X570 and AMD 3700x with possible ram cause bluescreened.

Discussion in 'Processors and motherboards AMD' started by Truemello252, Jan 11, 2020.

  1. Truemello252

    Truemello252 New Member

    Messages:
    3
    Likes Received:
    0
    GPU:
    Gtx 1070 of gigabye
    For the past couple weeks I've been tortured with plenty of bluescreens with different codes and cannot pinpoint why.it feels like it might be a clash between my ram and the cpu there is also no consistent point of crash accept that I'm gaming. Most of my data has already been posted on Tom's hardware so I will link that thank you I'm advance for your help!!!

    Moderator: links removed, if you have an issue post your data here on this forum and not point towards toms hardware forum. "Dont see why its removed when i just need help there is alot of information that I have to repost."

    CPU | AMD Ryzen 7 3700X 3.6 GHz 8-Core Processor | $327.20 @ Amazon
    Motherboard | Gigabyte X570 AORUS ELITE ATX AM4 Motherboard | $189.89 @ SuperBiiz
    Memory | Corsair Vengeance RGB Pro 16 GB (2 x 8 GB) DDR4-3200 Memory | $99.99 @ Amazon
    Storage | Samsung 970 Evo 500 GB M.2-2280 NVME Solid State Drive | $89.99 @ Amazon
    Video Card //Has a warrenty till 6/27/20| Gigabyte GeForce GTX 1070 8 GB Windforce OC Video Card | $479.99 @ Amazon
    Case | Cooler Master MasterBox MB520 RGB ATX Mid Tower Case | $85.99 @ Amazon
    Power Supply | SeaSonic EVO Edition 620 W 80+ Bronze Certified Fully Modular ATX Power Supply | $76.98 @ Newegg
    Operating System | Microsoft Windows 10 Pro OEM 64-bit | $117.85 @ Amazon
    Wireless Network Adapter | [Gigabyte GC-WB867D-I PCIe x1 802.11a/b/g/n/ac Wi-Fi Adapter]

    fresh install of windows and i just bluescreened while streaming monster hunter all temps are goos service exception was the bluescreen code

    some current stress test of cpu DURING FULL LOAD ON BIOS VERSION F11View: https://imgur.com/a/T1bd7pg

    RESULTS FROM PRIME 95 5 HOUR TEST:

    FATAL ERROR: Rounding was 0.5, expected less than 0.4
    x3

    Hardware failure detected, consult stress.txt file.x2
    They only started after 3 hours or so into test
     
    Last edited: Jan 13, 2020
  2. 386SX

    386SX Master Guru

    Messages:
    661
    Likes Received:
    698
    GPU:
    AMD Vega64 RedDevil
    Tbh I skipped most of the crash report because I had trouble to dig into that. Please use CODE tags and try to reformat it so others can easily read it.

    SERVICE EXCEPTION can happen with basically anything. Bad software (badly coded), bad RAM, bad CPU, bad PSU, ... you name it. But I guess from what I see on your screen shot temp may indeed be an issue. 92,xx degrees (celsius?) should be the issue. This is critically high!

    Did you try to unmount your cooler and reset it, with an eye on correct spreading of thermal paste in correct thickness? Or do you run a lower quality CPU cooler? Bad airflow in your case?

    Sadly I cannot read the screenshot, because on mobile if I zoom in I mostly see big pixels only. But I guess the 92 degrees in the middle left of this shot come from CPU, right?
     
  3. Clouseau

    Clouseau Ancient Guru

    Messages:
    2,426
    Likes Received:
    270
    GPU:
    ASUS STRIX GTX 1080
    If you are not going for a manual all core overclock, there is no need to ever run Prime95 on the 3000 series. They regulate themselves very well. AMD has truly killed off manual overclocking with these CPUs. One will actually get better performance from these CPUs by just letting them do their thing. Only thing to do is to set the voltages for the SOC and leave all the others be. With my 3800X all I have set is the SOC voltage with an offset of -0.01875 and the ram voltage. Overclocking the ram required me to play with the CLDO_VDDP, VDDG voltages. The CPU voltage is left on auto.
     
    Last edited: Jan 12, 2020
  4. GREGIX

    GREGIX Master Guru

    Messages:
    571
    Likes Received:
    54
    GPU:
    MSI 1080 /AMD v7
    It can be everything. But seems to me may be memory related.
    1. What is ur PSU?
    2. Are you run mems on xmp? If yes, disable it, leave all settings as it is after profile applied, set LOWER voltage for SOC and memory related settings. Xmp on all platforms set them way to high. Ram voltage should be as specified.
    3. I would try to run mems at 2400 just for sure just to eliminate this from equation, and run prime and some memory testing software. If u have errors with this, either is something wrong with them, or just CPU...which u can set as in post above. Personally I had on b450 and 3700x offset on CPU voltage, -0.1v. Temps went down, boost up...
    4. If this is memory related, but works on default (like2400 Mhz), find overclockers site with Ryzen Dram Calculator, read it, download sw, try it. Proper settings can solve your problem.
     

  5. SweenJM

    SweenJM Master Guru

    Messages:
    686
    Likes Received:
    356
    GPU:
    Sapphire 590 nitro
    Waaaay too hot. Get that under control and do some memory testing running defaults.....basically what they said.
     
  6. Kool64

    Kool64 Master Guru

    Messages:
    498
    Likes Received:
    181
    GPU:
    Gigabyte GTX 1070
    Is that 92c under idle?
     
  7. Astyanax

    Astyanax Ancient Guru

    Messages:
    4,449
    Likes Received:
    1,291
    GPU:
    GTX 1080ti
    is the latest bios installed (with a cmos clear)

    is the ram on qvl?
     
  8. Truemello252

    Truemello252 New Member

    Messages:
    3
    Likes Received:
    0
    GPU:
    Gtx 1070 of gigabye
    https://docs.google.com/spreadsheets/d/1d60x31rX-ALqM8YnWGx9MmesaQRWxCKjljuWEEjkekQ/edit?usp=sharing ===THIS IS A SPREADSHEET WITH THE CRASHES

    The test where my CPU went to 93c was during a prime 95 test full load on bios version F11 after my windows had critically crashed about 2 crashes ago I had to wipe everything reinstall windows and downgraded to bios f10 test became more stable around 65 now and won't break 78 but I did find that odd to be that high as well, I have been reading a lot of people have a temp like that during test even with a water cooler. But this system is at stock setting and would be crashing at this point it seems more stable I have only crashed 1 time on this install of windows while I was streaming and playing but my temps don't go past 66 while gaming and streaming no code either as it was a hard stop i guess.

    @
    GREGIX-I have a [SeaSonic EVO Edition 620 W 80+ Bronze Certified Fully-Modular ATX Power Supply] Not sure if this is a brick of supply or not but reviews were good.

    At this point I do believe my ram is what may be causing it as the CPU doesn't like it when the voltages go up this is a brand new pc built-in September last year and all the parts seemed compatible. will try a mem test with X.M.P when I get a chance ill to you to how it goes
     
  9. SweenJM

    SweenJM Master Guru

    Messages:
    686
    Likes Received:
    356
    GPU:
    Sapphire 590 nitro
    Still too hot! Are you using stock cooler?
     
  10. primetime^

    primetime^ Master Guru

    Messages:
    897
    Likes Received:
    12
    GPU:
    XFX Radeon RX580
    just relax the timings by a couple and see how it does...I have to do that with mine so i can go all year with zero crashs now..its not rocket science lol
     

  11. Dazz

    Dazz Master Guru

    Messages:
    865
    Likes Received:
    89
    GPU:
    ASUS STRIX RTX 2080
    That CPU is damn hot, looks like 73C idle and 93C under load none of which is ideal at most idle should be around 50C and 85C max i am asuming you are using a stock cooler possible the cooler is not making sufficent contact with the CPU. I would surgest removing the cooler and replace the TIM with some cooling paste and reseat the cooler. Before replacing the TIM check that there is indeed contact.
     
    GarrettL likes this.
  12. GarrettL

    GarrettL Active Member

    Messages:
    83
    Likes Received:
    43
    GPU:
    Asus Strix 2070S
    My 3800x would hit mid 70’s with the stock cooler and idle in the low 40’s.
     
  13. Truemello252

    Truemello252 New Member

    Messages:
    3
    Likes Received:
    0
    GPU:
    Gtx 1070 of gigabye
    I'm going to run the Prime95 test tonight to see if it will go that high again but I'm sitting at about 45c idle. I have not gotten a crash yet but I'm running X.M.P on the ram been streaming fine but I tried to mess with timings and I kept getting set to the factory after the 3 bios restarts and this is with the calculator safe settings. XMP has ram timings at 15 15 15 36.
     
  14. bobblunderton

    bobblunderton Active Member

    Messages:
    76
    Likes Received:
    33
    GPU:
    16g 2400mhz (2x8)
    100% too hot for the CPU. Never ever should crest 85C with the stock cooler.
    I've had my 3700x for 6 months and the hottest it's gotten is 83C, when it was 75F in here, and I was running the cpu at max for over an hour with physics stuff. NOTHING in properly built PC should ever hit 90C*... aside of maybe rarely some very, very, very busy voltage regulators / chokes etc...
    I am using the stock cooler.
    Make absolutely sure PBO is OFF. That will drop a few degrees down especially under load.
    Prime95, while a good test, will really make that CPU *HOT*. It's a bit unrealistic, but a stress-test is a stress-test. I can run it on mine here fine.
    Make sure your BIOS is up to date on that board, a rushed release-day BIOS could cause an issue like this with keeping the CPU boosted longer than it should by feeding it too much voltage, and not cutting back soon enough. Use the in-BIOS updater, not a Windows one, you'll need an empty USB drive.
    Better thermal interface material will help shave off maybe a few C max... so consider buying a bigger cooler if you cannot seem to get the heat under control and some good Thermal Grizzly konductonaut (spelling on that?). Don't need the best cooler, a medium sized 12-series Noctua will do plenty, you can sometimes score one for under 50$ depending on sales, stock, market conditions (haven't checked lately).
    You shouldn't need a better cooler though to keep under 85C, so I'd definitely go over those BIOS settings with a fine toothed comb.
    Do not artificially set voltage lower in BIOS on Ryzen 3xxx, it will harm performance / boost ability depending on how much you change it.

    *80~85C is generally the warmest you ever want your PC components to ever hit on a regular basis.

    -Best of luck
    If anyone has corrections to this; please cite the source so I can learn something, too. I'd offer you a cookie for doing so, but the dog and I ate them all.
     

Share This Page