Problems With My Computer
#21
Posted 16 February 2015 - 02:24 PM
Gonna try with the other pc at home and report back
#22
Posted 16 February 2015 - 04:45 PM
#23
Posted 16 February 2015 - 05:07 PM
Max doesn't use Flash, or Java, and you can't just say "well we already know part of the GPU is broke, but couldn't possibly be the GPU, because we can just automatically explode those into two different problems here!".
http://en.wikipedia....Occam%27s_razor
Sure, it could be two completely different problems, but why go with two and you can go with one? So far testing has eliminated every other cause investigated, and while it certainly might not be the card, decreeing that it's definitely not, based on reasoning that is entirely useless in establishing that (corrupting only part of the screen does not mean it's not the GPU), is not a particularly good way of going about conclusions.
Usually one reaches conclusions semi-definitively after testing is done, not before.
NRP, on 16 February 2015 - 04:45 PM, said:
Yeah I looked, and even assuming you could find a fairly small box to stay withing cheap shipping options and keep the weight down, USPS is saying it'd be like $30 to ship, minimum.
Edited by Catamount, 16 February 2015 - 05:11 PM.
#24
Posted 16 February 2015 - 06:17 PM
I've seen similar issues to this a few times and have a few different solutions. One is that several things the OP uses rely on the same foundations, something like specific DLLs or packages (like Java or .Net), that end up getting corrupted and thus you see the visual corruption as the system struggles to interpret+output. Another thing was a GPU's soldering starting to crack, and the solution was to bake the card in the oven (seriously, lots of success stories with this). Another was a dog having used the monitor cable as a chew toy, and while not broken was damaged just enough that output would randomly just go nuts for several minutes. Kicking the machine and jarring the cable would fix it temporarily usually, but obviously replacing the cable was the actual solution. Another time, it was terrible power delivery fluctuations causing severe memory corruption in VRAM.
No philosophy there.
#25
Posted 16 February 2015 - 06:44 PM
Thanks for the consideration NRP, appreciate it.
I'm thinking Im gonna have to pick up an gtx 960, although it sucks since they retail for 300 dollars here, its almost like buying a 970 (and those cost 480 dollars down here!)
Gonna keep searching.
#26
Posted 16 February 2015 - 07:58 PM
Just a thought, have you checked the heat level on the card while operating these programs that cause the video issues?
- Do you hear and see the fan speed change any?
- Are you running the correct program to allow you to ramp the fan speed as heat increases?
(this is found on the Nvidia site, look under the 'system tools' item, they recommend the vid card manufactures specific programs over there own. I use Zotac Firestorm for the 2 560Ti Zotac SLI cards.)
- Next if you have the connectors, try a different connection type to the monitor.
- You stated you did run (at one time) 2 monitors but the signal died for the second. (that is never a good indicator)
~ swapping the monitor connector port on the 560Ti and check for effective signal to the 1 monitor.
-Check for all the correct settings within the Nvidia control panel, showing the connection type and anything specific for it.
(may require a reboot, for the system to see the new connector port? WARNING, hot plugging while powered is NEVER a good idea.)
Are you running single DVI to the monitor? Do you have the option of HDMI connector?
And with the case power off and PSU plug disconnected, carefully remove the vid card. Check the 16x rail for dirt, damage, and clean the card contact base. (isopropyl alcohol to clean)
- In some cases, with poor manufacturing QC, the DVI plug connectors solder points may be 'suspect'. Or if the plug and cable 'stressed/pulled' that connector. (Check very carefully the bent signal connector solder points, if you don't know what your looking for, or have never done PCB solder work, sorry can't help here.)
- Be very careful with compressed air on any fans while cleaning them, you can over rev them and cause 'issues'.
And as some have stated, do a clean install of the Nvidia drivers, latest v347.52.
Final check;
- change the power connectors to the vid card from the PSU, should be 2 x 6pin plugs. And should be on a dedicated rail just for the video. (don't tap them off of other connectors.)
Just some suggestions,
9erRed
Edited by 9erRed, 16 February 2015 - 08:07 PM.
#27
Posted 16 February 2015 - 08:49 PM
Jesus DIED for me, on 16 February 2015 - 05:48 PM, said:
Funny, but Occam's razor actually says to the effect that the unprobable conclusion that the video card is somehow selectively renders incorrectly is outlandish in it's reach. It's a lot more probable, and, in line with Occam's razor, that the simpler conclusion, that of software rendering in the web browser being flawed, is correct. Why bring up Occam's razor if it bites you in the heel? Your own argument and conclusion is fundamentally flawed.
A.) Wut?
B.) Occam's Razor states that the conclusion that should be preferenced is that which makes the fewest assumptions. It has absolutely, positively, nothing to do with selective rendering of video cards. You're relying on an explanation that requires two problems, two assumptions of what's going on. I'm suggesting it may be a single problem.
That makes your solution the less parsimonious. I'm sorry if you have no idea how that works. You've given no reason why a video card, a single, malfunctioning video card couldn't be malfunctioning in such a way as to cause this problem, you've given no reason why we should assign different causes to different aspects of a problem, and so far you've offered nothing but blind assumptions. What's even more hilarious is that you're quick to latch onto Wired's post, when three out of the four causes of "similar issues" he's seen are hardware related, and two of them because of the GPU.
It seems you don't even know what you think.
Quote
If basic scientific reasoning can address the likelihood of a hypothesis then I think it's quite useful. Contrary to your claim, you're no less bound to "philosophy" than anyone else. Just because you don't have formal understanding of the logic you make use of, or why it works, doesn't mean you somehow get by without using logic.
You almost always make use of scientific reasoning, and it works for a reason. It's not progressions from assumption to conclusion that you simply draw out of a hat. Well, I hope it's not, anyways.
Edited by Catamount, 16 February 2015 - 08:49 PM.
#28
Posted 16 February 2015 - 09:00 PM
Karpi, on 16 February 2015 - 06:44 PM, said:
Thanks for the consideration NRP, appreciate it.
I'm thinking Im gonna have to pick up an gtx 960, although it sucks since they retail for 300 dollars here, its almost like buying a 970 (and those cost 480 dollars down here!)
Gonna keep searching.
I actually looked up the USPS charge to send it. It's pretty steep
Anyways, if you're having a software problem, then you do run the risk that that'll just carry over to your next GPU, which means you'd have to fix it and the purchase of a new GPU would be unnecessary. If you're out for an upgrade anyways, then maybe that's not really a big deal (software problems could be fixed after the fact if they persisted). Still, bumping to another computer is probably the fastest way to peg the problem down reasonably definitively.
9erRed has also offered some useful suggestions, imo.
#29
Posted 16 February 2015 - 09:05 PM
#30
Posted 16 February 2015 - 09:20 PM
#31
Posted 17 February 2015 - 01:41 AM
As a check of the card,
- before you pull it for testing on another machine,
- or install a different card to check cables and connectors.
Can you go to the Nvidia site and run one of the 'benchmark' videos to see what it does.
- If it runs just fine, great.
- If it causes similar problems with 'pushing' the card, then we've narrowed down some issues.
Did you change or alter any Dx11 installs?
Are you running the game in 9 or 11 mode?
Are you running the game in the 64bit mode?
Does Dxdiag say everything is ok?
Does the event manager say there are any 'conflicts' or problems?
Just narrowing down items before you haul parts around.
9erRed
#32
Posted 17 February 2015 - 04:36 AM
The only cure is to get a new one.
The longer you wait, the worse the artficating will get, until you reach a point when the computer won't even turn on.
#33
Posted 17 February 2015 - 08:52 AM
I tried mwo on directx 9 and 11, same problem. Im running it on the 64bit client.
Did a fresh install last night, havent played though.
#34
Posted 17 February 2015 - 09:56 AM
#35
Posted 17 February 2015 - 10:25 AM
#36
Posted 17 February 2015 - 10:59 AM
Karpi, on 16 February 2015 - 11:36 AM, said:
IF you still have warranty on the card, get a replacement from your dealer ... otherwise try another GPU card !
Karpi, on 17 February 2015 - 10:25 AM, said:
I am sure that a new installation wont help you ... it never helps by HARDWARE problems like a faulty GPU card.
Edited by Brother MEX, 17 February 2015 - 11:01 AM.
#37
Posted 17 February 2015 - 04:22 PM
Just to be sure, we deactivated the directx on firefox and the glitch went away confirming our suspision. So, I'll be buying a new card it seems.
Visually the card looked fine and didnt smell burned or anything like that. Fine was working fine aswell.
#38
Posted 17 February 2015 - 09:01 PM
The other computer just had a fresh install of windows 8.1 and had everything up to date. We literlly built it yesterday for my mom boyfriend.
I have my windows 7 up to date with everything. I have an original version too.
Edited by Karpi, 17 February 2015 - 09:15 PM.
#39
Posted 17 February 2015 - 09:35 PM
#40
Posted 17 February 2015 - 09:59 PM
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users