@@ItsBHE Just find the SSTV Encoder and Robot 36 apps seen in the video, find some images and enjoy. (I got them off some apk download sites, as I'm not in the habit of putting Google accounts onto what is essentially burners, and at least one of them might not even have Play services installed.)
SSTV was used in Portal by Valve as part of a new achievement/hidden ARG. When you brought the seemingly randomly placed radios to certain areas they played an audio tone that you could feed into an SSTV decoder to generate images relevant to the ARG.
that was my introduction to SSTV and i thought it was so amazing, and the fact one of the images had a phone number that was a dialup FTP run on a PC sitting on the kitchen counter of one of the Portal devs (iirc) was equally as cool
@@ch12ua yes i think it had copies of all the images from each radio/location in-game and some extras which were all related to hinting at portal2 and its release date
It is almost unfair to task viewers with both the calculation of the bitrate of the recorder and drooling over the Curta at the same time. I can’t multitask these.
The mid 00s Sony digital handheld recorder brings back memories. I used to bring one to my lectures and record them with a directional microphone pointed at the professor speaking. Each class got its own memory card. I would upload them to my iBook G4 and make them into 128kb MP3s (more than good for just public speaking recordings) and organise those into playlists, I would listen to those playlists constantly on my iPod and in my car (using this cassette tape looking adapter that gave me a AUX line in). I got my undergraduate degree by memorisation 😂
Yeah, something that most people don't understand is that compression is all about probabilities. Every compression algorithm has, implicitly or explicitly, a *prior expectation* about what its input should look like. Inputs that match that expectation well compress well. Inputs that don't match that expectation compress poorly (for a lossless algorithm, they take a lot more bits; for a lossy algorithm, they get reproduced poorly). So if you develop an algorithm that has a very high compression ratio and is very highly optimized for speech, that means it "wants to hear speech" no matter what. Give it music, and you get a garbly mess that sounds kind of like a room full of people having a million conversations. Give it random noise, and you get something that sounds *more like speech* than an uncompressed recording would. It will sound like whispers, or ghosts, especially if that's what you want to hear. Although that Sony codec is proprietary and undocumented (which is typical Sony), being LPC makes it a cousin of many codecs that have been used in cell phones (original GSM), VoIP apps, and older streaming formats like RealAudio. I suspect the reason LPC makes such a hash of SSTV is that it has decent frequency reproduction, but it's subject to "temporal smearing" - which not only blurs the image horizontally, it also makes the sync pulse hard to localize (like the tape flutter only 10x worse), which means that successive lines hardly ever align with each other.
I was looking at the pixel spread and thinking about the 'long-term' in the codec name. Thanks for that detailed description. That clears a few things up.
Slow Scan TV is super popular on ham radio. People send photos all over the world from their house radio stations or even in the park, with a wire tossed up in the tree and the radio operating off of battery power. 73 de WU2F
My dad did ham radio all his long life. However, he was never into SSTV - he thought it was rather silly and preferred just talking😄 73s from the son of VK6VU (and in the 70s VK3 AWN)
I mean its basically a modem. There was a virus back in the day where a computer, with the virus, would transmit a sound over speakers, and any computer with a microphone could hear it and get the virus. Like an airborne computer virus. I am fascinated with the idea that these speakers can produce and microphones can hear outside the range of humans. So you could give, for example, Alexa commands that humans can not hear.
This explains why audiophiles love "lossless" music files. You can really see what is lost with this experiment! But also, how good mp3 128 looked. So some lossy codecs maybe "Good enough" for good sound quality. .Wav or .Flac (lossless) would give the best results. Super interesting. Great video.
Exactly. I didn't realize how much these things make a difference until my Marshall Monitor headphones broke and my phone broke. My replacement earbuds are good for the price but can't handle lossless audio or better codecs and the replacement phone doesn't have the greatest codecs either so while the speakers themselves are very ok, the digital bottlenecks rob the recordings of their "high resolution" and with it erase a lot of detail. There's Jacob Collier tracks I can't really enjoy anymore because the elegant, gorgeous strings in the background are suddenly all flat fiddledies in a vacuum.
"Audiophiles" would indeed prefer lossless to 10kbps pre-Mp3. What? Lossy is perfectly fine. ~256kbps Opus is good for anyone, 320kbps is probably indistinguishable from source. Lossy is much smaller than lossless. Use lossless for masters and high quality archival, use high-bitrate modern lossy for playback.
UA-cam has been recommending this video to me for a week now and I finally watched it. Hands down one of my favorite videos on here...love what you're doing man! Something that came to my mind while watching - I think the sample rate (not necessarily just the bit rate) of the audio might have a lot to do with how accurately the audio is captured and then reproduced. The sample rate of the older digital recorder is very low, so it's not capturing the full picture. The newer recorder is capturing at 44.1khz, so it manages to get a much more accurate sample of the image. This is such a fascinating topic!
I'm really glad you enjoyed this. I had lots of fun doing these experiments. I'm pleased that I get to share then with others. Thank you for letting me know.
When I saw that thumbnail I thought you're going on a journey into that weird time in the voice recorder industry when everyone tried to push features into them super hard. As a result I used to have a voice recorder from Olympus in 2004 which did indeed have a camera and take pictures! You couldn't listen or view them but you could drop them onto your PC after the fact
This is quite fascinating. You can encode and decode anything in any form. It was also fun to see an analog calculator. As retrospective as it can be. It's truly amazing.
I'm glad you enjoyed seeing this :) I'm amused you say analog calculator, because technically it is digital in the way it works. But because the gears and levers in the mechanism are so large, the underlying analog layer is extremely apparent to us humans. Unlike our phones and computers where the analog world of electrons energy states and waveforms that support the digital layers above them are so small we don't ever think about it. I really enjoy exploring the boundary between the analog and the digital, hence this very video about saving analog sstv signal on a weird codec digital recorder!
@JanusCycle i think one of the reasons for me to watch the stuff you're uncovering in the channel is information or knowledge that many may be lacking. As an electronic technician, i would definitely say that the videos you post always seem to be on point and educative to others.
Not the same but when i was doing the final project for a degree in IT back in 2009 i didn't have an internet connection at home, so me and my friends send each other word files compressed into .rar, added them at the end of picture files using the '+' command on CMD to make them into images you could send through MMS. Felt like hackers
I tried recording with an old cell phone and it worked really well. And that. We in 2024 making comparisons and changing settings of both applications is very difficult. Imagine the time in which this technology was created. wonderful video. New subscriber from Brazil.
Utilizing mechanical sound waves instead of electromagnetic waves for data transmission isn't that novel but being able to hear the signal is an intriguing thing to behold regardless. Love the recorder.
3:05 That wavyness is not necessarily because of the tape; A bit of it is because of "Multipath" noise! Since you're sending it out on a speaker instead of direct cable, The sound 'echoes' back to the receiver at multiple different times. EDIT: 10:30 Quite a lot actually, SSTV is very sensitive to stuff like harmonic distortion and frequency response so it could very well be most of the quality loss. 14:13 This absolutely would be multipath effects, You can see how the recorded one is sort of 'ghosting' to the right where the direct connection has none of that, That's multipath.
Sounds like those ACELP? Codecs from US cellphones in the late 90’s and early 2000’s. They model the human trachea with a filter, the input being seudo white noise. Then they just send the filter coefficients. On the other end an actual filter is implemented. Yields crazy low bit rates, and very good human voice audio. Will not model anything else with good quality.
It may be more robust to transmit a digital image file using a low baud rate FSK modulation mode, but you'd likely need to reduce the image resolution significantly, and use a lossy compression codec
First time seeing this channel. I always wanted to do this as a kid (90s peak recorder era) but forgot about it, saw the thumbnail. And you sealed the subscribe button once you busted out the pepper grinder
The first MP3 player I ever owned (a Creative Nomad, the tiny little thing with a SmartMedia slot from 1999, not any of the later devices they slapped the same name on) was also a voice recorder. It used 32kbps ADPCM for the voice recording mode, which probably would handle SSTV just fine because there's nothing very "clever" about it.
Excellent! I did some experimentation once around streaming ZX Spectrum tapes from the web. Looking at the source, I went with ''-q 5' for ogg vorbis (around 160 kbps) and '-V 1.5' for mp3lame (which I think offhand is almost 256 kbps). Psychoacoustic compression is really terrible for encoding slow audio bits - the tag line for the site was "turning kilobytes into megabytes"
How are you so knowledgeable about stuff like these? Nothing hinders you and just always comes up with possible solutions, converting stuff, using different devices and methods, and calculations. It's crazy.
Very kind of you to say, thank you. But there are many things I don't know. Simple things like how to use an Apple Mac properly. I also make many mistakes and get things wrong sometimes. I need to show more of this. Because allowing yourself to make mistakes is where the best learning comes from.
@@JanusCycle As a millennial who appreciates the old and new technology, I feel like a kid once again, watching in awe, the nostalgic devices you post and how you modify and customize them. So much memories.
Great idea! It allowed us to SEE the difference in sound quality. I wondered why I disliked the low bitrate lossy compressed sound (like in MP3) so much, compared to just lower bitrate uncompressed sound or analog sound with limited bandwidth. Cut the bandwidth alone and the encoded image will just be more blurry. Use lossy compression with too low bitrate and you get a huge amount of noise and artifacts that makes the original image nearly unreadable. Weirdly, the speech remains intelligible.
Some of these distortion patterns look awesome. I gotta give this a try myself if I can find a good recorder for it. Imagine how cool it would look if you ran every frame of a video sequence through one. It would look great played back on a little screen for a grungy scifi/cyberpunk movie.
That RLM video mentions how unserieous it is to use voice recorders for ghost hunting, you should use something designed for raw CD quality PCM like the Zoom recorder they used. I want to believe myself but not enough to find it buried in compressed artifacting and audio pareidolia. I recorded a great example of this from a Discord stream of my friend using a shop vac. The vacuum cleaner sounds like it has voices in it because that's what the encoder is looking for, it's eerie!
I still think this voice based codec would be useful for 'hearing' EVPs. And I think Mike would have fun listening to these sort of recording artifacts until Rich calls out his bullcrap.
That last recorder worked best. I also get those apps on every device I can. Sstv encoder and robot 36 are a companding set that go together. Scottie dx is the only one that can do my profile picture without ghosting.
I've often had this idea but the one thing that has put me off this idea is the fact that every voice recorder I've come across hasn't got an input jack, which not only means that I have to listen to the hellish noise of the SSTV for the entirte length of the recording, but I also have to contend with the analog distortion of the microphone (not distortion in the usual sense, but the frequency response of the mic altering the final signal). although I have toyed with both SSTV and NBTV onto audio cassette. there is even an NBTV encoder that supports color, though finding an encoder that will take mp4 files has been quite the task, and I don't quite understand the color encoding myself to be able to make my own encoder. though I have made one for some of the monochrome ones, and I may in the future make my own standard that interlaces the frames for a balance of framerate and resolution.
I've done a bit of NBTV experimentation. But I had to play the mp4 on Windows and use NBTV software to screen grab it. I wish there was NBTV software for Android.
So what you did there is kind of demonstration of analog vs digital. Low quality analog is better than high compression digital. Also; considering NASA didn’t have to take cassette speed wobble and flutter into consideration. But other things like distance and weak signal. I’ve heard this is the same technology the voyager missions use to send pictures from far away planets back to Earth.
Nice to see you working a Curta to calculate things. Never knew tat such mechanic calculators are doing that job with sucn large numbers so accurate. 🙂
Compression maybe-- it might technically have the bandwidth it claims, but not at uniform resolution. The algorithm was probably tuned for human voice.
Bruh whipping out a curta calculator is a huge (rich) nerd flex. I only know tech moan has one. You should do a separate Tutorial and review video on the curta if you haven't already, such a fascinating piece of mechanical engineering technology. Edit: oh wait nevermind i saw your thumbnail on the curta calculator, ill watch that too!
The mp3 192kbps looked flawless! I wonder if there would be any differences at all compared with a lossless recording. It seems that beyond a certain threshold, the SSTV signal gets completely transported.
That’s an amazing technology this is responsible of all the images and data sent from the moon, mars and all those deep space missions like the voyager, and it’s still in use helping sailors to have a daily forecast weather report, really useful for vessels that won’t have access to satelital services for any reason, and a great chance for radio enthusiasts to experiment on data transmission
Using Curta to calculate is a big flex especially if one has fixed it... put a big smile on my face this one!
@@JPkerVideo obviously he’s more of a wizard and will calculate precisely when he means to.
Once again, I grabbed two random Android phones and started making them yell pictures at each other for no reason than the enjoyment of it.
How to do it
@@ItsBHE Just find the SSTV Encoder and Robot 36 apps seen in the video, find some images and enjoy. (I got them off some apk download sites, as I'm not in the habit of putting Google accounts onto what is essentially burners, and at least one of them might not even have Play services installed.)
@@ItsBHEhe stated the apps in the video Robot36 and a SSTV recorder
😂🤣 😝 ROTFLMFAO
I just did that lmao
SSTV was used in Portal by Valve as part of a new achievement/hidden ARG. When you brought the seemingly randomly placed radios to certain areas they played an audio tone that you could feed into an SSTV decoder to generate images relevant to the ARG.
It was SSTV?! That's really cool
that was my introduction to SSTV and i thought it was so amazing, and the fact one of the images had a phone number that was a dialup FTP run on a PC sitting on the kitchen counter of one of the Portal devs (iirc) was equally as cool
@@kneel1 did someone save anything from the pc?
@@ch12ua yes i think it had copies of all the images from each radio/location in-game and some extras which were all related to hinting at portal2 and its release date
Came here to talk about SSTV
This is why i love youtube. It's the only place where you can find interesting things like this
Loved the unexpected Curta Calculator Cameo
Respect for using a freaking Galaxy Note 7 😂
Nice spotting. Not sure if anyone would realize.
i thought it was one but why would you be crazy enough to use it? Turns out I was wrong
@@JanusCycle i wonder if they check for this phone at airport
But it you want to fuly charge it you can change battery that been fixed ( new)
6:55 it sounds like from jigsaw movie I WANT TO PLAY A GAME thats what you neded to say
When he said "I really don't want it to blow up" cracked me up!
I love the casual use of a portable mechanical calculator as if that’s a totally normal thing to use
I've never wanted something more in my life
It is almost unfair to task viewers with both the calculation of the bitrate of the recorder and drooling over the Curta at the same time. I can’t multitask these.
Meh, it's kinda cringe.
The mid 00s Sony digital handheld recorder brings back memories. I used to bring one to my lectures and record them with a directional microphone pointed at the professor speaking. Each class got its own memory card. I would upload them to my iBook G4 and make them into 128kb MP3s (more than good for just public speaking recordings) and organise those into playlists, I would listen to those playlists constantly on my iPod and in my car (using this cassette tape looking adapter that gave me a AUX line in).
I got my undergraduate degree by memorisation 😂
I'm glad you made it through :)
"I wanna put this into my 'Top 10 test modes of all time' list."
This channel is truly one of a kind. Never change. 👍
I wholeheartedly agree
Yeah, something that most people don't understand is that compression is all about probabilities. Every compression algorithm has, implicitly or explicitly, a *prior expectation* about what its input should look like. Inputs that match that expectation well compress well. Inputs that don't match that expectation compress poorly (for a lossless algorithm, they take a lot more bits; for a lossy algorithm, they get reproduced poorly).
So if you develop an algorithm that has a very high compression ratio and is very highly optimized for speech, that means it "wants to hear speech" no matter what. Give it music, and you get a garbly mess that sounds kind of like a room full of people having a million conversations. Give it random noise, and you get something that sounds *more like speech* than an uncompressed recording would. It will sound like whispers, or ghosts, especially if that's what you want to hear.
Although that Sony codec is proprietary and undocumented (which is typical Sony), being LPC makes it a cousin of many codecs that have been used in cell phones (original GSM), VoIP apps, and older streaming formats like RealAudio.
I suspect the reason LPC makes such a hash of SSTV is that it has decent frequency reproduction, but it's subject to "temporal smearing" - which not only blurs the image horizontally, it also makes the sync pulse hard to localize (like the tape flutter only 10x worse), which means that successive lines hardly ever align with each other.
I was looking at the pixel spread and thinking about the 'long-term' in the codec name. Thanks for that detailed description. That clears a few things up.
I'm getting Posy vibes from this video, I like it.
I'm in great company then :) I'm honoured, thank you.
+1
Agreed
Slow Scan TV is super popular on ham radio. People send photos all over the world from their house radio stations or even in the park, with a wire tossed up in the tree and the radio operating off of battery power. 73 de WU2F
My dad did ham radio all his long life. However, he was never into SSTV - he thought it was rather silly and preferred just talking😄 73s from the son of VK6VU (and in the 70s VK3 AWN)
i think ive heard these transmissions before when ive used those online tuneable radio receivers, i forget their proper name atm.
@captaineldeezee1336
webSDRs
@@captaineldeezee1336 WebSDR
I mean its basically a modem. There was a virus back in the day where a computer, with the virus, would transmit a sound over speakers, and any computer with a microphone could hear it and get the virus. Like an airborne computer virus. I am fascinated with the idea that these speakers can produce and microphones can hear outside the range of humans. So you could give, for example, Alexa commands that humans can not hear.
Here to officially request the _"Janus_ _Cycle_ _best_ _of_ _test_ _mode_ _sounds_ _list"_ please.
Me too
+1
Can we run doom in it?
Surprisingly good results for mp3 codec.
This explains why audiophiles love "lossless" music files. You can really see what is lost with this experiment! But also, how good mp3 128 looked. So some lossy codecs maybe "Good enough" for good sound quality. .Wav or .Flac (lossless) would give the best results. Super interesting. Great video.
Exactly. I didn't realize how much these things make a difference until my Marshall Monitor headphones broke and my phone broke. My replacement earbuds are good for the price but can't handle lossless audio or better codecs and the replacement phone doesn't have the greatest codecs either so while the speakers themselves are very ok, the digital bottlenecks rob the recordings of their "high resolution" and with it erase a lot of detail.
There's Jacob Collier tracks I can't really enjoy anymore because the elegant, gorgeous strings in the background are suddenly all flat fiddledies in a vacuum.
@@schmuiare you feeding lossless audio into the phones or a lossy source? There might be a generation loss or two, reducing quality further.
"Audiophiles" would indeed prefer lossless to 10kbps pre-Mp3. What?
Lossy is perfectly fine. ~256kbps Opus is good for anyone, 320kbps is probably indistinguishable from source.
Lossy is much smaller than lossless.
Use lossless for masters and high quality archival, use high-bitrate modern lossy for playback.
@@schmui Your earphones sucking have nothing to do with lossy codecs. For playback high-bitrate lossy formats sound the same as lossless.
@quarteratom Bluetooth headphones create another step of generation loss, especially those that use the SBC codec only.
I just found your channel, and I consider it a treasure. It deserves millions of views. A big thank you from Morocco!
Well thank you for watching. I'm glad you are here.
And I would like to say hello to everyone watching from Morocco :)
I'm Moroccan Also... biiig fan 🎉 @@JanusCycle
UA-cam has been recommending this video to me for a week now and I finally watched it. Hands down one of my favorite videos on here...love what you're doing man! Something that came to my mind while watching - I think the sample rate (not necessarily just the bit rate) of the audio might have a lot to do with how accurately the audio is captured and then reproduced. The sample rate of the older digital recorder is very low, so it's not capturing the full picture. The newer recorder is capturing at 44.1khz, so it manages to get a much more accurate sample of the image.
This is such a fascinating topic!
I'm really glad you enjoyed this. I had lots of fun doing these experiments. I'm pleased that I get to share then with others. Thank you for letting me know.
When I saw that thumbnail I thought you're going on a journey into that weird time in the voice recorder industry when everyone tried to push features into them super hard. As a result I used to have a voice recorder from Olympus in 2004 which did indeed have a camera and take pictures! You couldn't listen or view them but you could drop them onto your PC after the fact
Interesting, thanks for mentioning that.
@@JanusCycle I've even found an old pic of it, and it was called the Olympus W-20. Very sleek bit of kit even in the modern day
Top quality as always. Thank you for posting :)
This was so frakking cool. Thank you for making the video.
I'm pleased you enjoyed this :)
This is quite fascinating. You can encode and decode anything in any form. It was also fun to see an analog calculator. As retrospective as it can be. It's truly amazing.
I'm glad you enjoyed seeing this :)
I'm amused you say analog calculator, because technically it is digital in the way it works. But because the gears and levers in the mechanism are so large, the underlying analog layer is extremely apparent to us humans.
Unlike our phones and computers where the analog world of electrons energy states and waveforms that support the digital layers above them are so small we don't ever think about it.
I really enjoy exploring the boundary between the analog and the digital, hence this very video about saving analog sstv signal on a weird codec digital recorder!
@JanusCycle i think one of the reasons for me to watch the stuff you're uncovering in the channel is information or knowledge that many may be lacking. As an electronic technician, i would definitely say that the videos you post always seem to be on point and educative to others.
Maaaan that is so cool. At the end where it all comes together like the slowest cable TV you've ever seen - beautiful.
I'm glad you had fun :)
Small button-operated battery-driven segmented-LCD devices are so cool. They are designed to do one thing well. Segmented-LCD is so energy-efficient.
Oh yeah, more importantly, thank you for making excellent
intelligent and interesting videos on obscure topics.
The International Space Station ISS is also sending SSTV some times. Thanks for sharing.
I tried to find out to see if I could get the signal. Then I missed the December 2023 passover.
@@JanusCyclePassover in 2023 was in early April, your Rabbi must be extremely disappointed in you
@@RaytheonTechnologies_Official Its currently sending SSTV out, been doing it since Tuesday and ends on Monday
@@Ben1551 the festival of Passover is unable to do that, you must be picking up signals from a different Jewish festival
If I had used two separate words pass and over this may not have happened. But I did enjoy reading this.
This is the kind of videos that Blows Up on UA-cam after 15 years... not because it features the Legendary Note 7.
Love the technical detail while still appreciating the feel and style of old gadgets.
THis is why I watch youtube. Channels and Videos like this. The most random fascinating stuff. Keep it up !
I had a lot of fun doing this. I'm glad you enjoyed it :)
So uhhhh... I think this is how I'll be making all of my album art going forward... Really neat!
This will work great for my project. A setup with an old portable TV/Radio fed with a Minidisc player to display images.
This is my new favourite way to test wow and flutter in tape decks I’ve repaired 😂
Not the same but when i was doing the final project for a degree in IT back in 2009 i didn't have an internet connection at home, so me and my friends send each other word files compressed into .rar, added them at the end of picture files using the '+' command on CMD to make them into images you could send through MMS. Felt like hackers
That is awesome! I love hearing about this kind of stuff, thank you.
Mad respect for owning a Note 7, I'd recommend putting a note FE battery in it as even if it's not fully charged, the battery could still blow
It's the danger that attracts me to using it
I tried recording with an old cell phone and it worked really well. And that. We in 2024 making comparisons and changing settings of both applications is very difficult. Imagine the time in which this technology was created. wonderful video. New subscriber from Brazil.
international space station sends picture to earth all the time for hams.
Utilizing mechanical sound waves instead of electromagnetic waves for data transmission isn't that novel but being able to hear the signal is an intriguing thing to behold regardless. Love the recorder.
5:40 - Yes, please.
I love everything about this video and I want to buy all the things you used here. Extremely interesting and satisfying.
Neat to literally see just how awesome mp3 encoding is
He has a Curta calculator...my eyes are blessed
That, coupled with the Note 7. I don't want to get out of this video
The idea seemed strange to me at first, but then I remember that modems existed and were often used to transmit pictures.
3:05 That wavyness is not necessarily because of the tape; A bit of it is because of "Multipath" noise! Since you're sending it out on a speaker instead of direct cable, The sound 'echoes' back to the receiver at multiple different times.
EDIT:
10:30 Quite a lot actually, SSTV is very sensitive to stuff like harmonic distortion and frequency response so it could very well be most of the quality loss.
14:13 This absolutely would be multipath effects, You can see how the recorded one is sort of 'ghosting' to the right where the direct connection has none of that, That's multipath.
Sounds like those ACELP? Codecs from US cellphones in the late 90’s and early 2000’s. They model the human trachea with a filter, the input being seudo white noise. Then they just send the filter coefficients. On the other end an actual filter is implemented. Yields crazy low bit rates, and very good human voice audio. Will not model anything else with good quality.
The old Speak & Spell toy used this kind of linear predictive coding for storing vowels and consonants.
When I was a kid, Sony made all the coolest stuff
Cool, expensive, desirable.
A freaking mechanical calculator wtf lol
my old metronome had a calibration mode documented, but if you held some extra buttons the test mode would come up and play a melody
This sounds intriguing
Joy to watch. Thanks for real inspiration
it would be cool to see how well the tape recorder works through the direct connection
Me too
It may be more robust to transmit a digital image file using a low baud rate FSK modulation mode, but you'd likely need to reduce the image resolution significantly, and use a lossy compression codec
First time seeing this channel. I always wanted to do this as a kid (90s peak recorder era) but forgot about it, saw the thumbnail. And you sealed the subscribe button once you busted out the pepper grinder
Thanks for watching this. I hope you find some interesting videos on here :)
Yeah I watched quite a few already. I do have to ask what’s your day job? Software engineer?
I was working in IT full time, now I'm part time doing other things.
@@JanusCycle hell yeah, sharpening up your pepper grinder to shred numbers clearly
That is good idea to send congratulations postcard to my friends, just record this kind of trek :)
Second recoderd is digital so it uses compression, sstv is for analog devices
The first MP3 player I ever owned (a Creative Nomad, the tiny little thing with a SmartMedia slot from 1999, not any of the later devices they slapped the same name on) was also a voice recorder. It used 32kbps ADPCM for the voice recording mode, which probably would handle SSTV just fine because there's nothing very "clever" about it.
I'm so happy a random Northern Exposure clip wound up in here with Ruth-Anne Miller smashing Maurice's device. I miss that show!
Ended up
Cicely is such a beautiful town. I live there in my mind whenever I see an episode :)
We use digital SSTV in the amateur radio world. It would be interesting to see how these fare under a voice optimised CODEC.
the moment i clicked here, i knew i was gonna see the legendary SSTV tech being used
Why not just swap out the battery with a "fixed" small size battery and flash a ROM that disables the warning/bricking?
You deserve every single second of my time ❤
Excellent!
I did some experimentation once around streaming ZX Spectrum tapes from the web.
Looking at the source, I went with ''-q 5' for ogg vorbis (around 160 kbps) and '-V 1.5' for mp3lame (which I think offhand is almost 256 kbps).
Psychoacoustic compression is really terrible for encoding slow audio bits - the tag line for the site was "turning kilobytes into megabytes"
Of course. We used cassettes to store programs, codes, and images are codes.
This may be an interesting way of visualizing sound quality loss by different phone service providers.
i wasnt expecting it to be in color.
also, it feels as slow as downloading a picture in early 2000
Whipping out a Cutra was a massive flex
New ARG tech just dropped
BRILLIANT ! congrats !! 4 real
this is the kind of video youtube is meant for!!!! love it
Excellent choice in music for this video.
Yeah, nice vibes to match
You broke my nerd meter. It pegged out and bent the needle cracking the glass. ;-)
We already have tape-based data storage.
How are you so knowledgeable about stuff like these? Nothing hinders you and just always comes up with possible solutions, converting stuff, using different devices and methods, and calculations. It's crazy.
Very kind of you to say, thank you. But there are many things I don't know. Simple things like how to use an Apple Mac properly.
I also make many mistakes and get things wrong sometimes. I need to show more of this. Because allowing yourself to make mistakes is where the best learning comes from.
@@JanusCycle As a millennial who appreciates the old and new technology, I feel like a kid once again, watching in awe, the nostalgic devices you post and how you modify and customize them. So much memories.
Great idea! It allowed us to SEE the difference in sound quality. I wondered why I disliked the low bitrate lossy compressed sound (like in MP3) so much, compared to just lower bitrate uncompressed sound or analog sound with limited bandwidth. Cut the bandwidth alone and the encoded image will just be more blurry. Use lossy compression with too low bitrate and you get a huge amount of noise and artifacts that makes the original image nearly unreadable. Weirdly, the speech remains intelligible.
Some of these distortion patterns look awesome. I gotta give this a try myself if I can find a good recorder for it. Imagine how cool it would look if you ran every frame of a video sequence through one. It would look great played back on a little screen for a grungy scifi/cyberpunk movie.
Nice Alan Watts drop, definitely my favorite! Alan is to philosophy what Janus is to hacking
Lmao using RLM as the reference video for ghost hunting is great.
Those guys :) I'm happy they make these kind of videos.
That RLM video mentions how unserieous it is to use voice recorders for ghost hunting, you should use something designed for raw CD quality PCM like the Zoom recorder they used.
I want to believe myself but not enough to find it buried in compressed artifacting and audio pareidolia. I recorded a great example of this from a Discord stream of my friend using a shop vac. The vacuum cleaner sounds like it has voices in it because that's what the encoder is looking for, it's eerie!
I still think this voice based codec would be useful for 'hearing' EVPs. And I think Mike would have fun listening to these sort of recording artifacts until Rich calls out his bullcrap.
That last recorder worked best. I also get those apps on every device I can. Sstv encoder and robot 36 are a companding set that go together. Scottie dx is the only one that can do my profile picture without ghosting.
There is some numeric typo in subtitles when calculating audio bitrate. Like 16:19
I've often had this idea but the one thing that has put me off this idea is the fact that every voice recorder I've come across hasn't got an input jack, which not only means that I have to listen to the hellish noise of the SSTV for the entirte length of the recording, but I also have to contend with the analog distortion of the microphone (not distortion in the usual sense, but the frequency response of the mic altering the final signal). although I have toyed with both SSTV and NBTV onto audio cassette.
there is even an NBTV encoder that supports color, though finding an encoder that will take mp4 files has been quite the task, and I don't quite understand the color encoding myself to be able to make my own encoder. though I have made one for some of the monochrome ones, and I may in the future make my own standard that interlaces the frames for a balance of framerate and resolution.
I've done a bit of NBTV experimentation. But I had to play the mp4 on Windows and use NBTV software to screen grab it. I wish there was NBTV software for Android.
I have tinkered with making apps for android, as well as obviously nbtv. if I make something, I'll be sure to let you know.
@@AB-Prince If you do I'll promote your software in a video. Even if it's a paid app.
So what you did there is kind of demonstration of analog vs digital. Low quality analog is better than high compression digital. Also; considering NASA didn’t have to take cassette speed wobble and flutter into consideration. But other things like distance and weak signal. I’ve heard this is the same technology the voyager missions use to send pictures from far away planets back to Earth.
Voice codecs in CDMA and GSM phones also operated at ~ 10kbps range. Maybe they used similar technologies on them, too.
11:00 I would never imagine the digital recorder would do worst.
Well if you listen closely it sounds like extreme compression and noise reduction is the cause
Would’ve been cool if you yet the sound play without commentary so we could’ve downloaded the app recorded at home.
Good suggestion, next video with SSTV will include something like this.
"Open the cassette bay doors HAL . . "
It’s a really good visual demonstration of what a dac can do to a signal.
incredible.
thank you
Nice to see you working a Curta to calculate things. Never knew tat such mechanic calculators are doing that job with sucn large numbers so accurate. 🙂
So much more satisfying than an 8 digit calculator, or an app on your phone.
Amazing video,, thank you for the breakdown 💫
Neat, gonna run my album cover through this
We used to save basic programmings made by Commodore on a magnetic audio tape.
minimodem is a terminal app that can do similar but with bell modem, rtty, EAS etc
Compression maybe-- it might technically have the bandwidth it claims, but not at uniform resolution. The algorithm was probably tuned for human voice.
ROBOT36 is one of the fastest modes, but also one of the lowest quality.
Bruh whipping out a curta calculator is a huge (rich) nerd flex. I only know tech moan has one. You should do a separate Tutorial and review video on the curta if you haven't already, such a fascinating piece of mechanical engineering technology.
Edit: oh wait nevermind i saw your thumbnail on the curta calculator, ill watch that too!
I get to have this Curta for a while so I'm going to use it. I'm eating rice and beans to afford it one day.
@@JanusCycle ha ha ok
This calls for a "SSTV to Speech" and "Speech to SSTV" converter, to give us the best image quality with that voice optimized audio codec. 😅
Came for the music, stayed for the music. Learned quite a bit though.
That's an interesting looking "calculator" you got there
A 1954 Curta. Maybe it's time I bought one of those new electronic ones.
Great video! Those images look so retro and cool 😎
The mp3 192kbps looked flawless! I wonder if there would be any differences at all compared with a lossless recording. It seems that beyond a certain threshold, the SSTV signal gets completely transported.
That’s an amazing technology this is responsible of all the images and data sent from the moon, mars and all those deep space missions like the voyager, and it’s still in use helping sailors to have a daily forecast weather report, really useful for vessels that won’t have access to satelital services for any reason, and a great chance for radio enthusiasts to experiment on data transmission
Sstv is still used by ham radio ops on the hf radio.