1090XL remake

[Mathy]

Hello Brian

 

6 hours ago, reifsnyderb said:

Thinking this through, the downside is that MIO addressed it's memory through a 256 byte window.  Did the redesigned MIO do this differently?

 

It doesn't, AFAIK.  How much different is it to access memory through a 256 byte window than it is through a 16 kbyte window?  And you still have to read byte for byte.

 

I seem to remember that reading an address in the 256 byte window multiple times gets you consecutive bytes (or was that windows?).  I never had a MIO, but have wanted one for a long time.  The only thing I miss in the BlackBox is the extra memory the MIO has.  What I'm trying to say with this is, I never had one, so all I know about is what I read and heard about it.

 

Sincerely

 

Mathy

 

 

 

[reifsnyderb]

30 minutes ago, Mathy said:

How much different is it to access memory through a 256 byte window than it is through a 16 kbyte window?

 

There would need to be 2 banking registers instead of one.  So, there would need more decoder logic and another register.  It would be a lot easier to do 16k at a time and only have one banking register.

 

30 minutes ago, Mathy said:

 And you still have to read byte for byte.

Absolutely.

30 minutes ago, Mathy said:

I seem to remember that reading an address in the 256 byte window multiple times gets you consecutive bytes (or was that windows?). 

The bytes would be consecutive as long as you increment the bank registers in order.  The 256 bye memory region is sometimes called a window.

 

Best Regards,

 

Brian

 

 

[Mathy]

Hello Brian

 

Which methode (256 byte or 16 kbyte windows) would need more/less CPU cycles to access the memory?

 

Sincerely

 

Mathy

 

[+kheller2]

Maybe we are getting off course.  Is this card to be just "another" bank memory expansion?   That's fine and we should use some standard, and I see that the Axlon method won't work... so pick another

 

As for the MIO concept, besides the P, R, and HD interfaces, the RAMDisks (with the ability to swap around drive letters) and print spooler was heavily used.  I only mentioned this because the MIO FW has been tweaked before for the reproduction run of the board.  Tweaking it again to be a 1090 card might be easy.

 

I love all the work you've put into the 1090 remake and am considering diving into this. But if there was a MIO board people would be tossing money at you.

[_The Doctor__]

10 hours ago, reifsnyderb said:

After some more thought about the 4MB board, it appears it would be only usable as a RAM disk, with possibly a decent print spooler.  I am concerned about board real-estate as it's pretty tight already.  Removing the base memory will free up some space.  Adding a ROM chip and battery back-up chip may take up too much space for a 1091.  I'll have to try to lay it out and find out.

 

I don't think the code to use the board for a RAM disk will be much of a problem.  Since there isn't a ROM on the board, I can configure the current 4MB board to just use one of Atari's reserved banking addresses, another address to enable the banking, and run the ROM off of a firmware board.  So the current board won't be a total waste for R&D purposes.

there is enough space between cards you can populate in a limited fashion on both sides of the PCB

[reifsnyderb]

14 hours ago, _The Doctor__ said:

there is enough space between cards you can populate in a limited fashion on both sides of the PCB

True.  I could put through-hole components on one side of the board and surface mount on the other.  The problem is the board.  I am trying to keep with 2 layer boards so as to keep the price down.  Going to a 4 layer board would add another $35 for a run of boards due to an "engineering fee".  I am trying to avoid that.

[reifsnyderb]

16 hours ago, kheller2 said:

Maybe we are getting off course.  Is this card to be just "another" bank memory expansion?   That's fine and we should use some standard, and I see that the Axlon method won't work... so pick another

 

As for the MIO concept, besides the P, R, and HD interfaces, the RAMDisks (with the ability to swap around drive letters) and print spooler was heavily used.  I only mentioned this because the MIO FW has been tweaked before for the reproduction run of the board.  Tweaking it again to be a 1090 card might be easy.

 

I love all the work you've put into the 1090 remake and am considering diving into this. But if there was a MIO board people would be tossing money at you.

I just looked up the MIO board.  From the manual, it has:

 

"a parallel printer interface, a serial interface for printer or MODEM, a printer buffer, a bootable RAMDISK, and an SCSI/SASI interface for hard disk drives"

 

A bootable RAM disk and a printer buffer can all be done with memory.  Does anyone still use the old parallel printer interfaces (db25?), serial interfaces, and/or SCSI interfaces?

 

[+kheller2]

56 minutes ago, reifsnyderb said:

I just looked up the MIO board.  From the manual, it has:

 

"a parallel printer interface, a serial interface for printer or MODEM, a printer buffer, a bootable RAMDISK, and an SCSI/SASI interface for hard disk drives"

 

A bootable RAM disk and a printer buffer can all be done with memory.  Does anyone still use the old parallel printer interfaces (db25?), serial interfaces, and/or SCSI interfaces?

 

-- it also was supposed to have an 80 column card .. but never happened.

I think that some people still use SCSI interfaces, today of course it wouldn't be the driving factor.  However, the ability to take my old MIO drive and plug it into an expansion box and go is pretty nifty. The TI 99/4a PEB was something I thought was incredibly sexy at the time.  For me the MIO SCSI interface and the ability to have multiple RAM disks that I could swap around and boot from was very important.  

 

And let's be real here: exactly how many people will derive benefits out of re-creating 1090s and the old cards, even with new tech or newer functions?  *I* would still buy a 1090 to run floppy drives, scisi drives, serial and parallel ports, RAM disks etc..  but I'm weird like that.

 

[_The Doctor__]

3 hours ago, reifsnyderb said:

I just looked up the MIO board.  From the manual, it has:

 

"a parallel printer interface, a serial interface for printer or MODEM, a printer buffer, a bootable RAMDISK, and an SCSI/SASI interface for hard disk drives"

 

A bootable RAM disk and a printer buffer can all be done with memory.  Does anyone still use the old parallel printer interfaces (db25?), serial interfaces, and/or SCSI interfaces?

 

The MIO is still in use in my home, and has never not been

The same for the Black Box

The MIO allows you to use the serial port for printers as well, it is redefinable and works. So parallel and serial printers, and other such fun. The were going to expand those abilities to allow for R: devicel assignments 1,2,3,4 for the serial port as well so it would co exist with other serial solution... PR connection  1 and 2 MIO 3 or 4, stuff like that.

IDE and cf adapters work but I am using real spinning hard disks. Just got some replacements recently and have to wait on a 50 pin ribbon cable and connectors to crimp on

 

The way the RAM disks work is indispensable, I have use them to save to when the DOS disk was incompatible or even now when fujinet tells me it won't save to the .atr for no good reason. Just save the RamDisk, reboot Fujinet and then copy from ramdisk to atr or real disk.

The RamDisks held much BBS info that does not change like menu screens commands etc. This makes for a fast BBS experience. But everyone prolly knows this noise I am spewing

 

The MIO has bailed me out so many times with it's little basket of nice touches, I wish the next revision to the rom came out.

 

I wished the Black Box had those MIO ram disks and much wanted assignable R: device numbers for both PBI solutions would have been so nice... such dreams!

Edited August 15, 2023 by _The Doctor__

[reifsnyderb]

2 hours ago, kheller2 said:

-- it also was supposed to have an 80 column card .. but never happened.

There is an HDMI 1090XL card now. 

2 hours ago, kheller2 said:

I think that some people still use SCSI interfaces, today of course it wouldn't be the driving factor.

My thought as well.  I just can't see a lot of demand for a SCSI interface.

2 hours ago, kheller2 said:

 However, the ability to take my old MIO drive and plug it into an expansion box and go is pretty nifty.

That would be cool.  But what's the demand?

2 hours ago, kheller2 said:

For me the MIO SCSI interface and the ability to have multiple RAM disks that I could swap around and boot from was very important.

I like the idea of multiple RAM disks or even bootable RAM disks.  Being able to turn on my Atari and have a booted OS, without messing around with floppy disks, would be a nice "modern" feature.

2 hours ago, kheller2 said:

And let's be real here: exactly how many people will derive benefits out of re-creating 1090s and the old cards, even with new tech or newer functions?  *I* would still buy a 1090 to run floppy drives, scisi drives, serial and parallel ports, RAM disks etc..  but I'm weird like that.

 

It sounds like there is some interest in the 1090 parallel/serial card.  Atari had a schematic for a parallel/serial card.  I don't know if any pictures exist of it, however.  I'll have to check when I am able.  Chip availability could be a concern as well.  Either way, if enough people are interested, re-creating this card is possible.

 

Maybe a SCSI and IDE driver interface could be created as well.

 

After some thought, it would be possible to make a MMU upgrade that would also monitor the R/W signal.  By only enabling the ROM chip selects on a read signal, Axlon compatible banking would work.  Of course, this does defeat the idea of expanding the computer by only plugging in the 1090XL.  (I really hate the idea that my 800 has more memory than my 600XL with the 1091.)

 

I was reading up a little on the MIO.  A possibility for powering the RAM would be to add an external USB connector.  Then put a jumper on the board to select how the SRAM chips are powered.  This would avoid the necessity of that extra expensive chip and battery holder.

 

Edited August 15, 2023 by reifsnyderb

[_The Doctor__]

The MIO is already externally powered, the battery back up is the thing.

[+kheller2]

1090 Serial/Parallel schematics and even ROM exist, including the specs from Mr. Decuir.   I'm not sure one would want to follow those specs vs building a better faster card that COULD use existing drivers for BB/MIO/PRCon etc..  Or one could start using the specs as is and building drivers for 1090 ... using D800-DFFF. D1xx, and GPDH.

 

I'm just saying that if the 1090 could do some of the great features of the MIO/BB in a similar way or better ... then you got a product.

 

[reifsnyderb]

8 minutes ago, kheller2 said:

1090 Serial/Parallel schematics and even ROM exist, including the specs from Mr. Decuir.   I'm not sure one would want to follow those specs vs building a better faster card that COULD use existing drivers for BB/MIO/PRCon etc..  Or one could start using the specs as is and building drivers for 1090 ... using D800-DFFF. D1xx, and GPDH.

 

I'm just saying that if the 1090 could do some of the great features of the MIO/BB in a similar way or better ... then you got a product.

 

Looking for the MIO schematic....

 

Edit to add:  This would have to go on a full-size 1090 card.  It would be too big for a 1091.

Edited August 15, 2023 by reifsnyderb

[+kheller2]

20 minutes ago, reifsnyderb said:

Looking for the MIO schematic....

 

Edit to add:  This would have to go on a full-size 1090 card.  It would be too big for a 1091.

Well, maybe....    you don't have to build everything into one card.  A "Ramdisk" card would be nice, along with a P:S: card, along with...

 

The MIO 80c card I think is pictured here: 

and here:

 

 

The schematics should be floating around for the main MIO somewhere, as it's been reproduced once already.

 

[Mathy]

Hello Brian

 

@pancio developed a PBI/ECI card that has a bi-directional centronics port.  Maybe that's a start of a 1090XL parallel printer port.

 

Sincerely

 

Mathy

 

[kenames99]

I have actually started a serial/parallel card board design. The original one is pretty outdated and undocumented. It would take a long while to decipher the design. I also have not seen the .rom file for it.

 

Ken

 

[_The Doctor__]

serial,parallel,usb,ethernet,wifi,scsi/ide,ramdisk,spoolers,video and so on and so on and so on... like the shampoo comercial

[kenames99]

Well, of course!😁

[Mathy]

Hello Doc

 

23 minutes ago, _The Doctor__ said:

serial,parallel,usb,ethernet,wifi,scsi/ide,ramdisk,spoolers,video and so on and so on and so on... like the shampoo comercial

 

The 1090 is an expansion board.  Without expansions it's just a board.

 

Sincerely

 

Mathy

 

Edited August 16, 2023 by Mathy

[reifsnyderb]

13 hours ago, Mathy said:

Hello Doc

 

 

The 1090 is an expansion board.  Without expansions it's just a board.

 

Sincerely

 

Mathy

 

That's why we are making expansions.   🙂

[+kheller2]

I can't wait to plugin that Apple II card!

It will sit right next to the C64 card.

They will shun the CP/M card.

Eventually the 1090 will be the controller, and the Atari just an "External" card to it.

 

[_The Doctor__]

maybe a game of stuff the bus will come to pass where the 1090/91 will feed data to specific chips within the Atari, the supporting code to help pulled from PBI then executed and the 1090 takes control / able to use the custom chips in the Atari

Edited August 17, 2023 by _The Doctor__

[reifsnyderb]

22 minutes ago, _The Doctor__ said:

maybe a game of stuff the bus will come to pass where the 1090/91 will feed data to specific chips within the Atari, the supporting code to help pulled from PBI then executed and the 1090 takes control / able to use the custom chips in the Atari

Some lines, like R/W and Phi2 are buffered in the 1090.  So, the 1090/1 can't take over.  The CPU could, and with the right software, take instructions from a card to become a "slave" to the card.

Edited August 17, 2023 by reifsnyderb

[reifsnyderb]

Ok.  I took the Axlon card (attempt) and greatly modified it.  The 128k SRAM chip was removed and a Flash ROM put in it's place.  The ATF16V8 connections were re-worked and jumpers for device selection were added.  A socket for the SRAM battery back-up chip was added and a battery holder for a CR2032 was added.  An external battery could also be used to keep the SRAM powered.  I also modified the banking such that the ROM can even bank 16k into the usual banking location so there is a total of 18k of space for drivers, setup firmware, etc.  I figure that a small section of the main 4MB of SRAM could be used to save any settings as well.  With the right firmware it should be possible to boot from this card.  I figure that some of the memory could also be used as a print spooler.  (Could you imagine a 4MB print spooler for an Atari 1025???      )

 

As an experiment, of sorts, I tied the /RESET line to an ATF16V8 to use to reset it's registers.  Using the Happy Enhancement schematic as a guide, I also added a capacitor to the chip's clock signal so there should be a slight delay between the signal and the clock pulse being received.  So, the system reset button should now reset the latches on an ATF16V8....if it works as planned.  If not, the power will have to be cycled as usual.

 

This will also fit the 1091!   🙂

 

Here's a couple pics....

 

 

Edited August 18, 2023 by reifsnyderb

[reifsnyderb]

Here's version 0.94 of the "The Comprehensive Guide To The Atari Parallel Bus".  I've added a complete section on the MMU as it affects what can and cannot be directly controlled through the PBI.  (I should probably add some info about how the ECI affects this in the next release.)

 

I also added the MMU section because of getting burned on my 4MB Axlon card attempt.  I figure that's another lesson learned for anyone who wants to make PBI devices.

 

Parallel Port Guide v0_94.pdf