Computer totally locks up when "loading in" things.

[Omega Pi]

Um, hi. (This is my first post here, I kinda feel bad about my first post being something that I could use some help with, but I am at a total loss)

 

Basically my issue is that whenever I load a game, or the game has to load in new elements, such as switching menus, loading screens etc) my pc will, softlock I guess you can call it.

Often animations and sounds will keep playing, I can still move my mouse, hover over for example the taskbar icons to highlight them, but when I try to click anything my pc will completely become unresponsive.

The issue is very easy to reproduce for me by starting up forza horizon 3 and giving my CPU a 4.4ghz OC (this is a guaranteed way for me to reproduce the issue, it will lock up after the initial loading screen just before it switches to the actual game) in other games, like Hitman 2016 it may take a while, but eventually on a loading screen the pc will lock up, could take an hour, could take 2.

Also, on lower end games the issue will reproduce exept it just takes a lot longer.

 

Generic browsing, watching youtube, movies etc, all works fine. The issue ONLY happens when games need to load in recourses, Once the game loads gameplay is fine on any game. When I press escape to go to a menu, or scroll through menus (such as the vehicle upgrade ones) in forza, it's hit and miss with the lockups. Sometimes my mouse will lag a little for a few moments before the system stops responding.

 

My Graphics card and CPU stressed seperately in either Furmark, prime95 etc. Everything runs as desired, no errors, but once they have to start working together It's as if my gpu just kind of drops load for some reason.

Next to that, I have tried pretty much anything:

-Reinstalling windows, rebuilding the computer (to make sure everything is seated properly, no shorts being created etc) cleaning the GPU's heatsink, re-flashing my bios, turning off cores, turning off turbo core on the cpu.

 

The specs of the system are:

CPU- AMD FX-8320 (used to run a stable 4.7ghz OC, cooled adequately on water, not overclocked at the moment, turbocore turned off) - increasing the clockspeed will reproduce the issue quicker.

GPU - AMD RX580 4gb (sapphire pulse) - Never been touched when it comes to overclocking etc.

RAM- Kingston HyperX Fury ddr3 1600 2x4gb

SSD- Kingston SSDNow V300 120g (I think)

HDD- Freecom USB hdd - 500gb (not the cause of the issue)

Motherboard - MSI 970a-gaming pro carbon. (has had an issue or 2 with recognizing my ssd after things went sideways, never really had trouble with the board)

Power Supply- MS-tech Val(something) claims to be 750Watts. I am well aware of how horrible this power supply is and that I should replace it ASAP, the voltages on the power supply seem off and change every time I check.

OS - Windows 10 x64

 

I am well aware that the issue might be caused by the power supply, although my problem at the moment is that I'd rather bet my money on the best bet for short term and delay the rest a little, concidering I do not have 500 or so euro laying around to spend on everything I'd like to replace in the system, and fixing this issue has more priority for me than replacing something because it might fail in the future, and since I am torn between all the components in my system being the issue, I feel like having a 2nd opinion on this might be a smart first step.

[paddy-stone]

21 minutes ago, Omega Pi said:

Um, hi. (This is my first post here, I kinda feel bad about my first post being something that I could use some help with, but I am at a total loss)

 

Basically my issue is that whenever I load a game, or the game has to load in new elements, such as switching menus, loading screens etc) my pc will, softlock I guess you can call it.

Often animations and sounds will keep playing, I can still move my mouse, hover over for example the taskbar icons to highlight them, but when I try to click anything my pc will completely become unresponsive.

The issue is very easy to reproduce for me by starting up forza horizon 3 and giving my CPU a 4.4ghz OC (this is a guaranteed way for me to reproduce the issue, it will lock up after the initial loading screen just before it switches to the actual game) in other games, like Hitman 2016 it may take a while, but eventually on a loading screen the pc will lock up, could take an hour, could take 2.

Also, on lower end games the issue will reproduce exept it just takes a lot longer.

 

Generic browsing, watching youtube, movies etc, all works fine. The issue ONLY happens when games need to load in recourses, Once the game loads gameplay is fine on any game. When I press escape to go to a menu, or scroll through menus (such as the vehicle upgrade ones) in forza, it's hit and miss with the lockups. Sometimes my mouse will lag a little for a few moments before the system stops responding.

 

My Graphics card and CPU stressed seperately in either Furmark, prime95 etc. Everything runs as desired, no errors, but once they have to start working together It's as if my gpu just kind of drops load for some reason.

Next to that, I have tried pretty much anything:

-Reinstalling windows, rebuilding the computer (to make sure everything is seated properly, no shorts being created etc) cleaning the GPU's heatsink, re-flashing my bios, turning off cores, turning off turbo core on the cpu.

 

The specs of the system are:

CPU- AMD FX-8320 (used to run a stable 4.7ghz OC, cooled adequately on water, not overclocked at the moment, turbocore turned off) - increasing the clockspeed will reproduce the issue quicker.

GPU - AMD RX580 4gb (sapphire pulse) - Never been touched when it comes to overclocking etc.

RAM- Kingston HyperX Fury ddr3 1600 2x4gb

SSD- Kingston SSDNow V300 120g (I think)

HDD- Freecom USB hdd - 500gb (not the cause of the issue)

Motherboard - MSI 970a-gaming pro carbon. (has had an issue or 2 with recognizing my ssd after things went sideways, never really had trouble with the board)

Power Supply- MS-tech Val(something) claims to be 750Watts. I am well aware of how horrible this power supply is and that I should replace it ASAP, the voltages on the power supply seem off and change every time I check.

OS - Windows 10 x64

 

I am well aware that the issue might be caused by the power supply, although my problem at the moment is that I'd rather bet my money on the best bet for short term and delay the rest a little, concidering I do not have 500 or so euro laying around to spend on everything I'd like to replace in the system, and fixing this issue has more priority for me than replacing something because it might fail in the future, and since I am torn between all the components in my system being the issue, I feel like having a 2nd opinion on this might be a smart first step.

Well, it could be the RAM, sounds iffy, run memtest on the individual sticks to see if any errors come up. But mainly I think it could be the PSU to be honest... it's better to replace something that's defective than to hold off on it and possibly watch your PC go up in flames, yes it's a worst case scenario, but possible. If you get somehting like an EVGA g2/g3 550/650 then it should last you a good while for sure and is warrantied for 7-10 years depending on model... so overall a cost of approx $8-14 a year maybe in all.

Just have to keep testing stuff if you can to try and whittle it down... borrow parts if you can.

 

[Omega Pi]

7 minutes ago, paddy-stone said:

Well, it could be the RAM, sounds iffy, run memtest on the individual sticks to see if any errors come up. But mainly I think it could be the PSU to be honest... it's better to replace something that's defective than to hold off on it and possibly watch your PC go up in flames, yes it's a worst case scenario, but possible. If you get somehting like an EVGA g2/g3 550/650 then it should last you a good while for sure and is warrantied for 7-10 years depending on model... so overall a cost of approx $8-14 a year maybe in all.

Just have to keep testing stuff if you can to try and whittle it down... borrow parts if you can.

 

I was planning on getting a good one, wouldn't want to skimp on the power supply indeed, my problem at the time I bought this one was mainly that my old one wasn't sufficient and this was the only one that would allow my pc to boot/run that I could get the same day, does indeed sound iffy, will run a memtest overnight, if all comes out good I will order a new power supply tomorrow, even if that doesn't fix the problem, can't hurt to replace this one. Just wanted to hear some extra thoughts on the issue since I am at a total loss, I've spent a week or so troubleshooting and it just seems to get worse, if that helps anything.

 

Sadly I can not borrow parts, I could however go to a local computerstore who would have the supplies to test the system.

[paddy-stone]

2 minutes ago, Omega Pi said:

I was planning on getting a good one, wouldn't want to skimp on the power supply indeed, my problem at the time I bought this one was mainly that my old one wasn't sufficient and this was the only one that would allow my pc to boot/run that I could get the same day, does indeed sound iffy, will run a memtest overnight, if all comes out good I will order a new power supply tomorrow, even if that doesn't fix the problem, can't hurt to replace this one. Just wanted to hear some extra thoughts on the issue since I am at a total loss, I've spent a week or so troubleshooting and it just seems to get worse, if that helps anything.

I'm leaning towards RAM now after a re-read of your first post. The freeze screens and lag are definitely pushing me towards that way... although it could be attributed to thermal throttling too maybe... so check thermals for definite too. The new PSU certainly won't hurt at all, and is a good step anyway... I get why you have it, we've all been hard pressed for money at some point and gone with a cheaper option.

Could be SSD/HDD even, there are so many variables in PC troubleshooting, and it's hard even for experienced trouble-shooters to do this too... just that they usually have some more  spare parts around to test with, lol.

 

One thing I would do for sure, is create a system image in case  you need to re-install to a different SSD/HDD, as you've recently re-installed anyway.. then at least it'll be quicker to re-image or write to new disk of you have to at some point. I use AOMEI backupper, it's free and pretty useful, you can create a USB recovery drive for re-imaging and stuff, like a LIVE build OS to use for certain tasks. I find it quite useful.

 

Anyway, hopefully we can get to the bottom of the problem.

[Omega Pi]

20 minutes ago, paddy-stone said:

I'm leaning towards RAM now after a re-read of your first post. The freeze screens and lag are definitely pushing me towards that way... although it could be attributed to thermal throttling too maybe... so check thermals for definite too. The new PSU certainly won't hurt at all, and is a good step anyway... I get why you have it, we've all been hard pressed for money at some point and gone with a cheaper option.

Could be SSD/HDD even, there are so many variables in PC troubleshooting, and it's hard even for experienced trouble-shooters to do this too... just that they usually have some more  spare parts around to test with, lol.

 

One thing I would do for sure, is create a system image in case  you need to re-install to a different SSD/HDD, as you've recently re-installed anyway.. then at least it'll be quicker to re-image or write to new disk of you have to at some point. I use AOMEI backupper, it's free and pretty useful, you can create a USB recovery drive for re-imaging and stuff, like a LIVE build OS to use for certain tasks. I find it quite useful.

 

Anyway, hopefully we can get to the bottom of the problem.

Thermals are completely fine, the cpu doesn't push over 60 degrees C on a thermaltake water 3.0 GPU also appears to be totally fine, it uses the default fan curves and sticks at around 70-75 when pushed on full load with the fans not spinning at their full speed.

SSD also came to mind, concidering it only happens when the system loads in something new, although I'd have no clue where to begin testing the SSD rather than pushing a lot of load onto it, which it gladly accepts. I could however push some load onto the ram and cpu by rendering something and see what it does, will report back when done.

 

[edit 1]

But the issue reproduces too consistently for it to be thermals.

(also, I must add, this motherboard was initially bought for it's VRM heatsink concidering my old motherboard would throttle due to overheating, my case is an nzxt h440 (currently running without either of the sidepanels) and the watercooler pulls air from the case and dumps it out through the top) After that incident the first thing I ever check is thermals, and eventhough I am not a pro on things like this, I am not a total noob either, I would like to think.

 

[edit 2]

Also, as mentioned in the initial post, increasing the clockspeed, even enabling turbo core would make the issue worse, for me that takes the Graphics card out of the equation and draws my attention to the fx8320 (which I abused more than I'd like to admit) the ram, like you implied, or indeed the power supply. (I must also admit that I forgot to mention that I have managed to make my pc completely shut down more than once due to high powerdraw, especially with rather silly overclocks, although back then I was rocking an r9 270x, which draws quite a bit more power than the Sapphire pulse rx580.

 

[edit 3]

pc totally locks up and the mouse lags, completely reproducing the problems I am having, with the only exeption being that it recovers. (this is with the gpu at zero load whatsoever, the cpu at full load and about 70 percent ram used)

[paddy-stone]

15 hours ago, Omega Pi said:

Thermals are completely fine, the cpu doesn't push over 60 degrees C on a thermaltake water 3.0 GPU also appears to be totally fine, it uses the default fan curves and sticks at around 70-75 when pushed on full load with the fans not spinning at their full speed.

SSD also came to mind, concidering it only happens when the system loads in something new, although I'd have no clue where to begin testing the SSD rather than pushing a lot of load onto it, which it gladly accepts. I could however push some load onto the ram and cpu by rendering something and see what it does, will report back when done.

 

[edit 1]

But the issue reproduces too consistently for it to be thermals.

(also, I must add, this motherboard was initially bought for it's VRM heatsink concidering my old motherboard would throttle due to overheating, my case is an nzxt h440 (currently running without either of the sidepanels) and the watercooler pulls air from the case and dumps it out through the top) After that incident the first thing I ever check is thermals, and eventhough I am not a pro on things like this, I am not a total noob either, I would like to think.

 

[edit 2]

Also, as mentioned in the initial post, increasing the clockspeed, even enabling turbo core would make the issue worse, for me that takes the Graphics card out of the equation and draws my attention to the fx8320 (which I abused more than I'd like to admit) the ram, like you implied, or indeed the power supply. (I must also admit that I forgot to mention that I have managed to make my pc completely shut down more than once due to high powerdraw, especially with rather silly overclocks, although back then I was rocking an r9 270x, which draws quite a bit more power than the Sapphire pulse rx580.

 

[edit 3]

pc totally locks up and the mouse lags, completely reproducing the problems I am having, with the only exeption being that it recovers. (this is with the gpu at zero load whatsoever, the cpu at full load and about 70 percent ram used)

 

Yeah, it's a puzzler alright. Let's see what happens with the RAM tests. If no errors occur it's not necessarily a sign that they're OK, I have had RAM pass memtest yet still be faulty... but it might throw up some errors if it's faulty in this case because of the consistency in which you get problems.

If it passes tests, I'd go ahead and either buy a new GPU or borrow one if you can just for testing, unlikely, but worth a try I guess as someone you know might have a spare.

I also recall some of the kingston SSDs being in question, not sure if yours is one of the models or not, but might be worth investigating that as a possibility too... if you have a spare HDD to load windows onto and test it would be a good idea. You don't have to do anything to the SSD, if testing with a HDD still throws up errors, you can just plug the SSD back in and change the boot order if needed.

 

Oh, one other thing I meant to ask is, did your system all of a sudden just start doing this, or had it been a problem since you built it? or when you changed something?  I don't recall you mentioning how long you've had the system... if you did then I apologise as my memory isn't the greatest at times.

[Omega Pi]

33 minutes ago, paddy-stone said:

 

Yeah, it's a puzzler alright. Let's see what happens with the RAM tests. If no errors occur it's not necessarily a sign that they're OK, I have had RAM pass memtest yet still be faulty... but it might throw up some errors if it's faulty in this case because of the consistency in which you get problems.

If it passes tests, I'd go ahead and either buy a new GPU or borrow one if you can just for testing, unlikely, but worth a try I guess as someone you know might have a spare.

I also recall some of the kingston SSDs being in question, not sure if yours is one of the models or not, but might be worth investigating that as a possibility too... if you have a spare HDD to load windows onto and test it would be a good idea. You don't have to do anything to the SSD, if testing with a HDD still throws up errors, you can just plug the SSD back in and change the boot order if needed.

 

Oh, one other thing I meant to ask is, did your system all of a sudden just start doing this, or had it been a problem since you built it? or when you changed something?  I don't recall you mentioning how long you've had the system... if you did then I apologise as my memory isn't the greatest at times.

 

I did run a ramtest overnight, no errors returned, bit the bullet and ordered a new power supply, I noticed that before the last freeze I had both my gpu and cpu load dropped to 0. Also, eventhough I do know that voltage readouts aren't necasserily the most reliable things to go off software wise, but add to that that the power supply I have is nothing more than a piece of crap my voltages are WAY off, the power supply itself is been known to give up on me when pulling rather high amounts of power through it i/e high overclocks (4.9-5ghz) will have to wait til tomorrow before the power supply arrives.

 

The SSD itself always ran fine without any trouble, I do think the SSDnow ssd's are the ones you're talking about when it comes to questionable performance, it's just mainly that at the time I bought the parts that are of lower quality was mainly because these were the only ones I coukd physically get my hands on within a day, and concidering my computer is the only device I have, I can't really spend a day or 3 without it.

 

I indeed didn't mention how long I had the system because none of the parts are as old as the others. The cpu is fairly old, up for an upgrade to ryzen including a new motherboard and ram when I can. (around 3-4 years?) The gpu is recently bought, I think I was one of the first few people in my country who had one before the prices skyrocketed. The ram is also around a year, maybe a year and a half, the motherboard is around half a year old and the power supply also something like that.

 

Will report back once new power supply is installed.

 

[edit] Did it suddenly start, honestly. It's a hard question to answer. I had a bit of trouble with Windows defender a while ago, which totally ruined my windows installation to a point that anything that required my graphics card to use it's drivers it would just turn my whole screen green for a split second and shut down my gpu completely. (that has been resolved by re-installing windows)

 

Before that I did indeed have some minor issues that were pretty much neglectable for me at least, such as the computer completely shutting down under high load very rarely, or not getting past the MSI logo where it's supposed to load in windows under certain circumstances. (which I at the time passed off as a bios bug) I don't think I mentioned this, but I do run the latest bios update. But the issues just got worse, for example a week or so ago I could play Forza Horizon 3 without issues, it'd freeze so every now and then rarely, and that's pretty much where it began til it's at the point where it's now, where the freezes become more frequent and common to a point I don't even bother starting up any game that puts the slightest amount of load on the whole system.

 

So I would say the issues like this started a week up to 2 weeks ago and gradually got worse.