Screen tearing not sure of cause

[Aquinus]

my screen would freeze and instantly screen tear and look like it was put together after being through a paper shredder.

Hello. First of all, I think you're using the term tear wrong. Just telling us that the machine is locking up is sufficient. There is a lot of info and it's kind of all over the place so it's a little hard to understand.

I guess the first questions (which should always be asked is,) are you overclocking? I see that you have a 925 @ 2.8Ghz which isn't an overclock but you're running DDR3-1600 which, for the 925 is an overclock since the stock memory speed is 1333Mhz.

I would start by dropping your memory speed to 1333Mhz and testing it again to see if it makes a difference. I've seen quite a few Phenom II owners who have run into memory instability and upping the CPU/NB voltage or dropping to 1333Mhz usually smooths it out. My rigs (both my SB-E and P2 rigs,) have behave like this when when the memory clocks weren't completely stable.

 

[Eon the Wolf]

Everything is factory default in terms of OC; No OC at all.

I think there was actually a memory issue before, and I think my sticks ARE set to 1333 instead of their factory default of 1600....

As for tear, what I've been describing it as is this;

Imagine a paper going through a paper shredder...now imagine the layers are out of order or misaligned. That's what my screens looked like....Imagine...older arcade machines freezing, or the well known 'death screens' and how they tear...now imagine its the ENTIRE screen with NO unaffected spots.

As for voltage, I don't know what it's set to, but I think my bros tech friend or my tech friend may have adjusted it awhile ago....I can try n check for you if you can help me run through it via skype on another pc or something...

 

[Bo$$]

This may seem like a crazy suggestion, have you tried reformatting your windows all those crc errors may have caused some sort of corruption to some core files. Could try a checkdisk also. I doubt it'll fix it but it's worth a try!

 

[TC-man]

Okay, I'll try that if need be, but I'm sure it's not the cause of my problems. I doubt it would cause screen tear/freezing, especially since I've been using dual monitors for about a year prior to this without issue; 6 months of those 12 were with a CRT, and the other 6 with this LCD tv as a secondary LCD monitor.

So on the topic of other tests....do you or anyone else have any other suggestions/tests I can run? Or know how to run memtestCL on a larger scale? I read I should do a multi thousand iteration test of memtestCL to be accurate/reliable, but I couldn't see an option to set how many iterations it'd do, or how much memory to test like was suggested on the site

Hi,

You can set memtestCL to test the amount of memory and how many test iterations via Command Prompt (cmd). Extract MemtestCL in C drive for easy access the program via cmd.

For example: MemtestCL 512 1000
That's a test of 512 MB video memory in 1000 test iterations.

Or make a shortcut of the memtestCL.exe and put e.g. 512 1000 in the "target" and click "apply", then double click on the shortcut to do the 512 MB in 1000 test interations.


As for the freezing, did you check the temperature of the CPU and your videocard? Overheating could cause freezing. On the other hand, your power supply unit may be failing; power deficiency could cause freezing. How old is your PSU?

 

[Eon the Wolf]

PSU is bout 1-1.5...maybe 2 years old. I listed heating earlier in thread. Vents/case were/was cold at spots it should've been warm at at both times of tearing, and if it were a heating failure I think Furmark would've stressed heating to the point it would've crashed, especially i let it run for 1.5 hours.
Thank you for telling me how to set up memtestcl, I'll try that later...out of curiosity, since I'm running a radeon HD 5770 GPU...And I know one shouldn't/can't use ALL graphical memory to test...what amount should I set, and how many iterations do you folks suggest?

A friend (matt) suggested a possible OS corruption, but this is the same friend who thought an electric powered air COMPRESSOR was as good/safe to use as a can of air when it came to cleaning out case dust...Also, matt's not nearly as knowledgeable as another friend (james) who regularly fixes/builds/updates computers n such. James...and erik for that matter...anyway the two of them both say it's likely not a OS problem for THIS issue. My ever slowing up PC is likely OS being clogged, but they both said OS should not be able to cause the tear/freezing I described. (I plan to reformat OS soon, but need to get/do a few things before I do the reformat)

It ran a checkdisk after the first (or second?) crash, don't think it caught/reported anything. It also ran a system start up restore after the 2nd tear/crash, and since then I've not run games to see if the start up restore fixed it cause I'm worried/pretty sure it wouldn't have been an issue fixed by a start up restore/repair. I also want to be as sure as possible if it's fixed/been found before I start running games or testing again...Don't wanna run the risk of corruption or loss of data before I've had a chance to fully move/back up all my files :X Can't fully back up all my stuff yet till I can get a HDD to move my 320 and 160 files into and then disconnecting my 500s lol. Trying to work with you guys to see if there's a way to find/test for the exact culprit(s) for this issue before I start tinkering with OS installations/reformats or hardware replacement or playing games and 'seeing if it happens again'....

Also crc errors? You referring to the errors I got from my HDD health check? It just told me there was interface communication errors during the health test, I never actually got error popups during normal use....

 

[Eon the Wolf]

http://puu.sh/2aBEx

Welp me and friend found problem o.o All dem errors. We're gonna take it out and try to get a replacement or some such. Newegg tells him it's more than 30 days so replacement RMA isn't available, yet page says 2 year warranty :/ Will let you guys know if/when we get a replacement so I can play games n see if issue is gone. Would it be safe to assume then that GPU is likely the culprit for my issue at present?

Can I run memtestCL on onboard GPU? My friend suggests taking out GPU to test onboard GPU just in case, and if its fine, then he'll lend me an 8800 GTX or 2 of them until we can get a replacement of some kind...unless he lets me have those 2 lol...dunno if my current mobo can run sli though....

EDIT - Sorry for doublepost, just figured it'd be better than editing a post n HOPING you guys saw it.

 

[Aquinus]

Imagine a paper going through a paper shredder...now imagine the layers are out of order or misaligned. That's what my screens looked like....Imagine...older arcade machines freezing, or the well known 'death screens' and how they tear...now imagine its the ENTIRE screen with NO unaffected spots.

That sounds more like artifacts rather then tearing. Tearing is what happens when you turn v-sync off and the second frame is writing over the first in the frame buffer so you get an image between two frames. What you're describing sounds like it could be temperature related. You said the fans aren't as warm as they should be? How hot is your GPU running?

 

[Aquinus]

Welp me and friend found problem o.o All dem errors. We're gonna take it out and try to get a replacement or some such. Newegg tells him it's more than 30 days so replacement RMA isn't available, yet page says 2 year warranty :/

Newegg warranty is 30 days.

Manufacturer warranty is 2 years.

Therefore you need to contact the manufacturer not NewEgg.

Would it be safe to assume then that GPU is likely the culprit for my issue at present?

Not to say it isn't the GPU but you need to gather more info before I would be willing to say that.

EDIT - Sorry for doublepost, just figured it'd be better than editing a post n HOPING you guys saw it.

You should also figure that we're subscribed to the thread. If you know you're double posting you probably shouldn't. Just as a friendly warning, the rules are pretty specific. I don't recall a "loopholes" section for double posting.

 

[Eon the Wolf]

Newegg warranty is 30 days.

Manufacturer warranty is 2 years.

Therefore you need to contact the manufacturer not NewEgg.

Ah, Okay. Yeah n 5770 is on it's way to manufacturer already by now :X. In a week or 2 I should be getting a new card and/or newer card back

Not to say it isn't the GPU but you need to gather more info before I would be willing to say that.

Fair enough. Let me explain this; I took 5770, put in friends 8800 gtx. Tested 2 or 3 LoL games. None of the artifacting/freezing/crashing before. I took it as a good sign. I've done like 10-15 ARAM games n 6-8 ranked, usually 2 ranked n 3-5 aram a day. The last 2-3 days I've had a new problem. It's not a freezing issue, but the last few days my screen has had flickering black lines and/or 1-2 second freezes. However, this started a few days after I tested GPU....and found the memory issues IT has...so I'm playing it safe till I get my new GPU back n I'll test that just to make sure and if issues still come up....then I can rule out GPU for sure n get back to testing other parts.

You should also figure that we're subscribed to the thread. If you know you're double posting you probably shouldn't. Just as a friendly warning, the rules are pretty specific. I don't recall a "loopholes" section for double posting.

I didn't know that subscribing notified of edited posts. Apologies :X

Also, sorry for super late response.

http://imgur.com/a/Fts7p

This is an imgur album of both GPU test results for memtestCL.
I did have another question...Since the brand new GPU will allow me to once and for all rule out GPU....IF there's still issues....

*IS* there a way to test Mobos and/or CPUs fully...or is changing them to see if problems persist the only way to see if mobos/cpus are going bad? And how bout PSUs? I ask because I've got a few spare MOBOs n PSUs I forgot to contact manufacturer for, and while some of em may still be under warranty, I want to know if I can check somehow for SURE that they're bad/going bad before I try n get some em replaced and/or use or throw the spares out...

 

[Aquinus]

Fair enough, lets see what the replacement GPU brings.

 

[Eon the Wolf]

Okay so here's the update;
I've been playing LoL on my friends backup 8800 GTX the past week or two...eventually sticking to ARAM mode and the ARAM map. Originally I could play without issue, but I started noticing artifacting and moments where it'll flash black screens or even freeze a few times for a few seconds....This only happened on the main map (go figure both times in ranked :/).
However, I noticed it NEVER happened, either on full graphics or not, on the ARAM map. To be safe, after my last ranked match, I changed to windowed graphics with everything on medium (Then everything on low eventually) in a 1360 x 768 resolution. No freezes, graphical spazzs or anything....That's how I've been playing the past so many days.
Replacement GPU is here and installed.
Played a few ARAMs on borderless fullscreen with max graphics, no issues SO FAR...But to be safe, I tested it on MemtestCL. This was supposed to be a 'replacement'/aka 'new' GPU.

http://imgur.com/a/Fts7p#0

This is the gallery with all 3 results. The 33k failure is my first gpu, the 100k failure is friends backup gpu....the 27k failure is my 'new' gpu.

Does this mean there may be an issue with the mobo or the wire(s) I'm using? Does this mean nothing? Did they replace my failing gpu with another failing gpu?

Like I said I've not noticed the prior freezing issue YET...But I've not done ranked LoL, I've yet to play anything OTHER than LoL, or do anything really that could've matched to the amount of graphical stress it took the other card to show problems...and this memtest of my 'new' gpu have me worried I may not have fixed problem, or I didnt really get a working replacement....

EDIT - Can ANYONE please help? I have a few other questions, but I'd like to try n clarify/double check some things with new GPU before I ask new questions or do new tests if needed. I was told editing updates subscribed people, so I'll edit for now, and if no one helps by 30th, I'll make another post. Kinda need some info/help here to clarify/check some things...

 

[Eon the Wolf]

Bumping because it's been over two weeks and I tried an edit and no one responded.

Update: I've tried playing more games besides LoL (Diablo 3, SC2 WoL/HOTS, RIFT) and so far no issues....

But I was still wondering why this 'new' or 'replacement' GPU gave such a presumably high error result, and why it could give such a high error return but not show any issues....

I also had some questions about a few other hardware parts not directly related to my current PC I was wondering if I could get answers to...Well one relates to my PC. The rest relate to my spare parts...

 

[Aquinus]

So it only occurs in LoL but no other application?

Sounds like a LoL or driver problem to me. Beyond that I'm not sure. If everything is at stock it's possible that your NB is struggling with PCI-E but I doubt that is the case.

Does your computer ever randomly shutdown or restart on you? If it does, it could be the PSU.

 

[Eon the Wolf]

Nonono, I've had no issues with the new replacement 5770 that Sapphire sent me.

I had issues with LoL on the 8800 gts temp card my friend gave me, but after I got sapphire's replacement 5770, the freezing/artifacting issue I originally had has not returned and the issue(s) that the 8800 seemed to have have also not returned.

HOWEVER....The GPU memtest, which is that imgur gallery I linked there, still had 28k errors if i recall. The 8800 had 100k, my original 5770 had 33k, the 'replacement' had 28k...I've not noticed graphical issues yet, but i thought it strange that a new/'working' gpu would have such a high error count....yet no issues have shown so far...was wondering if anyone knew why that was

Also had a few questions about parts;

1. Is there a way to test PSUs/MOBOs other than just using them and seeing? I have some spare Mobos/PSUs Id like to test n see if they're still working and/or if they're failing/bad...but I'd like to know if there's another way to test them beyond "hook it up n find out" lol

2. I have this occasional problem where my second monitor will flicker black...it used to flicker black lines on the bottom part of the screen as well on the old 5770, but now it just flickers every once n a while, sometimes for a few seconds it'll go black....Why? It's only on the second monitor that it happens.
For clarification my second monitor is actually a 'broksonic' TV with pc monitor capability and it's using display port...could that be the cause? Because it says it can handle 1440 x 900, yet sometimes it'll stop displaying if turned off and/or after restarts and i have to reset it to 1440 x 900 n it may sometimes say "this ratio is unsupported by this monitor"...Why would it say it can't support a ratio that's supposed to be it's native, and why am I getting this flickering issue? (And I can't recall offhand if the flickering only started after I switched to display port cords, but I think it could be a possibility...can't recall offhand if the tv had the flickering issue when it ran a VGA cord with VGA to DVI adapter on it)

In regards to the 2nd problem btw...I heard tell the 5700 series has/had issues with running multiple monitors...is there any chance it's not the tv monitor's fault at all, but just my 5770 being bad for dual monitors? I'm only asking cause I'm not so sure it's the fault of the GPU...I've used this tv for about 9+ months now, the flickerings only been around the past 6 or so, and before that I used a CRT monitor n it NEVER flickered like this tv/monitor has...