New GUI for the Atari 8-bit

[Matej]

Only 256colors on VBXE??? On VBXE there is also 640x480/64 colors or 320x240/1024 colors mode too???

Cant wait too see it! Will be beautiful!!!

[Prodatron]

Only 256colors on VBXE??? On VBXE there is also 640x480/64 colors or 320x240/1024 colors mode too???

Cant wait too see it! Will be beautiful!!!

Please don't forget that it's a lot of work to add all the different conversion routines for bitmaps with different colour depths.

In SymbOS we have displays with 2, 4 and 16 colours. Bitmap graphics can have 4 or 16 colours. You need the follwing routines:

- display a 4 colour bitmap on a 2 colour display

- display a 4 colour bitmap on a 4 colour display

- display a 4 colour bitmap on a 16 colour display

- display a 16 colour bitmap on a 2 colour display

- display a 16 colour bitmap on a 4 colour display

- display a 16 colour bitmap on a 16 colour display

Now imagine what will happen if you have even more different colour depths?

It would be possible to write a generic routine which could convert everything into everything, but it would be terrible slow, if you don't have some hardware, which will do it "for free".

[Bikerbob]

Does anyone know or any historical articles about GUI creation in 8-bit systems?

 

I know bits and bobs of all of it, and thanks Jon for the SymbOS link, I read a bunch on there as well.. very interesting.

 

I have always been an ATARI man, and because of that had the idea that our little "cult" of users were pushing boundaries wider and further than all those other computers I know nothing (out of my ignorance) about.

 

I knew of Amstrad, but here in Canada, I am not sure if it ever came to sale. Impressive that SymbOS, just as I was not a huge C64 fan, but wow, they had GEOS so early.. and since I was fixated on Atari, I never knew. I am sure there is much and more I know nothing about.

 

its interesting how your perspective limits your view.. just wondered if anyone knew of any more 8-bit histories, that might give me an idea of how the various surviving communities developed.

 

James

[flashjazzcat]

There's some stuff in my blog and this is quite a nice site about GUIs:

 

http://www.guidebookgallery.org/guis/

 

There's not a lot regarding the nuts and bolts of actually coding the things, which is why I had to write everything from scratch (including re-implementing the SymbOS window manager on the basis of a description of the rectangle lists provided by Prodatron). GEM source code is interesting reading, too, as is the story of the original Mac's OS (I forget the site URL). Not all 8-bit stuff, but since there were so few decent 8-bit window managers, we have to look to 16-bit platforms for inspiration.

[Stormbringer]

At the risk of getting yelled at - whats the recent gossip on your GUI?

[+mytek]

I can answer that .

 

Nothing much to say since I stole all of his free time over the last year or so.

 

It all began when I sent him a complete 1088XEL kit in 2017. This opened up the rabbit hole that he then fell down into, and only found his way out a few months ago. If it had been nothing but a hardware build project, he would have gotten free only a couple of months after agreeing to do the online assembly videos. However once he started using the new hardware, and through discussions that he and I had, he embarked on writing a custom version of the BIOS to give it an XEL personality. If this is where it stopped, then only another month or two would have been all that was needed. But unfortunately for Jon I had other ideas brewing that would require a much more extensive use of his unique abilities, and that was the introduction of a built-in CF drive. We went through 3 different versions of hardware to get to the present day XEL-CF3 design. And as the hardware evolved, so did the XEL PBI BIOS and the new XEL Loader that got integrated into the U1MB firmware update. Of course Jon being his ever thorough self not only made changes to the previous versions of the PBI and Loader to suit the XEL-CF3, but went above and beyond by adding in dual CF drive support, while also doing numerous refactorings of the original code. And all the while taking what was learned along the way, and integrated it into the non-XEL U1MB Bios as well. It became a massive undertaking, but in the end I really think it is an accomplishment that Jon should be proud of, and benefits a lot more then 1088XEL users.

 

So lack of work on the GOS really comes down to me as the major source of distraction that put a wrench into that project .

[+MacRorie]

I can answer that .

 

So lack of work on the GOS really comes down to me as the major source of distraction that put a wrench into that project .

 

 

 

Oh, but what a beautifully useful wrench it is!

[flashjazzcat]

I'll add to the above that I started writing new U1MB, SIDE and Incognito firmware in March 2015, shortly after introducing pre-emptive multitasking to the GOS the previous year. The requirenents of the GOS itself were the primary motivating factors behind the new U1MB BIOS, and the loader and PBI BIOS were likewise rewritten. The v.2.0 alt firmware was released in July of this year.

 

Aside from this and the 1088XEL, I was also involved with the Ultimate Cart and UNO Cart. Not to mention the PDM player. That was coding time 100 per cent accounted for during the past three years.

 

I'm currently procrastinating over the massive changes required to the driver model in order to move the GOS to the next phase. I have to completely break it - probably for some weeks - in order to move things forward.

[pixelmischief]

I'll add to the above that I started writing new U1MB, SIDE and Incognito firmware in March 2015, shortly after introducing pre-emptive multitasking to the GOS the previous year. The requirenents of the GOS itself were the primary motivating factors behind the new U1MB BIOS, and the loader and PBI BIOS were likewise rewritten. The v.2.0 alt firmware was released in July of this year.

 

Aside from this and the 1088XEL, I was also involved with the Ultimate Cart and UNO Cart. Not to mention the PDM player. That was coding time 100 per cent accounted for during the past three years.

 

I'm currently procrastinating over the massive changes required to the driver model in order to move the GOS to the next phase. I have to completely break it - probably for some weeks - in order to move things forward.

 

Better minds than mine will choose their directions and make their magic. I would say, however, that new hardware development has so completely eclipsed complimentary software development that I find it hard to imagine what I might do with it all. This might be a good time to give us plain old user types something we can be excited about running on all the new gear.

[Timothy Kline]

There's some stuff in my blog and this is quite a nice site about GUIs:

 

http://www.guidebookgallery.org/guis/

 

There's not a lot regarding the nuts and bolts of actually coding the things, which is why I had to write everything from scratch (including re-implementing the SymbOS window manager on the basis of a description of the rectangle lists provided by Prodatron). GEM source code is interesting reading, too, as is the story of the original Mac's OS (I forget the site URL). Not all 8-bit stuff, but since there were so few decent 8-bit window managers, we have to look to 16-bit platforms for inspiration.

 

"...so few 8-bit window managers..."

 

Which among the rare breed would you consider ahead of the rest, insofar as 8-bit window managers? DiamondGOS? Another one?

 

--Tim

[Guest]

At the risk of getting yelled at - whats the recent gossip on your GUI?

 

I think Eddie Izzard probably put this kind of question into context

 

 

Do it like Microsoft - "lt'll be done, by Saturday, Tuesday, next week. - We'll bring it out when we're fucking ready, right?"

 

 

 

 

Edited September 25, 2018 by Guest

[Stormbringer]

 

I think Eddie Izzard probably put this kind of question into context

 

 

Do it like Microsoft - "lt'll be done, by Saturday, Tuesday, next week. - We'll bring it out when we're fucking ready, right?"

 

 

 

 

 

and of course there's always this response.... I know it will be DONE when its ready, I just wanted to hear if there was any news. its been a year and I think that sufficient time for a nudge.

[Kernal]

SOON!!! and am excited about this beyond words.. i cant wait to get the gui running 100% this will revolutionize how we use our ataris.. and make work flow sooo much easier.. kudos to flash and his motley crew.. i have an 800xl he modified with ultimate1mb and stereo pokey and side2

[flashjazzcat]

Which among the rare breed would you consider ahead of the rest, insofar as 8-bit window managers? DiamondGOS? Another one?

This is a great question. I've seen nothing else which touches Diamond GOS, although its window manager makes no attempt to optimise redraws based on obscured sections of underlying windows, etc. Everything gets redrawn from the back to the front of the scene every time you move, resize or close a window. But you can still move windows (albeit on byte boundaries), resize them and use scroll bars, which quite easily beats all other implementations on the Atari, which usually draw some immobile representation of a window on the screen and take a ponderous amount of time to fill it with content. Of course, having thrown several years at creating something more efficient, I think Diamond's window manager now seems rather slow and clunky, but Diamond was a commercial product coded within time restrictions, and at the time Diamond was written, even some 16-bit GUIs weren't all that great. Respect for Diamond should always be assumed.

 

On other 8-bit platforms, SymbOS is easily miles ahead of anything else. The Atari GOS window manager works in much the same way as the SymbOS manager thanks to SymbOS's author convincing me (in 2014 or so) to drop the masks I was using (something akin to 'Regions' on the early Mac operating systems) and replace them with a more traditional rectangle based window manager such as one will find in GEM, etc. This method involves recursively splitting windows into the smallest possible whole rectangles and then deciding what's visible and what needs to be redrawn, and resulted in the Atari window manager gaining a 100 per cent or so speed increase over the earlier versions.

 

GEOS is of course somewhat better presented than Diamond (proportional, scalable fonts, etc) but the windows don't move and it is extremely slow. Later customised versions of GEOS gained movable windows and other improvements.

 

But the combination of responsive, good-looking UI and multitasking in SymbOS is unbeatable in the 8-bit realm and as a consequence SymbOS is the benchmark by which I judge what has so far been accomplished on the A8.

 

and of course there's always this response.... I know it will be DONE when its ready, I just wanted to hear if there was any news. its been a year and I think that sufficient time for a nudge.

 

The reason there's always this response is probably because periods of inactivity always have the exact same cause (namely: I was busy with other things), and as soon as there is news, it will be reported here in this thread by me.

 

Of course, I appreciate any encouragement and the knowledge that people still want to see the project completed is just about the sole incentive for actually completing it.

 

Better minds than mine will choose their directions and make their magic. I would say, however, that new hardware development has so completely eclipsed complimentary software development that I find it hard to imagine what I might do with it all. This might be a good time to give us plain old user types something we can be excited about running on all the new gear.

I agree, and that's a really good point. Since the mid-Nineties all I have done with the Atari is write productivity software which I have personally rarely ended up using for said productivity purpose. Instead I just write more software.

 

The constant worry is that the entire 8-bit scene simply fades before such enormous projects can reach completion, and when one is working on something for so long on a hobbyist basis, one has to contend with everything happening in one's life, and that includes changing interests and priorities.

Edited September 25, 2018 by flashjazzcat

[Timothy Kline]

This is a great question. I've seen nothing else which touches Diamond GOS, although its window manager makes no attempt to optimise redraws based on obscured sections of underlying windows, etc. Everything gets redrawn from the back to the front of the scene every time you move, resize or close a window. But you can still move windows (albeit on byte boundaries), resize them and use scroll bars, which quite easily beats all other implementations on the Atari, which usually draw some immobile representation of a window on the screen and take a ponderous amount of time to fill it with content. Of course, having thrown several years at creating something more efficient, I think Diamond's window manager now seems rather slow and clunky, but Diamond was a commercial product coded within time restrictions, and at the time Diamond was written, even some 16-bit GUIs weren't all that great. Respect for Diamond should always be assumed.

 

On other 8-bit platforms, SymbOS is easily miles ahead of anything else. The Atari GOS window manager works in much the same way as the SymbOS manager thanks to SymbOS's author convincing me (in 2014 or so) to drop the masks I was using (something akin to 'Regions' on the early Mac operating systems) and replace them with a more traditional rectangle based window manager such as one will find in GEM, etc. This method involves recursively splitting windows into the smallest possible whole rectangles and then deciding what's visible and what needs to be redrawn, and resulted in the Atari window manager gaining a 100 per cent or so speed increase over the earlier versions.

 

GEOS is of course somewhat better presented than Diamond (proportional, scalable fonts, etc) but the windows don't move and it is extremely slow. Later customised versions of GEOS gained movable windows and other improvements.

 

But the combination of responsive, good-looking UI and multitasking in SymbOS is unbeatable in the 8-bit realm and as a consequence SymbOS is the benchmark by which I judge what has so far been accomplished on the A8.

 

Thank you for the response, I appreciate it!

 

And keep up the good!

 

--Tim

[+DrVenkman]

I'll add to the above that I started writing new U1MB, SIDE and Incognito firmware in March 2015, shortly after introducing pre-emptive multitasking to the GOS the previous year. The requirenents of the GOS itself were the primary motivating factors behind the new U1MB BIOS, and the loader and PBI BIOS were likewise rewritten. The v.2.0 alt firmware was released in July of this year.

 

Aside from this and the 1088XEL, I was also involved with the Ultimate Cart and UNO Cart. Not to mention the PDM player. That was coding time 100 per cent accounted for during the past three years.

 

I'm currently procrastinating over the massive changes required to the driver model in order to move the GOS to the next phase. I have to completely break it - probably for some weeks - in order to move things forward.

 

Well, your work is wonderful and even if the GOS as it exists today isn't very useful, it shows what is possible with the right man moving the bits around and pushing the pixels:

 

 

(Recorded via an S-video capture dongle into my Win10 laptop running OBS).

Edited September 27, 2018 by DrVenkman

[Gunstar]

 

The constant worry is that the entire 8-bit scene simply fades before such enormous projects can reach completion, and when one is working on something for so long on a hobbyist basis, one has to contend with everything happening in one's life, and that includes changing interests and priorities.

I've seen the retro 8-bit community come to life pretty much with the advent of Atari Age, and like 18 years later, I think it is now safe to say that the community will be around until all the current community dies off, I plan on that being another 20-30 years for myself, God willing.

 

I think most here would agree, for their part. The community we have today is far stronger, and lasted longer than the original life span of the computer itself and far more active, in a more contributive way I think than it ever was as a commercial concern. So I really don't think the community fading away is an issue for another decade or two, but most of us would like to see this your GUI with some great apps, be in our hands to use before we go senile.

[+mytek]

...be in our hands to use before we go senile.

 

Based on that criteria, it better happen within the next 5-10 years. Otherwise I'll forget what it is that I'm suppose to be waiting for .

 

Good thing Jon's younger than I.

[Gunstar]

 

Based on that criteria, it better happen within the next 5-10 years. Otherwise I'll forget what it is that I'm suppose to be waiting for .

 

Good thing Jon's younger than I.

I'm getting curious as to the average age here on AA 8-bit forum, what the percentage is above and below the 50 mark, etc. I just turned 50 myself last month...man time flies...17 when I got my first Atari 8-bit, 27 when I got my Jaguar...37 when I got my first Falcon and Amiga 2000...

Edited September 27, 2018 by Gunstar

[Goochman]

We're not that far apart - I got you by 3 months.

[_The Doctor__]

It' already been 9 or so years. To be honest, I forget so often now, it might be new every time I see it! Although I might be in the cold cold ground as well... I forget which they are betting on happens first. Such as it is in life... and death. It's nice to know it will happen though... a pleasant thought to take with you. It will hard to get it delivered to my address at that point.

Edited September 27, 2018 by _The Doctor__

[Kyle22]

Don'

 

It' already been 9 or so years. To be honest, I forget so often now, it might be new every time I see it! Although I might be in the cold cold ground as well... I forget which they are betting on happens first. Such as it is in life... and death. It's nice to know it will happen though... a pleasant thought to take with you. It will hard to get it delivered to my address at that point.

 

Don't be morbid.

[+mytek]

I got my first Atari (400) when I was 25, and lost track of what came next or when, but in a few more months I'll be 63.

[flashjazzcat]

IMHO, the GOS didn't get really interesting until the 2014 preemptive multi-tasking re-write and the arrival of Jorn Mika (SymbOS) in an advisory capacity, and given the amount of stuff I've churned out since then, I'm relatively unapologetic about time-frames. The GUI was initially supposed to be a UI library for applications (begun properly in January 2011), and the next three years just saw a staggering amount of feature-creep until it became a complete multi-tasking graphical OS. Window manager was completely re-written three times, etc.

 

I don't know about users going senile, but my mental stability probably remains in real jeopardy for the entire tenure of this and other projects.

[_The Doctor__]

I just slobber and pet the cat when it gets too bad.. be better if I had a dog again... he can go fetch whatever I forgot I needed. I will call him dog. That way I won't have to worry about the name

Edited September 27, 2018 by _The Doctor__