Book(s) will probably need to be added for subject areas that aren't covered by the SDK but don't "fit" into an existing book. For example, the FEFS filesystem format and various file formats. Some of these are covered in the Psionics Files, so they could be included in their own book.

I'm pretty settled on #AsciiDoc, which will make it easy to generate HTML, PDF and EPUB versions.

Original versions of the documentation will be made available, of course.

#Psion #RetroComputing

So. SIBO/EPOC16 docs.

My current feeling (subject to change, of course) is to use the C SDK structure as a basis. Keep subject areas in "books" - PLIB manual, Hardware manual, HWIM manual, etc.

Changes will be:

- A style guide for consistency across the docs
- Merge the HDK with the SDK's Hardware Manual (they share chapters anyway)
- Info about mx machines added
- New "book" for the OPL Programming Guide
- Updates for discovered info
- Updates for modern use

#retrocomputing #Psion

Of course, if you want truly era-appropriate, genuine #Psion artwork, look no further than these gems.

#RetroComputing #Psion3 #Art

Also, I do have a paper copy of the #Psion Series 3a Programming Manual. It's been very helpful with all the bold, italics, and combined bold-italics.

I know it's meant to be a beginner's guide, so it's written in a friendlier manner than the SDK. All well and good. But it really could've done with an editor to neaten things up.

#RetroComputing

While #Psion in the early 90s seems to have used the same Word template for writing their public #documentation, there doesn't seem to have been a consistent "style guide."

Different parts of the SDK have clearly been written by different people. The HDK is not as polished. The 3a Programming Manual is different again, both in voice and in its heavy use of bold and italics.

As I go through the documents converting them to AsciiDoc, I wonder how much I should "tidy" the text.

#DigiPres

Now that the TopSpeed TechKit with its #assembler is out in the world, I'm considering porting the #TASM and #MASM code in one or two of the open source #EPOC16 apps over to TSASM.

Hardly an urgent project, but it would mean that the apps could be built with one #compiler ecosystem with a single project file.

Of course, the code would still be restricted to the #16bit world, but it's no worse off than we already are in #Psion land.

#DOSBox #RetroDev #RetroComputing

Earlier today, after some hunting, I found the text version of the #OPL Programming Manual that came with the Series 3a. So, while I was sitting in front of the TV this evening, I thought I'd start converting it to #AsciiDoc.

First chapter is pretty much done.

I think (hope) I've got a paper copy in a box somewhere, so I can compare it to the original.

#Psion #RetroComputing #Documentation

Reading through the first manual in the SDK has reminded me what the "pure small memory model" is.

The ES register is never corrupted (DS=ES=SS).

#TopSpeed C implemented this, which was one of the main reasons why #Psion used it for the SDK.

I've heard that it's possible to implement this with #Borland C (and maybe #Watcom?), but I don't know how.

#16bit #retrocomputing #epoc16 #x86

Reading through the first manual in the SDK has reminded me what the "pure small memory model" is.

The ES register is never corrupted.

#TopSpeed C implemented this, which was one of the main reasons why #Psion used it for the SDK.

I've heard that it's possible to implement this with #Borland C (and maybe #Watcom?), but I don't know how.

#16bit #retrocomputing #epoc16

This is a #Psion hill that I will die on.

The OS was never called SIBO.

SIBO was the name of the hardware platform. The OS was called EPOC (or Epoc/Os, or similar). I've seen internal docs going back to 1988; they stick with this basically from the start.

The distinction is worth remembering, because there is one SIBO machine that didn't run EPOC16. The MC600 ran DOS.

I use the EPOC16 retronym because it makes it distinct from EPOC32, which is really a totally different OS.

#retrocomputing

HIVE MIND: Can you give me some examples of good #retrocomputing documentation websites? Or any doc websites about an entire computing platform.

I'm looking for examples on which to model the #Psion #Documentation Project (https://doc.psion.info/). I want to give it a proper clean-up, formalise the structure somehow, and have some guidelines.

AFAIK this is the most complete version of the #Psion SIBO C SDK.

It includes multiple versions, the full TopSpeed C including the TechKit, all the documentation I can find, plus some handy config files for you to get started in @DOSBox_Staging and with VS Code.

There is evidence online that, by 2012, Psion were fine with the SDK and HDK being distributed.

Have at it!

https://archive.org/details/psion-sibo-c-sdk

#retrocomputing #retrodev

I've never had a proper look at this folder before. It's like some sort of parallel version of the SDK's include folder.

Some things seem newer, some older. Version numbers and file authors don't tie up. Some parts are broken up into separate files.

The folder conforms better to #Psion C standards. It includes a file called p_msdos.h, like you could use PLIB on something that isn't #EPOC16.

Is this the missing link to the long lost v1.00 of the SDK, designed for the MC range?

#retrocomputing

I don't think there was a newer version of the *software* part of the SDK. 2.20 is the latest one I've heard about and it's the latest version I have.

However, the documentation part of the SDK definitely goes up to 2.30, so it's possible that there was a newer release.

If anyone has something newer (for EPOC16, not EPOC32), drop me a line!

#retrocomputing #psion #16bit

I'll also be making a "2.21" with a few fixes, such as checking for includes properly.

I'm going to boldly say that my 2.21 version is the one you should use, but of course I'll distribute the older versions. I've definitely had projects that will compile with 2.00 but not with 2.20. I think these projects are more likely to compile with 2.21.

#Psion #epoc16 #psion3 #retrodev #retrocomputing

Spot the difference.

Three releases of the #Psion SIBO C SDK: 2.00, 2.10, and 2.20 (mostly).

The copy of 2.00 I was given is missing a few folders of examples, but the libraries and tools are all there. Helpful for some older projects.

I've never had to use 2.10, but it's there just in case.

The copy of 2.20 doing the rounds includes the 2.10 version of CLIB, but CLIB is only there to help with porting apps to EPOC16. New C projects should use PLIB or the OO libraries.

#retrocomputing

Oh, FFS, how long have I had this set incorrectly?! 🤦‍♂️

I'd guess it's not actually that important, then! 😂

#Psion #DOSBox #RetroComputing

Spending far too long wondering/worrying about the repercussions of moving the SIBOSDK folder from C: to D:, so that switching between SDK versions can be done with a MOUNT command in DOSBox.

Make a decision. Make a decision. MAKE A DECISION!

#Psion #retrocomputing

OK, I've worked out which of my "topspeed" archives is TopSpeed C 3.10 - AFAIK the final release for #DOS, and the one recommended for the #Psion SIBO C SDK. I've also found PsionTeklogix's instructions on how to install TSC and the SDK.

I got this all running years ago for myself, of course. But I'm preparing to write the instructions on how to set up the SDK, so I really need the ability to make a clean build. And I want to test it with multiple SDK versions, just in case weird things happen.

@hpower holymoly, einen #Psion hatte ich mir damals nicht geleistet. Heutige Smartphones sind günstiger, haben 10000 mal mehr Speicher und sind Welten Leistungsfähiger. Damals hätte ich gerne so ein Teil gehabt, aber die junge Familie hat das nicht erlaubt.
@Mina @yvesmoret @johnmacintosh