|
I just load the AviSynth file into VirtualDub now. It's actually sort of what's supposed to happen — when ffdshow-tryouts was switched to LAV Filters, the raw YV12 filter got modified in a way that MPC-HC can't handle it. DirectShow is cool.
|
#
?
Jul 30, 2014 13:36
|
|
|
|
| # ? May 10, 2024 09:59 |
|
|
I use VDub explicitly for two things. Exporting the audio so I can put it under the commentary track, and getting frame numbers for avisynth.
|
|
#
?
Jul 30, 2014 22:18
|
|
|
Not strictly LP related, but I didnt want to ask in the stream thread. Out of curiosity, what is the "regular" setup with streaming and skype calls? Do the callers just have to watch the stuff mute? I want to try to stream for some friends while chatting at the same time. Using OBS usually.
|
|
#
?
Aug 1, 2014 17:22
|
|
|
FrickenMoron posted:Do the callers just have to watch the stuff mute? Yes.
|
|
#
?
Aug 1, 2014 18:50
|
|
|
Edit: NEVERMIND, FIXED Hello I have a question that I'm pretty sure is about interlacing. The game is State of Decay recorded with fraps. I'm using Premeire Pro and exporting with the Lagarith AVI codec, then using megui to mp4 it. When I export the media with Premiere, I have the option to export as progressive, upper first, or lower first. Choosing upper or lower first results in what I'm pretty sure is interlacing, which messes up any movement with horizontal lines, two movement frames spliced together? But progressive seems to show movement with two frames ghosted/blended together, so there's no clear action if I pause on a frame. The progressive looks a lot better when watching the video, but it bothers me that pausing it results in double-image people or whatever. Is that normal, should I worry about it? I saw the aviscript thing for deinterlacing video in the OP, but I don't really understand if that's what I need to do, or how avisynth works. Some advice would be appreciated, I wanna get the LP going. Thanks! interlaced: http://i.imgur.com/TaWeWVK.jpg progressive: http://i.imgur.com/IbL5Tun.jpg skoolmunkee fucked around with this message at 23:27 on Aug 1, 2014 |
|
#
?
Aug 1, 2014 22:46
|
|
|
skoolmunkee posted:Hello I have a question that I'm pretty sure is about interlacing.
|
|
#
?
Aug 1, 2014 22:50
|
|
|
Niggurath posted:Is there not an option to not use any of those options? I seem to recall Vegas had the option of just turning that off when exporting. Nope.
|
|
#
?
Aug 1, 2014 22:54
|
|
|
skoolmunkee posted:Hello I have a question that I'm pretty sure is about interlacing. Are you changing the framerate of the video at all? Exporting as progressive is probably the right thing to do, since you don't want interlacing, and the frame blending sounds to me like a bad algorithm for changing the frame rate. Does the AVI you export as progressive have the same problem as your final MP4? If it does, then Premiere is doing something wrong with the video; if not, then MeGUI is. How you solve the problem will then depend on what the problem is.
|
|
#
?
Aug 1, 2014 23:21
|
|
|
Nidoking posted:Are you changing the framerate of the video at all? Exporting as progressive is probably the right thing to do, since you don't want interlacing, and the frame blending sounds to me like a bad algorithm for changing the frame rate. Does the AVI you export as progressive have the same problem as your final MP4? If it does, then Premiere is doing something wrong with the video; if not, then MeGUI is. How you solve the problem will then depend on what the problem is. Yes this is what happened, I'd set something to 29.97 when all the others were set to 30. It's fixed now! The re-exported AVI looks fine. Thanks! (and judge reinhold too) 
|
|
#
?
Aug 1, 2014 23:29
|
|
|
skoolmunkee posted:Yes this is what happened, I'd set something to 29.97 when all the others were set to 30. Can you elaborate on that a bit? I've got some footage that's 29.97fps and I'll be using Premiere to edit it, so were you changing from 29.97 to 30 and getting ghosting from that? In fact, as a general question to the thread, will I need to change the frame rate to a clean 30 to export it for YouTube? When I did a PS2 game that ran at 50Hz, uploading 25fps videos didn't cause any issues (no one said anything, anyway); does YouTube not care about frame rates the same way it kind of cares about video resolutions?
|
|
#
?
Aug 2, 2014 15:26
|
|
|
frozentreasure posted:Can you elaborate on that a bit? I've got some footage that's 29.97fps and I'll be using Premiere to edit it, so were you changing from 29.97 to 30 and getting ghosting from that? If you have 30 fps footage, but your sequence in Premiere is 29.97, it attempts to conform, usually by blending frames. You want to match your source and project frame rates whenever possible. It's ok if you do 60->30 fps since that's evenly divisible, and that's what Youtube downgrades the framerate to anyway. Keep your frame rate the same. Youtube will keep frame rates below 30 the same, if it's above it'll downgrade to 30. https://support.google.com/youtube/answer/1722171?hl=en judge reinhold fucked around with this message at 15:36 on Aug 2, 2014 |
|
#
?
Aug 2, 2014 15:31
|
|
|
Supposedly YouTube added support for 60FPS videos. Looking at that chart it looks like they only support them for 720p videos. According to various other articles on it, the support is extremely browser/connection dependent and is basically limited to Chrome and how you good you are at making your saving throw against DASH. And that's just watching the videos. For uploading videos at 60FPS, that's apparently limited to YouTube employees. Or maybe not. Who knows? It's YouTube.
|
|
#
?
Aug 2, 2014 16:34
|
|
|
It's a limited test to special partners. Rollout is happening slowly, but my source at YouTube tells me it's planned to happen before the end of the year.
|
|
#
?
Aug 2, 2014 16:41
|
|
|
I've recently begun recording Playstation 2 footage in M2TS containers using a Hauppauge HD-PVR and that much is going great, but in editing and encoding, VirtualDubMod and MeGui crash often but inconsistently due to access violation errors. They'll typically work for a while, then give up and shut down. If a job finishes before a crash occurs, the resulting file works fine, and everything starts back up right away after a crash like nothing happened. This is mostly a problem when encoding long videos with high bitrates in MeGui as it cannot finish the job before crashing and so yields nothing. I attempted to use MediaCoder instead, which also failed halfway through (due to a Avsindex.exe crash, which might be unrelated but I doubt it), but MediaCoder did make that half-finished video into a usable file so I just encoded the other half and merged them. It sure would be nice not to have to rely on such a slap-dash method with many uncontrollable variables since even this probably won't work every time, plus it's making a mess of my workflow. I recently replaced my hard drive, so it's possible I'm missing something I had on the old one or set something up improperly. This never happened with the Xbox 360 footage I recorded and edited the same way, though I haven't had to edit or encode any long Xbox videos since changing hard drives. I've read a lot about access violations now and have ruled out most of the common causes: memtest86 didn't find anything wrong with my RAM, I make sure to run as administrator, and the relevant Windows updates are installed, but nothing yet has determined a cause or solution. Anyone know of anything else that might be happening here before I start retrying everything I've already done up to this point?
|
|
#
?
Aug 5, 2014 11:45
|
|
|
Are you using AviSynth and FFMpegSource? If so, do you have threads=1 in the FFVideoSource statement? By default FFMpegSource attempts to use multiple threads, which is silly, because its threading support has never loving worked. So you need to use threads=1 to tell it to only use one thread. Note that in this case "access violation" means "program crashed." It's an access violation in the sense that the program is trying to access memory it never allocated, because of a bug somewhere in FFMpegSource. (Almost certainly to do with synchronization between threads given that it doesn't happen when single-threaded.)
|
|
#
?
Aug 5, 2014 11:51
|
|
|
Avisynth, yes, but not FFMpegSource. MeGui's file indexer spits out a .dga file with an associated avs script that has the AVCSource plugin loaded already, so I've always started from there. Should I use FFMpegSource instead for these recordings, and if so, how? EDIT: Tried FFMpegSource out of curiosity, it plays the video at double speed, making the 29.97 fps source into a 59.941 fps output that's half as long. My attempts to bring it back down to 29.97 fps just result in every other frame being dropped, not actually slowing the playback down to proper speed. The ChangeSpeed function I use for intentional speed ups and slow downs seems to be able to correct playback, albeit with a redundant frame between each necessary one. I'm using ChangeFPS to drop those frames, but this seems like a really hideous solution. Gonna test encoding now to see if that's at least solved with FFMpegSource. Fiendly fucked around with this message at 21:24 on Aug 5, 2014 |
|
#
?
Aug 5, 2014 13:12
|
|
|
Hi TSF, I'm running into a fairly strange problem with AviSynth and MeGUI that I think is related to the dissolve function. Basically, MeGUI won't render the video file properly. It will get to a point right around where a dissolve cut happens and then it gets hung up and I can't even cancel the process without closing the program. I've run into this problem in the past and it has always been in a video with a dissolve cut. I didn't give up immediately this time around, though. Instead I wrapped the AviSynth file into another one (just using the import function and nothing else) which I remembered hearing in the past could help resolve the issue. Strangely enough, it did, but only for the video file. When I tried to render the audio file it got hung up at the same point in the video as the video file had in the past. Here's a pastebin of the .avs file that puts all the footage together: http://pastebin.com/pQD2mzAb And here's the editing file: http://pastebin.com/ZrbU5vS3 I hope you guys might have an idea what's going on because I just don't know any more! 
|
|
#
?
Aug 5, 2014 22:13
|
|
|
JordanKai posted:Hi TSF, I'm running into a fairly strange problem with AviSynth and MeGUI that I think is related to the dissolve function. I've looked at your script, and I've certainly never had this problem myself, but I have a few suggestions that may help narrow down the problem: 1) It may have to do with the giant mess of splices and trims in the final result. I highly recommend using an accumulator clip variable and splicing/dissolving parts onto it one at a time, line by line, just for sanity's sake. Probably not the root of your problem, though. 2) Does it crash if all you have is a dissolve with two clips? If that works, then you can try exporting each of the dissolved portions as a lossless AVI, reimporting, and just splicing everything from there. 3) If it does crash on just a dissolve, then you've got a much shorter test time to work with. Try changing things like the colorspace, audio rates, etc. to see if any of those changes fix the problem. 4) If all else fails, you might try exporting your "video" clip as it is, lossless, and working with that instead of the conglomeration of FFVideoSource calls - it would eliminate the source filter as a source of errors. It's not the ideal workflow, but you may be able to find an alternate source filter to FFMS that would solve the problem if that's where it lies. EDIT: Two other questions based on a memory error I've had in the past: Can you encode this script (say, into Lagarith) in Virtualdub, or does it also have the problem? If you can't encode it, can you preview it successfully? (Albeit slowly.) I had a memory issue when I sourced a large number of HD video clips, but I was able to export my file as an Uncompressed Direct Stream Copy, then import just that and transcode it in MeGUI. Nidoking fucked around with this message at 23:42 on Aug 5, 2014 |
|
#
?
Aug 5, 2014 23:39
|
|
|
I rendered this at 1600x900 and I'm wondering if it's resulting in any Youtube video rescaling weirdness before I go ahead on making videos or if I need to rerecord with a different resolution. Thoughts? https://www.youtube.com/watch?v=dL1hiMTO7W4
|
|
#
?
Aug 6, 2014 00:16
|
|
|
JordanKai posted:And here's the editing file: http://pastebin.com/ZrbU5vS3 Try threads=1 in your FFVideoSource statements. (See above about FFVideoSource going all screwy if you don't.)
|
|
#
?
Aug 6, 2014 00:22
|
|
|
Youtube's probably just going to re-encode the video at 1280x720. The next step up Youtube supports is 1920x1080, so unless you can go that high, you're probably better off just encoding your source in 1280x720 and skip the re-encoding step (as long as the video is x264/AAC, Youtube should use the original data).
|
|
#
?
Aug 6, 2014 00:23
|
|
|
Xenoveritas posted:Try threads=1 in your FFVideoSource statements. (See above about FFVideoSource going all screwy if you don't.) Do note that this issue is supposedly fixed as of version 2.18. Unfortunately, googling for "FFMS2" still brings up the old Google Code page as the first result instead of the new Github page, so many people are not aware that there's versions past 2.17. https://github.com/FFMS/ffms2/releases
|
|
#
?
Aug 6, 2014 00:25
|
|
|
If you're using the version that comes with MeGUI, it most certainly wasn't fixed as of a month ago. Edit: AviSynth isn't MeGui. Xenoveritas fucked around with this message at 00:34 on Aug 6, 2014 |
|
#
?
Aug 6, 2014 00:30
|
|
|
Any ideas on why my 30 fps clips called with FFVideoSource play at double speed with 60 fps by default? Nothing seems to crash now that I'm not using AVCSource, but all the functions I'm applying to halve the speed of the working clips seem to have more than tripled encoding times. Is this just how it goes or is there anything more elegant I can do here?
|
|
#
?
Aug 6, 2014 00:48
|
|
|
FFVideoSource has fpsnum and fpsden parameters to fix variable framerate files, try that maybe. Or the AssumeFPS function. Though I wonder why that happens in the first place.
|
|
#
?
Aug 6, 2014 00:51
|
|
|
Tried fpsnum and fpsden already, which didn't work, but AssumeFPS accomplished exactly the same thing as the multiple functions I had been using and brought encode times back down to what they should be, so that's a perfect solution. I'm still curious about the root cause as well, but it doesn't matter now. Thanks for the help!
|
|
#
?
Aug 6, 2014 01:20
|
|
|
I'm trying to see if my machine can record/run smoothly enough for video. My graphics card is not that great, and the game I'm running (Miasmata) has an engine built by all of two people and wasn't overly optimized either. I did a test using Action! from Mirillis, as it seems to capture the best for this and has decent compression (by which I mean high bitrate but enough not so big as to tax the hard drive). Here's a few samples. The first is taken at 30fps, but the system doesn't consistently maintain that. https://www.youtube.com/watch?v=Qy_52PU48e0 Here's another from the same file (this is just after loading the game), using 24fps. https://www.youtube.com/watch?v=gNqmGuknvAA In both of those I move a bit more spastically than during most moments of the game. This is taken from the middle of a longer session, for comparison. https://www.youtube.com/watch?v=ktKboaWBA7U (All of these compressed to 5MB/s using ffmpeg/libx264.) My question is which if any is acceptable for viewing? I'd rather not put up something that's too hard to view and just wait until I get a better machine if the answer is 'none'.
|
|
#
?
Aug 6, 2014 20:28
|
|
|
I have a rendering related question. I've noticed that when my videos are rendered out, in the final version there is a fair bit of interlacing when I'm moving quickly, like so: https://www.youtube.com/watch?v=tP1iXsMti-I&t=146s (it's easy to see when I'm turning and stuff) It's hard to convey what I'm talking about, but basically it makes the video look choppy even though it's perfectly smooth. It's kind of annoying to watch once you notice it happening. The interlacing is not there in the source video. I haven't got any idea why or how to solve this (because I am a lost cause when it comes to video rendering), but it's not present- or at least not as bad -in other videos I've watched which is what confuses me about it. I'll just dump my settings here, point out anything that sticks out as odd to you I guess:   edit: Maybe it's the games' motion blur messing with it? Help me I'm completely useless 
CJacobs fucked around with this message at 13:02 on Aug 10, 2014 |
|
#
?
Aug 10, 2014 12:54
|
|
|
That's not interlacing, that's frame blending or motion blur. Motion blur is an intentional effect from the game engine's side, you can probably turn it off if you don't want it. Frame blending is a side effect of frame rate conversion during encoding, whether it happens depends on what the frame rate of your original recording is, what your target encoding frame rate is, and what method you use for frame rate conversion. If you want to avoid blending entirely you will need to use a simple frame duplication/decimation conversion, the main drawback of that is that the result can appear more choppy in playback.
|
|
#
?
Aug 10, 2014 13:22
|
|
|
In your case, the frame blending looks like it's coming from Sony Vegas. It does frame blending if: ~Your render setting is set to good or best and ~Your clip is set to smart resample Either set the clip to disable resample, or render your video using draft or preview. Read more in the guide to preventing Sony Vegas from ruining your video.
|
|
#
?
Aug 10, 2014 21:19
|
|
|
Please could someone tell me if this is possible: I'm capturing with ShadowPlay and would like my audio in different streams for easy editing in Sony Vegas, so; 1. Game audio 2. Music from foobar 3. My USB microphone input 4. Ventrilo audio Each on its own separate stream. I assume I'm going to need to use VAC to get this working if it's possible? I also have dxtory with Lagarith and can use that instead if it would make things easier.. But I'd prefer ShadowPlay because of the zero performance hit and lower file sizes. Thanks Edit: Ok, I've managed to set up VAC correctly... but how I'm doing it right now is capturing the video using ShadowPlay, then using dxtory to capture a 10fps 1x1 resolution video with all the audio streams, then I'm adding the SP capture and dxtory capture to Sony Vegas and deleting the dxtory video so I'm left with the SP Video and 4 audio streams that I can edit. This isn't really ideal though. Is there a standalone audio program I could use to do capture the audio alone or a better way of doing it over all? uhhhhahhhhohahhh fucked around with this message at 20:30 on Aug 13, 2014 |
|
#
?
Aug 13, 2014 08:50
|
|
|
uhhhhahhhhohahhh posted:Please could someone tell me if this is possible: I'm capturing with ShadowPlay and would like my audio in different streams for easy editing in Sony Vegas, so; Game audio should be captured fine with Shadowplay and I believe Shadowplay only captures game audio if you set it to that, so that's easy to take care of. You can dub in music after the fact so you don't really need to worry about trying to capture that. Audacity can catch your USB microphone quite nicely and easily and I'm assuming you have a bunch of people over Ventrillo so you should ideally just have them record their voices in Audacity and send you their recordings for editing.
|
|
#
?
Aug 13, 2014 23:34
|
|
|
ChaosArgate posted:Game audio should be captured fine with Shadowplay and I believe Shadowplay only captures game audio if you set it to that, so that's easy to take care of. You can dub in music after the fact so you don't really need to worry about trying to capture that. Audacity can catch your USB microphone quite nicely and easily and I'm assuming you have a bunch of people over Ventrillo so you should ideally just have them record their voices in Audacity and send you their recordings for editing. Yeah I am using the ShadowPlay game audio, don't need the audio from the dxtory capture. The reason I wanted the music from foobar in a separate stream was so that I could listen to music while playing but then remove the music stream from the render, even if you set ShadowPlay to In Game it still captures the music playing. It's only one person on ventrilo but I wanted their audio separate from everything else so I could clean up the mic hiss that's on it. Even if they send me their audacity recording I have to sync it somehow and I still think I need dxtory + VAC to not have my music stream present in the file render... although I might be able to use VAC without dxtory to remove it from the ShadowPlay capture somehow?? not sure. Is there really no standalone audio program that can capture audio in separate streams like dxtory can? Also: If I'm putting a ShadowPlay capture in Vegas, chopping it up, editing the audio streams and then rendering it again am I going to destroy the quality? It's only going on youtube. I'm still on a 16:10 screen at the moment also... not sure how I'm going to get around this. The only other issue I have is my mic picking up my mechanical keyboard clacking the entire time, not sure how to fix that without getting a new different microphone either.
|
|
#
?
Aug 14, 2014 00:04
|
|
|
If your Foobar music is getting captured as part of the game audio, try switching Foobar to a different output plugin and/or device. If you're using DirectSound for Foobar maybe try using WASAPI instead, and if you're using WASAPI then try DirectSound. For getting rid of sound from a mechanical keyboard, I'll recommend getting a barrier to place between the keyboard and microphone. Having the mic very close to the mouth can help improve signal/noise ratio, since the signal (your voice) will be louder than the noise (keyboard clacking). The only way a different microphone can help with this is if it has some sort of active noise cancelling, by having another microphone pointing away from you that captures the ambient noise and cancels that out from the main signal. High-end headsets are the most likely to have this, but they will typically be intended for telephony and not broadcast/production, audio-quality-wise.
|
|
#
?
Aug 14, 2014 07:48
|
|
|
I think the reason I'm not using WASAPI in foobar already was that it messed with volume control or something. Maybe that was ASIO, can't remember. Its a desktop microphone, sits above my keyboard. I don't have anywhere else to put it. I was thinking of getting a mic that hooks onto the headphone wire instead
|
|
#
?
Aug 14, 2014 07:58
|
|
|
Re: Syncing up lots of voices: Just count down from 3 and have everyone say 'sync', and have all of your co-commentators and yourself click record when you say it.
|
|
#
?
Aug 14, 2014 08:28
|
|
|
CJacobs posted:Re: Syncing up lots of voices: Just count down from 3 and have everyone say 'sync', and have all of your co-commentators and yourself click record when you say it. It's easier if you just all start recording before that, sometimes audacity can wig out a bit before starting to record on slower machines. It's best to err on the side of caution.
|
|
#
?
Aug 14, 2014 09:29
|
|
|
Yeah, usually when I have co-commentators, we turn on audacity a little bit before we actually wanna start the recording to make sure it's not wigging out or anything. Then we'll count to or from 3, say "sync" and then I'll do a thing to line up my commentary with the video, usually just shooting or slashing with a sword three times while counting or just moving up and down a menu.
|
|
#
?
Aug 14, 2014 14:50
|
|
|
Worth mentioning before you start, also have each of your commentators record test audio to make sure their audio looks good--clear distinction on waveform between talking and room noise, test different talking volumes, making sure any potential yelling doesn't clip, etc.
|
|
#
?
Aug 14, 2014 14:56
|
|
|
|
| # ? May 10, 2024 09:59 |
|
|
Just wanted to say thanks for the help I have gotten here getting my audio working right.
|
|
#
?
Aug 16, 2014 01:37
|
|