VBXE - Can't switch boot core (VBXE IO error)

[woj]

9 minutes ago, electron said:

If anyone is still interested in new VBXE cores, tell me about it from time to time - just for motivation

Well, since you mentioned it, one thing to fix is the off by 1 scanline NTSC bug.

 

Otherwise, a feature request, difficult I guess - nibble based blitter stuff for the high res mode. I mean, the collision detection is done on nibble basis, what would be cool is to move / copy things around with nibble, rather than byte, resolution. This would allow animating high res objects with one high res pixel precision, and easily make mirror images with the -1 (well, -0.5 technically) x step.

 

When working on my stuff I survived without these in the end, rather noticed this deficiency, however, for the x mirroring of high res objects I had to implement this in software ;).

 

Just a thought...

[AA_ron]

1 hour ago, flashjazzcat said:

I ran the VGA core very successfully on an Incognito 800 a year or two back (there's a video), but I wasn't sure that register instability would be problematic without the 28MHz crystal. Thanks - that explains a lot, and makes this even more of a corner case.  I suspect, then, that if the OP's machine (and mine) had the right crystal, the U1MB plugin would have ID'd the FX core and all would have been well.

FWIW, my machine is NTSC, but you probably guessed that.

[AA_ron]

1 hour ago, flashjazzcat said:

I ran the VGA core very successfully on an Incognito 800 a year or two back (there's a video), but I wasn't sure that register instability would be problematic without the 28MHz crystal. Thanks - that explains a lot, and makes this even more of a corner case.  I suspect, then, that if the OP's machine (and mine) had the right crystal, the U1MB plugin would have ID'd the FX core and all would have been well.

FWIW, my machine is NTSC, but you probably guessed that.

[Piotr D. Kaczorowski]

@AA_ron,

 

Sometimes, for flashing VBXE, it's better to connect it with a short wire directly to pin #9 of the 74LS138, rather than via U1MB. After successful flashing, you can reconnect it to the VBXE.

 

It is also possible that something is not okay with this particular unit. The vendor uses a pick-and-place machine with lead-free solder, which can make the connections more fragile. It happens. Sometimes, I have to resolder the FPGA under a microscope and clean the VBXE in an ultrasonic cleanser afterwards.

[AA_ron]

11 minutes ago, Piotr D. Kaczorowski said:

@AA_ron,

 

Sometimes, for flashing VBXE, it's better to connect it with a short wire directly to pin #9 of the 74LS138, rather than via U1MB. After successful flashing, you can reconnect it to the VBXE.

 

It is also possible that something is not okay with this particular unit. The vendor uses a pick-and-place machine with lead-free solder, which can make the connections more fragile. It happens. Sometimes, I have to resolder the FPGA under a microscope and clean the VBXE in an ultrasonic cleanser afterwards.

Yikes. We did get this back to a reliable VBXE core.
 

... but I was trying to get certain VBXE demos to run, like Mr. Proper. Several of the demos aren't exactly working right, or out-right freezing.
Maybe I'll start a thread about the results I'm getting of the fancy FX features, and why they aren't working.
I wanted VGA most of all, but I don't want to give up on the fancy features I paid for.

Thanks

[flashjazzcat]

10 hours ago, Piotr D. Kaczorowski said:

Sometimes, for flashing VBXE, it's better to connect it with a short wire directly to pin #9 of the 74LS138, rather than via U1MB. After successful flashing, you can reconnect it to the VBXE.

No need to do this at all.

[AA_ron]

On 3/23/2024 at 2:52 PM, flashjazzcat said:

It's hardly burdensome; the code is slight and plugin-based, and if the handful of people with special requirements want an alternative plugin which works in the way you describe (rather than burdening the average user with yet more settings they don't need), I'll happily provide it.

What do you think about the possibility of BIOS allowing change of the bootable core?

In trying various VBXE demos, I've had to switch the VBXE core a few times, and it's a series of steps to change ATR, enable SDX, load FC.COM, set alternate core as bootable.
Often to then run the program, I'd revert much of that setup (ex: to boot a different ATR).

(BTW, I did figure out the most-often magical VBXE demo combination for demos is FXRambo core, +Rambo 1088, +PIO, -Basic, -SDX, +XEX boot)

Really nice that U1MB is VBXE aware, for enable/disable and even NTSC color pallet.
It feels natural from user perspective that the BIOS could allow marking of a core as bootable from the existing core slots.

 

I'm a sw developer, and like that you've got plugins, but I know everything comes with an effort-cost (from you).

I won't ask for adding/removing cores in BIOS, but it would be nice to skip all the steps necessary just to mark a different core as bootable.

[+Stephen]

2 hours ago, AA_ron said:

In trying various VBXE demos, I've had to switch the VBXE core a few times

Can you post some examples please?  I've to the best of my knowledge ran every VBXE demo there is, including my own code and never once changed cores in the past 4+ years.

[flashjazzcat]

6 hours ago, AA_ron said:

What do you think about the possibility of BIOS allowing change of the bootable core?

Sounds like a great project for a software developer with a need and desire to make it happen. It would be useful to get some more feedback on the usefulness and clarity of the plugin developer documentation, anyway.

 

One stumbling block might be the algorithms necessary to actually manipulate the core booter, since none of this is documented (AFAIK) and the source code to FC.COM is not available. Testing might be a little fraught as well, since core flashing, etc, isn't implemented in Altirra so you'd need to go bare metal all the way.

[AA_ron]

On 3/25/2024 at 2:05 PM, Stephen said:

Can you post some examples please?  I've to the best of my knowledge ran every VBXE demo there is, including my own code and never once changed cores in the past 4+ years.

Reasonable point.

 

I think I ended up at the same magical combination. Published anywhere?

FXRambo core, +Rambo 1088, +PIO, -Basic, -SDX, +XEX boot

 

I'm only a few months back into the 8bit scene, and have had to figure things out.

Early on I couldn't get demos like Mr. Proper to run, and assumed that maybe, being an older demo, maybe needed an older core.

That's where I got into trouble, since I didn't know which core it wanted, and tried them all.

 

Once I figured out I was having a memory capacity issue, I found bumping core to FXRambo, using full 1MB, and disabling basic gave me most of what I needed.

I was also running them from SpartaDosX, so it was using some memory/resources and had to be eliminated.
So, almost every demo needed to be set to autorun with everything else turned off. I didn't try smaller dos boot menus.

 

I feel like there were 1 or 2 in the ~15 I recently tried that needed me to boot into SDX to start them... and maybe even one that wanted Basic?!?

I tried so many combinations that my recollection is a bit fuzzy... and I'm not going back to find specific examples.

I guess the next person will have to figure it out on their own too. Maybe they could post the exceptions when they find them.

 

... but like you said, one combination tends to work across most demos.

I'm curious if your stable settings are the same as where I ended up.

Please post your settings.

[AA_ron]

On 3/25/2024 at 6:43 PM, flashjazzcat said:

Sounds like a great project for a software developer with a need and desire to make it happen. It would be useful to get some more feedback on the usefulness and clarity of the plugin developer documentation, anyway.

 

One stumbling block might be the algorithms necessary to actually manipulate the core booter, since none of this is documented (AFAIK) and the source code to FC.COM is not available. Testing might be a little fraught as well, since core flashing, etc, isn't implemented in Altirra so you'd need to go bare metal all the way.

Thanks for proposing the fun idea!

 

I would absolutely love to, but it's not going to happen in the foreseeable future.

Other projects in the way, and I'm only mediocre at assembly now and was never good at it on my Atari back as a teen. Though it would be fun to learn.

 

Interesting about source not being available for FC.COM (and probably others).
I know that above you were only talking about the VBXE plugin, but I'm curious now... is your U1M bios source code published & available anywhere?

[+Stephen]

17 minutes ago, AA_ron said:

Reasonable point.

 

I think I ended up at the same magical combination. Published anywhere?

FXRambo core, +Rambo 1088, +PIO, -Basic, -SDX, +XEX boot

 

I'm only a few months back into the 8bit scene, and have had to figure things out.

Early on I couldn't get demos like Mr. Proper to run, and assumed that maybe, being an older demo, maybe needed an older core.

That's where I got into trouble, since I didn't know which core it wanted, and tried them all.

 

Once I figured out I was having a memory capacity issue, I found bumping core to FXRambo, using full 1MB, and disabling basic gave me most of what I needed.

I was also running them from SpartaDosX, so it was using some memory/resources and had to be eliminated.
So, almost every demo needed to be set to autorun with everything else turned off. I didn't try smaller dos boot menus.

 

I feel like there were 1 or 2 in the ~15 I recently tried that needed me to boot into SDX to start them... and maybe even one that wanted Basic?!?

I tried so many combinations that my recollection is a bit fuzzy... and I'm not going back to find specific examples.

I guess the next person will have to figure it out on their own too. Maybe they could post the exceptions when they find them.

 

... but like you said, one combination tends to work across most demos.

I'm curious if your stable settings are the same as where I ended up.

Please post your settings.

I just run the standard 1.26 Fx core, since the machine with my VBXE also has U1MB RAM expansion.  I do run the latest version of SDX and by default, my machine boots with BASIC off.

[flashjazzcat]

13 minutes ago, AA_ron said:

I know that above you were only talking about the VBXE plugin, but I'm curious now... is your U1M bios source code published & available anywhere?

No, but the source code for the plugins is available on the firmware download page.