I need some help converting my old VHS, VHS-C, and 8mm/Hi8 tapes. Many Q's!

[jagabo]

I'm doing so right now for the last 15min and I can't get anything to work, colors simply aren't matching.

They're not supposed to match what you see on TV. The TV picture is a pumped up caricature of what's on the tape.

It also doesn't help that whenever I enter this proc amp settings my video capture that's playing and that I would rely on to see a live preview of what I'm changing, disappears.

You can probably get around this by using GraphStudio to adjust the proc amp while viewing the preview in VirtualDub. In GraphStudio, add the capture, crossbar, or whatever filter that include the proc amp. At the main window right click on the filter and select Properties. You will see the proc amp controls in one of the tabs. Adjust the controls there while previewing in VirtualDub.

And once again it looks like there's frames dropped or something, it's not as smooth as on the camcorder and TV at all. Edges of even faces are jaggy and noisy, and now I'm noticing lines breaking in vertically occasionally, like a mirrored sword slash for half a second.

The AVI sample you uploaded didn't have any of those problems. In general, capture devices aren't as tolerant as TVs of the type of signal problems common to video tape formats. You may need a line TBC and a full frame TBC to get around those problems.

You should also disable sharpening filters and auto contrast etc. filters in the capture device and the VHS deck. All those will do is screw up the video before your capture it.

Not even using a VHS deck yet

My point was that you should disable whatever you can in what ever player you're using.

Keep in mind that your computer monitor is not a TV. It normally receives a perfect Desktop image from the graphics card and it does no processing on it. Any processing it did would ruin the deskop image. Your TV does lots of processing because analog video is not not perfect and the manufactures know that people prefer too much contrast, too much saturation, and an over sharped image.

What you want to do is get a good capture, not an over processed cap. You then filter to get the picture the way you want it. And check by watching the resulting video on a TV.

Another thing you can do, though I don't really recommend it, is adjust your graphics card's video proc amp so that video in a media player (or VirtualDub, depending on VirtualDub's settings) looks more like the picture on your TV.

A bigger issue is how much work you are willing to put into your videos. Making VHS look half decent requires a lot of filtering. For the best results you'll need to learn how to use AviSynth and many of its advanced filters. If want a quick, easy, moderate quality result just get an old DVD recorder with a line TBC, noise reduction, sharpening, etc. and record 1 hour per DVD. The resulting DVD played via a composite connection will look very similar to what you're seeing playing the tapes directly to the TV. But if you open one of the DVD's VOB files in VirtualDub you'll see a picture that looks a lot like what you're capturing now (or once you get the proc amp adjusted properly).

[CZbwoi]

They're not supposed to match what you see on TV. The TV picture is a pumped up caricature of what's on the tape.

How is it a pumped up 'caricature' when it looks great? It doesn't look amazing DVD quality, but it looks like a great piece of tape. On every TV I've played it through.

You can probably get around this by using GraphStudio to adjust the proc amp while viewing the preview in VirtualDub. In GraphStudio, add the capture, crossbar, or whatever filter that include the proc amp. At the main window right click on the filter and select Properties. You will see the proc amp controls in one of the tabs. Adjust the controls there while previewing in VirtualDub.

I'll try that out later tonight, never heard about it like that though, that individual settings can be stored inside of the capture device. I thought it was individual settings stored in each program's interpretation of the capture device.

The AVI sample you uploaded didn't have any of those problems. In general, capture devices aren't as tolerant as TVs of the type of signal problems common to video tape formats. You may need a line TBC and a full frame TBC to get around those problems.

The camcorders have built in TBC's in them if that means anything and they play fine everywhere but through this Elgato.

Your TV does lots of processing because analog video is not not perfect and the manufactures know that people prefer too much contrast, too much saturation, and an over sharped image.

As I've said, I know this and turned off all those feautures. Still looked way better and didn't play choppy or with seemingly dropped frames, it is significantly better than anything Elgato gives me, color and motion wise.

What you want to do is get a good capture, not an over processed cap.

Yup, but I know now that it won't be with this. So what's the best device for people to capture composite/s-video? USB, device, it doesn't matter. I don't care about the price anymore, anywhere up to $200 sounds like it should be reasonable.

A bigger issue is how much work you are willing to put into your videos. Making VHS look half decent requires a lot of filtering. For the best results you'll need to learn how to use AviSynth and many of its advanced filters. If want a quick, easy, moderate quality result just get an old DVD recorder with a line TBC, noise reduction, sharpening, etc. and record 1 hour per DVD.

That's what my uncle did, and I might do that...but I first want to try out capturing the VHS and see how it goes. But that's the last step like I said, right now my focus is on capturing Hi8/8mm and VHS-C tapes from my camcorders, so what I need right now is a better capture tool and the best preferably (just not the thousand dollar Hollywood stuff). That's what I need help with, I don't want this Elgato anymore...

[jagabo]

You keep bringing up the fact that you're not working with VHS now but the issues are the same for VHS, 8mm. Hi8, Beta, etc. Crappy time base, lots of noise, low resolution, wonky automatic gain controls, etc. It's only when you move into digital formats (D8, DV) where many of those problems go away.

The capture device's video proc amp is a part of the capture driver, not the software doing the capturing.

Beware of the difference between a line TBC and a full frame TBC. They do different things.

I'm not familiar with the device you're using now but I suspect you won't get much better results with any of the other common video capture devices. ElGato is primarily a Mac company and they really don't know what they're doign in Windows. These forums are full of threads asking what to use and threads comparing different devices. I don't do analog capture any more but Hauppauge (HVR 1255, USB Live2, etc.) and AverMedia are usually mentioned. EZCap's device is decent too -- but beware of the fakes. A $1500 BrightEye won't get you any better results. The HD capture devices from Hauppauge, Avermedia, and others can capture SD but they don't do any better (indeed, often worse) than the dedicated SD capture devices. The problem with SD video capture now is that few people are doing it and the influx of crappy $10 Chinese devices has killed most of the decent manufactures. And the remaining decent manufactures are cutting corners to stay alive.

[dellsam34]

Here is my advice to you, Go to your local electronic store and pick up a decent capture card or USB device, Brick and mortar stores usually have a return policy that won't cost you money in return shipping fees, try it and see for yourself.

[CZbwoi]

I don't do analog capture any more but Hauppauge (HVR 1255, USB Live2, etc.) and AverMedia are usually mentioned. EZCap's device is decent too -- but beware of the fakes.

Do you know if all these capture in YUY2? I know that the Elgato I have doesn't and people tell me to find one that does (even though it does UYVY).
And is there actually like a ranking list of the best USB ones somewhere?

Here is my advice to you, Go to your local electronic store and pick up a decent capture card or USB device, Brick and mortar stores usually have a return policy that won't cost you money in return shipping fees, try it and see for yourself.

I've tried Best Buy, Radio Shack, the department stores like Target, no one has one for me to just buy and try out and compare to this Elgato.

I don't mind ordering online again, I just want to make sure it's the one that people over here or that have daily experience say is the best one possible.

[dellsam34]

I personally use Pinnacle 500-USB device, It is a 8 bit 4:2:2 native PAL/NTSC so do most of the non professional cards and USB devices and that's the best for consumer analog video capturing, anything above that is just overkill, You can get it for cheap due to later versions of windows driver unavailability, However there is a trick that most people don't know, pinnacle software is built in a driver just install the latest Pinnacle Studio trial to force the driver to Windows, You don't have to use the software, Works very well with my computer in Win 7 with virtualDub, I had it since windows XP days and I didn't see the need to replace it with another one due to the fact that analog video is an old technology and its standards has not been changed since, So there is really no advancement in analog capture devices technology.

http://www.ebay.com/itm/Pinnacle-Studio-500-USB-Capture-Editing-Cartridge-with-USB-AV-...sAAOSw8RJXCj36

Oh by the way it is not made in China, It's made in Germany.
Here is a big file from the other forum shows a TV broadcast capture with the 500-USB:
http://www.mediafire.com/download/j7caxovs7f7alea/500-USB+CAP.zip
If the file is big for you I will try to post my own when I get time.

[Brad]

Do you know if all these capture in YUY2? I know that the Elgato I have doesn't and people tell me to find one that does (even though it does UYVY).

People on forums use YUY2 as shorthand for 4:2:2 8-bit digital video. The data held in each form of storage is 100% mathematically identical, only the structure differs (packed YUY2, UYVY, YVYU; planar YV16). I thought you were clear on this point already.

No one here has told you to ditch a device because it uses UYVY instead of YUY2. Anyone that tells you that doesn't know what they are talking about.

There are two different versions of the Elgato Video Capture hardware. The first, from 2008, uses eMPIA hardware. The second, from 2010, uses Conexant Polaris (same chip as the Hauppauge USB-Live2 and many others).

Go to Device Manager -> Sound, video and game controllers. Locate your Elgato and double-click on it. Choose the Details tab. Change from Device description to Hardware Ids. If it says VID_0FD9&PID_0037&MI_01, keep the Elgato. (The older one is 33 rather than 37.)

[jagabo]

Do you know if all these capture in YUY2? I know that the Elgato I have doesn't and people tell me to find one that does (even though it does UYVY).

They have YUY2 or some other YUV 4:2:2 format like UYVY. They're basically the same thing just in a different order. The point is that you want to avoid devices with hardware that encodes to MPEG 2 or h.264. That encoding loses a lot of detail and creates artifacts. Such devices usually can't deliver the raw uncompressed YUV video.

And is there actually like a ranking list of the best USB ones somewhere?

Vaporeon800 has been posting samles here for the last few years. He doesn't rank them, you'll have to decide that for yourself.

https://forum.videohelp.com/threads/360704-2013-my-video-capture-device-comparison-screenshots
https://forum.videohelp.com/threads/329016-2001-2010-my-capture-cards-comparison-screenshots
https://forum.videohelp.com/threads/376526-VHS-waviness-in-digital-conversion-compariso...hots?p=2438043
https://forum.videohelp.com/threads/376837-DV-conversion-vs-lossless-capture-comparison-%28Hi8%29-[WARNING-auto-load!]?p=2432635#post2432635

But keep in mind that those comparisons may not mean much for VHS:

Small differences in sharpness make no difference for VHS/8mm/Hi8. The resolution of those tape formats is far below the resolution the device can capture.

Small difference in noise don't matter. All the consumer capture devices save 8 bit samples, though some capture higher bit depths internally. But consumer tape has only 5 or 6 bits of signal, the rest is noise. The ability to capture 10 or 12 bit samples will only get you more "accurate" noise.

He doesn't often discuss the proc amp controls. Some device have none. Some devices may crush darks and brights before proc amp controls are applied. When that happens no amount of adjustment will restore the details in those areas.

He doesn't often mention stability. The drivers for many devices lock up or stop capturing when the time base isn't good.

[Brad]

Proc Amp issues are noted (somewhat) in the Google Sheet linked in the 2013 thread.

I don't recall experiencing lockups from a capture driver over the past several years. Not that my memory is any good. And I never captured a full tape with most of the devices I own...

[CZbwoi]

Here is a big file from the other forum shows a TV broadcast capture with the 500-USB:
http://www.mediafire.com/download/j7caxovs7f7alea/500-USB+CAP.zip
If the file is big for you I will try to post my own when I get time.

I saw it, thanks for that.

There are two different versions of the Elgato Video Capture hardware. The first, from 2008, uses eMPIA hardware. The second, from 2010, uses Conexant Polaris (same chip as the Hauppauge USB-Live2 and many others).

Go to Device Manager -> Sound, video and game controllers. Locate your Elgato and double-click on it. Choose the Details tab. Change from Device description to Hardware Ids. If it says VID_0FD9&PID_0037&MI_01, keep the Elgato. (The older one is 33 rather than 37.)

USB\VID_0FD9&PID_0037&REV_4002&MI_01
USB\VID_0FD9&PID_0037&MI_01

That is what is inside of my Hardware Ids.

Are you telling me that essentially the exact same thing will be inside USB-Live2 and that they perform and deliver the exact same way, only the casing is different?

A user here has told me otherwise constantly and the device is not recommended for a reason: http://www.digitalfaq.com/forum/video-restore/7278-questions-converting-vhs.html#post43437

No surprise. That device is not recommended for doing what you want.

Capture devices like the ATi and Hauppauge are designed for analog interlaced capture and can capture to lossles media. Elgato doesn't have those characteristics. No one here would recommend Elgato lossy encoding for analog, and especially not for noisy interlaced sources.

"Now I'm no expert in this, but the same device did way better in vdub than in Elgato's own thing, you can notice it immensly if you look at the text comparison in the first 2 files. But you know more about what to look for than me. So seeing these files and what can be done in vdub would you say this is great and to stick with it..?"

I wouldn't say that. The captures don't look good. I'll demonstrate below.

"For what its worth my process right now would be JVC Camcorder and Sony Hi8 Camcorder > S-video and 1 audio jack > Elgato USB capture device > VirtualDub with Huffyuv compressor > Premiere Pro export to .mp4 (with target bitrate of 4 from the default 6, and either 1 pass or 2 pass VBR)."

Again, no one here would recommend Elgato. It's causing too many problems and unnecessary work.

I spent a few hours working with these nightmare captures -- something I'd normally avoid, but you keep asking. Some of the problems are coming from your capture gear, others are due to inappropriate levels during capture. The image below, resized to 4:3 to save a little forum space, is an otherwise untouched frame 42 from test 1b.avi.

Replace the Elgato with one of the recommended devices. Make short captures to get the hang of things.

USB didn't cause your problems. The Elgato you plugged into it is the culprit, not USB.

----------
As you can see I'm being pulled in several different directions, some are saying the Elgato is bad and its not on these recommended lists for a reason, others like you are saying it has the same chip as *insert here* and to stick with it...I appreciate it all but I don't know what to believe.

He doesn't often discuss the proc amp controls. Some device have none. Some devices may crush darks and brights before proc amp controls are applied. When that happens no amount of adjustment will restore the details in those areas.

Is there a recommended proc amp control guide for the Elgato or other devices? Like the recommended settings?

He doesn't often mention stability. The drivers for many devices lock up or stop capturing when the time base isn't good.

Oy vey, that seems like it would be a problem...does this happen on any of the popular ones mentioned?

I don't recall experiencing lockups from a capture driver over the past several years. Not that my memory is any good. And I never captured a full tape with most of the devices I own...

:/

[Brad]

Are you telling me that essentially the exact same thing will be inside USB-Live2 and that they perform and deliver the exact same way, only the casing is different?

Pretty much. With the same hardware, there can still be differences due to actual construction, it seems.

A user here has told me otherwise constantly and the device is not recommended for a reason: http://www.digitalfaq.com/forum/video-restore/7278-questions-converting-vhs.html#post43437

Well if you don't believe me, you can always open it up and look at the chip itself.

Is there a recommended proc amp control guide for the Elgato or other devices? Like the recommended settings?

The settings are entirely dependent on the tape you're playing.

[CZbwoi]

Pretty much. With the same hardware, there can still be differences due to actual construction, it seems.

Hm. Have you ever yourself used the newer Elgato model like what I have and tested/compared yourself? Because so far I haven't seen it in any of your tests you've ever done despite it being known and popular, any reason for this?

The settings are entirely dependent on the tape you're playing.

That's unfortunate, oh well.

And since you have so much experience and devices, what would you say was your best personal USB experience after all this time? Quality, least amount of errors, fluidity of the capture, etc. Just your best overall favorite at the moment, what would your choice be?

[manono]

Is there a recommended proc amp control guide for the Elgato or other devices? Like the recommended settings?

As vaperon800 stated, it depends on the tape. I do a few test captures of 30 seconds or so from different parts of the tape and check the levels (I do almost entirely black and white caps) in AviSynth using a histogram and/or a specific filter line in my script:

ColorYUV(Analyze=True).Limiter(Show="Luma")

A few adjustments and I'm good to go.

It's quite easy to get it the way you want. It won't be uniform, of course, and I refine it - scene-by-scene, if necessary - later on. The important thing is not to begin with crushed blacks and blown out whites in your first capture or you'll never be able to properly fix it.

[jagabo]

When capturing with VirtualDub disable audio playback. You can still capture audio, just don't leave it playing while you capture. That causes dropped frame problems with many devices.

[Brad]

Have you ever yourself used the newer Elgato model like what I have and tested/compared yourself? Because so far I haven't seen it in any of your tests you've ever done despite it being known and popular, any reason for this?

1. No.
2. I'm taken aback by the question. "Why wouldn't you buy something that's twice the price of its equivalent competitors, so that you can prove to the Internet that your deductions are accurate?"

And since you have so much experience and devices, what would you say was your best personal USB experience after all this time? Quality, least amount of errors, fluidity of the capture, etc. Just your best overall favorite at the moment, what would your choice be?

I use the Diamond VC500 when I do non-HDMI captures. Because adjusting its proc amp in Win7 is annoying, I would buy the Hauppauge USB-Live2 or one of the other devices on that Linux Wiki list instead, if my VC500 suddenly fell into a black hole. All of these utilize Conexant Polaris.

To more easily access the VC500 proc amp, I tried force-installing the USB-Live2 drivers, but that only gave me a black-and-white signal and I haven't bothered going back to see if further driver-hacking could get color while using the "wrong" drivers.

[CZbwoi]

When capturing with VirtualDub disable audio playback. You can still capture audio, just don't leave it playing while you capture. That causes dropped frame problems with many devices.

Audio > Enable audio playback? It was never even selected since I first started using it, I just figured that this program didn't play audio from the capture and just recorded it. Enable audio capture is checked though. It's still not as smooth as I'd like it to be (the Hi8 captures), like it doesn't move.

As vaperon800 stated, it depends on the tape. I do a few test captures of 30 seconds or so from different parts of the tape and check the levels (I do almost entirely black and white caps) in AviSynth using a histogram and/or a specific filter line in my script:

ColorYUV(Analyze=True).Limiter(Show="Luma")

A few adjustments and I'm good to go.

It's quite easy to get it the way you want. It won't be uniform, of course, and I refine it - scene-by-scene, if necessary - later on. The important thing is not to begin with crushed blacks and blown out whites in your first capture or you'll never be able to properly fix it.

I shall learn how to do this.

No. I'm taken aback by the question. "Why wouldn't you buy something that's twice the price of its equivalent competitors, so that you can prove to the Internet that your deductions are accurate?"

I don't know, maybe it IS different and you will see if you get it, nothing is a sure thing until you try it out yourself. I'm just surprised you never did considering you have everything else lol (and you could return it after testing).

I use the Diamond VC500 when I do non-HDMI captures. Because adjusting its proc amp in Win7 is annoying, I would buy the Hauppauge USB-Live2 or one of the other devices on that Linux Wiki list instead, if my VC500 suddenly fell into a black hole. All of these utilize Conexant Polaris.

I have Windows 7 as well so I'm guessing this means to stay clear of the VC500 if I ever want to easily go into proc camp and to go with the USB-Live2. Unless you really prefer the Diamond over it besides the proc amp settings bit, its quality.

[jagabo]

Audio > Enable audio playback? It was never even selected... It's still not as smooth as I'd like it to be (the Hi8 captures), like it doesn't move

Are you talking about the preview window or the actual captured file? Like I said earlier, the sample captures you uploaded didn't show any duplicate or missing frames.

Do you realize that players will not automatically deinterlace huffyuv AVI files during playback? You have to force them. Without deinterlacing you will see comb artifacts and less smooth motion. What you're describing sounds different than this though.

VirtualDub's capture module is very flexible but that flexibility can make hard to figure out. Check VirtualDub's A/V sync settings via Capture -> Timing. What works best depends on the particular device but try playing with the Resync Mode settings.

For judging brightness and contrast you should learn to use a waveform monitor. In AviSynth you can use Histogram() or VideoScope(). The default mode of Histogram() is a waveform monitor, not a histogram.

https://forum.videohelp.com/threads/340804-colorspace-conversation-elaboration?p=212156...=1#post2121568

VirtualDub's Capture Module's Histogram feature is helpful but not always easy to read.

You should also be aware a that not all YUV combinations are valid even if all three are within their expected limits (16<=Y<=235, 16<=UV<=240). Though this topic is more advanced and not something you need to worry about during the capture phase.

[CZbwoi]

The way many people do this , is either frameserve out (using debugmode frameserver or advanced frameserver) or a lossless intermediate (e.g. lagarith, ut video codec, magic yuv), then either A) resize to a square pixel format in something like avisynth /encode x264 or some GUI for it, or B) keep the same dimensions and set the AR to whatever you want either with encoder flags, or container flags.

Both, more so for the Hi8 camera. Matter of fact, I think I only see it on the Hi8 captures. Like I'm saying, it's just nowhere near as smooth as it is on TV and on the camcorder, and I'm not talking about all the filters that TVs embed onto them. I turned everything off, same for the camcorder (which didn't have any to begin with). And no, smaller screen =/= smoother playback, I shrink the crap out of my player screens and it still happens. Matter of fact when I shrink them everything becomes wavey, is this normal?

Do you realize that players will not automatically deinterlace huffyuv AVI files during playback? You have to force them. Without deinterlacing you will see comb artifacts and less smooth motion. What you're describing sounds different than this though.

I decombed once and then deinterlaced once with Handbrake and the decombing instance is the one that made it more smoother, albeit the quality got worse, so that's interesting.

VirtualDub's capture module is very flexible but that flexibility can make hard to figure out. Check VirtualDub's A/V sync settings via Capture -> Timing. What works best depends on the particular device but try playing with the Resync Mode settings.

For judging brightness and contrast you should learn to use a waveform monitor. In AviSynth you can use Histogram() or VideoScope(). The default mode of Histogram() is a waveform monitor, not a histogram.

https://forum.videohelp.com/threads/340804-colorspace-conversation-elaboration?p=212156...=1#post2121568

VirtualDub's Capture Module's Histogram feature is helpful but not always easy to read.

You should also be aware a that not all YUV combinations are valid even if all three are within their expected limits (16<=Y<=235, 16<=UV<=240). Though this topic is more advanced and not something you need to worry about during the capture phase.

Will do.

[jagabo]

it's just nowhere near as smooth as it is on TV and on the camcorder, and I'm not talking about all the filters that TVs embed onto them. I turned everything off

You only disabled the filters the TV let you. With analog inputs it's still doing some filtering over which you have no control.

when I shrink them everything becomes wavey, is this normal?

Interlaced video needs to be resized properly or you will get those wavy artifacts.

I decombed once and then deinterlaced once with Handbrake and the decombing instance is the one that made it more smoother, albeit the quality got worse

To retain the smoothness of interlaced video you need to double the frame rate as well as decomb/deinterlace.

[CZbwoi]

To retain the smoothness of interlaced video you need to double the frame rate as well as decomb/deinterlace.
Attached Files
File Type: mkv test2b.mkv

Dude, wow! That looks great, even better than the original! The only thing that's off-putting is that the white text in the right is now light blue, was it necessary to change the hue that much, what would it have looked like if the color didn't change but you did everything else?

Also, please, let me know everything you did to get that final clip. Every step by step, every filter added, what program(s) you used. Please, it looks great. Near perfect.

I was expecting you to do it to test1b as that's the one I was talking about that looks less smooth, if you could do your magic for that with those same steps so I could see what it could look like I would appreciate it!

And any reason for exporting it as .mkv over .mp4 or something else?

[jagabo]

I used AviSynth for all the video processing.

AviSource("test 2b.avi") 
AssumeTFF()
ColorYUV(gain_y=30, off_y=-55, gamma_y=100, cont_u=20, cont_v=50)

# some white balance for brights, darks are ok
brights=ColorYUV(off_u=28, off_v=-26)
Overlay(last, brights, mask=last.ColorYUV(cont_y=200, off_y=-30))

ColorYUV(cont_u=50, cont_v=50) # a little more saturation
ConvertToYV12(interlaced=true)
SeparateFields()
Dehalo_alpha(rx=3, ry=2) # remove oversharpening halos
Weave()
Spline36Resize(360,height) # downscale width for a little smoothing)
QTGMC(preset="slow") # double frame rate deinterlace
MergeChroma(last, CNR2("xoo")) # strong noise filter on chroma
TemporalDegrain(SAD1=400, SAD2=300, sigma=16) # moderate noise filter
nnedi3_rpow2(2, cshift="Spline36Resize", fwidth=720, fheight=height) # scale back up
Sharpen(0.3, 0.0) # sharpen horizontally
Crop(8,0,-8,-8) # remove black borders and most of the head switching noise

Color manipulation is usually better done in an NLE but I don't have one.

I encoded the script with x264 by dragging and dropping the script onto a batch file:

start /b /low x264.exe --preset=slow --crf=18 --sar=10:11 --output %1.mkv %1

I extracted the audio and encoded as MP3 with Audacity. And finally muxed the new video and audio with MkvToolnix.

MKV because that's my preferred container these days.

I adjusted the levels and colors for the main video. It's possible to "protect" the time stamps from the color changes but it's not worth the effort.

[jagabo]

Here's the first video (no audio) with just the double frame rate deinterlace so you can see the motion isn't jerky (except for the camerawork):

AviSource("test 1b.avi") 
AssumeTFF()
ConvertToYV12(interlaced=true)
QTGMC()

[CZbwoi]

Wow, that second one is great too, that is silky smooth! You are awesome. So I'm guessing everyone that deals with Hi8/8mm/VHS-C tapes always doubles the framerate? No one has mentioned that until you.

Why the no audio though, once you use avisynth it takes away the audio? What other reason could you have had for extracting the audio and muxing it back with the video, ergo would I have to do that too?

And if I copy that first code of text when I learn how to use avisynth, is the info like "remove black borders and most of the head switching noise" necessary, did you only add it for me to know what they do and does the # make the text irrelevant to avisynth when it processes it?

[jagabo]

So I'm guessing everyone that deals with Hi8/8mm/VHS-C tapes always doubles the framerate?

It depends. If you're staying with natively interlaced systems like DVD you can leave the video interlaced and get smooth playback. DVD players and TVs know how to handle interlaced video if you handle it properly. If you're only interested in computer or media player playback you can double speed deinterlace.

Why the no audio though, once you use avisynth it takes away the audio?

AviSynth can handle the audio but the x264 command line encoder only handles video.

would I have to do that too?

If you use the x264 cli encoder, yes. But some editors accept AviSynth scripts as input and will encode the audio too.

And if I copy that first code of text when I learn how to use avisynth, is the info like "remove black borders and most of the head switching noise" necessary, did you only add it for me to know what they do and does the # make the text irrelevant to avisynth when it processes it?

Everything on a line to the right of a # sign is ignored by AviSynth. I added those comments to give you an idea what was going on.

[CZbwoi]

It depends. If you're staying with natively interlaced systems like DVD you can leave the video interlaced and get smooth playback. DVD players and TVs know how to handle interlaced video if you handle it properly. If you're only interested in computer or media player playback you can double speed deinterlace.

Say that you end up also playing the files on the TV via connecting your computer to it with the HDMI and streaming them through Drive/the actual file or plugging a USB with them into the TV, will the fact that you double speed deinterlaced it and whatnot alter how its played on the TV or Blu-ray player since we're doing this essentially for playback on PC and mobile? Or will it just play exactly on TV like it does on the PC?

AviSynth can handle the audio but the x264 command line encoder only handles video.

This might sound stupid to you but what's the point of encoding the script with x264? Is that what makes it smaller (how did you make them so small?)

And if instead I wanted to have the final file be .mp4 instead of .mkv what would I have to do? Without losing any quality by re-encoding it with some other program of course. Basically everything this video says? I'm assuming you use AvsP too? (I'm new to all of this)

[jagabo]

Modern devices can handle 60 fps video up to 1280x720, at least. QTGMC() (what I used for deinterlacing) does a better job than any deinterlacer built into a TV or DVD/BD/media player. 30i (your source) and 60p can display the same motion smoothness. x264 is the compressor I used to reduce the size. You can mux the audio and video to MP4 instead of MKV by using, say, MP4Muxer instead of MkvToolnix. Or you can use any other encoder/editor you want as long as it supports AviSynth scripts. I usually keep the original audio but since your audio was only on the left channel I moved it to the center and compressed it to MP3.

[CZbwoi]

Modern devices can handle 60 fps video up to 1280x720, at least. QTGMC() (what I used for deinterlacing) does a better job than any deinterlacer built into a TV or DVD/BD/media player. 30i (your source) and 60p can display the same motion smoothness.

So it should run fine on TV too right even though I "already did the deinterlacing job the TV was supposed to do?" And why do I always hear people say online and through browsing these forum threads to not deinterlace because it ruins the film? I always hear people say that, that it kills the original or something. Unless I've been reading wrong.

x264 is the compressor I used to reduce the size. You can mux the audio and video to MP4 instead of MKV by using, say, MP4Muxer instead of MkvToolnix.

None of this should get rid of any quality when I use x264, it still retains everything? And nothing lost if it's to MP4 instead of MKV, since I'm ultimately still 'muxing' them, right?

I usually keep the original audio but since your audio was only on the left channel I moved it to the center and compressed it to MP3.

Ah yes, that makes sense considering they're mono camcorders. Premiere Pro would initially do that for me on it's own but after seeing what's possible with Avisynth scripts and all this I'm not going that route anymore.

[jagabo]

So it should run fine on TV too right

Yes.

even though I "already did the deinterlacing job the TV was supposed to do?"

QTGMC is better than any TV's deinterlacer today.

And why do I always hear people say online and through browsing these forum threads to not deinterlace because it ruins the film?

There is no perfect method of deinterlacing. If you deinterlace now the video you make will "lock in" the quality of the deinterlacing forever. So even though QTGMC() is better than your TV can do now, someday in the future TVs may be better than QTGMC. That's one reason people recommend leaving the video interlaced. But I don't think TVs will get better than QTGMC for a very long time. You can always save your original video caps if your concerned about that.

Also, some people refer to single speed deinterlacing when they tell you not to deinterlace. Ie, don't convert 30i to 30p. That loses half the temporal resolution, giving jerky flickery video. And several years ago, many players couldn't play 60p smoothly. Or they have been using the inferior deinterlacers found in most commercial products.

On the other hand, some filters only work on progressive frames. And even those that can work with interlaced frames can introduce artifacts related to the interlacing. And interlaced encoding with high compression codecs (MPEG 2, Divx/Xvid, h.264) can add artifacts and deliver lower quality than progressive encoding. So interlaced encoding is no panacea either.

None of this should get rid of any quality when I use x264, it still retains everything?

x264 is a "lossy" codec. It always loses some quality (except when using its lossless mode which gives very big files, like your huffy avi -- and few players can play lossless x264 video). But you get to decide how much quality is lost. If you use slow settings and a low CRF you won't lose much quality. If you use fast settings and a high CRF you will get a smaller size but the quality will be worse. I encoded at the slow preset with a CRF of 18. At that you will barely see a difference between the encoded video and the uncompressed source at normal playback speeds. If you step the CRF down to 12 you will barely see any difference between the original and the encoded video, even viewing an A/B comparison of enlarged still frames.

And nothing lost if it's to MP4 instead of MKV, since I'm ultimately still 'muxing' them, right?

Right. It's just the "box" you are putting your video in. The box has no effect on the quality of the contents.

[CZbwoi]

Okay, so bear with me here please.

I'm trying to insert some of your stuff from the first video clip you made and those codes into test 1b to try it out for myself but with the 2nd clip (where you barely did anything). I inserted the cropping line of yours and it cropped something, maybe too much I don't know, it was meant for the first clip you did. But every time I try to change one of the dimensions I get this error:



But now as it appeared that I finally got the hang of it, I'm not able to erase the final bit of black border lines. Every time that I try to go a number up or down in the way it should to further continue the last bit of cropping I get the error:





Does this mean it can be cropped in variables of 2 all the time? I.e. 2,4,6,8,10? Is that how all cropping is done in avisynth scripts, always by 2's?

Also, could you explain how to use the x264.2692.x86.exe file I downloaded for that next step? How do I insert my script and original .avi into it? What exactly do I do and what did you mean by batch file? Basically, how do I turn it into x264 with this?

*and are there other popular formats to turn it into when I have the completed avisynth script and other methods to do so when you have the avisynth script and original avi?

[manono]

Does this mean it can be cropped in variables of 2 all the time? I.e. 2,4,6,8,10? Is that how all cropping is done in avisynth scripts, always by 2's?

You should familiarize yourself with the crop rules. To answer your question more directly, yes, progressive video in YUV has to be cropped by multiples of 2. And it's not just AviSynth.