System Crash/ Gaming

[Pez]

Hi all.

Wasn't sure which forum thread to put this, but since it involves my OS crashing, I guess here is just as good a pick as any.

OK, let me give you the background: I recently built a custom desktop system; all new parts except for one that I carried over and saved from my previous system build: the PSU I carried over and put in my new system build is a Corsair HX 1050W [and I always thought that was kind of a strange number of watts: 1050!! Why not 1000 even?!?! Or 1100?!?! ]

Now, for the rest of the parts of my new system build:
- Aorus (Gigabyte) Gaming 8 Z270
- Samsung 850 Pro 512GB Solid State Drive
- Intel Core i7 7700K Processor
- Corsair Vengeance LPX 32GB Memory
- EVGA GeForce GTX 1080 Superclocked
- Windows 10 Pro

I put this system build together sometime around the end of March, 2017. Everything was fine. Somewhere around the end of April, I loaded up a game (one of those "indie" games, a Norwegian one called "Through the Woods").

One day while playing the game, my computer system suddenly clicked off: No BSOD, it was just as if somebody pulled the power plug. But then my system is configured to automatically reboot, so it did.

I thought perhaps this was a "one-off", as it didn't happen again.

Recently I installed another game; not exactly "new", it goes back a few years: the racing game "DiRT: Showdown". I've had a couple more of these system "crashes/ reboots".

So maybe it wasn't a "one-off".

As far as all the latest drivers/ software for my hardware? Yes....and especially my graphics card (I download the latest from Nvidia's web site).

So, since these crashes only seem to happen during gaming, I'm thinking graphics card issue, or overheating.

But......I have a sealed system liquid cooler for my CPU (idle temps for the Core i7 is high 20's C or low 30's). And my case is a CoolerMaster 932 HAF (it's got lots of "breathing room" inside and many case fans, so, lots of air flow). I also check the GPU with Precision X; idle temps in the low 30's C.

And yes, I understand that temps go up during intense gaming, but these are all basically brand-new components (except for the PSU, but it's still a good, quality solid one).

After one of these crashes, if I go to Control Panel/ Administrative Tools/ Event Viewer, and then scroll though Administrative Events, yes, I see that it has noted an improper shutdown. It will have phrases like, "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.", "The previous system shutdown at such-n-such was unexpected."

But these don't tell me the specific cause of the shutdown.

When doing some general searching on the 'Net, or, specifically here on these forums for an issue similar to mine, I see suggestions like running Prime95...MSI Afterburner....SpeedFan....CoreTemp, etc.

Since I'm not getting a BSOD, how about something I've heard of called "Dump files"? Is there a way to read these that they can tell you what the cause of your crash was?

Will any of those items specifically pin down exactly what is causing my system to shutdown/ reboot?

Before this started happening, just to let you know, I also use the PowerDVD program to watch Blu-ray discs on my computer; I also use it to watch digital downloads of video files that are in 4K resolution (I have an LG 4K monitor). I imagine these various videos (discs or digital files) are giving the system somewhat of a workout (CPU & GPU), but I experience no crashes with them, only when I loaded up a game.

And....with the games I mentioned, especially the recent one I installed of "DiRT: Showdown", it's not as though I was playing the game for hours & hours, really heating things up; I had just loaded up the game and started to navigate my way through the on-screen menu when *poof*, shutdown/ reboot.

So....with all the information I've given here - system specs, etc. - what can I try? How can I pin down what's causing my crashes?

Thanks for any info or suggestions,
Pez

 

[Seanie280672]

Hi all.

Wasn't sure which forum thread to put this, but since it involves my OS crashing, I guess here is just as good a pick as any.

OK, let me give you the background: I recently built a custom desktop system; all new parts except for one that I carried over and saved from my previous system build: the PSU I carried over and put in my new system build is a Corsair HX 1050W [and I always thought that was kind of a strange number of watts: 1050!! Why not 1000 even?!?! Or 1100?!?! ]

Now, for the rest of the parts of my new system build:
- Aorus (Gigabyte) Gaming 8 Z270
- Samsung 850 Pro 512GB Solid State Drive
- Intel Core i7 7700K Processor
- Corsair Vengeance LPX 32GB Memory
- EVGA GeForce GTX 1080 Superclocked
- Windows 10 Pro

I put this system build together sometime around the end of March, 2017. Everything was fine. Somewhere around the end of April, I loaded up a game (one of those "indie" games, a Norwegian one called "Through the Woods").

One day while playing the game, my computer system suddenly clicked off: No BSOD, it was just as if somebody pulled the power plug. But then my system is configured to automatically reboot, so it did.

I thought perhaps this was a "one-off", as it didn't happen again.

Recently I installed another game; not exactly "new", it goes back a few years: the racing game "DiRT: Showdown". I've had a couple more of these system "crashes/ reboots".

So maybe it wasn't a "one-off".

As far as all the latest drivers/ software for my hardware? Yes....and especially my graphics card (I download the latest from Nvidia's web site).

So, since these crashes only seem to happen during gaming, I'm thinking graphics card issue, or overheating.

But......I have a sealed system liquid cooler for my CPU (idle temps for the Core i7 is high 20's C or low 30's). And my case is a CoolerMaster 932 HAF (it's got lots of "breathing room" inside and many case fans, so, lots of air flow). I also check the GPU with Precision X; idle temps in the low 30's C.

And yes, I understand that temps go up during intense gaming, but these are all basically brand-new components (except for the PSU, but it's still a good, quality solid one).

After one of these crashes, if I go to Control Panel/ Administrative Tools/ Event Viewer, and then scroll though Administrative Events, yes, I see that it has noted an improper shutdown. It will have phrases like, "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.", "The previous system shutdown at such-n-such was unexpected."

But these don't tell me the specific cause of the shutdown.

When doing some general searching on the 'Net, or, specifically here on these forums for an issue similar to mine, I see suggestions like running Prime95...MSI Afterburner....SpeedFan....CoreTemp, etc.

Since I'm not getting a BSOD, how about something I've heard of called "Dump files"? Is there a way to read these that they can tell you what the cause of your crash was?

Will any of those items specifically pin down exactly what is causing my system to shutdown/ reboot?

Before this started happening, just to let you know, I also use the PowerDVD program to watch Blu-ray discs on my computer; I also use it to watch digital downloads of video files that are in 4K resolution (I have an LG 4K monitor). I imagine these various videos (discs or digital files) are giving the system somewhat of a workout (CPU & GPU), but I experience no crashes with them, only when I loaded up a game.

And....with the games I mentioned, especially the recent one I installed of "DiRT: Showdown", it's not as though I was playing the game for hours & hours, really heating things up; I had just loaded up the game and started to navigate my way through the on-screen menu when *poof*, shutdown/ reboot.

So....with all the information I've given here - system specs, etc. - what can I try? How can I pin down what's causing my crashes?

Thanks for any info or suggestions,
Pez

I would say its your PSU personally, I know its plenty of power, but it sounds to me like its got a small fault, like one of the safety triggers is faulty, like the over voltage protection, or over current protection etc is tripping itself, anything else would give you a BSOD, but complete power off is a different kettle of fish, Games and heavy tasking would put a lot of preasure on the PSU too demanding more power.

Do you have another one you can try ? or access to another one.

I guess its still under warranty ?

 

[Tom Tancredi]

A 2D display (Blu Ray movie) is NOT as demanding (even if your trying to push 4K) as a 3D program (gaming) that is generating vertices, polygons, overlaying textures as fast as you see them fly by (imaging driving past a town then looking back) and so on.

While your running 'NEW' parts, it doesn't mean anything. That just means those are LESS likely, but doesn't rule them OUT from being defective nor 'you screwed up' during installation.

As you done a great job in isolating down WHEN it occurs, eliminating temps (normally the issue) on everything BUT the GPU. As this happens when the GPU is stressed (and your using the OEM fan on it) . Idle temps tell us nothing as your not stressing it. I would stress test with graphics and have a OVERLAY like MSI or such to WATCH the temps as you "build up". If you 'overclocked' I would put it at NORMAL and try also the stress test.

THIS IS NOT SOFTWARE ISSUE, so no 'dump files' will say anything to help as your alluding to.

But ultimately the other test would be with another PSU. Considering how HIGH END your build is, did you confirm your WATTS usage when under FULL LOAD? You might actually be doing MORE THAN 1050W with that particular rig (watercooled, superoverclosed, etc.) I know SOME rigs (usually with SLI) at 1250-1500W . So before testing with another PSU, CONFIRM your MAX WATT LOAD USAGE first based on your TOTAL parts (that includes drives, etc.).

 

[Pez]

Seanie280672 & Tom Tancredi: thanks for your replies.

Seems like both of you might be leaning toward my PSU

Seanie: you said, "like one of the safety triggers is faulty, like the over voltage protection, or over current protection etc is tripping itself, anything else would give you a BSOD, but complete power off is a different kettle of fish, Games and heavy tasking would put a lot of pressure on the PSU too demanding more power".

Is there a way to "test" what you're describing on my PSU? Over voltage protection? Over current protection?

And Tom: You mentioned Overclocking; I'm not doing any of that, everything is at "normal" settings. (Yes, my GPU is called "Superclocked", but that's just the name!! I'm leaving it at defaults). Same thing with CPU & RAM: no tweaking; defaults.

And you also mentioned, "did you confirm your WATTS usage when under FULL LOAD? You might actually be doing MORE THAN 1050W with that particular rig (watercooled, superoverclosed, etc.)". I'm not too schooled on this: How would I go about doing that? How do I determine WATT usage when under full load?

Thanks for any more info,
Pez

 

[Seanie280672]

Seanie280672 & Tom Tancredi: thanks for your replies.

Seems like both of you might be leaning toward my PSU

[b]Seanie[/b]: you said, "like one of the safety triggers is faulty, like the over voltage protection, or over current protection etc is tripping itself, anything else would give you a BSOD, but complete power off is a different kettle of fish, Games and heavy tasking would put a lot of pressure on the PSU too demanding more power".

Is there a way to "test" what you're describing on my PSU? Over voltage protection? Over current protection?

And Tom: You mentioned Overclocking; I'm not doing any of that, everything is at "normal" settings. (Yes, my GPU is called "Superclocked", but that's just the name!! I'm leaving it at defaults). Same thing with CPU & RAM: no tweaking; defaults.

And you also mentioned, "did you confirm your WATTS usage when under FULL LOAD? You might actually be doing MORE THAN 1050W with that particular rig (watercooled, superoverclosed, etc.)". I'm not too schooled on this: How would I go about doing that? How do I determine WATT usage when under full load?

Thanks for any more info,
Pez

To be honest, i'll be surprised if you are pulling more than 1000w, id say off the top of my head with no overclocking, probably in the region of 600w max, I know graphics cards are juicy, but not that juicy, even the top end cards only pull 250w and your CPU without overclocking only pulls about 130w, everything else is next to nothing, but you can check here by adding all your system components to the lists available: http://www.coolermaster.com/power-supply-calculator/

Only way to test really if the PSU is at fault is try another one and see if the problem still exists.

One other thing that did catch my attention though is your RAM, is that 4 sticks of 8gb or 2 sticks of 16gb ?

 

[Pez]

Seanie: My 32GB of memory is comprised of 2 sticks of 16GB.

Is that significant? Have any bearing on my "crash" issue?
Pez

 

[Tom Tancredi]

To be honest, i'll be surprised if you are pulling more than 1000w, id say off the top of my head with no overclocking, probably in the region of 600w max, I know graphics cards are juicy, but not that juicy, even the top end cards only pull 250w and your CPU without overclocking only pulls about 130w, everything else is next to nothing, but you can check here by adding all your system components to the lists available: http://www.coolermaster.com/power-supply-calculator/

Only way to test really if the PSU is at fault is try another one and see if the problem still exists.

One other thing that did catch my attention though is your RAM, is that 4 sticks of 8gb or 2 sticks of 16gb ?

Hey thanks for answering about how to calculate. Another way is also to look up your parts in www.pcpartpicker.com and it tells you based on ALL parts picked what your normal load should be.

In MY experiance to date, since the iCore 4xxx Gen / GTX 9xxx series came out, the power on / extreme load demand tend to gravitate toward 2x the 'normal load' rating (as your saying) occurs, and like the OP and MYSLEF(I had similar circumstances) had issues when building and booting initially. For example, while he has in 1kW PSU, and rated at 600W, the full 'power on' / high demand load is 1200W, which when powering on the 1kW causes it to error off by not fulfilling the complete power demands of some components (GPU, CPU, RAM, HDD) and causes different 'issues'.

Normally the way I seen that this is 'true' is, the GPU works in another system with a proper PSU, and the new system works fine in IGP, but when GPU loaded poof issues. Swap out PSU for higher quality / power and magically it all working.

 

[Seanie280672]

Seanie: My 32GB of memory is comprised of 2 sticks of 16GB.

Is that significant? Have any bearing on my "crash" issue?
Pez

No 2 sticks should be OK, its when you move to 4 sticks issues can start to arise, I just wanted to check.

You can find out truly what your graphics card is pulling by running MSI after burner along side one of the latest 3DMark bench tests.

Everything now-a-days though is built to save power, I have a 1000w PSU too, with 1200w peak, and not pulling anywhere near that, its just to keep myself safe with plenty of headroom so I dont need to worry about what im adding to my system.

 

[Pez]

So I'm good with two sticks of RAM, eh?

And, it's come up more than once in this thread, but, no, I don't have another PSU just to "test" things.

And also....I checked out that Power Supply Calculator link you gave: I entered in all my components (and even rounded up to the high end on CPU Utilization TDP 90%), and it only had me pulling around 500 watts! Go figure!!

Now......

I just contacted EVGA (the maker of my graphics card) with an inquiry and am waiting to hear back; there's something that I noticed with my GPU, and I hope this does NOT turn out to be the solution. This would be sooooo embarrassing.

OK, a little background: I've been putting together my own custom desktop builds for years. I remember when the graphics card slot on the motherboard was an AGP slot, then it went to the PCI-e.

And then I remember as graphics cards got more powerful & demanding, they started needing more power from the PSU...and sometimes more than one PCI-e cable from the PSU.

The GPU I had right before my current one was also an EVGA card; it was the GeForce GTX 770 SC. It required two PCI-e cables: one 6-pin, and one 8-pin.

When I went to put together my current system build with the EVGA GeForce GTX 1080 SC, I was kind of surprised to see the card only required one PCI-e cable: one 8-pin. So, I plugged in one 8-pin cable and that's what I've been using since.

Then I remembered something......

When I had unpacked the box that the 1080 SC came in, I remember coming across in the supplies/ accessories, a "Y" connector/ adapter: it has two 6-pins on one end, then one 8-pin on the other. I figured it was just an adapter in-case a person did not have the latest & greatest power supply.

But it got me to thinking....

Was I supposed to be using this all along? It was THIS that I just wrote to EVGA about....and am awaiting a response.

But it would have seemed kind of strange if using this connector/ adapter was what I was supposed to be doing all along. Since the "Y" connector/ adapter has two 6-pins on one end.....I could have taken two 6-pin PCI-e cables from my PSU......but, like I mentioned, this graphics card only has one 8-pin plug on it.

And the instructions that came with this particular EVGA graphics are a bit "general": It shows pictures (not actual photos, just drawings) of more than one graphics card; some pictures/ drawings show graphic cards that have more than one PCI-e plug on them (6-pins & 8-pins, kind of like my previous 770 SC card). But like I said, this 1080 SC card only has one 8-pin plug.

I'll let you know what EVGA Tech Support says when they write back.....
Pez

 

[Seanie280672]

So I'm good with two sticks of RAM, eh?

And, it's come up more than once in this thread, but, no, I don't have another PSU just to "test" things.

And also....I checked out that Power Supply Calculator link you gave: I entered in all my components (and even rounded up to the high end on CPU Utilization TDP 90%), and it only had me pulling around 500 watts! Go figure!!

Now......

I just contacted EVGA (the maker of my graphics card) with an inquiry and am waiting to hear back; there's something that I noticed with my GPU, and I hope this does NOT turn out to be the solution. This would be sooooo embarrassing.

OK, a little background: I've been putting together my own custom desktop builds for years. I remember when the graphics card slot on the motherboard was an AGP slot, then it went to the PCI-e.

And then I remember as graphics cards got more powerful & demanding, they started needing more power from the PSU...and sometimes more than one PCI-e cable from the PSU.

The GPU I had right before my current one was also an EVGA card; it was the GeForce GTX 770 SC. It required two PCI-e cables: one 6-pin, and one 8-pin.

When I went to put together my current system build with the EVGA GeForce GTX 1080 SC, I was kind of surprised to see the card only required one PCI-e cable: one 8-pin. So, I plugged in one 8-pin cable and that's what I've been using since.

Then I remembered something......

When I had unpacked the box that the 1080 SC came in, I remember coming across in the supplies/ accessories, a "Y" connector/ adapter: it has two 6-pins on one end, then one 8-pin on the other. I figured it was just an adapter in-case a person did not have the latest & greatest power supply.

But it got me to thinking....

Was I supposed to be using this all along? It was THIS that I just wrote to EVGA about....and am awaiting a response.

But it would have seemed kind of strange if using this connector/ adapter was what I was supposed to be doing all along. Since the "Y" connector/ adapter has two 6-pins on one end.....I could have taken two 6-pin PCI-e cables from my PSU......but, like I mentioned, this graphics card only has one 8-pin plug on it.

And the instructions that came with this particular EVGA graphics are a bit "general": It shows pictures (not actual photos, just drawings) of more than one graphics card; some pictures/ drawings show graphic cards that have more than one PCI-e plug on them (6-pins & 8-pins, kind of like my previous 770 SC card). But like I said, this 1080 SC card only has one 8-pin plug.

I'll let you know what EVGA Tech Support says when they write back.....
Pez

Sorry for the late reply, my computer has been down all yesterday afternoon and evening having a rebuild into a new case and leak testing.

I know you say this cable has only 2 x 6 pins, this kinda makes sense to me but at the same time doesnt, are they 6+2 pin connectors ? whats on the other end of this cable ? (connector type), can you post a picture of it ? the reason it has 2 lots is so if you are running SLi, rather than running 2 lots of cables from your psu to the graphics cards, you can use 1 and use that cable to supply power to both cards.

 

[Pez]

Hi Seanie; thanks for getting back to me.

Dang, your computer was down?!?! And you were testing for leaks? I take it you have some liquid cooling going on in your system build? Is that what you meant by "leaks"? Or are you talking "power/ voltage/ wattage" leaks?

I've got a liquid cooler for my CPU, a Corsair H75 sealed system.

Anyway, this "Y" cable I mentioned? I'm pretty sure I don't need it; I think it's for people who don't have an 8-pin power cable coming out of their PSU. They would take two 6-pin cables from their PSU, plug it into the adapter, then the other end is an 8-pin, and this would plug into the graphics card; here's a pic:

So....I heard back from EVGA Tech Support; here's what they said:

"I do apologize for any inconveniences you may be experiencing. To answer your question about your power supply, yes that power supply is more than capable of handling your system's power needs.

There could be any number of things that can be causing your system to suddenly shut down, it can be your memory, graphics card or your power supply. If you are doing any sort of overclocking on your memory or if you are using a XMP profile for your memory I recommend that you load the optimized defaults in your BIOS and restart your computer. Proceed to play a game, if your computer does not shut off, there may be a memory setting that may need some tweaking.

If that is not the case, I recommend that you check the 12v rail on your power supply to rule out any potential power issues that may be causing your system to not work properly. What you are looking for when checking the 12v rail is any unusual fluctuations. If the 12v rail's rating is fluctuating constantly or if the rating is higher than 12.9 or lower than 11.4, then most likely there is an issue with the power supply that is making your graphics card to act abnormally. Here is a link on how to check the 12v rail on your computer through the BIOS screen, link HERE.

Finally if the above two does not solve your shut down problems, I do ask that you uninstall and reinstall your drivers to the latest version. To do that in the recommended way, you would first download a program called, DDU. A download to DDU can be found HERE. When you start up DDU for the first time, select on the Clean and Restart option. After your computer has been restarted, go onto our EVGA driver downloads page, a link to that page HERE, and download the latest driver file. Once you start the driver installation you will be asked to select an installation method, select the advanced option and choose the clean install option. Once you finish installing your drivers, restart your computer and see if this helps your issues."

They also mentioned in their reply that if none of these suggestions work, that I could look into warranty options (I think they're referring to an RMA; I only purchased this GPU about a month and-a-half ago).

Of the items suggested, I started with an easy one: I downloaded and used that DDU (Display Driver Uninstaller). I then installed the most recent display drivers from Nvidia (just released on 5/4/17). Then I fired up the game of "DiRT: Showdown". I played for a little while and no shutdown/ reboot. But....

But I'll say this: When this shutdown/ reboot thing started to happen recently, it did not happen consistently, like every time I played a game. Sometimes I could play for a half-hour/ 45 minutes, and no problem. Another time I started the game, was navigating my way through the on-screen game menu, and *poof*: shutdown/ reboot.

So, I'm not saying that this complete uninstall & then clean install of the display drivers "fixed" the problem; time will tell.

They also mention overclocking & XMP: nope, none of that going on with my configuration in my BIOS; it's all system and components' defaults.

While in my BIOS, as suggested by them, I checked my 12V rail voltage: 12.168. Normal, acceptable voltage, correct?

So hey: if the driver uninstall/ clean install did the trick, then great!! But like I say, time will tell.

You have any other thoughts?
Pez

 

[Seanie280672]

Hi Seanie; thanks for getting back to me.

Dang, your computer was down?!?! And you were testing for leaks? I take it you have some liquid cooling going on in your system build? Is that what you meant by "leaks"? Or are you talking "power/ voltage/ wattage" leaks?

I've got a liquid cooler for my CPU, a Corsair H75 sealed system.

Anyway, this "Y" cable I mentioned? I'm pretty sure I don't need it; I think it's for people who don't have an 8-pin power cable coming out of their PSU. They would take two 6-pin cables from their PSU, plug it into the adapter, then the other end is an 8-pin, and this would plug into the graphics card; here's a pic:

So....I heard back from EVGA Tech Support; here's what they said:

"I do apologize for any inconveniences you may be experiencing. To answer your question about your power supply, yes that power supply is more than capable of handling your system's power needs.

There could be any number of things that can be causing your system to suddenly shut down, it can be your memory, graphics card or your power supply. If you are doing any sort of overclocking on your memory or if you are using a XMP profile for your memory I recommend that you load the optimized defaults in your BIOS and restart your computer. Proceed to play a game, if your computer does not shut off, there may be a memory setting that may need some tweaking.

If that is not the case, I recommend that you check the 12v rail on your power supply to rule out any potential power issues that may be causing your system to not work properly. What you are looking for when checking the 12v rail is any unusual fluctuations. If the 12v rail's rating is fluctuating constantly or if the rating is higher than 12.9 or lower than 11.4, then most likely there is an issue with the power supply that is making your graphics card to act abnormally. Here is a link on how to check the 12v rail on your computer through the BIOS screen, link HERE.

Finally if the above two does not solve your shut down problems, I do ask that you uninstall and reinstall your drivers to the latest version. To do that in the recommended way, you would first download a program called, DDU. A download to DDU can be found HERE. When you start up DDU for the first time, select on the Clean and Restart option. After your computer has been restarted, go onto our EVGA driver downloads page, a link to that page HERE, and download the latest driver file. Once you start the driver installation you will be asked to select an installation method, select the advanced option and choose the clean install option. Once you finish installing your drivers, restart your computer and see if this helps your issues."

They also mentioned in their reply that if none of these suggestions work, that I could look into warranty options (I think they're referring to an RMA; I only purchased this GPU about a month and-a-half ago).

Of the items suggested, I started with an easy one: I downloaded and used that DDU (Display Driver Uninstaller). I then installed the most recent display drivers from Nvidia (just released on 5/4/17). Then I fired up the game of "DiRT: Showdown". I played for a little while and no shutdown/ reboot. But....

But I'll say this: When this shutdown/ reboot thing started to happen recently, it did not happen consistently, like every time I played a game. Sometimes I could play for a half-hour/ 45 minutes, and no problem. Another time I started the game, was navigating my way through the on-screen game menu, and *poof*: shutdown/ reboot.

So, I'm not saying that this complete uninstall & then clean install of the display drivers "fixed" the problem; time will tell.

They also mention overclocking & XMP: nope, none of that going on with my configuration in my BIOS; it's all system and components' defaults.

While in my BIOS, as suggested by them, I checked my 12V rail voltage: 12.168. Normal, acceptable voltage, correct?

So hey: if the driver uninstall/ clean install did the trick, then great!! But like I say, time will tell.

You have any other thoughts?
Pez

Leak testing as I upgraded my water cooling setup and put the hole system into a new case, Im custom water cooling, and everytime you change something, its advised to run it for 12 hours checking for leaks without power to anything else except the pump, its that or take a risk of a drop of water finding its way out and killing something in the computer.

That cable is to work the other way around, it converts 2 x 6pin connectors from your PSU into an 8 pin incase you dont have 8 pin with your PSU, as you do, you dont need it.

Your 12v rail is at a perfect voltage where it is.

Glad its sorted now, I guess a bad driver or a bad install or download caused the original problems.

 

[Pez]

OK, it's been a while, but I'm back.

Grrrrrrr

It seemed like things had been good for a while (no crashing); then again, I'm not that heavy of a gamer, I don't game every day.

In my last post, the last thing I had done was use DDU (Display Driver Uninstall); I then did a "clean" installation of the most recent Nvidia drivers; things seemed good for a while.

Then recently, a couple of things:

I launched the racing game "DiRT: Showdown". As the game starts to load, I hear what I think is a buzzing sound coming through my computer's speaker system (I have a quality, 5.1 Surround Sound System by LogiTech). Usually when the game appears onscreen as it launches, there's the game's music and some sound effects. And then I realized: I had forgotten to turn the speakers on, so, this "buzzing" sound that I'm hearing is not coming from the speakers, but, rather, from my computer's tower case that's on the floor.

I immediately quit from the game and exit back to my desktop. As soon as I do, that buzzing sound stops.

My monitor is on, displaying images properly, so yes, the graphics card seems to be working, but I get down on the floor where my computer's tower case is and look inside (it has a window on the side). I can see the graphics card, and then I look down lower, and then look up from underneath the graphics card. As some of you are already aware, the GeForce GTX 1080 SC has two fans on the underside.....and as I am looking at them....both are standing still and not spinning at all.

I write to EVGA tech support, and here's what they said:
"For the ACX 2.0 and 3.0 cards, it's perfectly normal for the fans to stop when the card is under load."

Was that a typo by them? Did they mean to say, "...it's perfectly normal for the fans to stop when the card is NOT under load"? To me, what they said is counter-logical, or counter intuitive: if the graphics card is under a heavy load (from a graphics-intensive game), then you'd think the fans would speed up to help keep the card cool.

Or am I reading it all wrong and I don't get it.....

Then, as far as that buzzing noise I heard? Here's what they said:
"The buzzing is coil noise, commonly called "coil whine". Coil whine is something that can happen with any high-power card or power supply, or anything that uses solid-state inductors for voltage regulation. While the whine, clicks, or buzzes won't affect the performance of the card, you can try to reduce it. Check first that the whine is not coming from the power supply. Use a rolled-up paper or cardboard tube to isolate the sound. You can also try enabling v-sync or G-sync to prevent the card from rendering over 60, 120, or 144Hz, as it's the extremely high frame rates that contribute to the noise."

OK, evidently this buzzing sound, too, is "normal"; perhaps it was there all along but I never heard it because I had my speakers on and the game's music/ sound effects covered it up.

But....that suggestion of theirs of using a "tube" to isolate the sound? Do they mean put one end of the tube to the GPU and the other to my ear? Then put the tube to the PSU then the other to my ear? Try to pinpoint where the buzzing is coming from?

And then what about the vsync suggestion?

In many of the games I've played through the years, the in-game graphics options usually let you either to Enable or Disable Vsync, but what they EVGA tech guy said was to "try enabling v-sync or G-sync to prevent the card from rendering over 60, 120, or 144Hz, as it's the extremely high frame rates that contribute to the noise". How do you enable it on the graphics card itself? Do you have to use something like the Nvidia Control Panel?

OK, speaking of Vsync in general: If the EVGA tech did indeed mean to tell me to enable vsync on the card itself (which, if he did, I hope you'll tell me how to do!!), is there any significance to this when displaying to a 4K monitor?

I have an LG 27UD58 Ultra HD 4K monitor; it's set at its native resolution of 3840 x 2160. Could some of my issues be related to this? Something happening between my EVGA Nvidia GeForce 1080 SC graphics card and the LG 4K monitor?

And then, besides the non-spinning fans on the graphics card and the coil buzz/ whine (which the EVGA tech said was "normal"), I launched "DiRT: Showdown" again and, while only on the in-game menu and clicking around......my system crashed/ and rebooted again.

I tried using a little program called "WhoCrashed", but it's not finding anything. It says that Crash Dumps are in-fact enabled on my computer, but, no valid crash dumps have been found.

Any more insights anybody? Any other suggestion or ideas from the info I gave above?

Thanks,
Pez