The Quest To Save the World's Largest CRT TV From Destruction (arstechnica.com) 119
A rare Sony KX-45ED1 television, considered the world's largest CRT TV, has been preserved from destruction in Japan, marking a significant moment for gaming history preservation. The 440-pound display was salvaged from an Osaka restaurant days before its scheduled demolition, following a two-week international rescue operation.
Gaming enthusiast Shank Mods, aided by local contacts and industrial shipping experts, secured the functioning 45-inch unit, which originally sold for $40,000 in the late 1980s. The TV, valued by retro gaming enthusiasts for its authentic, lag-free display capabilities, could potentially become a public exhibit pending future funding.
Gaming enthusiast Shank Mods, aided by local contacts and industrial shipping experts, secured the functioning 45-inch unit, which originally sold for $40,000 in the late 1980s. The TV, valued by retro gaming enthusiasts for its authentic, lag-free display capabilities, could potentially become a public exhibit pending future funding.
"Authentic, lag-free display capabilities" (Score:3)
"Authentic" display capability? What the heck does that even mean?
Marketers and salesmen really love to overuse that adjective.
Re:"Authentic, lag-free display capabilities" (Score:4, Interesting)
I'm assuming "authentic" means "as intended by the game designer."
Re: "Authentic, lag-free display capabilities" (Score:5, Insightful)
or as in "authentic" scanlines and not generated by emulation
Re: (Score:3)
I would think it means "historic" as in "in the past things were better" or such nonsense.
Re: (Score:3)
Re:"Authentic, lag-free display capabilities" (Score:5, Informative)
Sprites for old consoles were designed with the phosphor bleed over of CRTs in mind. This functioned as a sort of in-built anti-aliasing. This is why many of these old games look like shit when emulated naively on a modern LCD. They were meant to be viewed through the particular sort of "bloom filter" that CRTs produced. Sure you can emulate that with software and it can look pretty good, but it's never quote the same, just like digital photos are different (not better or worse, just different) than film. Or like if you've ever played asteroids on an an actual analog vector display, nothing else is quite the same. You can't emulate the intense phosphor pinprick glow of the projectiles because it was an actual physical property of the display producing a spectrum of light that isn't reproducible on a modern display.
That's not to say everyone should run out and dump 10s of thousands of dollars into old junked CRTs, but it's nice that people are doing the work of preserving them so they can be used in museums and such.
Re:"Authentic, lag-free display capabilities" (Score:4, Funny)
In this case, of it's for retro games... CRT was the only tech in town.
But for real retro feel, you have to use an adapter to connect RCA cables to rabbit ears.
It's not about the retro feel (Score:3)
For casual gamers you won't notice but if you're doing speed runs, practicing intricate fighting game combos or trying to do 1cc score attacks in bullet hell shmups even an extra 1-2ms of lag can kill it.
Modern "run ahead" emulation can all but eliminate that lag with powerful enough CPUs but not every console
Re: (Score:3)
Demo/Intro coders would follow the rasterizer all the way down the screen, changing the palette every single scanline immediately before the next scanline was sent to the crt
Any tv that also does upscaling and whatever, you can add another frame of latency minimum but its going to be more like 4 frames more because thats how temporal ai upscaling bullshit works
Re: (Score:3)
Indeed, and the delay can be no better than 1.5 frames (average) of latency.
Why? I mean yes, I know that a lot of existing hardware does this because it captures an entire frame and scales it, but in principle, there's no reason you couldn't do a scaler that has only a little over one field of latency (1/60th of a second) — just enough to start capturing the intervening scan lines — or, for that matter, disable scaling entirely, do dumb pixel doubling, and scan the LCD lines in a different order, resulting in zero added latency.
And even without a true native interlaced
Re: (Score:2)
You are a fucking joke
One scanline... 1/28200th of a second latency on the WORST hardware
Take your ignorance and your lack of an internal monologue and shut the fuck up until you at least have an internal monologue.
Re:It's not about the retro feel (Score:4, Informative)
it's about input latency. LCDs by design do digital image processing on every frame which adds a delay between when you press a button or move a joystick and when you see the result on screen.
Overall latency is what actually matters. With a 60hz console running on a CRT there is still the 16ms scanout to actually update phosphors with the full frame.
Native refresh of a modern panel is 240hz (4ms scanout) or more... even if you presume several ms of input latency this is still half of the overall latency of the CRT.
Re:"Authentic, lag-free display capabilities" (Score:5, Interesting)
It means it doesn't use significant post processing on incoming video signal like LCD and LED based displays do before displaying the image.
Re: (Score:3)
Analog "post processing" is nothing like "post processing" being referred to. This is why lag on CRTs is measured in single digit nanoseconds. On LCD based screens, it's usually in tens of milliseconds. I.e. hundreds to thousands of times worse. They're not even in the same ballpark. Heck, not in the same game.
Re: "Authentic, lag-free display capabilities" (Score:2)
Re: (Score:2)
This is more of the "I don't understand that weaknesses of LCDs are only inherent to LCDs and not other technologies" babble.
Lag doesn't measure what you think it does. It measures how long from a time when signal was sent to the monitor does it take for monitor to draw the frame. For CRT, that takes nanoseconds. For LCD, tens of milliseconds. Notably for OLEDs, it's closer to CRT than LCD, because as noted above LCDs have unique weaknesses inherent to the technology that neither CRTs nor OLEDs suffer from.
Re: (Score:3)
Lag doesn't measure what you think it does. It measures how long from a time when signal was sent to the monitor does it take for monitor to draw the frame.
To draw a line not a frame. A frame at 60hz still takes 16ms to draw presuming 0 input lag. Lines at the top of the display have less "lag" than lines at the bottom that have to wait in line the full 16ms for their turn to be scanned out.
For CRT, that takes nanoseconds.
It takes longer than this just to collect enough electrons to charge up the phosphors.
Notably for OLEDs, it's closer to CRT than LCD, because as noted above LCDs have unique weaknesses inherent to the technology that neither CRTs nor OLEDs suffer from.
Even with input latency a modern LCD panel with a 240hz panel driven at 240hz will have half the overall frame to photon latency than a 60hz console connected to the CRT.
Re: (Score:2)
>To draw a line not a frame. A frame at 60hz still takes 16ms to draw presuming 0 input lag.
You need to stop talking, and google how CRTs work. You have demonstrated to be making multiple wrong assumptions such as for example one I quote above, and then going from these silly conclusions towards even sillier outcomes.
Hint: CRT unlike LCD does not display a permanent image. There is no frame. There's merely a scanline.
Re: (Score:2)
You need to stop talking, and google how CRTs work. You have demonstrated to be making multiple wrong assumptions such as for example one I quote above, and then going from these silly conclusions towards even sillier outcomes.
Hint: CRT unlike LCD does not display a permanent image. There is no frame. There's merely a scanline.
Why does it feel like I'm having a conversation with ChatGPT? If you have a specific objection to specific information please feel free to quote it directly and state specifically with any necessary supporting credible objective evidence why you believe it is incorrect otherwise I can't respond to empty and or nebulous claims.
You said "Lag doesn't measure what you think it does. It measures how long from a time when signal was sent to the monitor does it take for monitor to draw the frame". Now you are sa
Re: (Score:2)
I suspect the reason for it is that just as ChatGPT, I can offer you snippets of facts that go against an apparently popular delusion that CRTs work just like LCDs.
It's somewhat hilarious to me, considering the time I originally spent tweaking CRT monitors to get them to look just right, so I have a bit more information about how they work than an average guy who grew with them. And yet so many legitimately believe in things like "it displays frames" and "is has at least same if not more lag because it take
Re: (Score:2)
I think you need to stop talking and listen to what people are telling you.
You're insisting that the correct measurement for lag is based upon the time for the time taken for a signal to be received and a frame drawn. You're sort of correct, but when people are talking about the implications of this you're lah-lah-lahing and ignoring what they have to say.
From the point of view of an application, the time taken is the time between changing a pixel in a framebuffer and that change showing up on screen. For m
Re: (Score:2)
The people are telling me that CRTs have pixels. They are telling me that CRTs display frames.
They are objectively, factually wrong on the most fundamental basics of technology. I do not take advice on planetary orbits from flat earthers for the exact same reasons I have on interest in taking advice from people who believe aforementioned things about CRTs.
The fact that you had to run "but muh framebuffer", an issue that is completely and utterly irrelevant to the issue of display induced lag demonstrates th
Re: (Score:3)
Re: (Score:3)
May also be interlaced. Then it is 2x that.
Re: (Score:2)
Re: (Score:3)
Lag and scan rate are two different things. The latter is how long it takes the current frame to be displayed, where as lag is how long it takes the *correct* frame to be displayed. Any lag at all by its nature will result in a frame delay of at least scan rate + 1 for fully synchronous frames.
But frames don't need to be vertically synchronised, you don't need to wait to the end of the frame to start displaying the next one.
Re: (Score:2)
This is another issue that people used to LCDs don't understand. Scanning CRTs just update frame as soon as it's sent. And in rapid motion on analogue tech with imperfect geometry, that actually doesn't look as awful as it does on perfect geometry of LCD. As a result tearing is much less of an issue on a CRT.
Re: (Score:2)
Re: (Score:2)
No, that is inherent to trying to get three separate beams to converge at the same spot without exciting any of the surrounding area. This is why this happens even with digital input into CRT.
Latter point is irrelevant to the subject.
Honestly the amount of people who assume CRTs function like LCDs in fundamental ways is astounding. It's very visible that people just forgot how CRTs work in the twenty years of their absence, and are now projecting LCD basics on CRTs. Because you're far from alone in these hi
Re: (Score:2)
Re: (Score:2)
I mean if you're going to argue that then we should also include the speed of light limiting electron movement across the cable. Unlike everything else discussed in this thread the delay to end of field is by every sense of the word insignificant.
Re: (Score:2)
Re: (Score:2)
What? Lag on a typical CRT is 12-17 milliseconds. More like 35 ms for an actual TV. The phosphors themselves had a response time of several milliseconds.
Re: (Score:2)
"Phosphors" are not pixels, and "they" do not have a "response time". This is more of projection of inherent features of LCDs upon technology that doesn't have these features.
Re: (Score:2)
Ha ha. Sure. Did you never look at a CRT with a magnifying glass? Or sit close to one? Or have a shitty one from the early nineties?
Here's a picture:
https://commons.wikimedia.org/... [wikimedia.org]
And another:
https://it.wikipedia.org/wiki/... [wikipedia.org]
Totes not pixels man. It's all a conspiracy by big LCD!
Re: (Score:2)
Correct. Not pixels. Dots. That aren't arrayed like pixels either. That's the backporting of terminology from LCD age. Just like all the people who claim that CRTs display frames, and take a long time doing so because it takes a while to draw a frame. They don't know that CRTs are inherently incapable of displaying frames, and instead display scanlines with most of the screen staying dark at any given time.
Re: (Score:3)
Dots. Lol.
You're the one who mentioned pixels. And no, they weren't "backported" from LCDs. The world "pixel" was first published in 1965 in connection with scanned images from space probes. Similar words are even older.
CRTs display frames. They paint the screen once per set time period by scanning an electron beam over phosphors. That's quite similar to an older passive matrix LCD that paints the screen once per set time period by electrically scanning a row/column addressing system over each element once
Re: (Score:2)
>CRTs display frames.
I encourange you to try to find an image of a reasonably fast camera taking images of a CRT displaying a frame. Hint: unless it's fairly high exposure time, you won't be able to. Because CRT is inherently incapable of displaying a frame. Most of the screen is dark at any given time, and only scanline and faint trail after it is visible.
And no, CRT dots are not pixels, and they do not match pixels from the image being sent to them to display. This is why CRTs do not have native resolu
Larger Screen = More Lag (Score:2)
This is why lag on CRTs is measured in single digit nanoseconds.
At that point how far away you sit from the screen will be a significant contribution to your lag since light only travels 30cm in 1ns. A larger screen here will therefore have more lag because you have to sit further from it to see all of it.
Re: (Score:2)
but then you have to wait for the refresh scan, that aint happening in nanoseconds silly boy.
Input lag and refresh rate are two different terms describing different things. By definition lag is the wrong frame displayed during the scan. A CRT will display the frame as soon as it's available.
Re: (Score:2)
That's good because there are no pixels in CRTs. Though a lot of people apparently backported LCD terms into CRT after the fact, so you can find babble about dots as "pixels" in a lot of articles.
Fucking hilarious for someone who actually grew with the tech.
But you even believe that CRTs display frames. Like a lot of other people here. When they are literally incapable of displaying frames. Technology is fundamentally unsuitable for it. It can only display scanlines.
Re: (Score:2)
Re: (Score:2)
That is because I am not "blathering bullshit". What is actually happening is that you are without insight.
Re: (Score:3)
Re: "Authentic, lag-free display capabilities" (Score:2)
What marketer or salesman is in the article?
Re: (Score:2)
It is just bullshit. As many people are clueless and unable to fact-check, they get impressed by such nonsense.
Re: (Score:2)
Re: (Score:2)
Must come with an NFT, certificate of authenticity.
Re: (Score:2)
"Authentic" display capability? What the heck does that even mean?
Blurry. Blurry with barrel distortion.
Re: (Score:2)
Ageing CRTs do lose their sharpness - I took my Trinitron to an electronics recycler when it became too fuzzy.
Apparently they can be re-calibrated by tweaking the internals but Googling that suggests "There are parts inside of a CRT monitor that can kill or seriously injure you, even if the monitor has been unplugged for months".
Eeek!
Re: (Score:2)
Ageing CRTs do lose their sharpness - I took my Trinitron to an electronics recycler when it became too fuzzy.
Apparently they can be re-calibrated by tweaking the internals but Googling that suggests "There are parts inside of a CRT monitor that can kill or seriously injure you, even if the monitor has been unplugged for months".
Eeek!
The tube itself is basically a giant capacitor. I once got knocked halfway across the room by a 12-inch black-and-white, and learned a healthy respect for electricity — just another one of the many reasons I can't stand CRTs.
Re: (Score:2)
Yes.
If you've ever worked on a CRT, the first step is to discharge the anode to ground. If you don't, there's several hundred thousand volts sitting there to surprise you.
Re: (Score:3)
It means, for starters, black is black, and not just a really dark grey.
Re: (Score:2)
Most modern laser projectors do the same
colors not experienced by this generation (Score:5, Funny)
Re: (Score:2)
Yeah, I remember those greenish flesh-tones from the 1970s. Can't say I miss some of those colors!
Re: (Score:2)
Among about a dozen other picture adjustments, our first color TV had a helpful "Tint" knob. It allowed you to select any skin tone you preferred from the available choices of purple, green, or purplish-green.
It also featured "instant on" technology. It achieved this by keeping about 150W of vacuum tube heaters turned on 24x7.
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
I do miss illegal NTSC colors.
Illegal? Enlighten us, please (really!).
What I remember from the good old days of Amiga/Atari ST fights, there was also the eternal PAL/NTSC battle. I believe at the time, PAL was considered superior - with NTSC having alternative meaning "Never Twice the Same Color"
Re: (Score:3)
Re: (Score:2)
Re: (Score:3)
It's a riff of the Tears in the Rain monologue from Blade Runner (Rutger Hauer...RIP). It's a thinker...but funny.
Re: (Score:2)
The first gen of lcd monitors (those 5:4 ones) were lower resolution (1280x1024) than the previous generation of crts (1600x1200)
Re:colors not experienced by this generation (Score:4, Insightful)
Maybe not LCDs,
Let me stop you there. The OP's joke is either ignorant, outdate, or purposefully incorrect to try and get the funny side out. The original sRGB gamut was modelled on the capabilities of CRTs of the day. We've been making LCDs which exceed the colour gamuts of CRTs for 25 years now, and these days you can get a DCI-P3 gamut display for a couple of hundred dollars which leaves even wide gamut high end CRTs of back in the day in the dust in colour performance terms.
The only area LCD doesn't outperform CRTs in visual quality is in black level (contrast ratio). People have this rosy view of CRTs that isn't at all reflected in how the technology used to work, largely because early LCDs were rubbish and that sort of cemented the reputation.
X-Ray Colours (Score:2)
Re: (Score:3)
The NTSC gamut has been met and exceeded for over a decade now by common LCD displays. Even your phone likely has a better gamut. The iPhone 4S was 99% NTSC gamut, the only reason i
If modern displays were built for 480i (Score:2)
They would be lag-free too.
Re:If modern displays were built for 480i (Score:5, Informative)
First of all, no. LCDs inherently require some post processing to display the output, which means that there's no way for them to be "lag free" i.e. have nanosecond grade lag. They're pretty much stuck in a few millisecond range at best. OLEDs are better but still not a match for CRT.
Also CRT displays don't have fixed resolutions the way LCD and LED based displays do. They're analogue.
I used to overclock all my CRT displays back in the day for higher refresh rates for example, and most could push them very high. Though many would give off annoying coil whine when they were close to the limit, not to mention heating issues. Ability to display resolutions was also usually more than the spec, and since they didn't have a native resolution they would display all resolutions equally accurately.
Re: (Score:2)
In some respects, such as the processing time of LCDs, you're right. But no, CRTs couldn't display "any" resolution because they were analog. The colors were analog, yes. But driving the screens at higher-than-supported resulted in blurry pictures. There certainly was a limit to their resolution, and that limit was pretty small compared to today's 8K monitors.
Re: (Score:2)
Any resolution, no. But they could generally do "resolution x at n hz, resolution y at m hz, resolution z at o hz" where resolutions would go up and refresh rates go down. Essentially the problem was that speed of scanning was about the same, so you could choose to do less sweeps for lower resolution but more times per second, or more sweeps for higher resolution but less times per second.
This obviously had an upper limit, but that was usually pretty high. Especially for screens at the end of CRT cycle. Big
Re: (Score:2)
But it got blurry at the higher resolutions. The beam amplifier had limited bandwidth, the beam has a finite size and of course the phosphor dots were patterned on the screen which also limited the maximum resolution.
Re: (Score:2)
It was always blurry. Inaccurate geometry is one of the inherent flaws of CRT technology. Another one is convergence. That's when component colors of any single point at the corners of the screen diverged in terms of position, resulting in very specific kind of separation of geometry for each color that is utterly absent in pixel perfect technologies that actually reserve a pixel for each input pixel displayed in their native resolution.
It's equally funny how nowadays people are used to calling phosphor dot
Re: (Score:2)
It was always blurry.
Blurry as in relation to the underlying tech. For example on a good workstation monitor, e.g. an SGI, it would do pretty decent, sharp text at 1600x1200 which was IIRC the design resolution. At lower res, e.g. when y resolution switched for an old game (I had it hooked up to a PC running Linux), you could clearly see the individual pixels of the text.
It's equally funny how nowadays people are used to calling phosphor dots "pixels" even though they don't ever map to pixels being display
Re: (Score:2)
The thing with blurriness is that it's actually inherent to CRT technology. Yes, it can still be sharp enough to see input pixels, absolutely. But it will never be as good as native resolution LCD, because there will always be minor geometry issues. This actually got much worse towards the end of CRT era as flat screens became fashionable requiring more and more trajectory correction being applied at corners of the screen. Earlier CRTs had screen shape close to being a perfect distance from the gun at every
Re: (Score:2)
Lol. Are you a vinyl enthusiast too?
Unless you were sitting in front of your CRT with the mask and phosphor layers somehow removed and shining the electron beam into your eyeball, yes they absolutely have a native resolution. You just don't notice so much because you get automatic "antialiasing" courtesy of the bleed between phosphors.
Re: (Score:2)
Ah yes, "the phosphors". Because as we all know, all technology has pixels, they're just arrayed differently.
Truly, "what do you mean different operating principles, it was always touch screens, yes trackball obviously has multitouch... somewhere" analogue s great to listen to.
Re: (Score:2)
Ah yes, as we get older our memories smear together into a general "the golden age of yore" hey?
I remember my first CRT. Pixels big enough and a shitty enough shadow mask that the screen door effect was clearly visible from the working distance.
Re: (Score:2)
No, you just forgot the period when we switched, and everyone was raving about "being able to display pixel perfect geometry on LCDs, while our CRTs can't display pixels at all".
Re: (Score:2)
Your posts suggest that you don't really know much about the technology. I'm not really surprised you were hanging out with and listenting to people who also didn't know much about the technology twenty years ago.
Re: (Score:2)
Flat earthers also think that explaining to them that it's not in fact flat suggests that person doing the explaining doesn't understand much about physics.
Re: (Score:2)
Well, keep digging that hole and I guess you will eventually prove that the Earth is indeed round, by coming out in China.
Re: (Score:2)
Also CRT displays don't have fixed resolutions the way LCD and LED based displays do. They're analogue.
That's not-quite-right in a way that's misleading. CRTs have fixed refresh rates, sometimes for the more advanced multisync displays, they are forgiving, and you can drive them at rates that are close enough that the circuitry will sync up for a surprisingly liberal interpretation of, "close enough."
But then in color CRTs, there is the shadow mask, and that most definitely sets a pixel pitch. Look closely enough on a tube that has one of the venerable Sony Trinitron tubes in it, and it looks rather like a
Re: (Score:2)
>CRTs have fixed refresh rates
They do not. You can in fact find refresh rates at input resolutions supported in the manual of pretty much every CRT monitor since at least 1990s. By late 1990s, it became a selling point.
P.S. Trinitron is proprietary Sony technology, specifically aimed to solve a problem of geometry on more flat CRTs.
Re: (Score:2)
CRTs have fixed refresh rates, sometimes for the more advanced multisync displays,
Not the multisync ones. They had a very wide range f H and V frequencies.
Re: (Score:2)
Scroll down to the different console versions, it shows all the resolutions they are capable of.
Re: (Score:2)
I'm sorry to inform you that you have failed to measure half a dosen LCDs. Lag in question is built into technology itself, and well documented. Including by the very manufacturers of the panels in question.
If you truly believe that you have somehow managed to get your LCDs to show 1ms or lag, you should immediately patent your methods, and then sell it to one of the display majors. You'll become a multi-millionnaire overnight.
Heavy (Score:2)
Early 2000s, I had a 32" Aiwa widescreen CRT TV. Thing weighed as much as a big block Chevy engine block (220 lbs). It was a freakin' beast and I left it behind when we moved cross country.
I hope some retro gamer has lucked into it.
Lag-free? WTF? (Score:2)
Apparently, these people have no clue how a CRT works.
Re:Lag-free? WTF? (Score:5, Interesting)
unless you are claiming that there is lag literally between scanlines, because the process was: video card reads a byte of memory, produces and pixel sends pixel out to a crt that is completely in sync with that sending, altering its electron guns based on the live signal and not any imaginary buffer of ignorance
Re: (Score:2)
Ah, so for fake values of "lag free". Sure.
Re: (Score:2)
Think about what you just said. Sure, if you wrote a byte to memory at just the right time, then it might end up on the screen pretty fast. Might, if you can actually write a byte that close to it being read, and the video memory isn't double buffered and etc.
Even then, who cares how long it takes a carefully timed pixel to get to the screen? If you missed your write then you've got to wait a whole scan to get another try.
Re: (Score:2)
Re: (Score:2)
Sure. And trying to render in a hurry like that generally causes nasty artifacts, which is why actual graphics are usually double buffered. Nasty artifacts due to haste like the one that seems to be running in your head.
Not Lag Free (Score:2)
yes, lag free
No, because with a screen that large you'll have to sit far enough away that you'll be getting ~10+ns of lag from the light travel time.
Menu burn-in (Score:4, Funny)
Imagine the burn-in after displaying the restaurant menu on-screen for 40 years.
Anything special? (Score:2)
Some retro things really ... (Score:2)
... ought to _stay_ retro.
I get that rescuing this CRT totally makes sense for historic reasons and I applaud the effort.
However, claiming that shitty steam-age technology is the only "real authentic" way to experience modern conserved media is plain and utter nonsense. The vinyl retro-craze is a prime example. The amount of hardcore hacks and tricks the record industry had to go through to make vinyl even remotely feasible is mind-boggling. Anyone claiming that vinyl is superior in any metric vis-a-vis mod