Two separate comments? Why not edit the first one? I don't know but here's another one and I'm gonna pin it! Firstly, apologies for the clipping audio. I set the audio level higher to record the sound of the VCR and forgot to set it back. And here's that link to the T-shirt. It was in the description the whole time. But now here's another one. Cool! www.redbubble.com/people/fuzzclone/works/36782468-cromaclear-slot-mask-crt-pattern?p=classic-tee&ref=explore-recently-viewed
I remember anolog being washed out. It was kinda the whole reason I never really watched our home videos, they just look really bad. I'm more the kind of person who when they buy a TV I boost the contrast and turn the color to vibrant or dynamic for extra color vibrancy. I remember only 10 years ago having a tube TV that supported digital channels and I thought they looks so much more colorful while at the same time being clearer. Unfortunately my cable company ended up ending their digital channels only a year after I learned about them, now requiring a decoding box so that means digital was no longer an option. Later they would even force analog channels to have a digital decoder. What a scam. It seems like tube TVs with digital support is incredibly rare but unfortunately the TV no longer works. I somehow ruined the light causing it to flicker. With that, I finally moved up to a flat screen in 2013. That was a really short lived TV.
I'm speculating, but could the hdmi capture look better because it also does Gamma Correction? Analog TV is meant to be shown on CRT and modern LCD is using HDMI. These technologies have different Gamma curves. There is probably also some brightness levels that should be tuned in addition...
What I find funny, and I know that this isn’t the point of this video, is that we can’t see things as well as we think we can. What we see is thanks to our brains filling in what we don’t see.
As an engineer who started his film/video career at the beginning of the Beta/VHS age I can say that VHS should never look washed out, that is simply ludicrous and is the result of a terrible quality transfer. Your easy to use method to perform transfers is awesome, easy, reliable and delivers quality better than what most people at home are getting otherwise, thank you!
I have a feeling that this recollection people have of washed out VHS video comes entirely from poorly set up televisions. Television with default settings for the showroom with a setup to look good under bright florescent lighting would look washed out, and I had plenty of friends with that issue on their television. My own parents had this strong belief for quite some time that the colour rendering on the television was too strong and vibrant when I had set it. I suspect this came from being used to earlier colour televisions perhaps not being as vivid, or equally poorly set up. The old BBC test screen did help make my point about colour rendering though. So, short story long, I suspect it is likely these people are conflating the washed out colour of their poorly set up television with the output from VHS.
Professionally produced videos should look ok on VHS (shot on studio grade cameras, with proper lighting, etc). Home movies always looked washed out as camcorders at the time had cheap image sensors and rarely the best shooting conditions for those sensors.
@@NJRoadfan None of my home videos ever looked washed out. Between 2 Sony 8mm , 1 Sony Hi-8, and 1 JVC VHS-C cameras I never had any issues with videos that had washed out colours. Obviously they were not studio quality videos, but colour rendition was absolutely fine (except when lighting conditions were suboptimal).
@@TheRealColBosch Not guessing. I have both shot and transferred video on period 80s and 90s studio cameras and consumer camcorders of various quality (from tube to MOS and CCD). I remember being very disappointed with the Panasonic AG-456 SVHS camcorder in particular. Despite being from their professional/industrial line and having a higher quality CCD sensor to take advantage of the extra SVHS resolution, the colors were always on the dull side, even on daytime outdoor footage.
Think the best way to sum up your sentiment is that for 95 percent more effort, you can make it 5 percent better. And for most sane people, it's not worth it.
@@user-vn7ce5ig1z Youre the reason he had to make this video. Sure, there is going to be shit like the moon landing that needs to be completely redone because of advances in tech, or because of the value of the footage, we are talking about the best prosumer level solution opposed to the consumer solution of pointing a camera at your TV
@Pusalieth dude, you do see this is the only comment he has agreed with. Im thankful to be a patron to this guy, cuz i know at the end of the day it offsets the BS he has to deal with.
@@FierceDeityLink1 less effort after you have learned the entire process right? This is what he is getting at. Those from the outside looking in don't even know what that is called, and if you're someone looking for a simple solution, you do just that, go for the simple option.
"You trying to convince me that I shouldn't be happy is one of the most frustrating things that the internet does." That's almost T-shirt- or coffee mug-level pithy.
it is absolutely on this level. I could use a coffee mug with these sentiments scrawled on it. I don't even drink coffee, but I would be more than happy to stare at my empty mug and think "Yes. This is very true" and start my day on the right foot, with profound knowledge straight from Technology Connections.
I own a TV comercial production company and I work with old materials and I found the same problems as you when digitizing old analog video. And as you say analog video has great image quality if you find a nice and CONVENIENT way of digitizing. I'm bored of all these know-it-all guys that has allways something to say. So..keep your good work and your great way of explaining complex technologies in an easy to understand manner.
exactly, there's a thing that's worse than digitizing the old footage at 95% quality: not digitizing it at all because it requires more effort than just pressing two buttons
ARGHH! I was gonna go into this, but forgot. S-video was most commonly used in S-VHS VCRs, and while there are a few standard VHS machines with S-video, that's on the whole even rarer than SVHS, in my estimation. In fact, S-video is so synonymous to S-VHS, that people often confuse the two. Laserdiscs *do* store composite video, and this may be why dot crawl can be so bad. But all consumer videotape formats that I know of used a color-under encoding scheme to get around their own bandwith limitations and present a reasonable color image after translating it into real NTSC or PAL.
Technology Connections 2 its a shame that you guys in the USA don’t have the SCART connecter like I do in the UK which transmits an RGB signal rather than composite. Combined with PAL being slightly higher resolution, means that analog video looks surprisingly good. In fact a good quality tape on my 6 Panasonic VCR looks almost as good as early DVDs.
@@doctormac123 most PAL VHS players just output composite trough scart and even if they output RGB the only thing that have improved color with RGB scart is the the on screen menus/overlays produced by the VHS player itself. The only benefit with PAL VHS is no dot crawl.
I learned the hard way to capture laserdisc from composite and VHS from S-Video (yay, extra ADB cables) when my laserdisc capture looked awful! After watching the previous video I thought, "hrm, this setup isn't for me but I really didn't know about those awesome 10¢ chips, so maybe I should head over to Amazon and see what's new since 2012." I currently have a $49 Hauppague (linux support advertised) on my electronics list for when I get some time that has all four inputs and USB out. I am somewhat confident that I will finally be able to get the quality I need out of some of those old tapes that I've been keeping because "argh, this is quality?" I think future generations may appreciate the work. So, even though I'm not using your specific setup, and I may even use ffmpeg (bum, bum, buuuuum) I still benefited greatly from your video and I absolutely love your channel. Thanks for all that you do, and congrats on the lovely new studio!
Starlight Yeah I’d say it’s the lack of dot crawl that makes it look good, I was convinced is was an RGB signal because of the lack of the usual composite artefacts. Also the TV I’m using is a Pioneer Kuro KRP-500a plasma, 50 inch in size, very high end model from the late 2000s it might just be very good at playing analog video.
@@Stjaernljus, My experience is exactly the opposite. The SCART has separate pins for RGB and Composite and it outputs both at the same time. Whenever RGB or Composite is desired is commanded by another dedicated pin.
KISS... You did just that. It's a simple process, gets quality results, reasonable costs. Done deal. If someone knows a better way, then they can make a video about it.
Your saying how you’re tired of rabbit holes and wanted a solution that just works made me think of this: “Do I look like I know what an FFMPEG is? I just want a video of a gosh dang hot dog!” -Alec Hill
People with a high level of knowledge in this area just get excited when normal people express interest. In your other content, you have always welcomed additional information in the comments because you seem to genuinely enjoy learning about this pedantic, nerdy stuff. You wouldn’t be the first for whom the intricacies of video processing was their limit, though.
I sorta agree, it's odd that this is the line lol, but I also got highly interested in analog video capture when I was in middle school and even carried the weight of the 1st broadcasting class in my High School. I thoroughly enjoyed it, mostly, but there were aspects of it that could be a real slog to get through. Maybe someone should just create an analog video capture Linux distro with wine and everything you would want, including premade scripts. I think it actually would be possible for people to more or less condense down all the knowledge in videohelp and other places and create something that could just work in general. I also still contend that finding a used VCR with firewire built in would be the simplest method - and possibly not cost much more than all the gear he listed. It would certainly save him a lot of time from not having to join all those separate clips. You could also feed the composite or svideo into a DVD recorder, as many of them had firewire built in and that'd be even better and cheaper than finding a VCR that had firewire natively. Firewire was also the interface that was actually designed for uncompressed SD video in the 90's and despite all of technological progress I don't think any modern capture device you find will out do a firewire captured SD video. Of course you will need a firewire pci card, but those are extremely cheap now. It also allows you to capture without needing to merge video clips later. Also ilo DVD recorders with Firewire sell for $50-60 on ebay, they work great. I bought one brand new for only like $100 back in the day.
Ben Reaves I don’t think it’s weird at all. I’ve been studying video encoding and restoration for 10 years and I still feel like a noob sometimes. I totally agree that there aren’t many good learning resources either.
Hmm. Weird. I thought you made it clear that the video was about an easy and fast "good enough" capture of analog video. So definitely not your whole audience.
Same, in fact I almost made a comment myself about other methods he could try but then I was like "I'm sure that's not what he wants" and decided not to. I think a problem with youtube is that many people comment before they even finish watching a video.
That's hilarious. And there's the rub: video from that era was crushed all over the place to begin with. There's only so much effort you want to expend polishing the turd.
This is somewhat what they did with D-VHS. They sendt back a forum guru that commanded a engineer that made the best way to film video with technology at that time. Soon we will get a guru who can capture vhs at the best method.
As a developer-type person, I agree with you about FFmpeg. It's a wonderfully-powerful tool, and one that I do use from time-to-time, but untangling the options is a nightmare every time. I'm sure there's a logic to it for someone (sound and video engineers, perhaps?) but the seemingly inconsistent order of commands (some are positional while some are not) and all of the esoteric terminology is confounding. I don't blame you for avoiding it. The existing GUIs that I've tried (and it has been a while, so this may have changed -- but I doubt it) either simplify things to the point of removing a lot of the power benefits of the tool, or just shift the confusion to a new form. To be fair, it's a hugely complex tool and making a coherent interface for that must be daunting, but that doesn't change the fact that it's extremely difficult to use effectively without help. The solution you give in your video was a perfect example of "this gets me within personal tolerance levels of my goal with a minimum of effort, and anything else doesn't provide enough benefit to offset the additional cost". I thought it was a very good explanation.
I agree. ffmpeg is very powerful and can get excellent results if you know what you're doing, but the user-friendliness and usability just aren't there. I've also found it a headache because it can be so dumb that it will mishandle things unless you pass it a slurry of options. As an example, I remember having to re-encode a bunch of stuff because I discovered that transcoding from one resolution to another can confuse video players into displaying colors wrong because they expect a colorspace conversion… and ffmpeg does *not* handle that automatically. And how on Earth would the average user know they need to do that? I sometimes think ffmpeg was really aimed at the sort of person to spend a number of years working on a college degree in video encoding theory. It's pretty hard to recommend it to the average, nontechnical guy. But for simple cases (or at least those in which you finally figure out the proper option magic) it can be very handy, especially when you need to process a whole batch of videos in exactly the same way automatically, which is a lot of what I personally use it for.
Well once you've used it a few times, it's easy to use. But in the end ffmpeg is a tool and just that and not an end user application. It provides an API for that exact use, so other applications can actually use all the technology in ffmpeg and provide it to users in various ways, like in Chrome or OBS.
I mean the only thing I'd use ffmpeg for in this context is to trim the video or stich them together, lol. The scope of the video was obvious from the get go. And is honestly something I really appreciate. I never would have thought that using an analog to HDMI _and_ HDMI capture would work out better than a direct analog capture device. So I appreciate what you shared.
This tbh. You don't have to be a programmer to use ffmpeg. It's a handy tool that can save you some work, not to "take the fun away". It's like that XKCD comic of doing something manually vs automating it.
You could also use ffmpeg to correct the aspect ratio. If you were doing a production line digitization then setting up an automated process with ffmpeg could make sense. But for just grabbing a bit of analogue footage here and there, I agree that you just use whatever's most convenient.
One of the main advantages of using HDMI > HDMI capture is that you're not limited to the bitrate and format of consumer analog capture devices. A lot of them are limited to 8mbps MPEG-2 for DVD compliance reasons. I personally like capturing at 480i H.264 and let video cards do the deinterlacing on playback, but I can understand why the method in question makes it simple for a lot of people, especially if they're using players like VLC which don't deinterlace by default and suck at deinterlacing when you enable it.
@@hmich176 So what if there is a learning curve? Do you want to stay stupid and not learn anything anymore because things could be slightly complicated?
Alec, I agree with your assessment about the "black crush": 90+% of the time, it improves the overall image quality-or at least the perception of it-by decluttering the image, minimizing low value content in shadows, and expanding the luminous details. What is lost in dark content is more than made up for by expanding the dynamic range of the luminous content (less "washed out"). I'll take an easy, fast, and not expensive solution which improves the vast majority of content.
It depends on the footage. Crushed blacks in animation is always horrendous. Stuff that was drawn and always meant to be clearly visible ends up gone in most modern releases. "Rrmastered in HD", but really it's just a highly processed, blurry, dark, uneven upscale of 480p footage scanned from a copy of a copy of the original film prints. Dragon Ball Z and Moomin are like that. It's awful. Especially since the original film prints exist. Just sitting there...waiting to rot away...
Hey - I'll tell you this... As someone with a bunch of VHS-C tapes of long lost loved ones that are sitting in a box because I also didn't want to go down the rabbit hole of which X is the best........ I thank you for the video. It's exactly what I was looking for. An easy to use process that I can follow and achieve great results. Thank you for your research and help.
Steven Barnes I transferred my VHS-C tapes by running the VCR thru my mini-DV camcorder in pass through mode to my PC, but the terrible de-interlacing I get on the PC has me wanting to re-do them all using a setup like this!
EXACTLY my issue. I want to get a somewhat good digital save of my old home videos of the kids growing up, and this seems like a good enough solution before the VHS degrades and will be lost before the perfect method comes along. However ... I might have missed that boat. I moved from the UK to the USA in 1994, and while I used to have a PAL-NTSC VHS player, it (not surprisingly) died. These home videos are PAL, but I live in the USA and even back in the 90s it was tricky to find a PAL-NTSC VHS player (that plays all the speeds), so I doubt it's going to be easy today. My solution is likely to be send the home videos to my children in the UK, they digitize them ... but that brings other anxiety from shipping losses to a feeling they might not do as good a job as I would, or they might put it on THEIR "I'll do it one way" backburner, and it never gets done. Sigh...
I thought your first video was spot on. While we didn't have the best TV or VCR in our house, what we watched on pre-recorded, off air and, cam-cordered looked much closer to the results you achieved in that video. One can only imagine that the folks who take issue with your process, may have never seen a CRT during the era when they were king. While certainly not as high resolution as 720p and the like, they had great color and fantastic response times. The comment section is usually full of trolls who get the vast majority of their information from the internet and not actual experience. They also cannot be pleased or appeased. I really dig your videos and usually learn something watching them. I'm seriously still geeking out over that automatic toaster.
But have you tried capturing the footage with a turbo encabulator? It uses a base plate of pre-famulated amulite surmounted by a malleable logarithmic casing. I hear that helps keep the pixels from leaking out during the conversion.
@hqqns I've found ffmpeg used in tandem with a turbo encabulator works pretty seamlessly, but the pre-famulated amulite base plate will only polarize bit stream converter. Now, I've heard that if you re-bias the heads using a magneto-flux inhibitor, you might prevent the pixel leak a little more effectively.
Welcome to people arguing over why the law of diminishing marginal returns doesn't apply to them. Or the Pareto principle. While they are endlessly stressing about eking the last .5% of quality out of one video you'll have yours all done and most of humanity won't know or care about the difference, if any.
Keeping in mind, of course, that this technique will be applied for the most part to old home tapes in varying state of signal decay (or whatever you call it). Signal, mind you, which originally carried significantly poorer image than early DVDs. It's not even RGB, it's the raw composite signal that goes straight into the TV. That signal is converted to a digital image in pretty much real time by an algorithm you don't understand or have influence over, so by the time a video compression algorithm comes into play - most of the damage has been done. Any *real* videophile dedicated to doing the job properly will sample the raw composite signal at 7MHz (64-bit depth ought to suffice), then write their own signal interpreter to produce the image, then store the whole thing raw in its interlaced format :P
@@whophd I just learned some things. I don't really mess around with video editing but I have a tendency of perusing advanced controls of pretty much any application I use including apps like VLC. I generally won't touch anything if I don't know what it does but I remember seeing some of those terms and now I have some idea of what they mean to the quality of the video.
Black crush could have something to do with the PC vs TV output range. Hardware boxes you use could just be hard coded to use TV range 15 to 250 or something like this.
I was going to say this. If your capture assumes 16-235 levels but the source is full range then anything below 16 becomes full black, likewise for white. But alternatively, it could just be the maker of the composite to HDMI converter wanted the image to pop and weren't worried about losing detail
The limited range probably comes about analogously to CVBS signal anyway, there's a wider range of baseband luminance values which are all 'black' and do not result in any emission from the CRT whatsoever. I don't think it's necessarily a straight up limited vs. full range luminance issue in this instance, because the whites are full scale even in lower-contrast examples. But it could still be a remnant of the analogue pedestal. Something else to note, digital 8-bit RGB displays and outputs of video codecs use sRGB encoding which is a crude imitation of a part of CRT luminance emission transfer function. It is a sectionwise approximation, with a tiny linear section in the blacks and a big exponential section covering the rest, while the true CRT luminance function is a continuous S-curve. This is a video capture and processing inconvenience, unless you're working fully in extended dynamic range, linearised formats. If you just capture luminance voltages from your baseband and scale them to a valid range and then pretend they're sRGB values, the result is close to perfect for the vast majority of the signal except deep blacks where there's a few values of weirdness.
Great explanation! Go with what works for you! Those that are looking for simplicity with great quality understand where you are coming from. Let everyone else follow their own rabbit hole and do what works for them..
BTW, related to differing standards of transferring video data from your old timey video hardware (like VHS or console) to your TV, have you ever considered doing a video on SCART? SCART was basically 'the' TV plug before HDMI in Europe, capable of handling composite, s-video, YPbPr component and analog RGB.
@UCZpAwg9LtA2a1OwS9hrtGYQ You misunderstand the question. I'm not asking him why he's not using SCART. I'm asking him to do a video about SCART, the question was merely prompted by the mention of S-Video.
SCART is just a type of multipin connector. The signal standards sent down those pins are irrelevant, the connector is just a way of getting electrical signals from device A to device B. Of course you could use NTSC via a SCART cable, if only any NTSC hardware actually had SCART connectors on it.
I use FFmpeg and similar tools all the time in my work, in lots of different ways, and I thought the last video was great! It was a really neat solution, that produced excellent results and could be used by everyone. Some people just need to understand that new ideas & solutions like this don't have to replace old ones, they can be used along side them. Keep up the good work!
I feel your pain, I felt you made your point crystal clear in the first video. I love your up beat humor and delivery. Please don't let anyone take the wind out of your sails. You are Awesome!
Hey Alec, by looking at the footage that the capture card crushed the black levels I think it's probably related to the RGB Full/Limited Range, try capturing a regular black level test pattern and see if it still crushes the blacks or if it's only doing it with the VHS player.
you are likely correct, I have no idea but i dont care ... keep it simple - that is the fundamental issue under discussion. Great pair of videos Alec thank you
@@rogerbeck3018 yes, his workflow is quite amazing for the quality he gets out of it, but understanding what happens in the hardware *is* what this channel is about, so you know... more content :D
well im an eye surgeon, and we were looking for the best method to capture composite video digitally out of our microscopes. we tried alot of game captures, converters; you name it we, probably tried it. in my experience and you might find this weird, the best knock out pictures we found out was using DVRs for cctv cameras. the picture is so clear, sharp and vivid. ofc our dvr doesn't record audio as we dont need it. but you can buy one that does. you should give it a try and tell us your thoughts.
21:00 About Premier missing frames. It is because of the PREVIEW. Look at your preview's resolution (bottom down of screen). It says 1/2, which technically means you are missing half the lines, or in this case, half the fields. This is intended for quicker render during preview, improving the workflow. The entire image would be rendered in the final result. You can also notice that Premiere tends to output lower quality images during preview playback, but will render a better image after the image has remained still for a little while. Consider this a "preview artifact". :-)
@@ThePrinzKassad SCART was such a pain. The plug is so huge that it can be disassembled in order to fit through tight spaces. Many people also used cheap SCART cables that only transmitted Composite video. This is evident if you look at the plug: If most pins are missing, it's a cheap one. I tried to explain this to someone when I was a kid and they blanked out immediately and decided to ignore me because of my age...
@@no1DdC I know, that's why I had a Multimeter to check the continuity of everything because at some point you realize that the plug may have every available pin but there aren't connected
Bottom line: please do NOT feel you need to reply to every pedantic comment and explain yourself. You will go insane. And I like your content. You’re just the right amount of insane. Don’t worry about the nerd/peanut gallery.
By the way, I'd love to see a video on color under...seems very interesting... And here's another unwanted software suggestion : HandBrake is pretty convenient for rescaling, etc. It's like a frontend for FFmpeg, and it's GUI and very easy to use. Might be a little faster for bulk stuff as it supports presets. (I use terminal all the time and FFmpeg is still a pain in the ass, so don't worry about not wanting to use it.) Notably, HandBrake also has a Decomb option, which can help remove those deinterlace artifacts. Of course, depending on how fast you're already using your current think you probably won't need this. I just thought I might be interested in the decomb option.
Thank you so much for this :-) I was getting to start archiving all my family's videos and quickly started to see all the different rabbit holes all got frustrated I would never get this done. I did want the best method, and now this helped me realize I just want the best PRACTICAL method that is both simple, easy, and good quality. I'm glad this and the other video popped up in my feed!
Thanks for addressing the S-Video question. I have a SVHS machine, super high end thing for its time and very rare. But all the tapes I've digitised from it (through my USB TV tuner with S-Video input) have been traditional VHS format.
Your original video was very clear about the purpose. This video brings up two fascinating issues: 1) Restoration is interpretation...would you WANT a painting subdued for generations restored to original colors? It's an artistic choice that, yes, can step on the original artist's toes. 2) Utility has two legs: First, will people create (here, convert video). Second, will people consume (here, watch the converted video). Here, clearly it's better for hundreds of families to easily have their video archive that is flawed (first leg) than for one family with a video enthusiast to see the needles in the trees in the background--provided that the archive is at least good enough that it is enjoyable to watch so that (second leg) it does get watched by those families. Your method is a great balance. Note also that major release films are color/tone adjusted for each edit if not frame by frame, even when starting with professionals running professional equipment with professional lighting, etc.
I got this equipment and I’m very happy with the results so thank you Alec. Out of curiosity I went to check FFMPEG and I found it very useful. I used it to correct the aspect ratio, fix the audio to mono, because it was just on left channel, and rise volume. All on one single command per video and keeping the quality: ffmpeg -i Video.MP4 -crf 18 -filter:a "volume=2", -vf "scale=1440:1080,setdar=4/3" -ac 1 Video_corrected.MP4 To use it on Windows; search and download ffmpeg, copy ffmpeg.exe inside the folder where the videos are (or add it to the windows path variables). Then with Shift pressed right click on an empty space in the folder and select “Open powershell window here”. On a notepad copy the code and replace Video.mp4 by the name of your videos. Paste the command on the powershell window and Enter to run it. Then just go changing the correlative number. Enjoy!
“I just want something that Works, Looks Good and is correct.” Maybe add “is Simple” to that list too. This is by far the simplest method when you consider the workflow of converting all mum & dads old home videos. Thanks this is awesome!
I have been using a method like this for years now. I use a Panasonic VCR/DVD burner with HDMI out and use an Avermedia Live Gamer Portable to capture the 480p60. The audio is always just-right and the video and deinterlacing quality is practically second to none. I don't want to plug my channel, but there are examples there. I tend to avoid the up-scalers, but I am going to try it since I need a better solution than the Hauppauge USB capture devices that never seem to work and wash out the video -- this is the method I use for more difficult tapes and a Panasonic 1980s editing deck. I think it'll work great, my only concern is having to re-encode the videos with ffmpeg to return the video to a proper 4:3 resolution and set the aspect ratio. Your video was extremely well done and demonstrates methods that are, in a word, EXCELLENT, for digitizing analog video and keeping it in its proper format. Software de-interlacing sucks and QTGMC is a huge pain since the software and config you have to do for it is clunky and very slow. ffmpeg is great with yadif and scaling, though... @Technology Connections 2, if you want a demonstration on an easy way to use ffmpeg, please DM me and we can do a Hangouts session
I believe I agree with mostly everything in this video here: When I saw the initial video, to me it was obvious that this was for a simple solution that didn't involve expensive equipment, heavy fine tuning of settings per video, analysis of interlacing, or post processing to get it right. Just something that a person who understands how to use a flash drive and drag and drop videos into a video editor to combine them can understand and do. Pretty sure you even briefly explained that was your goal. As for ffmpeg, my only nitpick of a qualm there is that you referred to yourself as a non-programmer. ffmpeg doesn't require to understand programming, but that's why this is a nitpick of a qualm: it's just semantics. You're not comfortable enough with a complex command line tool is what you're getting at, and that's perfectly fine. Hell even as a programmer I get confused with ffmpeg, especially when they keep changing how it works! Anybody that insists that ffmpeg is THE way to do it, well, ask your average grandma to do it. See how well they do and get back to me. As soon as you said people suggested S-Video, I did a double take and said "wait, people think there's a lot of devices with S-Video on it?!", and then you mentioned VCRs rarely having it. So I'm in 100% agreement here. I mean, if you HAVE S-Video capable equipment, I say definitely use that over composite! But really? "Why aren't you using S-Video"??? What reality are you in that this exists so easily? Finally I half agree on the black crushing. There's probably a lot of times the black crush will be negligible. But to me if it's more noticible, I think it can be an important missing piece, perhaps as important as the colors being washed out. Short of it all: I agree! This is just supposed to be simple for everyone and not overthought. But not perfect. Just enough to archive your VHS tapes for digital video. Not a studio trying to make analogue available in digital format.
I wanted to say I think you’re a freaking legend. I’m a massive nerd in video compression and used to do everything through avsynth and virtual dub and a myriad of other options and techniques evolved over the years and spent thousands of hours trying to get the absolute best but it just ended up wearing me down. I ended up losing the interest and wasn’t converting old video or post processing anymore so I’ve been out of the loop for a while. Having recently gotten back into it I’ve been getting frustrated and looking at some old vhs-c tapes I was dismayed at the thought of having to relearn it all. It’s just such a lot of effort for no guarantee of any gain. Your technique is bloody marvy and solves my dilemmas. I unlike a few realised that your technique is sound regardless of the cable you’re using but because I have a high end editing recorder with svhs it just occurred to me to see on my hunt for the upscaler if one supported s-video and Lo and behold found one without issue. Why anyone would criticise you for not doing that on your video is just sad and it’s a shame you got hounded so hard for that as well as the other issues. People just need to have something to moan about. Nameless idiots hiding behind aliases on here and slinging shit at people doing us a great service is a shame. I found some component video to hdmi upscalers as well and wondered if that would be worth trying to find a player for but ultimately if there is one given the cost and the low quality analog signal anyway there was a clear line I wouldn’t cross. I looked up the benefits of this extra level of quality and it appeared it’s only HD content that it’s good for anyway (perhaps some studio quality gear) so there’s little point expensing effort. Every technique has compromise and you have to be willing to know when to stop. Thanks for doing the follow up as the extra details from your examples has been really helpful. I still am working on finding a good hdmi capture box here in Australia, one with good options as I’d like 30mbps or more and good quality level control so as to give me the highest possible quality for editing. Again I really appreciate your efforts and think this channel is one of the best things I’ve found on RU-vid.
Thanks for showing a simple method to capture. I agree with you that it was easier with your method. I guess there are enthusiasts out there that want to get into the nuts and bolts. It reminds me of how I started to get right down into the specifics of fine tuning an overclock on my PC years ago...yet most people would be happy just to go with the basic tweaks to achieve the overclock.
It's funny, the only thing I took away from the original video is "I'm sure there must be a device that doesn't break up the video captures." Overall, I thought your solution was genius. I always felt that vhs captures on youtube always looked a bit bleh, especially considering I DID grown up in the vhs era. I wore out my vhs copy of "Chitty chitty bang bang" when I was really young, and I sure don't remember the colors looking so washed out. Hell, I didn't get DVDs until I was a teen, and I was working for Dell's technical support when BlueRay Disc came out.
Thank you for this video. Especially the beginning. Although this particular matter means little to me, I cannot count the number times I've tried to find a way to do something and all I can find are the most technical solutions out there. I generally don't need absolutely perfect; I generally just need good and easily. Thank you for saying what I've felt so many times!
Having tried numerous and various capture methods, as well aa having numerous crt tvs and monitors, I can't argue with your results here. Simple and effective
I love what you do, I love your explanations. I know about law of diminishing returns. I am an engineer but appreciate the old phrase “Better is the enemy of good enough”!
The only thing I really need for my VHS captures is the TBC. Everything else to me is like trying to overclock a videocard to get 103FPS instead of 101FPS in games.
@@kanalnamn I came here just for a comment like this. Is it 99% common or 99% uncommon? Really hoping the first one. Nobody seems to talk about it anymore!!
@@whophd I can't really tell. Many devices talk about nonsense like "locked audio" and other made up stuff that's not real. My guess (and it's only a guess) is that most of the better DVDR recorders had line TBCs. When it comes to more recent game recording devices I think it's not as common. The game console signals really don't need TBCs. Then... the things that converts to HDMI... who knows? I've never dealt with them, and the spec sheets can be hard to find.
I purchased the devices and used them as you recommended. Other than the low audio level (easily fixed in my editor after the fact and now fixed with small audio mixer boosting the levels in the line), the results blew away the analog capture device I had been using for years. The images looked awesome. I am happy. Thank you very much. Just remember, "You can't please everyone". But, you made my day.
There might be better ways out there, but your method is what works for my dad. It's a simple solution requiring little computer skills and one touch button recording. He loves it! He's been backing up all of his old VHS home videos onto usb sticks for me to stitch together, thanks big time for your method.
Speaking of rabbit holes, the last video got me looking at what is state of the art in video processing. Like I suspected, there are a few papers using deep learning to perform somewhat better deinterlacing, but what really got my attention was video super resolution technologies. Would be pretty interesting to see how it works on VHS footage. There is demo code for both papers on github, but I'm not sure I'm motivated enough to dig up unprocessed VHS footage and hack up some scripts to feed it through these algorithms... If someone had enough motivation, I guess there is possibility for a fun paper in taking a bunch of HD footage, feeding it through a VHS and then training a neural network to reverse out any VHS artifacts.
I've thought of the idea of making a specialized VHS -> HD neural network. Considering the amount of movies available in VHS and DVD/Blu-Ray, a dataset isn't the issue. The approach I've thought of is first reversing the telecine process applied to the VHS version of the movie to obtain a 24fps image sequence, then matching that to the same frames in the HD version. For TV shows available in VHS and DVD, you could deinterlace both sources into a 50fps/60fps image sequence and match them. I think it's doable by someone with experience on AI image processing, but I'm currently unable to take on such a project.
You can buy an AI solution now . I use it, and its amazing for grainy, blurry VHS to denoise , deinterlace and upscale it. Look up Topaz Video Enhance AI .
MAN! I am at the @8:10 mark and nearly have tears in my eyes BECAUSE YOU TOTALLY GET IT!!!! In the past week, I have endured so many hours on reddit, EVEN MORE DAYS ON ANCIENT FORUM THREADS, and watching numerous guides, where ALL the creators sum up their research and advice with "You really should own a VHS player with S-Video or Firewire ports, if your childhood memories matter to you" The same guides will spend 30 additional seconds suggesting to just get an EasyCap, if I cant be bothered to hunt down a $500 professional VHS player or circa year 2000 capture device. It's such a frustrating rabbit hole of "GET THE BEST, or just settle with garbage"
Both of these videos were great. Conveniently, my dad had just asked me about converting some old VHS so I went ahead and bought the boxes you used. Excited for when they come! I might end up re-capturing my Star Wars VHS to really see the difference between the way I used to do it.
• I for one understood and appreciated the point of the video, to the point that I intend to -blow- invest the $100 for the devices so that I can digitize our home-videos before it's too late. • I too am exhausted from obsessing over quality and wondering if it's worth -spending- wasting another 10 hours to re-render the whole thing in case it might be potentially possible to maybe get a very slightly better quality or not. OCD can be crippling.
LOL. Good luck trying to curtail this dispute. This is precisely the issue driving the eternal Mac vs PC debate. Some people like to dig down into the nuts and bolts of their computers, customizing and enhancing. Other people just like it to work "well enough" straight out of the box. Which is better? That's unanswerable. It's personal preference. Always will be. You found a solution that works for you. And my mother could likely follow those steps no problem - and that's saying a lot.
I really appreciated seeing a solution that “just works”. I’ve always been a bit nervous to try converting analog video, because the topic is so overwhelming. Thanks!
Whoever convinced you that using FFMPEG is coding did you a massive disservice in either making the technology sound more scary than it is or misleading you about what FFMPEG is. Which is not to say that FFMPEG is the right tool for the job for you - your setup looks just fine to me.
Its basically coding. In that you need to be constantly looking up how to use it, the commands need to be typed correctly, and if you do it wrong you will get error messages or incorrect results for something that could just be a missing space ffmpeg is also often used in shell scripts, which is literally coding
@@huttyblue No, it really isn't. There is no logic you can put in the commands. Its just "Tell me what you have, and tell me what you want". Its not coding, its not programming, its just using software.
@@Tritium21 It is coding because you are writing code for the computer to interpret. Yes you are also using software with the code, the things are not mutually exclusive. And its more involved than tell me what you have and tell me what you want. As how you tell ffmpeg what you have and what you want has very specific syntax that impossible to guess for anyone who isn't an experienced user. Have an external reference available as well as trial and error steps for the inevitable mistakes is required to make any use of the software.
@@huttyblue No, you are not writing code. At all. Ever. Giving arguments to the executable on the command processor (cmd.exe or bash) is not giving the computer instructions to execute. You are giving the executable data. that's it. If you were to use ffmpeg in a shell script workflow or write a wrapper around the executable, it would be programming, but thats not what we are talking about. It really is just tell it what you have and tell it what you want - its a declarative model. It is NOT programming. Do not continue to insult programmers by saying using FFMPEG is programming, since it absolutely is not.
@@Tritium21 Giving a program execution arguments is no different than setting the inputs to a function when you are programming. Its text the computer interprets to do stuff, which is code and a type of programming. You really are going full gatekeeping with that last statement though.
The first 2 minutes of this video - that's exactly the message I got from your previous video. You found a solution that worked for you. Not necessarily the best solution but one that was simple and easy and whose output quality was reasonable. It came through loud and clear. Ignore the haters - you're doing just fine. I love your videos and your method of delivery is spot on. 👍
I'm late to the party here but I love watching/listening to you. You're to the point while being thorough yet easily digestible. Keep up the GREAT work!
There's a big difference between the quality needed for a professional video asset vs. one used for a RU-vid channel. You are correct that the marginality of the quality isn't worth going down the rabbit hole of pedants. Don't let commenters get on your nerves. We truly enjoy your content and personality. Keep it up!
I think most people sharing their ffmpeg knowledge or similar alternatives were not "trolling" or saying you were wrong. As a commenter of the previous video, I just wanted to share another alternative for anyone inclined. I guess that's a comment function. I liked the previous video and enjoyed it, and don't think you have to "justify" yourself in any way. It was clear that you showed "your method" and it is quite straighforward, in a way. I personally don't use Premiere or Final Cut and find them quite daunting and expensive pieces of software. A cheap linux netbook attached to my media center is my tool of choice, and the "amount of control" discussion is not always about "getting that .5% increase", but indeed about cases like buying some capture device out of Amazon, finding it does whatever it wants with the footage and spending extra time on post processing and/or returning it and buying a new alternative. You mentioned removing those black bars: imposed 16:9 format is one case of "lack of control" from this point of view, quite more urgent that "what is your remuxing algorithm". I really love your content and the quality of your videos, though sadly you make this follow-up sound more like a rant about the comments than your usual "let's discuss this subject further".
I'm in the same boat as you, and left a similar comment. Just because some people have some other methods they suggest to get a similar effect doesn't mean they were trying to crap all over what TC was trying to do. FFMpeg is free, runs on virtually anything, performs, and works well as an ingredient of a larger utility, while also allowing you basically any level of control that you'd like. It's definitely not for everyone, and hardware encoders definitely have their place, but I really don't see why TC was so surprised and annoyed that a lot of engineers (presumably an audience he's trying to attract in his videos) are going to have alternative ways of achieving something similar.
Hello, just to say I'm late to the party and have only just recently been binge watching all your videos. There's some great material, thank you! I've watched both your videos on VHS conversion with interest. For the past 10 years I've owned a Panasonic VHS and DVD combo unit which I originally used to burn copies of my old VHS tapes to DVD so I could watch them on my DVD player. Obviously technology has moved on and I decided to convert to MP4 instead. The Panasonic unit has a built in upscaler as well as an HDMI output. I basically connect this to an HDMI to USB3 capture card and use ORS software to capture on my computer. I configure all the necessary settings within ORS (e.g. Aspect ratio, resolution etc). The process works a treat. I occasionally use Premiere to edit the output and then upload to my Vimeo account so I can keep the files for posterity as well as have the benefit of being able to share the videos with family members. Please keep up the great work! Best wishes from London, Jonathan
Hey Alec, just to let you know, your quick easy method with the upscaler is allowing me to archive two decades of broadcast TV recordings that my father made, from the late 70's through the late 1990's. granted i'm largely automating it with an old industrial Sony S-VHS player with an RS232 control interface, and one of those Chinese HDMI over IP converters people wrote a Linux kernel driver for, just in the interest of saving time. all I have to do is load tapes and name the output files. I'll probably dump all the scripts on github later if anyone is interested. though the VHS player controls is going to vary from ine industrial VCR to another, it should be simple to adapt...
Basically, with his method, you lose some detail on the things that don't matter, things in the background or in dark shadows. But in return all the things that do matter, the things you actually filmed to be able to watch later looks a lot better. Seem like a fair tradeoff too me.
I loved how simple and easy your idea for the quality is. I went out and ordered what you suggested and now waiting for them to arrive. Like you I tried to wade through all sorts of methods and software and gave up due to the hassles and time it takes. Thanks for figuring this out.
I really appreciate the original method. Because in my head I have 2 voices: 1. A technical perfectionist nerd that likes to go for "the best" and 2. A very lazy pragmatist, and this voice tends to say "whatever works" and "I don't have time to go down that rabbit hole" ... So I really appreciate your simple process, that still feel professional.
No, the bandwith is not sufficient. To get an idea, most audio soft captures 44100 samples per second... and even the color in the most washed out VHS tape has its Shannon limit at 3579540 "samples" per second
If you want to record raw video out of the tape you want to read a bit about Domesday86 VHS-Decode that I am using too (warning, it's for perfectionists and you have to use linux).
I would love to go to a weekly lecture taught by this guy. It could honestly be about anything because his knowledge and more importantly passion for the inner workings of things that we don't give a second glance in our day to day lives is something I have been coming back to more and more lately and it is just very pleasing if not addictive in some way. Love your channel man, thanks for all your effort and dedication to teaching some random man from NH more than he ever thought he'd learn about many many things
your method is just easy and makes good looking video? for the average person who wants to copy their home tapes from vhs to digital video, it honestly works perfectly fine. Once you have it set up, it records at the push of a button and yea okay for longer things you gotta stitch Em together but that's honestly the hardest part?? and it's not that hard once you know how??? the pursuit of perfection is never ending and so long as you keep digging, the rabbit hole will keep going. You found a method that not only works, but works for you and your needs and thought to share it with everyone so people can enjoy digital copies of their tapes without them being washed out.
Hey, don't let the twerps in the comments get to you! Quick and dirty solutions have their place. Keep up the good work, I'm proud to be a Patreon supporter.
I found your video to be extremely helpful for my purpose. Your description seemed pretty clear to me. I very much appreciate your work, thanks for all the great content.
I can't figure out how to do a reply of my own (I never usually comment on a YT video) so I'll leave what I have to say here and hope it is seen. The video capture method he is using is mostly fine and on the right track. His problems with what he calls "black crush" is actually caused by an RGB + RGB partial mismatch. see this YT video for a good explanation ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-7kjZZNT5js4.html You can also see the difference with a nvidia card on HDMI connections on the "change resolution" page of the driver settings change the "output dynamic range" drop down between "full" and "partial" to see the difference. What you need to avoid this is to set the HDMI capture device to capture as RGB Partial (could be called "console" or "tv" mode on gaming capture types) and not to RBG full. If the HDMI capture device isn't capable of doing this, get one that can.
As a devout follower of the KISS (Keep It Simple Stupid) philosophy, and also having a large collection of crappy video digitizing hardware, I thought your solution was both elegant and clever, and seemed to solve the same problem I've had over the years, and I plan on duplicating your setup before transferring all my analog video to digital. Thank you, thank you, thank you.
Surely the important thing to consider is WHY the video is being captured. The impression I got was that these captures are just for the purposes of extracting clips to edit into other videos. In that case, who cares if there is loss of detail and it's not the absolute best possible capture in the world! What matters is that it satisfies the creative intent of the person editing the video. Looks good? Done. (Incidentally, that was some great camera work back in the day, for an amateur video.) If the video is being captured to archive and preserve the videos, then yes, you might want to be a bit more careful about not losing things in the process. But if they are just family video clips, of limited general interest, then really it's nobody else's business how well they are preserved. At least the videos are being preserved at all! The recordings could just as easily have been destroyed by poor storage conditions or simply thrown away in a clear-out. About the only time I can imagine the general public having a justifiable reason to be concerned about the way somebody else's VHS tapes are being digitised is if the tapes are of enormous historical significance or if they are being digitised using public funds. It's not reasonable to insist that every scrap of video footage in existence be hoarded and preserved with the same level of care and diligence as you would expect for, say, the footage from the Apollo missions or a well-known music video. Most of it is simply not worth the effort. People need to learn to let go. Future historians scanning through some nobody's home videos from 1996 are probably not going to be saying "Wait! Go back! Enhance! Goddammit! Somebody used an inappropriate capture method in the 2010s and lost the detail! Now we'll never know what carpet patterns were popular in badly-lit corners of the 90s."
Yes, on point! These were also my thoughts throughout the video. I think it's important to differentiate the goal of digitisation of depending on if it's for demonstration and/or archiving purposes. But I can sympathize with those who doesn't have the time and/or energy to take that extra mile for some marginally better results as it can get pretty ridiculous how much effort one has to make to achieve said results. As long if you're not paid well enough to do so or have a genuine interest in general that is. With that said, I can strongly recommend StaxRip, as it has almost anything and everything that one would need for re-encoding videos with all kinds of filters etc. No need for writing lots of scripts/commands as the GUI covers most of that already by giving its user clickable settings, simple textboxes for some manual adjustmemts and much more quality-of-life implementations. Even though I use FFmpeg as standalone for alot of things, my go-to is almost always StaxRip to streamline batches of encodes which saves me tremendous amounts of time 😁
I am one of those people who want the best transfers I can get from my tapes. And by best, I mean the best I can get after investing no more than a few hundred dollars and tens of hours optimizing the process. If I want to extract the maximum detail possible, I’ll send that tape to a archival-focused digitization service and spend hundreds of dollars for a top-notch, fully time-base corrected transfer of the contents of that tape. I will do this for my family’s recordings, because I value getting 99.9% of the detail that is currently on the tapes, including faithful restoration of damaged / degraded segments. I don’t need to spend thousands of hours to become an expert at analog video restoration in a digital environment to digitizing 8 hours of personal & family memories from 6 tapes. For my TV recordings, I am transferring them with the equipment I currently have. My goal is to have the means to make faithful transfers of countless hours of home recordings. I will spend tens of hours and a few hundreds dollars getting my analog-side setup “good enough”. I have hundreds of tapes, and plan to transfer peoples collections by the box full. You never know, you might have the last surviving copy of a local TV broadcast, and I to rescue as many of these recordings as I can. In the future I will scale up to a professional rackmount system, one that fully automates the analog-side & digital-side processes. Tapes go in, files come out. I’m not currently at that level of wealth. I can afford to spend days to fine tune, not months. My next purchase is a shielded audio cable, and a RadioShack style ground-loop isolator (this specific one: www.amazon.com/gp/aw/d/B077Y5DLBB/). to cut perceptible noise out of the signal going into my Elgato USB analog capture device. I found the upscaler & s-video signal amplifier I bought to have a diminished picture. Not only did the amp and upscaler amplify graininess, it cropped the image by a few percent, as if it zoomed in slightly. This cropping does not happen with output from my Elgato capture card using Elgato’s included software. I would recommend the Elgato capture card as it does a fantastic job in interpreting the analog signal, especially with S-Video. I will warn you that you do need at least a mid-spec PC to capture HQ lag-free analog video from a capture card. I use Elgato Video Capture that is free when using their capture card, at “High Resolution” with “Preserve Source Format” checked. This capture card has produced the best quality transfers from my tapes to date. I would take great pleasure in spending countless hours perfecting a method if: -I were to get into the business of archive-grade magnetic tape transfers. -I were to bring forth a new all-in-one prosumer-grade VCR product onto the market that “Captures at least 97% of the details present on any VHS tape”. -Develop a mid-tier process improvement for businesses / companies (because the top-end is probably already served well enough).
I just finished my first 8MM --> Digital transfer following your instructions. This is a 25 year old tape and the results were fantastic. I had to do it in two steps because my RCA to HDMI didn't have the setting to not stretch the video. So I used AVS Video Converter to mush it back to 720p with the black bars on the sides. I played the original VHS on my TV and then the MP4 from my laptop. Honestly, I like the double converted MP4 better. 18 tapes to go. Christmas present for my wife. Thanks!
Or use the black boost (or whatever it was called) and throw a gamma filter on after that. But, again, you're adding complexity to a process where he wants to minimize complexity.
