MiSTer Feature Expansion Idea

[Hackshed_Carl]

This has been rattling around my head for a couple of weeks so I thought I'd share the idea and see how feasible it is...

I'll use the C64 as an example.
The C64 floppy drive isn't a simple device and the circuitry already exists within the C64 core.
How possible / difficult would it be to build a custom core that could link to the C64 with an IEC cable (or simple circuit) and use it to emulate a floppy drive?

This could also be useful for other systems that require messy storage options.
ZX spectrum would make a fine candidate for some MiSTer features too.

Before anyone says it, yes, I know you can get Pi1541, SD2IEC etc for C64 and similar devices for other platforms but I think this would make a nice added feature.

If I had the slightest clue about HDL / FPGA I'd look into this myself but I simply don't have the knowledge.

What do other folk think? Good idea or one that should be fired into space towards the sun?

[FPGA64]

Already covered by a multitude of cheaper alternatives than purchasing a DE10

[Hackshed_Carl]

Exactly the dismissive kind of answer I was expecting.
The discussion isn't about whether it should be done, it's if it could.
I did say that there are already solutions out there and that this isn't to compete with them but as an additional feature for the MiSTer.
For people that already own a MiSTer and also have real hardware, it would just mean an extra option.

[FPGA64]

just because you own a hammer doesnt mean eveything is a nail

Anyone with real HW is more than likely going to have a Real 1541 or the already existing alternatives.

You asked for opinions , you apparently meant I want people to validate my own opinion.

[akeley]

It's not a bad idea, but as usual the problem is finding somebody to execute it. This is what has just occured with the out-of-the-blue Groovymame core - a very niche thing, for which there already is a solution. and yet some enthusiasts made it happen.

In this case it could be useful because MiSTer owners also often collect old micros. And some of these storage devices can be either expensive and/or have limitations. Also, it could be perhaps possible to have a more universal core, like Gotek, or the BackBit cartridge. Though then you'd still need a heap of different connectors. So, rather unlikely to happen, but in the world of MiSTer I learned to never say never (which is a good thing).

[Flandango]

It's possible to write a core, along with using the User IO port as an IEC port the way it's done now with the C64 & C128 cores to behave as a floppy drive.
There are few things to consider though on how you approach the situation.
First and foremost, you'll need to decide how you want to navigate through the floppy images.
If you go the already included route of using the menu to select a floppy image from the SD card, then you'll need an additional display and input method for the Mister.
If you want to go down the route of sdiec or some of the other devices where you can use the C64 to instruct the virtual floppy drive (the new core) to change floppy images and what not, then you'll need to implement something in MiSTer's Main to manage that or you'll have to implement something using the 2nd SD card on the IO board (if so equipped).

I can't speak about other systems because I don't know the details off the top of my head and most of them will likely need more I/O lines than the User I/O port has.

So is it possible? Sure
Is it feasible? Probably not but the great thing about MiSTer is, it's open source and anybody who has the time and the proper equipment (real C64, IEC Snac....etc) can attempt to make one and see how far it goes.

[throAU]

Exactly the dismissive kind of answer I was expecting.
The discussion isn't about whether it should be done, it's if it could.
I did say that there are already solutions out there and that this isn't to compete with them but as an additional feature for the MiSTer.
For people that already own a MiSTer and also have real hardware, it would just mean an extra option.

Could it be done yes?
Is it a developer priority? Probably not
Would it be a reasonably simple project to start off with if you're looking to get your hands dirty learning: probably

I suspect the market of DE-10 Nano owners wanting to dedicate one to pretending to be a c64 disk drive is.... small to say the least.

Not to dismiss the idea entirely, but whilst someone is working on this, they aren't doing things like building something that doesn't have viable alternatives already.

[aberu]

Lots of things could be done in theory.

Whether or not it should be done is still relevant because at the end of the day someone with development skills with free choice to do what they please would be the one to implement it.

[Hackshed_Carl]

just because you own a hammer doesnt mean eveything is a nail
Anyone with real HW is more than likely going to have a Real 1541 or the already existing alternatives.
You asked for opinions , you apparently meant I want people to validate my own opinion.

My apologies if my response seemed terse.
No, I didn't want my idea validating for usefulness, I wanted to know if the idea itself was possible.
Just because something exists doesn't mean that it's not worth pursuing. Anyway, I'll leave that alone. Again, not wanting to stir anything up.
My intention isn't to start grief but I was looking more for discussion rather than negativity.

I'll highlight what brought this to mind.
I'm one of those people who collect old micros.
I've just dug the C64 out and wanted to test it.
First step was to make a safe PSU and from there test the datasette.
My KFF cart hadn't arrived at that point so in order to load software, I knocked up a IEC2PC with an arduino. It worked well enough for what I needed.
I've since ordered a Pi1541 PCB which is yet to arrive.

This all happened over xmas so shipping etc was delayed.
Also, retro computing can be an expensive hobby so to keep buying equipment isn't sustainable for everyone.

If the MiSTer had FDD emu functionality at the time, I would have used it and been happy.
I would probably end up with a Pi1541 anyway but could be a nice feature to use in a pinch.

I also wouldn't expect people to keep a MiSTer spare for this purpose but thought it might be a nice "value added addition" for current / future owners.
I agree that developer interest will play the most crucial factor with this but for all I know, someone with intimate knowledge of a particular core may think "that'll take 10 minutes" and then maybe include it in future updates.... who knows.

It's also just as likely that someone responds with "impossible due to x/y/z" and that would have also been useful to know the reasons why it's not possible.

[NCR]

Check this project...probably this is what you want...
https://github.com/venice1200/MiSTer_SNAC2IEC

[AngelicLiver]

Please do not interpret the responses you've received as negativity or drama. The vast majority of the feedback I see on this forum is positive and constructive.

Developers of any ilk have to pick their battles and identify value in what they're working toward; this is often entirely subjective. What you are suggesting isn't impossible but is incredibly niche, likely to be impractical to the end-user and would involve a wholly disproportionate amount of work. That said there is always value in doing something just because you can and learning something in the process.

Is it practical to play DOOM on your smart refrigerator? No. Could you do it?

[FPGA64]

I have 12 C64s of various types. 5 odd 1541. All of them have now been in the Attic for years. Its simply more of a pain to get them out and set up, hope the power supply isnt in kill the C64 mode than simply fire up winVice or the Mister.

If I ever do get the real C64 set up I always use a pi 1541. Its just so much more reliable than the 1541 and the increasingly rotting disks. Its also much more useful to have all the disks on one sd card.

My point is that setting up a Mister, loading a 1541 core, which is essentially a complete microcomputer that communicates over a serial interface with the C64 seems massive overkill. It may be possible but its using a sledgehammer to crack a nut.

[Hackshed_Carl]

Admittedly, the C64 core is excellent as it is and there's little reason to break out the real machine.
I actually only got the real thing out of the loft because I'm trying to teach myself 6510 assembly and was using the MiSTer primarily and a software emulator while out and about.
I kept seeing different results on MiSTer and emu and got really confused.
I couldn't work out why both methods would play complex demos perfectly but both crash differently when running my poor code.
C64 came out, got sorted and set up.

Only then did a realise the error of my ways.
Long story short. I started using a cart to fill the C64 memory with zeros priory to working on my code.
Turns out that the garbage data in memory was confusing me when trying to debug.

I'd be happy to go back to the MiSTer alone but now that I've sorted the PSU and got a KFF, I want to keep using the real thing and rebuild my keyboard muscle memory.

I'm planning a series of articles related to C64 assembly (and the C64 in general) which is how this whole can of worms got opened in the first place.

[Newsdee]

Yes, your idea is possible. If you just use the DE10 (without MiSTer extensions) you will have plenty of GPIO pins available.

If you use a MiSTer setup you will be restricted by the GPIO pins available (I think it's 8? and there is unofficial hardware using 9 pins). Might be enough if the c64 uses a serial protocol (with low pin count on the connector).

Either way, you would still need to come up with the right cable and electronics to interface properly, and the core to drive the whole thing. But yes, theoretically you have the hardware already to prototype.

[cursedverses]

Well, you could definitely follow the progress of RobSmithDev on YouTube. He got an Amiga floppy working: https://youtu.be/_MYw0lLerBA

Definitely possible, just ask yourself - is it worth it to have that many disks floating around? I for one would be happy with disk "sounds"

[djsquare]

Well, you could definitely follow the progress of RobSmithDev on YouTube. He got an Amiga floppy working: https://youtu.be/_MYw0lLerBA

Definitely possible, just ask yourself - is it worth it to have that many disks floating around? I for one would be happy with disk "sounds"

Disk sounds are amazing. That's one the of reasons why the Ultimate 64 and Ultimate-II+ are so amazing.

[FPGA64]

Well, you could definitely follow the progress of RobSmithDev on YouTube. He got an Amiga floppy working: https://youtu.be/_MYw0lLerBA

Definitely possible, just ask yourself - is it worth it to have that many disks floating around? I for one would be happy with disk "sounds"

This isnt about connecting a Disk Drive to the Mister. The idea was to connect a Real C64 to a Mister running a 1541 core.

You can already connect a 1541 via the user Port.

[cursedverses]

Well, you could definitely follow the progress of RobSmithDev on YouTube. He got an Amiga floppy working: https://youtu.be/_MYw0lLerBA

Definitely possible, just ask yourself - is it worth it to have that many disks floating around? I for one would be happy with disk "sounds"

This isnt about connecting a Disk Drive to the Mister. The idea was to connect a Real C64 to a Mister running a 1541 core.

You can already connect a 1541 via the user Port.

Oh I see, well I mean, we have a graphics adapter for an emulator in Groovy_MiSTer now, what the MiSTer can do is in the eye of the developer...

[Armakuni]

The 1541 emulation on the MiSTer C64 core is not cycle accurate and doesnt have support for all the original disk protections unlike other C64 FPGA cores which are sadly closed source or even Pi1541 so would need specific development