• Please review our updated Terms and Rules here

Sphere-1 Replica Project

Thank you for offering, but I'm not sure I could do much eyeballing that would be any more helpful to you than your already careful tracing! YOLO a power up and see what it does!
 
Haha!

What are some common issues you've run into with the Sphere? Or do you have that posted somewhere already?
 
Hi guys,

Am I correct in thinking that, if one was to obtain a Sphere CPU/2 board, populate it correctly, and get a copy of the original EPROM code, then all that would be needed is a keyboard and a video monitor?

If so,

1. Are there any difficult to find chips on the board?
2. Is a copy of the original EPROM code available?
3. Was there a special keyboard, or is there documentation on connecting a keyboard?

Thanks vey much in advance for your patience with my naive questions.

smp
 
Some idiot called Dave Roberts re-entered all of the source code listings (comment spelling errors and everything) to recreate the ROM images.

All available on Ben's website: https://sphere.computer/software.html.

The KBD/2 layout and schematics are also on Ben's website.

The parts list (BoM) is on the schematic. Double check the BoM with the actual schematic and layout to identify any errors...


Dave
 
Last edited:
Hi guys,

Am I correct in thinking that, if one was to obtain a Sphere CPU/2 board, populate it correctly, and get a copy of the original EPROM code, then all that would be needed is a keyboard and a video monitor?

If so,

1. Are there any difficult to find chips on the board?
2. Is a copy of the original EPROM code available?
3. Was there a special keyboard, or is there documentation on connecting a keyboard?

Thanks vey much in advance for your patience with my naive questions.

smp
You'd need a teletype or terminal I think if just the CPU/2 board. For video (and keyboard?) they came out with the CRT board. That's the one I'm currently struggling with as no one has pictures of the artwork or an unbuilt board..

Ben can correct me but yeah, I think the CPU/2 can run as an SBC with appropriate terminal. I'm not sure if they have the listing for the TTY/SIO ROMs though.
 
The source code listings for the sphere-1 are online as fan-fold listing scans.

All I did was to manually enter the original source code and comments from the scanned listing and re-assemble the source code to ensure I could re-create the correct binary EPROMs.

If all you have is a PROM of some type, all you can do is to disassemble it and then try to recreate the functionality in the form of comments.

Dave
 
The monitor EPROM code assumes a CRT/1 and a KBD/2 are fitted.

I don't think I have seen the listing for the teletype listing.

If the PROM exists somewhere, I can disassemble it...

Dave
 
The monitor EPROM code assumes a CRT/1 and a KBD/2 are fitted.

I don't think I have seen the listing for the teletype listing.

If the PROM exists somewhere, I can disassemble it...

Dave
Looks like the *real* idiot on this forum will have his work cut out for him trying to reverse engineer the CRT/1. 😆

Ulp.
 
The monitor EPROM code assumes a CRT/1 and a KBD/2 are fitted.
Yes, Dave is right on here. The Sphere firmware that has yet been recovered always assumes a Sphere CRT video board is connected, and an ASCII keyboard is plugged into the PIA at socket X4.

There was an alternate form of the firmware that did a bit banging tty interface directly off the CPU board making it into a SBC but despite my best efforts I have not yet found this program. (I do think it would be kind of a fun project to try to create! Fun challenge!) But note that even if you had this, you would not get the PDS environment from it— it was a totally different and much simpler monitor program.

So you do sort of need a CRT board or (another fun challenge) a more modern device that simply listens to the correct addresses and produce output in whatever form is useful (vs building an NTSC signal)

A keyboard can be easily rigged up using an eg mike willegal ps/2 to ascii keyboard converter board with the pins wired for the Sphere layout. https://www.willegal.net/appleii/appleii-kb-int.htm
 
If all you have is a PROM of some type, all you can do is to disassemble it and then try to recreate the functionality in the form of comments.

Thanks @daver2 Once I read your reply it occurred to me that maybe one of the AI tools could disassemble and comment it and it has to some degree. I'll have to play around with it a little. I should have thought of that earlier.
 
1. Are there any difficult to find chips on the board?
Nope. It’s all original 7400 TTL stuff and some Motorola 6800 parts. DRAMs are eight TMS4060 or equiv. All are obsolete and some will be pricier than others to buy but nothing is unobtainable or anything line that. The EPROMs are 1702As which are tricky to program but I can assist there if needed.
 
Last edited:
  • Like
Reactions: smp
At first glance (and they can be deceptive of course) I am just wondering whether we could modify the PDS GETCHAR and PUTCHAR routines to use the TTY ACIA on the SIM/1 card?

I might have a go at that on my simulator.

Some programs may still use the memory buffer of the CRT/1 interface though...

Dave
 
Here's the more or less final artwork of my replica. Still tuning a few things as I get an order organized.

The hardest part for me was the text. It seems like they were using multiple fonts or something, different sizes, etc. I'm sure the etching process probably changed some of them. The taping for traces also seems to be inconsistent widths. I can't be sure if it's just tricks of photography/scanning or not, but it's hard to replicate it exactly in KiCAD.
 

Attachments

  • sphere1-top-042026.png
    sphere1-top-042026.png
    275.8 KB · Views: 11
  • sphere1-bottom-042026.png
    sphere1-bottom-042026.png
    291.5 KB · Views: 11
Here's the more or less final artwork of my replica. Still tuning a few things as I get an order organized.

The hardest part for me was the text. It seems like they were using multiple fonts or something, different sizes, etc. I'm sure the etching process probably changed some of them. The taping for traces also seems to be inconsistent widths. I can't be sure if it's just tricks of photography/scanning or not, but it's hard to replicate it exactly in KiCAD.

Remember that these boards were originally laid out by hand.
I've seen it done, and it is tedious work. The red tape was available to the layout person in different widths for what they were trying to do or avoid doing. I imagine the same would have been for the lettering. The layout person could have been inconsistent with the labels.

smp
 
Cool project!
I'd like to add that if there is any question of under which chip a signal is connected between IC pins, if there are any cases of where it would be ambiguous, you might be able to use a four wire ohm measurement instrument to determine where the connection actually is.
(Aslo although not that likely to result in success, it seems worth a shot to shine light in under an IC and possibly see the traces through the PCB if it's bare fiberglass with no coating/paint).
 
In the hopes that it will be useful both to those attempting replica work as well as anyone trying to restore vintage equipment, I've written up, formalized, and posted a big pile of work notes I have kept while working on my own Sphere.

- Sphere bringup guide incl hints & tips and a way to load programs without cassette hardware
- CPU module info & troubleshooting & repair notes
- CRT/1 module ditto
- Several more articles like this on the documentation page

(I also continue to add material to the software and news posts sections of the site.) HTH!
 
Back
Top