CRT Manufacturing
vintagetek.org
vintagetek.org
A short excerpt:
"Scope faceplates used to be round, simply because glass CRT bottles were round, and the faceplate was part of the bottle. Today, most faceplates are rectangular, to provide maximum display area; thus they require tubes that are rectangular in cross-section. In ceramic tubes, the glass faceplate is fused (or 'fritted') onto the ceramic envelope. That is, glass bottles are truly 'bottles'; ceramic envelopes are more like sleeves, or funnels, open at both ends. Such a rectangular envelope requires great structural strength; to make one of glass would mean very thick walls, thus a much heavier tube. (TV cathode-ray tubes, for example, are not rectangular, but bowed on all four sides to achieve the necessary strength. That's why your screen is the shape it is.) ... A glass bottle is spun, or blown, inside a cavity; a ceramic one is formed outside of a mold. This gives us control over the internal geometry of the tube - particularly important in post-accelerator (helix-type) CRTs. ... the large-screen CRTs required for our display units and computer terminals rely on ceramic envelopes for the necessary strength, fidelity of image, close-tolerance geometry and reasonable cost."
Is OLED manufacturing easier?
What is the easiest display to actually make?
In all likelihood the ability/knowledge to manufacture high quality CRTs has been irrevocably lost and won't be regained unless somebody wants to sink large amounts of cash (hundreds of millions?) into reviving it. In contrast, vinyl never actually went away.
retro-fashion
The appeal to many of these outdated technologies extends beyond "fashion."
There might have even been a discussion on here about this actually.
Digital Foundry has done some excellent videos on CRTs for retro and modern gaming. High end CRTs are actually fantastic for modern games, too, especially at 120hz. (Although, good luck finding/affording one...)
https://www.youtube.com/results?search_query=digital+foundry...
While purely subjective, many would agree that low-rez pixel art looks much better on a CRT:
https://www.reddit.com/r/interestingasfuck/comments/9xg8w3/t...
There might have even been a discussion on here about this actually.
In reality, sometimes these old technologies have unique qualities. I have to laugh at the idea of CRTs being "retro-fashion." I've got a half-dozen CRTs in my home and basement and let me tell you, nobody thinks it's cool ha ha ha ha.
It's not that we literally don't know how to make CRTs any more. But the manufacturing ecosystem is gone: the machines, the tooling, etc. That was the product of 80 years and many billions of dollars of R&D. Making high-quality CRTs requires some pretty specialized propietary gear.
We could probably manufacture a 1950s-style black and white TV without too much effort. But it would be a tremendous leap from there to creating e.g. a high-end Triniton monitor capable of doing 1080p@120hz.
You could make one in your garage for sure. /s
Reasons why I preferred it over flat panel:
1. Zero lag (lag annoys me a lot when I am gaming)
2. Arbitrary resolutions (for example I was running several recent games using a medium-range video card, with graphic options on ULTRA or above, because the games would look gorgeous on CRT with low resolution)
3. Arbitrary resolutions (want to use content for other format than widescreen? It is covered! Content designed for pixels that are not square? It works too!)
4. Stupid high refresh rate if desired (while lowering resolution of course)
5. Stupid high resolution if desired (while lowering refresh rate).
6. Better contrast (one of the reasons I got CRT back was that I got annoyed with flat panel when I tried to play SuperHOT and watch Game of Thrones on the same day, if I avoided SuperHOT becoming a white blob, Game of Thrones became black blob, if I avoided Game of Thrones becoming black blob, superhot became white blob)
The last one sadly is only "theoretical" for me because I never had the right equipment and software, but CRT could do HDR images, for example Silicon Graphics for a time was shipping computers for digital art that had a fancy video card that could output 48 bits of colour, that computer came with a ginormous widescreen CRT too... I believe that CRT is the one John Carmack is using in a famous photo.
Also I heard before Windows Vista that introduced a lot of DRM-related restrictions on video and sound drivers, you could do a lot of fancy shenanigans with CRTs, seemly intended for engineers and architects, but I never got too much details about it, I never seen it in person and nobody ever described it to me, I only saw people bemoaning it was removed.
Also I heard before Windows Vista that introduced a lot of
DRM-related restrictions on video and sound drivers, you
could do a lot of fancy shenanigans with CRTs, seemly
intended for engineers and architects
From what I understood the non-official reason for its removal was to prevent people from using it to circunvent HDMI restrictions in a way that would allow them to duplicate movies without using DRM.
The official reason was that it caused driver instabilities and thus Microsoft was against its inclusion in the new driver model for Vista.
Why it was popular with CAD, I have no idea.
We can replace aging consoles with highly accurate FPGA clones, but we can't replace the displays. The best modern displays still effectively lag behind a CRT by about a frame (~15ms), which really does matter.
If you're a top-level speedrunner of a retro game,
there's just no good alternative yet.
Similar to how most folks, in a blind test, would prefer a fresh-baked pastry from a skilled baker that uses quality ingredients vs. some kind of shelf-stable junkfood pasty. They would likely not know or care about the fact that one uses high quality grassfed butter and the other uses partially hydrogenated oils and that one uses cane sugar and the other uses corn syrup. But they would enjoy one more than the other, even if they weren't able to articulate why.
Frankly, a well made tube amp coupled with a quality phonograph cartridge has a great sound! It is a sound many people, myself included, crave.
Some of that sound happens because of production constraints. There are limits to what can be put into a groove with good results and those impact how the music is produced. A vinyl mix often stands right out.
A similar thing exists with AM radio. Back in the peak time for AM music listening, many stations were broadcasting 8 to 10Khz of bandwidth. It is a far cry from the brutal 5khz brick wall so many stations use today.
Mixing music so it would sound good on AM as well as FM, stereo and wi-fi, tape, vinyl, involved some production constraints that again color how music is produced.
That distinctive 70's era sound is that way because of these things, in part. Obviously, there is more to the story, but where there are limits, there is art.
As those limits were removed, we got loudness wars and some albums shipped clipped and brick wall tiring to listen to. Expose someone growing up on that to a banging vinyl production and it can blow their minds and they go looking for more!
And let me be clear:
The tube amp and vinyl source material are not better than a CD, though they may be better in some respects than a stream is, and digital audio is superior. Period.
It is better than our ears are and has been since we got the storage to produce it and distribute it properly to people.
So that gives?
It is all about the limits and the art. Sounds people crave.
And for what it is worth, each vinyl play is unique. Heard only once and then forever gone. People know the record is consumable, an experience to be valued and so they do.
I like to capture them after a few plays. The occassional rattle from the cartridge, pop, click, all speak to the time I recorded a record I love while in it's prime.
Here is a sort of digital parallel experience that might help:
Right now, I am typing to you in the little HN text input window. I could take it off line, use other tools, then paste it right in. Sometimes I do. Depends on what it is and whether I feel the window is risky...
And the difference on your end? NONE! It was for me. A bit of theatre, immediacy, connection to a "conversation", or "person" created in my mind that does help what I say flow differently.
These little things matter to people. They are why apps will signal, "X is typing..." in an attempt to convey more than text, but less than people in close proximity.
People run little radio stations for it too. The station has very low power. But, it is "on air" and that sense of a unique experience colors what people in the studio do. Others may listen on a radio, and a few will. Locals. Most will stream, or pick up a replay later.
And there will be a subtle energy in that missing from just recording in a studio.
Once a person begins to look for this, it is fascinating! And that is why I am writing this. Music from before the computer editing and processing was not quite as perfect as we can make it.
But ask around and many pick up on the energy to be had from a great show performed decades ago. Maybe from before the listener was born even. And they will talk about those humans who did it in real time, no bullshit, just everyone in the moment, experiencing something familiar, yet unique to them.
And that is all. The analog age has a sparkle to it people crave. And it will still work today too. All people have to do is embrace the art. Take the limitations and roll it all up onto good sounds.
I'm pretty sure one person can make basic CRTs in their shed because that's pretty much how early CRTs were made.
Of course really good or very fast CRTs are a tad different, but the fundamentals (i.e. electron optics) are still used in various places, so it's not like we've collectively forgotten how to shape an electron beam. Tektronix also published tons of literature about stuff like this.
> What is the easiest display to actually make?
Discounting electro-mechanical displays, I'd say VFDs.
I recall an article from ~20 years ago where some researchers made an OLED display using an inkjet printer. I'm sure the consistency and reliability were not what you get out of modern manufacturing, but still...
I've been waiting for inkjet based printing of simple circuits and lighting but it's just not happening. Imagine printing light-up signs. c'mon Y-combinator we need this tech everywhere.
According to a paper from Samsung at IEEE IRPS three years ago, the best OLED screens only have about 2 years (600 hours) best case continuous use but that is extended by "screen savers" and estimates of only 25%-50% customer use in a 24 hour period to about 5 years (extend by not using)
https://ieeexplore.ieee.org/document/8720529
Part of the reason for this: low energy manufacturing materials/processes, by definition (physics), will have low energy failure mechanisms (which always means they will fail more quickly). This is about chemico-physical reaction activation energy. Anything ink-jet created or organic is a low energy process/material.
I don't have access to paper, but rough estimate is that there are ~17k hours in 2 years.
If screen has actually 2 years 24x7 continuous use (and I assume they use a semi-arbitrary brightness threshold or something?), I imagine for TVs that'll be an easy decade of real-world use, depending on usage patterns. For some phones maybe less. I'd be curious what the specs for LCD & LED are then.
If the whole display was expected to die in anything like this amount of time, I would anticipate widespread consumer outrage.
That is one of the most obvious-in-hindsight things I've learned lately. Thanks!
https://www.youtube.com/watch?v=GDvF89Bh27Y
The transistor would start to displace tubes during the 1960s (when I got into electronics in the early 1970s, you still would learn or use or have as option the user of vacuum tubes but by the end of the 1970s, that was 100% gone for professionals and hobbyists though in 1980s as a freshman EE student we had the last gasp of vacuum tube education corresponding to having to use FORTRAN and punch cards that one last semester for programming as well). At the same time, discrete transistors were arising as the de facto choice in commercial (the adoption "Chasm" for them occurred in the late 1950s). ICs were still "early adopter" in the 1960 and crossed the chasm in the late 1960s, early 1970s which is why the 1970s were the boom decade for Silicon Valley and IC manufacturing.
As I watch tube manufacturing with 40 years of experience in semiconductor, I just cringe at the non-reproducibility and labor involved - it's insane. By contrast, semiconductor (and by commonality of process, OLED) is far simpler. The reason: the planar photolithographic cycle is repeated over and over again in semiconductors which in contrast to the "different at every single step" of tube manufacturing is radially simpler.
Also, the much higher amount of materials, the bulk etc would make them unviable now.
Not really. CRTs are more bulky, more fragile, require lead for the front glass (which means issues with environmental protection laws), complex power supplies and they have a significant quality loss in the analog conversion. Why should anyone buy a CRT these days, unless for authentic replacement of historic appliances?
> Is OLED manufacturing easier? What is the easiest display to actually make?
That depends on your definition of "easy" - these days I'd assume TN LC displays are the easiest to manufacture at scale, the technology is decades old (and thus, no patents in play) and there are lots of manufacturers. OLEDs may be easier to manufacture (JOLED aka Sony/Panasonic have shown that one can print a display using an inkjet printer), but that stuff is heavily under patents for the next two decades so it's not accessible enough to meet the "easy" definition IMO.
My current employer (based in Taiwan) makes an "LCD Flat Panel Manufacturing Plant" product.
You literally just need a building shell (the plans for which we can provide to qualified sales prospects). Then you place your order and a few months later dozens of shipping containers arrive and you put the contents into the shell like a child's construction toy or Ikea furniture. All the cables are snapped together and you start feeding it the inputs, and out pops flat panels. It's 100% turn-key.
This product is part of the reason why flat panel TVs are so cheap now - and Chinese customers LOVE how it's so simple for them and requires zero actual knowledge of making flat panels - sort of the ideal manufacturing for them (we also sell a support contract service to run them).
Also one of the very last CRT manufacturers on the planet (which now manufactures flat panel screens as well with CRT production ending in the mid 2000s) was in Taoyuan TW (their logo still has a CRT in it last time I drove by their plant).
Taoyuan and Hsinchu have been glass centers for TW for the last 200-300 years due to natural reserves of highly pure silica and natural gas in the area. Before semiconductors, these places made scientific glass products (chemical/medical glassware, lenses, etc.) which is why this former CRT plant was located there. Note that the supplier of lenses for iPhones is located in this same area. The common supply chain of silica is why Hsinchu is the semiconductor center that it is. Old fashion geography still applies to why the world is the way it is.
If you were to start a CRT manufacturing plant in the US, the only best place would be in NY or PA near Corning NY because Corning Glass is still there and doing leading edge work in glass still. There's no company nor area better for glass or glass dependent products in the USA at this point in history.
If I had to guess on pricing of a single line capable of generating ~1-3000m^2 a day, it would be in the 10s of $million excluding the factory building or land itself.
1. Designing a shell that is unique enough but simple enough to pass muster for the average person. Aesthetics matter, after all.
2. Incorporating a remote that had the features you want without directly copying another design, and making it fully functional with the TV brains
3. Finding a microcontroller that can be purchased at the appropriate quantity that is able to handle the latest features that you want without requiring internet connectivity. (HDR, Post Processing, Dolby/Atmos/Surround, etc)
4. Getting the appropriate inputs to handle all of the expected inputs for at least 90% of your target audience's expectations
5. Getting the finished product UL Listed and fully evaluated by the appropriate licensing bureaus.
6. Doing all of the above without causing the TV's price point to explode well past the point where your average target purchaser can or would choose to afford it when less expensive brands exist that have reliable warranties or customer service or a name to live up to.
Should you pass those bars and have a modicum of success, then you will have to fight off all of the competitors who will leap at the target audience you've uncovered, releasing dumb tvs at price points you may not be able to match with the goal of driving you out of business even if it costs them money so that when the next cycle comes around customers will have no choice but to purchase their products again.
Basically they had similar panels to the then-current 1440p 27" Apple Cinema Displays but were usually "Grade A-/B/etc." instead of A+. They had the flimsiest, crappiest cases and most primitive controller boards (i. e. no OSD or multiple inputs on many models, just keep tapping the button and trying to get the brightness where you wanted it.). I think the original market for them was domestic, and you pretty much had to order them via eBay or sketchy small retailers.
I suspect what did it in was collapsing prices for mainstream-brand monitors, and a choice-paradox problem (there ended up being dozens of seemingly identical units to choose from with different cases, so the market was probably sliced too think for anyone to succeed).
But the concept is completely viable here. Remember that the name brands have major cost centres you don't-- advertising, a high-touch retail sales channel, the R&D on smart features.
TBH, I'm surprised you don't see more higher-end TVs without speakers. There's always going to be physics-related problems with trying to have really decent speakers in a 5cm deep set. You can target the demographic of people who already have competent audio systems AND save on the BOM!
Probably not.
> Is OLED manufacturing easier?
Define "Easier".
Is it easier to manufacture a transistor or is it easier to manufacture a vacuum tube?
If you were stuck out in the wilderness with nothing but rocks, sticks, and a pile of sand and a team of engineers you would probably have a much easier time building a primitive CRT display than a OLED one.
But if you have a goal of having to manufacture a reliable display at minimal cost then OLED wins hands-down.
Depending on your constraints one is 'easier' then the other.
------------
Old CRT displays were nice (inherent advantages in terms of resolution, refresh rates, color depth, view angles, and color accuracy), but they were fragile and required significant amount of labor to construct. It is unusual for CRT-based televisions to last decades without requiring repairing and tuning.
Where as modern modern flatscreen displays are much cheaper to produce, are much more efficient in terms of resource usage, and are much more reliable. People tend to replace flat screens because they become obsolete. However they replaced tended to CRTs because they stopped working.
So no. I don't think it will be back. Modern displays are much easier to make also, more high-tech but use less energy and materials.
https://en.wikipedia.org/wiki/Phosphor
You'd also have to scale strontium mining and refining back up.
This is completely wrong while technically correct. Sure, electron guns throw electrons around. So you are looking at a beta-emitter. But those electrons can't leave the tube. Even if they could - CRTs have, depending on size and application, an acceleration voltage of somewhere between 10 and 30 kV - this translates to a range of a few centimeters in air at best.
CRTs do produce radiation, namely X-rays by way of bremsstrahlung due to the electrons getting, well, bremsed at the screen. This is part of the reason why CRTs are heavy (aside from faceplate structural integrity); leaded glass is heavy.
> Flashing
Visible flicker is obviously undesirable, but low duty cycle displays are preferable for moving images.
True, I didn't want to get into that level of complexity. The electrons are also quite easy to capture with conductive glass. But the bremsstrahlung can be significant.
She got the hell out of microelectronics and is now a data scientist. The former is a dead end job in the USA and the latter pays better.
The trouble is that solar panels in lots of areas are vulnerable to weathering, accidents, theft and sometimes vandalism. The ones I've seen are just not very durable. I drive by a field full of them frequently and a couple times a year I see workers replacing a panel or two. I've been told they're supposed to last for a minimum of 15 years.
https://sometimes-interesting.com/electronic-waste-dump-of-t...
I wonder if or when it'll ever be economic to make a display that rivals CRTs in terms of characteristics that now make them seem so unreal for video games. Put another way, if we had to create a new CRT-rivaling display tech with more modern manufacturing capabilities, what would we come up with? Maybe little micro-arrays of electron guns? A spinning array of micropulsing lasers? I'm so habituated to new technologies feeling magical, that clear evidence of a back-slide feels... wrong.
The last crt I regularly used was the shared tv in my dorm, which I moved out of in 2008.
I absolutely have no memory of them being superior.
https://www.eurogamer.net/digitalfoundry-2019-modern-games-l...
Are we sure this is a lcd issue and not a frame rate?
It would sell to a niche if it was affordable; I just don’t think the manufacturing infrastructure exists anymore to make this at a price people would pay.
Let’s not forget the weight either - a modern CRT running something approaching a size and resolution we expect in 2022 is going to easily exceed 20kgs. Apple’s last 21 inch CRT before the switch to an all LCD line was 35 kgs.
I am not nostalgic for the days when 3 hours of using your PC felt like frying your eyes with lasers.