MiSTer FPGA Forum

I used to have a Commodore 1084 for my Amiga. What would be the shadow mask / scanline filters that most closely approximates a monitor like that?

Have you tried TrashUncle's "Commodore 1084 Round" shadow mask? Last time I saw one of those monitors in person was probably ~25 years ago, so I can't really rate the accuracy of the mask myself.

I collect CRT photos but don't have any high res of Commodore 1084 yet. This guy's blog post has the best I've seen so far on the net: https://electricthrift.com/2013/06/17/c ... 4-monitor/

The 1084 monitors aren’t all built alike, there are many variants even if they look alike from the outside, and you can even tweak its look by changing the sharpness on the flyback transformer, so the 1084 doesn’t really have its own discrete look.

You can say that for most CRTs But there is a certain reference frame you can use for particular brands, models, techniques used.

I took a macro photo of the display of my 1084 and it has a slot mask pattern (also referred to as JVC in the shadow mask patterns). I saw a number of macro shots of arcade machines displays that also use slot masks. It was a common type of mask in the eighties and nineties. (this photo, below, is vertical, with the monitor on its side). My nano/subpixelslot pattern, seen above, comes pretty close on a sub pixel level. This is a macro shot of my TFT display showing the pattern in photoshop. Unfortunately it is not displaying correctly on MiSTer at the moment.

LamerDeluxe wrote: ↑Thu Dec 02, 2021 3:02 pm I took a macro photo of the display of my 1084 and it has a slot mask pattern (also referred to as JVC in the shadow mask patterns).

Thank you, this is fantastic, better than I hoped for! I will test out some of the JVC masks to see how close they get.

LamerDeluxe wrote: ↑Thu Dec 02, 2021 3:02 pm My nano/subpixelslot pattern, seen above, comes pretty close on a sub pixel level. This is a macro shot of my TFT display showing the pattern in photoshop. Unfortunately it is not displaying correctly on MiSTer at the moment.

Your recreation looks great. I would love to see it when it's working.

I was playing around with a few shadow masks on the Super Nintendo core, but on the ones I tested, I had to set integer scaling to avoid crawling artifacts when scrolling horizontally. Is this a fact of how the shadow masks are implemented?

syltefar wrote: ↑Thu Dec 02, 2021 3:25 pm

LamerDeluxe wrote: ↑Thu Dec 02, 2021 3:02 pm I took a macro photo of the display of my 1084 and it has a slot mask pattern (also referred to as JVC in the shadow mask patterns).

Thank you, this is fantastic, better than I hoped for! I will test out some of the JVC masks to see how close they get.

The available JVC masks should get you pretty close.

LamerDeluxe wrote: ↑Thu Dec 02, 2021 3:02 pm My nano/subpixelslot pattern, seen above, comes pretty close on a sub pixel level. This is a macro shot of my TFT display showing the pattern in photoshop. Unfortunately it is not displaying correctly on MiSTer at the moment.

Your recreation looks great. I would love to see it when it's working.

Thank you. I hope it will be possible to have it working correctly in the near future, I'm looking forward to seeing it in action on MiSTer. Right now I'm not sure what is causing the problems with it.

I was playing around with a few shadow masks on the Super Nintendo core, but on the ones I tested, I had to set integer scaling to avoid crawling artifacts when scrolling horizontally. Is this a fact of how the shadow masks are implemented?

That is indeed a risk with these patterns, they are applied to the native resolution of the physical display, which is not the same as the scaled pixels of a core display. This is similar to how it works with a real CRT. The patterns are really sharp and combined with the different scales this can cause interference patterns, also in combination with scan line filters.

LamerDeluxe wrote: ↑Thu Dec 02, 2021 4:26 pm That is indeed a risk with these patterns, they are applied to the native resolution of the physical display, which is not the same as the scaled pixels of a core display. This is similar to how it works with a real CRT. The patterns are really sharp and combined with the different scales this can cause interference patterns, also in combination with scan line filters.

That depends what kind of "crawling" or artifacts in general we're talking about (which is a recurring problem in these kind of discussions because they are hard to describe). When you display via RGB or S-Video on a CRT image is pretty much rock solid, things get a bit worse on composite.

akeley wrote: ↑Thu Dec 02, 2021 4:34 pm

LamerDeluxe wrote: ↑Thu Dec 02, 2021 4:26 pm That is indeed a risk with these patterns, they are applied to the native resolution of the physical display, which is not the same as the scaled pixels of a core display. This is similar to how it works with a real CRT. The patterns are really sharp and combined with the different scales this can cause interference patterns, also in combination with scan line filters.

That depends what kind of "crawling" or artifacts in general we're talking about (which is a recurring problem in these kind of discussions because they are hard to describe). When you display via RGB or S-Video on a CRT image is pretty much rock solid, things get a bit worse on composite.

This is an interference problem with some MiSTer shadow mask patterns displayed over HDMI. With a CRT you don't have this problem because the slot mask is physical and the raster lines are smooth.