Bad VCR timing?

[bmichaelb]

You didn't see it in the camera pan on the crowd at the end of the clip? Maybe that has to do with GPU's display settings as well then.

-- merged --

OK... I recorded my screen with my camera for the first 10 minutes of a capture. It's recording a 29.970 capture at 30.091, but still shows what I'm seeing.

https://drive.google.com/file/d/1WtM...ew?usp=sharing

The opening has the damage I mentioned earlier, but I'm planning on splicing in the footage from the dvd trial. Or, I can just leave it in and say it's part of the nostalgia. It's only ~20 seconds worth. It has some vertical jitter, but that can be fixed with avisynth. But... the worst of the damage is that line scratched halfway up the frame. It pops in and out for the first 8 minutes, then fades to a blueish line for the rest of the tape. Not the whole tape though, as it fades in and out. I'm freaking heartbroken over that. It's not always in both fields at the same time though, so I may be able to fix some of it with a script by johnmeyer that fixes bad fields. But, it's still only 2 hours of footage out of 65 in total, so it's no 'too' bad I guess.

What's really strange though is the kink at the top of the frame... I'm guessing that's from the GoVideo deck that did the dubbing of my friend's sources. I never saw it on a CRT TV, only on an LCD while trying to edit it. Her tapes were SP, recorded on high end Sony decks, with Sony tapes. I used Fuji HQ160's for all my sources, and Fuji HQ120's for her tapes. I'm able to crop and shift that kink, but there's a colored line that appears when there's a colored background. Camcorder Color Denoise in VirtualDub removes almost all of it at full strength, but then it also reduces color everywhere else.

I'd like to save that top area as it's 13 pixels worth... top 5 pixels are black and can be cropped, so there's still 8 pixels worth of real estate. My sources are fine at top, but have 7 pixels worth of head switching noise at the bottom. To even everything up between the sources, I'd have to crop both the same, so that's 20 pixels worth in all. 720x480 just became 720x460. 680x460 total viewing area. Then looking at the dub of my friend's sources, the edges are spikey. If I want to clean that up, I'm looking closer to 670x460. It's still a decent size, I'd just like to squeeze as much as possible out of it.

[latreche34]

I couldn't see any stutter, which means it could be a dropped frame somewhere that I didn't catch, The top of the frame is normal it is not meant to be seen, it falls in the overscan area of television back in the day. I see a lot of people complain about captured files thinking they should look like modern videos, They don't, especially consumer tape formats, It is what it is. Check your capture program if it is dropping frames, less than 10 frames in a 2hr tape is normal.

I've also downloaded the file in post #1 and played it and it played perfectly, not sure what's the concern here.

-- merged --

Just put the clip into vdub2 and skipped it frame by frame of the last 40 seconds that you said problematic, not a single frame drop, you have a playback issue to address.

[bmichaelb]

Quote:

Originally Posted by latreche34

I couldn't see any stutter, which means it could be a dropped frame somewhere that I didn't catch, The top of the frame is normal it is not meant to be seen, it falls in the overscan area of television back in the day. I see a lot of people complain about captured files thinking they should look like modern videos, They don't, especially consumer tape formats, It is what it is. Check your capture program if it is dropping frames, less than 10 frames in a 2hr tape is normal.

I've also downloaded the file in post #1 and played it and it played perfectly, not sure what's the concern here.

There's no dropped or inserted frames reported in VirtualDub, and with 3 captures to use with Median, all 3 clips have the same amount of frames (after lining up in AviSynth), and display the same frame when choosing a frame near the end of the timeline.

That kink at the top is definitely not normal. It may be in the overscan, but every other tape I've ever captured, or seen uploaded to YouTube, that top area should be flat on the sides... no shifting. And when played through my JVC with the TBC turned on, its screws up even more, with the individual lines of the kink sticking out unevenly. And as I said, my sources, from the same ppv, don't have that kink. It's only the tapes that were dubbed.

As for the file I just uploaded... there's tons of jitter, starting from the drummer at 0:02:15... it's jerky as the camera moves upwards. then the crowd pans... look at 0:02:42. It jerks a few times in that scene. It's not just my eyes.

[latreche34]

On my computer there is no jerking on the file in post #1, which means you have a playback issue as I said. FYI vdub is not a player, Use a dedicated player and see.

[lordsmurf]

Quote:

Originally Posted by latreche34

Just put the clip into vdub2 and skipped it frame by frame of the last 40 seconds that you said problematic, not a single frame drop, you have a playback issue to address.

Sometimes frame-by-frame is misleading.

When I played the file, I saw would could have been dropped frames. But unknown when that happened. It could be the source (pre-tape), the ES15, the capture card. I played it forwards at realtime, halt-time, 2x time, and backwards realtime. I also went frame by frame.

After all of that, still inconclusive.

The main issue here is the computer playing these has issues. It could even be the version of VLC, as not all have been good, and sometimes the new version has problems. This is why most of us have multiple players for testing, and that depends on OS. I forget the Mac and Linux softwares, but for Windows it's MPC-HC (or MPC-BE).

[bmichaelb]

Quote:

Originally Posted by latreche34

On my computer there is no jerking on the file in post #1, which means you have a playback issue as I said. FYI vdub is not a player, Use a dedicated player and see.

I've already mentioned VDub is not a player. And as for the issues I was seeing, yes, some of it was fixed by swapping my GPU. That issue was microstutter on a timed interval, where it would like:

........!.!.!!!!!!!!!!.!.!!........

As I also said, I later discovered that one of the 2 GPU fans wasn't connected, so wasn't being powered on. I originally had done a Noctua mod, but decided to replace the original fans back on. The wire plug is still on the pins, but seem to be pushed to the top of the pins, so not really making contact. It's only an RTX 3060, stock settings, but my guess is it was hitting one of the Nvidia GPU Boost temperature limits where it starts to decrease slightly, then boost back up once it dropped below that threshold. I also said that while viewing in VirtualDub, if I saw a flutter/microstutter, I'd stop and rewind a few frames and watch it again... it was gone. But... it would then happen a few seconds later on a stretch that normally didn't flutter.

But the other issue was stuttering when the camera panned over the audience, or even sometimes on stage. Even after lordsmurf deinterlaced it, it still had a single hiccup at the end of the crowd pan. It's the exact same on both my capture pc, and sitting here on my VR gaming/Editing PC with an RTX 3080. But as I also said, as soon as I deinterlaced that, it was fine. But to say there's absolutely no jitter while watching the files I uploaded may be 'your' eyes. I'm fairly sensitive to tiny stutters.

Edit: Also, on the original dvd trials, it's 100% smooth, even on crowd pans.

[latreche34]

That's the nature of interlaced video, it is not smooth, period, While it looks stutter to you, it is the way SD video has been for last few decades compared to modern progressive 60p. Having another version on DVD doesn't make the capture bad, it was probably recorded from a better master.
Upload another file that looks worse and I'll take a look at it, The one in post #1 does not have stutter.

[bmichaelb]

I realize that's what interlaced video on a progressive screen is like, we already covered that, and I said I felt like an idiot for not realizing it a lot earlier... like years earlier. This is my 4th attempt, each time either not happy with the results, or as I stated above, my last attempt didn't have the levels corrected before capturing, and after reading about the histogram in VirtualDub, I ran a test capture with my previous settings... it was far in the red on the contrast side, and bit red on the brightness side. So this is my next, and hopefully last attempt. But... my previous attempt also has strange black dots appearing down a single line, about 20 pixels in from the left side. I thought it was satellite transmission artifacts, but they weren't there in the DVDs. I chalked it up to my Hauppauge HVR-1850 finally giving up the ghost. The ATSC chip died a few years before that. That's when I got the Hauppauge USB-Live2, but the thing wouldn't keep it's proc amp settings between captures or reboots, and needing to keep all 65 hours looking the same, I decided to get the Pinnacle USB-510 from lordsmurf. It's timing is rock solid, and allows me to use VirtualDub. And it keeps it's proc amp settings. Hauppauge doesn't play nice with VDub, so I was using AmarecTV. But both cards, and both apps still give the same issues.

The masters for the DVD are these exact same tapes. I originally wrapped each tape box in saran wrap, then boxed them up until I was ready to transfer. Then after each attempt, they got rewrapped. That's why I'm p*d at this attempt... they were pretty much pristine until I started again. But as for the DVDs... I'm just pointing out that whether the VOBs are played on the pc, or the disc is played through the ES16, the videos are completely smooth. (ES16, not ES15... it's the Canadian version of the ES15). But seeing as the DVDs are still interlaced, it boggles me as to why the vhs captures are choppy until deinterlaced.

I posted the full 2min clip that the first clip came from in post #3.

https://www.dropbox.com/scl/fi/g22bp...=emr53qbx&dl=0

I also uploaded that 10min screen capture of the vdub capture window, where there's quite a bit of stutter, especially where the camera pans over the crowd. It's not always the entire pan, or every crowd pan. Sometimes it's near the end of the pan, or starts 1/2 way through. And as I said in my first post, I see the same thing in Netflix as the camera pans sometimes. And I know that's not just my eyes, as that's what the 'smooth' settings in the tv configs are for, but people complain about the soap opera effect. But it's the same type of jitter I'm seeing. And it's not display hardware, as a rewind still has the same jitter during those shots, and on both PCs.

Edit: And the screen cap file shows the stutter when played back on my phone that it was recorded with, so it's definitely not just my PCs.

[lordsmurf]

Never seal tapes in plastic wrap, it taps moisture.

MPEG interlacing is different than lossless interlacing due to the compression. There's a slop in it, and it can go either way in terms of visual quality, either hiding issues, or highlighting issues. DVD-Video is not the measuring stick of quality.

[latreche34]

MPEG-2 playback is much easier on most media players compared to lossless AVI, it is not about how powerful a CPU or the graphic card, it is how stable the player is, I've had VLC stutters on a powerful modern PC and MPC-HC played smooth on Win 7 2008 machine, I've seen threads like this before where the OP thinks the capture card is at fault, but the file playback is the issue.

[bmichaelb]

Quote:

Originally Posted by lordsmurf

Never seal tapes in plastic wrap, it taps moisture.

MPEG interlacing is different than lossless interlacing due to the compression. There's a slop in it, and it can go either way in terms of visual quality, either hiding issues, or highlighting issues. DVD-Video is not the measuring stick of quality.

But the tapes are sealed to begin with, and even when people store batches of tapes in plastic bins, it's the same concept. Heck, they even sold individual closable plastic containers for VHS tapes... which is what any video store stored them in. It's also what tv stations store their tapes in. Yes, if it's humid out, then maybe you might get moisture. But not in the middle of July. And being a chef, I wrapped them fairly well, both individual boxes, then in stacks of 5. So unless the wrap was done poorly, then no moisture should ever enter the tapes. And, always stored them in a warm place... not hot, but definitely not cool. That's where mold starts to grow. Also, mold needs oxygen to live, so being sealed prevents that.

That makes sense about the MPEG vs uncompressed avi, as I said in pm that converting did seem to fix it as well, even when still interlaced. I seriously wish I'd come here 10 years ago... lol.

[lordsmurf]

Quote:

Originally Posted by bmichaelb

But the tapes are sealed to begin with, and even when people store batches of tapes in plastic bins, it's the same concept. Heck, they even sold individual closable plastic containers for VHS tapes... which is what any video store stored them in. It's also what tv stations store their tapes in.

People do dumb stuff all the time, sometimes even a majority of the time. (Give me time to ponder, and I'm sure I can think of something dumb I was doing, before I learned better. EDIT: Yep, here's one: Reusing random old rags to clean the car, then wonder why it has scratches. Youtube has good car cleaning advice, unlike videotape related advice. Bought a pack of microfiber disposable clothes, no more issues!)

Sealed new tapes had an intended shelf life, not to be hoarded/warehoused and sold 20+ years later as "new" (because they're not at all new anymore, merely unused for recording).

Plastic bins offgas, that's very damaging, and a common source for tape condition issues.

Quote:

Yes, if it's humid out, then maybe you might get moisture. But not in the middle of July.

Time of year has nothing to do with it. Humidity is 24/7/365. In fact, in many places (mine especially), summer is vastly more humid than winter.

Quote:

And being a chef, I wrapped them fairly well, both individual boxes, then in stacks of 5. So unless the wrap was done poorly, then no moisture should ever enter the tapes.

You didn't wrap it down to a molecular level. In fact, the plastic wrap itself is porous. Saran Wrap, all kitchen plastic wrap, is especially porous, to let steam escape (microwaves excite water molecules).

Quote:

And, always stored them in a warm place... not hot, but definitely not cool.

Very, very bad.

Quote:

That's where mold starts to grow. Also, mold needs oxygen to live, so being sealed prevents that.

Mold grows in all environments.

Quote:

I seriously wish I'd come here 10 years ago... lol.

You're here now, that's what matters.

And I think you need some good storage advice, not just playback or interlace advice.
Topic for a new thread.

[bmichaelb]

Quote:

Originally Posted by lordsmurf

Sealed new tapes had an intended shelf life, not to be hoarded/warehoused and sold 20+ years later as "new" (because they're not at all new anymore, merely unused for recording).

Plastic bins offgas, that's very damaging, and a common source for tape condition issues.

And yet that's how the tv industry has done it since the days of betamax using individual plastic boxes.

Quote:

Time of year has nothing to do with it. Humidity is 24/7/365. In fact, in many places (mine especially), summer is vastly more humid than winter.

I definitely disagree with that one. Here on the coast, it's more humid in the fall/winter. Books left in a garage are dry in the summer, moist in the fall/winter. And I'm beside the ocean, so we're naturally more humid than the interior.

Quote:

You didn't wrap it down to a molecular level. In fact, the plastic wrap itself is porous. Saran Wrap, all kitchen plastic wrap, is especially porous, to let steam escape (microwaves excite water molecules).

And yet chefs have been doing this for close to 100 years already. Have you ever had meat in the freezer, and gotten freezer burn? It's because it wasn't wrapped properly. This is literally what they teach you in chef school, or by the sous chef in a restaurant. You wrap an item top to bottom... 5 or 6 times, then you rotate 90 degrees and do the same. It's also how they make gravlox... if it's not wrapped properly, it won't cure properly due to the outside air and moisture seeping in. Chef would tear your hide if his gravlox was soggy.

Quote:

Very, very bad.

Mold grows in all environments.

You're here now, that's what matters.

And I think you need some good storage advice, not just playback or interlace advice.
Topic for a new thread.

Thanks, but we're definitely going to have to agree to disagree on this one bud. But cheers... I'm going to bed.

-- merged --

Damn... forgot another example... Hollywood. Every single film archive is kept in those film cannisters, even to this day. If mold was an issue by storing tape in a sealed container, then they would never have done it.

[lordsmurf]

Quote:

Originally Posted by bmichaelb

And yet that's how the tv industry has done it since the days of betamax using individual plastic boxes.

I'm not aware of anybody in the TV industry that stores videotapes in plastic kitchen wrap, nor plastic storage bins. That's not a thing.

Quote:

I definitely disagree with that one. Here on the coast, it's more humid in the fall/winter.

And at other times, it's less humid. But the humidity exists. More is more damaging, less is less damaging, both damage.

Quote:

And yet chefs have been doing this for close to 100 years already. Have you ever had meat in the freezer, and gotten freezer burn? It's because it wasn't wrapped properly.

Plastic wrap is porous, and allows in moisture, which then freezes. Freezer burn is caused due to moisture wicking out of the meat. You'll get initial freeze-time water molecules that freeze up, plus any that form from power dips that lower freezer temperature temporarily. Plastic wrap actually induces freezer burn, because it allows in air, while freezer paper does not. I don't need to be a chef to know science, nor to see how this plays out in my own freezer.

I store meat in freezer paper, which is what professional meatpackers do. They don't store in Saran Wrap -- which was made by Dow Chemical, and the food storage version was made for temporary fridge storage. Not freezer, not long-term. Plastic wrap was later used to nuke/microwave food, because it was convenient to reheat leftovers, which was often already on the plate/container.

Due to dangerous chemicals, Saran Wrap was reformulated 20 years ago, also made thinner, and has been crap ever since. We (family) quit using it for food long ago (we now move all leftovers to plastic storage tubs), and the modern version of plastic wrap is pretty miserable for most things. We use a paper plate, upside down, to nuke our leftovers.

Quote:

This is literally what they teach you in chef school, or by the sous chef in a restaurant. You wrap an item top to bottom... 5 or 6 times, then you rotate 90 degrees and do the same. It's also how they make gravlox... if it's not wrapped properly, it won't cure properly due to the outside air and moisture seeping in. Chef would tear your hide if his gravlox was soggy.

This may be true, but I would wager it's "old school" knowledge that never got the memo from 20 years ago. Things changed, they didn't. You see that in many industries.

I never heard of gravlox. Wow, that looks yummy!

Quote:

Damn... forgot another example... Hollywood. Every single film archive is kept in those film cannisters, even to this day. If mold was an issue by storing tape in a sealed container, then they would never have done it.

Film is not tape. Vastly different.

[bmichaelb]

Quote:

Originally Posted by lordsmurf

I'm not aware of anybody in the TV industry that stores videotapes in plastic kitchen wrap, nor plastic storage bins. That's not a thing.

lol... I didn't say that. I said wrapping is the same as using those plastic boxes made for VHS tapes, and the industry puts all tapes in those plastic VHS boxes. The lid folds, and it seals fairly well.

Quote:

And at other times, it's less humid. But the humidity exists. More is more damaging, less is less damaging, both damage.

Yes, but the fact of the matter is that whether the air already in the plastic VHS shell is wrapped in plastic or in a carboard sleeve, it's only going to get affected by the outside humidity if that air gets into the plastic shell. It's not a living breathing item. If you seal it properly, the outside humidity will never affect it. Period. And as I said... humidity needs oxygen to even begin. A sealed box won't allow any more to ever enter it. And as long as the tapes are kept within the acceptable temperature, you should be fine.

Quote:

Plastic wrap is porous, and allows in moisture, which then freezes. Freezer burn is caused due to moisture wicking out of the meat. You'll get initial freeze-time water molecules that freeze up, plus any that form from power dips that lower freezer temperature temporarily. Plastic wrap actually induces freezer burn, because it allows in air, while freezer paper does not. I don't need to be a chef to know science, nor to see how this plays out in my own freezer.

Freezer burn is caused by cold, dry air getting in. Period.

Quote:

I store meat in freezer paper, which is what professional meatpackers do. They don't store in Saran Wrap -- which was made by Dow Chemical, and the food storage version was made for temporary fridge storage. Not freezer, not long-term. Plastic wrap was later used to nuke/microwave food, because it was convenient to reheat leftovers, which was often already on the plate/container.

And yet every single food supplier delivers plastic sealed meats without issues.

Quote:

Due to dangerous chemicals, Saran Wrap was reformulated 20 years ago, also made thinner, and has been crap ever since. We (family) quit using it for food long ago (we now move all leftovers to plastic storage tubs), and the modern version of plastic wrap is pretty miserable for most things. We use a paper plate, upside down, to nuke our leftovers.

And yet restaurants have absolutely no problem with it.

Quote:

This may be true, but I would wager it's "old school" knowledge that never got the memo from 20 years ago. Things changed, they didn't. You see that in many industries.

What they teach in chef school is techniques that have been honed for centuries. If there was a better way, they would have done it.

Quote:

I never heard of gravlox. Wow, that looks yummy!

It's cured salmon. It's huge with the Jewish community... 'lox and bagels'. It is really good indeed.

Quote:

Film is not tape. Vastly different.

True, but it does get mold as well. I actually just read an article about film restoration from nyu. but... those reels were pulled from tins, so that blows my theory out of the water. But as I said, the air in the tape shell is no different in a cardboard sleeve than if it's wrapped. It's only the air that seeps in that causes issues.

-- merged --

I also remember having Fuji tapes that came with a plastic sleeve, not a cardboard sleeve. That would protect the tapes far more than cardboard. Cardboard can suck up the moisture in the air and transfer to the tape on it's own, while plastic sleeves would prevent that, as long as the sleeve has a tight enough seal with the tape inserted.

[lordsmurf]

The key here is that it's not just any random plastic. Kitchen food wrap isn't the same materials or specs of videotape cases.

Storage needs to be indoor temperate, preferably on the cooler side (60s to low 70s F), with the RH% as low as possible under normal conditions (50% target range, less is great).

TV stations, museoms, etc -- this is what they do. I should know, I work with many.

The cardboard cases are generally fine, because those are slick "cardboard" (more like thick paper stock), not corrugated cardboard. The coated outer layer can act very much like plastic, because it often is a form of plastic that creates the coating.

You have to realize this is something I have about 30 years of knowledge on. I actually knew quite a bit about paper/cardboard at least a decade before I ever did anything with video.

[bmichaelb]

Quote:

Originally Posted by lordsmurf

The key here is that it's not just any random plastic. Kitchen food wrap isn't the same materials or specs of videotape cases.

But when plastic is wrapped in many layers around the object, then it's no different. If it prevents freezer burn when wrapped properly, then it would work with tapes as well. It's not rocket science.

Quote:

Storage needs to be indoor temperate, preferably on the cooler side (60s to low 70s F), with the RH% as low as possible under normal conditions (50% target range, less is great).

This is why I said kept within a proper temperature, and not out in the garage. It's also why I said that whether the air in the plastic shell is then boxed in cardboard or plastic makes no difference to the air already inside. Only any air or moisture that leaks into it will make a difference.

Quote:

TV stations, museoms, etc -- this is what they do. I should know, I work with many.

Cool... but TV stations still keep them in those plastic cases. So after that, it comes down to storage temperature.

Quote:

The cardboard cases are generally fine, because those are slick "cardboard" (more like thick paper stock), not corrugated cardboard. The coated outer layer can act very much like plastic, because it often is a form of plastic that creates the coating.

Yes, they have a coating on the outer surface of the cardboard, but not the edges of the opening where the cardboard was cut in the factory, or the inside of the cardboard sleeve. I spilled coffee once... outer surface only beaded up, but the edge got soaked, and so did the inside. The open edge will be a wick in a humid environment.

Quote:

You have to realize this is something I have about 30 years of knowledge on. I actually knew quite a bit about paper/cardboard at least a decade before I ever did anything with video.

Cool. You have to realize I'm not new to VHS as I'm pushing 60, and besides the physics I already know from school, my 20 years in the kitchen gives the experience and knowledge about the use of plastic wrap. Cooking is science, both physics and chemistry. Saran wrap is not 'porous'. It doesn't leak on it's own. Fill a jar with water, lay Saran over top, and use an elastic to seal it. Then support it upside down. It's not going to leak.

I'm done. Cheers, and good night.

-- merged --

Well, I found the issue in the sequence with the drummer... one of the frames was glitched in the top field causing the jitter. It wasn't just my eyes. But using johnmeyer's script, I was able to fix that frame. But... it's slightly higher than the original, and cropping a single line off the bottom doesn't do it. If fixing bottom field, then the frames are the same height. So I may have to splice by sequence, or at least find a spot where the camera's not panning up or down. I'm planning on using this method to fix a few dropouts as well. They aren't always in both fields at the same time. It may not always work, but it's definitely better than a filter that blurs.

#Replace Bad Field With Motion Estimated Field from Good Field.avs
#John H. Meyer
#Copyright 2015
#

source = AVISource("D:\Tape 1v1.avi").assumetff()
fields = SeparateFields(source)             

even = SelectEven(fields) 
odd  = SelectOdd(fields)  

#Change the following to point to the good field
#good = odd ## top
good = even ## bottom

# Double the height to create spatial interpolation
double_height = Spline36Resize(good,good.width,good.height*2)  

#Double the frame rate to create temporal interpolation
estimated_clip = DoubleFPS2(double_height) 

#Keep only the motion-interplated fields
#replacement_fields = estimated_clip.selectodd().assumeframebased().separatefields().selectodd() 

replacement_fields = estimated_clip.selecteven().assumeframebased().separatefields().selecteven()

#Interleave the synthesize field that will replace the bad field, with the good field
Interleave(good,replacement_fields)
final=Weave()

#Return the fixed video
return final


#-------------------------------
#Debugging functions

#Before you start, enable this next line temporarily so you can see which field is the good field
#stackvertical(even,odd)

#When you are almost finished, enable this next line temporarily to check your work
#return final.SeparateFields

#------------------------------------
function DoubleFPS2(clip source) {
   super = MSuper(source, pel=2)
   back = MAnalyse(super, chroma=false, isb=true,   blksize=16, overlap=4, searchparam=3, plevel=0, search=4)
   forw = MAnalyse(super, chroma=false, isb=false,  blksize=16, overlap=4, searchparam=3, plevel=0, search=4)

#  MBlockFPS(source, super, back, forw, num=2*FramerateNumerator(source), den=FramerateDenominator(source), mode=0,thSCD2=130,thSCD1=400)

#  Alternate motion estimation. 
  MFlowFPS(source, super, back, forw, num=2*FramerateNumerator(source), den=FramerateDenominator(source),thSCD2=130,thSCD1=400)

}

[lordsmurf]

Quote:

Originally Posted by bmichaelb

Well, I found the issue in the sequence with the drummer... one of the frames was glitched in the top field causing the jitter. It wasn't just my eyes. But using johnmeyer's script, I was able to fix that frame. But... it's slightly higher than the original, and cropping a single line off the bottom doesn't do it. If fixing bottom field, then the frames are the same height. So I may have to splice by sequence, or at least find a spot where the camera's not panning up or down. I'm planning on using this method to fix a few dropouts as well. They aren't always in both fields at the same time. It may not always work, but it's definitely better than a filter that blurs.

Interesting. I have a complicated video with similar jitter issues. Personal (family) video, 10 minutes long, not too important, but I do want to fix it eventually. I need to tuck this script away for my next attempt. Honestly, I may have already tried it, but forgotten.

[bmichaelb]

It woks quite well, as long as there's not too much movement... like drum sticks with tracers due to moving so fast. But when trying to fix the upper field, as I say, it's not lining up identically to the original. I guess it's shifted slightly due to using the bottom field as the guide to interpolate the top field. I have at least 1 streak where it spans 3 frames... starts in the top field, ends in the bottom field, but the middle frame it's in both fields. So the plan was to repair 1st and 3rd frame, then interpolate the middle frame with the newly created neighbor frames. Hopefully.

I've been trying to play with the script lollo2 supplied in a previous thread of mine to shift fields in individual frames, but when he first shared it, all the code was kinda Greek to me. If I see it in an example, I can apply it to what I need. But looking at it last night/this morning, it's kinda makings sense, but it's giving me an error in my settings:

Avisynth open failure:
Script error: expected a parameter name
(D:\Tape 1v1.avi, line 3, column 28)

The frame I testing with is 58, bottom field shift by up by 1:

AviSource("D:\Tape 1v1.avi")
#function shift_fields_GMa(clip c, int frame_number, line_shift_even, line_shift_odd)
function shift_fields_GMa(58, 0, 1)
{
# separate fields tff
c_tff_sep=c.AssumeTFF().separateFields()

# separate fields tff even
c_tff_sep_even=c_tff_sep.SelectEven()

# separate fields tff odd
c_tff_sep_odd=c_tff_sep.SelectOdd()

# shift field even
c_tff_sep_even_rep = (line_shift_even > 0) ?\
	c_tff_sep_even.trim(0,frame_number-1)\
	++c_tff_sep_even.trim(frame_number,frame_number).crop(0,0,0,-line_shift_even).addborders(0,line_shift_even,0,0)\
	++c_tff_sep_even.trim(frame_number+1,0)\
:\
	c_tff_sep_even.trim(0,frame_number-1)\
	++c_tff_sep_even.trim(frame_number,frame_number).crop(0,-line_shift_even,0,0).addborders(0,0,0,-line_shift_even)\
	++c_tff_sep_even.trim(frame_number+1,0)

# shift field odd
c_tff_sep_odd_rep = (line_shift_odd > 0) ?\
	c_tff_sep_odd.trim(0,frame_number-1)\
	++c_tff_sep_odd.trim(frame_number,frame_number).crop(0,0,0,-line_shift_odd).addborders(0,line_shift_odd,0,0)\
	++c_tff_sep_odd.trim(frame_number+1,0)\
:\
	c_tff_sep_odd.trim(0,frame_number-1)\
	++c_tff_sep_odd.trim(frame_number,frame_number).crop(0,-line_shift_odd,0,0).addborders(0,0,0,-line_shift_odd)\
	++c_tff_sep_odd.trim(frame_number+1,0)

# repaired video
c_rep=interleave(c_tff_sep_even_rep,c_tff_sep_odd_rep).Weave()

return(c_rep)
}

Line 3, column 28 is between 5 and 8 of 58.

So... what the heck am I doing wrong?

[bmichaelb]

Here's more info from lollo2:

Quote:

If you see in the frame X a shift of Y line up in the even field you use shift_fields_GMa(X, Y, 0)
If you see in the frame X a shift of Y line down in the even field you use shift_fields_GMa(X, -Y, 0)
If you see in the frame X a shift of Y line up in the odd field you use shift_fields_GMa(X, 0, Y)
If you see in the frame X a shift of Y line down in the odd field you use shift_fields_GMa(X, 0, -Y)

If you need to replace a bad frame with an interpolated frame from frame-1 and frame+1 you can use this function

function interpolate_frame_GMa_1(clip c, int "frame_number")
{
	# interpolate bad frame
sup = c.MSuper()\
bv = MAnalyse(sup, isb=true, delta=2)
fv = MAnalyse(sup, isb=false, delta=2)
interpolated = MFlowInter(c, sup, bv, fv)

#stackhorizontal(c, interpolated)
#return(interpolated)

	# replace bad frame with interpolated frame
video_repaired = \
	c.trim(0,frame_number-1)++\
	interpolated.trim(frame_number-1,-1)++\
	c.trim(frame_number+1,0)

	# video repaired
return(video_repaired)
}

if you see a bad frame number X you use interpolate_frame_GMa_1(frame_number=X)


When dealing with multiple consecutive bad frames you need to replace the additional bad frames with a duplicate of the good frames and then interpolate one by one; in this example how to proceed for 7 consecutives bad frames

# interpolate frame GMa_1 2921-2927
# a x      x     x    x    x    x     x      b
# a a      a     a    x    b    b     b      b
# a a      a     a    iab  b    b     b      b
# a a      a     iaab iab  b    b     b      b
# a a      iaaab iaab iab  b    b     b      b
# a iaaaab iaaab iaab iab  b    b     b      b
# a iaaaab iaaab iaab iab  iabb b     b      b
# a iaaaab iaaab iaab iab  iabb iabbb b      b
# a iaaaab iaaab iaab iab  iabb iabbb iabbbb b

# replace frames 2921, 2922 and 2923 with frame 2920 and frames 2925 and 2926, 2927 with frame 2928
video_org_rep1=video_org.trim(0,2920)++video_org.trim(2920,2920)++video_org.trim(2920,2920)\
++video_org.trim(2920,2920)++video_org.trim(2924,2924)++video_org.trim(2928,2928)\
++video_org.trim(2928,2928)++video_org.trim(2928,2928)++video_org.trim(2928,0)

# interpolate bad frames
video_org_rep2a=video_org_rep1.interpolate_frame_GMa_1(frame_number=2924)
video_org_rep2b=video_org_rep2a.interpolate_frame_GMa_1(frame_number=2923)
video_org_rep2c=video_org_rep2b.interpolate_frame_GMa_1(frame_number=2922)
video_org_rep2d=video_org_rep2c.interpolate_frame_GMa_1(frame_number=2921)
video_org_rep2e=video_org_rep2d.interpolate_frame_GMa_1(frame_number=2925)
video_org_rep2f=video_org_rep2e.interpolate_frame_GMa_1(frame_number=2926)
video_org_rep2=video_org_rep2f.interpolate_frame_GMa_1(frame_number=2927)

My interpolate_frame_GMa_1 function is based on http://avisynth.org.ru/mvtools/mvtools2.html. In doom9's forum you can find alternative procedures written by several users for frame interpolation.

edit: to avoid audio problems with AlignedSplice ++ AviSynth command, better to dub with the original audio after the shifting and the replacement:

# use audio from original clip
video_restored=audioDub(video_org_rep2, video_org)