Here's simple chart for GPU mining temperatures

bakibaki Member Posts: 26
hey, I found it on dodgecoin reddit (thanks to dogebiscuit):

Googling didn't give me a simple chart, so I'm going to make one. Nobody has a definitive answer, so here is an attempt to create a broad understanding what temperatures mean based on many hours of research.
Please Note: All GPU have different safe ranges. This is an attempt to generalize temperature ranges for mining any coin in general. Because I may say a temperature is generally safe, doesn't mean it definitely is.
0C - 29C You might want to turn your computer on first!
30C - 49C Idling. No major tasks are running. Are you sure your miner is even on?
50C - 59C GPU is idling in a warm room, or GPU is underclocked. Might be able to crank a bit more power.
60C - 69C There's action happening. GPU is processing at a very acceptable rate!
70C - 75C Okay things are getting really warm. High graphics gaming or optimized mining. We shouldn't get any hotter.
75C - 79C Alright, this is questionable area. Some GPU can handle this range easily, but I'd rather play it safe and back off.
80C - 89C Like playing with fire. At this range, your GPU might seem like it's doing just fine (you don't smell burning plastic, right?) but always assume you're doing long-term damage to your GPU. Only run this high if you don't expect to use that GPU for another year or two!
90C - 99C There's definitely some damage happening. Don't breath in the burning smell, it's probably bad for you!
100C - 109C If you're lucky you can use your GPU after it get this hot, but treat it nice like an abused puppy.
110C+ Congratulations, your GPU is now a paperweight. However, let it cool off before you hang it on your wall.
Again, these are general estimates aimed to help the community understand what different temperatures mean. If you have any personal experience or professional opinion I'd love to hear them and make edits if needed!
Dig hard, dig proud, dig cool, dig safe!


  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    Sounds about right. Thanks for your effort!
  • Dr_House86Dr_House86 Member Posts: 566 ✭✭✭
    How does it come that my asus Strix R9 390 with factory settings (for gaming etc..) go up to 95 Celsius and the fans dont regulate less?

    Current they running betreten 62-72 Celsius (mining)
  • bakibaki Member Posts: 26
    edited May 2016
    my 5 gpu open rig:61 62 59 66 50 degrees (all r9 390) 2 msi, 2 sapphire, 1 gigabyte
    the last one (sapphire with 3 fans) is almost 50 cm away from the rest. it is 21-22 degrees inside the room.
  • wirelessnet2wirelessnet2 Member Posts: 385 ✭✭✭
    MY rule of thumb is to try avoid going over 75c, 80c max
  • adasebadaseb Member Posts: 1,043 ✭✭✭
    Useless list pretty much.

    Different cards are suppose to run at different temperatures.

    An older 6990/7970 reference with quality components can run night and day at 80-90 C without any issues.

    While some cheap Diamond/Sapphire cards can burn out running at 70C.

    Its matters how they are built, what type of components are used, AIB manufacteur, etc.
  • HelioxHeliox Member, Moderator Posts: 634 mod
    Got a few 290 reference cards, they've been running months over 80 degrees, they even started off running at 88 back in the litecoin days :). Still doing fine! It all depends i guess, but a good reference none the less!
  • JPSJPS Member Posts: 63
    This list is bullshit. End of story.
  • mmxmmx Montreal, QCMember Posts: 110 ✭✭
    Agreed, this chart is meaningless since every processor has a different TDP level, and manufacturing processes will determine the resulting ASIC quality of the card and how it handles overall power dissipation.

    What's more important to monitor than GPU core temperatures are the Voltage Regulator Modules (VRMs). A lot of miners seems to neglect cooling them separately, or users simply don't even know they exist. Two applications that offer the ability to monitor VRM temperatures are TechPowerUp's GPU-Z and HWiNFO, but I'm sure there are others out there (Trixx?). Not all cards will actively report it's VRM temps, so overclock or mine with caution.

    Watching a Sapphire HD 7950s catch on fire on its next power up is a nice lesson to keep an eye on all temps.

  • koorpakoorpa Member Posts: 73

    my vrm are great here, but core temp is bad, maybe I should re apply thermal paste?
    it's xfx dd r9 290
  • SteelBSteelB Member Posts: 1
    Anything under 80c is acceptable for me, the reference 290's and 290x's can run hotter as they are made to.
  • tytanicktytanick PolandMember Posts: 66 ✭✭
    edited May 2016
  • liquidtrollliquidtroll Member Posts: 8
    nice post thanks !
Sign In or Register to comment.