Instability (reboot cycle) only while gaming - HWinfo, CCleaner provide a clue?

I'm thinking a PSU issue as it seems to be crashing quickly even outside of games. Do you have a spare PSU to try? Or one you could borrow?

Another possibility is memory. You could download memtest and run it. Let it run like 8 hours. If the issue is PSU then it will likely crash before this.

Sorrry when I seen "and I couldn't even get into the start menu..." I assumed you meant the Windows Start Menu.

You could stress you CPU fully. Something like Cinebench on the CPU stress, or Prime 95. This will isolate the GPU from the equation. If it's PSU related, it should reboot when you put a heavy load on the CPU and thus the PSU. If it's stable, then it could be GPU related. There isn't any software that specifically tests the PSU. PSU problems manifest most often during high load conditions, and much less common during very low power states. It's very difficult to isolate the PSU from the rest of the system. The best way to eliminate the PSU is using another known good PSU and if the problem is gone, then the PSU is at fault, if the problem remains then you keep troubleshooting.

If it's only unstable during GPU related task, then most it's most likely GPU related. That's not to say it couldn't be software / driver related. If you find that an old GPU works, then you might want to try the "faulty" GPU in a known stable system. It could still be power related as your 1080Ti is by far the most demanding component with respect to power usage.

I'm leaning towards the GPU, but you want to be sure, especially when it comes to warranty.

Do not discount heat as an issue. That's not to say the component isn't faulty, but increased heat could be the symptom. You don't mention which component is 72 degrees, is this the CPU or GPU? Remember it doesn't need to be either the GPU or CPU, hot PSU's can shutdown too.

I definitely don't like tower cabinets in a desk. Typically it traps air around it, so while your case is blowing the hot air out, it doesn't get a fresh supply of cool air in and just ends up sucking that hot air back in through the intake. Ambient temperatures are important to cooling efficiency. It's hard to keep your parts cool when the "cool" air is 45 - 50 degrees already. I've seen all sorts of issues both in my personal and professional life caused by restricted airflow.

I would also look at one of my previous suggestions. If you think it could be a flaky power connection, if you have unused PCI-E power cables for you PSU, replace the ones you are using now with the spares. You could have a contact or contacts that are faulty. The faulty connection could be at either end of the cable, either the PSU side or the graphics card side.

That symptom 1 really makes me think that it could be temperature related, though it doesn't necessarily narrow it down to the GPU. If the PSU is overheating or sensitive to heat, power (most likely 12V) to your GPU could be fluctuating.

The first thing I would do to test the overheating theory is to remove it from the tower alcove in your desk. Take the side off of the case as well. Try to not disturb the cabling and system as best as you can. If the system is stable after a good run, put the side back on the case. If it continues to run, then you know it's the limited airflow where you put your case that is the issue.

It should also go without saying that you need to make sure that the radiator on your AiO is be clean and so should your heatsink on your 1080Ti. If they aren't, you need to blow the dust out of them. Also if your case has filters, make sure they are blown out. This is the most commonly overlooked cause to issues like these. Most people mistakenly think that because the instability came on all of a sudden, it can't be heat related as they think it should have slowly become worst over time.

I would do all this before introducing your friends GPU to your system. In fact if your system remains unstable after the above, I would test your 1080Ti extensively in your friends computer before putting his in your system. We still haven't determined your issue and you don't want to potentially damage his GPU if say your PSU is faulty. If your GPU works perfectly in his system, then I would look to your PSU. In fact you could see if your friend would loan you his PSU (huge pain in the ass for both of you) to test your card in your system.

Everyone's approach to troubleshooting is different. One rule of thumb I stress is, if you are using someone else's components to test, do as much as you can without endangering their components. You could turn a once helpful friend into someone that doesn't talk to you.

You won't be able to monitor the temps inside your PSU. There are a couple very high end PSU's that support this, but most do not.

Yeah changing out PSU's is a pain especially if you've done a meticulous job of cable management.

While I know you hope that it's not your graphics card, it's the easiest one to test. In this case if it works in your friends system, then you can be certain it's not at fault. Though to limit the time it takes to troubleshoot, it might be better if it fails in his system. This way you know right away and can start working towards a solution.

You may see a flashing when Windows is booting up. This can occur when the OS is booting up. The graphics card behaves like a basic VGA adapter with features and functions that all graphics cards support (kept in the BIOS of the card), when the OS loads the driver for your actual card, flickering is common when this driver takes over. It could even become more pronounced after a driver update if something was changed that affects how the driver initializes the hardware.

Air circulation is important. It's odd that your case is designed so that your air intake is impeded by having the front panel closed. For instance my case has a door over the front, but my intake fan is on the bottom of the case.

As for fan arrangements I tend to like positive pressure inside my case. This means that you have more fans drawing air in than fans exhausting. It's really half dozen one, six the other, I'm not sure it makes much of a difference to cooling. It can have an affect on the amount of dust gets in your case though. Most good cases have filters on the intake fans so air drawn in by them is generally speaking, dust free. However if you have negative air pressure inside the case, it will pull air in from anywhere it can which means it can come in from any crack or vent that isn't filtered.

As for where your exhaust fan was plugged in, academically (meaning maybe not necessary) it should be in a Chassis fan header. That said the speed of the pump header is typically controlled by the CPU temperature, so it should have been speeding up and slowing down in response to temp changes in your CPU which could in affect act quicker to load changes than a Chassis fan header that could be using a sensor located on the motherboard itself.

As for putting your liquid cooler in the top fan positions, this arrangement is pretty typical. Hopefully you have the fans on this cooler so that they are blowing the hot air out the top of the case. If you have them pulling air into your case, then you might want to turn them around. This way you are pushing that heat out of your case.

If you have filters on your intake fans, make sure they are clean. Make sure you've blown the dust out of your AiO, as well as the heatsink on your graphics card. Also blow the dust out of your PSU, there are heatsinks in there for the active power components (maybe some passive components too) that can get dusty and it's not as easy to see as the rest of your components. Dust is the enemy here.

Wattage wise 650W should be enough from a good quality PSU. Do you know what model of SuperNova you have? Is it a G2, P2, etc?

In general EVGA makes pretty decent supplies, however that doesn't mean you can't get a dud.

As for going with a higher wattage supply that is up to you. According to my quick calculations (making some assumptions about the rest of your system) I come up with 480W peak, so something in the range of 550W would suffice if it's good quality. Now if you are overclocking anything, you need to go up from there. The 12V amperage is really what matters. The SuperNova G2 can supply 54A on the 12V rail which is plenty providing the total system power usage doesn't exceed the rating of the PSU.

So I'd try to RMA the PSU first. I don't know if you have to go as far as 850W, maybe 750W if you want a little more wiggle room. That said I'm riding a 4770K with a single GTX980 with a SuperNova 1000W G2, so who am I to talk. Actually I was running it originally on an old PC&C 750W Quad Silencer, but the EVGA supply was on sale and I couldn't pass it up.

Power limit on your GPU really doesn't concern me. Essentially it's the ceiling of power the card can use before it starts to down clock to stay under that limit. It's part of TurboBoost, it works alongside the thermal limit. So lets say you have a game that exceeds your set power limit (old games that pump out huge framerates are a good example), TurboBoost will step in and start lowering the GPU clock until it gets within the limit. It's really an attempt to protect the VRM's on the graphics card from heat. This is one of the sliders that get adjusted when overclocking. For instance nVidia lets my GPU run a power limit of 125%.

Hey sorry for the delayed reply, I haven't logged in in awhile.

Good to hear, and hoping since you haven't updated it since Nov 18th that it's still working well for you.

A couple things regarding you GPU temps:

First you mentioned cable management, if you have cables obstructing good airflow, you might want to have a look at this first.

Next, make sure you have the heatsink thoroughly clean. Focused blasts of canned air is usually good at removing most of the dust and cruft that builds up. Be careful with the air around your fans though. I usually try and keep the fans from spinning as a result of sustained blasts of air as it can cause the fan to spin very fast which could potentially damage the bearing in the fan(s).

Lastly look at your case fan placements, the direction of airflow, and the possibility that you may need to add fans if you have the mount spots available.

Though that all said, 75 degrees isn't that hot.