|
Historical note:
In the 1980s, CERN bought mini/supermini computers from two vendors that experienced the same thing: The VAX-780 had a bug (I never knew the detail of that bug) that DEC proudly announced that would be fixed in the upcoming VAX-750. To which CERN replied that in that case, the 750 would never be on their shopping list: They had developed their software working around that bug, and they were not going to maintain one software version with the workaround, and another without. So the bug was not fixed.
Not long thereafter, Norsk Data made a similar announcement for the ND-100, the successor of Nord-10, which had a well known bug in the MOD calculation when both arguments were negative. They received a similar message from CERN: If the ND-100 was not 100% software compatible with the Nord-10, CERN wouldn't buy it. So Norsk Data changed the bug to a feature, i.e. they documented the old behavior, and stuck to it for the ND-100.
|
|
|
|
|
i recently learned of boost::safe_numerics . my worries of C++ int arithmetic overflow are over .
[edit] but not really .
[edit] i rolled my own i consider superior assuming no bugs as it is rather crude/clumsy/complex and have no idea how to test . can provide source to those interested .
modified 26-Sep-23 13:30pm.
|
|
|
|
|
Cyberpunk and BG3 all crash on repeat for me now, shortly after starting.
Even Overwatch seems to crash (though less often).
I don't see other people complaining and I suspect something with my system....
But.. I ran Memory Diagnostic: nothing found. DxDiag: not sure how to read it, but seems fine.
Looking at Event Viewer, the error info is unhelpful.
The only thing I know that is kind of strange with my system, and makes me slightly nervous, is if I use the power button to stop it, I can't start it for like 5 minutes.. if I try it glow briefly, start the fan and then stops. Looks like static electricity issue.. but no idea what to do about it or whether it is releated. I have this problem from day 1 (in April) but the game crashed only started this weekend... :/
I recently updated the NVidia driver, currently with latest 537.42, no one else seems to report that... :/
|
|
|
|
|
Hey,
Go setup a user-mode dump. Then start your application, let it crash and we can read the dump in WinDbg to try and determine the cause.
Edit: Use 0x00000121 as the CustomDumpFlag
|
|
|
|
|
Great idea, on it! 
|
|
|
|
|
Use 0x00000121 and don't do a full dump. Mini is fine.
|
|
|
|
|
ditto
"A little time, a little trouble, your better day"
Badfinger
|
|
|
|
|
Dumps are fine when it's a software issue, but for hardware oddities they generally don't tell you much.
Do you have an oscilloscope or multimeter? I'd start by looking at the simple stuff like PSU voltages. If it's all fine until you start a graphics intensive game, it's quite possible that the PSU can't keep up with the load and voltages are dropping out.
What GPU card are you running, how many watts is it rated at, and what's the PSU rated at? I like to run with at least a 25% margin on the PSU - so if my processor, drives, and GPU are rated to need max 400W, I'll get a 500 or 550 watt PSU.
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
|
The 4080 is a power hungry card, but the recommended specs are for a 750w PSU and you have 1Kw so you should be fine, even with a top end processor and several big HDDs.
But if the problem is getting worse it's unlikely to be software. Does anything else that uses significant GPU die, or do they work? If nothing else that tests your card fails, at least you can pretty much eliminate that ... which would be nice, given the cost of those things!
Have you tried something like this: Open Hardware Monitor - Core temp, fan speed and voltages in a free software gadget[^] - it can monitor voltages, temperatures, etc while your game is running which might help to eliminate problems.
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Only used NVIDIA performance overlay.
I dunno waht to look for, but as far as I can tell, temperature and memory usage are fine...
Gotta look at that thing you suggested!
|
|
|
|
|
|
Looks like the MSI PRO X790-A doesn't report voltages (or OHM can't get them for some reason), which is a pity.
It does for me: 2023-09-24-OHMGriff hosted at ImgBB — ImgBB[^]
Sorry, but I'm out of ideas from this distance!
Good luck. ![Rose | [Rose]](https://www.codeproject.com/script/Forums/Images/rose.gif)
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
OriginalGriff wrote: I like to run with at least a 25% margin on the PSU - so if my processor, drives, and GPU are rated to need max 400W, I'll get a 500 or 550 watt PSU.
Not only is good for that, but for possible future upgrades too.
I calculated my total need in high comsumption to be something less than 700W and I bought the 850W, next option would have been 1000W and I decided against it, because too much "unused" PSU should be avoided.
M.D.V. 
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
OriginalGriff wrote: Do you have an oscilloscope
That escalated quickly. 
Software Zen: delete this;
|
|
|
|
|
It's one of the handiest hardware debugging tools I know! Being able to just watch power lines and see if they drop under load ... checking serial coms are happening at all, never mind what speed ...
And you can buy one (albeit a cheap and crappy one with a tiny display) for under £25 on FleaBay if you can tolerate Chinese electronics!
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Oscilloscope for checking for faulty power supplies is a good idea - I ended up buying a cheap second hand one for exactly this purpose.
A big alternative is a bin of spare parts. You can quite quickly isolate a faulty component if you can swap out components that are hopefully themselves good. Trouble is most people, even if they have multiple computers like me, the machines span different hardware generations, each of which have incompatible connectors between CPUs, memory modules, draw different amounts of power etc.
|
|
|
|
|
I found this thread in german. If you have problems with the translations tell me, I can help you on that.
Gigabyte Rtx 4080 stürzt ab bei 3DMark und Spielen unter Last über längere Zeit | ComputerBase Forum[^]
TL;DR; As OG said, looks like a consumption problem. In the thread are a couple of tools that might help. Recomendations avoid overtact, reduce MPO, check if the card in other system gives the same problem...
A guy says that in other forum people was complaining about DOA (dead on arrival) cables, how long have you been using it and when did it start to give problems?
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Bough it in April... started to have crashes.. this weekend?
The power consumption seems fine... only used up to 300W in the captured image... (And I have an 1kW PSU)
Hardware-Monitor hosted at ImgBB — ImgBB
|
|
|
|
|
Based on what other people are saying, my takeaway is your PSU is not supplying enough power.
You said you had 1000W, which with my motherboard and the same GPU is yours, is *just* to spec, because my mobo and processor are hungry animals.
But if your PSU is giving you trouble - and it sounds like it might be - it might not be able to work under significant load say, > 300W.
Replace your PSU and see if that fixes it. If it doesn't, return the PSU. It's a lot of work, but it's one of the first things i'd look at based on your statements and the comments on this thread.
Check out my IoT graphics library here:
https://honeythecodewitch.com/gfx
And my IoT UI/User Experience library here:
https://honeythecodewitch.com/uix
|
|
|
|
|
I use the shutdown command ... I use the power button when nothing else will shut it off; and I don't like doing it.
I look at the file dates / update history of my game(s). MS is mucking with store and XBox titles (to force you to the cloud, IMO). Some of my titles are in a "suspended" state that won't run; won't update ... I'm back to (old) DVD's and my external Blue-Ray drive for "reliability".
"Before entering on an understanding, I have meditated for a long time, and have foreseen what might happen. It is not genius which reveals to me suddenly, secretly, what I have to say or to do in a circumstance unexpected by other people; it is reflection, it is meditation." - Napoleon I
|
|
|
|
|
-1) Do your wall socket/extension/all the way to the PSU wired properly?
0) Does it crash only while gaming? Crash or shutdown?
1) As they said check the PSU. Not every output cable created equal, check that your video card gets the dedicated lines (rails).
2) Not starting after turned off may be a motherboard problem, too. Try to re-socket the video card.
3) Check that all fans are unblocked and spinning. Heatsinks are connected, tightened? Crashing in the same time under same load may be overheating, too.
|
|
|
|
|
-1) I mean they build and ship it to me, but, hmm.. I only removed the foam and plugged it, looks alright...
0) game stop and I got an error report window to send to the game company
1) I tried to job the various cable in the box a bit, but everything seems very firmly plugged, no give... it's all I can say...
2) ... Increasingly... I think I might have to buy some compressed air and replug everything! ^_^
Also idly wondering if 1000W PSU is enough, latest most powerfull gaming box by MWAVE have 1200 PSU (and I have 3 SSD, and i9, etc..)
3) yea all fans working fine, not over heating happening...
|
|
|
|
|
Looks like your power supply is well within range:
Do I need to upgrade my PSU for the RTX 40 Series?
The RTX 40 Series doesn’t require a new power supply if you already meet the PSU wattage recommendations. The RTX 4090 TGP is 450 W and the minimum recommended PSU is 850 W. The 4080 16GB TGP is 320 W with a minimum recommended power supply of 750W, and the minimum recommended power supply for the 4080 12GB is 700W.
Can you borrow a similar high-power PSU for testing?
If you want to buy a new PSU, maybe it should not be Gigabyte (they had even exploding problems lately). Here are some offers and a lot of power requirement calculators, advices
But the best would be testing with another PSU before buying a new one.
Are some of the SSDs installed by you?
|
|
|
|
|
I'm going with everyone else and say something is going on with your power supply. I would suspect that it has bad capacitors in it. I'm running a RTX 4070 with a 750W Seasonic power supply and haven't had any issues.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
