Please be specific, no one was ignoring you. You weren’t just specific about the meter or the whole thing, what meter, which meter, where can I get it, can I do It all by myself? how can I use it?
you are talking about that a 12 year old can do this but not everyone is smart in this world.The Shop I went for testing is the best in my area but yet they’re stupid if we compare them to others.they didn’t said anything about the metre or anything there were just swapping and testing each parts.
And yes not everyone knows how a computer works, how hardware works.
If you want to know my PC specification then here it is
Core i5 6400
8GB DDR3 1600Mhz
Gigabyte h110m series
240gb pny ssd
500GB WD blue HDD
Sapphire rx590 8gb oc edition
Thernaltake smart modular 630W
And also what kind of number you’re talking about? Are you telling me to test the power supply with that meter? Is there any link of any video about this?
You sounds like an expert, then be specific about this and help me out.
Thanks
Edit: are you talking about digital multimeter? Telling me to test the psu with it?
Or those voltage numbers from BIOS is enough?
Had instructions been requested, then all those questions were answered.
Those numbers from a digital meter on the motherboard are not yet calibrated and are only half of necessary information. Get a digital meter, request instructions, and perform the labor. Then those questions and more were answered.
Many techs do not even know of that power controller, what it does, and why it causes power offs. Most techs only understand shotgunning - keep replacing good parts until something works. Then one never learns why a failure happened, how that subsystem works, or how to avert future problems.
A consumer magazine intentionally created numerous defects. Most of those computers has perfectly good power supplies replaced. Few actually had a defect repaired. Many only cured symptoms by shotgunning rather than first identify a defect.
First indication that a shop knows the technology: he also says why (defines a defect) and explains it with numbers.
Same applies to another necessary solution - a comprehensive hardware diagnostic. From what I see, most do not even know what that is. Some even confuse HWInfo with a diagnostic. A comprehensive diagnostic must operate without even Windows loaded. Since Windows can mask defects or complicate analysis. Another test to determine if that tech really knows how a computer works.
Meanwhile, nothing useful can be concluded until known is that power controller and what it sees. Even a comprehensive diagnostic might report defects (that do not exist) if the power subsystem is first not confirmed good.
Something like this, yes.
But remember to be careful, PSU components can store a charge and you can get shocked(literally) if you touch or do something wrong.
That paperclip he uses there in the video is an obvious example of a potential hazard, he is plugging it in while the PSU is completely disconnected, something you should absolutely not touch once the PSU is connected and switched on.
Also notice the gloves on the assistants hands there. Offers some protection, but not just any gloves should be chosen. Plain cotton would likely be an acceptable choice.
At least this one isn’t opening up the PSU so its not as dangerous, if you are going to test anything on your own, this might be the thing to test, but again, it’s the paperclip part that I would say bothers me.
At the very least wrap some electrical tape around most of it before plugging it into the connector to make it safer. And make sure to google and check/read through multiple guides before you attempt this.
Also, if you don’t already have one, the tester can easily be as costly as a replacement PSU, well a good tester anyways. You could get by with a cheap one too though.
Again, as others have stated before, this is something best left to a professional. Not a computer store but more like a electronics repair shop, one operated by a proper electrical engineer, with the papers to back it up. That’s the theory anyways. This is clearly not the cheapest option though, and there might not even be one available in your immediate area for all I know.
Annoying intro aside, here’s a better(as in more thorough, goes into more detail) video:
If I remember correctly you said you tried a different PSU and the problem persisted. It doesn’t make much sense for you to make electrical measurements. In addition, it is possible that the electrical measurements on PSU must be performed with a connected load to correctly measure the voltage drops. If you don’t have much idea about electronics, you better take your hardware to a professional technical service to avoid damage to you and your hardware.
@YAFU @Felix_Kutt Yeah i tried with different PSU, still it was restarting. But what @westom is saying isn’t clear at all. he’s still asking me to test the PSU with a digital meter xD
But after testing everything i though now it’s irrelevant. Cause how can i test it? i watched the whole video and looks like he’s trying to test if the PSU is providing enough voltage.
My pc isn’t restarting randomly or even when i’m gaming. This is a different matter.
Yesterday again i ran Heaven benchmarks for 20min straight. using pc for 8-10 hour straight in a day. still it doesn’t restart unless i do those two things, so can’t be a PSU related issue.
Again, perfectly good parts connected together can act defectively. Just one of so many reasons why shotgunning creates confusion. Furthermore that fear about stored charges is bogus - if one learned how this stuff works AND relevant numbers. 3, 5, and 12 volts does not threaten anyone.
That video shows how to identify some defective supplies. And does accurately describe connectors and rails. It mistakenly says setting the selector switch wrong or installing a defective supply can damage the computer. ATX Standards say It cannot; must not.
That video does not say a supply is good. Even some defective supplies can pass those 5% tolerances. That supply is not connected to other parts of a power system. So a meter can only identify some defective supplies, cannot say the supply is good, and says nothing about the entire subsystem. Measurements only say something useful when nothing is disconnected.
Also damning in that video. Nobody’s computer is so hot as to also toast bread. It is not consuming anywhere near 800 watts. But that is what we tell computer assemblers because so many have no idea what those rail numbers mean, what is necessary, and how to determine what is sufficient. Computer assemblers are told to buy at least an 800 watt supply for a computer that does not even consume 400 watt. So that help lines are not clogged teaching basic electrical concepts to assemblers. Just another reason to know why that video provides insufficient facts.
Does the PC restart? That was the original issue. PSU alone says nothing. The entire power system must be known as a system. Describing one part as good does not, for example, even discuss a power controller - another part of that subsystem. That controller is why a computer powers off.
So if it not restarting, then what is the current problem? Meanwhile, normal is for shotgunning to cure a symptom. A defective part remains. And sometimes, many month or year later, the system fails again. But a digital meter using requested instructions and two minutes of labor can identify some defects even when a computer still operates.
That video does not say how to use a meter and how to find other defects - that might has still existed inside that PSU.
Again, get a digital meter, request instructions, and perform two minutes of labor. Those resulting three digit numbers, that mean nothing to you, say everything to other informed assistants. Why? Too complicated to explain. But the better informed can say nothing without those numbers. And yes, it can still be a PSU issue. Or some other part of that power subsystem.
The symptom from the video looks really like hardware overheating caused by blender, like other users had said.
If you can, try removing your gpu and use intel cpu’s graphic (remove your HDD as well just in case) and try it again. If your pc restarts, then now you know for sure it’s not your gpu. Next use your linux hard drive (presuming the installation was separate from the start) and test the same again. If the pc also restarts, it is likely not related to the OS.
From there, you will only have to check mobo, cpu, ram and psu. Since you did say that you’ve already tested a few different ram and psu, it leaves only the mobo and cpu as the remaining hardware issue suspects.
If after changing to a complete different cpu and mobo combination still didn’t help, then it confirms it’s a blender issue. If not, then it’s should single out either the cpu or the mobo as the final hardware suspect.
and for ffs don’t immediately try to change wattage for bios and psu. Single out the issue before following dangerous procedures that could damage you and your hardware.
Hi…is the course good. Cgboost?
All kind of troubleshooting is done, also tested with another cpu and mobo still it’s restarting,
I’m guessing it’s something with AMD gpu’s, Can’t find any Nvidia gpu to test with, but i tested with another rx 580 8gb gpu and still pc restarted.
Maybe this rx series GPU can’t take some kind of pressure or something.
cause you know, it’s not restarting when working with 2px tiles size and 4px causing pc to restart while 4px is lower, and also in substance painter doesn’t cause restart even when i export in 8k, but it restart’s when i add HSL prespective, color correction and color balance with so many layers.
I think so,
i didn’t follow the course of whatever, after watching the course trailer, I tried to make that short animation on my own.
I know this will not make any difference to anyone… but I have the feeling that this rather extensive matter is about to reach “Barrymaritime”-status.
…I thought I would mention this.
carry on
In that case it’s probably worth opening a bug report on the blender tracker as well as contacting AMD support.
Confusion remains because that strategy is called shotgunning. Keep replacing good parts on speculation until something works. “I guessing” is the first indication that one is not “following the evidence”. Good diagnostic procedure has been replaced by speculation.
A power system defect can make other good parts act defectively. That power system defect is also why a system would suddenly power off. The reason for sudden power offs has not been solved. Symptons may have been cured.
Nothing yet says that power system is good. Only way to identify that defect is to remove nothing, disconnect nothing, request instructions, use a digital meter as defined in those instructions (and not by those other websites), and then report the resulting numbers.
Nothing has yet been defined good - without doubt. The so many “it could be this” or “might be that” speculations keep returning to many previously swapped parts.
Troubleshooting means eliminating parts as suspect - without any doubt. Thing that suddenly turns off power is the same thing that makes other good parts act defectively.
Two minutes of labor could have replaced those few hundred posts (with no resolution).
No pressure exists. Computers work synchronously. Components work just as hard or just as easy - all determined by master clocks. Pressure on a GPU is just another example of wild speculation not based in how computers work. Worse, that speculation is not tempered by any numbers. No numbers is another indication of why confusion reigns.
So what about that thread calling for RX 580/590 users to test with your .blend file at 4px size in volumetrics?
Here it is
Hi Thank you.
The idea was to make clear the call for users of AMD RX500 to test with your blend file, for example with a title: “I need AMD RX500 users to test with this .blend file”
Otherwise this new thread runs the risk of becoming another kilometer thread with multiple suggestions that have already been mentioned in this thread.
So if you can edit the title of the other thread for something better…
Oho, Thank you. I edited the title