24" Widescreen CRT (FW900) From Ebay arrived,Comments.

Here are results of dot pitch measurements. A lot of work went into proper image processing to calculate the dot pitches. It took the code about 4 minutes to analyze each image and spit out a dot pitch. Basically I used a pure luminance image, and only analyzed the green phosphors. I fit the processed images to an oriented sine wave grating, defined as 1 if >0, and -1 if <0 to emulate a square wave grating. Thing is, you can't change the duty cycle of a sine wave. In future, I might fit an actual oriented square wave grating as duty cycle is a parameter in a square wave. I have a feeling that the fitting procedure will be much faster in that case (but not sure). Duty cycle will also give more insight into whether the dot pitch is increasing due to thicker phosphor strips, or to thicker spaces between the strips, or some combination of both (i.e. is the thickness of the grille wires increasing and/or the spacing between the wires). My eyes tell me that the black spaces are the same width, and its the phosphor strips that are getting wider. I could always measure with an image editor and counting pixels, but it's cooler to do it automatically :p

2zs4qhs.png


Couple things to note:

The increase in dot pitch is non linear. The measurements are taken at three equal spacings, yet the midcentre pitch is about 13 microns larger than centre, and the edge is about 24 microns larger than the mid centre. Also, these images were taken towards the bottom of the screen. I may redo them in the vertical centre to see if anything changes.

Also, the measurements are a tiny bit larger than the listed specs. This is because the listed specs indicate the aperture grille pitch, which is in between the electron gun and the phosphor mask. The aperture grille pitch is always smaller than the dot pitch, as the electron beams continue to diverge after going through the grille. See image below:

2vj9xkx.png
 
Last edited:
I get the impression that we had neared the limits of what was possible in terms of standard CRTs (I dunno much about FED and SED). Perhaps new phosphor technologies would have emerged, allowing wider gamuts and lower amounts of afterglow. I dunno how much more could be squeezed out in terms of electron optics - To achieve sharper focus you need to induce steeper angles so that the beam can be refocused at a sharp point. This requires sufficient physical clearance in the tube, specifically the neck. So focus is in large part limited by the neck diameter.

Then there is the deflection angle issue - with larger screens, the deflection angles are greater, and more distortion needs to be accounted for by the optics and circuitry. One solution is to have a deeper tube (i.e. increase distance from triode to phosphor screen), as this decreases the deflection angle necessary to cover the whole screen.

So yea, potentially we could have gotten more resolution with larger display sizes, but the size of the tubes would be fucking massive.
 
There are some papers out there with proposals for more DSP control over the deflection yoke to make it perfect. Who knows how that would have panned out. My opinion though, is that in the end, it's still electromagnetic and would probably still have some distortions.
 
roberta said:
2nd don't forget to adjust your focus pods with this tool thru the air vents in rear

Good Luck to all enthusiasts
Click to expand...


Great post and yes - I totally recommend these alignment tools. GC Tools also makes degaussers too, for all of those interested in doing OTHER monitors (don't use the degausser on the GDM's while they're powered on).

But I want to emphasize that in my experience with my GDM's, focus adjustment is not a regular procedure. I find myself almost never needing to use it. So unless it's poorly focused, I would just leave the pots alone. MEME pattern (spacediver should have a copy of a very fine MEME pattern - pm him) is your best bet. If you can read the M's and the E's without issues (especially on the pattern that he has), then you're good.
 
@igsux3
issue with adhesive are air bubbles that are unavoidable without proper tools like the ones used to repair LCD's. Maybe it would be possible to apply it manually without those pesky air bubbles but I doubt it. Without adhesive using something like water+sugar solution (which some people recommend for LCD) it is easy to push those air bubbles out.

I doubt that without any water solution it would look good. Come to think about it UV glue can be problematic if it is too hard to remove. Water + sugar might be ok to try for first time. Worst case scenario you will need to reapply it after some time.

As for 73% it is suspicious for polarizer to be this transparent. It should block about half of light, right? If it had 53% then it would be more believebable. Polarizing propeties... it is hard to say how it would look without them. Even if there were some other plastic darkening sheets I dount they would have better light properties than polarizers.

Je&#347;li chodzi o sam monitor to jak b&#281;dziesz w lubelskim to jest do obejrzenia u mnie w domu. Mog&#281; go nawet sprzeda&#263; za kilka tysi&#281;cy z&#322;otych jak si&#281; mocno spodoba ;)
 
spacediver said:
So yea, potentially we could have gotten more resolution with larger display sizes, but the size of the tubes would be fucking massive.
Click to expand...

Yeah, the 40-inch tubes made by Sony were basically impossible to move with less than three people.

I wonder about what could have been with rear or front projection though. From the sounds of what you're saying though, you'd still need to build bigger tubes to get greater resolution, which would make the projection systems heavier than they already are.
 
From what I understand, both rear and front projection technologies simply optically magnify the image created on the phosphor screen(s). The resolution of the magnified image is limited by the resolution of the original image.
 
Been doing some more work on image rendering. There's room for improvements - I'm too lazy to generate a new calibration matrix based on camera filter data, so this is using the old one I created way back. But it's pretty close. And I remeasured the XYZ values of my display primaries, and created an XYZ - RGB matrix from that information, and am using that instead of the standard XYZ-sRGB matrix

The toughest part is figuring out the best exposure to use, and then intelligently normalizing the images.

Here's the full size image of the phosphors, taken at vertical centre of screen. Keep in mind I've traded half my sensor resolution for image accuracy. Also keep in mind that this is best viewed with a gamma of 2.4.
 
EnhancedInterrogator said:
Yeah, the 40-inch tubes made by Sony were basically impossible to move with less than three people.

I wonder about what could have been with rear or front projection though. From the sounds of what you're saying though, you'd still need to build bigger tubes to get greater resolution, which would make the projection systems heavier than they already are.
Click to expand...

Sony's G-90 projector can scan up to 150khz. GDM-FW900 can only go up to 121 kHz. No idea what the realistic resolution limit of the G-90 is but from what I've read 1080p is absolutely a joke to the projector and it handles it with no issues.

Edit - and some of the last Mitsubishi rear projection TV's had 9-inch tubes - same size as the G-90 tubes. Again - I don't know what the realistic resolution limit is on the television but something in my mind tells me that 1080p would be nothing for it.
 
jbltecnicspro said:
MEME pattern (spacediver should have a copy of a very fine MEME pattern - pm him) is your best bet. If you can read the M's and the E's without issues (especially on the pattern that he has), then you're good.
Click to expand...

Here's a copy. Created from scratch in Matlab, using this code:

Code: 
M = [ones(1,10);repmat([1,1,0,0,1,1,0,0,1,1],4,1)];
E = [ones(1,10);1,1,zeros(1,8);ones(1,10);1,1,zeros(1,8);ones(1,10)];
W = flipud(M);
RevE = fliplr(E);
HorGap = zeros(5,2);
VertGap = zeros(1,24);

MemeUnit = [M,HorGap,E,HorGap;VertGap;RevE,HorGap,W,HorGap;VertGap];

MemePattern = repmat(MemeUnit,100,80);
MemePattern = MemePattern*255;

MemeImage = uint8(MemePattern);

imwrite(MemeImage,'MEME_1920.png', 'png')
 
Last edited:
Last edited:
spacediver said:
I get the impression that we had neared the limits of what was possible in terms of standard CRTs (I dunno much about FED and SED). Perhaps new phosphor technologies would have emerged, allowing wider gamuts and lower amounts of afterglow. I dunno how much more could be squeezed out in terms of electron optics - To achieve sharper focus you need to induce steeper angles so that the beam can be refocused at a sharp point. This requires sufficient physical clearance in the tube, specifically the neck. So focus is in large part limited by the neck diameter.

Then there is the deflection angle issue - with larger screens, the deflection angles are greater, and more distortion needs to be accounted for by the optics and circuitry. One solution is to have a deeper tube (i.e. increase distance from triode to phosphor screen), as this decreases the deflection angle necessary to cover the whole screen.

So yea, potentially we could have gotten more resolution with larger display sizes, but the size of the tubes would be fucking massive.
Click to expand...

My understanding is that SED/FED were PWM driven like PDP and DLP, instead of being sent discrete values like CRT, LCD, LCOS or OLED which killed any interest that I had in the technology.
For televisions, I think that once there are RGB OLED displays on the market with backlight strobing options, no-one will be longing for CRTs, CRT-like technology, or Plasmas any more.
I doubt there are many people using CRTs for video at all now anyway. Most CRT enthusiasts seem to want them for gaming, not video.

Using CRTs as general-purpose monitors seemed like a dead end to me.
You get perfect geometry, a brighter image, higher resolutions, and no flicker at all from a flat panel.
I'm surprised that LAGRUNAUER claims to be selling/have been selling a lot of FW900s to post-production houses. The ANSI contrast is so bad on CRTs that I wouldn't attempt to do any kind of post work or video/photo editing on them.

The reason to use a CRT today is that it keeps moving objects in focus, instead of blurring like a flat panel does. And it does so with very few artifacts, unlike a strobed LCD.
Is the distortion that a CRT monitor has really that much of a concern for gamers? It never seemed like a big problem to me on even the cheapest CRT monitors. They usually offered enough control that you wouldn't have problems in a game, even if things are not perfect.

flod said:
is there a standard meme pattern? for instance see pg 22 and pg 50 of this: http://ftp.cis.nctu.edu.tw/csie/Sof..._Document_Center_Monitor_Interface/dstp10.pdf
Click to expand...
That seems like a far more useful pattern than the one posted above, and was easy enough to recreate.
Just set this as your background and tile it:

 
flod said:
is there a standard meme pattern? for instance see pg 22 and pg 50 of this:

http://ftp.cis.nctu.edu.tw/csie/Sof..._Document_Center_Monitor_Interface/dstp10.pdf

pg 36 of this has another
http://easp.optomausa.com/ASP/uploa...DDP2230-USB, 36.88N03G001) Service Manual.pdf


more:
http://www.quantumdata.com/pdf/isa_quick_start.pdf
no images here but apparently there's "MEME Sony" and "MEME Plus" patterns

http://www.manualsdir.com/manuals/144482/sencore-vp300.html?page=15
Click to expand...

MEME Sony is just an MEME pattern that's spaced a little more than the one I gave spacediver. MEME plus is just an MEME pattern with plus signs in between each set.
 
SED/FED have nothing to do with CRT other than fact that in both types of technology electron is exciting phosphor. Those were promising technologies but OLED (and related types) seems to have best potential for achieving ultimate picture quality.

CRT are absolutely terrible for desktop and brilliant for games and watching movies. That part was always true and still is. After buying my first LCD I stopped using CRT in desktop using it only for games and movies and what a relief it was for my eyes !!!

It is funny to see people use CRT for desktop and obsess for their sharpness and geometry. If those things are an issue then just use 4K LCD and be done with it. Is it that hard to get? :confused:
 
XoR said:
SED/FED have nothing to do with CRT other than fact that in both types of technology electron is exciting phosphor. Those were promising technologies but OLED (and related types) seems to have best potential for achieving ultimate picture quality.
Click to expand...
People seem to think that they were a "flat CRT" but if they were using PWM that's more like an alternative to PDP.

XoR said:
CRT are absolutely terrible for desktop and brilliant for games and watching movies. That part was always true and still is. After buying my first LCD I stopped using CRT in desktop using it only for games and movies and what a relief it was for my eyes !!!
Click to expand...
I'm not sure that I'd agree with CRTs being a good choice for movies now.
They have a good black level and motion clarity, but 24 FPS on a CRT is going to judder pretty badly.
Size, resolution, and sharpness are all a lot worse than a good flat panel too, and local dimming TVs are arguably higher contrast than a CRT now.
 
zone74 said:
That seems like a far more useful pattern than the one posted above, and was easy enough to recreate.
Just set this as your background and tile it:
Click to expand...

Agreed. I created mine based on the following photo, which jbl sent me of a pattern generated by his quantum data generator. I like the one you and flod posted as it has a wider variety of spacing between the elements. If I have time soon, I'll code it and upload it.

wcelq8.png
 
zone74 said:
I doubt there are many people using CRTs for video at all now anyway. Most CRT enthusiasts seem to want them for gaming, not video.
Click to expand...

Agreed that the ANSI contrast is far worse than the static contrast, but the FW900 is stunning for watching video, so long as you're not more than a few feet away from the screen. Apparently you can get 1080p at 96 hz on an FW900 for a judder free experience.

There's also something beautiful about the way that pixels are rendered on a CRT that makes them so great for photos and videos. The luminance profile of the pixel (due to the intensity profile of the cross section of the electron beam) means that there's some natural smoothing that takes place.
 
zone74 said:
Is the distortion that a CRT monitor has really that much of a concern for gamers? It never seemed like a big problem to me on even the cheapest CRT monitors. They usually offered enough control that you wouldn't have problems in a game, even if things are not perfect.
Click to expand...

If you're referring to the distortion of the pixel at the edges, it's certainly not noticeable. I'll be taking measurements of pixels soon so I can characterize it if it does exist- I wonder how pixel distortion due to wide deflection angles relates to geometry distortion.
 
spacediver said:
Agreed that the ANSI contrast is far worse than the static contrast, but the FW900 is stunning for watching video, so long as you're not more than a few feet away from the screen. Apparently you can get 1080p at 96 hz on an FW900 for a judder free experience.
Click to expand...
Movies are 24 FPS so with a low-persistence display you need 24Hz to eliminate judder. 24 FPS at 96Hz on a CRT will have bad judder.
I did a test recently with a CRT monitor (not an FW900) playing back movies at 72Hz with black frame insertion for an effective 24Hz and the results were very unexpected.
As you might expect, there is a significant amount of flicker at 24Hz. No sane person could tolerate watching a whole movie like this.
What I did not expect is that playing movies at 24Hz (effective) on a CRT makes motion super smooth, like an LCD with interpolation turned all the way up. Of course, unlike an LCD, the motion is crystal clear and there are no interpolation errors. A very surprising result.

For me, 24" is just too small to enjoy movies. I sit a couple of feet back from a 46" TV at the moment and would like something bigger. Having a huge screen (e.g. a 10ft projected image) filling your vision is a very different experience from sitting close to a small screen, even if it fills the same amount of your vision.

spacediver said:
There's also something beautiful about the way that pixels are rendered on a CRT that makes them so great for photos and videos. The luminance profile of the pixel (due to the intensity profile of the cross section of the electron beam) means that there's some natural smoothing that takes place.
Click to expand...
For viewing pretty images, sure. For editing/content creation, that's not a good thing.

spacediver said:
If you're referring to the distortion of the pixel at the edges, it's certainly not noticeable. I'll be taking measurements of pixels soon so I can characterize it if it does exist- I wonder how pixel distortion due to wide deflection angles relates to geometry distortion.
Click to expand...
I was really meaning geometry distortion. For desktop usage, imperfect geometry can certainly be annoying.
For gaming I don't think it's a big deal. The FW900 can be set up to have extremely good geometry, but even the cheapest crappiest CRT monitors are generally good enough that it shouldn't be a problem in games.
 
Last edited:
zone74 said:
Movies are 24 FPS so with a low-persistence display you need 24Hz to eliminate judder. 24 FPS at 96Hz on a CRT will have bad judder.
Click to expand...

Ah right, I forgot about frame tripling - yes, it would be noticeable when the camera pans with speed. For what it's worth, I watch all my stuff at 85 hz, and judder doesn't rear its head that often. But it is noticeable when it does, although I don't yet have enough experience to always distinguish the effects of judder from low source frame rate.
 
hi everyone.

I'm a proud new owner of a fw900. it had some scratches in the antiglare film, so I had to remove it. I used ispropilic alcohol and it came out almost completely, but there are some spots that it doesn't matter how much I rub it, it doesnt come out.

Dou you know anything stronger that the isopropilic alcohol for removing the las pieces of antiglare film???

Other option would be replace the glass. does anyone know where cloud I get a new one?

thanks!
 
14175mq.jpg


As for removing the remaining AG coating, I remember someone talking about using a hair dryer to help with the removal. Someone else can probably offer some advice.
 
hahaha

The dryer didn't worked. Thanks for the advice anyway :)

I think I need something stronger than the alcohol, which can disolve the remaining dots of the AG coating. Could athetone work?
 
zone74 said:
That seems like a far more useful pattern than the one posted above, and was easy enough to recreate.
Just set this as your background and tile it:
Click to expand...

May I ask what is wrong with the MEME pattern I gave spacediver? Nothing personal - but if there's a reason not to use it, then I'm all ears.

The thinking behind my using that MEME pattern is that the M's and E's are so compact that if your focus was good enough to resolve that pattern, then it was good for anything. If my thinking is wrong on this though, I'd love to be corrected. After all - we all are going to benefit from this kind of information. Thanks!
 
try Goo Gone thats what I used to get it all off ,get at hardware store order online its stinky works well .
acetone might work well too try it... your down to just glass now anything goes .go spacediver go
 
Last edited:
jbltecnicspro said:
May I ask what is wrong with the MEME pattern I gave spacediver? Nothing personal - but if there's a reason not to use it, then I'm all ears.

The thinking behind my using that MEME pattern is that the M's and E's are so compact that if your focus was good enough to resolve that pattern, then it was good for anything. If my thinking is wrong on this though, I'd love to be corrected. After all - we all are going to benefit from this kind of information. Thanks!
Click to expand...

well the horizontal features are 2px wide
 
If I'm not mistaken (would have to double check), in the MEME pattern I created, the only vertical gaps are 2 pixel wide, and the only horizontal gaps are 1 pixel wide.

The VESA recommendation has horizontal and vertical gaps of 1, 2, and 3 pixels each.
 
amlett said:
I think I need something stronger than the alcohol, which can disolve the remaining dots of the AG coating. Could athetone work?
Click to expand...
try some kind of label remover
If I remember correctly I used it to clean screen after removing AG
 
I was really afraid not to damage glass after removing AG, but my girlfriend was pretty certain that water with vinegar would help. She was actually as certain as I was afraid so I closed my eyes and let her do the duty. To my surprise 10 minutes later there was no single mark of glue :) of course YMMV, but I can surely tell that did the trick for me.
 
Thanks everyone for the advices. I have to try with a adhesive remover. Tried whit venegar and didn't do anything.

This are the dots I'm not able to remove no matter what I use.
http://i.imgur.com/Xy9pzx6.jpg

Funny thing is that the whole AG coating could be removed with alcohol, these parts... I don't know what to do.

Anyone selling a front caover with glass included? :D
 
XoR said:
CRT are absolutely terrible for desktop and brilliant for games and watching movies. That part was always true and still is. After buying my first LCD I stopped using CRT in desktop using it only for games and movies and what a relief it was for my eyes !!!

It is funny to see people use CRT for desktop and obsess for their sharpness and geometry. If those things are an issue then just use 4K LCD and be done with it. Is it that hard to get? :confused:
Click to expand...

Why do you say this? What is wrong with a CRT for desktop use? I can't stand the LCD motion blur; it makes scrolling a browser or document very unpleasant. If I got a strobed LCD, it would either be extremely expensive or TN-based. Either way, it would be in the horrid 16:9 ratio.
 
I am looking for a 334P06701 flyback. It is for a Gateway VX1120, which is a Mitsubihi 2060u rebrand. Does anyone know where I can obtain it?
 
Last edited:
amlett said:
Anyone selling a front caover with glass included? :D
Click to expand...

You can't just replace the glass. Look up CRT tube on google images. The entire tube is sealed together. Even if you could remove the faceplate part of the glass, you'd destroy the vacuum inside!
 
Ok last phosphor post for a while (I hope). This time, I remeasured the camera filters' responses to primaries, generated new custom transformation matrices, and normalized the min and max levels in the RGB domain. So this image provides good contrast, accurate colors, and preserves a good amount of detail.

I've also included an accurate scale indicating how big one millimetre is.

Here you go!

edit: because I've used a custom XYZ to RGB matrix, rather than the standard sRGB matrix, this image will look most accurate on a GDM Trinitron monitor :)
 
Last edited:
Hello all, hope everyone here is doing well and Happy Holidays!

Unfortunately I have lost interest and desire in this CRT due to some misfortunes upon receipt (which lead to being disillusioned with what I was getting myself into).

Anyway, any progress on real replacements for the CRT yet? How's OLED coming along? Any desktop news on that end?

How long are we going to wait, until 2017 at the earliest?

I just would like a comparable replacement already, the high end choices in the TN/IPS/etc world are so dismal and disheartening. It is crucial I have no input lag whatsoever for some of the games I play.

Sigh.
 
You must log in or register to reply here.
Back
Top