CRT luminofor fading simulation

Discuss emulation of the Nintendo Entertainment System and Famicom.

Moderator: Moderators

User avatar
Eugene.S
Posts: 317
Joined: Sat Apr 18, 2009 4:36 am
Location: UTC+3
Contact:

Re: CRT luminofor fading simulation

Post by Eugene.S »

Googling for "CRT phosphor emulation" and I've found this video:
Emulation of CRT Phosphor + Curved Screen + Fade
But I can't test this GPU shader on my old gma4500

Can anyone test this filter on Stars - Field demo (PD).nes?
zzo38
Posts: 1096
Joined: Mon Feb 07, 2011 12:46 pm

Re: CRT luminofor fading simulation

Post by zzo38 »

Do you have some Csound program which can emulate luminofor fading for a single pixel? If it is treated as an audio signal, I wonder what it will sound like?
(Free Hero Mesh - FOSS puzzle game engine)
lidnariq
Posts: 11430
Joined: Sun Apr 13, 2008 11:12 am

Re: CRT luminofor fading simulation

Post by lidnariq »

CRT phosphors emit light as either a first or overdamped second order exponential decay¹ after the incoming electron beam radiation has stopped. There will be some complication for the electron beam size, such that a "single" phosphor may actually be struck multiple times in quick succession as the screen is redrawn. Vectorscopes have a different retrace pattern, but are subject to the same caveats.

¹ 'Properties of Fast-Decay CRT Phosphors' by Pfahnl, ftp.helpedia.com
User avatar
Zepper
Formerly Fx3
Posts: 3262
Joined: Fri Nov 12, 2004 4:59 pm
Location: Brazil
Contact:

Re: CRT luminofor fading simulation

Post by Zepper »

What do you think?
Attachments
demo02.png
demo02.png (18.71 KiB) Viewed 6919 times
demo01.png
demo01.png (15.38 KiB) Viewed 6919 times
User avatar
tokumaru
Posts: 12427
Joined: Sat Feb 12, 2005 9:43 pm
Location: Rio de Janeiro - Brazil

Re: CRT luminofor fading simulation

Post by tokumaru »

Looks like motion blur instead of light trails...
User avatar
Eugene.S
Posts: 317
Joined: Sat Apr 18, 2009 4:36 am
Location: UTC+3
Contact:

Re: CRT luminofor fading simulation

Post by Eugene.S »

Looks like motion blur instead of light trails...
Yes.

On the my first video you can see that white stars have greenish phosphoric tails.
This effect is similar to "afterglow" but no blur.

I need to make more quality video.
When you see live on real CRT traces looks much longer, brighter and phosphoric.
User avatar
Zepper
Formerly Fx3
Posts: 3262
Joined: Fri Nov 12, 2004 4:59 pm
Location: Brazil
Contact:

Re: CRT luminofor fading simulation

Post by Zepper »

I see. At anyway, even for a single pixel, it's brighter and "bigger" due to his properties in a CRT monitor.
User avatar
tokumaru
Posts: 12427
Joined: Sat Feb 12, 2005 9:43 pm
Location: Rio de Janeiro - Brazil

Re: CRT luminofor fading simulation

Post by tokumaru »

Zepper wrote:At anyway, even for a single pixel, it's brighter and "bigger" due to his properties in a CRT monitor.
Yes... this will be pretty hard so simulate in emulators.
lidnariq
Posts: 11430
Joined: Sun Apr 13, 2008 11:12 am

Re: CRT luminofor fading simulation

Post by lidnariq »

Having done a very simple simulation assuming a half-life of 33ms, without any of the electron beam simulations, the effect is fairly subtle:
f029.png
f029.png (7.85 KiB) Viewed 6907 times
This is just "current displayed frame" = maximum("previous displayed frame" * .7,"current notional frame")

(This is a still from the loop that bisqwit made to demonstrate animmerger. Any obvious errors are from it.)
User avatar
Zepper
Formerly Fx3
Posts: 3262
Joined: Fri Nov 12, 2004 4:59 pm
Location: Brazil
Contact:

Re: CRT luminofor fading simulation

Post by Zepper »

I'm shifting right (div by 2) the previous RGB pixel to simulate an exponential decay. Then, I'm adding the new pixel. Well, I see it produced a similar effect from the previous screenshot... ;)
User avatar
HardWareMan
Posts: 209
Joined: Mon Jan 01, 2007 11:12 am

Re: CRT luminofor fading simulation

Post by HardWareMan »

You're missing a few points:
1. If the new point brighter than the old one - the new one is drawn just as they are.
2. If the new point darker than the old one - then the strength of the effect will depend on the difference in brightness of the old and the new point. Therefore, the maximum effect will be viewed as a new point - black.
3. Each one has a different color phosphor decay rate. I have already given a link to a video showing it. First fades blue phosphor, then green and red kept the longest. Therefore, it is necessary to work on the sub-pixels separately.
lidnariq
Posts: 11430
Joined: Sun Apr 13, 2008 11:12 am

Re: CRT luminofor fading simulation

Post by lidnariq »

HardWareMan wrote:3. Each one has a different color phosphor decay rate. I have already given a link to a video showing it. First fades blue phosphor, then green and red kept the longest. Therefore, it is necessary to work on the sub-pixels separately.
It looks to me like the half life depicted in that recording is on the order of 100µs for blue, 300µs for green, and 2ms for red. With modern phosphors with those halflives, there will be no visible interframe blurring from the phosphors, only from your eyes.
User avatar
Zepper
Formerly Fx3
Posts: 3262
Joined: Fri Nov 12, 2004 4:59 pm
Location: Brazil
Contact:

Re: CRT luminofor fading simulation

Post by Zepper »

So, it's not a matter of interpolating frames or blending pixels at every X ms.
Should I say "outputting pixels whiter than white"? ;)

Funny things ^_^;;
Attachments
funny2.png
User avatar
mikejmoffitt
Posts: 1353
Joined: Sun May 27, 2012 8:43 pm

Re: CRT luminofor fading simulation

Post by mikejmoffitt »

Zepper wrote:So, it's not a matter of interpolating frames or blending pixels at every X ms.
Should I say "outputting pixels whiter than white"? ;)

Funny things ^_^;;
This looks nasty, but it's closer than the other approximations above, which are more like motion blur.

Most of what makes the fading phosphor look look like it does is phosphors not going dark instantly.
Eugene.S wrote:Googling for "CRT phosphor emulation" and I've found this video:
Emulation of CRT Phosphor + Curved Screen + Fade
But I can't test this GPU shader on my old gma4500

Can anyone test this filter on Stars - Field demo (PD).nes?
This looks more like an LCD response time emulator, ironically, from the video.
User avatar
Eugene.S
Posts: 317
Joined: Sat Apr 18, 2009 4:36 am
Location: UTC+3
Contact:

Re: CRT luminofor fading simulation

Post by Eugene.S »

Image

Oh, mein gott :shock:
It's the LSD emulator with psilocybine mushrooms! :lol:

I want to see it on motion
Last edited by Eugene.S on Wed Jun 19, 2013 3:28 pm, edited 1 time in total.
Post Reply