Jump to content

mikegray

Member
  • Posts

    172
  • Joined

  • Last visited

Awards

About mikegray

  • Birthday Aug 23, 1972

Contact Methods

Profile Information

  • Gender
    Male
  • Location
    Winterthur, Switzerland
  • Interests
    Husband, Dad, Pastor, Dog Keeper, Philosopher of Religion, Gamer, Crypto-Dabbler, Fantasy-Reader, used-to-could Crossfitter

System

  • CPU
    AMD Ryzen 7 3700x
  • Motherboard
    Asus ROG Strix X570-E Gaming
  • RAM
    G.Skill Trident Z Neo 32GB (2x16GB)
  • GPU
    Gigabyte GeForce RTX 3090 Gaming OC 24GB with Bykski Waterblock and active VRAM cooling, Gainward RTX 3080 TI with Bykski Waterblock and active VRAM cooling
  • Case
    DIY bench in a wardrobe
  • Storage
    EVO 960 500GB system drive, 1 TB RAID 0 SATA3 local Storage, 3TB hard drive, 6TB hard drive
  • PSU
    be quiet! Dark Power 12 1000 W
  • Display(s)
    Samsung crg9 (49"), janky USB thing mounted in the wardrobe
  • Cooling
    Custom loop, except for the new GPU. (4x BeQuiet Magicool 140mm Pure Wings / 3x SilverStone AP183 / Magicool 180 Triple / Alphacool VPP755 / Aquabox Professional / EK Supremacy Clear Acetal / Bykski N-GV1080TIG1-X with Backside Water Block B-3090TC-X
  • Keyboard
    HyperX Alloy Elite, German Keyboard
  • Mouse
    g502 Hero
  • Sound
    nano iDSD Black Label + Sennheiser X Massdrop HD 6XX + Creative Gigaworks T20
  • Operating System
    Windows 11
  • Laptop
    ASUS ROG Zephyrus G14, 2021
  • Phone
    Huawei p30 pro / Samsung Galaxy Tab S6

Recent Profile Visitors

1,472 profile views
  1. Old thread, but I surfed in with this exact problem - the fix above worked. In case anyone else surfs through, here's some more information about my situation: My system (with a new GPU, RTX 4090, bought on release) had been working fine since I got it - my Samsung CRG9 doing RGB/8 bpc / full / 120hz with no problems at all. (Resolution: 5120×1440, DP cable certified 2.0 - but 5m long.) About a week ago, I went to the kitchen to make coffe, came back, and the screen looked wrong. Sure enough: Windows had deactivated HDR, and Nvidia Control Panel would only allow YCbCr422 - no other options shown. Same as OP. When this happened there was no crash immediately before, no error messages, no games running, just Windows with a few Chrome tabs and Word documents open. HOWEVER: I HAD just had a bunch of stability issues with the next-gen version of Witcher 3 - so I issued Witcher 3 was at fault. I tried the usual stuff, no beans: Restarted the GPU (shift-ctrl-windows-b) Rebooted a few times Substituted a different DP cable, also high end. Plugged DP cable into a different socket. Then I ran DDU, reinstalled everything from scratch, even reinstalled Witcher 3 - seemed like the problem was solved. Today, the EXACT same thing happened again, EXACT same circumstances - the only difference was that I've had no more big problems with Witcher 3. (Hotfix + turning off all overlays solved my issues there!) I spent a while surfing around, found this thread, tried the steps above: pull out the Display cable with the monitor turned on and the PC turned on with it displaying a picture. Then turn off the monitor and plug the cable back in whilst the monitor was off, then turn the monitor on and for some reason this has now fixed it !!! Problem solved. I can get normal settings in Nvidia Control Panel again. My best guess is that the problem has SOMETHING to do with signal integrity. Here are a few thoughts for posterity - in case somebody else surfs into this same problem: I run high quality cables, and I've been using these two cables for about 2 years now with this exact same monitor, same settings, ZERO problems. However, the cable runs are VERY long - I think they're actually longer than DP norms stipulate. This could create a situation where otherwise minor issues could cause signal degradation - and possibly cause Windows and/or Nvidia drivers to flag the monitor as non compliant with whatever it looks for to run at full settings. Sort of like booting in safe mode? I ALSO have a new GPU, so maybe there are some variations in DP socket quality there, too - like you add 2 & 3 together and have a weak point you didn't have before. I'm assuming that the fix above forces windows/drivers to re-recognize the monitor. So ... if this happens again, I'll also try uninstalling the MONITOR in control panel - that might ALSO force the same effect.
  2. Hey, I messaged you on reddit - but I should update this thread for future reference. In fact, sorry Internet: I should have done it back then. The issue turned out to be the PSU - which was high quality and powerful but VERY old - built around 2013. It was a Cooler Master v1200 Platinum. I can't find the site/threads that led me in this direction, but I found some indication that some MUCH older gen PSUs may have had trouble with 30xx gen GPUs. I can't confirm whether this is true, and I can't remeber WHJY older PSUs would have caused problems (could POSSIBLY have been the difference between digital and analog) - it could ALSO be that my particular unit was just old and tired. Anyway, I splurged on a be quiet! Dark Power 12 (1000 W) and never had another problem of this kind.
  3. I've been meaning to a do a little video on a project I started back at the very beginning of the pandemic: A full on wardrobe PC - with the PC on the other side of the wall from my desk and work space. It's kind of difficult to explain, so I just took some super wobbly cell-phone footage, rambled along about it and whacked the video on youtube for your viewing (dis-)pleasure. If you're curious about my components, check my profile page or just ask.
  4. Hmm. I've been running the thing for years now with nary a problem. But I have Corsair rm1000 I could swap in. (Nothing from Gigabyte - that would be fun!!)
  5. P.S. I just discovered WhoCrashed, It gave me the following - which looks more like a GPU problem. Edit: I just realized: The system crashes reported below are NOT the ones I'm getting when gaming. I wasn't even aware of these crashes - I run nicehash at night, and it looks like it has crashed the system a couple times.
  6. Could you guys help me make sense of a recurring system crash when gaming in in Windows? My system specs are in my signature line and on my profile. (Basically: x570, 3700x, RTX 3090, 32 gig, high-end but old PSU) Crash - The system simply shuts down, EVERYTHING goes dead - keyboard. screen, etc. However, the RGB on the MB and RAM sticks glow as in system off state. (Slight glow.) - The only way to reboot is to shut off the PSU, wait ca. 30 seconds, turn the PSU back on, press the power button. Power button alone does not reboot. (Highlighted because I've never had a crash where I had to power down the PSU!) - Windows does not report any errors - event viewer just notes that the system shut down unexpectedly. Almost like I had power blackout. Occurrence - Crash happens once every 2-3 days, and ONLY when gaming - specifically, in RDR2. - It has never happened in any other test or benchmark I've run (like, Heaven, Furmark, Windows Memory Diagnostic) Substitutions, Process of Elemination - I did a completely fresh install of Windows - re-formatted the system drive. (Running 11 now - but I was having the problem in Windows 10, too!) - I use aquasuite to run my fans and custom loop, and I'm positive that the temps are fine! (I mine with nicehash when I'm not gamign or working, never had any trouble at all.) - I substituted the MB, CPU and RAM with an older Ryzen 1700 system. I kept the other system running for about two weeks, had no more crashes. This PROBABALY eliminates the GPU, PSU and peripherals - HOWEVER, I also didn't game all that much during this period! (Though I did leave rdr2 running for over 24 hours ... ) Suspicions - My motherboard (same specs!) suddenly died about 3 months ago, RMA'd with ASUS, they replaced, no problem. (Maaaaybe the new MB is at fault?) - OTOH, whatever killed the old MB may ALSO have damaged some thing else! - My first bet is the CPU. Unfortunately, I don't have another CPU I can substitue on this board. (The board won't accept my older Ryzen 1700!) - My second bet is the RAM. However: Windows Memory Diagnostic didn't show any problems. - My third bet is the PSU - it's powerful and high quality but also very old. Everything seemed fine when I switched in the old MB-CPU-RAM, but maybe that was just luck?
  7. So ... how would you guys proceed if some idiot had emptied an XL mug of coffe over your mechanical keyboard and into your brand new LTT Deskpad? Specifically, the deskpad. (Let´s just assume that the person in question has some experience cleaning keyboards.) Just wipe it down with a damp cloth? Use detergent? Chunk it in the washer with the keycaps? (Asking for a friend, obviously.)
  8. I'll be contacting them anyway - but since Monday is a holiday in my part of the world, that won't be till Tuesday. In the meantime: SUCCESS! I switched from SATA mode to RAID in Bios, DL'd some the MB's Raid drivers, installed RAIDexpert2 - and to my utter shock ... it WORKS!! I didn't expect it to be that easy. I'm not actually sure whether the Raid 0 array will mount automatically in the future or whether I'll have to mount it manually on re-boot - but basically, I'm a happy camper. This will make rebuilding waaaaay easier. P.S. Rebooted, and my D drive mounted like a champ. It's like ... PCs have become almost boring. This was supposed to be basically impossible ...
  9. Well, shoot! My MB died on me - and a very nice MB, too: An Asus Strix x570-e with a ryzen 3700x riding in it. I was running two 1TB SSDs in a Raid 0 array. I managed to remount the system on a Gigabyte x370 gaming 5 running a ryzen 1700. I'm able to get Windows going, but Windows isn't seeing the Raid 0 array. For everything that's mission critical, I have backups - but recovering it would be a major hassle. Two questinos: - What are my chances of saving the Raid 0 array on this MB? Basically ZERO? - I think I'll be able to get the MB RMA'd. Assuming they replace the broken one with the exact same model MB, will I be able to get my stuff back then, or will a different MB STILL cause problems?
  10. Thanks for both of your inputs, guys. I think I'll "stuff 'n zip" as a temporary solution, just to get the system up and going. I mean to order some parts from Aliexpress anyway, so we aren't talking about having to use it for that way for ... umm ... any more than it takes Aliexperss stuff to arrive ...
  11. I just picked up an old aquatuning aqualis reservoir for my wardrobe project. It came with fittings that would be perfectly serviceable - only they're designed for 13/16mm tubing, whereas the rest of my system is setup for 10/13mm. I tried hitting some 10/13mm tubing with a hairdryer, and I was able to work it onto the 13/16mm barbs; however, I'm wondering whether this is a good idea. On one side, the tubes we be on SUPER tight. OTOH, the collars wouldn't be able press into tubes to fix them in place, and I'd rather not have a garden hose spraying in my rig. I guess I could maybe use a zip tie? BTW, it's a g1/4 system, so buying 10/13mm fittings wouldn't be the end of the world. What do you guys think?
  12. 100% noise reduction - that's where the project comes from. I close the door between the rooms and it's like my peripherals run on pure magic. I have an air-cooled Gigabyte rtx 3090 mining 24/7 except when I'm gaming or rendering videos. I've since repacked the 3090 with better thermal pads, but for about a month every fan in my rig was running at full throttle and the noise level was spectacular. By drilling a hole and moving my case to the other room I brought the noise down to zero. The next step is to drop the case entirely and move the pc into a bench test case that'll sit in the wardrobe.
  13. One step at a time, this truly nutty idea is becoming reality ...
  14. You know, I - err, he - didn't see any such sticker.
×