Clean the dust out of your CPU fan (and the rest of the computer), remove the CPU and renew the thermal paste.
Comment has been collapsed.
Good suggestion, but this pc is totally clean. I don't think its from my cpu temps as I raised them higher with prime95 then shadow of mordor could do and it didn't turn off then. The fans on both cpu's run fine and hit about 4500 rpm when under load.
Comment has been collapsed.
Try running another game, see if you still get the shutdown.
Do you have any monitor software installed? Anything Overclocked?
EDIT - missed the realtemp post, I prefer HWID it gives a lot more info
Comment has been collapsed.
I can play 7 days to die without problem but I dont believe that game pushes my graphics as hard as shadow of mordor.
I did install HWMonitor also cause I wanted to check on the voltages of my PSU but it only shows me the 3.3v which is normal.
Nothing is overclocked.
Comment has been collapsed.
Have you tried running FurMark or anything like that yet?
edit - ok I missed that also, I blame the beer. I think tzaar is on to the right thing and sober listen to him.
Comment has been collapsed.
Its all good :) But he could be right. Im not sure off hand but I think my old pc's psu had like 30a on the 12v rails so that could be it.
Comment has been collapsed.
Not sure if you checked this but you split up the 12V rail connections properly right?
Comment has been collapsed.
I was reading about this earlier, but I'm not sure exactly how to do this as I dont know which connections are on which rails.
I did try switching some around with my hard drives and dvd but it had no effect. It still could be the problem but I have no idea how to go about doing this correctly.
Comment has been collapsed.
I don't know how your particular power supply is split, but basically each rail has a maximum amount of power it can supply. So ideally it would be something like motherboard power on rail 1, CPU power on rail 2, graphics card power on rail 3. You probably can't choose the motherboard or cpu cable but the PCIE cables might be on separate ones.
Comment has been collapsed.
Ya this PSU just has a huge mass of wires coming out of 1 spot, so I have no way of knowing just by looking at it. I did google the psu trying to find some info on that but no luck.
Comment has been collapsed.
5 12 volt rails at 16a isn't a gaming PSU. You'd be better off with a single/dual high-amp rail(s). If you have GPU power connectors on two different rails, try connecting two rails to the GPU.
Outside of that, try pulling/disconnecting everything you possibly can (extra RAM, HDDs, soundcards, etc) and re-add them one by one to determine the problem.
IMHO though, it's your PSU.
Comment has been collapsed.
The PSU does have connectors for 2 graphic cards, two 6 pin and two 6+2 pin.
I'll try connecting one of each to the card and give that a go.
Also, it was the first psu I had that had 2 16a and 3 18a rails. The current psu im using has 18a on all 5. This still too low?
Comment has been collapsed.
As long as they are split, that should be fine. The idea of one gigantic 12v rail is a myth that often results in fires. I am an electrical engineer and have studied these things.
Comment has been collapsed.
Hello StonerSmurf, if you're sure that the problem is not due to overheating, then you may have problems with driver or be due to malware. Do you notice strange load or unfamiliar processes?
Comment has been collapsed.
No, I just recently set up this PC and its a brand new install of windows. The only programs I have are chrome, steam, the temp monitor programs, teamspeak and media player classic. Everything is actually lightning fast and no unknown or odd processes running.
Comment has been collapsed.
I see. Actually, chrome, for me personally, classed as malware. This is rather a joke, although I would never installed it on my Linux machine. However, it is better to do a quick check with Malwarebytes Anti-Malware https://www.malwarebytes.org/.
Comment has been collapsed.
Is it just directly turning off without any error messages? I'd personally check the event viewer for any errors. But if it's just turning off directly, it's probably the power supply and tzaar is your guy.
Comment has been collapsed.
The power turns off, but the power light on my case stays on. But everything is shutdown. Im going to look at the event viewer now.
EDIT: The event viewer just registers that the PC was not shutdown properly.
"The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly."
When I look for logs around the time of the crash, there are none.
Comment has been collapsed.
Hrm, in my experience when it's the power supply everything will go off, including lights even if only momentarily. Definitely take a look at the event viewer and see if anything is awry.
Comment has been collapsed.
If it is power supply, why computer does not switched off and when he run tests?
Comment has been collapsed.
That's what I was thinking. You'd think it would shut off while running furmark since it's shutting down while gaming at 100% GPU load.
Comment has been collapsed.
If I am overloading a 12v rail, it could be the one connected to my gpu and hard drives. The game is installed on my raid setup (the 4 hd's) and maybe when those 4 draw power and the gpu at the same time? I'm not sure.... I think my next test will be to run furmark and then copy a bunch of files from the hd and see if that can crash me.
Comment has been collapsed.
You shouldn't be using Furmark ... no one should. Reviewers stopped using it years ago because of its very limited uses and destructive potential. Real world gaming is the single best test for system stability, as it stresses more components with more features.
Download something like the free version of 3DMark and run Firestrike at the highest settings you can. It's still my guess that your PSU isn't "feeding" the GPU what it needs for feature-intense games. It sounds like you have a weak/bad rail or possibly bad ripple under full load, which could cause the PSU to kick into standby mode.
If you have a multi-meter, you might want to check the GPU power connectors to be sure they're within spec and delivering the power they say they are.
Comment has been collapsed.
I tend to recommend 3DMark as well. I don't like furmark, I haven't used it in a few years. I was only stating it since he said he tried it.
Comment has been collapsed.
I installed the free 3dmark and ran firestrike like you said.... in about 1 min it turned off again... exactly the same way as shadow of mordor. I don't think the PSU's are faulty... there just not made to power my video card.
Given my setup, which would you recommend? Also figure in a few external hd's, charging my phone, webcam...
My motherboard is "Skye - 2P Intel Socket LGA771, Seaburg, ESB2" and requires a 24 pin, 10 pin, and an 8 pin power connectors.
I'll also need at least 6 Sata power connectors, PCI-E 6 pin & 8 pin.
Comment has been collapsed.
I'd recommend a good 80+ Gold PSU from any of the following manufacturers -- Corsair, EVGA, Seasonic, Silverstone, or Antec (TruePower line preferred). You don't need anywhere near 1000w, either - 750w or so would be more than enough (and even a little overkill) for your server/gaming rig.
Your primary concern is going to be the 12v rail(s). You want something with a single (or dual) with at least 60a on it. Your hard drives, dvd drives, and whatever else draw negligible power, so don't concern yourself with their power draw (a HDD draws about 5w on average)
I didn't check every connector you listed to be sure it has enough, but I think this power supply would be an excellent place to start, and will only set you back $140.
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139137
EDIT: I just did a quick search on Newegg for that one - there are many, many others that will probably fit your needs, so have a look around some. Good luck, and I hope you get things back up and running soon! Happy Gaming!
Comment has been collapsed.
Ah wow, I looked a little closer, and you need a ten-pin PSU connector for a socket 771 mobo?
You may be out of luck, my friend ... those 771 motherboards haven't been made for 6 years or more, and finding a power supply to fit that motherboard AND a contemporary GPU might be an issue. Your only option may be to upgrade to a more recent mobo/CPU. On the plus side, that gear you have now would make an excellent home server to go along with a nice gaming rig.
Sorry to bring you bad news like this ...
Comment has been collapsed.
Ya... I was just looking at the PSU you recommended and at first I thought it had the correct connections but after counting pins, it doesn't.
I need a 24-pin, a 10-pin, and an 8-pin just for the motherboard. I never seen these before until I got these systems (my first server system) and I really might be SOL. :(
Comment has been collapsed.
Well, that's just kinda a WTF moment there...
At first, after reading your issue, I was thinking heat. Then you ruled heat issues out for the most part, so my best guess was that your GPU under full load was likely browning out your PSU because you were drawing too much power across a single rail, but if that were the case it should've done exactly the same thing under full load from benchmarkers too. So now I am just wondering wtf could be the issue when it's exclusive to Shadows of Mordor, and that leaves me assuming some kind of issue with the game itself and how it's interacting.
I would note that nothing about that system is designed for gaming, as that system is running a server design, down to running dual server CPUs. Also, how on earth does it have 14GB of RAM, but that's 8 sticks? Crazy damn hardware manufacturers using improper RAM setups... 8 sticks should be 16 GB (or 32 if 4GB sticks), and if it were tri-channel instead it should have 6 or 9 sticks, and be 12 or 18 (or, ya know, 24 or 36 if it was all 4GB chips instead). So that's just wonky as hell on it's own. I don't even know what they did to get to such an odd number there. I can only assume that it's set up with something like 6x2GB and 2x1GB or something strange like that...
Also, is it fully powering off during this, is it restarting itself, or is it doing something odd like ending up in Sleep/Hibernate mode? Have you checked your power saving setup too? I mean, the power saving stuff isn't SUPPOSED to kick in when anything is actively running, but it's a possibility - eg: the system is somehow not registering that SoM is running, so you play it for a while and then it tries to turn off the screen and/or hard drives for power saving, but the program is still running, so crap just freaks out and then the system ends up powered down, etc.
All I can really assume at this point is that it sounds to be software related rather than hardware related since the issue isn't replicated under other near identical conditions with other software responsible for taxing the system.
Good luck, this one sounds kinda tricky.
Comment has been collapsed.
I was actually looking into the specs myself just now. I was looking at what Lenovo offered for ThinkStation P options, and none of this is really selectable. The CPU is from '07 according to Intels website, the memory doesn't really add up either, and most ThinkStations seem to come with Nvidia Quadro. I'm wondering how much of this is prebuilt, and how much of it is a custom build gone wrong.
StonerSmurf, could I get a model number for this? None of it really adds up, so it might help.
Comment has been collapsed.
Yeah, I was wondering about that myself. When I went and looked up older models with the Xeon E5430, all I found were single CPU options too, rather than dual. But I didn't exactly dig that deep, so that could easily be inaccurate. Mostly I noticed the E5430 was quite old, and clocked at only 2.66Ghz, but it doesn't sound like it would necessarily be an issue so I discounted it, more or less, and moved on.
The age of the hardware makes me question the viability of the PC as a whole, and I'm confused as to how it could be new with such odd and aged components, but it doesn't necessarily relate directly to the issue at hand here, given that other programs with similar load don't create similar issues.
One thing I would like to add as needed info though is what Windows he's using. He mentioned a brand new install, but not what Windows OS it is. That could also be creating some software issues, if, say running it in Vista which it's not designed to run in, for a random example.
Comment has been collapsed.
I actually wouldn't be surprised if he's running the Windows 10 pre-release. If that's the case, problem solved. I know someone who has had similar issues with it.
Comment has been collapsed.
sorry forgot to post that info in the op, Im running Windows 7 Professional 64bit
Comment has been collapsed.
Judging by the fact that your PSU seems to be non-modular, and it only turns off at extremely high loads along with the temps being fine, I'd definitely say it's the issue. I'll leave it to tzaar to help you, power supplies aren't my strong point.
Comment has been collapsed.
I actually received 2 of these thinkstations, its model is MT-M 6493 - WHZ. I took the cpu from one and put it in the other. Some of the ram also and I think your right about the config as some of those are 2gb sticks and I think only 2 of them are 1gb. I did put the 1gb sticks in the same channel. I didn't think having an odd number or mix in sizes would be bad? I also have the quadro cards that came with it but those aren't gaming cards which is why I pulled them and put my R9 280 in it.
When it crashes, its as if someone flipped the power switch. My monitor says no signal, all the fans stop, the front panel usb lights go out. The only thing that still has power is the power switch light. In order to reboot the pc, I have to hold the power button so the light turns off, then press the power button to restart the system.
Comment has been collapsed.
That RAM config could be your issue right there actually.
You've mixed and matched RAM from wholly separate machines, which are likely not at the same clock speeds and almost guaranteed to be different voltages and latency timings. What could very well be happening here is that the benchmark programs are not hitting your RAM at nearly the same level that Shadows of Mordor does, so when you start playing SoM and it inflates more files to the RAM, further filling it, it ends up encountering an issue with the discrepancy in sticks, or just that a stick may actually be bad in general, all hell breaks loose, and the system crashes out.
Alternatively, the same could be true of one or the other CPU. Though that feels less likely since you've hit similar usage levels on the other programs, it's still quite plausible. Most games are coded up to Quad Core usage at most for multi threading. The benchmark programs may be hitting similar usage levels, but spread fully across both CPUs. SoM will basically be focusing entirely onto one, so it may be maxing all 4 cores of one CPU, tripping a fault, and BAM insta crash, while the other programs don't cause the issues because they are functioning differently with the CPUs.
I'm not sure where you got these from, but I would warn that used components are always highly risky, and combining two different machines is just as risky on it's own, if not more so. I hope you are able to resolve your issue, but I would advise, as a long time tech, never to do something like this.
Comment has been collapsed.
If the ram config was a problem, wouldn't other programs also be affected?
Comment has been collapsed.
I think you need to tighten the Flux Capacitor....and you need to upgrade your powersuply to 1.21 gigawatts.
Comment has been collapsed.
I was thinking bad memory, but if that was the case, I think you would have Windows crashing and not having the computer powering off.
Maybe a problem with the power or reset buttons on the computer case? I think my power button has some dust in it and my computer will randomly turn itself on so I have to manually flip the power switch off on the back of my PSU after I turn it off.
You can try this as a test to see if it is GPU or power related and maybe use it as a temporary workaround. Limit your fps while in that game so that your GPU isn't maxed out. I'm not sure what program you use to control your video card, but I run MSI Afterburner and it has an option to manually limit your fps to any number you want. You can have MSI Afterburner track your FPS and tell you what you are getting. Then you can manually set it to something lower like a value that would only use 50% of your GPU. If it still crashes, then it's most likely not related to your video card or PSU being maxed out. If it does not crash, you can turn the fps limit up until you get it to crash again and then turn it down a little. This will not fix the problem, but will help diagnose a little more and let you play the game if you don't find a solution.
Edit: If you can run it at a lower fps without crashing, I wouldn't keep it running very close to the crashing point for prolonged periods of time. Something is wrong and if you stress the part that is failing to the point where it doesn't fail, but is constantly running at the max amount of stress it can endure, you may cause more serious problems.
Edit 2: You can also try turning down a bunch of the in game video settings and see if that has an affect.
Comment has been collapsed.
I've dealt with both RAM and CPU faults that cause machines to act as if they have lost power - either just turning fully off, or restarting themselves. Occasionally issues that cause it to go to a black screen, but still be semi-powered up where fans and the like run but the system itself isn't fully. You are most likely on the right track there thinking it is memory related, imo.
Comment has been collapsed.
Partially why I recommended checking the event viewer as well. Normally memory errors causing stuff like this will have something show up there.
Comment has been collapsed.
Ya I checked the event viewer and there is nothing at or near the time of the crash. I wouldn't think a faulty cpu or memory would have such a specific and repeatable failure. You would think if one of those parts was bad then any program running would be affected, including windows. But I can browse the web, watch netflix, play videos and music locally and no problems.
Comment has been collapsed.
I wouldn't rule out RAM error though. If you have that 12-ish hours, boot up a MemTest stick and let it run everything. Because what you described happened to me twice. Once was a faulty mobo where one circuit just couldn't handle the load somewhere around the memory bus. (I never really investigated it, nor pursued the workshop to give me a full report, they just changed my mobo). The other one was a bad memory stick, which only happened after it reached almost 80% of RAM usage.
Comment has been collapsed.
Ya, I should run memtest anyways just to be absolutely sure. Thanks :)
Comment has been collapsed.
It really does sound like the gpu overheating. I had a very similar problem that I solved with reseating the gpu heatsink. Mine was only with one of the batman games though - other games seemed not to worry it.
Comment has been collapsed.
A lot of people seem to have problems with this game crashing. Some of them have the game crashing to a black screen and they can still hear audio and others have the same problem as you with their computer turning off or restarting. From the limited amount I have read the general consensus seems to be to limit your fps to 30. Pretty crappy solution, hope there is a better one.
Comment has been collapsed.
I just realized you are running a computer with 2 cpu's and 8 sticks of memory? I don't know anything about computers that run multiple cpu's, but maybe the 2 cpu's and 2 banks of memory is confusing the game somehow. Is there a bios option to disable 1 of the cpu's and 4 stick of memory temporarily to see if that helps?
Comment has been collapsed.
There is only 1 memory bank that both cpu's share. I'm not sure about the bios option but I don't recall seeing anything like that.
Comment has been collapsed.
So, I turned shadow of mordor down to the lowest settings and lowered the screen resolution and so far so good, no crash after 20 minutes of playing. According to HW Monitor the max gpu utilization was 95%. Oddly cpu utilization is a tad higher at the lower settings by about 10%, but that might be what i'm doing in the game.
Comment has been collapsed.
Would it be safe to connect my video card to the power supply in my other pc for a short time to test? I know I would have to turn both on for it to send power but just to test the game at max settings which is where the crash seems to be happening for a short time.
Comment has been collapsed.
You have PSU with 5x 12V rails. 1rail = 18A - This is not enough for your GPU.
When you get PSU with 1x12V rail about 80A = 960W or lets say it is 1000W power supply you will be just fine. Not sure how your CPU's, ram and 5 HDD's will work but your GPU will work just fine.
Comment has been collapsed.
That's why I'm asking if its safe to connect my video card to a different PSU while the rest of my pc is on the current PSU. This video card worked fine in my old system and I put many hours into shadow of mordor on it.
Comment has been collapsed.
Sure that is safe, you can try that out.
But I don't think this is a multi-rail/single-rail matter. For once, the power ouput on the 18A rails is 216W. The rails 3 and 4, each with 18A, have a combined power output of ~430W. Iny case, an R280X card needs about 200 - 220 Watts, exclude some board-partner designs that are overclocked/tweaked a little and have 2 x 8-pin connectors. Even in that case you can / should (have to, regarding your power supply) connect both of the 8-pin connectors, as if the power draw of the card can go higher, the connectors for the basis of more power draw have to be there.
If not, i.e. the card has 1 x 6 pin and 1 x 8 pin, you're fine running it off one rail, for the rail has enough power to give. (Theoretical maximum power draw of the card inside the specifications with 1x6 and 1x8 is 225W, rail gives 216W - let's not take 9W into account, as these are easily flowing)
If the card has 2x8 pin, I deduct you've already connected them, which means the card was already connected to two rails, beaing able to draw more than 400W - NO 280X card without volt-mod can draw that much power!
So we can safely deduct, it's not a single/multi-rail failure here. Also, this can happen to single rail PSUs.
Not meaning that Single rail is always crap, but even with OCP (which the user didn't have, to be fair), rails over 30A can have that kind of failure and impact. Problem is that the OCP in PSUs can't hold and stop currents over 30A - effectively, OCP is useless over 30A and rails over 30A are pretty much a hanging sword over the PC. Also, rails greater than 30A are impractical, because there's litereally nothing on the market and the sprcifications that has a power draw that big that a 30A rail can't still from the connectors (2x8 pin is maximum on one rail) connected to the rail itself.
1) Have you tried overclocking the card or is the power limit set higher than 0% currently? I've had this when overclocking a few R9 280Xs at work -> the cards tried to draw more power than they could due to the limit of their PCB designs, thus leading to a crash.
2) Still, both power supplies are very old.
a) after 5 years, there's absolutely no warranty that the fail-safes will work properly.
b) After such a long time, the electronics often have taken damage. One or more rails might not give out the required power, or under load do have such high spikes and drops in current or voltage, that the fail safe of the power supply kicks in. Even just a failure of the pg-time might lead to such crashes. Therefore: Definitely change your PSU. I mean, you can use that for the next 3 - 5 years for the next parts, your CPUs aren't getting any younger.
So, as for a recommendation, I'll say this one first: People here tend to helplessly oversize power supplies! Oversizing the power supply is often a loss in quality (spending money for a 1000Watt-label, while 600 suffice, for example) -> getting a 1000W power supply for the price of a good 600W is a loss in quality. Often means a loss in efficiency, as PSUs are most efficient at around 40 - 60% (mean) and generally, under 70%
Also, there's no 10-pin connectors in computers. Theres EPS 8 Pin, or ATX 4+4 pin. Anyway, since your board is a dual-xeon, I guess you're going to be needing two 8-pins/4+4 pins. You also said 6 SATA connectors, which narrows it down.
Regarding the power draw of your PC:
Xeon E5430: 90W under full load
times 2 = 180W
Graphics card: Let's be veeery generous here and say 300W
Little extra-stuff: 50W
100W for "headroom"
= 750 Watts would be perfectly fine.
Also in a new PC, you're likely to have one CPU less, the graphics card power draw is generously calculated anyway, good 750W power supplies are enough to power a GTX980 SLI and an i7, overcloked - just for size
When it comes down on what you wanna spend there's a few options.
From lower to higher prices:
- Antec Edge 750W (which doesn't have 2x8 pin EPS or 4+4 - only one. Maybe give us a picture of your motherboard, yeah?)
- Gougar GX800V3 (same model is available with 1050W if you really want...)
- Enermax Revolution87+ 850W ATX 2.4 (definitely my favorite!)
- Antec High Current Pro 850W
- BeQuiet Dark Power Pro P11 850W
The last two are, too, of superb quality, but their price seems a bit out of measure for what they're built of and qhat features they offer - one more reason the Revou87+ from Enermax is my favourite.
Also, a little unrelated, but playing in low details always taxes the CPU a little more, than playing in high details.
Comment has been collapsed.
Thanks for the info, and yes this psu is probably 5+ years old so its not a total surprise that its power output is starting to degrade.
I agree with you, a good quality yet lower wattage psu would be better then some cheap high watt one.
You are wrong about the 10-pin connector... Right next to the 24-pin connector is the 10-pin. There is also an 8-pin connector near the processors. Here is a picture of the 10-pin.
I also have another dual xeon system that requires two 10-pin connectors in addition to the 24-pin.
Comment has been collapsed.
Ugh, proprietary connectors. That means, you'll have to find a proprietary PSU for it, which I can't help you much with...
You could also buy a PSU-tester, like this one to rule out errors or spikes/drops in the PSU for the time being.
How about the graphics card, did you try to overclock it? And in line with other comments, I also would recommend running a MemTest sometime, although I don't believe it's the memory currently.
Comment has been collapsed.
I've never overclocked the GPU. I was happy with its performance and didn't see a need to do it.
That tester is pretty neat, but it doesn't have a 10-pin so I cant test that connection, but I could test my PCI-E connectors.
Any idea how much those are? I don't see a price.
Comment has been collapsed.
Should be safe, but you have 2 similar PSU's, so you will not get any progress. You still have weak rails on both PSU's.
Like i said you need 1 strong 12V rail to power up this GPU. AMD is not power consumption friendly device.
Comment has been collapsed.
It appears my PSU is at fault here, thank you everyone for helping me.... greatly appreciated.
Given my setup, which PSU would you recommend? Also figure in a few external hd's, charging my phone, webcam...
My motherboard is "Skye - 2P Intel Socket LGA771, Seaburg, ESB2" and requires a 24 pin, 10 pin, and an 8 pin power connectors.
I'll also need at least 6 Sata power connectors, PCI-E 6 pin & 8 pin.
Comment has been collapsed.
Get a Seasonic, Corsair or Thermaltake(TT has some really good PSUs that most people don't know) with at least 80+ Bronze spec. Also, have you tried underclocking your GPU ? I had the same issue once, I increased the power limit %20 and underclocked it 150/250 on CCC and it solved my issues. I only had that problem on specific games though, I revert it to default when playing other games. Anyway, try underclocking ^^
Comment has been collapsed.
Unfortunately I can't find a PSU with the connectors I need for this mobo. I need one with a 24-pin, a 10-pin, and an 8-pin.
Finding a PSU with the 10-pin is the problem.
I've never over/under clocked the GPU. When you say you increased the power limit 20%, what do you mean?
Comment has been collapsed.
Open AMD Catalyst Control Center and under Performance Tab, go to AMD Overdrive, Enable GPU Overdrive and slide Power Limit to %20. Also decrease the GPU clock and memory clock 100-150 each ( don't worry it doesn't damage or effect GPU in any physical way, it'll downgrade the performance a little but you can revert it anytime you want ).
Even if it doesn't work, underclocking will delay the shutdown problem a lot. You can find it beneficial until you buy a new PSU. Feel free to ask any questions ^^
Comment has been collapsed.
ok this is weird.... I go into the amd ccc...click on performance tab... click accept on the warning and then the entire tab disappears from ccc..... no more performance tab at all.
I tried to reinstall the driver and same thing happens.
I'm using AMD Catalyst Omega 14.12
Apparently ccc doesnt have access to some of the registry keys
in case anyone else has this problem this is how I fixed it.
- Using regedit go to HKCU\Software\ATI\ACE\Settings
- right click on settings key, select permissions, then click on advanced
- make sure your account has full access and then check "Replace all child object permissions with inheritable permissions from this object"
- click ok and ok again and done.
Comment has been collapsed.
Have you ever used DDU, Display Driver Uninstaller ? You should try it, delete your AMD Drivers from Programs and Features and open DDU. It'll ask your permission to switch to Safe Mode, allow it. Then delete Intel and Nvidia drivers without restarting, delete AMD drivers at last and click "delete and restart" on AMD drivers. Also I suggest installing 15.7 drivers, they're really stable and good.
Comment has been collapsed.
I had the same problem as you when I built a new rig months ago.
I did the same test as you too, but my PC got auto shut-off every time I play Saints Row 4. My BIOS says something about power surge detected (It's an ASUS mobo btw), and I tweaked a lot of settings in the BIOS. I even tried under-clocking my CPU and GPU. It's a brand new tower so there is practically zero dust in it.
I also tried swapping out my new PSU and use back the old PSU from my old rig. It worked fine.
I called the hardware store I bought the parts in, they ask me to send in the PSU for them to troubleshoot. Since it's quite far away and I can't be bothered to wait weeks for the warranty, I gave it a final try again. In the end, I solved the problem by unplugging the 24-pin power connector and plug just 20-pins into it (left out the last 4-pins). Works perfectly till now.
TL;DR: Maybe you can try unplugging all the power connectors, and plug them in again. Also try using 20-pins instead of 24-pins. Your GPU have its own dedicated power source, so the mobo doesn't really need 24-pins plugged in.
Comment has been collapsed.
On top of the 24-pin to 20-pin change, I also did this.
My GPU needed 2x 4-pin power. Originally I used a single cable that has 2x output to power both ports of my GPU, I changed it to use 2 separate cables to power them instead.
Comment has been collapsed.
Thanks for that info, but unfortunately I cant unplug part of the 24-pin connector as its all one unit. I have completely removed the original PSU and put an almost identical one in and still have the same problem. In the end this PSU just doesnt have enough amps on the 12v rails to run my video card at full power.
I have been able to play Shadow of Mordor by lowering my video settings but I really don't want to keep stressing this PSU and risk it blowing up and taking out my pc with it.
Comment has been collapsed.
620W should be more than enough for most setups. Hell my old rig only used 550W.
PSU having a 24-pin separated into 20+4 pins are very common nowadays. If not, if you know how, you can cut out the 4-pins by yourself if yours is one single long plug. But anyway, good luck.
Comment has been collapsed.
Sounds rather weird ... and like the mechanism that forces hardware to auto-shut down before it remains in critical status (too high temp) < which also could be a wrong reading induced by the hardware itself or other components (motherboard).
Even more surprising that it didn't shut down when you were running furmark ... have you tried running a similar stressing
game @ appropriate high Settings?
Comment has been collapsed.
Yes, I ran 3Dmark test and after a minute it shutdown... I was also able to get it to shutdown playing 7 Days to Die if I max out all the video settings.... after I lower the screen res and video quality, I can play it without it turning off.... It sounds like a heat issue with my video card but my temps are fine. I also tested the card in a second pc and even at max settings, it works fine. That leaves the PSU which isnt faulty, it just cant power my card at a max load.
Comment has been collapsed.
Seems so then (PSU related) - if you hadn't mentioned 3D mark and shutting down after 1 minute i'd had assumed
' its a solid software <> driver <> hw related error, but that leaves the functionality of the PSU in question ...
' Could be caused by the atypical setup - Server-PC (2 cpu's ... )
http://www.tomshardware.co.uk/answers/id-2320528/radeon-280x-random-shutdowns.html
https://steamcommunity.com/app/241930/discussions/0/606068060833501392/?insideModal=1
http://linustechtips.com/main/topic/240476-computer-shutting-down-during-shadows-of-mordor-possibly-other-games-and-i-think-its-my-gpus-also-driver-problems/
Comment has been collapsed.
I read thru those links you posted, they also seem to be in the same boat as me with an insufficient psu. In fact the steamcommunity link guy solved his problem by changing out his psu. Given that this pc works totally fine all the time and only shuts down when I really put my gpu to work at max, and the gpu works fine at max in another pc, im pretty confident the problem is my psu.
btw, shadow of mordor utilizes both processors and each core shows work when I alt-tab and check the task manager.
Comment has been collapsed.
This could help you with the proprietary 10 pin connector:
In fairness, it's not just us (Lenovo) that drove the use of the 10 pin connector. Check out the appropriate Intel CRB, and you'll likely see the same connector. There are reasons that connector was used, and most of them were Intel driven. In hindsight, it was probably over-designed, but at the time, that wasn't a risk we were going to take.
That's not really a proprietary connector either. The 10 pin connectors going to the board are all 12V and grounds if I recall correctly (don't have a D10 board handy to verify). The more common 8 pin connectors consist of the same thing. I'd venture to guess you could get an off the shelf PSU to work with the appropriate interposer cable.
A standard 8-pin EPS connector is similar, 4 12V pins and 4 grounds. You can use the 10 pin connector from the Delta PSU to create an adaptor if you are good with electronics. Or you can try to find a premade molex to 10-pin adaptor.
Something like this but with 10 pins instead of 8:
Comment has been collapsed.
I thought about creating an adapter myself and still might. I looked for one online but couldn't find one, but it should just be a matter of running a link from one of the 12V pins and a ground kinda like the PCIE 6pin to 8pin adapters.
Thanks for that link.
Comment has been collapsed.
I know these aren't tech forums, but there are a lot of knowledgeable people here so... here goes...
I recently got a new pc and when I play Shadow of Mordor, the pc turns itself off after 5-10 minutes of playing.
The PC is a Lenovo ThinkStation. It has 2 quad core cpu's (E5430). 8 Memory sticks (14GB), 5 Hard drives (4 in a raid stripe), MSI R9 280 graphics card, creative x-fi sound card, and a dvd burner. The power supply is a Delta 1000W. It has 5 12V rails, the first two are rated at 16 amps, and the other 3 are rated at 18 amps. The only USB devices I have connected are the keyboard, mouse, and xbox 360 controller.
The first time this happened I thought something might be getting hot, so I installed realtemp and had it log my cpu & gpu temps, fan speeds, and everything seems to be within range. At the time of shutdown, The shadow of mordor game only used 20-30% of the cpu and 100% of the gpu.
My cpu never went above 50c and my gpu topped out at 70c but mostly stayed in the 60-65c range.
I ran prime95 to stress the cpu and even when it hit 70c, it never turned off. I then stressed the gpu with furmark and again... no problems.
After that I ran both prime95 and furmark at the same time... and again... no problems.
Then I thought the power supply might not be enough with everything in this so I swapped it for another Delta 1000W but this one is rated at 18 amps across all 5 12V rails. Yet the PC still turns itself off when playing shadow of mordor.
My graphics card used to be in my old pc, which had a quad core amd, 4gb of ram, 4 hd's, 2 dvd burners, sound card, a 620W psu and it never had a problem. So I don't think the card itself is bad.
I'm kinda at a lost now as to what to try next... any suggestions or help is appreciated!
Comment has been collapsed.