TeaJayAye

[Deleted double post]

Two sticks of 8GB DDR4 rated @ 3200Mhz 36-18-18-18-16. I swear this computer is haunted. I've never had to do so much trouble shooting with a build. If you're interested in the back story, her it is. My friend picked up all new components (other than a PSU and the pair of RAM sticks mentioned above, which came out of a working system without issue and have tested good since). We put the system together, and everything was fine other than the crazy CPU temps we were getting with the stock Ryzen 5 3600. I recommended him a better air cooler, but he didn't want to wait for the long shipping times, so he bought a Hyper 212 EVO from a local retailer that was doing curb-side pickup during COVID. Shortly after I installed that cooler, the system started shutting down randomly. Whether the system was stressed or not, it would just randomly shut down. After a couple of times shutting down, it would fail to post. I figured it must be the PSU that he carried over from an overpriced, out-dated, POS computer he bought off of the FB marketplace. All the other parts were either new or known good and the 12V rail of the PSU was putting out less than 10V on most software measuring tools (I know those aren't perfect, but I didn't have the hardware to test the PSU). I swapped out the PSU, and the issue persisted. I then tested his CPU, RAM, and even his SSD in my own system. No shutdown/boot issues. We exchanged the CPU (thanks Amazon) and the issue continued. We then exchanged the motherboard (with an upgrade from a B-450 non-WiFi to a B-550 WiFI - both were Asus ROG Strix) and the issue persisted. Each time I swapped components, I double checked all connections. I then did a repair of windows. Then, I did a fresh install of windows. No fix. Then I noticed when the system wouldn't post, it kept giving an LED notification that the issue was with the RAM. I thought that was odd, knowing the RAM to be good, and having tested in my system, I started moving the sticks around and trying different configurations. That's what solved the random shutdown/no post problem. I figured maybe Asus' MOBO manuals were just incorrect about what slots were primary. The system appeared stable in short-term stress testing (an hour each) with P95 and RealBench at stock CPU and RAM speeds. So, I figured, let's get the RAM to its rated speed and leave well enough alone. That's when I realized I couldn't overclock the RAM at all. Now, for some odd reason, the system will post/boot with RAM in the slots that the MOBO indicates are primary. I'm running a windows memory diagnostic on that RAM as I'm posting this 85% complete without errors. Then, I'm going to run some other stress testing on the CPU/RAM and see if it blue screens or shuts down. If not, I guess I'll try to overclock again. I swear this computer is haunted or my buddy has the worst luck of all time. I've built at least a dozen computers for myself and friends over the last 5-6 years, and I've never had this much trouble with a build. I'm no expert in this stuff, but I'd like to think I'm fairly good at trouble-shooting. I'll report back if I encounter any other issues, but the problem seems to have just disappeared. Maybe I just didn't seat the RAM perfectly last time I moved them to the other slots, but I think that's unlikely.

Yes, I believe the latest BIOS is installed, but I will double check. This is a B-550 board. I don't think there is anything but the initial release BIOS, which dates to late April. CPU and MOBO are brand new. No bent pins on the CPU. I always check, even on new CPUs. System has been running a RealBench test for an hour without issue with RAM in the non-primary slots.

After troubleshooting my friend's PC (swapping out basically every part into a known good build and exchanging the parts we couldn't verify - CPU and MOBO) to solve a random shut-down/failure to post issue, we're now experiencing another odd issue where the system won't post with known good RAM in the primary slots according to the MOBO manual (B2/A2). MOBO manual says slots are laid out left to right like this (B1/B2/A1/A2). I find that the system posts without issue (and passes stress testing) with a single stick of RAM in A1. Both sticks work in A1. I also find that the system boots with sticks in B1/A1. You might be thinking, okay, well then what's the problem? Just use B1/A1! Well, with these slots utilized, I can't get my buddy's 3200Mhz RAM kit to run at its rated speed. Knowing that even 2nd Gen Ryzen (he has a 3600) is RAM frequency and timing sensitive, I don't want to leave him with a machine that is under-performing. You might be wondering why i think the RAM slot issue and the RAM OC issue are related. Reason is that I'm seeing other posts online relating to not being able to OC RAM on this motherboard when the DIMMs are in the non-primary slots. At least, after several weeks of troubleshooting, I have his system working. Now I just want to get the rated speed out of his RAM and let him have at it.

Solved. For anyone reading in the future, two Asus ROG Strix boards a B450 and a B550 wouldn't post with known good RAM in slots A2/B2, which the motherboard manual recommends for a two stick dual channel configuration. I ran a single stick in A1 and it worked. Then I relocated the fan on the Hyper 212 Evo to the other side of the heatsink and installed the second stick in B1. Worked fine. Going to run Realbench overnight to check stability with a RAM overclock. Man, what an adventure.

Well, we swapped out the CPU to no avail. Then we swapped out the motherboard to no avail. I unplugged the front panel connectors, just to see. As expected, that did didn't fix it. I swapped the CPU cooler from a Hyper 212 back the stock cooler, again just to see. Obviously, no change. Now, all that's left in my playbook is a Hail Mary. I'm going to do a fresh install of windows on his SSD. Cross your fingers for me.

Well, I checked everything except the motherboard and and CPU, but I have no way of testing those two components independently. My friend has ordered replacements from Amazon. My best guess is the motherboard is the issue here, but I'll report back for anyone searching for this issue in the future. The symptom was sudden power loss, particularly when stressing the CPU. The system would simply shut down. No BSOD, no system hang, nothing. Just a sudden loss of power and image. I tested the PSU, GPU, RAM, and SSD in my own rig to check, and could not recreate the problem. This was on an ASUS ROG Strix B450-F and a Ryzen 3 3600 with a Hyper 212 Evo.

Double checked his RAM in my system. It's good. Going to swap his SSD into my system tomorrow to see if that works. I'm guessing it's the motherboard, but he can RMA both to be sure.

Thank you. I now have his 2060 Super in my system, and it is running just fine. I'm writing this from my system with his GPU in it. That leaves the motherboard, CPU and SSD. I don't really have an option for swapping in another CPU or motherboard into his system, as both of my rigs are 8th generation Intel. Perhaps he should just RMA both the CPU and Motherboard? What a frustrating experience. Thank you again for the encouragement and kind words.

I helped a friend build a PC recently. It was an upgrade to an old POS system he bought off of FB Marketplace for far too much money. The only part we salvaged from the old system was a serviceable EVGA 80+ Bronze PSU. I also supplied him with some Corsair Dominator DDR4 out of a known working system. Otherwise, all parts were bought new. After building, we noticed CPU temps were out of control with a Ryzen 5 3600 on the stock cooler. I undervolted the CPU temporarily (for a few days of use) until my friend could buy a new CPU cooler. After installing a Hyper 212 Evo, the system ran much cooler. Shortly after the new cooler install, the system began to randomly shut down. No BSOD. Just a sudden loss of image and power, though the RGB LED on the motherboard never shut off. I restored the BIOS to stock settings, figuring maybe it was the undervolted CPU or something like that. No change. I then thought it must be the PSU we borrowed from the other system. Moments ago I swapped out the PSU and the problem appears to have gotten worse somehow. Now I can't even get the computer to post. Since the issue has occurred with two PSUs, and the RAM came out of another perfectly operational system, that leaves the CPU, Motherboard, SSD and GPU as the remaining culprits. I am admittedly VERY weary of swapping in my 1080ti to this system to see if it fixes the problem, though it would eliminate one culprit. My best guess is the CPU or the Motherboard. Any tips for solving this problem? My friend is understandably frustrated as a newbie to PC gaming, and I really want to be able to fix this problem for him. Any thoughts are appreciated.

Put your parts in your brother's computer. Does his system boot with your Ram? What about with your GPU? If so, both components are fine. This will help you narrow it down.

Do you have a friend or family member who may have the ability to help you test the components piece by piece? Maybe a friend will let you load your GPU or RAM into his or her system to test that they work? I think the only way to determine it's your motherboard is to eliminate everything else. Start with the easy stuff like the RAM and GPU and then test the CPU. If all of those components work, it may just be the motherboard.

Yeah, if you're sure everything is connected properly and the BIOS update was done correctly, it has to be a bad component. Dumb point, but be sure that your monitor is connected properly and is turned on.

Gotcha. Wasn't aware the new Ryzen chips didn't have onboard graphics. Make sure all the connections to your motherboard are correct. Make sure you haven't reversed the positive and negative connections for your front I/O. Are the components getting power? Double check all connections. Rebuild the computer if you have to.

If you're positive the mobo has the proper bios to support that Ryzen 5 3600, you may have a bad component. Will it post if you run with onboard graphics? Do you have other DDR4 you can try?