1088XLD - Custom A8 Computer in 1050 Drive Case

[Curt Vendel]

Michael,

 

Since your bored, you may want to start layout of a version that will fit into a 1450XLD case, because that's gonna be next on the agenda... I've had a friend 3D scan a 1450XLD case a few months back, the data is ugly but its cleaning up and see 3D models of the 1450XLD case parts will be ready and no reason why we couldn't look to have a run of cases made...

 

 

 

Yep Bryan mentioned that to me on my last UAV order, but I've got 4 or 5 in stock which will easily get me by for quite sometime, especially since this project probably won't even go to a PCB house until the end of the year.

 

--------------

 

So just a heads up. Since the 1050 comes with two SIO ports I opted to not integrate an SIO2PC, and instead will be going with an external one like the one Lotharek makes (got 3 of those). Other than that small omission, the 1088XLD will come with everything that the 1088XEL has except for the following...

• No DIN-13 Video Pass-Thru
• No ECI (although those signals are on the MPBI header)

 

[+mytek]

Michael,

 

Since your bored, you may want to start layout of a version that will fit into a 1450XLD case, because that's gonna be next on the agenda... I've had a friend 3D scan a 1450XLD case a few months back, the data is ugly but its cleaning up and see 3D models of the 1450XLD case parts will be ready and no reason why we couldn't look to have a run of cases made...

 

I'm about as bored as you are Curt . Seems like you and I just can't sit still for long before we're getting ourselves in trouble with more Atari stuff in the works.

 

Yeah I saw your post about this over in the 7800 arena. Looks like a very ambitious project for sure . And I know of at least one individual that'll be cr_pping his pants when he sees what you are proposing (Kyle). However this is not the direction my projects are heading, so I don't think I'll be designing a motherboard anytime soon for that case (although it was suggested that I do so about a year ago). I just like the whole separate keyboard aspect that the PC and many others adopted, and that is the road I have decided to take. But I know for many users on AA that they'll absolutely love something like a rebirth of the 1450XLD. Best of luck on your new venture .

[Kyle22]

Michael,

 

Since your bored, you may want to start layout of a version that will fit into a 1450XLD case, because that's gonna be next on the agenda... I've had a friend 3D scan a 1450XLD case a few months back, the data is ugly but its cleaning up and see 3D models of the 1450XLD case parts will be ready and no reason why we couldn't look to have a run of cases made...

 

 

 

I LOVE the idea! I sure hope this goes forward.

[AtariNerd]

I do like the purity of the 1400/50XL design, but the interesting part of Michael's current board designs, is that they would probably fit under just the keyboard area portion of those cases, by themselves, leaving the area beyond, toward the squared, boxy portion of the case, clear. Plenty of room to stick other goodies. For example, on the 1400, the right and left sides are large enough to encapsulate an ST-style drive or MIDI ports. Or on the 1450, the rear could be modded for slots, and one of the two card bay projects crammed in there.

Edited September 26, 2018 by AtariNerd

[ivop]

Personally, I'd be more interested in a 600XL motherboard. SMD and most expansions included in the design instead of plugin boards. But well, there's no Open Source U1MB equivalent, no Sophia, so I guess that's another cook

 

Anyway, I also like this 1050 design and I'm seriously considering building one when it's finished. Perhaps an empty 600XL could be repurposed as a keyboard. BITD (early 90's) F#Ready had his XL build in a tower and used a thick cable with a ton of wires to connect to an empty 600XL as a keyboard.

Edited September 26, 2018 by ivop

[Kyle22]

Just a thought: If the PIC chip was also attached to the /Command line, would it be possible to make MIDI show up as a proper SIO device? Something that (using a small handler) would enable a BASIC program to do something like this:

 

10 open #1,8,0,"M:"

20 M$="ATASCII Character string containing raw MIDI data"

30 ?#1,M$

 

or possibly use PUT.

 

Does this have any merit?

[+mytek]

Yep not everything is open source. So making it all inclusive on a new motherboard is problematic. Although with foft's designs who knows? Of course anything is possible with the right kind of determination behind it .

-----------

I finished my firmware overhaul, going with a CASE structure in the Intro Music Player (so much easier now to follow the flow).

;**********************************************************************
;*  Name    : MIDIctrl_PIC12F1572.gcb
;*  Author  : Michael St. Pierre
;*  Notice  : Copyright (c) 2018 Mytek Controls
;*          : All Rights Reserved
;*  Date    : 9/26/2018
;*  Version : 1.0 BETA9
;*  Notes   : MIDI clock, gated MIDI IN & OUT, and Intro Music Player
;*          : for use with Dream Blaster S2 wave table module.
;*          : Based on Microchip PIC12F1572 interfaced to Atari SIO
;*          : Compiled in Great Cow Basic version 0.98.02
;*          : Source: <https://gcbasic/sourceforge.net/>
;*
;*
;*  This program is free software; you can redistribute it and/or modify
;*  it under the terms of the GNU General Public License as published by
;*  the Free Software Foundation, either version 2 of the License, or
;*  (at your option) any later version.
;*
;*  This program is distributed in the hope that it will be useful,
;*  but WITHOUT ANY WARRANTY; without even the implied warranty of
;*  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
;*  See the GNU General Public License for more details.
;*
;*  You should have received a copy of the GNU General Public License
;*  along with this program.  If not, see <http://www.gnu.org/licenses/>
;*
;**********************************************************************

;Default to 16 Mhz internal factory calibrated oscillator

  #chip 12F1572

;Port Equates

  #define MIDI_OUT PORTA.0
  #define DATA_OUT PORTA.1
  #define MIDI_CLK PORTA.2
  #define ENABLE PORTA.3
  #define MIDI_IN PORTA.4
  #define DATA_IN PORTA.5

  #option Volatile MIDI_OUT
  #option Volatile DATA_OUT

;Set PWM parameters and activate Midi Clock on PORTA.2

  #define HPWM16_3_Freq 31.250   '31.250khz
  #define HPWM16_3_Duty 50       '50% duty
  HPWM16On ( 3 )

;Set-up gated MIDI output port and force it high

  Dir MIDI_OUT Out
  MIDI_OUT=1

;Set-up UART for MIDI output

  #define USART_BAUD_RATE 31250
  #define USART_TX_BLOCKING

  wait 2250 ms  'allow for Synth and monitor set-up time

;Set-up variables for use by PlayTune Player

  Dim length As Word
  Dim index As Word
  Dim delay As Word
  ReadTable intro,0,length  'retrieve length of table
  length=length+2           'and adjust value for read routine

;Custom PlayTune Player (sends intro music to Synth upon power-up)

  For  index = 1 to length

  ReadTable intro,index,data 'Retrieve data
  MidiStat=(data AND 0xF0)   'pull out status nibble

    Select Case MidiStat

    Case 0x80  'Status = Note Off
        data=data+0x30       'Create Mode Change byte (0xBx)...
        HSerSend data        'Send it!
        data=0x7B            'Load ALL NOTES byte...
        HSerSend data        'Send it!
        data=0x00            'Load Velocity OFF byte...
        HSerSend data        'Send it!

    Case 0x90  'Status = Note On
        HSerSend data        'Send it!
        Gosub GetByte        'get Note Value & Send it!
        Gosub GetByte        'get Note Velocity & Send it!

    Case 0xC0  'Status = Program Change
        HSerSend data        'Send it!
        Gosub GetByte        'get Instrument & Send it!

    Case 0xF0  'Status = End of song
        GoTo Done 'Yes, leave!

    Case Else  'Status = Delay
         tmpH=data           'store high byte of delay
         index=index+1       'get low byte of delay
         ReadTable intro,index,data
         delay=(tmpH*256)+data  'combine to create word...
         wait delay ms       'and use it for in-between note delay

    End Select

  Next

  Done:
  TXEN=0  'Disable UART TX output --- give up port

;====================================================
; MAIN PROGRAM LOOP --- BEGIN
;====================================================

Do Forever:

    Do While ENABLE=1 'Keep looping if  Motor Control is enabled

    ;activate SIO related outputs (take over SIO bus)
    Dir DATA_OUT Out
    Dir MIDI_CLK Out

    ;transfer DATA input to gated MIDI output
    temp=DATA_IN
    MIDI_OUT=temp

    ;transfer MIDI input to gated DATA output
    temp=MIDI_IN
    DATA_OUT=temp

    Loop

;float SIO related outputs (free up SIO for other devices)
  Dir DATA_OUT In
  Dir MIDI_CLK In

;and force gated MIDI output high
  MIDI_OUT=1

Loop

;====================================================
; MAIN PROGRAM LOOP --- END
;====================================================

;Load Intro Music from file into ROM when flashed

  'PlayTunes binary format created with midi2tones.exe from MIDI file
  'Source: https://github.com/MLXXXp/midi2tones
  'Execute from command line: midi2tones -v -b -t16 basefilename
  'Optionally edit intro.bin for missing parameters from MIDI file
  'This would be especially true for a Program Change (new instrument)
  'or to adjust volume by changing velocity parameters

Table intro from "intro.bin"


  Sub GetByte
     index=index+1              'point to next byte...
    ReadTable intro,index,data  'read it...
    HSerSend data               'and Send it!
  End Sub

[ivop]

@Kyle22: There's already a MIDI CIO handler from the 80's that's MidiMate/Max/Muse and XEL-MIDI compatible:

 

https://github.com/ivop/midimuse/tree/master/software/midimon

 

I reused this handler for midimon.

Edited September 26, 2018 by ivop

[+mytek]

Now to bring this topic back to the 1088XLD Motherboard. Here is an update showing where things are currently at in the PCB layout (please pay no heed to the component ID numbers, since they will get re-sorted before routing).

 

 

A place for P-COXOX to connect it's audio and select signal will be coming back. I just need to really think through where it will go for optimum accessibility.

 

The MPBI connector is gone, and has now been replaced with the more traditional '130XE like' CART/ECI connection for parallel access. I did this mainly because only one thing has ever been made that used the MPBI, and that device has been integrated into this board design (IDE). And from a practicality standpoint, there are already several devices that use the more standard parallel port, or could more easily be accommodated with this more traditional setup. I personally don't plan to have an actual Cartridge Port on my version of this, but I believe it would still be quite possible to use something like Dropcheck's Cartridge Slot Extender and a custom front panel to get this back for those that really want it.

 

As this project progresses, I find myself leaning more and more towards a public release so that others may enjoy . But be warned, it's still a long ways off before something like that happens in reality.

 

[ivop]

Funny how an image like this can bring the same joy as art. Although I guess most "sane" people wouldn't understand

[+mytek]

Funny how an image like this can bring the same joy as art. Although I guess most "sane" people wouldn't understand

 

Too funny, because I feel the same way .

 

I'm working on an idea to also extract a VSYNC signal from the VGATE chip (I have one unused pin), and if I can do that, then the LM1881 will not be needed. Also thinking of moving the MIDI MCU in it's place since a lot of the signals required by the MIDI DIN Interface board will originate from that chip as well as the SIO connector. I will also scoot it a bit more over to the right, since I just noticed its a little bit too close to the DIN interface header (always something to tweak).

[Level42]

 

As this project progresses, I find myself leaning more and more towards a public release so that others may enjoy . But be warned, it's still a long ways off before something like that happens in reality.

 

You can’t go back now.....teasing us sooooo much with this brilliant idea

[+mytek]

You can’t go back now.....teasing us sooooo much with this brilliant idea

 

Yeah you guys should know by now that I don't tease about this stuff. If I say I'm going to do it, then it will happen .

[+mytek]

Now back to MIDI Synthesizers...

 

You know we can probably build a pretty cheap Dream Blaster board by using the same chip, and just designing a board that is a drop-in replacement for the original S2 board.

 

Chip: SAM2695 AliExpress: 10 for $51.15 with FREE shipping ($5.12 each)

 

SAM2695 Break-Out Board available from OshPark

 

Schematic

PCB

Download: Eagle Files

 

The OshPark offering isn't what I had in mind, since it just makes more sense to use the standard Wave Blaster connector and pinout. But it does present a pretty good way to do it, with a very uncomplicated layout.

 

I figure that a drop-in Dream Blaster S2 version could be done for around $10-12 total (vs. $40 for Dream Blaster S2), assuming that at least a 10 board run from ALLPCB was done, and the chips were purchased from AliExpress. Problem is the soldering, because that SAM2695 chip has a pretty fine pitch (QFN48 package).

 

[+Stephen]

Now to bring this topic back to the 1088XLD Motherboard. Here is an update showing where things are currently at in the PCB layout (please pay no heed to the component ID numbers, since they will get re-sorted before routing).

 

1088XLD_pcb.png

 

A place for P-COXOX to connect it's audio and select signal will be coming back. I just need to really think through where it will go for optimum accessibility.

 

The MPBI connector is gone, and has now been replaced with the more traditional '130XE like' CART/ECI connection for parallel access. I did this mainly because only one thing has ever been made that used the MPBI, and that device has been integrated into this board design (IDE). And from a practicality standpoint, there are already several devices that use the more standard parallel port, or could more easily be accommodated with this more traditional setup. I personally don't plan to have an actual Cartridge Port on my version of this, but I believe it would still be quite possible to use something like Dropcheck's Cartridge Slot Extender and a custom front panel to get this back for those that really want it.

 

As this project progresses, I find myself leaning more and more towards a public release so that others may enjoy . But be warned, it's still a long ways off before something like that happens in reality.

 

Damn that is pretty! I will soon be making photo frame like "glass boxes" for some PCBs I have, and will be hanfing them on the walls around my work bench.

[+Ripdubski]

Midi Tune Playback on Power-Up now Implemented.

 

 

 

Demo Video

https://www.youtube.com/watch?v=pYsZOUHzFlY&feature=youtu.be

 

 

Agent Orange!

[+mytek]

Agent Orange!

 

Yep you got that right .

 

There are quite a few more Depeche Mode songs in MIDI format that I've stumbled across, but I really like this particular tune.

[seastalker]

I really like the R2-D2 type beeps when you see the U1MB logo. @mytek, Maybe asking if the board would work with a Sophia is not the right question (it's a A8 after all), but is there room in the back if someone wanted to put a jack for one?

[ivop]

Now back to MIDI Synthesizers...

......

I figure that a drop-in Dream Blaster S2 version could be done for around $10-12 total (vs. $40 for Dream Blaster S2), assuming that at least a 10 board run from ALLPCB was done, and the chips were purchased from AliExpress. Problem is the soldering, because that SAM2695 chip has a pretty fine pitch (QFN48 package).

 

I also considered this early on, but decided to try new designs with the CS9236/CS4333 combo and the VS1053b instead. Mostly because the SAM2695 is really impossible to solder by hand (at least by me). I have a spare chip and it's about the size of the nail on your pinky-toe Sockets cost $80 and up.This meant I would have had to have my prototype design assembled in China, which is pretty costly for a small run of 1. I also thought it a little rude to serdashop to sort of copy their Dream Blaster. And, the more the merrier I'm really curious how the other chips sound, as most of the 90's MIDI chips have their own character.

 

[+mytek]

I really like the R2-D2 type beeps when you see the U1MB logo. @mytek, Maybe asking if the board would work with a Sophia is not the right question (it's a A8 after all), but is there room in the back if someone wanted to put a jack for one?

If you want to use the Sophia RGB version (or a VBXE), then that support is already built-in via the DIN-13 audio/video connector and RGB header. So that would give you a standard SC1224 monitor connection, or an easy SCART connection with the Cool Novelties 1088XEL cable (other break-out cables are possible). BTW that same DIN-13 jack also serves up the composite and S-Video as well, and I will be looking into seeing if Cool Novelties can also produce a cable with that in mind.

 

For a Sophia DVI a hole would need to be made in the case somewhere.

[+mytek]

I also considered this early on, but decided to try new designs with the CS9236/CS4333 combo and the VS1053b instead. Mostly because the SAM2695 is really impossible to solder by hand (at least by me). I have a spare chip and it's about the size of the nail on your pinky-toe Sockets cost $80 and up.This meant I would have had to have my prototype design assembled in China, which is pretty costly for a small run of 1. I also thought it a little rude to serdashop to sort of copy their Dream Blaster. And, the more the merrier I'm really curious how the other chips sound, as most of the 90's MIDI chips have their own character.

 

Yes you are right on both points. Near impossible to hand solder chip, and not nice to Serda. So maybe it's good that it wouldn't be an easy DIY possibility.

[seastalker]

If RGB, S-Video and Composite all are native to the DIN-13 port, do you mean that these video standards are provided out of the box or that you'd still need a UAV or Sophia RGB to actually provide the ability to output as such? I'm assuming the latter since your jpeg literally shows a place to put in a UAV. That and you specified the DVI version of the Sophia. It may sound daft, but good to document here for our future Atarians that may find this thread. If one can wire both a UAV or RGB Sophia to the DIN-13, that would be the ultimate Atari video.... playing on an 80's CRT with S-video or through RGB via SCART/COMPONENT connection.

[+mytek]

If RGB, S-Video and Composite all are native to the DIN-13 port, do you mean that these video standards are provided out of the box or that you'd still need a UAV or Sophia RGB to actually provide the ability to output as such? I'm assuming the latter since your jpeg literally shows a place to put in a UAV. That and you specified the DVI version of the Sophia. It may sound daft, but good to document here for our future Atarians that may find this thread. If one can wire both a UAV or RGB Sophia to the DIN-13, that would be the ultimate Atari video.... playing on an 80's CRT with S-video or through RGB via SCART/COMPONENT connection.

 

This always seems to be a problem in comprehension of what is provided and what needs to be added for the various video standards to exist. So don't worry because even a few members of the beta test group initially had trouble with this.

 

So here's the break-down...

 

UAV is a normal must have item, and often times considered to be built-in. However it is not an absolute requirement if you can forego standard video. Meaning you could just decide to go with either a VBXE or a Sophia as your sole video source. However I wouldn't recommend that, because at some point you will most certainly want to hook this up to something where the RGB and/or DVI will not work, whereas composite is pretty much universal.

 

BTW, your suggestion of having both the Sophia RGB and the UAV standard video on a single jack (DIN-13) is how it is intended to be on the 1088XLD.

 

For reference...

 

I still need to test out the Color Adjust circuit which will now derive it's bias voltage from a precision 10 V zener fed by the 12 V supply. What I need to do is establish what fixed resistor value will be added to one of the variable resistor legs to produce the correct color pretty much at the center point of the trimmer resistor. Then add this to the schematic.

[+mytek]

Unfortunately it's looking like this project as well as a few others I had planned may not become a reality due to the US/China trade tariffs that have been recently imposed. My PCB supplier is based in China, and if negotiations do not go well, I am hearing that tariffs will increase on January 2019 from the recently proposed 10%, up to as high as 25%.

 

I will continue to work on the idea, but unless I can find a PCB manufacturer with a similar cost as it once was, the reality of having samples made for testing is becoming a no go situation, especially considering what may happen by early next year when I would likely be ready.

 

I'm not going to get too political other than saying I know who to thank for these recent developments, and he lives in the same country as I .

[+Stephen]

That sucks.