[Ffmpeg-devel-irc] ffmpeg.log.20160919

[burek]

[00:19:30 CEST] <debianuser> Hello. I'm trying to encode extremely-low-bitrate audio with libvorbis. According to https://www.ffmpeg.org/ffmpeg-codecs.html libvorbis supports options "minrate" and "maxrate", but I can't find how to actually pass those from `ffmpeg ...` command line. :(
[00:28:23 CEST] <kepstin> debianuser: it should be as simple as '-minrate 16K' on the command line
[00:35:21 CEST] <debianuser> kepstin: I can't find that in ffmpeg.org/documentation.html But https://trac.ffmpeg.org/wiki/Limiting%20the%20output%20bitrate says that it's for video. Can I somehow set it for audio track only?
[00:36:54 CEST] <kepstin> the codec-specific avoptions aren't listed in the overall ffmpeg docs
[00:37:11 CEST] <kepstin> to check what options a particular codec actually uses, you can run e.g. "ffmpeg -h encoder=libvorbis"
[00:37:44 CEST] <kepstin> s actually
[00:38:02 CEST] <kepstin> that'll only show non-generic ones actually, so -minrate and -maxrate won't show up there
[00:39:28 CEST] <kepstin> but yeah, the docs for the libvorbis encoder say what the options are, and you turn them into options for command-line ffmpeg by adding a '-' on the front
[00:39:35 CEST] <kepstin> so i imagine they'll work fine...
[00:40:16 CEST] <kepstin> (it is unusual to use -minrate and -maxrate on anything other than video, since the bitrate variation in most audio codecs tends to be fairly small)
[00:43:36 CEST] <gusto> hi
[00:44:10 CEST] <gusto> btw. I noticed that the vp9 works quite well, good pict on low bitrates
[00:44:49 CEST] <debianuser> kepstin: Ah, got it, thank you! Then, I hope, last question: `ffmpeg -i test.mp3 -minrate 2k -ab 8k -maxrate 16k test.ogg` fails with [libvorbis @ 0xb41dc00] encoder setup failed. What am I doing wrong?
[00:45:10 CEST] <gusto> bitrate variation is high, they use different blocks but in the end it averages to some bandwidth
[00:47:59 CEST] <kepstin> debianuser: hmm, just tried that myself, and it doesn't work for me either. It looks like libvorbis is just rejecting that combination of values. not ffmpeg's fault :/
[00:50:42 CEST] <furq> yeah oggenc won't let you use those values
[00:51:10 CEST] <kepstin> you're probably just using twoo low of a bitrate for the sample rate or channel count of your input
[00:51:49 CEST] <kepstin> you might consider using opus instead of vorbis - opus has a lot of automatic adaption features for low bitrate, it'll handle downsampling/downmixing if needed automatically.
[00:51:54 CEST] <furq> huh
[00:51:59 CEST] <furq> it won't accept anything lower than -b 64
[00:52:17 CEST] <kepstin> furq: even if you give it e.g. mono 11050Hz input?
[00:52:28 CEST] <kepstin> er, 11025 whatever :)
[00:52:36 CEST] <furq> i didn't try
[00:52:42 CEST] <furq> also the lowest is 45 weirdly
[00:55:03 CEST] <furq> huh
[00:55:11 CEST] <furq> i guess it's floor(rate / 1000)
[00:55:24 CEST] <furq> -b 8 works with 8000hz but not 11025
[00:57:33 CEST] <kepstin> debianuser: so yeah, add an "-ar 8k" to your command to downsample to 8kHz, and it might work? Might also have to add an "-ac 1" to downmix to mono.
[00:57:47 CEST] <furq> update: it's not floor(rate / 1000)
[00:57:49 CEST] <kepstin> it'll probably sound horrible tho :/
[00:58:31 CEST] <asturel> is there a way to apply delogo vf at specific positions?
[01:00:10 CEST] <asturel> i see https://ffmpeg.org/pipermail/ffmpeg-devel/2011-August/114050.html  {"file", "read times and coordinates from a file", OFFSET(file), FF_OPT_TYPE_STRING, {.str= NULL}, 0, 0 }, but the man doesnt mention it
[01:00:44 CEST] <kepstin> asturel: it takes a bunch of parameters with coordinates; see the docs: http://ffmpeg.org/ffmpeg-filters.html#delogo
[01:00:55 CEST] <kepstin> oh, you mean, move it around at different times?
[01:01:15 CEST] <asturel> no, just apply at different times
[01:01:20 CEST] <asturel> i want to 'mask' the hardcoded sub
[01:01:54 CEST] <asturel> but seems like that patch didnt merged :/
[01:03:16 CEST] <asturel> any suggestion?:)
[01:03:32 CEST] <kepstin> it doesn't look like there's any way to do that at the moment, I don't see any 'enable' option or any time-varying calculated options on the current filter version
[01:04:00 CEST] <kepstin> you could split the video into chunks and apply different filters to different chunks, but that's a bit of an annoying workaround :)
[01:06:11 CEST] <asturel> its more annoying that the blurried video when its unnecessary :D
[01:06:16 CEST] <asturel> than* :D
[01:18:06 CEST] <debianuser> kepstin, furq: Thank you for your hints! Now I know where to look next...
[01:18:33 CEST] Action: debianuser tries to understand how vorbis_encode_setup_managed() selects bitrate in https://git.xiph.org/?p=vorbis.git;a=blob;f=lib/vorbisenc.c#l947 ...
[01:23:17 CEST] <kepstin> debianuser: that function is mostly just making up numbers to put the nominal bitrate field if you leave it unset. If you pass all three of nominal, min, and max it just sets them as-is.
[01:26:05 CEST] <debianuser> kepstin: It's just oggenc fails with "invalid parameters for bitrate" only when vorbis_encode_setup_managed() fails ( https://git.xiph.org/?p=vorbis-tools.git;a=blob;f=oggenc/encode.c#l323 ). So it must be the function restricting bitrate somehow...
[01:28:12 CEST] <kepstin> https://git.xiph.org/?p=vorbis.git;a=blob;f=lib/vorbisenc.c#l633 is probably the responsible function then
[01:29:08 CEST] <kepstin> debianuser: looks like that function checks the parameters against a table of constraints
[01:53:09 CEST] <debianuser> Yeah. It seems that table restricts minimum possible bitrates to:  8kb/s for 8-9kHz mono;  12kb/s for 9-15kHz mono;  16kb/s for 15-26kHz mono;  30kb/s for 26-40kHz mono;  32kb/s for 40-50kHz mono. I guess my experiments would be rather limited.
[01:55:21 CEST] <furq> i take it you can't use opus
[01:55:22 CEST] <debianuser> kepstin, furq: Thanks again! I wouldn't have found that without your help.
[01:57:11 CEST] <furq> opus is quite happy encoding 44.1k mono at 6kbps
[01:57:40 CEST] <debianuser> furq: Ah, no, I probably can (and it's a good idea, thanks, I should try it). I just wanted to test how much I can push .ogg bitrate...
[01:59:43 CEST] <debianuser> furq: By the way, what video containers support opus? Looks like I can't put it inside .avi or .mp4, and I can't put h264 inside .webm?
[02:00:30 CEST] <furq> mkv
[02:01:09 CEST] <c_14> The answer to every question of "which format supports $ codec"'
[02:01:17 CEST] <furq> mkv and mpegts are your only options for h264+opus afaik
[02:21:37 CEST] <kepstin> debianuser: do note that webm and mkv are the same thing, except that the list of codecs allowed in webm is restricted.
[02:25:05 CEST] <debianuser> I have video-audio desync with low frame rate .mkv (steps to reproduce: `ffmpeg -i test.mp4 -vcodec libx264 -r 1 -acodec libvorbis -t 600 test.mkv`, open file and seek forward). No desync with .mp4 (replace "out.mkv" with "out.mp4" in the above command). Is that a bug in ffmpeg, my player or .mkv format?
[02:26:11 CEST] <debianuser> (i mean, replace "test.mkv" with "out.mp4" in the above command to get .mp4 with no desync)
[03:07:06 CEST] Action: Phantom-X5 waves hello
[03:11:20 CEST] <Phantom-X5> anybody around tonight?
[03:14:11 CEST] <debianuser> Why do you ask? ;)
[03:14:46 CEST] <Phantom-X5> because I am looking for someone who might be willing to help me solve a problem
[03:20:03 CEST] <Phantom-X5> ok, maybe next time
[03:20:38 CEST] Action: debianuser now wonders what was that problem...
[03:25:03 CEST] <vans163> Is there a way to encode really cheap low quality h264? Cheap in CPU? Quality level like 70% jpg
[03:25:10 CEST] <vans163> 70% compression on jppeg*
[03:26:17 CEST] <vans163> *wonders if the raspi costs 35$ and has a 1080p 60fps h264 encoder/decoder, how much would a board cost with 10-100 of these raspi decoders/encoders?
[03:39:34 CEST] <debianuser> vans163: `ffmpeg infile.mp4 -vcodec libx264 -preset ultrafast outfile.mp4` ?
[03:43:48 CEST] <debianuser> if you actually need low quality you can add `-crf 30` or reduce resolution with `-vf scale=-1:240`
[03:56:57 CEST] <debianuser> .mkv is buggy. :( ffmpeg produces incorrectly interleaved .mkv files. ( http://pastebin.com/JWTePGgQ - full output of johnvansickle.com/ffmpeg/ build encoding http://peach.themazzone.com/durian/movies/sintel-1024-surround.mp4 and `ffprobe -show_packets | grep ...` for it) Bug in ffmpeg?
[04:19:16 CEST] <DHE> vans163: hardware encoders are fairly common these days. nvidia cards, intel quicksync, etc are common enough
[04:22:44 CEST] <fling> when trying to play m4a files in mpd -> Sep 16 10:04 : ffmpeg: avcodec_send_packet() failed: Invalid data found when processing input
[04:24:16 CEST] <fling> What is the best hardware encoder btw?
[04:48:20 CEST] <DHE> fling: though question. I only have experience with nvenc. there's several chip versions on various revisions of the GPUs over the years. but 60fps 1080p is available at a minimum across the board, and 2 simultaneous streams are supported (at the risk of bumping into the chip's limits)
[04:57:35 CEST] <fling> DHE: I want to reencode a bunch of raw video without wasting a lot of cpu time.
[07:05:33 CEST] <ferdna> i need help capturing video from firewire device:
[07:05:46 CEST] <ferdna>  ffmpeg -i /dev/fw1 -target ntsc-dv -aspect 4:3 file.dv
[08:46:32 CEST] <ss_> Hi i've rtsp stream from IP camera and I want to save it to two different paths using code not commands. Do I need two diff AVFormatContext objs and need to allocate context using avformat_alloc_output_context2 on both paths?
[09:05:21 CEST] <sjums> Hey, I'm using ffmpeg to encode videos uploaded by users. Upon uploading a file I return a thumbnail of the video immediately. Problem is that ffmpeg still read the entire file, so a 500MB file will be read entirely, just so I can snapshot the first frame of the video. Is there a way to avoid this?
[09:06:42 CEST] <Mavrik> That shouldn't really happen, what's your commandline?
[09:07:43 CEST] <sjums> -i inFile -vframes 1 -y -an -s 1920x1080 -ss 3 outFile
[09:09:37 CEST] <sjums> I can read that placing -ss before the input will speed up the process
[09:11:44 CEST] <Mavrik> -ss 3 won't grab the first frame of the video
[09:11:58 CEST] <sjums> that was to simplyfy
[09:12:02 CEST] <Mavrik> You're demanding ffmpeg to skip part of the file which sometimes makes funny things happen.
[09:12:30 CEST] <sjums> depending on the length of the video I grap the first frame, 3s in, 10s in or 20s in
[09:12:32 CEST] <Mavrik> But yeah, moving -ss to the start helps with skipping data at the expense of accuracy.
[09:12:46 CEST] <sjums> And accuracy isn't that important in this case
[09:22:02 CEST] <sjums> testing locally I went from 5.1s to 1.1s grabbing a screenshot
[09:35:09 CEST] <ss_> Hi i've rtsp stream from IP camera and I want to save it to two different paths using code not commands. Do I need two diff AVFormatContext objs and need to allocate context using avformat_alloc_output_context2 on both paths?
[10:05:00 CEST] <ss_> is there a code example for tee muxer?
[10:34:24 CEST] <zephod> @durandal_170  sorry, that i didn't reply earlier... let me renew my question _how_ would a ffmpeg command look like to transcode dsdfile.dsf to pcmfile.flac ?
[10:57:32 CEST] <durandal_1707> zephod: ffmpeg -i dsdfile.dsf pcmfile.flac
[11:45:14 CEST] <zephod> thanks
[12:14:27 CEST] <Spring> welp, just realized after all the testing I did I didn't test videos that didn't contain an audio stream so ffmpeg outputs an error as the command expects an audio stream. I'm guessing the only way to avoid this is by checking for audio streams first?
[12:21:08 CEST] <Spring> perhaps I could run a command that copied audio to null and if it errors the video doesn't contains audio, or similar
[12:21:29 CEST] <Spring> (apart from ffprobe / ffmpeg -i input regex parsing)
[12:23:01 CEST] <Spring> that could introduce a delay though, may need to just use ffprobe...
[12:41:49 CEST] <xavier666> hello everybody. I'm streaming a file with ffmpeg. The stream is working but the received file is smaller in size. While sending, i'm using the following parameters -vcodec copy -acodec copy
[12:42:16 CEST] <xavier666> so shouldn't the received file be of similar size?
[12:42:51 CEST] <xavier666> the source video = 31.4 MB, received video size = 19.5 MB
[12:43:24 CEST] <xavier666> i'm streaming an mp4 file using rtp protocol
[12:55:04 CEST] <xavier666> also, i'm using the parameter -timelimit 120 on the client, that is, i'm instructing the client to shutdown the ffmpeg listener after 120 seconds but the client does not obey this parameter. Any idea?
[13:05:28 CEST] <ss_> can i write output to different locations using same AVFormatContext ofmt_ctx?
[13:05:39 CEST] <ss_> can i write output to 2 different locations using same AVFormatContext ofmt_ctx?
[13:15:29 CEST] <flux> I can't say I know, but I understand it would be possible to provide your own custom IO providers, that can multiplplex to however many files you want
[13:34:29 CEST] <vash> Hi guys
[13:35:40 CEST] <Guest50402> 5.1 ’ 6 × mono https://trac.ffmpeg.org/wiki/AudioChannelManipulation#a5.16mono
[13:36:44 CEST] <Guest50402> How can I can extract specific channel only (lets say FC or LFE) from 5.1?
[13:37:12 CEST] <Guest50402>  5.1 ’ 6 × mono https://trac.ffmpeg.org/wiki/AudioChannelManipulation#a5.16mono
[13:37:33 CEST] <bencoh> xavier666: check your stream mapping (in/out)
[13:37:53 CEST] <Guest50402> Above ^ extracting all 6 mono wav
[13:42:47 CEST] <xavier666> bencoh: could you tell me how to check my stream mapping?
[13:44:17 CEST] <bencoh> xavier666: ffmpeg prints it when remuxing/transcoding
[13:44:59 CEST] <xavier666> great. So which part of the log do I have to see?
[13:45:12 CEST] <xavier666> and what should I be looking for?
[13:46:40 CEST] <bencoh> everything? Input and following lines, same for Output
[13:46:44 CEST] <bencoh> just .... read it? :)
[13:48:33 CEST] <xavier666> this is my server script -> ffmpeg -re -thread_queue_size 4 -i ~/Videos/source.mp4 -strict -2 -vcodec copy -an -f rtp rtp://10.0.0.36:6005 -acodec copy -vn -sdp_file random_sdp_file -f rtp rtp://10.0.0.36:7005
[13:48:49 CEST] <xavier666> and my client script -> ffmpeg -i random_sdp_file -strict -2 -timelimit 120 random_video_file.mp4
[13:49:14 CEST] <xavier666> are there any errors in the script? If not, then I'll check the log
[14:00:52 CEST] <n4zarh> may I have a question here or should I go for mailing list or something?
[14:03:16 CEST] <xavier666> why not both?
[14:04:42 CEST] <n4zarh> so maybe I'll start here - I try to cross-compile ffmpeg to get android libraries and so far it worked for me; however I can't find a way to add speex without configure crash
[14:05:50 CEST] <n4zarh> everytime I try to do that I get "speex not found using pkg-config" error and I pretty much didn't find any solution to that
[14:13:38 CEST] <n4zarh> i can configure without enabling libspeex without problem or by enabling it with cross-compile, but can't find a way to make both work at the same time
[14:17:25 CEST] <n4zarh> without cross-compile* my mistake
[14:18:09 CEST] <ss_> flux: what i want is to duplicate output using libav
[14:19:55 CEST] <ss_> do i need two pairs of AVOutputFormat and AVFormatContext for this?
[14:29:05 CEST] <flux> ss_, now I recall FFmpeg has tee muxer. it seems it does what you want. I don't have experience with it, though. look how the ffmpeg tool does it.. https://trac.ffmpeg.org/wiki/Creating%20multiple%20outputs
[14:34:24 CEST] <ss_> flux: yes i was just checking this link but unfortunately there is no code example.. looking at tee.c in ffmpeg source
[16:06:21 CEST] <sxcdsds> I tried to join two .mkv files with latest version of ffmpeg static downloaded and extracted on my desktop but it errored displaying "unsafe file name 1stfile2b-joined.mkv" and "operation not permitted on list2b-joined.txt" but then I tried again with ffmpeg downloaded inside the "TheKMPlayer folder" and it did the job. now what am I doing wrong or whats the difference between new version on desktop and kmplayer installed version of ff
[16:06:48 CEST] <sxcdsds> I just wanna understand it so to make it clear for future use
[16:29:28 CEST] <sxcdsds> anyone ? help ? ^^^
[16:30:23 CEST] <ice9> how to play dv4 files or convert them to any other usable format?
[16:34:27 CEST] <sxcdsds> ice9: I'm myself a new user and no one else is replying
[16:35:50 CEST] <sfan5> sxcdsds: the error is coming from the "concat" decoder
[16:36:09 CEST] <sfan5> it's probably just that one of the two builds you had was older and without the "safe filename" functionality
[16:36:50 CEST] <sfan5> generally if you're writing input files for use with -f concat start then with "ffconcat version 1.0" (without the quotes) on the very first line
[16:39:38 CEST] <sxcdsds> sfan5: I used command -> ffmpeg.exe -f concat -i fileslist2b-joined.txt -c copy joined.mkv in both cases. http://www.tiikoni.com/tis/view/?id=a63e45a
[16:41:19 CEST] <sfan5> well just begin your file with "ffconcat version 1.0"
[16:41:28 CEST] <sfan5> then it will work
[16:42:31 CEST] <sxcdsds> means ? I'm a new user, plz explain
[16:42:38 CEST] <sxcdsds> sfan5: ^^^
[16:43:18 CEST] <sfan5> where did your list2b-joined.txt come from?
[16:43:33 CEST] <sxcdsds> I created it myself
[16:43:58 CEST] <sfan5> well then you obviously know how it works
[16:44:01 CEST] <sfan5> just do it like usualy
[16:44:14 CEST] <sfan5> but on the first line have "ffconcat version 1.0"
[16:44:27 CEST] <sxcdsds> I followed some page on web regarding joing files using ffmpeg
[16:44:50 CEST] <sxcdsds> joining*
[16:45:19 CEST] <n4zarh> sigh, I wish my question was answered that quickly :D
[16:45:27 CEST] <sxcdsds> how to have "ffconcat version 1.0" on first line ?
[16:46:01 CEST] <sxcdsds> n4zarh: plz dont joke man, I deeply thank and appreciate your help sfan5
[16:46:04 CEST] <n4zarh> so maybe i'll try other way around: do I have to subscribe to ffmpeg-user list to ask a question or not?
[16:46:34 CEST] <sfan5> n4zarh: it certainly helps to get the answers to your question into your inbox
[16:47:01 CEST] <n4zarh> okay, that's why I'm asking - never used list like that before
[16:49:09 CEST] <sxcdsds> sfan5: have a look at the link I provided. its a pic of cmd of both the versions of ffmpeg - left hand side newer version with safe filename and older kmplayer version which did the job. and how to bring "ffconcat version 1.0" on first line ?
[16:49:21 CEST] <sxcdsds> http://www.tiikoni.com/tis/view/?id=a63e45a
[16:49:35 CEST] <sfan5> open list2b-joined.txt in notepad
[16:49:42 CEST] <sxcdsds> ok
[16:49:47 CEST] <sxcdsds> opened
[16:50:01 CEST] <sfan5> then move the current first line down one line
[16:50:08 CEST] <sxcdsds> ok
[16:50:18 CEST] <sfan5> and type ffconcat version 1.0 in the now empty line
[16:50:19 CEST] <sfan5> save
[16:50:28 CEST] <sxcdsds> ohh ok
[16:51:52 CEST] <sxcdsds> but I remember making a txt of dir containing files to be joined and it gave errors coz the txt had many wrods and lines besides [file 'path\file.ext']
[16:52:45 CEST] <sxcdsds> I had to manually delete all other words and lines leaving only the [file 'path\file.ext] command in txt file
[16:55:22 CEST] <sxcdsds> and is there any way of just disabling it or turning it off temporary for current operation in command format seperatly or alongwith original work command ? sfan5
[16:56:40 CEST] <sfan5> i don't know of any way of doing that
[16:56:49 CEST] <sfan5> just add the ffconcat thing i have been talking about to your .txt file
[16:57:38 CEST] <sxcdsds> ok
[16:58:11 CEST] <sxcdsds> but why is this safe-filename thing implemented ? its just a filename, theres nothing to be unsafe.
[16:58:35 CEST] <sxcdsds> file can be unsafe and dangerous but filename cannot be.
[17:00:22 CEST] <kepstin> sxcdsds: there's some issues where if you e.g. play a playlist over the internet, it shouldn't be able to cause your computer to read local files
[17:00:24 CEST] <kepstin> for example
[17:02:41 CEST] <sxcdsds> but what does it has to do with joing two local files with no internet connection at all kepstin
[17:03:42 CEST] <kepstin> if you know the file is safe, then you just have to tell ffmpeg that you know the file is safe
[17:04:35 CEST] <sxcdsds> how to tell ffmpeg that its safe file ? how to disable safe filename feature on temporary basis or for current operation ? kepstin
[17:05:55 CEST] <kepstin> sxcdsds: use the input option "-safe 0"
[17:06:24 CEST] <sxcdsds> kepstin: btw what I get is some sites or playlists on sites or infact any file on net tries to grab local data, so this safe filename is something to do with avoiding local files sending local info on net ?
[17:06:54 CEST] <kepstin> could be, i dunno the full details.
[17:08:47 CEST] <sxcdsds> okay, 1 more question. I tried joinging 3 files before. 2 files had same bitrate but all three had minor resolution difference. after joining only 1st file played correctly while after that 2 files just forwarded to end while audio was fine. audio had same bitrate
[17:08:58 CEST] <sxcdsds> kepstin: ^^^^
[19:31:35 CEST] <MINIMAN10000> Huh I just learned that batch variables do exist.
[19:31:55 CEST] <MINIMAN10000> I should probably make myself some more batch scripts for ffmpeg lol
[19:32:35 CEST] <MINIMAN10000> Just made one for setting a start and end time.
[19:36:52 CEST] <JEEB> I recommend using something else than batch for anything longer than like three lines.
[19:37:15 CEST] <JEEB> there's powershell scripting and some javascript based thing on windows
[19:37:23 CEST] <furq> s/ for.*$//
[19:37:34 CEST] <JEEB> I don't disagree, but I tried to be nice ;)
[19:40:42 CEST] <MINIMAN10000> javascript I personally avoid
[19:41:05 CEST] <JEEB> yes, I was just listing scripting languages available in windows out of the box
[19:41:06 CEST] <MINIMAN10000> unless it does work out of a new windows install.
[19:41:21 CEST] <MINIMAN10000> however I have no idea whats up with power shell
[19:41:29 CEST] <MINIMAN10000> it just feels like some clunky third class cmd prompt
[19:41:59 CEST] <JEEB> it certainly got more love than cmd.exe ever got
[19:42:06 CEST] <MINIMAN10000> by who?
[19:42:08 CEST] <JEEB> MS
[19:42:23 CEST] <JEEB> cmd.exe was only finally touched probably for the first time since winxp by win10 devs :P
[19:42:28 CEST] <JEEB> and those changes were quite minimal
[19:42:40 CEST] <JEEB> meanwhile powershell kept on rollin'
[19:42:56 CEST] <DHE> cmd isn't really any better than DOS's command.com was, just made into a windows-friendly app
[19:42:58 CEST] <JEEB> although I must say I don't utilize powershell too much other than for accessing registry through the shell
[19:43:04 CEST] <MINIMAN10000> I enjoy working within cmd.exe because it feels familear and snappy but powershell feels like some odd colored large texted "thing" not really sure how to describe it
[19:43:23 CEST] <DHE> or now that bash is a thing in windows... :)
[19:43:30 CEST] <MINIMAN10000> people told me that
[19:43:34 CEST] <JEEB> that's a completely separate environment though
[19:43:36 CEST] <MINIMAN10000> I searched bash in files on my computer
[19:43:38 CEST] <MINIMAN10000> nothig existed
[19:43:42 CEST] <MINIMAN10000> nothing*
[19:43:47 CEST] <DHE> it does need to be distinctly installed. it's still "beta"
[19:43:52 CEST] <MINIMAN10000> ah
[19:43:57 CEST] <furq> it's still in developer preview isn't it
[19:43:59 CEST] <DHE> and it's part of a sort of linux emulation layer. it even provides a minimal /proc filesystem
[19:44:04 CEST] <furq> or you could just install msys
[19:44:05 CEST] <MINIMAN10000> so... then it "might"? become out of the box?
[19:44:13 CEST] <MINIMAN10000> I should have msys
[19:44:15 CEST] <JEEB> anyways, if you're interested in something less braindead than batch, https://msdn.microsoft.com/en-us/library/by8safft(v=vs.84).aspx
[19:44:15 CEST] <MINIMAN10000> how was it idk
[19:44:35 CEST] <JEEB> although I think powershell scripting might be much better now than the WSH stuff
[19:44:53 CEST] <MINIMAN10000> ah right
[19:45:01 CEST] <MINIMAN10000> msys and mingw have some form of relationship with each other
[19:45:07 CEST] <MINIMAN10000> to the point where I have mingw
[19:45:13 CEST] <MINIMAN10000> but have not a single clue how to use msys
[19:45:23 CEST] <JEEB> you usually have msys as a posix-like shell from which you then utilize mingw-w64 toolchains
[19:45:40 CEST] <JEEB> msys(2) is a fork of cygwin, basically
[19:47:34 CEST] <MINIMAN10000> why they don't just have a msys installer idk
[19:47:44 CEST] <MINIMAN10000> Things are never easy eh
[19:47:45 CEST] <JEEB> they do?
[19:47:53 CEST] <JEEB> https://msys2.github.io/
[19:48:15 CEST] <MINIMAN10000> so msys 2 isn't msys-base from mingw installation manager?
[19:48:31 CEST] <JEEB> no
[19:48:33 CEST] <MINIMAN10000> "things are never easy" Ill go with no
[19:48:34 CEST] <MINIMAN10000> right
[19:48:35 CEST] <MINIMAN10000> figured
[19:48:42 CEST] <JEEB> ye olde mingw/msys is still kind of alive
[19:48:47 CEST] <MINIMAN10000> go look up mingw and you get that
[19:48:49 CEST] <MINIMAN10000> er
[19:48:50 CEST] <MINIMAN10000> msys
[19:48:58 CEST] <MINIMAN10000> which is great for me because it's wrong.
[19:49:14 CEST] <JEEB> msys2 is a thing based on a newer version of cygwin and by a different set of people who seem to be more active
[19:50:06 CEST] <JEEB> just like mingw-w64 started as a fork of ye olde mingw to produce binaries for 64bit windows, but has no pretty much overridden ye olde mingw for any sort of usage for both 32bit and 64bit binaries
[19:50:12 CEST] <MINIMAN10000> this problem probably arose because msys in tagged onto mingw which makes it high up in the search results
[19:50:31 CEST] <JEEB> *has now pretty much overridden
[19:50:38 CEST] <JEEB> I fail at typing
[19:50:45 CEST] <MINIMAN10000> lol
[19:53:51 CEST] <MINIMAN10000> Gah went as expected. "$ second class turd? ^V^C"
[19:54:33 CEST] <MINIMAN10000> even command sucks with regaurd to bulk deletion
[19:55:52 CEST] <MINIMAN10000> alright good to know in the future Ctrl-C is cancel ( deletes the whole whatever you were typing )
[19:57:26 CEST] <MINIMAN10000> i wonder how power shell is in regaurds to copy/paste/select all and delete
[19:58:17 CEST] <MINIMAN10000> hmmmm
[19:58:26 CEST] <MINIMAN10000> actually powershell is better in that respect than command.
[19:58:39 CEST] <MINIMAN10000> but it took like 10 seconds for powershell to startup :|
[19:58:52 CEST] <JEEB> not sure you need to actually start it to run any scripts
[20:02:20 CEST] <MINIMAN10000> ...
[20:02:42 CEST] <MINIMAN10000> why is the .ps1 file type which is supposed to be powershell opening it in notepad
[20:03:23 CEST] <MINIMAN10000> LOL
[20:03:29 CEST] <MINIMAN10000> it's a security feature!
[20:03:38 CEST] <MINIMAN10000> it's intended that a script doesn't run!?
[20:04:01 CEST] <MINIMAN10000> I don't know what to say
[20:04:04 CEST] <kepstin> yep, because a ridiculous number of people have been infected by ransomware because they saved an email attachment then double-clicked to open it
[20:04:21 CEST] <MINIMAN10000> so then send a batch file?
[20:04:47 CEST] <MINIMAN10000> if morons are running malware theres no hope they'll just run anything
[20:04:49 CEST] <kepstin> batch files are typically filtered out by corporate email servers; powershell stuff is newer so it might be missed
[20:04:54 CEST] <MINIMAN10000> exes?
[20:05:06 CEST] <JEEB> and if you try to run it with `PowerShell.exe -Command thing.ps1` it will complain it's not signed ;)
[20:05:08 CEST] <CFS-MP3b> Does it make sense that using -r 25 (on a stream that in theory is already 25 fps) doubles CPU usage?
[20:06:09 CEST] <JEEB> thankfully you can just add -ExecutionPolicy Bypass and powershell will happily ignore the "must be signed" policy ;)
[20:06:16 CEST] <MINIMAN10000> Alright well I got my reasoning for not using powershell.
[20:06:38 CEST] <MINIMAN10000> I was up for any scripting that worked out of the box. powershell doesn't so it's out.
[20:07:37 CEST] <MINIMAN10000> Microsoft may have given it love. But they gave it the wrong type of love locked it down and made it useless for me.
[20:08:20 CEST] <JEEB> it's funny how the simplest way to run a powershell script is to actually call powershell from a batch script with the security disabling parameters :)
[20:13:56 CEST] <MINIMAN10000> lol
[00:00:00 CEST] --- Tue Sep 20 2016