 2 full rotations each fixed the issue. It now idles at 36°C. Can't wait to see the finished product and footage from Saturday's garage sale!
As a new first time PC builder...I have similar issue happening where my computer will just lock up slowly and the only way to turn off the PC is flipping the PSU power. I also had an issue with my DRAM light coming on (within 30 days of me building with brand new RAM) It locks up rough 2 times a month, randomly. Maybe it could be me over tightening the cooler. Thanks for the info!
@@GamingWithHybrid - Last system I built I picked up most of my parts at a MicroCenter 2 hours away and when I got home I dug out the CPU and it was floating around in the box completely loose, lots of bent pins (at least not sharp bends). I was able to straighten them all out by taking a box cutter blade (thin, strong and long enough to span an entire row of pins) and VERY gently and patiently wiggle the blade back and forth in line with the row, gradually straightening out all the pins on the CPU after a bunch of passes. Not for the faint of heart, though.
Ha, exact same thing with my 5900X and custom waterblock. Exact same symptoms, and memory related problems due to overpressure on the socket. Switched block to a heatkiller and noticed they had a different mounting mechanism, no more problems, and memory overclocking, etc all good.
Great video Jay! I was trying to help somebody in PCMR subreddit the other day, and never saw a resolution. Wonder if this may have been the issue? I recommended testing all RAM one stick at a time, and they all passed individually, but not when all installed. I feel kind of bad, because I suggested it might have been the memory controller or the motherboard RAM slot/trace.
Always excellent tips and outright mental tools explained so well! Especially for someone just starting to learn! Thank You Jay and crew!! Love the humor also!
Very interesting. Too much pressure on the CPU not the first thought that comes to mind. These troubleshooting videos are very cool and helpful. Thanks.
I had a similar issue but while overclocking my i7 4770k. I couldn't get the damn thing over 3.9Ghz (which is the max boost clock for this CPU) without getting random blue screens or even crashes while just booting up, even with a lot of voltage and good temps. Recently I changed my thermal paste cause I didn't do it for a few years, temps were a tiny bit better but I think that the cooler tension was the actual problem since now I can get the CPU to run 24/7 at 4.2Ghz without any problems.
I have an old PC that I had repurposed with Linux to run World Community Grid and I was getting totally random crashes just like this but I could never pin down where the problem was. Every single test I ran came out fine. I had tried multiple distros thinking maybe I had gotten a bad build. I decided to try with my old Windows 7 install but still had the random lockups. Watching one of your earlier videos I had decided to redo the CPU paste on my current rig and figured what the heck I'll do the old rig too. The random crashing went away and now, full circle, I think I finally know why. I've always tightened any screws or bolts in a cross or star pattern but I hadn't thought about how much torque I was using. This is why I like your channel. Because of watching your content I finally started overclocking and trying other things I was always interested in but just never pulled the trigger and because you're a tinkerer like me you figure out such random problems that someone might not have even known WAS a problem. It's those tiny bits of obscure knowledge that are truly helpful to have. Thank you for doing what you do.
DUDE! You are a God! Re-pasted my cooler & added another 16GB of ram to my laser engraving PC this morning. It booted then began to do random restarts followed by BSOD. Only way I could get it to run was to remove the new ram. So while I was engraving a quick test piece just to be sure the problem wouldn't reoccur I sat & watched your vid on my phone. Then when the test piece finished I loosened the cooler & reinstalled the new ram. Wooo Hooo! 32GB of ram & a stable system. Sweet. Cheers Jay.😀✌
Yeah, never try to tighten screws to the maximum on cpu/heatsink frames. Had this issue on my very first build like 15, 16, 17 years ago. It took me like 2 weeks to figure out what's wrong, and a lot of new ordered parts to send back. I could replicate the error and it felt really satisfying finding it out. Especially with Intel 12th and 13th gen i would be even more careful to damage something, as the whole cpu tends to bend easily without the usage of a third party cpu frame. And Intel doesn't care. 😑
Does this also apply to the amd processors? Because I also sometimes have random bsods and or game crashes. Not to mention the cpu (ryzen 9 5900X) is hooked up to a 360mm water cooler from nzxt and still gets up to something like 90 degrees C (I use kryonaut extreme paste)
I finally got around to turning XMP on after you saying it a dozen times. Everything was working great but my system would have a problem shutting down. Every once in a while after clicking shutdown,Windows 11 would stop normally but everything else would stay on. I had the latest BIOS installed, Windows was completely updated. Had no idea where to start other than googling the issue. Even though Windows said it was all up to date, the Intel Management Engine driver was two years old and I don't remember how many versions out. Got that updated and everything is working fine. In the 30 or so years I've been messing with computers I've never had one not shutdown like that. Fun learning experience.
That’s why I didn’t like when I built my first Ryzen system last year I found the new wraith stealth stock cooler with my R5 3600 didn’t use the included heatsink clip brackets on my motherboard but required I unscrew those and remove them and screw the heatsink into the backplate. And the same with my Hyper 212 Evo, even though the older and smaller Hyper TX4 for my previous system included and used those little lever clip hold down things. Screws allow you go tighter most of the time, sure, but sometimes too tight, and screws can strip, both their threads and the head, especially Phillips screws. I prefer the lever mechanism, reminds me of old computers anyway, even an old Celeron 600 MHz cpu used that mechanism for its tiny heatsink, pretty much all computers used lever mounts in the beginning.
Jay would be an amazing tech teacher because honestly when I was in school they thought it was bad drivers or even the memory channels never did they think to loosen the aio
I never would've thought of this! Mind you, I fixed a friend's computer the other month which kept power cycling during boot and we changed the psu, psu didn't fix the issue. After 20 minutes I unplugged the dvd drive and that's what was causing it. It honestly looked like a power supply issue but now at least her 10 year old setup has a new power supply. It's literally the nicest thing in her pc
UM OKAY... WAIT... Is THIS my issue??? My Kraken z73 has an LGA 1700 bracket for my 12900k... But this is like my EXACT complaint, AND I'm new to building. I've spent WEEKS trying to find an answer and it's been so difficult to even DESCRIBE to anyone because it's so intermittent and random. Thank you SO much for this lead. idk where I'd be without your channel.
You know what I like about this the comp? You can tell only the hardcore are still participating, if you divide your entries by the total and have been in it since the beginning, you should fine your odds have been slowly increasing.
I'm speechless I've been troubleshooting my random blue screen crashing for weeks now on my 12900k and literally 2 days before this video I concluded it was a CPU issue. Perhaps the new architecture isn't mature enough or I got a bad combination of software with bugs in windows. I immediately did a half a rotation unwind on the mounting screws and my rig hasn't crashed yet. I'm so happy thanks for making this video.
I had a similar issue years ago on a X299 board that was open box, one ram channel wasn't showing up. I reseated the ram and flipped them around a few times and the system wouldn't see all the channels. I then took off the cooler and CPU, then looked at the pins with a magnifying glass. One pin was slightly out of position by like half a millimeter. I very carefully moved it into it's correct position with a pair of tweezers and put it back together. Viola! The 4th ram slot showed up fine. Never had a problem with it as long as I had it.
Oh brilliant. I've started getting random bluescreens on my Intel 8700k rig that I use exclusively for streaming now from time to time. Thought it might be degraded liquid metal that I applied 2 years ago after delidding, but temps are fine, and well within normal ranges. Was starting to look at memory, but now I'm going to go check my air cooler mount to make sure it's not too tight! Thanks jay!
I built my first system back in March. Used the AMD stock cooler and literally from day one have always had this issue or memory issues. For the life of me I couldn't figure it out and asked everyone including forums, microcenter employees and my brother In law who's been building computers for years. Then I saw this right before I gave up and decided it was the CPU ( I have already replaced my motherboard). Tried it with stock cooler and still couldn't get it. Bought a cheaper single tower air cooler and immediately fixed everything!! I can't tell you how relieved I was.
I found this out the hard way not very long ago. As I was doing some upgrades to my old Asus P6T, LGA 1366 system and I ran into some crazy stability issues. I went from a I7 930 and a very old corsair H100 AIO to a Xeon X5690 and a custom loop using an ek supremacy evo block. But I couldn't find a compatible mounting bracket. So, I modified my old H100's mount to take the larger studs from the evo block's 115X mount. I didn't take into account just how much thicker that Xeon X5690 is over let's say an I7 8700K or realize how the additional clamping force would impact anything. To make a long story short, I wound up backing off the screws nearly 2 full turns from the bottom stop position and it works perfectly now. Thanks for this information, Jay. I only wish you could have posted this much sooner.
I actually just learned that you can over tighten and AIO mount, my PC was having all kinds of skipping issues and locking briefly. I also experienced higher temps nothing bad but not good for the AIO I have. So I thought, let me un tighten it once i did that. System now runs smoother and cooler temp wise. love the vids JAY keep! long time fan!
As someone who worked in the PC repair business, I have seen this numerous times when someone builds a PC using an AIO cooler. People don't realize that they only have to tighten it down until it's snug. They think they have to tighten it down until the screws stop turning. That was especially common in the era that the AIO Jay used in the video was from. BTW, I have a 12 year old Asetek 240mm AIO, I believe it's a 570 LX, that is still going strong and cooling my rig running an i7-875K. I use that PC to run some older retro games that won't run on my current desktop.
This video has taught me some more important and often overlooked complexities of cooling and general closeness of components like pins on an IC/CPU. Even a tiny shutdown is cooling the system enough to create some major failure but maybe only temporarily. And it cannot be expected that every computer user knows this. This video stands for itself and it's usefulness. Even if what is being taught (expansion and contracting of objects based on heat and the general proximity them in such a tight space.) is actually pretty common knowledge, the presentation of it in the way that Jay has lends such clarity. The shit out of it.
This video brought back a memory from when I did this on pentium 4/LGA775/stock cooler. It was really easy to over torque the downdraft cooler and then it would not post. I had just bought a new motherboard because the original(prebuilt compaq presario) motherboard caps blew up and was convinced there was some OTHER major problem like a DOA board or a psu issue. I didn't really understand what was going on, and felt like a genius when I finally figured it out.
I had a P4 that was in a tower case and I put a Zalman solid copper fin heat sink on it. You may remember it. It was fins pressed together about 4" spread out. The thing must have weighed a pound. Well, in a tower case, that eventually tweaked the mobo enough. If I hit my desk just a little, REBOOT. Took it off and all was good.
I had the same issue when I used a smaller drill to screw the AIO on. I had random blue screens and instability. I realized the cooler was over tightened when I thought to reseat the CPU. After a couple weeks my buddy was building his pc and had the same issue, I told him to back the tension off of his CPU cooler, sure enough it ALSO fixed the instability! This issue isn’t talked about enough! Lol awesome video
I had this same problem about a month ago and the kicker for me was that I'm using an EK block that has springs. As soon as I remounted the water block and left them looser it fired right up with both sticks of ram. Now I can make the problem happen on command by tightening the mounting screws. I tested this just to make sure I wasn't crazy and this was actually the issue since the remounting process might have been the thing that fixed it.
Had this exact same problem on my x99. I spent months swapping out parts trying to find what’s wrong, finally found someone talking about this on a 10 year old OC form. Glad someone made a video on this.
I'm glad this video was posted because I have all these exact same issues but I just have a 5600x with stock cooler. Sometimes won't post or will freeze after a few minutes or the shutdown screen also.
As newish builder, who has dealt with some real hair pulling issues like this already, these videos are wonderful and I really hope you make more of these even if they are in shorter form.
12:58 Jay, think about what you said there, it's the BIOS, you don't need RAM to access the BIOS, it's not how Windows accesses the RAM, but just the fact that it does...
10:07 I used to work at a cellphone store, the same thing happened with customers' phones as well. Some even claimed it works fine at the store because they know there is a cell tower in the store.
The thing is this isn't an issue for just desktop coolers, industrial grade server coolers for those xeon scalable cpus have a) torque order on the screws b) max torque on the screws. And I've had the phantom ram or pcie slots from the wrong torque on a socket with 3k USD cpus.
Just built my first system this week, but I am still waiting on my PSU that is coming in tomorrow. I have had my system sitting for like 5 days waiting with immense anxiety about the build process and if I executed each step precisely. I am worried about if it will even post when I finally put power to it tomorrow night. I am worried about the bios updating/flashing. I am worried about setting up all my bios settings correctly. I am worried about installing windows correctly. And now I am worried about harming another component that I can't afford to replace. Thanks for giving me more anxiety Jay. LMAO! Hope construction goes well though. Super excited for you and your team!
I dodged a bullet there. I went from 7th gen to 12th gen, so I didn't have the choice to use my old CPU cooler. If I was upgrading from a more recent generation and saw that my cooler fit, I totally would have just used it and would have likely run into this problem. Thanks for the PSA, Jay!
2013, quite old... We have a Corsair H100 (CoolIT, not Asetek) which is basically working 24/7 since 2011. I'm quite sure that it's mostly empty, but somehow still works
It was just a poof and I think it might be bc it was sitting for like 2-3 months, not fully assembled or used, so I think some dust might have gotten on the fans circuits
Oh man, that would be a nightmare to figure out what was wrong and there is NO way I would have thought the cooler was too tight and I've been building my own, my friends and my family's PCs since 2000.
My mind was blown when I had sporadic POST issues when I built my old system and somebody on a forum suggested the fix of loosening the CPU cooler screws a bit and it actually worked...
I've got almost 25 years hardware experience and I don't think I would have thought of that for that error. I'd have been tearing my hair out on this one. I've reused parts in upgrade builds since the 90's and never thought of this. Thanks Jay.
Thats the first thing were i would look since reinstalling ios is a lot more work
I begin with plugging and unplugging wires and then gpu memory and cpu ...
haha same i got like 17 years and i think i got this issue with one of my rigs
I always fixed those kind of errors with complete reassembly without knowing that was the issue
@@Ivantonkic2 I hear you. That would be very frustrating, not knowing what fixed the issue. In my career when I come across something like this I make a mental note and file it away for future use.
Remembering that it is a common error on some platforms, in the old days when there was still the extreme series from Intel, if you pressed too much or left too much slack, some memory channels of the usual 4 simply wouldn't boot, you could use the system but without those memory channels. In simpler systems that have fewer pins it is something much more difficult to happen.
As someone who constantly repurposes older hardware in newer builds I found this extremely helpful information. Thanks Jay!
Did you play battletech under the tag Hammer for FPG?
Jay, I literally just finished adding a water-cooling loop to my system and was dealing with some weird issues and was worried that socket pressure could have been one of the issues. This video was released at the perfect time.
I had the same issue with my Asus H410M-, Corsair LPX ram with the Deepcool L360 ARGB with the tubes from the cpu block touching/pushing the ram
when i mounter the tubes coming off the cpu block other way(without touching/pushing the ram) it works fine
Tolerance stackups can miss during the manufacture of a cooler/aio...something that jaybeebee1cent doesn’t get. Yes, you will have to tweak around with torques to compensate. Start with low uniform torques, wait for boot and work your way from there based on the temps you see.
I had this exact problem a year ago after I built my pc so I brought it to a local shop paid for the diagnostic and they said that this was my problem
hey could you come back to update this? I have a watercooled build as well and thinking this could be an issue I'm dealing with and never thought about
@@Z3RK3R2020 At this point, I'm leaning towards this problem more likely being cause by an aggressive undervolt setting rather than the socket pressure being too high but I'll need some time for stability issues to present themselves
This is LITERALLY me this past month. I’d never built a PC before and I’m currently on my first try. Brand new cooler, motherboard, ram, CPU, GPU, power supply, everything! I ended up having the error 55 issue and taking it in to a local tech shop to help me out. We’ve been trying to figure out exactly what the issue is for a couple of weeks now. I even replaced my CPU and Mobo after figuring out it wasn’t my RAM that was the issue. Incredibly frustrating and you cannot find any information about this anywhere online. Thank you for bringing this to the attention of people.
For a long time, friends have told me that you want to tighten your cooler JUST until it starts to resist with normal pressure, and I just thought it was to avoid threading. Now I know better. That's incredibly scary for someone who doesn't have dozens of Mo-Bos and CPUs, and now I'll be even more careful. Thanks, Jay! I love these little 'how to take care of your parts' videos from time to time, because longevity of GOOD components is incredibly important, and I'm all about this careful maintenance stuff.
Yooooo just last week my cousin was having this exact issue after building a pc that we could not track at all.
We did reuse the aio cooler and we did our best not to fuck up the tightening but we're all human.
We even checked out your videos then to see if you already talked about something like this but couldn't find anything exact until now.
You're a life saver Jay, thank you so much!
I personally would've swapped the AIO back to the original to begin with as that was the only recent hardware change. If it were a brand new system build, I can see how this would be frustrating. Excellent tutorial as usual jay!
And this is why, if you ever have multiple upgrades planned for a build, you should install them _one at a time_ and then _test them thoroughly._ That way, if something goes wrong, you know which part caused the issue.
@@TheRogueWolf Only problem is if you're making a new build from scratch
Thats why we get it to a testing state before throwing it in a case
Sort of funny as I had an issue yesterday. Swapped a 120mm AIO for a Hyper 212 Black. Finished up and the PC wouldnt boot with both sticks of RAM. Ended up being the Hyper 212 bracket doesn't work with my motherboard, as tightening it adds pressure. Swapped the AIO back and boom it ran fine.
Great video Jay. I used to see this all the time when I was a SunMicrosystems hardware engineer back in 2001-2006. Granted, a different platform but the principle is the same. If you would over tighten the CPU module onto the system board of an E4500, E6500, or tighten the torx screws in the wrong order, the host would never pass POST. POST on these Enterprise beasts was quite a process. We were issued with a torque driver precisely calibrated for the correct torque to guard against this. Love the office refurb too. Great channel. Please keep it coming! Best, Jonny
I love these troubleshooting videos. The more bizarre the issue, the more satisfying the final fix. Keep 'em coming!
I use to love them before i start to build PC as my job,i had so much weird things happen including exactly this what is happening in the video,i'm looking at this video and literally shaking my head.
This ^, have also the same thoughts
Thanks for this video. I've never had this problem myself in spite of the fact that I've build well over a hundred gaming rigs over the past 30 years and have fitted air and liquid coolers of many different brands including some that needed to be modified in order to fit. I never tightened any of them down with a lot of force, though but always just enough so they wouldn't move.
Now that I am aware of this being a possible problem when fitting an older cooler on a similar sized socket mount I'll be able to use this knowledge when troubleshooting problems because as you said, your first thought when you get errors like this is a problem with the RAM, once you eliminate that the motherboard, once you eliminate that the CPU, once you eliminate that the power supply.
Your first thought is not going to be "Hey maybe it's the CPU cooler."
This is a great reason to include stock coolers. Even if it's just the most basic one.
I've always prefered coolers that use springs for tensioning.
I experienced this very frustrating issue on my second build ever. Basically had to remove everything and then reseat everything twice to fix it. Crazy
Great job - I would have never suspected cooler over-tightness to be the problem
The "car noise" was such a good analogy/example!
I hate the stupid car noise!🤬😆
I had similar symptoms with a i5 2500k and a z68 long time ago when installing a HYPER212EVO. Unscrewing the HYPER212, reseating the cpu and carefully centering the cooler while cross tighting the screws fixed my issue. It all depends on how the manufacturer decides the pressure amount their product needs to apply on the socket to be efective
Oh shit, same hardware as me lol
This happened to me when I installed the CPU contact frame, two weeks ago. I followed the instructions from the kit and videos I watched. It appears that the 0.6Nm was too much for my setup. 0.15Nm was perfect and thermals are also great (no more crashes). It took me a couple hours to figure it out, but everything is perfect now. Thanks for your awesome videos, Jay!
I never knew over-tightening of one's CPU cooler was an issue.
Also enjoying the changes to your workplace thus far.
The title has me worried that I overtightened my components now 😩
Sus emoji
We only live physically once so we shouldn’t be afraid to do anything bro, i smoke weed on my TH-cam channel and i ate Mcdonald’s inside Walmart 😈😈, screw anyones opinion 👌👌
HONESTLY SAME
Every single rookie in the hobby do that, but its okay!
@@User-JT89WA1 🤨📸
That's exactly what happened with my AIO! I overtightened it and every so often it would just blue screen. After hours of nearly tearing my hair out as I troubleshot, I tried loosening the screws on the pump and voilà! No more random blue screens. After that, I just tightened the cooler down just until the cooler doesn't slide or move on the CPU when you try wiggling it around.
That was the method I used too and haven't had any problems. Haven't built a PC in a LONG time so was totally unsure of how torque would work with the springs in the mix. I settled on tightening the screws a little at a time in a criss-cross pattern until my heat sink quit sliding/moving with reasonable pressure from the sides and left it at that.
My first thought would definitely not be to loosen the aio... wow... this was enlightening... thanks Jay
LGA would be fine if they designed them as removeable modules that have 10 micron tolerances, because then it wouldn't cost as much as a motherboard to replace a socket.
Mix match aio hit diff 😅
Your biggest hurdle with what you suggested is that you would have to make sure the socket is mounted as perfect as humanly possible, or else it may not make proper contact with the motherboard circuitry and nothing will work at all and could possibly brick the socket or the board or the cpu or anything plugged into it, but, if you could make a system where this wasn’t an issue, hypothetically, I’d be down for a replaceable socket, but it’s not like you could change generations though, cause the board layout for a 10th gen Intel is not the same as the layout for a 12th gen Intel and they would not be cross compatible even in the slightest
@@What_In_This_World Like I said the tolerances would have to be within 10 microns, anything much larger and the actual connections might not always line up, however if all LGA sockets were designed for this concept from the ground up we would be in a much more reasonable place because it would make cheaper boards $60-100 reasonable to repair with a 20-30$ socket, currently replacement socket prices are $5-10 but the labor is upwards of $100.
@@dagarath I was going to say that sockets already are replaceable modules. It's just that it requires soldering. But your last comment cleared that misunderstanding when you referred to it as currently being a $5 socket and $100 of labor. So that's fair enough, but:
How would you securely mount this modular socket to the motherboard? Surely it would require a retention mechanism of it's own if you're not soldering the socket onto the motherboard. Adding an extra retention mechanism seems counterintuitive to me if the problem your trying to solve is the potential cost of replacement.
I don't think I have ever over tighten anything in the 27 yrs I have been working on PC's. Thanks for the video, good info.
GUYS, Jay, you have no idea how much I have been racking my brain for the past 3 weeks after my mobo upgrade trying to figure out why my pc was unstable! I never considered that the pump was screwed on to tight! This video was a saving grace lol thanks man!
Thank you Jay! I had an 12600k build for a client that would boot but bluescreen' randomly after a few hours. Well I loosened up the cpu cooler per your suggestion and it appears to be stable and has passed multiple stress tests! Glad I watched your video because it definitely saved me hours of frustration.
never in a million years would i have guessed the cpu being overtightened would lead to a memory error. Great info jay. keep that shit coming!
it aint the overtightening per se its the bending of said overtightend cpu/MB/socketr etc :D if you for what ever reason would have a Motherboard that has a 2cm steelplate stabilizing it and your cooler plate is also a 2cm steelplate with a small insert in the middle out of copper and we asume you can tightening it in a way that those 2 plates are perfeclty plane to each other, you can pretty much squeeze to the point where the silicon is almost crumbling(or the ihs starting to get flatend out like in a hydraulicpress) and wont have any errors what so ever
I'm the same as a bunch of others here. I too wouldn't have thought to think about the cooler tention. Really interesting. Always something new to learn. Thanks Jay.
I loved that sponsored segment, I could watch it all day.
@MrВeast why the stupid fake acc?
Fascinating ... I've never had such a problem but I'll add this to my checklist if I ever get a problem in the future.
This is why I've been subscribed to this channel for years, it's interesting to get troubleshooting tips from people who build systems on a regular basis.
Love the educational videos. And the "What not to do" videos too. Learning from mistakes is the best way to really get that lesson in. Thanks, Jay! :D
(To the viewers) You can also come across a situation where all your RAM sticks are individually working and each memory slot on your motherboard are individually working but as soon as you install a pair of memory stick where it activates the dual channel memory, now your system becomes unstable and crash. That's a thing and it's important to remember that the memory controller is in the CPU nowadays, among other things.
holy hell Jay, thank you so much for this video. My PC has been having that exact issue where it totally locks up and I have to hard reset it. I've been going batpoop crazy trying to figure out what has been causing this issue since April-ish and now thanks to this video I may have the fix. So again thank you sir because you might have just saved my sanity and wallet.
YES! Biggest problems I've seen people do with tech is WAY too much torque (this is not a car tire going down a bumpy road.) Than after that absolutely no focus on torque balance of things with multiple bolts. I had actually bought a micro torque driver not just to speed up the torque balance procedures but to show how far off the rails customers were with their DYI upgrade. I'd bet my last dollar that if you tested this, you would have found two corners relatively the same torque and than one corner the threads are barely grabbing while the opposite corner was suitable torque for a car part. Like a see-saw it will jack the CPU cockeyed into the socket.
Played Jay at 0.5 speed while he was giving codes... It makes him sound like he got into the Elijah Craig
This is immensely helpful. I built my first computer almost 20 years ago, and been regularly building since, and never once thought about this as a troubleshooting step.
Great info Jay. It never occurred to me that the Z-axis might be different from one socket to another that are otherwise compatible. I'd love to see you test a hardware fix such as adding washers to raise the height, which I would think would give more even pressure than simply loosening the screws.
Brings me back to the launch of Skylake, when Scythe had to update their coolers with thicker metal washers.
The Z-axis was actually the same (or quite close), but the substrate was thinner, so high pressure mounts would bend it.
This is going to help a ton of people, especially with how the new 12th gen has changed the Z Axis height... I noticed that one of my Corsair AIO manuals explicitly says to not over tighten the screws now. I guess they are aware of this issue, but I honestly just heard about this issue from one of your earlier videos. Now when I install a cooler that has the spring loaded screws, I make sure to not bottom out the screws. Just let the spring tension do the work.
We need a computer torque wrench
Thank you so much for including two codes today. That was very cool of you to follow through.
I had the exact same issue with putting a Corsair H50 on a socket 2011 E5-2680 v1. The original heatsink was allowing the CPU to idle at 75°C so I decided to put an older AIO on it. I freaked myself out because after restarting the system it would not post. Turned out loosening the screws about 2 full rotations each fixed the issue. It now idles at 36°C.
Can't wait to see the finished product and footage from Saturday's garage sale!
As a new first time PC builder...I have similar issue happening where my computer will just lock up slowly and the only way to turn off the PC is flipping the PSU power. I also had an issue with my DRAM light coming on (within 30 days of me building with brand new RAM) It locks up rough 2 times a month, randomly. Maybe it could be me over tightening the cooler. Thanks for the info!
Yeah check the CPU pins, if one of them is bent. do take it to a professional
@@GamingWithHybrid they are definitely not bent.
@@GamingWithHybrid - Last system I built I picked up most of my parts at a MicroCenter 2 hours away and when I got home I dug out the CPU and it was floating around in the box completely loose, lots of bent pins (at least not sharp bends). I was able to straighten them all out by taking a box cutter blade (thin, strong and long enough to span an entire row of pins) and VERY gently and patiently wiggle the blade back and forth in line with the row, gradually straightening out all the pins on the CPU after a bunch of passes. Not for the faint of heart, though.
Thanks Jay, helps me with my personal builds and issues that I may potentially see at work in IT. Just more tools under my belt.
Jay "forgetting" codes, so that people watch his videos twice looking for the code.
Ha, exact same thing with my 5900X and custom waterblock. Exact same symptoms, and memory related problems due to overpressure on the socket. Switched block to a heatkiller and noticed they had a different mounting mechanism, no more problems, and memory overclocking, etc all good.
Great video Jay! I was trying to help somebody in PCMR subreddit the other day, and never saw a resolution. Wonder if this may have been the issue? I recommended testing all RAM one stick at a time, and they all passed individually, but not when all installed. I feel kind of bad, because I suggested it might have been the memory controller or the motherboard RAM slot/trace.
My wife got me an ifixit pro kit.... she is the best and she showed an interest! Thank you
Always excellent tips and outright mental tools explained so well! Especially for someone just starting to learn! Thank You Jay and crew!! Love the humor also!
indeed, pretty much the most straightforward pc guide youtuber.
Very interesting. Too much pressure on the CPU not the first thought that comes to mind. These troubleshooting videos are very cool and helpful. Thanks.
I had a similar issue but while overclocking my i7 4770k. I couldn't get the damn thing over 3.9Ghz (which is the max boost clock for this CPU) without getting random blue screens or even crashes while just booting up, even with a lot of voltage and good temps.
Recently I changed my thermal paste cause I didn't do it for a few years, temps were a tiny bit better but I think that the cooler tension was the actual problem since now I can get the CPU to run 24/7 at 4.2Ghz without any problems.
As someone who recently acquired a second-hand 4770K/Z87 combo, what’s your voltage and other settings like?
@@veeli1106 I'm currently in Malaysia without my PC so I can't answer to that rn.
Once I come back home I will check that.
I have an old PC that I had repurposed with Linux to run World Community Grid and I was getting totally random crashes just like this but I could never pin down where the problem was. Every single test I ran came out fine. I had tried multiple distros thinking maybe I had gotten a bad build. I decided to try with my old Windows 7 install but still had the random lockups. Watching one of your earlier videos I had decided to redo the CPU paste on my current rig and figured what the heck I'll do the old rig too. The random crashing went away and now, full circle, I think I finally know why. I've always tightened any screws or bolts in a cross or star pattern but I hadn't thought about how much torque I was using.
This is why I like your channel. Because of watching your content I finally started overclocking and trying other things I was always interested in but just never pulled the trigger and because you're a tinkerer like me you figure out such random problems that someone might not have even known WAS a problem. It's those tiny bits of obscure knowledge that are truly helpful to have.
Thank you for doing what you do.
jay really tried to tighten them loose screws huh 🤣🤣
DUDE! You are a God! Re-pasted my cooler & added another 16GB of ram to my laser engraving PC this morning. It booted then began to do random restarts followed by BSOD. Only way I could get it to run was to remove the new ram. So while I was engraving a quick test piece just to be sure the problem wouldn't reoccur I sat & watched your vid on my phone. Then when the test piece finished I loosened the cooler & reinstalled the new ram. Wooo Hooo! 32GB of ram & a stable system. Sweet. Cheers Jay.😀✌
Awesome video, saw it all in 3 minutes
Yeah, never try to tighten screws to the maximum on cpu/heatsink frames. Had this issue on my very first build like 15, 16, 17 years ago. It took me like 2 weeks to figure out what's wrong, and a lot of new ordered parts to send back. I could replicate the error and it felt really satisfying finding it out. Especially with Intel 12th and 13th gen i would be even more careful to damage something, as the whole cpu tends to bend easily without the usage of a third party cpu frame. And Intel doesn't care. 😑
Does this also apply to the amd processors? Because I also sometimes have random bsods and or game crashes. Not to mention the cpu (ryzen 9 5900X) is hooked up to a 360mm water cooler from nzxt and still gets up to something like 90 degrees C (I use kryonaut extreme paste)
i have a overclocked 5900x on a 240 AOI and my max temps is 70c running cinebench so you deffo have a bad mount
@@ml33cg Alright, I've just loosened it up slightly and will see what it'll be
I finally got around to turning XMP on after you saying it a dozen times. Everything was working great but my system would have a problem shutting down. Every once in a while after clicking shutdown,Windows 11 would stop normally but everything else would stay on. I had the latest BIOS installed, Windows was completely updated. Had no idea where to start other than googling the issue. Even though Windows said it was all up to date, the Intel Management Engine driver was two years old and I don't remember how many versions out. Got that updated and everything is working fine. In the 30 or so years I've been messing with computers I've never had one not shutdown like that. Fun learning experience.
One reason i love AMD mounts, that lever mechanism is just like "hey, THIS is the tension you need"
AMD is going from PGA to LGA with the AM5 Socket, so the same issues can occur.
That’s why I didn’t like when I built my first Ryzen system last year I found the new wraith stealth stock cooler with my R5 3600 didn’t use the included heatsink clip brackets on my motherboard but required I unscrew those and remove them and screw the heatsink into the backplate. And the same with my Hyper 212 Evo, even though the older and smaller Hyper TX4 for my previous system included and used those little lever clip hold down things. Screws allow you go tighter most of the time, sure, but sometimes too tight, and screws can strip, both their threads and the head, especially Phillips screws. I prefer the lever mechanism, reminds me of old computers anyway, even an old Celeron 600 MHz cpu used that mechanism for its tiny heatsink, pretty much all computers used lever mounts in the beginning.
Jay would be an amazing tech teacher because honestly when I was in school they thought it was bad drivers or even the memory channels never did they think to loosen the aio
Thank you, Jay, for the double codes. When you have something like and you're the only one experiencing it, it can make you question your sanity.
The mechanic and the noise thing hit right in the feels man. Just happened to me as well. Sadge
I'm glad I got smart and my NH-U14S has tension springs and all it needs is a few turns to tighten
I never would've thought of this! Mind you, I fixed a friend's computer the other month which kept power cycling during boot and we changed the psu, psu didn't fix the issue. After 20 minutes I unplugged the dvd drive and that's what was causing it. It honestly looked like a power supply issue but now at least her 10 year old setup has a new power supply. It's literally the nicest thing in her pc
UM OKAY... WAIT... Is THIS my issue??? My Kraken z73 has an LGA 1700 bracket for my 12900k... But this is like my EXACT complaint, AND I'm new to building. I've spent WEEKS trying to find an answer and it's been so difficult to even DESCRIBE to anyone because it's so intermittent and random.
Thank you SO much for this lead. idk where I'd be without your channel.
You know what I like about this the comp? You can tell only the hardcore are still participating, if you divide your entries by the total and have been in it since the beginning, you should fine your odds have been slowly increasing.
I'm speechless I've been troubleshooting my random blue screen crashing for weeks now on my 12900k and literally 2 days before this video I concluded it was a CPU issue. Perhaps the new architecture isn't mature enough or I got a bad combination of software with bugs in windows.
I immediately did a half a rotation unwind on the mounting screws and my rig hasn't crashed yet.
I'm so happy thanks for making this video.
I had a similar issue years ago on a X299 board that was open box, one ram channel wasn't showing up. I reseated the ram and flipped them around a few times and the system wouldn't see all the channels. I then took off the cooler and CPU, then looked at the pins with a magnifying glass. One pin was slightly out of position by like half a millimeter. I very carefully moved it into it's correct position with a pair of tweezers and put it back together. Viola! The 4th ram slot showed up fine. Never had a problem with it as long as I had it.
Oh brilliant. I've started getting random bluescreens on my Intel 8700k rig that I use exclusively for streaming now from time to time. Thought it might be degraded liquid metal that I applied 2 years ago after delidding, but temps are fine, and well within normal ranges. Was starting to look at memory, but now I'm going to go check my air cooler mount to make sure it's not too tight! Thanks jay!
Thank you, I've had similar issues for so long till I ended up with the same conclusion.
I built my first system back in March. Used the AMD stock cooler and literally from day one have always had this issue or memory issues. For the life of me I couldn't figure it out and asked everyone including forums, microcenter employees and my brother In law who's been building computers for years. Then I saw this right before I gave up and decided it was the CPU ( I have already replaced my motherboard). Tried it with stock cooler and still couldn't get it. Bought a cheaper single tower air cooler and immediately fixed everything!! I can't tell you how relieved I was.
I found this out the hard way not very long ago. As I was doing some upgrades to my old Asus P6T, LGA 1366 system and I ran into some crazy stability issues. I went from a I7 930 and a very old corsair H100 AIO to a Xeon X5690 and a custom loop using an ek supremacy evo block. But I couldn't find a compatible mounting bracket. So, I modified my old H100's mount to take the larger studs from the evo block's 115X mount. I didn't take into account just how much thicker that Xeon X5690 is over let's say an I7 8700K or realize how the additional clamping force would impact anything. To make a long story short, I wound up backing off the screws nearly 2 full turns from the bottom stop position and it works perfectly now. Thanks for this information, Jay. I only wish you could have posted this much sooner.
I actually just learned that you can over tighten and AIO mount, my PC was having all kinds of skipping issues and locking briefly. I also experienced higher temps nothing bad but not good for the AIO I have. So I thought, let me un tighten it once i did that. System now runs smoother and cooler temp wise. love the vids JAY keep! long time fan!
Thank you JaysTwoCents!!! i just fix my system now after one year failure
This is literally me right now, with a 13 900 KS, top-of-the-line, Asus motherboard, and my second set of 128 GB of RAM! Praying that this solves it
As someone who worked in the PC repair business, I have seen this numerous times when someone builds a PC using an AIO cooler. People don't realize that they only have to tighten it down until it's snug. They think they have to tighten it down until the screws stop turning. That was especially common in the era that the AIO Jay used in the video was from.
BTW, I have a 12 year old Asetek 240mm AIO, I believe it's a 570 LX, that is still going strong and cooling my rig running an i7-875K. I use that PC to run some older retro games that won't run on my current desktop.
I cannot tell you how many times Ive had this issue. Now I know what to look for should it happen again. Thanks Jay
Even for an experienced builder this is very good info to keep in the back of our mind if we are having issues.
This video has taught me some more important and often overlooked complexities of cooling and general closeness of components like pins on an IC/CPU. Even a tiny shutdown is cooling the system enough to create some major failure but maybe only temporarily. And it cannot be expected that every computer user knows this. This video stands for itself and it's usefulness. Even if what is being taught (expansion and contracting of objects based on heat and the general proximity them in such a tight space.) is actually pretty common knowledge, the presentation of it in the way that Jay has lends such clarity. The shit out of it.
This video brought back a memory from when I did this on pentium 4/LGA775/stock cooler. It was really easy to over torque the downdraft cooler and then it would not post. I had just bought a new motherboard because the original(prebuilt compaq presario) motherboard caps blew up and was convinced there was some OTHER major problem like a DOA board or a psu issue. I didn't really understand what was going on, and felt like a genius when I finally figured it out.
releasing the screw tension on the cpu cooler helped my weird freezing issues in windows. i was going crazy! thanks jay!
I had a P4 that was in a tower case and I put a Zalman solid copper fin heat sink on it. You may remember it. It was fins pressed together about 4" spread out. The thing must have weighed a pound. Well, in a tower case, that eventually tweaked the mobo enough. If I hit my desk just a little, REBOOT. Took it off and all was good.
I had this problem at first. I tore the machine down twice and it "fixed on its own". This might have been my issue.
I had the same issue when I used a smaller drill to screw the AIO on. I had random blue screens and instability. I realized the cooler was over tightened when I thought to reseat the CPU.
After a couple weeks my buddy was building his pc and had the same issue, I told him to back the tension off of his CPU cooler, sure enough it ALSO fixed the instability! This issue isn’t talked about enough! Lol awesome video
good point... not considered it. thanks for the vid and info
I had this same problem about a month ago and the kicker for me was that I'm using an EK block that has springs. As soon as I remounted the water block and left them looser it fired right up with both sticks of ram. Now I can make the problem happen on command by tightening the mounting screws. I tested this just to make sure I wasn't crazy and this was actually the issue since the remounting process might have been the thing that fixed it.
Been building PCs for 10 years and this is very new to me. Thanks Jason
Had this exact same problem on my x99. I spent months swapping out parts trying to find what’s wrong, finally found someone talking about this on a 10 year old OC form. Glad someone made a video on this.
I'm glad this video was posted because I have all these exact same issues but I just have a 5600x with stock cooler. Sometimes won't post or will freeze after a few minutes or the shutdown screen also.
I love this video where something specific, rare, and yet help you understand the differents element that are implicated!
This is incredibly helpful for future builds. The more I know, the better. 👌🏼
As newish builder, who has dealt with some real hair pulling issues like this already, these videos are wonderful and I really hope you make more of these even if they are in shorter form.
12:58 Jay, think about what you said there, it's the BIOS, you don't need RAM to access the BIOS, it's not how Windows accesses the RAM, but just the fact that it does...
10:07 I used to work at a cellphone store, the same thing happened with customers' phones as well. Some even claimed it works fine at the store because they know there is a cell tower in the store.
Thanks for the video. Another thing to check when getting the memory error.
The thing is this isn't an issue for just desktop coolers, industrial grade server coolers for those xeon scalable cpus have a) torque order on the screws b) max torque on the screws. And I've had the phantom ram or pcie slots from the wrong torque on a socket with 3k USD cpus.
Just built my first system this week, but I am still waiting on my PSU that is coming in tomorrow. I have had my system sitting for like 5 days waiting with immense anxiety about the build process and if I executed each step precisely. I am worried about if it will even post when I finally put power to it tomorrow night. I am worried about the bios updating/flashing. I am worried about setting up all my bios settings correctly. I am worried about installing windows correctly. And now I am worried about harming another component that I can't afford to replace. Thanks for giving me more anxiety Jay. LMAO! Hope construction goes well though. Super excited for you and your team!
I dodged a bullet there. I went from 7th gen to 12th gen, so I didn't have the choice to use my old CPU cooler. If I was upgrading from a more recent generation and saw that my cooler fit, I totally would have just used it and would have likely run into this problem.
Thanks for the PSA, Jay!
2013, quite old... We have a Corsair H100 (CoolIT, not Asetek) which is basically working 24/7 since 2011. I'm quite sure that it's mostly empty, but somehow still works
Uuuh is it bad that when I started my pc that I built, some smoke came from a ventilation fan? Everything is working fine but I'm worried...
It was just a poof and I think it might be bc it was sitting for like 2-3 months, not fully assembled or used, so I think some dust might have gotten on the fans circuits
Had the exact same issue with my first pc build never figured out what happened till seeing this
Oh man, that would be a nightmare to figure out what was wrong and there is NO way I would have thought the cooler was too tight and I've been building my own, my friends and my family's PCs since 2000.
Me too.
I am So Glad you made this video. I had this issue and it was driving me nuts!!! I even changed the motherboard...
My mind was blown when I had sporadic POST issues when I built my old system and somebody on a forum suggested the fix of loosening the CPU cooler screws a bit and it actually worked...