Video editors, editing WMM, VSDC ect

[EEVblog]

There is a simple principle involved in ANY photographic pursuit: Get your raw shoot material as close as possible to finished quality as you can.  Post production processing is an inordinate waste of time if you have to correct for things you could have (and should have) captured properly at the shoot.  Getting it right straight into the camera will produce a better quality result, too.  I believe THIS is what Dave is saying.

That's exactly what I'm saying.
Another massively important thing to do is to shoot your clips in order. No messing about ordering stuff in editing.

[Radio Tech]

Basically it comes down to the quality available on youtube.
Even if you have the world's best video camera, and you render to 1TB of uncompressed raw video (some of my videos could be!  ) and upload that to youtube, the playback quality will be no better than a version heavily compressed (that is 1/100th the files size) with Handbrake and uploaded.
Handbrake is so good, even the standard quality factor, it's almost impossible to tell the difference from the original footage. More than good enough for Youtube and almots any other platform.

The standard constant quality factor on Handbrake is 21 or 22, and I use 19 which is really guilding the lilly, for no good reason than because I can.
The final file size will depend upon how much moving content is in your videos. My screen capture videos for example compress down to ridiculously small files sizes, with no loss in quality. On the other hand, one of my outdoors video with tons of motion and tress and whatnot won't compress well at all, in which case I usually don't even bother with Handbrake, I just upload the rendered version.
But most of my videos though get maybe a 5 fold decrease in file size by using Handbrake over what I render in.
I do not archive the rendered version, only the Handbrake version I upload.

Good information Dave.
So with everything said then I think I may have a plan in place.  My videos have very little action shots. Not a lot of movement unless I pan from the unit under test to a shot of the test gear, but even then I would like to use a separate camera to capture that shot and overlay that onto the video. See you do that a lot and the results look great.
So my plan is to continue using WMM. ( since I like the way it works)
Get a converter to change the format.
Study up on handbrake.
Thoughts

Thanks for all this information.

[madalf71]

A newbie to video editing as well.
I was recently shown Camtasia,  seems user friendly.
Anyone used it, have any thoughts?

[Brumby]

A newbie to video editing as well.
I was recently shown Camtasia,  seems user friendly.
Anyone used it, have any thoughts?

I haven't dipped my toe into that pond yet, but I am interested in what responses you get....

[edy]

There is a simple principle involved in ANY photographic pursuit: Get your raw shoot material as close as possible to finished quality as you can.  Post production processing is an inordinate waste of time if you have to correct for things you could have (and should have) captured properly at the shoot.  Getting it right straight into the camera will produce a better quality result, too.  I believe THIS is what Dave is saying.

That's exactly what I'm saying.
Another massively important thing to do is to shoot your clips in order. No messing about ordering stuff in editing.

To add to Brumby and Dave's excellent advice:

I've been using Womble for years to prepare videos for my YouTube channel. It will do DIRECT STREAM COPY editing which means no wasted time transcoding/re-encoding raw edits, saving a huge amount of time. http://www.womble.com/.  I believe it can handle MPEG-1 and MPEG-2 direct streamcopy editing, but will let you edit other formats as well since it just uses the system's codecs (but will need to re-encode).

My Sony camcorder outputs MPEG-1 directly to the built-in hard drive. With Womble, I can take all the videos and do chops and edits and transitions and I output to the same MPEG format/bitrate (Womble detects input and automatically lets you choose output to be same). The only re-encoding going on are the transitions (fades, edits, effects), or if you are playing with the audio (volume boosting or editing) it will re-encode audio but that is fast. Like this...



Meanwhile, all the video bits in-between gets copied unmodified. This is tremendously fast and also no change in quality from the source. At that point, you need to ask yourself this....

1. is it worth compressing it down to MP4 using Handbrake or FFMPEG before uploading...  or....
2. is my bandwidth and DL/UL limits big enough to just upload the original MPG and let YouTube handle it

I find most of the time #2... I would waste more time trying to re-encode or compress down my MPG files than just uploading the entire thing to YouTube. And I have enough bandwidth and monthly limits that I didn't care to upload a file that was 2x the size. YouTube would then do all the necessary compressions if needed. In fact, in the time it took for me to convert my video to MP4 and upload it, I could upload my original MPG video and re-download the resulting MP4 file from YouTube using a YouTube downloader!

Bottom line.... Your camera may output MP4, MOV or other format. Try to get an editor that does direct video stream editing if possible for that format to avoid wasting time transcoding/re-encoding, and upload the resulting file to YouTube without bothering to shrink it down if your bandwidth is faster than your computer (which is my case). If your computer is a speed-demon and have crappy slow bandwidth, it may be worth compressing it locally. I prefer using YouTube to do it.

Now.... I have several cameras I use besides the Sony.  Some output AVI, some output MP4, some output MOV... I still find myself converting my files with "ffmpeg" to MPEG-1 at a very high bitrate (same or higher than the source) so that I could use Womble to edit the MPEGs. I know MPEG-1 is not particularly good compression-wise, but if you use a high quality factor it shouldn't lead to any degradation of the video. Then I let Womble do all the streamcopy editing at blazingly fast speeds, add my transitions and so on, and then upload the resulting MPEG.

I still come out ahead this way than trying to use an editor that can handle AVI, MP4, MOV, MPG and other formats using the system codecs and transcoding/re-encoding everything, without telling me that direct streamcopy is available. At least Womble tells me before I hit "Process" a complete timeline showing me what portions of the video are going to be re-encoded, what portions are streamcopied for both audio and video. Streamcopy is just incredible if you can get away with it and get your sources all lined up to the same parameters...

No issue if using output from same camera, but if you have to combine videos from various cameras/sources with different resolutions, formats, etc... best to create some batch over-night conversion of all your raw project data into the same common stream so you can do all your editing that way. Just pick the largest bitrate and resolution of your intended output and get everything to be common.

Another reason this is important is because sometimes you spend 20 minutes rendering your video and you notice you have a glitch or you forgot to snip something.... Now you have to waste another 20 minutes to reprocess your video. With streamcopy editing, you fix up your glitch and within a few seconds the output is done again because it is just COPYING blazingly fast. No re-encoding! 

[mariush]

MPEG-1 is really ancient, above 352x288 it shouldn't be used.

You probably mean MPEG-2, it was normal for cameras to record SD content in MPEG-2. Nowadays higher end cameras still record HD in MPEG-2 but at high bitrates (60-100mbps) and with uncompressed PCM stereo sound. However, for Youtube and other not so sensitive material, it's easier to record in h264 at up to 20-25 mbps, especially if you want to keep the original footage besides the edited video.

h264 is much more complex compared to mpeg-2, and without knowing the exact encoding parameters, it's difficult if not impossible to direct copy a section of the stream and re-encode a segment.

If your editing software renders the project once to a file with h264, it would know the exact encoding settings and would be able to render a second output file by copying content from the original rendered file and re-compressing what changed (but this implies keeping previous rendered files just in case user wants to render again, which is not guaranteed).
Most editing products don't do this though, because to be able to do this, the encoding settings have to be... "looser", basically you get less quality per bits if you want to make it possible to do this.

x264.exe (the open source codec) has a specific command line parameter to allow this (besides API commands) : --stitchable  ... basically you can encode several segments and output to raw bytestream .264 file (not packaged in a container) and then you can just do  COPY /B file1.264+file2.264 output.264  and the file will just work as if it was encoded in one shot. An editing application could use the x264 API to render 1-5 minute long segments of the project and only re-render segments that have changed due to editing.

If the rendering takes a long time, you can always render the project using a lossless but fast codec like MagicYUV (renders in Vegas at 2-4x real time at 1080p)  and then recompress to h264 using handbrake or command line x264.exe ... the only problem is that such lossless codecs need a lot of disk space, gigabytes per minute of video.

If you use x264.exe, you can then choose on using higher quality encoding settings and slower re-compression if you have a slow upload speed (for example, encode 1h of video in 50 minutes at 1080p and 4-6mbps), or you could just use some fast encoding settings (like --preset superfast for example coupled with --crf) and you'll get larger files but compressed fast (for example, encode 1h of video in 20 minutes at 1080p with bitrate varying between 8 and 16 mbps, depending on scene complexity)(
Handbrake exposes fewer options but it should still work quite well to recompress huge

[edy]

MPEG-1 is really ancient, above 352x288 it shouldn't be used.

You probably mean MPEG-2, it was normal for cameras to record SD content in MPEG-2.

I'm using some old hardware so I think by now everything has completely changed.    We're talking computer and Sony camera from about 8 years ago, all meant to work with DVD resolution sources (480p or 720x480 NTSC). Womble is DVD-era software and perhaps ideal for that, but will suffer same issues dealing with newer formats requiring heavy processing. 

But I've been importing also SJCAM footage (h.264 MOV 1080p) and downloaded MP4 files, etc... Usually when I work with non-MPEG1/2 I would batch convert everything to MPEG2 using "ffmpeg" , then use Womble to edit and add transitions/effects, and output in the same MPEG2 format so it doesn't need to re-encode the whole stream... only transitions.

The total time for all of that is probably LONGER than someone just working on the original MP4 files, on a faster computer. I thought perhaps streamcopying was still available in the newer formats and that there were editors that could handle it on the native file. Otherwise from what I know "ffmpeg" prefers to copy stuff over to ".ts" files and then you concatenate them:

Convert the files to TS intermediate file first like this:

"c:\program files\winff\ffmpeg" -i 1.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts 1.ts
"c:\program files\winff\ffmpeg" -i 2.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts 2.ts
"c:\program files\winff\ffmpeg" -i 3.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts 3.ts

Then join them like so:

"c:\program files\winff\ffmpeg" -i "concat:1.ts|2.ts|3.ts" -c copy -bsf:a aac_adtstoasc output.mp4

 

I need a new laptop! 

[Someone]

Lots of different people work in different ways, its arrogant to claim yours is the one true way and every other possibility is wrong.

I claimed no such thing. Just saying that I know a thing or two about Youtube video production.

Still, a little time spent adjusting each clip could further improve the quality of your videos. Even in your purpose built and well lit space there are still exposure and balance problems.

[Richard Crowley]

You say its wrong, others disagree, manipulation of video is just another option for producing video. Only the very high end production cameras have enough in camera controls to produce broadcast deliverables straight up (TV uses these extensively to reduce expenses in the workflow) and you can improve the information in a video by utilising its possibilities rather than presenting "auto" from a camera.

I do everything from low-end, single camera capture/streaming all the way up to my HD digital flight-pack with six cameras, live switching/effects, etc.

There is a place for BOTH types of production. But if you are making the typical YouTube channel, Dave is absolutely correct. There is NO REASON to have to do extensive video processing. If you have to do ANY kind of post-processing manipulation of the video or audio then you are doing something wrong.

If you LIGHT the "set" well, you can use a $300 camera and make videos as good as anybody on YT, and WITHOUT post-processing the video. And likewise for audio. If you use the proper mic in the proper place and set levels properly, etc, you can produce decent audio with a <$50 microphone.

Some photos of my rig back when I had only 4 cameras. For this event we used 3 of the cameras on remote-control pan/tilt heads. Along with full camera/lens remote control.
http://www.rcrowley.com/rojas.htm

[Someone]

You say its wrong, others disagree, manipulation of video is just another option for producing video. Only the very high end production cameras have enough in camera controls to produce broadcast deliverables straight up (TV uses these extensively to reduce expenses in the workflow) and you can improve the information in a video by utilising its possibilities rather than presenting "auto" from a camera.

There is a place for BOTH types of production. But if you are making the typical YouTube channel, Dave is absolutely correct. There is NO REASON to have to do extensive video processing. If you have to do ANY kind of post-processing manipulation of the video or audio then you are doing something wrong.

Saying Shouting that any video manipulation is doing it wrong is just the same thing. Adjustments and looks are quick to add and improve the communication of the video, the above example flattening the content could be achieved with studio lighting and time spent setting that up for the shot or it could be achieved as above with a channel curve in editing. Which would be quicker and easier for the average YouTube producer?

[EEVblog]

Good information Dave.
So with everything said then I think I may have a plan in place.  My videos have very little action shots. Not a lot of movement unless I pan from the unit under test to a shot of the test gear, but even then I would like to use a separate camera to capture that shot and overlay that onto the video. See you do that a lot and the results look great.

No, I very rarely use a 2nd camera. I simply use the same camera and move position etc. Same with the closeup shot, I screw and unscrew my macro lens as needed.

[EEVblog]

Bottom line.... Your camera may output MP4, MOV or other format. Try to get an editor that does direct video stream editing if possible for that format to avoid wasting time transcoding/re-encoding, and upload the resulting file to YouTube without bothering to shrink it down if your bandwidth is faster than your computer (which is my case). If your computer is a speed-demon and have crappy slow bandwidth, it may be worth compressing it locally. I prefer using YouTube to do it.

You also have to think about archiving the footage. Rendered versions can be very large, that what Handbrake is good for. It saves archive space and speeds your upload time.
Transcoding with Handbrake for me is literally a few seconds work. I have script and I just drag my rendered video on that desktop icon.
Speed is typically 150fps or something, so 10's of minutes later at most I have a finished video ready to upload and archive.

No issue if using output from same camera, but if you have to combine videos from various cameras/sources with different resolutions, formats, etc... best to create some batch over-night conversion of all your raw project data into the same common stream so you can do all your editing that way. Just pick the largest bitrate and resolution of your intended output and get everything to be common.

I've never had the need to do that. My video editor handles anything I throw at it.
If you have to transcode video before you edit then once again, I'd day you are doing more work than you have to.

Another reason this is important is because sometimes you spend 20 minutes rendering your video and you notice you have a glitch or you forgot to snip something.... Now you have to waste another 20 minutes to reprocess your video. With streamcopy editing, you fix up your glitch and within a few seconds the output is done again because it is just COPYING blazingly fast. No re-encoding! 

I've never had an issue.

[EEVblog]

Still, a little time spent adjusting each clip could further improve the quality of your videos. Even in your purpose built and well lit space there are still exposure and balance problems.

That zero people complain about.
It's not worth the effort, really. It will bring in zero extra viewers, and just takes time away from either producing more content, running my business, or having a life.

[EEVblog]

If you LIGHT the "set" well, you can use a $300 camera and make videos as good as anybody on YT, and WITHOUT post-processing the video. And likewise for audio. If you use the proper mic in the proper place and set levels properly, etc, you can produce decent audio with a <$50 microphone.

Indeed.
My studio lighting is still pretty poor in the scheme of things. But the wide range and styles of video I make make it really difficult to get consistent lighting. So I just pretty much make do with what I have. Studio lights on stands etc are a real PITA. Fine if you do the exact same video in the exact same spot every time. But if you don't then they are more trouble than they are worth. There is a reason why I relegated them down to the bunker.

[EEVblog]

You say its wrong, others disagree, manipulation of video is just another option for producing video. Only the very high end production cameras have enough in camera controls to produce broadcast deliverables straight up (TV uses these extensively to reduce expenses in the workflow) and you can improve the information in a video by utilising its possibilities rather than presenting "auto" from a camera.

There is a place for BOTH types of production. But if you are making the typical YouTube channel, Dave is absolutely correct. There is NO REASON to have to do extensive video processing. If you have to do ANY kind of post-processing manipulation of the video or audio then you are doing something wrong.

Saying Shouting that any video manipulation is doing it wrong is just the same thing. Adjustments and looks are quick to add and improve the communication of the video, the above example flattening the content could be achieved with studio lighting and time spent setting that up for the shot or it could be achieved as above with a channel curve in editing. Which would be quicker and easier for the average YouTube producer?

You don't quite get it. The point is neither of those options is good for a Youtuber who produces a lot of content regularly.
People simply don't care how perfect the look of your video is provided it's framed well, in focus, and exposed reasonably well enough.
Your correction above is fine, but it's a tweak that no one will care about.
Some of my videos have over one hundred different shots (50-70 average), often with each shot exposed differently etc. Should I really go and tweak them to get it perfect? Half of them? even 10 of them? No, that's just silly.
If you want to do that, then hey, go for it, more power to you. But as a youtube who has produced over a 1000 videos and hasn't gone insane doing it, my opinion is it's a waste of time.

[VEGETA]

If you want a professional method, here it is:

1- when you finish editing, produce it using a lossless video codec like Lagarith (or uncompressed if you can handle it). this will give you a huge file size but quality loss is exactly 0.

2- use the best encoder of all times, x264, to encode it. Of course, you have to learn its commands (I do). Something like this will do most jobs (assuming you have the t-mod version of it):

x264.exe --preset veryslow --bframes 10 --ref 16 --crf 16 --aq-mode 3 --aq-strength 0.8 --qcomp 0.6 --b-adapt 2 --b-pyramid 2 --deblock -1:-1 --subme 10 --trellis 2 --psy-rd 0.75:0.05 --direct auto --output "your_output_file.mp4" "your_input_file.avi"

^ it is CLI, so stuff like handbrake doesn't do a damn thing at all! it is just a GUI no more and maybe less. so the line "handbrake is good and produces quality videos" is wrong because it doesn't do it, x264 does.

of course there is the 10-bit version of it that I don't recommend for youtube, at least for now. There is also the HEVC new codec with its x265 encoder (based on x264 code) going around and it is already giving quality results near x264 or maybe better.

the settings above can be modified and there are more, but no need to dig deep.

3- the left thing is audio... assuming you chose "PCM" or uncompressed WAV in step 1 (which you should), then you should use a quality encoder for AAC like QAAC (it requires itune to work), or even Nero AAC... nearly all of them work fine and all of them are CLI applications... so no GUI. it is not hard, just a very simple line that you can copy and paste all the time.

4- now you have an H.264 quality video with no audio (x264 doesn't encode audio, some mods have some options), and an AAC audio... you must mux them. use YAMB to do it which is free GUI for a tool called "MP4 box". The output is MP4 video with audio.

5- upload to youtube.

I plan to use this method in my next videos, because I didn't have some space in my hard drive to use lagarith codec.


____

Notes:

1- I am sure there are options in handbrake to use the x264 encoder directly without handbrake's silly presets, where you post the command line you want. Search for it.

2- Of course, HB must has the x264.exe in its folder, u can replace it with the current version of x264 or any mod (like tmod that I use), just keep the naming right. the option "aq-mode 3" doesn't work with normal x264, if you don't have the tmod, choose aq-mode 2 instead.

3- Blender, Camtasia studios,... etc don't support CLI encoders, but VFW versions of it... there is one for x264! just be sure to update the x264.exe in it to the latest version or mod then you can use it directly with your video editing software without the need of Lagarith and step 1!!! it allows command line stuff too!!



I hope this long post did help... Now I have these requests from you guys:

1- a good (preferably free) screen recording software other than Camtasia studio.
2- Is Canon Ixus 160 good enough? it is the absolute cheapest option in my country.
3- for recording the desk where I describe stuff or solder or breadboard... what is the tripod to be used? I think the normal one is just too big.

 

thanks!
 

[EEVblog]

2- use the best encoder of all times, x264, to encode it. Of course, you have to learn its commands (I do). Something like this will do most jobs (assuming you have the t-mod version of it):

You don't have the use the commands, that's what Handbrake is, it's a nice shell around the x264 encoder.
I do however use a command line script, just drag'n'drop my rendered file onto it.

[EEVblog]

There is also an x264 plugin for Video For Windows that allows you to use it with video editor to render directly.
https://sourceforge.net/projects/x264vfw/
But I've been able to get ti working properly, I have audio sync issues. Seems buggy.

[VEGETA]

2- use the best encoder of all times, x264, to encode it. Of course, you have to learn its commands (I do). Something like this will do most jobs (assuming you have the t-mod version of it):

You don't have the use the commands, that's what Handbrake is, it's a nice shell around the x264 encoder.
I do however use a command line script, just drag'n'drop my rendered file onto it.

I know, but not all of its stuff is good. For example you said "Constant quality". There is no such thing.. it is "constant quantizer" which is the "crf 16" that I wrote. people will chose constant bitrate most of the time which is a bad method.

For me, all x264 guis are just pointless. I use t-mod x264 which has some more good options like "aq-mode 3" and MixAQ. If you really really insist on a gui... You must then use x264 VFW that is usable in all packages once you install it. It allows writing command lines and also way much better than any other GUI. Here you won't need to render the video then encode it... it happens in one step.


audio sync problems doesn't have anything to do with x264 or any of its guis. if you use genuine tools (away from guis and "converters") you won't face any problems at all as I have never been able to during all my encoding journey.


thanks

[Someone]

You don't quite get it. The point is neither of those options is good for a Youtuber who produces a lot of content regularly.
People simply don't care how perfect the look of your video is provided it's framed well, in focus, and exposed reasonably well enough.
Your correction above is fine, but it's a tweak that no one will care about.
Some of my videos have over one hundred different shots (50-70 average), often with each shot exposed differently etc. Should I really go and tweak them to get it perfect? Half of them? even 10 of them? No, that's just silly.
If you want to do that, then hey, go for it, more power to you. But as a youtube who has produced over a 1000 videos and hasn't gone insane doing it, my opinion is it's a waste of time.

You both have said straight up that its doing it wrong, it really isnt. If you dont want to spend the extra seconds per clip to improve the visual content then thats fine, its not wrong either, but you end up with a video that has the look of an amateur videographer or someone who doesn't care much about the visual content. You put a lot of effort into the audio and content which really comes through well, so much so that I usually "watch" your videos without even following the imagery and just tab to the window when there is something that needs to be seen. You come across really well in the podcast format and this is reflected with your priorities and decisions in putting together videos. Play to your strengths but don't claim its the one true path and everyone else is doing it wrong.

[mariush]

Vegeta, Youtube will recompress the video anyway, so it makes no sense to use such slow options, you can just use a faster preset coupled with a higher crf

Let's analyze.. x264.exe --preset veryslow --bframes 10 --ref 16 --crf 16 --aq-mode 3 --aq-strength 0.8 --qcomp 0.6 --b-adapt 2 --b-pyramid 2 --deblock -1:-1 --subme 10 --trellis 2 --psy-rd 0.75:0.05 --direct auto --output "your_output_file.mp4" "your_input_file.avi"

--bframes 10 forces clip to be encoded with level high 4.0 or 4.1 (depending on resolution) which makes the video potentially unplayable on some devices ( preset veryslow configures bframes to 8 which is more than enough for youtube videos, and you're overriding it to 10)
--ref 16 is redundant, because it's set by --preset veryslow.  If you use better crf and faster preset (like --preset slow with --ref 5) you'd be fine and get pretty much same quality after youtube recompresses everything . Raising the --ref to 8 is basically the major difference between this profile and  --profile "slower".
--deblock -1:-1 is potentially a bad choice because it lets x264 slightly blur the image which may be bad if your Youtube videos have lots of screen capture... not to mention that you get the same result just using --tune film  ...
psy-rd 0.75 changed from the default of 1 is again something that makes sense for cinema movies... you're basically telling x264 that it's ok to blur some more and drop grain if it improves compression in frames that are really bitrate starved, which again may not be smart for youtube videos (and pointless because most youtube content is static scenes with Dave talking or showing up something in the case of EEVblog).
-aq-mode 3 is overkill for youtube videos, the default 1 is good enough... youtube videos don't have transitions to dark scenes and the other way around, so that you'd want x264 to take bits from some frames and allocate more to fade outs or darker scenes and so on.. that's what basically aq-mode 2 or 3 does, tweaks the adaptive .
--direct auto is redundant, it's included in all presets higher than medium

so basically the command can be shrinked to x264.exe --preset slow --tune film --crf 16 --output film.h264 input.avi which I'm sure you'll find in Handbrake with no issues, then mux audio with this file into a .mkv file using MKVToolnix  (or just mux using handbrake).

For youtube, you can just use --crf as low as you can afford , basically it's up to you and your internet connection, how much you're willing to upload for each minute of encoded video.

and PS. avoid Lagarith, since it's Windows only and relies on floating point implementation which is x86 dependant, some future processors may be different enough that you won't be able to decode your files or your encoded files won't be playable by people using older computers or ARM based systems. MagicYUV (and other lossless codecs like UTVideo or Huffyuv) are much much faster than Lagarith.

I know, but not all of its stuff is good. For example you said "Constant quality". There is no such thing.. it is "constant quantizer" which is the "crf 16" that I wrote.

No, --crf is short for constant rate factor, which is pretty much "constant quality", in a super simplified way it's constant quantizer on steroids... constant quantizer is sort of like treating each frame of the video as a jpeg picture and trying to compress each frame at a fixed quality preset (like jpeg 85%), ignoring how much space that frame will take. Constant Rate Factor also does some psycho-visual stuff, it analyzes each frame and sequences of frames to see where there's darker areas in the frames or stuff that your eyes won't notice easily and moves the saved bits towards the other.. and plays with the underlying qp varying in small steps the quantizer when needed to preserve quality

For me, all x264 guis are just pointless. I use t-mod x264 which has some more good options like "aq-mode 3" [...]

--aq-mode 2 and 3 are built into the default x264.

[EEVblog]

You don't quite get it. The point is neither of those options is good for a Youtuber who produces a lot of content regularly.
People simply don't care how perfect the look of your video is provided it's framed well, in focus, and exposed reasonably well enough.
Your correction above is fine, but it's a tweak that no one will care about.
Some of my videos have over one hundred different shots (50-70 average), often with each shot exposed differently etc. Should I really go and tweak them to get it perfect? Half of them? even 10 of them? No, that's just silly.
If you want to do that, then hey, go for it, more power to you. But as a youtube who has produced over a 1000 videos and hasn't gone insane doing it, my opinion is it's a waste of time.

You both have said straight up that its doing it wrong, it really isnt. If you dont want to spend the extra seconds per clip to improve the visual content then thats fine, its not wrong either, but you end up with a video that has the look of an amateur videographer or someone who doesn't care much about the visual content. You put a lot of effort into the audio and content which really comes through well, so much so that I usually "watch" your videos without even following the imagery and just tab to the window when there is something that needs to be seen. You come across really well in the podcast format and this is reflected with your priorities and decisions in putting together videos. Play to your strengths but don't claim its the one true path and everyone else is doing it wrong.

I'm claiming no such thing. I'm just pointing out what I do as a professional Youtuber, and what I suggest other Youtubers who produce similar content should do as well. Just like the OP is asking.
If you produce different type of content to me and you have different goals, then YMMV.

I have never heard of a single instance where someone has said they didn't subscribe or didn't want to watch my videos any more because I didn't tweak every clip to adjust the brightness and contrast etc, and trust me, I get a lot of feedback.
I'm serious when I say no one cares, because they don't, it's a fact, so why even spend "seconds" doing yet another step that won't really gain you any benefit?
BTW, it's not just "seconds", it's usually a but more than that. And like I asked before, what do you do when you have a hundred clips with different brightness and contrast? Tweak them all? Come on. It's a Youtube video, not a National Geographic documentary.
Can some of my clips do with some tweaking? Sure, I just shot a video today that had some horrible fixed exposure issues (again, white paper, dark meter), but did I bother to fix it? No, it's juts not worth it, no one cares.
Even if it's only a few seconds per clip, every added step adds to annoyance. Make 1000 videos and you'll understand that it's the little things that matter to keep your sanity and keep you enthusiastic about continuing to produce content.

BTW, you said "You put a lot of effort into the audio". If you are fussed over how I don't tweak my video quality during editing, you'd be shocked to learn how little effort I put into the audio, bordering on zero. Again, my audio is "good enough", so I rarely bother to tweak it.
Audio is however more important than video, because poor audio really stands out to people.
I get zero complaints about my video quality (ignoring hand held site footage), in fact, I only praise. I do occasionally get some complaints about audio though, that matters to people a lot more than video quality does.

[Someone]

BTW, you said "You put a lot of effort into the audio". If you are fussed over how I don't tweak my video quality during editing, you'd be shocked to learn how little effort I put into the audio, bordering on zero. Again, my audio is "good enough", so I rarely bother to tweak it.
Audio is however more important than video, because poor audio really stands out to people.
I get zero complaints about my video quality (ignoring hand held site footage), in fact, I only praise. I do occasionally get some complaints about audio though, that matters to people a lot more than video quality does.

An array of microphones to pick from, carefully positioned, pop filters etc, you do put effort into your audio and there are obvious edits and adjustments in the final video that aren't matched with similar detail in the imagery, we haven't seen you using bounces or fills in lighting for example. If you were going to get it right first time in camera you would probably spend a lot of time adjusting a lighting rig.

I'm claiming no such thing. I'm just pointing out what I do as a professional Youtuber, and what I suggest other Youtubers who produce similar content should do as well. Just like the OP is asking.
If you produce different type of content to me and you have different goals, then YMMV.

Lets see,

But if you have to do any colour, whitebalance, or other processing for youtube videos, then I must say that you are doing it wrong. You shouldn't have to do these things, get it all right in-camera.

There is that word any again, not just using zero qualifiers but trying to extend the statement to any YouTube video. You've found a balance of effort you're happy with and thats great but tone down the condescending attitude. You're not getting it right in camera, you're getting it good enough to "print".

Even without any interest in the content people will watch beautiful or interesting imagery, there is some of that going on with links into your content from people mesmerised by the details of teardowns, no interest in the electronics at all just the imagery.

[EEVblog]

An array of microphones to pick from, carefully positioned, pop filters etc, you do put effort into your audio and there are obvious edits and adjustments in the final video

An array of mics, yeah, but I mostly use the internal mics on the cameras.
Carefully positioned, hardly.
Pop filters, only on the podcasting mic, which of course anyone should do, basic stuff.
Edits? I tweak the manual level of maybe a few clips in every video, tops. That's out of 50-70 clips average, so down in the few percent range. No auto leveling, no filtering, I just eye ball the levels and go "meh, near enough".
I get complaints about the audio on my Mailbag because I use the internal shotgun mic on the camera and it's a bit "distant". I have a super high quality expensive wireless mic, but I don't use it for that, why? Because I couldn't be bothered, I think the shotgun audio is good enough.

that aren't matched with similar detail in the imagery

I continually get people asking me how I get such good shots.
Are they perfect, no, but good enough to get me praise for them and virtually no complaints.

we haven't seen you using bounces or fills in lighting for example. If you were going to get it right first time in camera you would probably spend a lot of time adjusting a lighting rig.

I explained that lighting rigs are a huge PITA, I can't image a bigger PITA during filming. I get by just fine without them.

I'm claiming no such thing. I'm just pointing out what I do as a professional Youtuber, and what I suggest other Youtubers who produce similar content should do as well. Just like the OP is asking.
If you produce different type of content to me and you have different goals, then YMMV.

Lets see,

But if you have to do any colour, whitebalance, or other processing for youtube videos, then I must say that you are doing it wrong. You shouldn't have to do these things, get it all right in-camera.

There is that word any again, not just using zero qualifiers but trying to extend the statement to any YouTube video.
[/quote]

Jeeze, get over it. I qualified it several time you have snipped out, and you know I was talking about people producing similar material to me on Youtube.

You've found a balance of effort you're happy with and thats great but tone down the condescending attitude. You're not getting it right in camera, you're getting it good enough to "print".

Of course that's what I meant 

I'm done on this, I'm not playing pedantics with you any more.

[VEGETA]

Vegeta, Youtube will recompress the video anyway, so it makes no sense to use such slow options, you can just use a faster preset coupled with a higher crf

I know, but if I am making an encode, I make it as best as I can afford. TBH, these settings I use them mainly for encoding anime BDs in my server... I put them here as an example and clearly said that.

--bframes 10 forces clip to be encoded with level high 4.0 or 4.1 (depending on resolution) which makes the video potentially unplayable on some devices ( preset veryslow configures bframes to 8 which is more than enough for youtube videos, and you're overriding it to 10)

I think 10 is good and playable by all current devices, even my very old bad laptop. Again, this is something you can modify to suite you... I might choose 8 for faster encodes but there is nothing wrong with 10 and surely I don't go below 8.

setting a preset doesn't mean I want everything from it... it is just a starting point then I modify what I want.

--ref 16 is redundant, because it's set by --preset veryslow.  If you use better crf and faster preset (like --preset slow with --ref 5) you'd be fine and get pretty much same quality after youtube recompresses everything . Raising the --ref to 8 is basically the major difference between this profile and  --profile "slower".

As I said before, I put the slowest accepted preset then put my settings... If one of them is in the preset, it won't be a problem. for youtube to recompress everything, first you should get a very good quality HD video... then when youtube does its thing on it, it will be better than if you posted a medium quality video in the first place. Ref values helps in that alot.

I can use your suggestion by making it high crf and mid-high ref to make it a very good quality video (IF I CAN AFFORD IT). If not, I stick to ref 8 or something similar.

putting crf 16 is enough for real world videos (as it is perfect for anime). I bet if you choose crf 20 it will also be great but I can afford uploading a big sized video so I go with best settings.

--deblock -1:-1 is potentially a bad choice because it lets x264 slightly blur the image which may be bad if your Youtube videos have lots of screen capture... not to mention that you get the same result just using --tune film  ...

this is weird, are you sure? what I know is + deblock blurs the details while - one doesn't. I don't use tunes at all and will never do.

psy-rd 0.75 changed from the default of 1 is again something that makes sense for cinema movies... you're basically telling x264 that it's ok to blur some more and drop grain if it improves compression in frames that are really bitrate starved, which again may not be smart for youtube videos (and pointless because most youtube content is static scenes with Dave talking or showing up something in the case of EEVblog).

I think you are right here, but the value 1 will produce big size video for no gained quality I guess... psych enhancements are not required for youtube videos like you just explained but I can not put it to 0, so yeah, I will make it 1.

-aq-mode 3 is overkill for youtube videos, the default 1 is good enough... youtube videos don't have transitions to dark scenes and the other way around, so that you'd want x264 to take bits from some frames and allocate more to fade outs or darker scenes and so on.. that's what basically aq-mode 2 or 3 does, tweaks the adaptive .

this is your own personal choice to have aq-mode 1, but it is better to have it 2 here. 1 will give you huge file size for nothing gained but 2 will give smaller size with nearly the same quality as there are no motion no grain in it. 1 is not adaptive and only recommended to be used in a high grain video where mode 2 will fail (at least for me).

it is not a crime nor wrong to use it too.

about mode 3, it is the same mode 2 with better handling of dark scenes, so using it all the time instead of 2 is safe (i think) because if it is not a dark scene, it works like 2 or nearly the same. I also said you can replace it by 2 with no problems as I will do.

I didn't check the official build of x264 for a long time, so I will check about this mode 3. last time I checked doom9 about it the posts said that mode 3 and 4 are experimental and won't go in the official build until they aren't.

this is not the only reason I use tmod, it is the mixaq oreaq thing with other patches.

 

--direct auto is redundant, it's included in all presets higher than medium

xD

For youtube, you can just use --crf as low as you can afford , basically it's up to you and your internet connection, how much you're willing to upload for each minute of encoded video.

for me it is not like that, I do want a perfect result as much as I can... at least for archiving purposes.

BTW, did you try 10-bit videos for youtube? I know it is bad but I want to try it xD. It will be a very long journey until we can use HEVC in youtube, even I still have a bad laptop that can not handle it properly.

and PS. avoid Lagarith, since it's Windows only and relies on floating point implementation which is x86 dependant, some future processors may be different enough that you won't be able to decode your files or your encoded files won't be playable by people using older computers or ARM based systems. MagicYUV (and other lossless codecs like UTVideo or Huffyuv) are much much faster than Lagarith.

good info, thanks!

for me I will use x264 lossless, it is the best for me. I will just install x264 vfw for it with updated tmod version manually put in it.

No, --crf is short for constant rate factor, which is pretty much "constant quality", in a super simplified way it's constant quantizer on steroids... constant quantizer is sort of like treating each frame of the video as a jpeg picture and trying to compress each frame at a fixed quality preset (like jpeg 85%), ignoring how much space that frame will take. Constant Rate Factor also does some psycho-visual stuff, it analyzes each frame and sequences of frames to see where there's darker areas in the frames or stuff that your eyes won't notice easily and moves the saved bits towards the other.. and plays with the underlying qp varying in small steps the quantizer when needed to preserve quality

the term "constant quality" is not achievable through crf only. it is constat rate factor as you pointed out and to get the exact same quality for all frames you have to use "qcomp 1" which is bad. so quality is not gonna be "constant" at all. As you said, the encoder will analyze the video frames and pick the right bits to each frame while not caring about final size.. unlike "bitrate" mode.

So if you call it constant quality, it can be accepted as a simplified term which these guis made it an official name, and people think it is gonna be that simple as their guis tell them.

Actually, the term constant quality applies to using "QP" option rather than "CRF"!

If you choose qp 20 it will always be 20 for all frames but crf 20 is not! it will vary depending on your settings!

_______________

mariush

you must be one of doom9 members like me xD. I just know that.

We had a good discussion, but let's not spam the post with x264 battles xD.