Using the 2nd Field of Interlaced Frame for Extra V-Blank in 480i Mode

Discussion of hardware and software development for Super NES and Super Famicom.

Moderator: Moderators

Forum rules
  • For making cartridges of your Super NES games, see Reproduction.
Post Reply
User avatar
Nikku4211
Posts: 381
Joined: Sun Dec 15, 2019 1:28 pm
Location: Bronx, New York
Contact:

Using the 2nd Field of Interlaced Frame for Extra V-Blank in 480i Mode

Post by Nikku4211 » Fri Dec 04, 2020 4:01 pm

YOOOOOOOOO I JUST GOT THIS MAD IDEA

You know how the SNES has the little-used interlace setting? Where each frame pair is combined into 2 fields of a single frame in any mode that isn't 5 or 6?

If you don't know what interlace is, Captain Disillusion has done a video on it.

What if we made the 2nd field completely blank, using the entire 2nd field for extra computations that require forced blanking, so that we can render on the 1st field?

Since it's interlaced, the screen shouldn't blink, but rather just display the 1st field of the frame with the 2nd field acting as a window blinds effect, reducing the epilepsy problem.
I have an ASD, so empathy is not natural for me. If I hurt you, I apologise.

lidnariq
Posts: 10265
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Using the 2nd Field of Interlaced Frame for Extra V-Blank in 480i Mode

Post by lidnariq » Fri Dec 04, 2020 4:25 pm

That's not what interlacing is.

At absolute best, on a modern TV that tries to deinterlace your input, you'll end up with "comb" artifacts and a 30fps result. On an old CRT TV you'll see a very visible flicker at 30Hz.

93143
Posts: 1314
Joined: Fri Jul 04, 2014 9:31 pm

Re: Using the 2nd Field of Interlaced Frame for Extra V-Blank in 480i Mode

Post by 93143 » Fri Dec 04, 2020 4:27 pm

The first and second fields are basically 240p frames timing-wise, just offset by half a scanline. The whole field is drawn, then VBlank happens and the other field is drawn. (This is the normal way for a CRT TV to operate. 240p signals just force the same field to be drawn every frame, which leaves the other field blank and results in what emulator video filter option pages refer to as "scanlines" and what you've referred to as a "window blinds" effect.) Your scheme is functionally identical to blanking the screen every other frame.

Sorry.

It might be possible to do a sort of 240i approach, where every other scanline is blanked and you switch which ones every frame. This would probably not look nearly as good as either 240p or 480i, but it would increase your blanking time massively...

User avatar
Nikku4211
Posts: 381
Joined: Sun Dec 15, 2019 1:28 pm
Location: Bronx, New York
Contact:

Re: Using the 2nd Field of Interlaced Frame for Extra V-Blank in 480i Mode

Post by Nikku4211 » Fri Dec 04, 2020 4:57 pm

lidnariq wrote:
Fri Dec 04, 2020 4:25 pm
That's not what interlacing is.

At absolute best, on a modern TV that tries to deinterlace your input, you'll end up with "comb" artifacts and a 30fps result. On an old CRT TV you'll see a very visible flicker at 30Hz.
Those HDTVs with their combing are pretty misleading then. Bruh.
93143 wrote:
Fri Dec 04, 2020 4:27 pm
The first and second fields are basically 240p frames timing-wise, just offset by half a scanline. The whole field is drawn, then VBlank happens and the other field is drawn. (This is the normal way for a CRT TV to operate. 240p signals just force the same field to be drawn every frame, which leaves the other field blank and results in what emulator video filter option pages refer to as "scanlines" and what you've referred to as a "window blinds" effect.) Your scheme is functionally identical to blanking the screen every other frame.

Sorry.

It might be possible to do a sort of 240i approach, where every other scanline is blanked and you switch which ones every frame. This would probably not look nearly as good as either 240p or 480i, but it would increase your blanking time massively...
You shouldn't apologise for telling the truth. Lel, I haven't used a CRT since my family threw our last one out in 2017, and I don't remember testing the difference between 240p and 480i.

Would this 240i approach increase the blanking time even more than blanking every 2nd frame? :thonk:
I have an ASD, so empathy is not natural for me. If I hurt you, I apologise.

lidnariq
Posts: 10265
Joined: Sun Apr 13, 2008 11:12 am
Location: Seattle

Re: Using the 2nd Field of Interlaced Frame for Extra V-Blank in 480i Mode

Post by lidnariq » Fri Dec 04, 2020 6:25 pm

Nikku4211 wrote:
Fri Dec 04, 2020 4:57 pm
Those HDTVs with their combing are pretty misleading then. Bruh.
It's unfortunate, because this "weave" deinterlacer is basically pessimal for every other content ever transmitted, and yet that's the thing that they test against.

In a way, HDTVs dropping their RF and composite inputs might actually be a kindness, instead of assuming that everyone only is watching DVDs of interlaced telecined content.
Would this 240i approach increase the blanking time even more than blanking every 2nd frame?
No, because of the overhead of turning rendering on/off every other scanline. Otherwise it would be comparable.

User avatar
Nikku4211
Posts: 381
Joined: Sun Dec 15, 2019 1:28 pm
Location: Bronx, New York
Contact:

Re: Using the 2nd Field of Interlaced Frame for Extra V-Blank in 480i Mode

Post by Nikku4211 » Fri Dec 04, 2020 7:19 pm

lidnariq wrote:
Fri Dec 04, 2020 6:25 pm
Nikku4211 wrote:
Fri Dec 04, 2020 4:57 pm
Those HDTVs with their combing are pretty misleading then. Bruh.
It's unfortunate, because this "weave" deinterlacer is basically pessimal for every other content ever transmitted, and yet that's the thing that they test against.

In a way, HDTVs dropping their RF and composite inputs might actually be a kindness, instead of assuming that everyone only is watching DVDs of interlaced telecined content.
HDTVs might suck at deinterlacing, but at least some of them are compatible with composite.

I'd still rather have mediocre composite support than no composite support, because if I wanted my consoles to work, I'd have to use a special adapter and it probably would have been just as terrible at deinterlacing because cheap.
lidnariq wrote:
Fri Dec 04, 2020 6:25 pm
Would this 240i approach increase the blanking time even more than blanking every 2nd frame?
No, because of the overhead of turning rendering on/off every other scanline. Otherwise it would be comparable.
Okay, then. Flicker it is.
I have an ASD, so empathy is not natural for me. If I hurt you, I apologise.

Post Reply