Atari's Plato Cartridge question

[_The Doctor__]

and the actual touch pad being back in would really take it to the top. You can see a prototype touch pad in some cyan engineering footage for sure. but the actual sold to the public touch pad is the goal.

Edited February 24, 2018 by _The Doctor__

[a8isa1]

We can easily push it to 16K.

 

-Thom

We could but it wouldn't be a linear 16K cart.

 

TLP uses $2000-$3FFF for normal mode display data and $4000-$9FFF for the zoom mode display data. display lists at $1000 and $1200, IIRC

[_The Doctor__]

I think he's talking about moving whole chunks to keep each within their own boundaries, a re-writing for sure. Not just a buffer hack.

[a8isa1]

I think he's talking about moving whole chunks to keep each within their own boundaries, a re-writing for sure. Not just a buffer hack.

I was trying to point out that TLP uses some of the address space normally used by a 16K cartridge, specifically the $8000-$9FFF region.

 

TLP needs 32K for its huge 512x512 virtual screen used by its zoom mode. The normal mode, the 320x192 scaled mode, uses another 8K.

Edited February 24, 2018 by a8isa1

[_The Doctor__]

yep that's 48k total give or take for memory in use. that leaves 16k for code in a cart.... 64k total... on todays Atari.. you are correct things would have to be move around

of course an 800 could handle that as well depending on memory and mods though something would have to be worked out 52k or more because that's room for 4k of code for most 800 users, but if it moves out of a cart format or even goes the route of XE banking....also compression could be used to extract itself from the smaller rom if 16k is wasting too much of the ram area... so a compressed version on an 8k? maybe... stick PLATO on newell ramrod board perhaps

((SILLY TIME Bonus Real or Not)or the good ole 8k 4k left right cart slot deal on an 800.. PLATO 800 the 2 cart set..))

Edited February 25, 2018 by _The Doctor__

[a8isa1]

We could but it wouldn't be a linear 16K cart.

 

TLP uses $2000-$3FFF for normal mode display data and $4000-$9FFF for the zoom mode display data. display lists at $1000 and $1200, IIRC

Too bad I can't count in hexadecimal. $4000-$9FFF is only 24K.

 

The lines are 64 bytes, so 512 pixels across.

Edited February 25, 2018 by a8isa1

[_The Doctor__]

I took what you said to be a rough look at it, not that big a deal, 16+24+8=48k it will still fit an 800 give or take as mentioned, roughly 44 up to 52 is just fine, a little silliness ensues with lack of sleep and other stuff for me. Like right now. Aching so bad I can't see clearly or think. so I'll say something like this...

but your working on it and you caught yourself, it is a battle... eventually you'll win the war.

Edited February 25, 2018 by _The Doctor__

[Bill Lange]

PLATO can be humorous.

 

[+Nezgar]

interesting note... while messing around I noticed the serial port is at 1200 baud, 7 characters, 1 stop bit, even parity, with no flow control while using the patched rom...

 

you mean 7 bits

 

The MSS100, even though configured for 8N1 should be fine if the server and client are talking 7E1 because the MSS100 is just passing bytes with all 8-bits anyway for the endpoints.

 

Do you know if Cyber1 uses 7E1 or 8N1? or maybe that's part of the terminal selection...

 

Edit: it could be interesting to see if TLP actually checks the parity of received bytes if using 7E1. It may have a mechanism to request a re-transmission of data where parity errors were detected. Occasional garbled data in the stream would have been a thing to account for in software back then... Also if TLP generates parity for outgoing bytes, or I'm not sure if the 850 does that itself.

Edited February 25, 2018 by Nezgar

[tschak909]

The ASCII protocol is literally straight ASCII, 7 bits. The parity bit was originally specified to be even parity, but is now ignored.

 

-Thom

[_The Doctor__]

Trust that setting the Lantronix MMS100 2400,7 bit chars,1 stop bit, even parity, with no flow control is helpful throughout the product line... I've hear the argument before while using these for BBS and other applications, Lantronix included it for a reason, and it detects and changes it for a reason. It also reports it for a reason as well...

 

some people can also go on about 2400 being the flow rate as opposed to baud or bps. I don't worry about that either. It's been baud forever in the modem world and often is assumed to be the label used and often left out or not even mentioned.

 

While you and I might immediately know what 2400,7,1,E is and set everything up accordingly, a new to this world individual might not have a clue what that is. I haven't seen a terminal program accept those numbers any other way though . Putting a 1 or 2 in the bit per char slot just doesn't work you see, nor does putting 7 or 8 in the stop bits slot, and certainly neither of those worked in the flow rate slot. We still tend to give the number in the same order through the ages and have the same discussion about them though...

Edited February 25, 2018 by _The Doctor__

[_The Doctor__]

@tschak909

When you are making your game remember limited sound is also implemented at the very least in the form of bell.

You might consider limited use of it for death or something like that.

 

Have you tested to see if Atari IST to IST Atari code is in the cart? possible R2: to R2: etc.. not sure if passed through the entire network or not, but I thought it was done to keep a cluster from eating up too much 'bandwidth' lol

Edited February 26, 2018 by _The Doctor__

[tschak909]

nope, that would require an 8080 simulator, there isn't one in the cart.

 

-Thom

[_The Doctor__]

Well I can confirm the beep (system bell) is still in here at least.

 

so die beep pause beep, back at bashing bad guys again.

 

As for linking together, I didn't think it needed the full 8080 but just the link/protocol. I'm not sure we need it all, I know we can't download full code and run it. I was of the mind there were limited functions available but perhaps not in the very tight confines it now sits.

Edited February 26, 2018 by _The Doctor__

[tschak909]

No, no such link sharing functionality is present.

The PLATO cartridge almost completely implements an IST-III terminal, minus the code push features which make Micro-TUTOR work. This would only work on terminals that implement the classic 20-bit CIU/NIU protocol used by the PLATO PPT, and would require a site controller. This is currently done so that the remaining Magnavox PLATO-IV terminals can function at the Computer History Museum.

 

-Thom

[_The Doctor__]

Nice, that's cool.

Not that this would be done, but maybe an emulation could be done even if it relied on perhaps indus cpm or atr800, or add on upgrade of the Atari (processor board etc. Rapidus VBXE, or whatever will help)

 

Sharing the SIO cuts into this a bit but could be done. The idea wasn't just to be a stand alone solution for both of those but also to in theory use them as co processors, Most just made the Atari the terminal and ran CP/M on the boxes, which also opens up the idea of using the CP/M of either but perhaps now using TLP terminal as the display for those... This would be quite nice

Edited February 26, 2018 by _The Doctor__

[tschak909]

Just reminding, I am tschak909 / inted on cyber1, if anyone needs/wants to term talk me.

 

-Thom

[_The Doctor__]

I went looking for the Gooch Cybernetic Synthesizer, my search fu must be weak... I found nothing.... it was an add on device. Wikipedia told me to go pound sand also...

 

tried again, got some results.. google must not have had it in it's cache.. it does now..

Edited February 26, 2018 by _The Doctor__

[tschak909]

There isn't much information on the GSW, the only implementation of it that still exists is in the PTerm implementation, and it's only available in the classic 20-bit protocol, which TLP does not do, at all.

 

-Thom

[tschak909]

Although this may very well be, because of the fact that the ASCII connection isn't timed very tightly...

 

Meanwhile, the disassembly is proceeding nicely, from the looks of things.

 

I'm also progressing on the Android implementation of the PLATO Terminal (a friend and colleague is currently starting work on the iOS implementation), Cyber1's systems staff have been unbelievably forthcoming with required information to properly implement a terminal (and I just received a copy of the Mac and Sun implementations from 1990, which are actually very clean C code, which helps with cross referencing proper implementation behavior. Cheers to cyber1, they are good people.

 

-Thom

[tschak909]

Ok, so guys... I have a complete copy of the CYBIS system running in an emulator. This isn't the complete copy running on cyber1, but close...it has all the officially published lessons (the ones that begin with 0)...would anyone be interested in using one that is set up just for us Atari PLATO users? All....4 of us?

 

I could e.g. customize every part of the system and add a menu router to make things easier to bounce around in...

 

-Thom

[tschak909]

Having a little fun. This was cooked up in the PLATO screen designer, which is a handy little graphics tool on the system to allow authors to visually draw a screen, which is then turned into TUTOR commands. I did this in about 30 minutes.

 

 

The result in both PTerm and the Atari PLATO cartridge.

 

-Thom

[tschak909]

Nobody interested?

 

-Thom

[Kyle22]

Nobody interested?

 

-Thom

Very interesting, just havent had time to check it out yet.

[troff]

You mentioned the screen designer saves the image you draw as TUTOR commands? How easy would it be to then import that into a new lesson? It would be fun to create a new lesson... something like Atari 101 that briefly covers the different Atari 8-bits. Talk a little about the genesis of the 8-bit line with pics of the 400 and 800. A little about the release of the short lived 1200XL. The 600XL and 800XL. The unreleased prototypes such as the 1450XLD. The Tramiel buyout and the XE line. Cool way to introduce people to the Atari computers.