I'm re-doing a lot of my meshes with this tool, it's great! Thanks

In future versions, is it possible to update the GUI so that it has a standard minimise/maximise/close buttons, or that opening the program doesn't pause the script? I'd like to open the program from the taskbar / the icon 'dock' but every time I do that, it pauses the script and I often forget to unpause it, since I'm only checking on the progress.

The answer is "not on my list of priorities" unless I get bored or a lot of people request it. At the moment it does a good job of what it is suppose to do, but is a bit clunky to use.

-dan

Just a quick question: It says that the newest version has improved starting and stopping. Is there anything special I need to do to resume the pyffi process halfway through? It's been running on my meshes folder for about 5 hours, and isn't even halfway done. If I want to stop it and start later, can I just close out and reopen it later (with the same pyffi folder and the same destination folder)? Or do I have to just pause the script and leave the process on my taskbar? Thanks!

Are you running alpha 4 or alpha 5 (only available from the link here in the forum)?

If you are running alpha 4 . . .
If you exit the process, and restart with the same source/destonation folder files that WERE NOT COPIED (ie thier new size was not smaller than their original size) WILL BE PROSSED AGAIN (waste of time).

If you are running alpha 5 . . .
If you exit the process, and restart with the same source/destonation folder then it will resume exactly where it left off with no redundant processing (it checks for files listed in filespyffied.txt in the output folder)


-dan

Thanks dan, I'm running Alpha 5, this is a relief to hear!

And, thanks for this terrific tool!

Will the first partial output log be ovewritten or just will new lines just be added to te first partial output log? Or will there be a separate output log for each continuining process?

This log is appended only. The only way files are removed is if you manually edit it or delete the log file (which happens I imagine when people delete their temp directory). Good point, I'll add an explaination into the readme.

-dan

I was having issues with the automation tool saying that the Pyffi log wasn't created. I looked at the code and then the batch file, and found out that the above was the problem. This isn't an issue if you installed Python to "C:\Python25" because there aren't any spaces in the path, but by installing it to Program Files will cause the issue. As is, it's trying to make a logfile named "Program" under C:\


Edit:
I couldn't see any missing quotes in the source. And after compiling it myself, the program deletes the test.bat and continues to run perfectly. I looked at the source again. This time I deleted the quotes and was able to reproduce the missing log file error and an indentical batch file. So it looks like the included exe is not using the latest compiled source.

Hmm I hate to think I could have been that stupid when releasing but it sounds bad . . . i'll check my backups here.

Which version are you using? Alpha 4 or Alpha 5 (only available in this forum at the moment). If you're using alpha4, I would appreciate it if you could try again with alpha5.
http://www.gamesas.com/bgsforums/index.php?showtopic=1007148&st=80&p=14768843&#entry14768843

-dan

turns out I can be that stupid . . . thanks for pointing out the problem.

I'll upload alpha5 to tesnexus

-dan

Alpha 5 released

See top post for full readme.

available at tesnexus

http://www.tesnexus.com/downloads/file.php?id=25423

Just checked the alpha5 and it seems to be working fine here.

would really love to see this work on .bsa files....hmmm

Well I'd originally posted an issue I'm having on the download site but then it occurred to me I should probably post it here instead.

I'm on Vista home premium x32, intel core2duo @ 3.0 ghz, 4 gigs of somethingorother ram. When I launch this thing my core temp skyrockets up from an average of 40-45 to 85+ and my computer shortly after starts shouting at me with obnoxious threats to shut down due to overheating. I've ran it in administrator and compatibility with windows XP, no change. I changed the core number in the .ini file from 2 to 1 and, running it on only one core, my temperature plateaus out at 74 degrees. This is a number I can live with comfortably considering it's within the safe operating range for my CPU, albeit higher than I like. What I'm more interested in is why the load is so incredibly massive on both cores and why this works my CPU to its absolute limits when NOTHING else I run on my computer has ever brought my temperature up past 60. Is anybody else having core temp issues as well, or is it just me? If it's just me, what might be causing pyffi to raise it so high when I can run several simultaneous applications at one time and my CPU doesn't break a sweat?

pyffi is a computationally intensive program so this will max out your processor

Well you are either overclocking your processor without adequate heat sinking on your processor, or your heatsink is not properly installed.

Read any of the bazzillion articles online about how to overclock properly.

-dan

(by the way, I use and E6750 at 3.8Ghz with core temps below 60*C at full load. just need a good heat sink)

I have used PyFFI and this alpha script. It was quite neat to see a single program take full use of my processor!

However, if running your processor at 100% causes overheating, then you have a hardware problem. Possibly incorrectly seated heat sink as above suggested. Possibly something else... blocked, inadequate airflow, maybe? Some of these things might not show up until you stress your system as PyFFi is doing.

But, your processor ought to be able to run at 100% without overheating.

gothemasticator

My CPU isn't overclocked and my heatsink is seated properly. Perhaps I'll look into just getting a better heat sink and thermal grease. I figured it wasn't the problem in and of itself, but at least knowing it's a CPU hog helps. Right now the temperature is under control. If it overheats again I'll turn it off, pull it out, and remount it. If this doesn't solve it I'll replace the dinky little stock heatsink that came with my CPU for something else.

will BSA be supported in 6?

Fun fact: replacing the thermal grease can drop the temperature 10 degrees AND at 100% work load on both cores I'm no longer topping 50 degrees C.

In the mean time I'm running back up at 100% on both cores and it's going very well. Can't wait to see how and if it affects my FPS.

I'm not going to release a version 6 unless there is a need for it.

It isn't hard to make BSA unpack/pyffi/repack script, I would just rather work on my other mods.
The source is included with the download so feel free to do it yourself.

-dan

I just recently had the tool quit in the middle of a session and an error code was generated, 0x03, I think it was. I didn't write it down.
After trying several times to run the tool with different selections, I finally rebooted my computer and then it worked.
I was checking Nexus and saw you had an Alpha 5 (I've been using 4). Any chance on also getting a text sheet with what any error codes mean?
Many thanks for this. Now, I won't install a mod now unless I run the meshes with this tool.

The error codes are all for debugging purposes, you won't get anything usefull out of them.

You can check the source code.

In alpha 4, this error code (0x03) means that the execution program failed to move the temporary working file to the output directory.

How this could happen??? I have no idea . . . I would think that something before would break first (like the pyffi process)

either way switch to alpha 5

-dan

Hi everyone!

I've been thinking about using this to PyFFI all of my .bsa's (vanilla, MMM, All Natural etc) and maybe my Data folder itself (once all the bsa's are done) And I was wondering whether the skip_list provided would be safe to use for all kinds of configurations??

I'd be mighty grateful if anyone could just post what their skip_list line looks like. Mine looks like so:

skip_list	="custom","characters\","creatures\","fx\","clothes\","hair","armor\","weapons\","sword","axe","mace","bow"

Thanks for all help!

Mine looks like this

size_test	= 2.00skip_list	="hair"

Seriously, I'm only skipping hair. Also note that I set size_test greater than 1 as per recommendation / testing by amorilia. Yes I know what the aim of setting it to smaller than 1 is, but in some cases optimized meshes will be slightly larger than originally, and I want to accomodate this. We discussed this in the PyFFie thread about 2 or 3 weeks ago.

At the beginning I was as conservative as you are, but I found out that could not detect any defects when being more aggressive. Also I have been told whatever defects arise after pyffie, they would usually be very minor. So nowadays I'm skipping only hair. Certainly you can pyffie armor (including helmets), weapons and creatures. But maybe you experiment yourself, start conservative and get more aggressive later.

Make sure you are using at least pyffie 2.0.5. I think that amorilia is now at version 2.0.6 you may want to check that. This was also in the PyFFie thread a short while ago. I just did not tested and used it yet as I don't have anything new to pyffie.

Woah!!! That's awesome More pyffication, less stuttering, right??

But seriously, only hair?? I read somewhere that PyFFi'ing creatures and weapons (basically anything animated) would cause serious issues etc. Maybe it's time I just read through the entire PyFFI threads (all 5 of them)

Btw, is using a version that's higher than the version in the thread inviting trouble?? I found a PyFFI 2.1.2 at sourceforge!