shark103

3 hours ago, will01gt said:

Only on these two games now

However windows has started freezing randomly as well every 4+ hours

I have not yet tried snappy and will now

Crystal disk said the bigger ssd windows is on is at 100% health and the older 970 evo is at 98% health

Also for notes I have 400gb free on bigger ssd out of 1800gb

and the other has 50gb out of 450gb

If you have at least 10gb on windows partition and updates up to date(no 10gb+ updates in que) then free space is not an issue for sure.

Randomly when it's on for more than aprox 4hours or randomly and 4+ hours is aprox from a day/two ? 

Is the psu under warranty ? 

10 hours ago, will01gt said:

Right I've done individual memory tests and nothing came up

And I let it run for a few days and it hasn't crashed yet

Until now

I guess I'm back here for any ideas

Still almost only crashing on those 2 games ?

Did you try with updating chipset drivers etc via snappy ?

Did you tried anything with peripherals ?

Did you look at drive health - eg Crystal Disk Info ?

Beside what's above all all earlier things - I'm out of ideas and leaning fully to PSU fault as it may not be able to handle spikes in load, problem with it only havving one 8pin

22 hours ago, Winnah5 said:

Last question and I should be okay to put some possible parts together before coming to a conclusion. Are there any other cards I should consider aside from the GTX 750 Ti/AMD Radeon HD 7850? I just want to expand my options before settling on a card since i' m buying used after all.

 

Anyway time to start prepping for my finals and then I can choose what I will actually go with for these upgrades lol

Can be anything higher (in power) than those if you can get it cheap enough. Example if you can get a GTX950 (or higher) in the same price or lower than 750Ti then go for it. Nothing lower as 750Ti/7850 is an optimal pair with your cpu - lower one and your new gpu will be the bottleneck for the cpu.

One additional thing as you might not now this a 670 is more powerful than 750ti and those you probably could also get cheap.

Nvidia gpu power you could say that if you take their number from gtx family as XY0 then

X|Y+2|0 >> X+1|Y|0

what i mean on examples

670 >> 750

660 around the same 750

650 << 750

XY0 << XY0ti

2 hours ago, Winnah5 said:

Thanks for the PSU recommendation! I'm starting to like the idea of fully modular PSU's since it would probably make cable management much easier for me being first time doing so and on top of that my case isn't as friendly for cable management solutions but its still doable lol. It makes sense though since even for my case fans something about the 3 pin to Molex connection just gave me a weird feeling, I guess even the noob in me knows adapters and cheap power sources don't go together really lol

 

I'm going to hold back on the RTX 2060 though, it would be nice to have it on hand for when I do build a new rig, but i'd rather not bottleneck my system and would like to fully explore my options after upgrading this rig before even considering stuff for the new one (well aside from the PSU depending on if I will utilize the same wattage in the new one). I'll definitely consider the other stuff for the new build though, after watching a few vids I do like the smaller form factor of a NVMe SSD versus the Sata one in which my case would be a pain in the ass to mount anyway. The Commander MS-I is a decent case but seeing it's a bulk bought part from a prebuilt company all the neat features it has instantly get stripped away lmao

Get a new PSU for sure. @cbigfoot gave you a nice suggestion with that. You can also look at this list if you can get a modular cheaper - suggestion look at tier A and PSU with long warranty like that Seasonic.

Getting a RTX 2060 now make zero sense, there is absolutely no point in that if you plan to keep current rig with upgraded GPU for 2-4 years. After that time 2060 could be a bottleneck for new CPU or could just be dead and you would never even use 50% of it's power.

So you need more RAM, ok probably best way is to just buy that in shop as DDR3 should be cheap as it's older gen - 60-70 bucks would get you 2x8gb from Patriot, Mushkin or G.Skill.
Sata SSD is obvious that's why I didn't even mention it. You'll be upgrading psu, gpu, ram quite a bit under your budget so you can spend that 35-50 bucks for a WD Green, Kingston A400, Crucial BX500. Later on when you will be building new rig you can take that ssd and throw it into cheap usb3.0 external enclosure.

PSU+RAM+SSD+GPU would on total be around 300-350 bucks from what i see online. PSU will be moved to new rig later on, SSD could be used as external.

@Winnah5
Hey,

Upgrading this machine hmmm... I wouldn't. That CPU is old and weak as hell it will be a bottleneck for any modern GPU. If don't want to buy a new rig just yet but feel the need to upgrade GPU then I would suggest getting a used GTX750Ti/AMD Radeon HD 7850 or something a bit more powerful that will still be cheap. 60fps on medium in GTA5 no way with that CPU, maybe 60fps on low.

RAM 16gb with that CPU is not worth it if you don't use memory hungry programs - you won't be using it in new future rig as it's DDR3.

1 hour ago, LethoX said:

Update: My little brother bought the exact same PC as I did at the same time too. Now he's the one getting lags, we updated windows 10 the same day (4-5 days ago). So it really makes me think it has something to do with windows. But then it doesn't make sense why it worked to let it sit overnight. Because if you just do a simple restart it still lags when it boots. Shutting it off for 30 minutes and it still lags. I checked the task manager when it happend for him and cpu is at 0-1%, memory is at 18%. Should we get an earlier version of windows?

What about disk usage ? Some programs/processes could use only a fraction of cpu/memory available but have massive disk usage.

8 hours ago, MrChemistryCow9 said:

I just did a quick cable clean up job on my pc. I guess I screwed it up again as it started crashing when it went over 100% power with cpu load. I tried fixes and none worked, and I can tell when they do as kombustor doesn't report GPU power if its not working. What would you do?

After fixing lose cable, crashing at max load and what would I do ? That's simple but probably not what anyone looking for help wants to hear - get a new PSU. 

With the current one you could "fix" the issue for a bit by going back to stock (zero OC) on both CPU and GPU or even undervolt CPU a bit but that would only be putting a bandage over a hole in a wheel - won't fix the problem only hide it for a bit. Get a new PSU asap and while picking it look at the tier list

Anything tier A or higher will be good (don't go for S instead of A as it won't give you anything, only do that if you can get it cheaper that lower tier one). For wattage use a psu calculator < this is the best one I know as it also allows to input OC, AIOs etc (don't buy recommended psu just use this to calculate how much W you need).

8 hours ago, will01gt said:

I'll try all this

 

Also somebody sent me this:

https://docs.microsoft.com/en-us/windows/client-management/troubleshoot-event-id-41-restart

 

This indicates I have scenario 3 which in my case is a hardware problem

But what do you think as i have sent the event viewer pages

I've seen all that on MS site - that's why I wanted you to hit it with synthetic load and wanted those logs. From what i see there it doesn't seem like PSU fault but I cannot be sure in 100% as you PSU is on the lower side on quality for such a system.

One problem with synthetic load compared to games is that the load gradually increases and in game it's a sudden spike. 

edit: Idea of unpluging all usb devices and using different ones is based on one of the errors from event viewer - "Error setting traits on Provider" if you would google the id after this text you would stumble on a lot of information about this being a problem with usb devices, usb controllers etc.

One more thing you could try is to get best possible drivers for chipset, usb controller and things like that  - for this task i would use a small piece of software called Snappy Driver Installer. I've used it with good effect on a lot of system, to be safe always install max 1-2 drivers and one time and create restore points.

1 hour ago, Royal7Legion said:

gpu driver update seems to have fixed the problem. I've had this issue for a while when using the the lenovo vantage program. Driver update in the passed didn't work till now.

So everything is ok now ?

Looks like GPU drivers problem or GPU temp/power problem.

Did this start recently or is this new machine ? Any driver/windows updates before this started, any hardware changes etc ?

First step unplug and replug every power cable PSU <> mobo, PSU <> GPU, PSU <> drives

If that didn't work take out gpu from pci-e slot and plug back in, if that didn't work do the same with ram, if that didn't work do the same with cpu.

If nothing from above helped we'll  think some more.

52 minutes ago, MrChemistryCow9 said:

A frys employee recommended it to me. This was before I had my current build as the PSU transferred over. I fixed it anyways though, it was a lose power cable. I didn't know that psus were really tiered back when I got it 2 years ago. Also isn't the Smart Pro RGB a bad PSU too, as this isn't really an entry level system?

Ahh so it's from an older system, that's why it was paired with those component. Lose power cable  a bit funny and good that nothing more serious  

Smart Pro RGB would be border line but could be if system would not be OCed and pushed to the limits. That's why at first i wrote that ony DPS G is suitable from "Smart" Thermaltakes.

1 hour ago, will01gt said:

I’ve tested them together with memtest 86 and no errors

all 3 ram "combos" ? 

50 minutes ago, will01gt said:

I have keyboard mouse mousepad headset that’s it

and all of that always attached to system ? 

ditch mousepad and headset for start and test later if you can use different keyboard / mouse

edit:
make sure that all power cables from PSU are connected all the way into mobo in both placed, gpu etc - different  topic someone just found that he didn't have one cable plugged all the way

2 minutes ago, will01gt said:

Another note is that if im talking to my friends on discord they can hear me when the screen goes black as audio still plays sometimes for 4 or 5 seconds but then it cuts out

This sound like a problem I saw on those game forums while searching for black screen - most cases got resolved by  bordeless windowed mode

34 minutes ago, will01gt said:

Ok so it crashed

This is the file

log help.CSV 1.44 MB · 2 downloads

PSU is ok. Every voltage is in range. Every temp is also ok so that nothing with thermals.

Now to fully test RAM, on every step make sure that ram in in the correct slots (1 stick > DIMMA2, 2 sticks > DIMMA2+DIMMB2)

1. Switch stick in slots > memtest > game some

2. take one out > memtest > game some

3. replace with the second one (still on in  mobo) > memtest > game some

If 1 no error, no crash - test some more and be hopefull

Any errors stop testing

Crashes test next step

Extra thing - do you have a lot of usb devices connected to your system like controlers etc ? Sometimes this kind on "events" can be usb related, if possible also worth a try to use different keyboard,mouse for testing as we are stepping closer and closer to - ffs i don't  have any more ideas whats wrong.

Log few hours or at least through 2-3 crashes if they happen  and we'll see what's what

One last thing to rule out the PSU cause it is not going away from my mind and that to chech 12V rail.

Download OpenHardwareMonitor - enable logging data under options - maybe for your mobo it will show 12v on my system it doesnt so i'm searching for different software. You could use HWmonitor but that doesn't have logging option so it system crashes then we have nothing  

2 tests:

- OCCT with power test 

- launch few games

EDIT:
Download HWInfo, even a portable one, launch sensors in it. Start logging > test , game etc and we'll see everything possible that happens  

Did it go into windowed mode ? I found a topic on game forum that it sometimes doesn't work and stays on fullscreen - to fix that you have to delete config.blk file

1 minute ago, will01gt said:

Fullscreen Window?

Maybe, i don't play War Thunder. 

Try borderless window mode - quick google for black screen and apex legends/war thunder and a lot of people have this problem and window mode solves it.

Just now, will01gt said:

Only in games

2 x 8gb 3200mhz

And do you play only those 2 /only on those 2 it happens ?

Attach details of those 3 errors above/below kernel-power.

Does this happen only in games or all fullscreen games/apps ? 

Your RAM is one 16GB or two 8GB sticks ? 

25 minutes ago, will01gt said:

I’ve run both these on an hour and the highest test and nothing crashed or went above 70 degrees.

So it’s not the ram or the cup or the gpu and probably not the ssd

Its either the mobo or the psu

Which one do you think, I think psu

I also have only 1 of 2 8pins connected on the motherboard as my psu only has one

Both for and hour and the same time and no crash with "100%" system load ? Then I would say that it's not PSU fault as it proved to supply enough juice and is stable. 

1 of 2 8pins for CPU could be problematic but if it works under hour long 100% stress test then that doesn't seem like it.

CPU fine, GPU fine, PSU fine, RAM fine, mobo fine, SSD ? did you test the drive ?

Have you tried disabling windows fast startup ? 

As windows 10 updates sometimes cause troubles - did the same thing happen with the system when it was relatively fresh (without all the possible windows updates) or is this a scenario not tested ? 

5 hours ago, will01gt said:

What is Synthetic Load?

And my PSU is Bronze

As I wrote already "For example kombustor and p95" those programs are synthetic load, artificial, programs that are created to create max load and that's all.

If you launch only kombustor and GPU will be under full load and nothing crashes after 15-20minutes then GPU is in fine.

If you launch only prime95 and hit it with default stress test after 30min-1h nothing crashes then cpu is fine.

While testing keep an eye from time to time at temps, hardware has failsafes against thermal damage - like underclocking itself - but still it's not an excuse to just leave it be unsupervised. 

If you launch both of them at the same time and after a short time or immediately system crashes then that is PSU fault.

PSU in rated Bronze and is Tier B which is good but not ideal for kind of components that you have. PSU might simply be running out of breath while a power peak happens while gaming

231th place with almost 24/7 14days without using cloud on anything like that - quite nice in my humble opinion  

almost as  i had to clean it once and one stop for update of cooling with 3 noctuas

7 hours ago, trainergames said:

Yes he did, but he has had no ransomware problem at all on that pc.

May be a case that on main one it's inactive. Worth a try creating install media on another machine, full clean boot drive (zero it from bootable tool) before install.