Enhanced ARX True High Resolution Display Driver

Any discussions related to the creation of new hardware or software for the ZX80 or ZX81
Moggy
Posts: 3269
Joined: Wed Jun 18, 2008 2:00 pm

Re: Enhanced ARX True High Resolution Display Driver

Post by Moggy »

Both work fine. :D
User avatar
stefano
Posts: 569
Joined: Tue Dec 11, 2012 9:24 am
Contact:

Re: Enhanced ARX True High Resolution Display Driver

Post by stefano »

It is good news, I suppose ?
I will leave the NOP in though, so it won't stumble when run on EightyOne.

I'm now trying to find a way to interlace two 64 pictures and get 4 grayscale levels (probably acceptable only on phosfor monitors, if I ever make it work).
It should exist already in WRX mode.

Would you mind to try also one of the samples built for the 64 rows mode ?
Moggy
Posts: 3269
Joined: Wed Jun 18, 2008 2:00 pm

Re: Enhanced ARX True High Resolution Display Driver

Post by Moggy »

You post it and I will try it! :D
User avatar
stefano
Posts: 569
Joined: Tue Dec 11, 2012 9:24 am
Contact:

Re: Enhanced ARX True High Resolution Display Driver

Post by stefano »

I posted a "sampler" one page ago (actually it is in page #3). It includes 64rows and 8K stuff (arx64*.p).
User avatar
stefano
Posts: 569
Joined: Tue Dec 11, 2012 9:24 am
Contact:

Re: Enhanced ARX True High Resolution Display Driver

Post by stefano »

While testing the new code on EO I noticed it could not work properly on 60hz sets. I don't remember how the old one was behaving, though.
User avatar
stefano
Posts: 569
Joined: Tue Dec 11, 2012 9:24 am
Contact:

Re: Enhanced ARX True High Resolution Display Driver

Post by stefano »

Updated timing values, they look correct on EO.

I'm also adding a new feature, at a very beta release level to deal with an intelaced picture, the artifacts are espected (see picture) and in this moment it gets useful to see black areas (compared to gray). Could anybody see how it looks on the real hardware ? I'm expecting a relevant flicker on a traditional TV and a better result on high persistency monitors (green phosphor, etc..)... if the picture keeps the sync :D
gray_arx.png
gray_arx.png (9.26 KiB) Viewed 445 times

The new test cases are with/without the break detection, 64/192 rows.
They all should be able to compensate the timing on 50hz and 60hz TV sets :oops:

This ideally implies 10 different tests (which should also be verified on different monitors), but obviously I'm not asking that much.
arx_test.zip
(10.77 KiB) Downloaded 20 times

EDIT: please ignore the greyscale mode, I think there's still too much missing for useful feedbacks.
Last edited by stefano on Fri Apr 12, 2024 1:53 pm, edited 1 time in total.
Moggy
Posts: 3269
Joined: Wed Jun 18, 2008 2:00 pm

Re: Enhanced ARX True High Resolution Display Driver

Post by Moggy »

I can test but only on LCD monitor as I no longer have a CRT monitor, also is this to be tested with UDG4ZXpand or just plain zeddy?
User avatar
stefano
Posts: 569
Joined: Tue Dec 11, 2012 9:24 am
Contact:

Re: Enhanced ARX True High Resolution Display Driver

Post by stefano »

Right, this makes a difference :( ..both?


..in the meantime I found a way to work this out:
zx81_arx_gscale.png
zx81_arx_gscale.png (8.24 KiB) Viewed 430 times

I'm afraid it won't be healthy for your eyes !
gmandel.p.gz
(2.19 KiB) Downloaded 26 times


I've already put the latest driver version online, I think it is already better than the previous ones.
User avatar
stefano
Posts: 569
Joined: Tue Dec 11, 2012 9:24 am
Contact:

Re: Enhanced ARX True High Resolution Display Driver

Post by stefano »

More about the gray library, I have 3 possibilities,:

the current driver shows a first frame 2 times, a second one only for 1 time.
This is in theory correct because the picture would allow a higher persistence to one of the two allowing a dark gray and a light gray.
By the way the 2 frames in a raster picture are interlaced. This uneven count will probably make the lines "bounce" up and down.
It can help in shuffling the overall result but perhaps it will bounce the frames making the picture more fuzzy.

Otherwise, I could avoid the unbalanced persistence. The result would be much like the screenshot I've pasted, two identical gray levels but slightly "shifted" plus black and white.

A further possibility could be to deal with a 6 frames counter trying to give every frame a preferred position. I think something would still need to bounce, but... well we should try and see.
User avatar
stefano
Posts: 569
Joined: Tue Dec 11, 2012 9:24 am
Contact:

Re: Enhanced ARX True High Resolution Display Driver

Post by stefano »

I think the best compromise is to cycle between 4 frames, a first one (light gray) appearing only 1 time and the second one (dark gray) being displayed 3 times (thus a white frame interposed only one time).

The result seems plausible, observing it on EO, (this is the best screenshot I could take, the dark and light levels look inverted because of the flicker)
z88dk_graytest.jpg
z88dk_graytest.jpg (15.77 KiB) Viewed 287 times
gray_arx.p.gz
(2.01 KiB) Downloaded 19 times
I've updated also the WRX version:
gray_wrx.p.gz
(1.97 KiB) Downloaded 14 times

Originally the WRX version was designed to reduce the flickering (sacrifying the 2 gray levels), but I think it was never tested on the real hardware.

Please let me know if the diamond looks dark or light.
Post Reply