PDA

View Full Version : Severe Tearing, Graphical Artifacts (at wits end!)



Dragoncub
12-12-2012, 01:26 PM
I'm using a custom-built rig that I had assembled for me and I am having some VERY frustrating visual problems.

At first things were running smoothly barring ONE issue: graphics driver crash while using the in-game merchants in Borderlands 2.
Updating to the latest (beta) driver fixed that, but introduced three new problems:
--> horizontal 'streaks' flickering on the screen, varying intensity. Most visible on bold dark colours, less visible on light or greyscale.
--> severe tearing, regardless of manually set refresh rates, CCC options, game options - apparent at ALL TIMES, including the desktop, even when booting from cold (ie has been off for over 12 hours)
--> artifacts appearing in programmes - mysterious geometrical 'shapes', triangles, jagged polygons, flash periodically (every half hour or so) across the screen. In Skyrim, these run constantly over far objects (most noticably sky, distant hills), where they may take upon other colours also (such as red) or 'lift' parts of the render and flicker it about in place. {EDIT NOTE: this is only when the game is actively running the in-game world - when paused and thus showing the interactive menus, the problem is temporarily alleviated).

Things I have tried:
--> reinstalling the driver from the disc that came with the card. No change.
--> using driver cleaner and sweeper to completely remove all video drivers (including PhysX), and updating to latest (official) driver (I suspected it was an issue with beta). No change.
--> Checking the refresh rate set in Control Panel was set to 60hz (monitor's natural). Was already set to such. No change.
--> Using Catalyst Control Centre (CCC), changing 3D application settings to use application settings for all. Catalyst AI was on, turned off. No change.
--> using CCC, triple buffering tried both on and off. No change.
--> using CCC, refresh rate tried both 'always off' and 'always on'. No change.
--> video card physical check. Was firmly seated (and replaced again as such), properly wired, cooling system running fully (not detached or rattling). No change.
--> New DVI-DVI cable for the monitor (previously analog with converter). Successfully solved streaking. Tearing and artifacts however remain.

Each driver install came with reboots and uninstallation of previous drivers as warranted (including the one that was also chased down additionally with sweeper).

I'm using Windows 7 x64 with an AMD Radeon HD 7000 series graphics card.

At this point I am thinking it is either of two things: (1) graphics card is bad, probably the memory, or (2) bad RAM, which I have yet to test (it is a time consuming process and I cannot get my computer to recognise the USB version of memtest86 and have no discs to write the .iso version to).

Specific specs as determined by dxdiag:

------------------
System Information
------------------
Time of this report: 12/12/2012, 13:06:26
Machine name: USER-PC
Operating System: Windows 7 Home Premium 64-bit (6.1, Build 7601) Service Pack 1 (7601.win7sp1_gdr.120830-0333)
Language: English (Regional Setting: English)
System Manufacturer: System manufacturer
System Model: System Product Name
BIOS: BIOS Date: 08/03/12 12:01:30 Ver: 08.12
Processor: Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz (8 CPUs), ~3.4GHz
Memory: 16384MB RAM
Available OS Memory: 16318MB RAM
Page File: 2870MB used, 29762MB available
Windows Dir: C:\Windows
DirectX Version: DirectX 11
DX Setup Parameters: Not found
User DPI Setting: Using System DPI
System DPI Setting: 96 DPI (100 percent)
DWM DPI Scaling: Disabled

---------------
Display Devices
---------------
Card name: AMD Radeon HD 7800 Series
Manufacturer: Advanced Micro Devices, Inc.
Chip type: AMD Radeon Graphics Processor (0x6819)
DAC type: Internal DAC(400MHz)
Device Key: Enum\PCI\VEN_1002&DEV_6819&SUBSYS_23201787&REV_00
Display Memory: 1727 MB
Dedicated Memory: 2017 MB
Shared Memory: 3806 MB
Current Mode: 1920 x 1080 (32 bit) (60Hz)
Monitor Name: Generic PnP Monitor
Monitor Model: Philips 236VL
Monitor Id: PHLC082
Native Mode: 1920 x 1080(p) (60.000Hz)
Output Type: DVI
Driver Name: aticfx64.dll,aticfx64.dll,aticfx64.dll,aticfx32,at icfx32,aticfx32,atiumd64.dll,atidxx64.dll,atidxx64 .dll,atiumdag,atidxx32,atidxx32,atiumdva,atiumd6a. cap,atitmm64.dll
Driver File Version: 8.17.0010.1151 (English)
Driver Version: 9.2.0.0
DDI Version: 11
Driver Model: WDDM 1.1
Driver Attributes: Final Retail
Driver Date/Size: 9/28/2012 14:41:40, 1120768 bytes
WHQL Logo'd: n/a
WHQL Date Stamp: n/a
Device Identifier: {D7B71EE2-2B59-11CF-4870-2A03BEC2C535}
Vendor ID: 0x1002
Device ID: 0x6819
SubSys ID: 0x23201787
Revision ID: 0x0000
Driver Strong Name: oem18.inf:ATI.Mfg.NTamd64.6.1:ati2mtag_R575B:9.2.0 .0:pci\ven_1002&dev_6819
Rank Of Driver: 00E62001
Video Accel: ModeMPEG2_A ModeMPEG2_C ModeVC1_C ModeWMV9_C
Deinterlace Caps: {6E8329FF-B642-418B-BCF0-BCB6591E255F}: Format(In/Out)=(YUY2,YUY2) Frames(Prev/Fwd/Back)=(0,0,1) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_PixelAdaptive
{335AA36E-7884-43A4-9C91-7F87FAF3E37E}: Format(In/Out)=(YUY2,YUY2) Frames(Prev/Fwd/Back)=(0,0,0) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_BOBVerticalStretch
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(YUY2,YUY2) Frames(Prev/Fwd/Back)=(0,0,0) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY
{6E8329FF-B642-418B-BCF0-BCB6591E255F}: Format(In/Out)=(UYVY,UYVY) Frames(Prev/Fwd/Back)=(0,0,1) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_PixelAdaptive
{335AA36E-7884-43A4-9C91-7F87FAF3E37E}: Format(In/Out)=(UYVY,UYVY) Frames(Prev/Fwd/Back)=(0,0,0) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_BOBVerticalStretch
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(UYVY,UYVY) Frames(Prev/Fwd/Back)=(0,0,0) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(YV12,0x32315659) Frames(Prev/Fwd/Back)=(0,0,0) Caps=
{3C5323C1-6FB7-44F5-9081-056BF2EE449D}: Format(In/Out)=(NV12,0x3231564e) Frames(Prev/Fwd/Back)=(0,0,2) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_PixelAdaptive
{552C0DAD-CCBC-420B-83C8-74943CF9F1A6}: Format(In/Out)=(NV12,0x3231564e) Frames(Prev/Fwd/Back)=(0,0,2) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_PixelAdaptive
{6E8329FF-B642-418B-BCF0-BCB6591E255F}: Format(In/Out)=(NV12,0x3231564e) Frames(Prev/Fwd/Back)=(0,0,1) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_PixelAdaptive
{335AA36E-7884-43A4-9C91-7F87FAF3E37E}: Format(In/Out)=(NV12,0x3231564e) Frames(Prev/Fwd/Back)=(0,0,0) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY DeinterlaceTech_BOBVerticalStretch
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(NV12,0x3231564e) Frames(Prev/Fwd/Back)=(0,0,0) Caps=VideoProcess_YUV2RGB VideoProcess_StretchX VideoProcess_StretchY
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(IMC1,UNKNOWN) Frames(Prev/Fwd/Back)=(0,0,0) Caps=
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(IMC2,UNKNOWN) Frames(Prev/Fwd/Back)=(0,0,0) Caps=
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(IMC3,UNKNOWN) Frames(Prev/Fwd/Back)=(0,0,0) Caps=
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(IMC4,UNKNOWN) Frames(Prev/Fwd/Back)=(0,0,0) Caps=
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(S340,UNKNOWN) Frames(Prev/Fwd/Back)=(0,0,0) Caps=
{5A54A0C9-C7EC-4BD9-8EDE-F3C75DC4393B}: Format(In/Out)=(S342,UNKNOWN) Frames(Prev/Fwd/Back)=(0,0,0) Caps=
D3D9 Overlay: Not Supported
DXVA-HD: Not Supported
DDraw Status: Enabled
D3D Status: Enabled
AGP Status: Enabled

Trev
12-12-2012, 01:47 PM
Who built it.
:)

Dragoncub
12-12-2012, 02:21 PM
Through a TradeMe representative of Onlinesale Computers Ltd. It DOES come with 18 months warranty and is only 2-3 months old, but I am loathe to have it freighted up to Auckland if I can avoid it since that means I'd be without a computer at all for an extended period. Considering I have four sticks of 4GB RAM, I confess I'm hoping it is that (I can get by on 12GB until a replacement is available).

Unfortunately the listing is now expired from my TradeMe so I am unsure of how to find it again to get specs (neither of us posted trade feedback) but I do have contact for the representative who built it and we have conversed already regarding the afore-mentioned convertor.

As a custom-built it has had other issues too - it doesn't have hibernate as an inate option, and when I modified sleep modes to make it available it boots itself back up after a few seconds. This is however a common 'feature' in computers not designed explicitly to work with W7, and something I can tolerate.

mikebartnz
12-12-2012, 02:27 PM
Considering I have four sticks of 4GB RAM, I confess I'm hoping it is that (I can get by on 12GB until a replacement is available).
To rule out memory issues run memtest on it preferably from a bootable CD.

Slankydudl
12-12-2012, 02:29 PM
Possible causes:

1. Overheating
2. Unstable clock
3. Bad drivers
4. Ones you suggested

Solutions:

1. Check for dust and proper airflow, increase fan speed or lower the core voltage
2. Basically the same as the previous, but i recomment getting a gpu stress test and if it cant pass that then the clock is unstabe and you may have to put more voltage in it or reduce the clock.
3. Get the latest drivers from the companys website.
4. RMA but make sure this is the problem otherwise ur without a gpu for a while and you gotta pay money for it.

If you are overclocking then it may be an over heating problem, when you overclock you want to reach the highest possible clock withought getting a high voltage as the higher the voltage the higher the temp. You will kind of be able to tell when ur getting to the clock limits of the card when you have to scale up the voltage alot to be able to increase the clock.

Speedy Gonzales
12-12-2012, 02:35 PM
So what version of the drivers are installed now?? 12.10 or 12.11 beta? If the beta drivers are installed, have you tried 12.10?

The Error Guy
12-12-2012, 02:36 PM
I'd try using a linux live CD to test it, or making a small partition to install windows onto to test drivers etc. I'm guessing it's the GPU though. If you test and find it IS the GPU you'd only have to send the part back (yay!)

With the RAM you could try just using 4 GB at a time, you should quickly see if one stick of RAM is bad .

Also, check for a BIOS update and check the GPU temps. Could be that the cooler isn't seated although I doubt it.

As I said before, my preference would be to start with a clean install of windows to see what happens. No artifacts? Install the drivers off the disk, still no artefacts? beta drivers, STILL no artefacts then it's a software config/bad driver install. If on a clean windows there are artifacts then it's most likely a hardware issue OR a bad hardware config in the BIOS somewhere. If you still haven't got that memtest working you can try windows mem test (just search memory diagnostic in start) or just swap out RAM modules (using of course, a clean windows install) I know it's a bit icky making a partition for winows just to test things but it will give you a fresh, untampered base through which to eliminate problems.

Quick q, tearing, is that occurring even during the boot process? If so that's all hardware in which case ignore all my stuff about making a nice clean windows install to sandbox.

pctek
12-12-2012, 03:12 PM
It DOES come with 18 months warranty .
18 months? What brand is the GPU? Asus and Gigabyte GPUs have a 3 yr warranty.

Anyway, the easiest test is to swap out the card with another and see what happens.
Fixed - card faulty, send the card itself back.

Not fixed, could be drivers, sometimes the latest isn't the answer, try the next oldest one instead.

Dragoncub
12-12-2012, 03:13 PM
To rule out memory issues run memtest on it preferably from a bootable CD.

As I have mentioned I cannot get my computer to regonise it. I went into the advanced BIOS menu at boot to set the stick in question as priority, but it would insist on booting normally. Though obviously memtest would save me SO MUCH TIME I don't have a CD on hand. I don't have any writable discs, so I'd need to make a special trip to get one. First I'm trying to evaluate my options when it comes to faulty hardware.


Possible causes:

1. Overheating
2. Unstable clock
3. Bad drivers
4. Ones you suggested

Solutions:

1. Check for dust and proper airflow, increase fan speed or lower the core voltage
2. Basically the same as the previous, but i recomment getting a gpu stress test and if it cant pass that then the clock is unstabe and you may have to put more voltage in it or reduce the clock.
3. Get the latest drivers from the companys website.
4. RMA but make sure this is the problem otherwise ur without a gpu for a while and you gotta pay money for it.

If you are overclocking then it may be an over heating problem, when you overclock you want to reach the highest possible clock withought getting a high voltage as the higher the voltage the higher the temp. You will kind of be able to tell when ur getting to the clock limits of the card when you have to scale up the voltage alot to be able to increase the clock.

Cooling is working fine. Not a heating problem - occurs immediately from startup at cold boot when the computer has been off for over 12 hours. Checked the fan physically when I pulled the card out - clean and turned smoothly. Right now it quietly hums in place.
GPU is not overclocked but working at stock speeds. If it is working incorrectly at stock speeds, I'd take that as an indication of a genuine fault that requires investigation regardless, as it could mean the part is unreliable in the long run.
Drivers... well, normally I'd have thought it were a driver issue. (Aren't these things always?) However, trying the beta, trying the official release, and trying the on-disc version... same problem in all of them. For this reason I'm turning my mind toward hardware.


I'd try using a linux live CD to test it, or making a small partition to install windows onto to test drivers etc. I'm guessing it's the GPU though. If you test and find it IS the GPU you'd only have to send the part back (yay!)

With the RAM you could try just using 4 GB at a time, you should quickly see if one stick of RAM is bad .

Also, check for a BIOS update and check the GPU temps. Could be that the cooler isn't seated although I doubt it.

As I said before, my preference would be to start with a clean install of windows to see what happens. No artifacts? Install the drivers off the disk, still no artefacts? beta drivers, STILL no artefacts then it's a software config/bad driver install. If on a clean windows there are artifacts then it's most likely a hardware issue OR a bad hardware config in the BIOS somewhere. If you still haven't got that memtest working you can try windows mem test (just search memory diagnostic in start) or just swap out RAM modules (using of course, a clean windows install) I know it's a bit icky making a partition for winows just to test things but it will give you a fresh, untampered base through which to eliminate problems.

Quick q, tearing, is that occurring even during the boot process? If so that's all hardware in which case ignore all my stuff about making a nice clean windows install to sandbox.

Since I have four sticks, I was going to test them in breaks of two. It is unlikely two will have the same fault. I'll also try swapping them in the sockets, but damn if I hope it isn't a socket error as replacing the whole motherboard will be a lot more cumbersome.

Tearing at BIOS? I truly cannot tell, or at least, I didn't actively notice any (not to say it isn't there, though). I don't have animation rapid enough to test it. Currently one of the more effective methods I've found to make it immediately apparent is in playing games, scrolling browser pages, and moving windows and icons around on the desktop. I'll see if I can have another look though.

As for a clean install of windows, I have feared that would be the next part of testing. I am at that ve.e.ery awkward point where I am momentarily without a backup external HDD, so I've delayed that for testing. Though I do have one, it is not as large as my desktop's HDD, so I would only be able to back up documents, music, saved files, app data, and the like (and potentially 'trim the fat' at that). And I'd need to reinstall and bunch of stuff regardless (copying over all the wrecked files back again would solve nothing). I am anticipating access to a 1TB HDD in about a month as a family member is upgrading her backup storage, and offered hers - though only 1TB (to my desktop's 2TB) it would cover me for now. I have to wait to get access to that, though.

Thanks for pointing out Windows diagnostics, I'll finish up this reply and try it now.

Dragoncub
12-12-2012, 03:16 PM
18 months? What brand is the GPU? Asus and Gigabyte GPUs have a 3 yr warranty.

Anyway, the easiest test is to swap out the card with another and see what happens.
Fixed - card faulty, send the card itself back.

Not fixed, could be drivers, sometimes the latest isn't the answer, try the next oldest one instead.

Yeah, people keep suggesting this to me. GPU fault? Swap with another! Monitor stopped working? Test it with a spare! I say, who are these people walking around with an entire spare computer sitting in their wardrobe? As for my friends, the only one nearby chose her laptop 'because it was pink' with an inbuilt chipset. (Despite my attempts to help her find a good one, the only influence I ended up having was that she got one with a CD-DVD drive.) Unfortunately neither of these are options. I could only test it with another GPU by getting another GPU, which negates the whole going-through-warranty thing.

If I was to take this route, it would mean sending the whole tower (potentially the monitor, also) back to the builder so he can run the swap himself.

The warranty is for the whole built from the builder, which is probably based on the minimum warrantly of one of its parts (not necessarily all of them). The GPU brand is ASUS.

wainuitech
12-12-2012, 03:20 PM
Check something please --- Click start, type in reli - click view reliability History - when it loads, look for any errors at the time it was playing it, it may actually tell you what was going on/wrong, or at least point to the fault.

Dragoncub
12-12-2012, 03:49 PM
As an update, I cannot detect any tearing at the BIOS, but it is difficult to tell (best I could do was switch to EZ mode and mouse-over highlight buttons rapidly to check the visuals). That would indicate a driver issue. (Thing is, I would have sworn it was driver-related anyway had I not tried three different drivers of completely different ages and gotten the same problems from all, even with driver sweeper).

Windows Memory Diagnostic (as suggested by The Error Guy) had nothing to report.

In the process of running the reliability monitor now.


So what version of the drivers are installed now?? 12.10 or 12.11 beta? If the beta drivers are installed, have you tried 12.10?

Missed this one, sorry. Answer: yes. I have tried the beta, I have tried the latest official release, and I have tried the on-disc version (which doesn't have a version number printed on it so I cannot provide one).

1101
12-12-2012, 03:53 PM
1st thing, get in touch with the seller & let them know ASAP . They may just suggest sending up the vid card ??(fingers crossed)

DO NOT tell them you have opened it up . Dont give them any excuse to try & avoid any warranties

Dragoncub
12-12-2012, 04:22 PM
If it does require a freight, I will delay it for another ten days (at least). I'm about to do a big shift, and it will likely be cheaper and faster to ship from my new location (and less cumbersome for return post, since I'd otherwise be moving during). Despite the attempts have installing different drivers, I haven't completely ruled out software. Keep in mind that these issues first became apparent (that is, were noticed) AFTER INSTALLING the latest beta, so it may actually be fragments that DriverSweeper and DriverCleaner missed or did not have permission to erase. It is possible they were present beforehand and I never saw (though.... I do doubt it, especially in the case of artifacts)... but that would imply the hardware mysteriously failed just sitting peacefully in its case.

wainuitech
12-12-2012, 04:29 PM
What does the reliability monitor report ? Should only take 30 seconds or less to load.

If its hardware or software it should tell you what played up and when.

Dragoncub
12-12-2012, 04:48 PM
Didn't report anything, at least initially. Said it was 'still compiling' on an empty graph despite an extended period.

I just opened it again now though after a restart and finally managed to get something.

Graph sits high and even until the 11th week, in which case suddenly it gets all jagged. The problem? Well, I get a few /!\ symbols marking 'unsuccessful driver installations' (including PhysX), but only for a period of three weeks (weekly view). That would say software, but I have reinstalled drivers several times since then.
WEEK ONE: Unsuccessful driver installation
WEEK TWO: Unsuc app reconfiguration (Google Update Helper); four failed Windows Updates, and MSXML 4.0 failed to install twice. There are some critical errors for programmes that stopped working, that is probably due to a hard reset (if memory serves on occurred around this time).
WEEK THREE: NVIDIA PhysX Unsuc app installation, twice (one 3hr after the other). Then few the next couple weeks I get a bunch of windows critical events in the form of "Video hardware error".

I did actually just upgrade to the beta AGAIN earlier today. AMD installer said there were errors while installing and to look at the log but when I did, it just said successful-successful-successful across the board and the 'Error Messages' had nothing following.

This computer has given me more trouble in the two months I've owned it than my laptop did in three years... this is so damn depressing. I just don't know anymore. I could do a complete purge of the PC, reset to the OS completely, but that would require a backup and then redownloading (on limited connection) and reinstalling (for who knows how long)only to have the problem occur again regardless.

wainuitech
12-12-2012, 05:36 PM
Going by all the faults, one possible cause could be the original install is damaged/faulty ( happens sometimes).

As for downloading everything, the only things you should have to redownload is any software you may have done so since it was purchased if the installers weren't kept, and any windows updates ( if you want them).

When you got the PC built, you should have been given the Original DVD containing the OS and Product key, along with any CD/DVD's that came with any hardware.

By the sounds of the problems since the beginning it wasn't done correctly, or tested in the first place (any components can be faulty right from new).

Sometimes if a OS is corrupted, all the work you do trying to fix it is pointless, as once broken it cant be repaired. The only way to be sure is a fresh install.

Slankydudl
12-12-2012, 06:00 PM
Thats how it usually goes, when all else fails delete everything.

Dragoncub
12-12-2012, 06:31 PM
Going by all the faults, one possible cause could be the original install is damaged/faulty ( happens sometimes).

As for downloading everything, the only things you should have to redownload is any software you may have done so since it was purchased if the installers weren't kept, and any windows updates ( if you want them).

When you got the PC built, you should have been given the Original DVD containing the OS and Product key, along with any CD/DVD's that came with any hardware.

By the sounds of the problems since the beginning it wasn't done correctly, or tested in the first place (any components can be faulty right from new).

Sometimes if a OS is corrupted, all the work you do trying to fix it is pointless, as once broken it cant be repaired. The only way to be sure is a fresh install.

Sorry, I think I may be misunderstanding here, because I'm not quite sure what you're saying.

...the only things you should have to redownload is any software you may have done... ...which I assume includes both the standards Java, Flash, Reader, and the clients Steam and Origin (including their games)? Yeah... that's the huge data I'm talking about. Some of them are on disc but their DLC and patches are not, and many are online-only. (The computer did not ship with a partitioned HDD, but I imagine I'd need to format anyway.)

...When you got the PC built, you should have been given the Original DVD containing the OS and Product key, along with any CD/DVD's that came with any hardware... I did indeed. A good thing too, since it is apparent I'll need them.

...Sometimes if a OS is corrupted, all the work you do trying to fix it is pointless, as once broken it cant be repaired. The only way to be sure is a fresh install.

So basically yes, reinstall the OS? Goddamn it.

wainuitech
12-12-2012, 07:26 PM
Basically whats written is correct.

Sometimes if a OS get damaged, theres no real way to fix it apart from a fresh install. Sometimes when they install things dont always go right. If this is what has happened then all the 'fixing" can be wasted.
Actually had one of those today on a Laptop- I installed what was meant to be the correct drivers, but something went wrong and it turned the new install into a total unstable mess.
Theres a difference though if doing lots of installs,you can install from preconfigured images, so mine only took 6-7 minutes from wiping the partitions, to installing windows latest updates and drivers.

What I'd suggest - save all your data. Wipe out all partitions, reinstall windows and drivers, ONLY load in what you have to so you can simulate /test, try to replicate the same problems that happening now (meaning any program or game if thats all thats playing up) then see if the fault is still there. If it is, unless you're really unlucky then its possible a hardware fault.

OR do a reinstall from fresh, then run Burnin Test (http://www.passmark.com/products/bit.htm) use the 30 day trial, then click on Configuration / Test And Duty Cycles tick the boxes as shown below:
4551

Then click OK, click the green arrow and agree and see if if causes the PC to fail the test. There will be a few windows open with various stress tasks happening all at once. ( the 2 and 3 D tests running at the same time really load up the graphics) :devil

Note: its a stress test, and WILL put the whole PC under a load, it can cause components to fail if they are faulty to start with. While not conclusive it will test the components. The test will stop on its own with either a pass / Fail or the PC will crash/shutdown if its to bad.

Edited: Actually, you could run the test as is now, be interesting to see if it fails or not.

Dragoncub
12-12-2012, 08:52 PM
Thanks for the help wainuitech, I'll try that soon. Would rather NOT try running the test until I've made a backup of more more valued data though (and games/programmes I don't want to have to redownload). Hopefully that works.

Iantech
12-12-2012, 09:15 PM
Going by the symptoms described in the original post, I'm going to suggest faulty video memory. Run a video memory stress test. Google vmt.zip, download, extract and run the exe file.

pctek
13-12-2012, 06:41 AM
Yeah, people keep suggesting this to me. GPU fault? Swap with another! Monitor stopped working? Test it with a spare! I say, who are these people walking around with an entire spare computer sitting in their wardrobe?
.
Us techs.....LOL.


But you could perhaps borrow one?
Or just buy some old GPU off Trademe, it doesn't have to be the latest, nor high end, nor even mid-range.

Even as a tech, my test GPUs weren't flash.

The problem is, with troubleshooting, unless it is a software problem - driver, something you installed, particular game etc - you do need things to test with. So you may have to send it back, after all, it is their problem, not yours.

You could try another O/S, Linux for instance. Problem remains, that eliminates any software issue.

Agent_24
13-12-2012, 10:38 AM
Going by the symptoms described in the original post, I'm going to suggest faulty video memory. Run a video memory stress test. Google vmt.zip, download, extract and run the exe file.

I think so too. Had similar graphical issues with a video card and it turned out to be faulty VRAM.

Chilling_Silence
13-12-2012, 12:46 PM
Yeh my money is also on checking the temperature. It'll be cold in the BIOS, coz it's not stressing it. In a game though ... different story.

Download Speccy, it's free, check what your temperature is. It may be something as simple as a loose GPU heatsink / fan, or needing additional thermal paste. Thats where my money is right now.

Slankydudl
13-12-2012, 12:50 PM
Gpus today tend to get artifacts at around 80c. Personally i use msi afterburner as it allows me to setup a fan profile so the fan spins at a desired speed for a certain temp, controll over clocking and has an ingame monitor that tells temp, gpu usage and vram usage.

GPU's have been getting more and more sensitive to heat, i never remember a voodoo getting artifacts.

stratex5
13-12-2012, 08:08 PM
Are you running crossfireX ?

Dragoncub
13-12-2012, 10:07 PM
--> No, not running crossfireX. I'm only using the one card.

--> People keep suggesting heat issues. IT HAPPENS IMMEDIATELY AFTER BOOT, ON THE DESKTOP, AFTER BEING OFF FOR OVER 12 HOURS. I know you are trying to be helpful but I have said this several times, including in the original post. The temperature also averages around 30-35*C under standard use because cooling systems all up in this.

--> Ran Speccy after playing Mass Effect 3 actively for about 4 hours and isn't saying anything about GPU temp, only CPU (which is currently at 50*C) and motherboard + hard drive (both 30*C). I was STILL GETTING ARTIFACTS during this period.

Basically it is looking at being either the VRM (new GPU required) or an improper initial install (OS re-installation required).

Slankydudl
13-12-2012, 10:26 PM
Is it getting artifacts in windows or ingame?

wainuitech
13-12-2012, 10:49 PM
The errors you have listed in all the posts can be caused by either a faulty card, as you mentioned, Some of the other errors you're getting may or may not be OS related, you really need to find out what the problem actually is.

Any of the stress tests mentioned in the previous posts will test the card, if it fails then problem solved, at least you'll know then.

The ideal test as PCTek mentioned is to drop in another card, but if you don't have one there is one thing you can try, it may or may not have any effect.

Example:
(Got one customers PC here, faulty card, if you start it in safe mode it boots and runs as expected, put that card into another good OS and it promptly crashes the PC or wont allow it to even boot - it just hangs, put in a good card and it runs nice.)

Boot the PC into safe mode ( keep tapping F8 on start up) if its a driver issue generally the problems wont show. Running in safe mode only loads the default inbuilt graphic drivers.

From the temps mentioned id be surprised if its over heating. Esp if it does it the immediately.

Chilling_Silence
13-12-2012, 10:54 PM
--> People keep suggesting heat issues. IT HAPPENS IMMEDIATELY AFTER BOOT, ON THE DESKTOP, AFTER BEING OFF FOR OVER 12 HOURS. I know you are trying to be helpful but I have said this several times, including in the original post. The temperature also averages around 30-35*C under standard use because cooling systems all up in this.

Still not convinced. Fire up a Linux Mint LiveCD, prove me wrong.