MPEG2 Profile and Level for Final Conversion for VHS

[skatingrocker17]

After capturing lossless from VHS and performing some clean-up in VirtualDub, I use XMedia Recode to compress the file to MPEG 2.

I'm wondering what if the profile/level will have an affect on quality. The lossless file is saved at 4:2:0 and the resolution is 720x480. This fits the MP@ML profile/level. My concern is that the maximum bitrate is 15 Mbps. If I compress my video using the 422P@ML profile, the bitrate is slightly higher but the sampling is converted to 4:2:2. 422P@ML has a maximum bitrate of 50 Mbps.

My question is, am I gaining more than I'm losing if I decide to use 422P@ML over MP@ML? I'm open to other suggestions as well, but this workflow has worked for me so far.

[poisondeathray]

Why are you using MPEG2 in the first place? What is the final format goal , or why are you doing this?

MPEG2 422 will not be compatible for things like DVD (that would be one of the common reasons for using MPEG2)

[DB83]

Perhaps a more pertinent question to start with is why you are compressing to mpeg2. There are much more efficient codecs available.

[skatingrocker17]

To answer both questions, I was reading through the forums and saw that there were other members that were using MPEG 2 as the final format and that it had advantages over H.264 / MPEG-4 AVC in regards to saving interlaced video.

I don't have plans to export the files to DVD or BluRay. They will be stored on my media server and uploaded to Google Photos to share with family. I'm not concerned with Google Photos portion, I'd just like to save the best quality version of each file without each file being 50GB+ (I'd estimate I have 100 home movies).

I've only made my way through about 8 VHS tapes and I still have all of the lossless captures of each, so I don't have a problem re-encoding them all to another format. What do you suggest?

[DB83]

I kinda hoped that others with far greater knowledge about these things would have come to the party. Maybe after this they will come from behind the sofa

I can not fault your desire to have the best possible quality. But that really means you leave the video at lossless and your make a fair point about file size. And there is always a trade off between file size and quality. Just bear in mind that encoding at 15 mbps is going to result in files of 8 gb for one hour (hope my maths are right) and I seriously doubt your family will thank you should they wish to download these.

There is no reason why H.264 / Mpeg-4 AVC can not be encoded as interlaced and, yes, an interlaced source should, whenever possible, remain interlace. There are 2 issues:

1. Many encoders can not encode as interlaced without tweaking and you end up with badly prepared progressive video.
2. Even when interlaced many users do not realise and think the video they see is damaged whereas you can simply set your player to de-interlace (and you usually have a choice of de-interlacers). In fact should you distribute these mepg2 files as interlaced this problem could still happen.

I would suggest a double handed approach. By all means do the HQ mpeg2 for your own purpose but also a high quality progressive mpeg-4 for distribution. Plenty of topics on here of how to go about that.

Also bear in mind that you only keep your video at 720*480 if the video has a DAR flag set at 4:3 (or 16:9 whichever the case may be). In other cases you must resize to 640*480 for 4:3 or 852 * 480 for 16:9.

[skatingrocker17]

I kinda hoped that others with far greater knowledge about these things would have come to the party. Maybe after this they will come from behind the sofa

I can not fault your desire to have the best possible quality. But that really means you leave the video at lossless and your make a fair point about file size. And there is always a trade off between file size and quality. Just bear in mind that encoding at 15 mbps is going to result in files of 8 gb for one hour (hope my maths are right) and I seriously doubt your family will thank you should they wish to download these.

There is no reason why H.264 / Mpeg-4 AVC can not be encoded as interlaced and, yes, an interlaced source should, whenever possible, remain interlace. There are 2 issues:

1. Many encoders can not encode as interlaced without tweaking and you end up with badly prepared progressive video.
2. Even when interlaced many users do not realise and think the video they see is damaged whereas you can simply set your player to de-interlace (and you usually have a choice of de-interlacers). In fact should you distribute these mepg2 files as interlaced this problem could still happen.

I would suggest a double handed approach. By all means do the HQ mpeg2 for your own purpose but also a high quality progressive mpeg-4 for distribution. Plenty of topics on here of how to go about that.

Also bear in mind that you only keep your video at 720*480 if the video has a DAR flag set at 4:3 (or 16:9 whichever the case may be). In other cases you must resize to 640*480 for 4:3 or 852 * 480 for 16:9.

Thanks DB83, I appreciate the info. My initial thought was that I was going to encode the lossless file as a high quality MPEG2 file, which would become my master archival copy. It plays pack well with Media Player Classic, if the interlacing is too distracting it's easy enough to turn on deterinlacing in VLC.

I know that some people deinterlace and produce another file for sharing and video playback, but that's one more step and I don't see the need to do so, at least not at this time. I can also come back later and do that. I believe when the file is uploaded to Google Photos that it will be deinterlaced and, at best, the quality is so-so when played back on there anyways. But, it's the easiest way to share with the family and they're just happy enough that they can watch the videos on their iPads. They had no complaints about the tapes I captured 6 years ago with no TBC and a cheap Toshiba VCR but I plan on re-capturing those as well while I'm at it.


Here's the info from one of my final conversions as an example:
Video
ID : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@High
Format settings : CustomMatrix
Format settings, BVOP : No
Format settings, Matrix : Custom
Format settings, GOP : N=18
Format settings, picture st : Frame
Duration : 2 h 7 min
Bit rate mode : Variable
Bit rate : 8 371 kb/s
Maximum bit rate : 20.0 Mb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 4:3
Frame rate : 29.970 (30000/1001) FPS
Standard : NTSC
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Interlaced
Scan order : Top Field First
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.808
Time code of first frame : 00:00:00:00
Time code source : Group of pictures header
GOP, Open/Closed : Open
GOP, Open/Closed of first f : Closed
Stream size : 7.48 GiB (94%)

Audio
ID : 192 (0xC0)
Format : MPEG Audio
Format version : Version 1
Format profile : Layer 3
Format settings : Joint stereo / MS Stereo
Duration : 2 h 7 min
Bit rate mode : Constant
Bit rate : 320 kb/s
Channel(s) : 2 channels
Sampling rate : 48.0 kHz
Frame rate : 41.667 FPS (1152 SPF)
Compression mode : Lossy
Stream size : 293 MiB (4%)
Writing library : LAME3.99.€

[LMotlow]

Now that we've seen some of the myth, superstition, and voodoo about interlaced vs non-interlaced, and likewise the same about MPEG vs h.264, if you want to follow the violent debate on these issues you're welcome to it (bring food and other supplies to last for several decades). But there is no debate about the following:

MPEG is still used for both BluRay and DVD, in standard definition and high definition, most of which issues are either interlaced or telecined (a few progressive HD formats excepted). Interlaced MPEG is still a world wide broadcast standard. The internet is mostly low quality progressive streaming, for which h.264 is favored usually as very low bitrate junk video or at somewhat higher bitrates for paid internet progressive streaming services over your TV (with wide variations in quality).

So, if you have lossless interlaced captures you are a step ahead of the more narrowly focused users who capture to 640x480 h.264 and have to take a really huge quality hit when they want to ship a DVD to grandma or do some cleanup of all that ugly nasty crummy annoying VHS noise and illegal signal levels. You are all set for DVD and standard definition BluRay/AVCHD if so desired. All you have to do, other than any losssless intermediate cleanup you might want, is encode to those formats. If the internet is your goal, it's no problem deinterlacing your captures and resizing/encoding them for 4:3 progressive internet mounting. You can also resize them to square-pixel format and encode to mp4 or something so that you can view them on your not-so-smart TV at lower quality than you would get with DVD or SD BluRay and a decent set top player.

As for which is "better", MPEG2 or h.264, the debate goes on and on. You can use either, although MPEG is a must for universally playable DVD. I don't know if you've checked any stores lately, but DVD still outsells BluRay and many classic DVD issues will never see a BluRay release.

[DB83]

if you want to archive then you keep the lossless original. Not a good idea to keep a mpeg2, even if it is high quality.

Never heard of Google Photos before but I guess it is related to Youtube and they DO re-encode which can make a mockery of your carefully crafted encode.

[JVRaines]

The lossless file is saved at 4:2:0 and the resolution is 720x480. This fits the MP@ML profile/level. My concern is that the maximum bitrate is 15 Mbps.

That is quite enough for nearly all SD video and more than enough for VHS. In the early days of MPEG-2, broadcasters figured that 6 Mbps was a good compromise between quality and economy.