Computer no longer dependable

So about a week ago my two year old ASUS G10 started acting weird. Crashes, lockups, and finally a BSOD followed by a hard drive check disk on D:.
C: is a 100 GB that only has Win10 and programs that only install to the windows drive while D: has all my different programs as well as all my trainz programs and TANE Local databases.

I was still able to play TANE up until four days ago but then it started to lockup the computer. No response to anything.

While my first thought was that the main 2TB hard drive D: was failing I'm see some odd behavior while attempting to copy all the data and etc to a new 3TB drive. Most of the time the data is transferring at above 30MB/s it occasionally drops to 0 for varying lengths of time.

It has been very hot here recently, is it possible that I've fried something (memory, motherboard, CPU)? Would this generate the hard drive error?

If I am successful in transferring everything to the new drive (currently E: ) can I disconnect the current D: drive and rename the E: drive to D: will all the shortcuts, program pointers and database pointers work?

Re-apply heat sink compound with Arctic Silver Thermal Paste, on both the CPU and video card coolers

You should also think in terms of a UPS such as a cyberpower 1500 VA 900 watts this is not so much to protect from surges but from brownouts. When we dropped in UPSes for a couple of hundred laptops we saw far fewer software problems. I used to be a fan of APC but when I purchased one of their newer models I discovered that it didn't work with "green" power supply PCs and their tech support was aware of it. It certainly didn't work with mine.

Cheerio John

Thanks for the suggestions. I was apparently successful in copying over the first of two 380+ GB TANE Databases, took forever (4 hours or so) but seemed to go OK.

I'll see if the new hard drive solves the problem as I'm not overly confident about disassembling a NVIDIA 970. Since I didn't build this system and documentation is very thin I'm also hesitant about striping out the CPU but if the HD replacement doesn't solve the problem I guess that would be my next step.

Does anyone know of any freeware hardware testing programs to check memory and or CPU? (something simple for this aging old guy)

normhart said:

Thanks for the suggestions. I was apparently successful in copying over the first of two 380+ GB TANE Databases, took forever (4 hours or so) but seemed to go OK.

I'll see if the new hard drive solves the problem as I'm not overly confident about disassembling a NVIDIA 970. Since I didn't build this system and documentation is very thin I'm also hesitant about striping out the CPU but if the HD replacement doesn't solve the problem I guess that would be my next step.

Does anyone know of any freeware hardware testing programs to check memory and or CPU? (something simple for this aging old guy)

Click to expand...

I might start by removing the GPU card then plugging it back in. I'd also reseat any other cards including the memory. Use an antistatic strap connected to an earth when you do so. A plugged PC metal frame is earthed. Technically for safety reasons you shouldn't work on it if its plugged in. I wouldn't touch the CPU. Very gently blow any dust you can see out of the case.

Does it do its nut only when playing TANE, but its OK when browsing the web? If so it could just be that it wasn't designed to handle the heat. GTX70 GPUs etc churn out a fair bit John Citron is the expert on monitoring software for heat. Look for his posts or PM him. That will give you an idea about the temp of the GPU and the CPU.

Cheerio John

Well I had to remove the GPU to get to the SATA ports on the mother board so that is done. I did grab a some canned air and blew out the CPU and GPU fins as best I could.

The problem seemed to be progressive, at first minor things like longer loading times for routes but then it got rough. Last night I couldn't even get the task manager to run. This morning I cleaned things and it seemed a bit more stable but I didn't even attempt TANE. I did run a hard drive check on D: and it came up with something like 57 bad sectors. I've been copying things and browsing the WEB for the last 5 hours without problems so it could just have been the dust accumulation. I also dropped the AC in the house by three degrees. (can't afford that for long) I guess that once everything is copied over to the new hard drive I'll try firing up TANE one more time before taking out the old hard drive. Who knows, it may work fine. I could add a second case fan I guess, I'd have to take off the face plate which is always a pain especially with this unit which has a lot of front face stuff.

Saw my name mentioned....

ASUS offers monitoring software for their systems. I would install that as that will work with your hardware.

At this point, Norm you've done the best thing which I would have suggested anyway if all this started with the increase in summertime heat. I no longer get the canned air and instead use a Rocket Blaster, which is a rocket-shaped squeeze bulb. It does the same, but with a little bit more hand work, though there's nothing to throw away either.

For memory testing, I recommend memtest86, which I've used since about 1998.

http://www.memtest86.com/

At this point it looks like my troubles were caused by heating since I was able to download and start the new Coal Drag session. I'm going to proceed on that assumption although I am going to complete the transition to the new drive.

Thanks for all your help!

I would also look at installing one of the system voltage monitors (you may have one built into the motherboard) - I have had progressive failures due to the PSU becoming unstable/faulty. Indeed, one time I added another hard disc and the additional load was a bit too much...

Cooling is always a potential problem, the stock heatsinks/fans for the CPU are OK, but not necessarily up to running the CPU at full power for extended periods, similarly with some GPU systems. The system I have at the moment was built for me, but includes a case with additional cooling fans, water cooled CPU and a beefed up NVidia GT 770. That seems to run Trainz for extended periods even with today's heat (it's 28C, the CPU is idling at 33C and the GPU at 44C) all the fans are controllable, though default settings seem fine.

Colin

Colin,

You might want to speed up your fans with the outside temperature being close to the CPU temperature. This can cause heat to build up inside the case.

In the past, I too have had wacky power supply do what you describe. Every time I did anything system intensive, I'd get a crash. I initially assumed it was a memory problem and chased that around for a few days. Then after checking out a bunch of things, I narrowed down the problem to the power supply, after suffering from some data corruption. The +12V would drop well below the minimum +/- 5% range it was supposed to be. Since the +12V is split off to feed the +5V, +3.5, and +1.5V, all these rails were not at their proper voltages. When the hard disk was writing, it pulled the +12V down and that caused the memory errors.

After spending quite a bit more for a name brand power supply, Thermaltake at the time, the problem went away.

I followed your suggestion John and ordered a CyberPower CP1500PFCLCD PFC Sinewave UPS 1500VA 900W won't be here for awhile.

I'm also searching for a compatible and more powerful power supply (current one is only 500W)

I'm seeing my VCCIN running consistently at 1.75 which appears high but below alarm range, other voltages are within normal range. CPU temp is running at 45C during data transfer with the ambient temp at around 27C.

I'm working on the last major data base transfer now.

I might go with https://www.newegg.com/Product/Product.aspx?Item=N82E16817139143&ignorebbr=1 but to be honest go to newegg.com, select power supplies, most reviewed, 630-700 watts and you should be fine. It's five cents off this week. The Corsair 650 watt model mentioned is highly rated, which means you can trust the outputs, some of the cheaper ones have been known to be a bit optimistic about how much power they can supply.

In theory a 500 watt power supply say made by Antec or Corsair should be fine with a GTX 970. I run a GTX 980 on a 650 watt power supply but without opening the case I'm unable to recall the brand.

Changing the power supply could be more fun but if you take it steady and use an antistatic strap you should be fine.

Cheerio John

Amazon had this on sale Thermaltake TOUGHPOWER 750W 80 PLUS GOLD Semi Modular Power Supply PS-TPD-0750MPCGUS-1 for under $90 so I grabbed it.

I'm seeing the GPU temp running around 80C with TANE SP2, that sounds awful high, 37C at idle. I guess I'm going to have to learn to disassemble a GTX 970.

I've gotten the old hard drive out and am running trainz from the new drive now. Seems stable but it will take a few days before I trust the system.

:hehe:, the new UPS reports that at idle my pc, monitor, and gateway pull 81-117 watts. With trainz and the new C&O Coal Run... 347 watts peak. :hehe:

normhart said:

:hehe:, the new UPS reports that at idle my pc, monitor, and gateway pull 81-117 watts. With trainz and the new C&O Coal Run... 347 watts peak. :hehe:

Click to expand...

Good news on the UPS. Now when you get those California brownouts, your PC won't suffer from the blips, and you'll actually see better performance due to the steady power. Recently I checked this myself and found that I pull about the same in watts while in Surveyor and route editing. It shows that the PC is doing a bit of work.

Regarding your GPU. The GTX970 will run at most at 83 C, which is not unknown when using a graphics intensive program, and will throttle down. You will notice a bit of stuttering too as this occurs because the card will cool off then speed up again until it gets hot only to slow down again to cool off.

I don't think putting on fresh thermal compound will help, besides, removing the cooler can be a bit risky. I don't know about you, but even as a technician, I was and still am a bit leery of removing the coolers from GPUs. The reason is in addition to the screws and clips, the heat sink (cooler) is usually epoxied into place as well on to the chips and board to ensure a tight hold. In the process of removing the cooler, there's the bigger chance of destroying your GPU or some other chips, or severely damaging the PCB its self by lifting and pulling up the etches in the process. I'm not saying don't do it, but proceed carefully if you want to do this. In the end I would be quite horrified if I ruined a perfectly working video card while trying to make things better. Now there are some video card manufacturers who cater to the hobbyist nerds who like to fiddle and replace parts. If you have one of those cards, well then you can do this with no damage at all. These usually accommodate those people who like to replace air coolers with water blocks, or bigger air coolers. Again not all cards have this capability, and this is usually a premium feature at a premium cost.

The alternative to this is simple. For your video card, you can use something like MSI's Afterburner, or even the software which comes with the video card from the manufacturer to control tweak the video card settings. The software is aimed at the over clockers to play with various settings, but the important thing you are looking for is the fan control. If you want to cool down your card, you can speed up the fans and let the air do what it's supposed to do.

You can actually get some pretty good results. With my video card, I use EVGA's PrecisioX for my EVGA NVidia GTX1080Ti card. I find that if I cool the card first, prior to starting up T:ANE or another graphics intensive program, there's better cooling since there's less of a chance for heat build up in the computer case.