The Falcon's confusing history

[AdamK]

GEM was not that great. Maybe in 1985, yes, but world moved on and GEM froze. I think DRI didn't want to invest in it after they lost Apple suit, and Atari did not have manpower to push it forward.

but how than you explain one or three man behind MagiC or Geneva? Were they Supermans?

Probably But there was no real improvements, and nor MagiC or Geneva implement everything that GEM (and later MiNT - I'm talking about before Atari let it go) included.

No not really, by any count. ST line was introduced in 1985 and officially ended in 1993. That is 8 years tops, but really, after 91 (92 tops) ST line was dead in the water, and Falcon did nothing to help it.

But there was lot of life in ST even past official dismiss... ST was truly remarkable piece of equipment!

No, there was not. Your view is skewed. Almost all commercial software development for ST stopped. Most of the users left the platform. No mass produced new hardware since.

[AdamK]

but how than you explain one or three man behind MagiC or Geneva? Were they Supermans?

Probably But there was no real improvements, and nor MagiC or Geneva implement everything that GEM (and later MiNT - I'm talking about before Atari let it go) included.

But there was lot of life in ST even past official dismiss... ST was truly remarkable piece of equipment!

No, there was not. Your view is skewed. Almost all commercial software development for ST stopped. Most of the users left the platform. No mass produced new hardware since.

[calimero]

No, there was not. Your view is skewed. Almost all commercial software development for ST stopped. Most of the users left the platform. No mass produced new hardware since.

You can go at http://milan.kovac.cc/atari/software/ and type e.g. "1994" in search and you will see what software were released in 1994. (I did not finish with cataloging but...)
German C't magazine keep writing about Atari well beyond 1993.
Yes, there was no mass produced hardware since there was never true ST successor.
It is shame that Falcon (or Sparrow) did not come to market sooner with lower price and that true Falcon did not at all come to market.
I really would like to know when these project starts (Sparrow, Texas Falcon) and how they go... Is it possible that project start in December 1991 and that they have finished PCB with chips until 1993?

---
John Horton is mentioned by Dad Hacker in his blog: http://www.dadhacker.com/blog/?p=1383

"To run Unix effectively we needed some hardware that was very fast, that was simple enough to put into a minor spin of the ST’s memory controller with little project risk, and that would still provide some kind of memory relocation and protection. The ability to have separate address spaces to isolate processes would be good, too.

“If you can come up with something that takes about a gate delay, I’ll put it in,” said John, the memory controller guy. He seemed dubious, but willing to listen.

I went for a bunch of walks.

...

John Horton, the memory chip guy, actually did this hardware for the Mega ST; I don’t know if it’s documented, or if he had to sneak it in or not. I do know that it was not used for Unix in my time at Atari; "...

[mfro]

... GEM however, is a truely elegant OS, and it is a shame that Atari did not continue to develop ... Gary died, but the system he created is truely a wonderful, lightweight, fast OS that is modular, suitable for platform change, and tight...

Would strongly disagree. Don't get me wrong, I personally like GEM (mainly for it's simplicity), but it's certainly not a single person's (at least not Gary Kildall's) masterpiece.

1. GEM is not the operating system, it's a graphics kernel
2. GEM is a more or less verbatim, stripped down (but otherwise mostly 1:1) copy of the earlier GKS (ISO/IEC 7942) standard, suitable for small machines
3. DR's merits go merely to getting rid of floating point operations in the kernel and to rip it out of scientist's ivory towers making it cheaply available

Some details in DR's VDI implementation/documentation even led me to believe they didn't fully understand the concepts of GKS ...

[AdamK]

No, there was not. Your view is skewed. Almost all commercial software development for ST stopped. Most of the users left the platform. No mass produced new hardware since.

You can go at http://milan.kovac.cc/atari/software/ and type e.g. "1994" in search and you will see what software were released in 1994. (I did not finish with cataloging but...)

So, currently about 30 titles, most of which are updates, after that most of them are shareware single person (devout developers) programs. That does not market made

Magazines strugling to remain afloat are not good example of a thriving market

[calimero]

No, there was not. Your view is skewed. Almost all commercial software development for ST stopped. Most of the users left the platform. No mass produced new hardware since.

You can go at http://milan.kovac.cc/atari/software/ and type e.g. "1994" in search and you will see what software were released in 1994. (I did not finish with cataloging but...)

So, currently about 30 titles, most of which are updates, after that most of them are shareware single person (devout developers) programs. That does not market made

yes, yes... it is (or will be): black on white what software come in late years as soon as I finish it but you can see that Maxon, ASH and others publish new software for Atari.

Magazines strugling to remain afloat are not good example of a thriving market

I am sure that you have not idea what is c't magazina
it is not certainly (or ever was) "magazines strugling to remain afloat".

[calimero]

I just stumble, on Atari HQ CD, what happened with Atari Falcon software (?) modem (file PSL_MEMO.TXT):

Code: Select all

"Panic Systems Ltd.
Dacom House
Presley Way
Crown Hill
MK8 0EF

Attn. Graham Adams, John Waddel

March 28, 1994


I have received the hardware prototype of the Falcon based modem which I 
initially became involved with in October of 1992.  Obviously, your contracted 
milestones of September and October of 1992 were not met, but we felt a 
reasonable amount of time afterward would be acceptable given such an 
important project. 

Because we believe that it was a low priority with PSL, since receiving it in 
mid-February I haven't had an opportunity to attach it to a Falcon to test its 
function. As we have not received much in terms of progress, the testing of 
this project has now been given  a low priority at Atari because its viability 
as a marketable product has been eliminated due to the lateness of its 
development.  When I became involved on a weekly basis in May of 1993, I 
brought a willingness on Atari's part to give technical support where we 
could.  We hoped this would accelerate the development in order to assist in 
the  acceptance of the Falcon as a fully featured proprietary platform.  
Indeed, the marketing campaign of the Falcon had been partially built on the 
concept of a software based modem and the lack of that product has been not 
only frustrating, but one of the tangible factors contributing to of the  
failure of the Falcon in the computer market.

So now it is more than a year since I have been involved with this project and 
I can buy a 14.4k modem for a little more than $100.  Therefore, at this time 
we see no reason to continue with the development of this project.  If there 
is any interest from any parties to pick up the project, I will let you know.

Regards,

J.Patton 
Atari Corporation

Previous:

Code: Select all

FAX TRANSMISSION

To:	Piers Uso Walter - i˙link	Atari Corporation
	Fax: 49 30 781 70 56	1196 Borregas Avenue
		Sunnyvale, California 
94089
Date:	5-5-93	(408) 745 - 2000

Subject:	NeXTWorld	Number
		of Pages
From:	J. Patton - Atari SUVL	including this one
	Fax: 408 745 2088
	Tel. 408 745 2135
		

Hi,

I wanted to see if you would be able to stop by Atari just before or 
after NeXTWorld so we could talk about some possible projects involving 
the Falcon.  

One of the projects we would like to talk about is a software modem. 
Royalties and hardware cost would have to be relatively small since we 
would like to sell this for about $100 US.

Please let me know at your earliest convenience when we could schedule 
something while you are in the Bay area.

Thanks,

J.Patton - Atari Corporation

[penguin]

Atari computers were purloined for industry (I got my TT from a knitting factory, where it ran an industrial machine), in the automotive industry they ran the show.

There was special ST in Germany in industrial case special for CNC and other purposes! (I always forget model name... :/)

Industrial STs were the first to include a 020 CPU and a mobile version even arrived before the Stacy. The most well known was the 190ST:

http://stcarchiv.de/stm1991/06/190st-020

Here's another device made by Rhothron (I should replace the images with better scans...):

http://stcarchiv.de/stm1990/08/high-tech

Both Rhothron and IBP were big players in the "industrial ST" market. There were other manufacturers as well, in fact, you could even build your own industrial ST, the Elrad ST (named after the German electronics magazine Elrad). The instructions were published in issues 6 through 9/92, with add-ons described in later issues (http://www.stcarchiv.de/tos1992/07/news).

Other custom hardware included a special keyboard used by publishing house Burda to help with the layout of their magazines. Germany's oldest women's magazine Brigitte was layouted on the TT.

ms1190.png

[Zarchos]

Very interesting thread.

Last week I received some Computer Shopper magazines, and there is an interview of an Atari representative expressing the idea Atari is probably going to use a powerful RISC chip for their next machines.
Any infos about that ?
I can find the issue and scan the article if you are interested.

[calimero]

@penguin great infos! Very pleasing to see ST in such variaty of applications.

Regarding RISC: only RISC like CPU in Atari product was Tom and Jerry in Jaguar.
Transputers (ABAQ and ATW) was... Transputers

[Zarchos]

@penguin great infos! Very pleasing to see ST in such variaty of applications.

Regarding RISC: only RISC like CPU in Atari product was Tom and Jerry in Jaguar.
Transputers (ABAQ and ATW) was... Transputers

OK so you mean they never thought of using a MIPS chip for example, or an Intel RISC chip whose name I forgot.(i960 ?)
Strange then, from Atari, to make such annoucement.
IIRC the representative was talking about a RISC chip as the main CPU for the next generation of Atari computers.
I won't say the Archie has impressed them (honestly I doubt it), but the guy was from Atari U.K., and this RISC thing was very commonly praised in the U.K press (from what I can read in these mags from early 90ies).

[Zarchos]

It is in French, but use Google translate.
About some French programmers, the Jaguar and the Falcon, and the name 'Sparrow' appears ...
I knew I had read it many eons ago, glad I found it back.
http://www.obsolete-tears.com/anecdotes ... er-67.html

[ryan]

1. GEM is not the operating system, it's a graphics kernel
2. GEM is a more or less verbatim, stripped down (but otherwise mostly 1:1) copy of the earlier GKS (ISO/IEC 7942) standard, suitable for small machines
3. DR's merits go merely to getting rid of floating point operations in the kernel and to rip it out of scientist's ivory towers making it cheaply available

Some details in DR's VDI implementation/documentation even led me to believe they didn't fully understand the concepts of GKS ...

Sorry, but I've read the source for both the VDI and the DRI GSX implementation of GKS and what you say is not really accurate.
Firstly it is VDI which is based on the GKS concepts, not GEM as a whole.
Secondly even VDI was not a copy but a reimplementation of similar concepts.
Thirdly, the AES itself is a quite well designed API (for its time). But it was clearly done by a different team or at a different time than the VDI. To me the real genius of GEM is the AES, not the VDI. The VDI itself is full of leaky abstractions and inconsistencies. And many of the problems of GEM program are related to the poor interaction between the three layers of the AES, the VDI, and the underlying DOS.
AES itself duplicates a lot of abstractions that the underlying DOS has (process IDs and task lists, etc.).

I think in the AES itself is the kernel of something very nice, if only they had the manpower time and mandate to make an entire OS from it, instead of pasting it over CP/M/GEMDOS/MS-DOS.

[mfro]

Sorry, but I've read the source for both the VDI and the DRI GSX implementation of GKS and what you say is not really accurate.
Firstly it is VDI which is based on the GKS concepts, not GEM as a whole.
Secondly even VDI was not a copy but a reimplementation of similar concepts.

Maybe "copy" is indeed too harsh a word here. Let's say it's a "simplified reimplementation".

One interesting aspect: you might know the VDI function v_cellarray() ? It's not implemented in any VDI I know of and the DR documentation is somewhat obscure at best:

»Cell array« - Colour setup of the screen by reference to coordinates of coloured cells.
I suspect they just didn't understand what it is supposed to do.

In GKS, a cell array is the most important concept to implement device independent, scalable rasters. Something that GEM VDI completely lacks. Probably just because they didn't get it.

Regarding the AES: IMHO, it's lightyears away from something that even remotely deserves the name "operating system" (and, to be fair, probably was never intended as such). It mainly implements a windowing system and the glue between the VDI and the "real" OS. And while we are at it (and everybody talked about multitasking that time), let's nail some half-hearted implementation of concurrency to it.
Remember: this was at a time when the X windows system on top of "true" multitasking systems already saw light. I don't want to bad-mouth it (it just was what we had that time and I personally like it for it's simplicity), but in my opinion, it's far from genius...

[calimero]

@mfro ryan wrote that "real genius of GEM was aes" - not that AES itself is genius but rather as best one of three components of GEM.

@zarchos thanks for link. Bill Rehbock is also on this forum!

From Zarchos link (http://www.obsolete-tears.com/anecdotes ... er-67.html) Google translated:

Thierry about a discussion between Duval (Mr. en dev atari), Thierry and Rudolph ... to end the show Microphone & Co of 92 (or 91);
---------------------------
He spoke especially us a detail that I thought hype at the time and that proved true after several overlapping: the first version of sparrow / FX1 were equipped with a chip video much better than the crap which ended on falcon, including a chunk mode 256 colors, it was still too inconsistent with the old modes video and mostly incompatible with the atari habits (plans instead of chunk), so it was replaced with a super 8 mode unusable plans.
Notice we fared well, one could also have a true color fashion
16 shots! He also talked about the microbox which was to be full 32bit finally he talked about so much good stuff ... to sort

so there was once a 8bit chunk mode, no wonder than there are trace of it in TOS!

[Zarchos]

I really thought the Falcon had a 256 colour mode, chunky ... exactly like the Archie, but with a true freedom for selecting them (unlike the Archie).
In reality this mode is planar ? What the heck they had in mind ?
Ok I have just checked : chunky is only for truecolour modes.
Ootch, that's a lot of data to move.

[calimero]

So on Archi you have 8bit chunk mode but you can not choose colors? It is like 8bit pallet... (half of ST color palete)?

[Zarchos]

So on Archi you have 8bit chunk mode but you can not choose colors? It is like 8bit pallet... (half of ST color palete)?

You have only 16 entries so the other 240 colours are derivated from your freely selectable 16 base colours.
Still, it is possible to have some interesting 256 colour palettes.
http://www.stardot.org.uk/forums/viewto ... tte#p48090
http://www.stardot.org.uk/forums/viewto ... 256+colour

[mfro]

@mfro ryan wrote that "real genius of GEM was aes" - not that AES itself is genius but rather as best one of three components of GEM.

A pointless question. Nothing I'd really like to judge on. Just as if you asked me which one I'd rather give away: heart, brain or liver .

And, just to clarify that: GEM is only two components: the AES and the VDI.

[Neurotoxic]

Indeed the history of the Falcon is a little bit confusing. At first there was developed the Sparrow that became the Falcon030. At the same time Atari tried to develope a computer that could have become the Falcon but was much more powerful and would have come with a 68040 cpu? So far we know now the history of the Falcon, I think. After releasing the Falcon030 the Falcon030 Microbox was developed by Atari that wasn't released anymore?

You may take a look at the Microbox here:

https://forum.digitpress.com/forum/show ... box-casing

The Microbox case had a similar look like the Sony Playstation case. But there are also photos on the web of Atari's (Microbox) Falcon040 prototype. You may take a look at them here:

http://www.maedicke.de/atari/hardware/falcon040.htm

http://www.atari-computermuseum.de/pics ... robox1.jpg

Is anybody able to reconstruct Atari's roadmap of the Falcon computer? Did a second or later prototype of the Falcon040 exist fitting in the bigger Microbox case that's not looking like the Sony Playstation? Or was only just one prototye of the Falcon040 developed at the same time the Sparrow was developed? Im not sure but I think Atari itself wasn't sure about what to do and what not after they released the Falcon030? They still had a few plans with the Falcon and produced perhaps some prototypes but can anybody tell what prototype was planned to release in series production?

[AdamK]

And, just to clarify that: GEM is only two components: the AES and the VDI.

That is not 100% accurate. GDOS is also part of GEM, it just did not fit ROM, so it was dropped from ROM TOS.

[mfro]

And, just to clarify that: GEM is only two components: the AES and the VDI.

That is not 100% accurate. GDOS is also part of GEM, it just did not fit ROM, so it was dropped from ROM TOS.

GDOS is just the part of VDI that didn't fit into the ROMs.

[AdamK]

And, just to clarify that: GEM is only two components: the AES and the VDI.

That is not 100% accurate. GDOS is also part of GEM, it just did not fit ROM, so it was dropped from ROM TOS.

GDOS is just the part of VDI that didn't fit into the ROMs.

Yes, it is also a valid classification, but this stil leaves one component out: Desktop

[wongck]

IMHO Atari squeeezed a cut-down lame VDI into the ROM.
The lame VDI can only load one device - the screen and nothing else.

[calimero]

Is anybody able to reconstruct Atari's roadmap of the Falcon computer? ...

It would be cool to have timeline of evolution of Falcon or any other Atari computer.
There are many problems with Falcon/Falcon040/Microbox history: many cases simple does not match with prototype motherboards so first we should compare which motherboard can be fit in which case.