ST Basic - inspired by QDOS?

[janbredenbeek]

IMO none were needed. For me TAlent's Assembler Workbench was simply magical. I never really looked back after that.

Most of my projects were assembler-based too (actually my first 68K project was a monitor/disassembler!). But there are lots of cases where BASIC would be a better choice for rapid development. Programs involving a lot of FP calculations (I've written some astronomical programs too) or string manipulations come to mind.

[NormanDunbar]

Given that you need SuperBASIC to EXEC any of the code you assembled using Talent Workbench, I would say that without SuperBASIC, everything you wrote was completely up shit creek.

No matter how brilliant the code you wrote, it was nothing more that s file of bits without SuperBASIC.

As for any good applications written in SuperBASIC, off the top of my head:

Turbo
Supercharge
QLiberator
The Editor
Sport
Almost anything sold by Digital Precision
WinBack
Gopher
Almost everything written by Dilwyn Jones
Super Disk Labeller
Printer Master
ARSE (ARchive Syntax Evaluator)
Adventure Creation Tool (ACT) -- probably.
...

I could go on. Oh yes, I lied about ARSE, that was written in C68.

So which commercial applications did you have on the market that we might have heard of?

Bear in mind, your idea of a useful application might not be what other people think of as useful. As an analogy, lots of people like drinking beer and watching football. I would rather drink piss and watch paint dry than do either. And quite honestly, I'd really rather not!

Cheers,
Norm.

[XorA]

Probably time to put the troll back under the bridge. It’s clear it has little knowledge of computers with the statement

Basic is today nowhere to be found

[Pr0f]

All your comparisons seem to compare what's possible and understood now, with what was largely accepted as the way to do things back in the early 80's.

Yes bootable systems for 68K were being made back then - but these OS's tended to be pricey to license - and we know Sir Clive was all about keeping costs down.

If you are going to make comparisons - I would stick to the same era - as comparing capabilities of machines some 40 years apart is frankly ludicrous.

[Derek_Stewart]

Given that you need SuperBASIC to EXEC any of the code you assembled using Talent Workbench, I would say that without SuperBASIC, everything you wrote was completely up shit creek.

THis is why Linux has elf - "eXecutable and Linkable File" format.
It can do FAR more than just load a bunch of bytes in memory and execute them.

You don't need a friggin BASIC interpreter to do that. Sinclair could have gone for simplified version of that.
One possible option:
After reset, make QL boot from any media where it could boot from, be it SER port, NET. MDV or any other device it could reach at boot.

With SER and NET, boot stream could just be a bunch of bytes with starting address, length and checksums.
With MDV, it could be sector 0 and subsequent sectors, if they are marked for boot.

With MDV, such a machine could always fully boot reliably under 15 seconds.
With NET, it would probably take less than 7.
WIth floppy, it would be less than 3.

That solution would be more than enough.

BTW, once they introduced SPC into machine, they could have omitted ROMs completely.
Starting few bytes could be served simply by 8049.

To dot the "i", QL would get Vector base in RAM, so all the vectors would be free to redirect and not poured over with concrete, like in
actual machine.

Case in point - PC. It loads the first sector on first disk it can find and executes it.

Look where PC DOS started and where it ended. Latest DOS 5.xx are MASSIVELY better than DOS 1.x.
Even discarding the difference the hardware made, pure SW evolution made many things possible.

It's even more pronounced with Windows. What would happen with them if they were baked in ROM ?
Win95 look and work massively better than Win1.0.
AND WIn10 is another universe WRT Win95.
Whole sequence was made possible by having it media that can be simply changed.

No one really cares how fast does PC boot. That problem has been irrelevant and even in that part it was solved long ago.

Hi,

This sounds great, can you post some example source code to achieve this.

I also used The Talent Workbench, but did not like the IDE based system.

Jan Bredenbeek's Monitor programme was much better than anything commerical on the market, coupled with the QMAC macro assembler, made the QL a better assembler development machine.

I think ST Basic was out before Superbasic, Hisoft wrote basic system for the ST, after the QL was launched.

As I have mentioned earlier, QDOS Superbasic is not good, it took Qjump to write the Super Toolkit 2 to enhance it, also Superbasic does not multitask, it took Minerva to introduce multitasking Superbasic without any overhead on the operating system.

But SBASIC in SSQ/E is by far the best new QL BASIC system available.

[QLvsJAGUAR]

Hi folks,

out of my summer hibernation, currently was having my first vacation this year, where on one or the other occasion I was getting a glimpse on one or the other thread using my smartphone.

The QL is a mid-1982 to end-1983 design (codename ZX-83) of a rather small company called Sinclair Research Ltd. where a bunch of young, very talented and passionate humans tried their very best to develop the best affordable micro at the time; while at the very same time they and their colleagues had to complete the promises the company did for the ZX Spectrum, such as the Interface 1, ZX Microdrive, Interface 2. Humans are individuals and tend to make mistakes. Even considering all of this, and believe me, I know all of the QL flaws and why they happened, the QL is not a bad drop.

When judging the QL, we must compare apples with apples and stay at the time given (1982/1983).

@Brane2: Branko, what are your posts contributing to the QL community? Anything useful, anything the folks here can solder/stick together and use with their QL systems, any piece of software, even a tiny SuperBASIC proggy, any valid and backed explanation of the reality, anything?

Relating this thread, yes Metacomco had an eye on the QL’s SuperBASIC while developing ST BASIC for ATARI Corporation and yes, even they did BASIC interpreters before ST BASIC and knew the QL and its SuperBASIC quite well, they failed to create something big, something flexible and something expandable as the QL’s SuperBASIC. Even bundled with the ATARI ST computers, ST BASIC was gradually superseded by much more powerful BASIC systems such as GFA BASIC. Sure, time pressure from ATARI and an unfinished computer at the time (the ST, its TOS/GEM being far from complete/documented at the time Metacomco had to do the job) is a good excuse. Comparing the ATARI ST BASIC timeline and story, Ian Jones did a great job with her SuperBASIC for the ZX-83, being assisted and coached by her boss Tony Tebby who was heading OS (GST’s 68K/OS and his Domesdos aka QDOS) and Firmware (mainly done by Aaron Turner) with some work done by Jonathan Oakley. In the end it all is judged by Time-To-Market and Market Acceptance (both Sinclair, ATARI and COMMODORE presented their machines and had enormous pressure to ship) and hence we got SuperBASIC as is with ROM Version PM in March 1984. Things remaining uncompleted on the To-Do-List were among others, the full screen BASIC Program Editor, the Error Trapping and Trace features and INTeger support in FOR loops. Tyche, its offspring Minerva, Toolkit 2 and SMSQ are the evidence what the very same humans were able to complete in one way or another after missing TTM and Market Acceptance.

Stay safe, healthy and positive!

Cheers, Urs

[bwinkel67]

...Tony Tebby who was heading OS (GST’s 68K/OS and his Domesdos aka QDOS) and Firmware (mainly done by Aaron Turner) with some work done by Jonathan Oakley.

Thank you for the context. I really love the history behind QDOS. I was reading a series of articles (I think it was QL World) about Domesdos and how it came to be. Kind of neat how Sinclair Research was forward thinking in their OS design and wanted to create a workable UNIX type OS and that's what QDOS became. I think Microsoft even tried to sell them an OS and luckily Sinclair declined.

[bwinkel67]

I would hope they bring in additional POV that help filtering out some misconceptions and bring in some fresh insight.
Constantly agreeing with every idea just to feel "safe" and "within one's group" might feel cosy at first, but very soon everything starts smelling stale.
Good ideas have to be constantly challenged in order to be sorted and filtered as good ideas, other wise they become just myths.

But your whole premise doesn't make any sense. This is 35+ year old technology. This forum exists for us to talk about the things we liked about it. It is there for us to not necessarily constantly agree, but for us to "reminisce about those ideas so we can feel 'joy' for its nostalgia" not 'safe' as you put it. I don't think there is interest in creating a modern QL, fixing perceived design flaws, and selling to a mass market. The closest that's happened in the Sinclair retro community has been the Spectrum Next and that was literally there to recreate modern hardware of the old platform, with an extensible environment that mimicked the one from 40 years ago. One could argue the Q68 is similar to that, sans a modern case.

This forum is for folks that got attached to the old beast, share their experiences, ask questions of each other, and find ways to augment that system to keep it going (either with emulation or through hardware projects like emulated MDV's, processor upgrades, FPGA implementations, etc). But to argue how the base design was bad and what should have been done doesn't add anything to the conversation (it's not a POV that is helpful). If this were a forum on design and we were focusing on bad old designs then your points could be very enlightening but instead you are beating a dead horse here, one that many of us grew up with fondly.

Why not focus less on what you didn't like about the QL, what you believe Sinclair got wrong, etc, and instead focus on how you can move things along. You had a thread of wanting to create a replacement ULA for the ZX8301/2...that's great, work on that. But badgering people about why you think SuperBASIC sucks (or as you've done in other threads how Sinclair shouldn't have used a keyboard case, or they shouldn't have done this or that) helps no one and it doesn't really qualify as fresh insight. In fact, it qualifies as the definition of trolling. I feel like I'm sitting here watching someone argue about how my favorite sports team lost the title match 35 years ago and that they should've used this player instead of that player, etc....it's 35 years ago, let it go.

A fresh insight would be "hey, how can we make QDOS multi-user" or "can we use the QLNet port to create a cheapo sound card through software" (something I'm looking into) or "should we have a smaller cheaper fpga-based QL for projects" (QZero that Peter is working on) or "should there be a PI Storm for the QL" which had some pros and cons...those are examples of relevant discussions because they focused on the present, not on the past.

[Pr0f]

But where are you going with this thread brane2 ? It was after all originally posted to ask the question was ST BASIC inspired by QDOS? I think somewhere in among all the unrelated stuff about SuperBASIC and now QL hardware, it may have actually been answered.

You had threads going to discuss some hardware ideas - if that's what you want to post on - put your effort there - you effectively spamming this thread to it's eventual death.

[pjw]

...Tony Tebby who was heading OS (GST’s 68K/OS and his Domesdos aka QDOS) and Firmware (mainly done by Aaron Turner) with some work done by Jonathan Oakley.

Thank you for the context. I really love the history behind QDOS. I was reading a series of articles (I think it was QL World) about Domesdos and how it came to be. Kind of neat how Sinclair Research was forward thinking in their OS design and wanted to create a workable UNIX type OS and that's what QDOS became. I think Microsoft even tried to sell them an OS and luckily Sinclair declined.

You werent by any chance thinking of these articles from QL Toady TTos.zip? (The link is just to the download page.) A great read! From those it seems more like Tony Tebby was trying to escape from Unix, MSDOS, and the rest, rather than make a better Unix.