New Atari BBS software.

[tschak909]

Thanks for calling in! It's gonna be very bare, for a while, until I get some temporary UI functions in place.

 

As for terminal differences, this one is a booger-bear, especially seeing as there are no fewer than THREE backspace mappings in the ASCII derived character sets in use today.

 

Although...it does...give me an idea...

 

maybe if I ask for <BACKSPACE> instead of <RETURN> at the beginning, maybe this is enough of a differentiator to be useful...

 

-Thom

[idavis]

Anytime. Can't wait to see it come alive.

 

I wouldn't worry about the backspace bit. I was just using my linux terminal. If I had been using putty I am sure it would have mapped the backspace to ctrl-h. And I didn't need it at the beginning. It was when I was typing in my email and flubbed it. As someone who has dealt with terminals for years, I just hit it naturally after seeing the wrong chars pop up when I hit backspace. I use hjkl to navigate in vi if that tells you anything.

[tschak909]

Wow, two calls today, keep this up, and I might get into double digit indie board call status!

 

-Thom

[Android8675]

Basically, as big as you can get it. I'm not sure whether you can use 512 byte sectors over APE with SDX, you can try, but...

 

65535 sectors, 256 or 512 bytes a sector.

 

use spartados x FORMAT to get it ready.

 

In Altirra, I set up an SIDE 2 hard disk, and pared it with the Ultimate 1MB expansion. It will beat the living shit out of any SIO driven disk in performance. ... and since you have an Incognito...don't you HAVE an SIDE 2 built on?

 

-Thom

 

yeah, couldn't figure out how to get the SIDE working Altirra, want a dev enviornment that I can work on then easily just load up APE and start my 800 with the same setup. Easier to enter stuff into Altirra then run it on the 800. Also when I'm at work I wanna goof around on stuff. Get a system setup that I can sync at home via cloud.

 

I'll figure it out.

[tschak909]

I can help you get the SIDE running under Altirra. Did you configure the U1MB in Altirra? You should do that first.

 

-Thom

[evilmoo]

maybe if I ask for <BACKSPACE> instead of <RETURN> at the beginning, maybe this is enough of a differentiator to be useful...

 

Perhaps accept both ^H and ^? as delete until log-in, then use whatever is set as the user preference? And ask for delete to be pressed at new user signup?

[tschak909]

I shall experiment.

 

-Thom

[tschak909]

And now, a new status: I've implemented enough of menus to be able to have TWO functions on the main menu:

 

User List

[G] Goodbye

 

More will come, I promise.

 

Everything has been checked into the git tree, and I've made a subfolder called menu, for menu files that need to be placed on D1 in the MENU directory.

 

Of course, you can log into the usual address to see the results, yourself:

 

 

-Thom

[TXG/MNX]

Hello,

 

Question, BB65 supports big message bases that's great. Is there on your todo list something like a check for integrity on the messagebase?

 

I had BBS Pro for a while, (the damn harddisk is stuck will no boot now) with lots of files, still trying to re-animate this harddisk. Also messagebase grew pretty big.

Occasionally I had an issue of corruption due to a bad sector or sudden powerloss.

[tschak909]

Well, that's the thing... Since each body has a header mark for length, I can rebuild the index, that way.. I could also potentially store the headers in there too, to allow for rebuilding.

 

Ultimately, what I _WILL_ do, is store the parts that I am working on, in a temporary file on the same disk. If it writes successfully, I will append it to the existing file.

 

As far as things like backup, there does need to be a maintenance cycle, where message boards, etc, are backed up to other disks. Since I am deliberately coding this for modern hard disk interfaces, we'll have plenty of drive specifiers, to be able to mark certain drives as backup drives to push copies of message boards to. This will take some thought.

 

But right now, I just want to get things working.

 

-Thom

[TXG/MNX]

O ok thats a good answer for me.

You got me thinking why no Atari 8-bit controller has raid-1 implemented in hardware :-) ? Maybe it can be made so that the controller automaticly writes 2 drives at the sametime.

[flashjazzcat]

Maybe it can be made so that the controller automaticly writes 2 drives at the sametime.

It can, although it'd half the speed of throughput. You'd just need a custom driver which echoed all primary transfers on the secondary device.

[tschak909]

Guys...

 

We're having an ADHD moment...

 

it's a BBS, running on an 8-bit computer... Need to keep in perspective. I will take a good look at this once I have a good idea of how things will work on disk. If I took off on _every_ rabbit trail, this thing would take a decade to complete the first version.

 

-Thom

[flashjazzcat]

Just answering the question as succinctly as possible. Carry on.

[tschak909]

Today I am adding pagination to terminal_*, so that we get a nice pause message.

 

It can be used by a couple of public functions:

 

terminal_enable_line_counter();

terminal_disable_line_counter();

terminal_get_line_counter();

 

terminal_send() will use the toggles above, comparing to the line counter (which gets incremented on EOL, depending on terminal type), and will show the pagination prompt when the number of lines has been reached. Currently, this is hard coded at 22 (to account for the prompt), but will be user-settable soon enough.

 

-Thom

[tschak909]

Pagination has been implemented in the terminal_* functions, and I have put them to use in the filemenu_show() function. Works great. Running the latest build on the test system.

 

-Thom

[tschak909]

Guys, I really need some intrepid souls to not only test, but dig into the code to help make sure that the serial port routines work as they should on a real Atari 850 or other PBI or serial interface.

 

Mostly because, the Altirra serial emulation patches the OS in real time, and doesn't use the SIO to send its data across, thus it's not susceptible to the idiocy of having to make sure that concurrent mode gets turned on and off at each disk I/O pass. This is the biggest hiccup, and while I am trying to make sure that I CLOSE the port every fucking time I try to do a disk access, I am 100% sure that there are those bits that are falling through the cracks.

 

And here's the rub:

 

* The TCP/IP modem emulation STOPS, when I use anything other than the patched R: emulation in Altirra.

* I do not have a MODEM anymore to test with. I would need to get a Lantronix UDS1100 to really test this out. (Donations?)

* I do not have a working Atari at the moment to test with, nor a place to even put it on my desk (I have no less than four computers on this desk, actively being used, including the ASUS eeePC 1005HA that is running Altirra that I am literally developing this whole thing on. I chose it, because it wasn't being used, and it is teeny tiny enough to fit in the unused sliver of wood plane on my desk.)

 

So, I need some help ferreting out the serial port hiccups that may be happening. Must not be afraid to dig into the C code.

 

-Thom

[phaeron]

Altirra supports emulating the 850 at SIO level -- change the SIO Emulation drop-down from None or Minimal to Full, and it will bootstrap a regular R: handler across the emulated SIO bus and use SIO commands instead of intercepting the R: device requests. You should be able to replicate some SIO bus conflicts or interrupt latency related problems that way.

 

Mind you, this uses the R: handler I wrote instead of Atari's, though, so you may still see a different set of problems. If you have a disk-based version of the 850 handler, though, you should be able to load that instead and use it with the hardware emulation.

[tschak909]

Altirra supports emulating the 850 at SIO level -- change the SIO Emulation drop-down from None or Minimal to Full, and it will bootstrap a regular R: handler across the emulated SIO bus and use SIO commands instead of intercepting the R: device requests. You should be able to replicate some SIO bus conflicts or interrupt latency related problems that way.

 

Mind you, this uses the R: handler I wrote instead of Atari's, though, so you may still see a different set of problems. If you have a disk-based version of the 850 handler, though, you should be able to load that instead and use it with the hardware emulation.

 

Yup, and I was doing this, problem is, if I do this, the emulated hayes modem that I am using to do the TCP/IP bridging seems to go away. I replicated this in ICE-T to make sure I wasn't hallucinating...

 

-Thom

[phaeron]

That's weird, it shouldn't. I just checked it with 2.50-final and Ice-T 2.76 and verified that it works.

 

There is one thing that it might be, though, since it looks like you are still using a 2.50 prerelease version. I did make a last minute fix to the R: handler to fix a bug where it wasn't selecting the internal input buffer when supplied an input buffer length of zero. That was tripping up BobTerm, but it's possible it's tripping up your case too.

[tschak909]

That's weird, it shouldn't. I just checked it with 2.50-final and Ice-T 2.76 and verified that it works.

 

There is one thing that it might be, though, since it looks like you are still using a 2.50 prerelease version. I did make a last minute fix to the R: handler to fix a bug where it wasn't selecting the internal input buffer when supplied an input buffer length of zero. That was tripping up BobTerm, but it's possible it's tripping up your case too.

Ok, I can now clarify this issue a bit more...

 

With full R: emulation turned on, I can indeed send ATDIblablabla, and connect outward.

 

This is, however, useless to me.

 

What doesn't work, is when I do something like

 

ATZ

 

I don't get an OK back, like I do if I just do the emulated handler. This is my sanity check, as it is in....just about every BBS software on Planet Earth. ... Am I missing something?

 

-Thom

[phaeron]

Think I know what it is... do you have "disable serial port throttling" on, by any chance? That setting won't work with SIO-level emulation, as it just causes the emulated modem to spam POKEY faster than it can accept bytes. Just tried it with that on, and indeed in that mode you can transmit but not receive reliably.

[tschak909]

Thank you, Avery. That was it.

 

Now I'm experiencing buffer issues when interleaving disk access and output. Something definitely goofy I'm doing wrong, but at least I can reproduce this now. I have a funny feeling I will have to give my terminal functions an overhaul...

 

But thanks, now I can mark this down on my todo, to deal with this, definitely some problems.

 

-Thom

[Android8675]

I can help you get the SIDE running under Altirra. Did you configure the U1MB in Altirra? You should do that first.

 

-Thom

 

Yeah, got U1MB running in Altirra, just can't get the hard drive to come up in fdisk (can't find any APT devices). I'm selecting a new HDD image and choosing SIDE 2 Hardware. Is there a firmware file I need for the SIDE 2? (I'm running the 2.60-test9 alpha at the moment) i saw someone running SIDE 2 in Altirra and it looked like there's a DEVICE driver needed, but I can't figure out where that file resides. Maybe it's on the SDX4.46 rom for the SIDE2?

 

I'm going camping this weekend so not much time to goof around with my hardware, but I'll run the system thru it's paces next week.

[flashjazzcat]

Yeah, got U1MB running in Altirra, just can't get the hard drive to come up in fdisk (can't find any APT devices). I'm selecting a new HDD image and choosing SIDE 2 Hardware. Is there a firmware file I need for the SIDE 2? (I'm running the 2.60-test9 alpha at the moment) i saw someone running SIDE 2 in Altirra and it looked like there's a DEVICE driver needed, but I can't figure out where that file resides. Maybe it's on the SDX4.46 rom for the SIDE2?

When using SIDE2 with Ultimate, you should use the PBI BIOS and enable "SIDE Hardware w/button" in the Ultimate BIOS setup screen. You can get the PBI BIOS here. The SDX image (on the same page) contains FDISK, etc, but nothing inherently required to get the HDD to work. If FDISK can't find the adapter, more thank likely SIDE Hardware isn't turned on in the Ultimate menu.