MiSTer FPGA Forum

Is audio in connected? I tried loading some tapes via my tape loader and nothing seems to be read by the MiSTer.
After I type CLOADM"" I get a high pitched noise, which I can still hear now everything is switched off

The tape loading only looks at the game folder on the SDCard at this time.
Maybe ADC support will come in future, but probably disk support will come before that (but it's not trivial, so no estimates on timeframe)

@dshadoff,

The ADC support has also another advantage, some cartridge need to have access to the audio in to work. The amazing Audio Spectrum Analyzer (cat. N°:26-3156) in one example. From my point of view (being moreover an user of this cartridge ) this feature is a quite important one to have as it will prevent to use a software, not to run it of course, but to use it.
Anyway thank you for your hard work on this beloved core

I think i see some hardware I left out of the dragon64..

Thank you guys so much for this core!!!
The Dragon32 was my first computer and I've wanted a core for it ever since getting my MiSTer
I'll be spending quite a lot of time with this core.

Thanks again, it's very much appreciated

Just tried the latest core and more and more games are working fine.
However I notices a graphic issue with Speed-Racer, don't know if it is my dump or an issue with the core but it hang after the track selection.

kathleen wrote: ↑Fri Nov 27, 2020 8:34 pm Just tried the latest core and more and more games are working fine.
However I notices a graphic issue with Speed-Racer, don't know if it is my dump or an issue with the core but it hang after the track selection.

I had the same problem. I think the tape didn't work on xroar, but the disk did.. maybe not until we get disk support working.

I presume at some point Coco3 will be worked in? (this would give us an open alternative and hopefully official coco3 core)??

thanks for making a CoCo2/Dragon32 core!
I have some issues with defining my gamepad buttons. it seems that only the horizontal axes from the analog stick are working, not up and down. will there be a digital pad support coming too?

teller wrote: ↑Sat Nov 28, 2020 1:49 pm thanks for making a CoCo2/Dragon32 core!
I have some issues with defining my gamepad buttons. it seems that only the horizontal axes from the analog stick are working, not up and down. will there be a digital pad support coming too?

Did you configure the Stick X: Tilt Right and Stick Y: Tilt Down when configuring your controller from the MiSter Main menu?
Here is a quick video configuring my PS controller.
https://youtu.be/Ck4BAt82lHM

teller wrote: ↑Sat Nov 28, 2020 1:49 pm thanks for making a CoCo2/Dragon32 core!
I have some issues with defining my gamepad buttons. it seems that only the horizontal axes from the analog stick are working, not up and down. will there be a digital pad support coming too?

I was having the same issue than you with my Inext pad and thanks to the great video of NML32 I managed to get it working

ericgus09 wrote: ↑Sat Nov 28, 2020 6:04 am I presume at some point Coco3 will be worked in? (this would give us an open alternative and hopefully official coco3 core)??

It's something we'd like to do, but there is a lot of work involved in implementing any feature, most importantly obtaining good technical information about the machine. And there are lots of other important things to add (and fix) before we start on the CoCo3.

Perhaps it's better not to talk about CoCo3 until after CoCo2 has disk support (which will probably take a while).

teller wrote: ↑Sat Nov 28, 2020 1:49 pm thanks for making a CoCo2/Dragon32 core!
I have some issues with defining my gamepad buttons. it seems that only the horizontal axes from the analog stick are working, not up and down. will there be a digital pad support coming too?

Yes, this caused me problems too; I even ended up getting a new controller because I had thought it was my controller.

I don't know why the analog stick settings take place AFTER settings for mouse. (And especially after one has already tilted the stick in the early part.) There are a few points in that process where you might think that you are done, before even getting to the analog stuff.

But, I'm pretty sure that you just didn't set up the stick completely yet.

Regarding joysticks, when I had my Dragon I built a digital to analogue converter to use Atari standard joysticks.

I am sure there are Dragon games that play better with analogue but many are Spectrum ports, that I found worked better with normal digital controls, plus my Dragon joystick was like a stick in a bucket! Modern thumbsticks are a dream!

Higgy wrote: ↑Sat Nov 28, 2020 5:43 pm Regarding joysticks, when I had my Dragon I built a digital to analogue converter to use Atari standard joysticks.

I am sure there are Dragon games that play better with analogue but many are Spectrum ports, that I found worked better with normal digital controls, plus my Dragon joystick was like a stick in a bucket! Modern thumbsticks are a dream!

Hi Higgy !

I've got an analog joystick and works fine with this core. Original one is based on potentiometers, like Apple ][ and own CoCo.

dshadoff wrote: ↑Sat Nov 28, 2020 3:48 pm
Perhaps it's better not to talk about CoCo3 until after CoCo2 has disk support (which will probably take a while).

Oh sure yes that makes perfect sense.. was more or less trying to get a feel for the long term roadmap.. thanks..

Thanks dshadoff & NML32 for helping me out getting my gamepad mapped correctly. now all the analog axes of my controller (ps4) work fine. I own a Dragon 32 and the original controllers are indeed analog but non self-centered, a feature modern analog controllers unfortunately don't have.
I think not all users have gamepads with analog sticks so this core will get many questions how to map the directions. Maybe in future there will be some option to also set the dpad as the left analog stick, so you could use both ways.

Back to speed-racer, I found the Microdeal K7 version (Dragon 32) and the game works on the Coco2 core.
I've now serious doubts about the validity of the dump of the K7 game that I found for the coco2.
Great news then.

The Alice core inspired me to do this

That's a great video !
I only knew a few minor points of this history; it was nice to see a more complete story !

kathleen wrote: ↑Sun Nov 29, 2020 12:52 pm Back to speed-racer, I found the Microdeal K7 version (Dragon 32) and the game works on the Coco2 core.
I've now serious doubts about the validity of the dump of the K7 game that I found for the coco2.
Great news then.

I am glad you tried the Dragon version. I noticed that the cassette I had wouldn't run on xroar. But the disk version would. I didn't try the Dragon version. I suspect the cassette might be bad.

Here are two nice add-ons to add to the Mister running with the Coco2 core for those who maybe want to put the Mister in a real Coco case
https://www.tindie.com/stores/fiscap0768/

I'm just wondering if the real analogue joystick could work with the Coco2 core

kathleen wrote: ↑Fri Dec 04, 2020 5:23 am Here are two nice add-ons to add to the Mister running with the Coco2 core for those who maybe want to put the Mister in a real Coco case
https://www.tindie.com/stores/fiscap0768/

I'm just wondering if the real analogue joystick could work with the Coco2 core

And if you want a CoCo keyboard to go with that USB converter... look here (but he runs out of them quite frequently):
https://thezippsterzone.com/keyboards/

Higgy wrote: ↑Wed Nov 25, 2020 1:57 pm Cool. I would like to test Dragon loading.
I have a CASdunio that I used with my Dragon64 for loading tapes.

Higgy, if you have a CASduino, could you run a test for me ?
Please try playing the output into the ADCTest core, with the "3.3V" scale, and take a screenshot. (Please play CoCo format)

I am wondering whether that device is AC coupled or not (it should be), and whether the amplitude is within spec.
This will determine whether it can be used with the ADC as-is. In building my own CoCo-type virtual cassette player (different design), I have met issues like these, and adjusted my design accordingly.

If it's correct, the waveform should be roughly in the middle of the scale, without touching either edge.
If you see only half of the waveform, but centered on "zero", that would indicate DC coupling.

No problem. I will have a go.
Funny I was on their website yesterday and spotted that they had updated the CASdunio firmware. I will test before updating firmware and again after updating.

Updated with photos. Is this good?
This is with CASDuino 1.17 firmware.

The position of the line indicates that it has a good capacitive coupling, and can be used.
But it doesn't appear to be playing anything in the pictures... there's no visible waveform.

I've submitted a pull request for tape input via ADC for the CoCo core, and it should turn into a new release soon, once the framework is updated.

In my ADC update, I detect transition from '0' to '1' when the signal overshoots the running-average value by 0.1V; this is fine based on my tests with my own virtual tape player device (which was itself tested by playback to a real CoCo2).

For reference, on the 3.3V scale of the ADCTest core, each of the solid green lines is 1 volt from the next, with 0.5V from solid green to dotted-green... so 0.1V is not such a big movement. I check for overshoot to reduce the possibility of noise causing a zero-cross.

So, soon it should be testable... Actually, if you can build the core for yourself, you can beta test now, since Alan merged the pull request.

I've advised Pierco of the change as well; the Alice/MC-10 core shouldn't be too far behind for real tape support.

There is something there just quiet. I can hear the 'music'
There is an onboard amp but I did not tweak it. I should have tested a MSX .CAS to compare output.
I only have a physical MSX now, but the CASdunio's output should not have changed from when I used it with a Dragon.
I got the latest firmware 1.24 installed on my CASduino last night, but did not get around to comparing output. Also I will try and use a TOSEC .CAS, I can't remember where I got my .CAS files. The new firmware is supposed to improve TOSEC compatability.

OK, the ADC updates are in the CoCo2 core now, and it has been released. Please give it a try.


@kathleen: you might want to try the Spectrum Analyzer program now too, but if you have any issues, check ADCTest to see the the signal is not half cut-off.

I've tried from various sources, and:
- analog sources such as old cassette players have capacitive decoupling, display in the centre of the ADCTest screen
- digital sources such as cellphones and sound cards tend to be DC-coupled, and won't work even on original machines, unless the signal is sent through a series capacitor. (Something between 1uF to 10uF should be fine).

dshadoff wrote: ↑Wed Jan 13, 2021 1:09 pm
@kathleen: you might want to try the Spectrum Analyzer program now too, but if you have any issues, check ADCTest to see the the signal is not half cut-off.

@dshadoff,

1st of all thank you for your work on this amazing core.

Here are my today's results:

I've just tested by connecting an old realistic tape recorder to the Mister
Then, I have tested the signal with the ADC test utility. See below photos for the results.
I can also hear via the audio out of the mister (connected to an amplifier) what is coming in. As long as I stay between the red line, the sound is OK as it should be.
I then loaded the new Coco2 core with the Audio Spectrum cartridge and as expected the audio in is working as I can see the bars dancing
but, the output signal is distorted despite during the test with the ADC utility it was OK.

I gave a try with the audio out of my PC and it is catastrophic, I mean it is completely saturated even with a very low input level.

When you say "the output signal is distorted", let me confirm that this is the scenario you are talking about:
- Using CoCo2 core
- Using Audio Spectrum Analyzer
- Listening to the MiSTer output

...if this is the case, I can understand that - I didn't implement a pass-through from ADC in to the MiSTer sound output on the CoCo core; there was already a bunch of other logic and levelling there, and the "audio pass-thru" option on the audio is not implemented.

I guess what I was hoping to understand, was whether the spectrum analyzer part of the software was functioning how you expect it to, visually.

...Of course, if I misunderstood the scenario, please help me to understand.