Atari-Forum

I need to spend some time into looking in to the STE before I'm able to have any real insights, sorry. But while on the ST the GLUE and Shifter are sort of independent, the same is not as true in the STE. This is because the GLUE timings are the ones that makes "room" for the STE pixel shi...

Yes, exactly

This is not yet another thread about possibilities of border removal in mono!! Instead, I'm interested in the possibilities on STe around synclocking and then changing the video pointer mid frame. I'm finding it tricky to work out how to time this so it takes place in non-visible screen areas... am...

When it comes to helper utilities, coders are usually not very keen to release them. Or to create them in the first place.

THAT'S IT! Now you'll get no greetings



https://github.com/troed/PNGconv/blob/master/README.md

j/k - looks really useful!

Okay, next question. Did anyone do a game/demo with split screen resolution (e.g. low res for game play then med for hud)? I am sure the STE is not too fussy about doing this (not sure about STF). I remember trying to do this in STOS but I can not remember the results, I think it ended with messy s...

I am sure it can be done but are there any examples? As in released medium res overscan demos or example code? For the latter - it's exactly the same code as for low resolution. As to released medium res overscan then there are a few. Flix/Delta Force did a medium res fullscreen here: https://demoz...

elliot wrote: ↑Tue Oct 18, 2022 11:52 pm What about medium res?

GLUE only deals in 50Hz, 60Hz and mono (71Hz) modes. Medium and low res are both either 50 or 60, so everything that can be done border-wise in low resolution can be done in medium resolution.

The new n0 depacker has an excellent depacking speed, outperforming m7 and m4 by a big margin: m7=40.855 s m4=48.498 s and n0=31.310s. The compression ratio is somewhere between the m4 packer and de m7 packer, for some files even worse as the m4 packer. The size of the n0 depacker is very small: on...

Data point: My two most recent packer needs have been for in-place fast depacking (upx nrv2b was the choice I made) and extremely small sizecoding compo depacker (24 bytes, but not general case). 64 byte general depacker sounds awesome.

(These images all show as broken links for me. Chromium on Linux)

Yeah I'm just now starting to learn the STE so I can't really weigh in on whether it's possible to disrupt the Shifter's loading of words through poking the pixel offset register. I looked at the new video and it's clearly "the same" effect - left over words in the Shifter registers from o...

OS calls in demo code - you're crazy

Interrupts is of course also my code - no problem handling.

Ok so the wraparound is now fixed but there's still some other glitches happening when using the pixel offsets?

ThorstenOtto wrote: ↑Fri Sep 23, 2022 5:38 pm That probably won't work, due to the different stack frame for exceptions.

Why would there be exceptions? And if there are, they're intended and I'm correcting the stack myself.

Actually I check if it's possible and which way I could do so. But be shure, that the processor couldn't change during operation. Reason for this is, that the OS set the machine type (e.g. setting cookie, ...). Changing the clock frequency could be possible. Thank you! Clock frequency is a good sta...

Yeah it's a 3bpl move.w (16 pix y-dist, looks 8 pix-flickering due to offset in buffering) scroll on top of 1bpl logo. Stars or:d on top of everything (no and:ing).

Well, that's my guess from just looking at it anyway Never disassembled another group's code.

The stars are 2 bitplanes, the Union logo is 1 bitplane but the scroller is also 2 bitplanes (1 colour where the rainbow rasters are mapped, 2 colours for the font border and 1 black colour) but are the logo and the stars mixed every frame? That's a lot of or-ing. Screenshot from 2022-09-23 09-14-0...

So hardware switching only? I was hoping for a register Is the switching clock-safe (that is, doesn't generate any glitch pulses) so it can be done during execution?

How's the switching between speeds and CPUs done? Reason for asking: As a demoscener I need to run my code in 100% compatible mode, but I'd love to run the editor and compiler at (much) higher speeds. Ideally this is handled by my own code setting compatibility when launched and setting back to high...

Right, yeah it needs to be done at the very end of the scanline - purpose is to make sure any left over words are cleared from the Shifter registers before the next line. No worries - it's fine to go to medium. And you also only need to do this once so you can run the clear in a single VBL and the c...

The easiest way to test my hypothesis, if you're sync locked anyway during this time, is to perform a Level 16 "right border stabilizer" at some line. Enough to do it once before the effect starts.

Alright, let's try to break this down. The issue, if I've understood things correctly, is this: Screenshot from 2022-09-07 10-22-36.png The final 8 pixels of graphics is displayed on the left instead of last on the right. Looking at this image in detail it's my opinion that the graphics on the left ...

Are we sure everybody talk about the same issue ? ;) My point was about this bug https://youtu.be/ZRU6bvv_gJU?t=86 On all tested machines with 1st revision of shifter you can see that 8 pixels from the right of the image appears on the left of the screen. It is not the case with ste having the 2nd ...

But then the glitches are likely due to ageing components. STEs, regardless of revision, didn't come with such glitches off the factory.

It does sound like there might be something to test here. Unfortunately I think it requires someone to themselves have both sorts available. I will check what I have myself when I find the time. (I think Exxos' buffer board allows the 300588-Shifter to work in a motherboard that previously had a 301...