And we're back to going rounds with the texture decompression error on my friend's machine. They had abated for several months, but now they're back in force, for no apparent reason. Have been getting them pretty frequently lately after several months of good behavior...
Tried deleting all the HOGG files and repatching 12GB.
Got launcher errors. got past those. Got errors while patching. Got past those. Got errors in load screens. Got past those. Got errors getting to character load screen. Got past those. Got texture decompression errors trying to log into the actual maps...
I've read this thread and it seems that it's different for everyone.
It's a different texture for each individual.
What I find unusual is that, after deleting and replacing the suspect HOGGs, the error remained exactly the same!
The same textures/objects cause the crash.
I'm sorry, but you're not correct. I can attest that deleting HOGGs has resolve the issue for ONE texture only to magically have a different texture become the issue...
The BASE ISSUE is the same, some random texture fails to decompress and completely crashes the game. Force Verify does nothing whatsoever and won't fix anything at all.
Deleting HOGGs and re-downloading 6-12GB generally fixes the issue temporarily until you hit the next corrupted texture some arbitrary time later when you run across it in-game. Sometimes you go a few weeks or a few months, but eventually have to patch again at which point you get "+2.5% chance of game failure." Not a fun passive ability, I assure you. There should be an accolade "Sainted" that you get after 100 texture crashes and complete 6-12GB re-patches.
Have you tried Options > Force Verify and let that complete ?
You can try disabling on demand patching and see if that helps too.
You can try logging into another character, lower the video/graphic effects, and then go back and login to the trouble character to see if that helps.
Read the whole thread? Done pretty much all of that.
Force verify does nothing whatever. Same issue persists with same texture not decompressing correctly. Only thing that seems to affect it is completely deleting and re-downloading the HOGG files for textures. But often a DIFFERENT texture file will start failing...
On demand patching or no on demand patching does nothing. A corrupted texture is a corrupted texture whether you downloade a 100MB of textures or 3GB. The question is WHEN and WHY was the texture corrupted? At time of download? After the fact? Is it actually a corrupted texture or is it an error in the decompression routine? Probably the former as it seems to error on different textures after re-downloading HOGGs...
So... Can you reproduce the issue on another machine.... If it is a coding issue, then it must be reproducible on more then a single machine...
It's a slightly maddening error. It seems to be machine specific, but it's unclear exactly WHAT associated with a specific machine is causing the issue, whcih is why we've been going round and round with it for months.
At my friend's house, their old machine which wasn't designed for STO but runs it (poorly) has NO ISSUES, while the new machine purchased specifically to run STO is the one having issues.
Both are plugged into the same wired Frontier FIOS router. Both run Windows 7 64-bit. One's an eMachines (old one). One's an ASUS (new one).
Again, don't know what the issue is, have gone through any number of troubleshooting steps over the last few months... It was good for a while but is now back to misbehaving...
Warranty expired on the unit (so, decided it was fine to crack the case for further testing & a fix). Started blue screening. One in particular was a big hint. I think it was basically a blue screen saying MEMORY_MGMT or something like that. Whatever it was I looked it up and came across some article relating to memory issues.
So, I decided to re-run some old memory tests and a new one.
Basically, run Passmark's Burn-in-Test. They let you use the regular or pro version as a 30 day trial. Also run Prime95's system stress test.
Whereas prior test didn't come back with anything, this time around Burn in Test errored out all over the place and Prime95's stress test "stopped -- Memory errors" almost immediately after starting on all processes.
So, I completely overhauled the memory for $80 worth of 4x4Gb (16Gb total) PC3-10600 Corsair memory, totally replacing the 8Gb that shipped with the machine. It was still running slow and having some issues (probably due to corrupted automatic updates, etc.), so I ended up nuking it back to spec (reinstall from built-in recovery partition).
The slowness and instability post-memory upgrade went away and there have been no further texture decompression issues.
So, basically, I'm 99.99% sure this error is an indicator of faulty memory modules corrupting downloads. Don't ask me why the verification routines don't catch the corrupted downloads and re-download the corrupted bits and pieces until it finally gets a verifiably 'good' copy. But there it is... Replaced memory. Issue resolved. Period. Fix has been in place for a month or so now with zero recurrence.
So, if you're having these texture decompression errors it PROBABLY means you have faulty memory modules.
It wasn't apparent at first, but it seems to have gotten worse over time to the point it became painfully obvious. Original memory tests didn't catch it (or maybe I misread them?).
So, yeah. Definitely have your RAM tested and replace as necessary. Don't know whether this might also happen if you overclock your memory, etc.? If it causes some kind of instability?
Can't say for sure that other network connectivity issues couldn't result in corruption of the .HOGG files, but I can pretty much say that *in my friend's case* it was 100% the fault of the memory, from all signs.
Last edited by mgmirkin426; 09-07-2012 at 01:03 PM.
These seem to do pretty well in ferreting out memory errors. If any errors come back, you should probably have someone test your memory modules at Best Buy or Frys or something and replace which ever are defective.
Hope this helps folks. My friend is happy his machine isn't puking up rancid textures anymore.
Now he can go back to complaining about the rampant 1-shot kills in the game...
I've followed this thread for a while and have I found an explanation in my case and thought that it wouldn't hurt to share it.
When running memory tests my system passed without errors regardless of the thoroughness of the test to the application used. On a hunch I did some research on the components in my system and found that my main board supplied a maximum of 1.5v to the system RAM and that the RAM in my system required 1.65v.
I replaced the memory with model of the same speed and size which required only 1.5v, I re-patched the texture files after the hardware change to ensure they were un-corrupted and my system has been stable for more than two weeks.