YM2151 questions

m00dawg · Post by **m00dawg** » Sat Feb 13, 2021 6:27 pm

Addresses will certainly change at least once the emulator starts being worked on again. I do worry as all the repos haven't been administered in some time. There's several PRs across the project and some of these would be really valuable. The emulator I might understand if it's not technically open source yet (though I think it is?) but the docs - I know there's one PR that provides the actual color swatches of the palette in line with the docs, and that would be SUPER helpful.

Not sure why the repos have seemingly been ignored by the team. I'm sure there's a reason for this but it does create some issues. We know, for instance, the kernal now has claim to a few zeropage addresses that are currently available in the emulator for instance, and we know from Kevin's last vid on the revision 2 board working that several addresses (FM included) will be changing.

Not trying to stoke any conspiracy theories here but Micheal Steil was doing tons of work on the project but hasn't made any changes since August so perhaps he may have gone onto other things and the team is busy working on the other stuff to have the time to circle back and work on the repos at the moment. That could make sense if the hardware design is still changing (better to go back and update the emulator once if there's limited resources than over and over again as the hardware changes)?

I do wish some of the PRs would be looked at though, as noted above, There's some good ones worthy of eval.

desertfish · Post by **desertfish** » Sat Feb 13, 2021 8:33 pm

Yeah- you can fork and build your own emulator an kernel but eventually there is going to be a Rom made for the real hardware....

BruceMcF · Post by **BruceMcF** » Mon Feb 15, 2021 5:43 am

On 2/14/2021 at 2:27 AM, m00dawg said:

Not sure why the repos have seemingly been ignored by the team. I'm sure there's a reason for this but it does create some issues. We know, for instance, the kernal now has claim to a few zeropage addresses that are currently available in the emulator for instance, and we know from Kevin's last vid on the revision 2 board working that several addresses (FM included) will be changing.

A lot of work would have been put on hold during the period when the revised board was not able to boot. It would not be surprising if there is a lot of work going on behind the scenes as we speak. That makes it quite plausible that it is not an effective time investment to update the repos based on previous iterations of work in progress hardware and Kernel when the finalized design may appear to be in sight.

Given the track record so far of announcing things when there is something to announce, including announcing that they had hit some kind of hardware snag when they were NOT able to get the board to boot up, I prefer to imagine a period of silence as representing the garage door being shut but clanging sounds coming from inside.

m00dawg · Post by **m00dawg** » Mon Feb 15, 2021 6:15 am

15 minutes ago, BruceMcF said:

A lot of work would have been put on hold during the period when the revised board was not able to boot. It would not be surprising if there is a lot of work going on behind the scenes as we speak. That makes it quite plausible that it is not an effective time investment to update the repos based on previous iterations of work in progress hardware and Kernel when the finalized design may appear to be in sight.

Given the track record so far of announcing things when there is something to announce, including announcing that they had hit some kind of hardware snag when they were NOT able to get the board to boot up, I prefer to imagine a period of silence as representing the garage door being shut but clanging sounds coming from inside.

Indeed those are good points and I agree with most of that. I do think at least the documentation PRs are well worth looking at and certainly if folks were looking at PRs, I would provide a few of my own (the section of the docs the mention the kernel routines, like CHROUT, could stand to be better formatted for example). I'd bet plenty of folks would be willing to contribute to those (and some have by way of some excellent external resources though these require knowing that they exist). There's a PR that adds nice color swatches for the built in palette for instance. That would be super convenient to have and I would really like to see that merged.

ZeroByte · Post by **ZeroByte** » Wed Feb 17, 2021 9:56 pm

I came here exactly because the official repo has been quiet for so long. I was thinking about updating the YM2151 support to move it to the base address and add read support to the emu, as well as adding its INIT routine to the system reset. I was wondering if there was any news here about why emu development seems to have stopped completely.

StephenHorn · Post by **StephenHorn** » Wed Feb 17, 2021 11:30 pm

1 hour ago, ZeroByte said:

I came here exactly because the official repo has been quiet for so long. I was thinking about updating the YM2151 support to move it to the base address and add read support to the emu, as well as adding its INIT routine to the system reset. I was wondering if there was any news here about why emu development seems to have stopped completely.

As far as anyone seems to know, Michael Steil is still part of the team, but may simply be busy with their day job and simply hasn't been able to keep up with the Github.

This comes, in part, from Mike Allison, who responded to a ping about this on the unofficial Discord a few days ago.

So... "patience, grasshopper".

Though I'll say I sure would love to help catch up the emulator on pull requests and features, and maybe catch up on some of the bugs from the actual emulator (and not the ROM). Of course, anyone can fork the emulator in GitHub, and the magic of Git is that (once you've learned how to do it, anyways), you can pull changelists from almost anywhere. Just keep in mind that there is only one official emulator repo, and it is the only repo that has been granted permission to host the X16's modified ROM files that are necessary for the emulator to do anything useful.

ZeroByte · Post by **ZeroByte** » Fri Feb 19, 2021 4:48 pm

On 2/17/2021 at 5:30 PM, StephenHorn said:

As far as anyone seems to know, Michael Steil is still part of the team, but may simply be busy with their day job and simply hasn't been able to keep up with the Github.

This comes, in part, from Mike Allison, who responded to a ping about this on the unofficial Discord a few days ago.

So... "patience, grasshopper".

It would definitely be good to go ahead and start using the new address locations for things like the RAM/ROM banking and YM2151 base addresses in our X16 projects.

On the emu side, I've also felt like taking a stab at doing a VIA emulation that works for the clocks/IRQs. Lacking those made it difficult to debug my program back when I was helping Kevin with the 'hello world' tests for the YM2151 back on prototype 1.

StephenHorn · Post by **StephenHorn** » Fri Feb 19, 2021 8:50 pm

4 hours ago, ZeroByte said:

It would definitely be good to go ahead and start using the new address locations for things like the RAM/ROM banking and YM2151 base addresses in our X16 projects.

On the emu side, I've also felt like taking a stab at doing a VIA emulation that works for the clocks/IRQs. Lacking those made it difficult to debug my program back when I was helping Kevin with the 'hello world' tests for the YM2151 back on prototype 1.

Well, the nice part about the emulator's design is that it's really easy to move around memory-mapped I/O. There's no reason someone couldn't work on cycle-accurate VIA emulation for the current addresses, and then move it to the proper addresses later.