Where Are Our Video Phones?

Videoconferencing has been around in one form or another for quite a while, but it took the pandemic to thrust into prominence with just about everyone. In a way, it has been the delivery of something long-promised by phone companies, futurists, and science fiction writers: the picture phone. But very few people imagined how the picture phone would actually manifest itself. We thought it might be interesting to look at some of the historical predictions and attempts to bring this technology to the mass market.

The reality is, we don’t have true picture phones. We have computers with sufficient bandwidth to carry live video and audio. Your FaceTime call is going over the data network. Contrast that with, say, sending a fax which really is a document literally over the phone lines.

So how did people imagine picture phones? And when did they become available? The answers might surprise you. We aren’t sure how far back people imagined such a device, but we know that one was built as early as 1930.

Even as early as the 1870s, science fiction mentioned telephonoscopes, an intention a cartoonist attributed to a future version of Thomas Edison. Alexander Graham Bell did think such a thing was possible and even wrote of using selenium as a way to sense light to convert it into electricity. He knew, though, that the sensor would have to be very tiny. This was years before electromechanical TV started to appear in the early part of the 20th century.

A Brief Survey of Picture Phones

From the Radio Shack comic: The New Science Fair Story of Electronics” from 1978.

You can argue what exactly constitutes a picture phone. AT&T offered a picture phone commercially in 1992 — we’ll talk about it soon — that showed black and white images at 10 frames per second. While that’s crude by today’s standards, most people would agree it was video. But some old attempts used something akin to what ham radio operators call slow scan TV. A camera would grab a black and white image and send it fax-like via audio tones. That means you got an 6- or 8-second delay and you could see the picture. Most people wouldn’t consider that video, but it is a slippery slope where you draw the line.

We remember Radio Shack briefly advertising these phones, although we can’t find any evidence of it now. But we think they were Mitsubishi Luma phones. Regardless, these kinds of phones were mostly novelties and didn’t sell well in the consumer space.

Ancient History

It is no secret that the phone company itself had the biggest interest in Picturephones (a trademark) and, in fact, everyone assumed that by 2001, every payphone would have a camera and a screen.  Not many people thought the payphone itself would be an endangered species. Even the 1927 classic Metropolis had a video phone call in it.

The phone company started working on video calling when it sent a speech by Herbert Hoover — then the Secretary of Commerce — from Washington to New York in 1927. The video was only one way and with a limited frame rate. The New York Times noted that Hoover’s face was not clearly distinguishable.

The equipment used probably looked like the 1927 prototype seen here. The scanning for the image was done with a mechanical disk, which is limiting.

Getting that much data across phone lines would be a challenge before there was any way to compress the signal.  In fact, the image traveled over multiple phone lines. With a mechanical system, the motors on the sending and receiving disk also needed to be synchronized, presenting another technical challenge. The phone company kept working on this system — dubbed Ikonophone — for several years with a team of around 200 people. Work slowed during World War II, but by the 1950s, the company would resume work on the system now known as Picturephone.

Meanwhile, in other parts of the world, Germans had public video phone service as early as 1936 between the post offices in Berlin and Leipzig. A dedicated cable 100 miles long connected the two phones that combined a flying-spot scanner and a cathode ray tube. The video was respectable, at 25 frames per second and 150 lines of resolution. The system eventually increased to 180 lines of resolution and other routes including Hamburg, Munich, and Nuremberg. Each city had two booths and a call cost about 1/15th of the average worker’s weekly wage. The French also had a similar system. These post office video phones, like the United States version, halted during the War.


By the mid-1950s, prototype picture phones using signal compression could transmit an image every two seconds over regular phone lines. The phone contained a storage tube or magnetic drum. By 1964, the Picturephone Mod I which had a better frame rate was seen at the New York World’s Fair and Disneyland. By 1964, there were several Picturephone booths, but they were not very popular, mainly because of the cost which could be as high as $27 for three minutes — over $250 today. By 1968, the phonebooths were gone.

The Mod II, however, found use for videoconferencing in 1970. The cost was high. The phone cost $150 along with a service fee of about $160 per month for the first phone and $50 for each additional phone. For that price, you got a whole 30 minutes of calls, after which you would pay a quarter a minute. Eventually, the price came down, but it was a lot of money in the early 1970s.

For that price, you got about 250 lines of resolution at 30 frames per second. It looks like you also needed a speakerphone for the audio and — by way of an additional box — signaling. There were two separate phone lines for the video. Early models were not color.

Final Analysis

Unfortunately for AT&T, the video phones they had developed over decades were a commercial failure. The official program alone was over 15 years and cost around $500 million. The phone company had predicted 100,000 phones in service by 1975. The real number was in the hundreds.

They tried again in the early 1990s with the VideoPhone 2500 marketed at consumers. At $1,500 a pop and anywhere from 10 frames per second to 1/3 of a frame per second, they didn’t sell very well. Surprisingly, they apparently sold about 30,000 units, but that was hardly enough to recoup their investment.

Other countries tried similar systems with similar results. France, Sweden, and the United Kingdom all had some form of video phone and all suffered from the same problems — low bandwidth over phone lines is simply not conducive to live video.

Villemard’s 1910 drawing of life in the year 2000 shows a video call in progress.

The answer would be digital networks and techniques, even though that wasn’t always obvious. For example, ARCNET — an early competitor to Ethernet — could — in some configurations — reserve bandwidth for analog video signals.

You can compress video with digital techniques, and network bandwidth has been on an upward slope. So as compression gets better, bandwidth goes up, and today it isn’t unusual to video conference with a large number of participants at once. While it doesn’t look much like Villemard’s 1910 imagining of videoconferencing in the year 2000, it isn’t that far off, either.

If you have the urge to hack something related, you can create a mirror system that lets you read your screen and make eye contact at the same time. Or, be a real hacker and do your conferences in text mode. We’re just glad today’s version of the video phone doesn’t let you see us pacing in the control room waiting for a call like the one in Metropolis (see below) does. We still aren’t sure what the paper tape coming out of those phones is for.

71 thoughts on “Where Are Our Video Phones?

  1. I use a DSL internet connection so all my video calls are over phone lines. Sometimes I even make modem noises to entertain my friends; I can do any baud rate up to 33600

    1. I should do that more often. Though my go-to word for an unspecified messaging technology is the “fax”, so “I’ll fax him” or “operations faxed me about…” – and it’s spreading in our company

  2. @Al Williams asked: “Where Are Our Video Phones?”

    I dunno about you but my video phone is in my pocket, and has been for years. It’s called a mobile phone. Do I use it for video calling? Not so much, it feels intrusive compared with voice-only communications.

    1. It’s a little bit funny, but this was something nobody really asked for. Even before this was technically possible, people were all like, “will I be able to turn the camera off if I’m not presentable?”. We were all used to the visual ambiguity that telephones provided, that allowed us to talk to our bosses while still in (or out of) our pyjamas. The video part only became important for group calls, so we could figure out at any given moment who was speaking. And so while you can use Zoom for a two-party conversation, we still don’t often do that.

      1. > so we could figure out at any given moment who was speaking

        Which became moot because the software brings your user icon to the front when you speak.

        I rarely use a webcam for my Teams conferences because you’ll forget that it’s open and start digging your nose or something. Without other people actually present in the room, it’s too easy to forget that they’re watching.

        Having to actively mind the fact is incredibly stressful, also because you never know who else might be watching if somebody records or relays the conversation onwards. It feels invasive, like having a security camera in the ceiling of your living room.

      2. In the robot novels, their video phones have filters that improve their appearance in real time. Which, even twenty years ago, seemed unrealistic and Impossible to a younger me.

    2. Yeah, I’m not sure what this article is about, but most people carry video capable phones on their person.

      While not all the calls I make/receive are video calls, the majority (in volume of calls and time) are video calls. And all calls with more than two callers are video.

      Maybe @Al Williams has been hanging onto his flip phone and fax machine, and just missed this.

      1. Do you have a video phone that will work with anyone else’s phone, regardless of what model they have, the same way you can place an audio call without worrying about what phone the other party has?

        1. Yes, I do. Mine will work with any phone or computer supported by Zoom, which doesn’t rule out anything but land lines, and even they can be included without the picture. Not really sure what you’re getting at.

          1. i dont think zoom is available for MIDP 2.0 by any chance???
            anyways, zoom requires an internet connection for video, no?

            but my biggest gripe is the fact that the article never mentioned the most recent version(s?) of ACTUAL video-phones.

            i dont know much about them execpt that they needed a specific subscriber service and were only available in select locations, they might have been using DSL or ISDN for the HARDWARE portion of the communication maybe? but certainly did NOT use include or provide access to anything other then what is available at the other end of the line, ie it was more like a faster version of dialup. think intRAnet, not intERnet, the rest is proprietary codecs and of course some sort of actual device.

        2. As long as they have Snapchat, or WhatsApp, or FB Messenger, or Teams, or Signal, or Zoom, or an Apple (facetime), or an Android, or Apple/Android (Duo is available for Apple and native to Android, or Apples can send a URL invite to non-Apples via SMS). Telco’s are now data carriers. Apps are now voice/video carriers. Meta would love to merge a few of these together. I’d like them not to. It’s not quite “without worrying what the other party has”, admittedly, but edging closer.

          1. You’ve made my point very well indeed.

            It’s fine that you can use any app that works. What is not fine is that I have to worry about what app you have. Why don’t all apps use a standard protocol to place the call, such that I can call with whatever app I want, and you can answer with whatever app you want? (That’s rhetorical: the answer is because companies want to try to control the market.)

            When that happens, we’ll have our video phones.

          2. @CityZen Given that pretty much all previous attempts required both parties to have specific hardware, it seems little like hair splitting to claim modern smart phones are not “video phones” because there are different apps you can use.

            I agree video calling still has a long way to go, as you point out a standard protocol would really help, but still, today people across the world are still making video calls using phones.

          3. Unless someone can come up with a peer-to-peer teleconferencing application that can run on multiple platforms, the big problem is that “videophone” services require servers, which means that even if there was a standard protocol, you’d still have to subscribe to the same service that the person calling has. Of course it is technically possible for there to be bridging services, but until all of that is in place, it will be the same cluster-dribble that SMS was in the bad-old-days.

          4. Back in the day (land lines), you got a “long distance” charge if you called somebody on a different telco (even if they where half a mile from you), but people weren’t say that phones didn’t really exist. Even now, although we appear to have peer-to-peer voice phone calls, we’re paying companies to allow that to happen (that host the servers, that allow bridging to different networks/services, etc.).

            Maybe there is a business opportunity here, bridging all the existing major video calling services.

  3. A flying spot scanner was a tv set for the scanning, and a photomultiplier tube for pickup. Hence it was only useful for fixed images. But simpler than building a whole camera. Hams didit with ATV, but it was really useful for sstv since the scan rate was so low that you didn’t get motion.

    Bell had Picturephones at Expo 67, a highlight. Only llater did I learn they had been at the 1964 NY World’s Fair. I don’t remember them there. But it’s a small world.

  4. I have a Polycom video conferencing system that has the option to bond four ISDN lines – at the time the equipment was new, that would have cost at least £400 gbp/mo. (Thankfully, these days it can use an ethernet connection as an alternative).

    The remote control is rather clever, and works a bit like the campfire notion of a talking stick – whoever has the remote control is deemed the speaker/presenter, and the camera head pans to the IR beam automatically. There’s a microswitch on the bottom of the remote so location is automatically triggered when it is picked up. Hey mom, no AI!

  5. There is simply no reason to bother trying to get the quality needed to do video over landlines. Bandwidth for mobile devices and PC is so plentiful it is simply not work the trouble.

    1. The industry were trying to make it happen all through the 90s, various capture cards and packages promising 15fps QVGA over 28k modem or better. Realmedia were into that as well as early streaming. But typically every time you fired up Realplayer it needed an update that took 15 min to download. Then there were some built into early Instant Messengers with various degrees of functionality. It seemed to be connection drops and long pings in the middle of the route that were the bugbear of them in those days. So into the early 2000s video chat didn’t not have a stellar reputation for reliability, I think we’ve only really pulled out of the slough of despair and into the productivity plateau in the last decade really.

      However, a kind of side channel development was that some national telecoms were offering video phone services to registered deaf people starting from the 80s. I believe it was generally subsidised but still quite expensive. I am not sure, but I think they basically needed two pairs to do it, much like ISDN.

      1. I remember sometime around 2002-2004-ish that one of my friends had a party where the living room 32″ CRT television was showing a webcam view of another party and he’d put a webcam on top of the TV so the other guys could see back sofa-to-sofa. I think it was Messenger video chat. It was a poor man’s version of the video wall with a terrible 12 fps not-quite-VGA camera because webcams at the time were just… not good. In order to do it properly, you would have needed a real video camera and a capture card and a fast PC to do real-time compression, which cost a lot of money for not a lot of utility.

        This was a time when 12/12 Mbit DSL was like, “Whoa dude! Your daddy’s rich!” – or you were one of those neckbeards living in a university campus apartment hooked straight to the mainline. Regular people were still stuck on 56k modems or 64k ISDN on the sub-urbs and 512k ADSL in town.

        The 56k and ISDN were also billed by the minute, so you wouldn’t really use them like that to hang out online and chat with people. You did what you had to do and then promptly turn it off. Having a poor quality video call was a complete waste of money. In attempts to save money, I actually picked up one of those early GPRS cellular modems because the carrier was offering 100 MB per month free – so you could hang around on IRC without racking up the minutes. That was about all it was good for, except the modem would crash multiple times a day and stop responding.

    2. I get 10x the bandwidth reliably out of my landline than I can trust the local LTE/5G network to deliver despite all the hype.

      The trick is that mobile bandwidth is seriously constricted on the upstream path. It’s a system optimized primarily to stream commercial content one way to your phone or tablet – not the other way around.

  6. Did April Fools day roll around twice this year somehow? My video phone is right next to me charging. I was using it just yesterday to talk to my son in Norway, (I am in the USA) and the video was frankly pretty amazing. I know it is a surprise to some people, but video phones are a thing of the present.

        1. Yeah it’s ticking me off that none of these so called smart watches have got a fracking* Dick Tracy mode.



          $Fudge, only I didn’t say fudge.

          1. Yah, even my derpy $30 watch with the 2 hour 300mAh battery does that,, like a BT headset strapped to your wrist, but it don’t do video both ways, neither does anything else I’ve seen..

  7. I wonder if some of this would have been more practical around the late 80s early 90s if they hadn’t overpriced ISDN. That would have made a decent transmission channel, and would have been practical to expect if everyone who had use for a modern chose to use ISDN instead.

    1. Oh yeah. ISDN allowed multiplexing data with the regular phone signal by pushing it to a higher frequency band, or use both for data to get more bandwidth (but you couldn’t pick up the phone) – so of course the telcos treated it as if you just bought two separate landlines and charged you for both.

      1. It really is a great example of an industry that profited itself almost out of existence. If they’d actually lowered prices appropriately they wouldn’t have been rolled over by cable, fiber and cell phones.

        1. Well, that, and not using obsolete technologies like ISDN/ADSL. You also forget WiMAX which was a thing before 3G became a thing.

          The problem of ADSL is that the signal quality drops very quickly over distance along the phone line, so the only way to get high subscriber speeds is by making it very asymmetric to the point that you have no upstream bandwidth for anything. It also suffered from the fact that the access concentrators didn’t have enough bandwidth for the number of customers and the speeds they were sold because the infrastructure was built to handle analog modems.

          The solution to both problems was to pull cable or fiber everywhere, to put more access concentrators closer to the customers, which would mean that there’s basically just a few hundred feet of actual phone line between the typical customer and the DSLAM just down the street or even in the customer’s basement – at which point you might as well hook them up directly with cable, fiber or ethernet.

          In fact, that’s exactly what they did. Now you only get ADSL in very old housing complexes where it’s not cost effective to tear up walls to retrofit ethernet cables, so the old copper pairs are still used from the utility room to the apartments.

          1. It’s interesting to remember that some P2P technologies like BitTorrent were purposely built with features to get around the issue of asymmetric ADSL speeds – since transferring data between any two customers was limited by the upstream rate.

            You yourself could have a “fast” 8/1 Mbit line but the other guy living a mile or two further down the road could only have something like 1Mbit/256k and you were stuck at his speed – except that you could download other parts of the same data from other people. The telcos hated this and started blocking and throttling torrents, because they would completely saturate their feeble networks that were not upgraded with the assumption that nobody would actually use the bandwidth they were sold.

      2. Although ISDN only used a single wire from your home to the exchange, it WAS taking up two 64k data slots beyond that.. you can argue about where the telco’s costs lie in that setup and personally I agree they priced it way too high.

        1. Yes, but it wasn’t circuit switched. The original reason why you’d pay per minute to keep the line open is because it would literally make a physical connection through the network all the way to the other end and reserve one cable pair per call. ISDN made the last mile digital, so you were actually packet switched from end to end, and the telco could support many more customers. Yet they still kept charging per minute because they could.

          It wasn’t until ADSL which was always on by design, that they dropped it – although a few operators did try to pull off a by-the-minute ADSL, especially where it came to competing carriers trying to operate through their networks under common carrier laws. They’d say “yes you can buy from provider X but you have to pay us 7.5 cents a minute for renting our lines” – which was quickly declared illegal.

  8. > “We still aren’t sure what the paper tape coming out of those phones is for.”

    Serious reply: Ticker tape for the latest news and quotes as it was fashionable for those times–it’s a communications center not just a phone.

    Snarky reply: Receipts for the continuously increasing phone rates and charges from the vidiphone company.

    1. It almost looked like he was looking up the coordinates for… maybe… someone who called before? I don’t know, but you have to wonder what was going in the guy’s mind that mocked that up.

      1. Probably something like “Line printers and fanfold paper won’t be around for 20 years or so, disk for 40 and flash memory for 60+, so I’ll have to write the log to tickertape I guess.”

      1. He said. In text. Further evidencing her point.

        Its true that voice is still very much what phones are for, but doesn’t it seem like email and sms are becoming more and more important? Is that just my perspective?

      1. In the old days there were said to be four kinds of managers: Memo, meeting, phone and wander/ambush.

        They all sucked because they all use one tool for every job. Like a pre scrum, scrum master. Just shoot them all.

        The rarest thing is the ‘understand the problem’ type manager. They all think they are, but 99.9% wrong.

        Nothing really changes. MBA school admission lobotomies will continue.

  9. If he did not have a video phone, would the machine attendant have left the door closed?

    Would a millennial be more afraid of their boss than the bloodthirsty mob outside?

  10. Here’s what went wrong: the telephone system had a product that was inherently very cheap, but they priced it at what the market would bear. This placed a baseline value on the 3 kHz bandwidth that you got with this technology. Now you want to send video, which requires about 1000 times the bandwidth. At the same bandwidth price, it’s a non-starter.

    Something similar SHOULD have happened with MMS. We were paying a nickel a pop for 160 byte SMS messages, so when they added the ability to send pictures and even videos, the price should have been prohibitive, like $50/picture. And then, if it turns out that the bandwidth is so cheap, it’s no big deal to send pictures, then why doesn’t the market ask “why can’t I send a 10 page text document, then?” But these days, nothing has to make sense.

      1. It’s a feature of monopolistic markets. Competition brings the price towards the marginal cost of production, which for data is essentially zero: once the network is built, sending one byte or one terabyte costs virtually the same.

        For cellphones, laws that require operators to let other operators use their network infrastructure opened up competition and brought the price of data down to the point that in most countries you no longer pay for it; you get non-metered access for a fixed monthly payment.

        MMS happened at a point in history where this was not generally the case, and the MMS messages are sent to the Multimedia Messaging Service Centre which then relays them forwards, and this service is proprietary to the carrier. You can’t use anyone else’s MMSC except your own carrier’s which means there’s no free competition over the price.

      2. What planet do you live on?

        Price is basically cost for commodities.
        That’s a test, e.g. Are Rolexes commodities? Is RAM? Bandwidth? Wireless bandwidth? Corn?

        Not understanding the difference between Rolexes and cell phones will bite you. iPhones want to be Rolexes, but no.

        It’s not entirely that simple, you have commodities with long investment return lags (e.g. petrochemicals, chips) where the price runs long term average cost, but is quite profitable for short periods.

        Marx thought that everything would be a commodity eventually and that would cause capitalism to eat itself. Not the dumbest thing he predicted, completely wrong testable prediction though.

  11. Now that video is ubiquitous, most meetings I attend with Teams or the numerous video capable collaboration programs are audio only and really most of them are not even that but are email or text.
    When given the option of live multipoint video or screenshots of code, pdfs, pwrpoints, spreadsheets, etc., it’s generally the screenshot that wins the content battle. Who would guess that the killer app for 21st century business would be the evolution of a text pager even when videophones are free?
    Says something about human nature and preferences, doesn’t it?

  12. When I read this headline I was thinking about the 3G phones from 20 years ago. I had an LG U880 flip phone that could do video calls but (a) I didn’t know anyone with a video phone; and (b) I had no 3G in my area.
    What happened to that? I use FaceTime a bit and occasionally Teams or Zoom for work, but outside of apps where is the carrier protocol?

  13. That’s a really widescreen for watching friends play tennis over the “phone”! Those headrests hamper widescreen viewer action. Oh, or are the 2 viewers evenly split on the players? As in to each a view-screen.

    Just like some phones today it has a notch cut in the screen edge, how ironic.

  14. The 1992 VideoPhone 2500 was actually entirely developed by Compression Labs, Inc. (CLI) under contract to AT&T. Worked pretty well over an ISDN link — provided you could get the ISDN link to work.

    1. Yeah, a big gap in the history of video phones if you leave out company’s like CLI, Picturetel, Tandberg and later Polycom and on the audio side Clearcom (now Clearone). These companies helped development of the H.264/H.265 video compression algorithims and Clearcom with their echo cancellation algorithims.
      Most of units could acheive 25 frames/sec video (PAL/NTSC/SECAM) with 4K audio quality (telephone quality audio). 6 x 64K isdn channels were multiplexed to give you a bandwith of 384K. As others have mentioned it was very expensive. From memory it was around the mid 2000’s when Tandberg and Polycom started doing video over IP.
      Cisco bought out Tandberg and rebranded them to telepresence units which are still sold today and have a bandwith up to 5Mbps. I can’t believe companys still buy these, most I guess are Cisco fanboys.

    1. Persistence of vision kicks in at around 10-12 fps. Below that, the brain can identify and process the individual frames as separate pictures, above looks like motion, although the brain can pick out an recognize a single image out of a series at up to 75 fps which is approximately the flicker fusion threshold.

  15. Sitting on my desk. along with a few thousand others in the same office, and tens of thousands of others across other campuses. Couple of different models (all Cisco) that talk to each other, and also to Zoom with some awkwardness. Softphones are replacing them in some cases, but everyone hates the softphone vs using an actual device.

  16. Yup – pseudonymous is completely right. Video phones are everywhere, and video screens even more! But because interoperability is a paid additional feature – most business don’t have it. And why would I want a video only phone? I don’t even have a single purpose phone… just my smartphone.

    1. For the most part I just wish mine was ‘phone’ and ‘text’ with an actual keys to press. The flip phone was close to perfect for a communicator. Really that is all I use (for home, every day use… Business is different) . But no, they got to load it with ‘apps’ one will never use, play movies, access the net, take pictures, etc. Silly. Actually the best phone would be a standard home phone with caller id with text capability…. Who actually needs individual phones? Or have one on you 24×7 (other than business on-call needs) … Just costs more anyway :) . Get back to the simpler life of just checking the phone when you get home for messages…. Why do some think we need to be always connected?

      So video phones? Not wanted. Not needed.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.