Hello, I'm wondering if there could please be a preset in order to merge cdg with mp3 to create a video file. Any format video is fine. I saw earlier in the thread a guy got it to work, but I'd like to know how to do it in batch form. I'm pretty unfamiliar with command, but I can try! Thanks for your time.
+ Reply to Thread
Results 211 to 240 of 277
Hi, you can use stream multiplex tab to achieve this, but batch creating videos from image is not available there. It may be possible in main tab, if the image file is always the same. For example, using the image Piant.jpg, and getting the audio file from the file list:
-r 1/5 -start_number 0 -i "C:\Users\Abel\Pictures\Piant.jpg"Code:
-c:v libx264 -vf fps=5 -c:a aac -b:a 160K
Unfortunately the "image" in this case is words. CDG+MP3 is basically a karaoke song. They have the same name, but that's about as far as that goes. Any other ideas?
You've got it the other way around. In this case, the Karaoke "words" are an image. "CDG" streams are images, timed to the music/audio, and placed into auxilliary sectors multiplexed along with the digital audio data in a CD. Very much like the DVD "sub" subtitles, or similar still image format for Bluray subtitles. It is an image overlay - like how greenscreen works.
The text is converted to images before being put onto the CD. When ripped from the CD and put into an "MP3+G" format, it is not yet converted from cdg pix format, only multiplexed with the MP3 file (the data is appended like an ID3 tag, IIRC).
So, you will need to first extract the CDG data stream from the MP3, then convert it from the arcane CDG format (which is ONLY used for CDG material) into a standard picture format (like JPG, PNG, etc).
From what I have seen of most (paid) apps on the net, they do this by rendering the CDG as raw pixels and encoding the render, rather than interpreting the characters and performing OCR, etc.
A quick look at cdg support in ffmpeg makes me think that it is geared only toward muxing+demuxing with MP3, rather than in decoding/rendering for use as a standard picture sequence format.
Last edited by Cornucopia; 18th Feb 2019 at 01:20.
Just come across this and it seems to be the answer to my woes in converting a load of ISMV files to MP4.
I am, however a bit dense at figuring things out. Basically all I'm trying to do it output the file into the same location as the original input, while renaming the file to video.mp4.
No files will be overwritten as each ISMV is in a folder on its own.
Could anyone kindly give me a nudge in the right direction? I managed to do a straight conversion by using
ffmpeg -i filename.ismv video.mp4
Hi, you should not have any problem achieving that if you process a single folder and select it as output folder too.
However if you add multiple input folders, for the moment you can't do that, because for relative paths, even checking "Recreate source path" a FFBatch folder will be created wiithin.
[QUOTE=ethanhunt202;2543825]How do I go about altering the output title in that case? I couldn't figure it out yesterday
You can enable "Rename output" and then write what will be added at the end of every input file name to generate the output file name.
Hi Eibol, is there a way that I can add a comment in the ini file so I can store a version of it? I am starting to make other coworkers to use it, so I give them the conf and they just need to update it, but now I've got a problem to know what version I gave them or on what of my machine there is the latest one that I've created.
As always, thanks for your great gui
Hi, there is not configuration versions control in the interface, you would need for the moment to do it manually.
You can add any text, like a comment or a version number, as a new line at the end of the configuration file, provided you don't start it by 'PR:".
The file path isCode:
I'm afraid the workaround I recommended below causes a runtime error, it expects a preset format, so the only way to add version information would be to add a "dirty" preset, stating the version in the name and the parameters, so at the end of the presets lists it would appear something like "Configuration v1.2".
I will look into it in a future release.
Last edited by Rajesh Singh; 9th Mar 2019 at 01:47. Reason: Duplicate post
Hi, I have implemented in this executable some manual configuration version control, as well as a new button to edit the configuration file inside the application.
You just need to press "Save settings" button once and the 'Version' line will be included at the right place in the configuration file, so you can then use the new edit configuration button to change it, or edit the ini file as you wish.
[Attachment 48353 - Click to enlarge]
-c copy mp4 yes Vs grid_yes keep_no subf_no Version 1.0 PR: Video: MP4 Stream copy & -c copy % mp4 PR: Video: Convert audio track to AAC HQ 2 channels & -c:v copy -c:a aac -cutoff 20K -b:a 256K -ac 2 % mkv PR: Video: Convert to H264 HQ + Source Audio & -c:v libx264 -crf 20 -c:a copy % mkv
Last edited by Eibol; 12th Mar 2019 at 09:19. Reason: Link & typo
New version 1.6.5
- Latest FFmpeg 4.1.1 stable (february 2019).
- New edit settings button, to manually edit settings file.
- Automatic settings backup on edit/restore features.
- Added settings version code.
- Improved refresh file list feature.
- Other Minor bugfixes.
many thanks for this useful program!!
two house-keeping features that I did not found in v1.6.5
a) an option to move input files to the recycle bin, after successful conversion
b) an option to have the output path the same as the input path
- I can have a "..\FFBatch" relative subfolder, or an absolute path, but I cannot edit the output path to "."
Hello, this tool is really useful, but it my opinion, it lacks to the possibility to add more files after the encoding has started
Hi Eibol, I built a new machine (Ryzen 7 2700, non-X, with 16gb ram) and when I try the auto multi thread after a bit it starts to fail every single clip, then I tested by manually setting the thread numbers and when I set at 16 it start to fail again. Is there a log for failed clips to see what is happening and share it here?
One more thing, could you please clarify to me how the Multi-file conversion button works? The number that you have to put in is referred to the thread count or the number of files that it has to convert? Because if it's the first case, well I could not take advantage of a processor that has a bunch of cores (eg. 8, 16 threads) and using multiple of them to convert a number of clips (eg. 4) with a codec that benefits from multi threaded encoding (case of the eg. 2 cores, 4 threads per clip).
Source was a mix of h264 (in 420 8 bit and 422 10 bit) and some h265 (420 10bit), mixed frame rates. This is actually just a testing, I actually converted this clips with multi file at 4 on a i7 4790k without any issue when I did this job some weeks ago.
P.S.: Sorry for forgetting again to talk a little bit about what I'm actually doing ahahah
Hi, I'll try to clarify the usage of sequential/multi processing, I think it can be a little bit confusing:
- For video encoding using cpu, sequential processing is the choice, because ffmpeg librearies (libavcodec, etc.) are already usually capable of using all cpu cores and threads, pushing cpu usage to 100%. Using multi-file processing in this case would be pointless and even counterproductive.
- The only cases in which either automatic or limited multi-file can make a differences are:
- GPU encoding, since there are gpus that can process more than one file at a time. In this cases automatic multi-file will fail if the number of files is greater than the processes. For 1050TI it is just two sessions at once:
- CPU encoding for audio or stream copy operations- Audio libraries are usually designed for monothread processing, so multi-file will "simulate" multi-thread by launching several file conversions at a time.
- Automatic Multi-file processing just launches as many ffmpeg processes as Windows considers necessary to use all system resources, it uses a library for automatic parallel processing.
So if you encode using libx264, libx265 or prores_ks, sequential processing should be using 100% of cpu resources. If that is not the case, you may try processing two files at once.
If you encode using NVENC, you may try processing two files at once too, a higher number would make the gpu to reject further simultaneous sessions.
I hope this makes things clear. Anyway, in newer systems, I can't say for sure this will be the behaviour, but it should not change much.
Last edited by Eibol; 2nd Apr 2019 at 13:00. Reason: typo
With the new cpu I'm not hitting 100% usage, and prores_ks is just ~14% usage (but still completes faster then my old cpu), and I know for sure I'm not filling the read/write bandwidth. I'll experiment a little bit then I'll write back here if there are any updates.
I was using an old 1.5.9 version,later updated to the latest version and now I'm disappointed with its performance.I compared this two versions with same file and preset, encoding speed of 1.5.9 is up to 7x and new version only 1.8x. What could be the reason?
I honestly don't know the reason, maybe the latest stable ffmpeg release is not working so good as the previous "nightly" builds included.
You can uninstall and reinstall 1.5.9, and then overwrite the executable with this beta.
You can also let me know the parameters you used so I may test.
Last edited by Eibol; 3rd Apr 2019 at 10:06. Reason: Typo