Jump to content
IGNORED

130ST TOS Boot Disk???


Recommended Posts

As we know, the Atari 130ST was never released and their prototypes was showed only during the Winter edition of CES Las Vegas in January 1985.

 

It's also known the TOS was still in development then. Those STs were running CP/M 68K and later soon the decision to move to new file system, GEMDOS, was made.

 

As we can read in the fantastic post from Landon Dyer's blog "The Atari ST (part 1)": https://dadhacker-125488.ingress-alpha.easywp.com/the-atari-st-part-1/

 

Quote

We had disk-based versions of the OS (called TOS, for “The Operating System” — catchy) booting, and that’s what we showed.  The hardware guys doubled the amount of RAM in the system so the OS could live in RAM with room left over for applications.

 

So, beside the curious fact that those 130ST prototypes were running with 256K of memory, we know that they booted the OS from disk.

 

Up to now, I've just found two sources where we can know something about the content of that disk and the look of that GEM Desktop.

 

1. "The First Atari ST Book" (http://www.atarimania.com/documents/FirstAtariStBook.pdf) from First Publishing, an english translation of the original german Data Becker edition from 1985. This edition was later revised and updated but the original version was written before the launching of the first STs and as you can read in the preface:

 

Quote

The information in this book is derived from hands-on experience which the authors received during their work with prototypes of the new Atari ST.

 

In Chapter 5, "Working with GEM", there are some pictures showing the screen of this proto GEM Desktop. It shows an opened window from A:\ and we can read the content of it:

 

Three folders: AFOLDER / GEMDOS / SRC_DIR

Seven files: DESK1.ACC / DESK2.ACC / DESKHI.IMG / DESKLO.IMG / DESKTOP.SRC / GEM.KOM / GEM.SRC

 

The content of all the files and folders occupy 216704 bytes

 

2. This youtube video recorded during the Atari's new line of systems presentation in the Winter CES'85. From 19' 20'' onwards we can see with more detail that GEM Desktop booted from those 130ST prototypes. And on a certain moment someone open the content of A:\ and we can see it's the same content we can see from the book.

 

Dos anyone know where to find this TOS Boot disk?. It's part of the early Atari ST history and would be great for the community to have it.

 

 

 

  • Like 1
Link to comment
Share on other sites

What was presented at the end of 1985 was not with CP/M, but with TOS on disk. There was small sized ROM of 16 KB in machines, for booting TOS and GEM.

TOSBOOT.ZIP

Here is floppy image with early TOS 1.00 US: TOS10US.ZIP

UK: TOS100uk.zip

Icons are not same, and they are changed because Apple .

Here I must say that things look pretty much that DRI was not so motivated (paid ?) for pushing 68000 CPU based platform - we know that GEM was done for PCs too. C compiler for 68000TOS10US.ZIP done by DRI (Alcyon) had some limitations, and they remained at least until 1989. While competition did huge progress. Actually, is there any SW done by DRI after 1985 what became popular ?

  • Like 2
Link to comment
Share on other sites

1 hour ago, ParanoidLittleMan said:

What was presented at the end of 1985 was not with CP/M, but with TOS on disk. There was small sized ROM of 16 KB in machines, for booting TOS and GEM.

TOSBOOT.ZIP 8.83 kB · 2 downloads

Here is floppy image with early TOS 1.00 US: TOS10US.ZIP

UK: TOS100uk.zip

Icons are not same, and they are changed because Apple .

Here I must say that things look pretty much that DRI was not so motivated (paid ?) for pushing 68000 CPU based platform - we know that GEM was done for PCs too. C compiler for 68000TOS10US.ZIP done by DRI (Alcyon) had some limitations, and they remained at least until 1989. While competition did huge progress. Actually, is there any SW done by DRI after 1985 what became popular ?

 

Hi!

 

Thanks for the answer but this is not exactly what I’m looking for here.

 

I know not many people know TOS and its code as well as you but I think I’m referring here to a proto-TOS that never was really commercialized in the form it was in that moment.

 

The 130ST was presented at the beginning of 1985 and it seems that effectively those prototypes were running under CP/M 68K. I mean only those prototypes and not the 520STs and 260STs that were finally sold later that year with the 16 KB ROM and the Tos Boot Disk you’ve shared here.

 

There is an article from START Magazine published in Summer ‘88, “Three years with Atari ST”, https://www.atarimagazines.com/startv3n1/threeyearsofst.html

 

There it says this:

 

Quote

The version of GEM first demonstrated to Tramiel Technology was actually running on top of CP/M-68K. Indeed, GEM's CP/M-68K incarnation was the only version of GEM available for the 68000 microprocessor at the time, and Atari would continue to plan for it to be the ST's underlying operating system in the months to come.

 

Later in the same article:

 

Quote

 

Miraculously, by the beginning of January 1985 all the pieces began to come together. Atari officials realized they would meet the CES deadline, and they did: they shipped five STs to Las Vegas, with GEM still running on top of CP/M-68K.

 

For Shivji, this was the climactic, energizing moment: "The really exciting thing was that in five months we actually showed the product at CES with real chips, with real PCBs, with real monitors, with real plastic. Five months previous to that there was nothing that existed. You're talking about tooling for plastic, you're talking about getting an enormous software task done. And when we went to CES, 85 percent of the machine was done. We had windows, we had all kinds of stuff. People were looking for the VAX that was running all this stuff."

 

 

And later:

 

Quote

Yet Atari still had much work ahead of them. In February, GEMDOS was nearly complete and Atari had to make a crucial decision: should they continue with CP/M 68K or to move on to GEMDOS? Leonard explains: "That was an extremely difficult decision to make. CP/M-68K had been around several years; it was a well-known, well-understood, relatively well-accepted existing operating system. GEMDOS was a completely brand-new, untried, untested, incomplete operating system. However, it also offered significantly higher performance and gave the full hierarchical file system that CP/M-68K simply did not have. It was quite a difficult decision to make, but I think we went in the right direction going with GEMDOS."

 

Quoting again a post from Landon Dyer, one of the Atari software engineers who were at Monterey together with the DRI SW engineers to begin the port of GEM (https://dadhacker-125488.ingress-alpha.easywp.com/the-atari-st-part-2/?

 

Quote

 

October: Work. We get (rented) houses in Monterey.

November: More work. We barely see those houses.

December: Much more work. The ST boots TOS for the first time.

January: CES (with STs running CP/M-68K). Decision made to move to new file system (GEMDOS).

February: 16K boot ROMs written (a couple-week side effort).

March: Even more work. Two weeks to crunch TOS to fit into 192K.

April: ROMs actually work (do you know how long it takes to burn 192K of ROM, not to mention UV-erasing older chips?)

May: ROM TOS 1.0 shipped.  Phew!

 

 

So, it really seems those STs showed in Las Vegas in january ‘85 were running CP/M 68K

 

Maybe someone has more information about the SW that was used to boot those early prototypes.

 

  • Like 1
Link to comment
Share on other sites

2 hours ago, Mr. Undo said:

So, beside the curious fact that those 130ST prototypes were running with 256K of memory, we know that they booted the OS from disk.

I knew the 130ST was shown.  But I never could understand how a 128K ST could be functional, especially in the disk TOS days.   Now I know :)

  • Like 1
Link to comment
Share on other sites

3 hours ago, Mr. Undo said:

1. "The First Atari ST Book" (http://www.atarimania.com/documents/FirstAtariStBook.pdf) from First Publishing, an english translation of the original german Data Becker edition from 1985. This edition was later revised and updated but the original version was written before the launching of the first STs and as you can read in the preface:

Very valuable book. I didn't know it.

Interesting is that, there is still a paragraph about CP/M (instead of GEMDOS) as Operating System.

thanks

 

 

Also interesting is that the Desktop has a black border!

https://youtu.be/nIKzXHNh_pQ?t=1388

 

Edited by Cyprian
  • Like 1
Link to comment
Share on other sites

Well, what I wrote is based on what I read back in time. All literature says that first (disk based) TOS was presented at CES at beginning of 1985. I don't think that it is possible that whatever CP/M or GEMDOS + GEM with Desktop (and that's longer part) can fit in 128 KB RAM. Even with 256 KB is very tight.

Let's see it:  screen 32 KB, OS workspace min some 20 KB. AES and Desktop workspace some 30-40 KB. Already min 80 KB. AES+Desktop - about 90 KB min.

So, very little left for GEMDOS, maybe little better with CP/M . But that's good only for presentation, not for running some serious SW. There were obviously misjudgements, and they even manufactured cases with 260 ST logo, and even used them ?

After reading all it here, I really can blame mostly leaders for this mess. It seems that at beginning they had no clue about what OS it should have. Or just seriously misjudged memory requirements - like 128 KB RAM - no way. Even 128 KB ROM appeared way too short.

"March: Even more work. Two weeks to crunch TOS to fit into 192K." - funny thing is that they solved it with Line-F emulation in AES, saved some 5-6 KB with it. While could save more by packing RSC +  Desktop.inf template data at end of TOS . Happens when people is under pressure.  Btw. Landon Dyer was not happy about how DRI people threaten them, programmers at Atari.

Little jump to 1992 - TOS 2.06 - biggest step in TOS evolution. As I know it was done completely in Atari.

 

I really don't know what is difference between CP/M and GEMDOS. Know what is difference between GEMDOS and PC DOS FAT16 filesystem. Would be good to read something about CP/M hard disk filesystem. Probably it was limited to small sizes, considering drive capacities of that time. And saying "GEMDOS was a completely brand-new" - nothing is completely new .

 

Back to topic: I think that chances to dig out those early, CP/M TOS versions are really small. It was used only internally, so never went out from Atari/DRI . There is still some commercial SW without good images. Not to mention TOS 1.xx sources ...

Link to comment
Share on other sites

Quote

Back to topic: I think that chances to dig out those early, CP/M TOS versions are really small. It was used onlyinternally, so never went out from Atari/DRI

Yeah, I know it’s not going to be easy to find this CP/M TOS and maybe it never will see the light. I’ve searched hard in internet without any success. But, who knows?

 

Maybe someone have it. Or maybe someone who worked in the GEM port to ST at Atari or DRI back then could some day read this topic and bring some light to this really not well known episode of ST history. Does not follow Leonard Tramiel this forum? ?

Link to comment
Share on other sites

1 hour ago, Mr. Undo said:

Maybe someone have it. Or maybe someone who worked in the GEM port to ST at Atari or DRI back then could some day read this topic and bring some light to this really not well known episode of ST history. Does not follow Leonard Tramiel this forum? ?

 

Yes, this is the most likely way this will see the light of day, if it ever does.

 

Something along the lines of, "I worked for Atari and I just found a bunch of my work disks in the attic. Would these be of any use to you guys?" or "My Mom used to work at Atari in the 80s. She just passed and I found all this stuff in her basement when I was cleaning it out. Do you think it's worth anything?"

 

I think the odds are still pretty long, but stranger things have happened. It will all hinge on this stuff getting into the hands of the right person, rather than going out with the weekly trash, and then that person knowing what the have their hands on. Otherwise, it could just wind up sitting in someone else's attic, even if it is saved.

 

Link to comment
Share on other sites

2 hours ago, ParanoidLittleMan said:

Well, what I wrote is based on what I read back in time. All literature says that first (disk based) TOS was presented at CES at beginning of 1985. I don't think that it is possible that whatever CP/M or GEMDOS + GEM with Desktop (and that's longer part) can fit in 128 KB RAM. Even with 256 KB is very tight.

According to that page 130ST was presented at CES Winter 1985 with 192Kb ROM and 128kB RAM

https://www.islandnet.com/~kpolsson/comphist/comp1985.htm

"At the Winter CES, Atari introduces the Atari 130ST computer. It features 128 kB RAM, 192 kB ROM including Digital Research's GEM operating system, 640x400 monochrome or 320x200 16-color graphics from a palette of 512 colors, 32 kB screen RAM, MIDI interface, and mouse. Price is US$399."

 

 

2 hours ago, ParanoidLittleMan said:

Even 128 KB ROM appeared way too short.

"March: Even more work. Two weeks to crunch TOS to fit into 192K." - funny thing is that they solved it with Line-F emulation in AES, saved some 5-6 KB with it.

yep, that was really short-term thinking to place ROM area at $FC0000. They should use $Exxxxx and bigger ROM from the beginning.

Another example of short-term thinking is the SHIFTER's color registers - they should use 4bit per each color component from the beginning or just use binary pattern %xxxxRRRxBBBxGGGx

 

 

2 hours ago, ParanoidLittleMan said:

I really don't know what is difference between CP/M and GEMDOS. Know what is difference between GEMDOS and PC DOS FAT16 filesystem. Would be good to read something about CP/M hard disk filesystem. Probably it was limited to small sizes, considering drive capacities of that time. And saying "GEMDOS was a completely brand-new" - nothing is completely new .

flat filesystem without folders.

Regarding the rest, GEMDOS is based on the CP/M and have similar functions.

 

There are some informations about GEM and CP/M. There is also "GEMDOS for the ATARI ST. Source included"

http://www.retroarchive.org/cpm/archive/unofficial/

 

 

Interesting is that DRI had a preemptive multitasking version of GEM - X/GEM. That's pit Atari didn't implement that in the ST.

 

 

 

Edited by Cyprian
Link to comment
Share on other sites

2 hours ago, ParanoidLittleMan said:

So, very little left for GEMDOS, maybe little better with CP/M . But that's good only for presentation, not for running some serious SW. There were obviously misjudgements, and they even manufactured cases with 260 ST logo, and even used them ?

After reading all it here, I really can blame mostly leaders for this mess. It seems that at beginning they had no clue about what OS it should have. Or just seriously misjudged memory requirements - like 128 KB RAM - no way. Even 128 KB ROM appeared way too short.

Well the management wanted a 16-bit Macintosh killer,   the rest is just details :)  

 

The Mac had a 128K model, and Atari wanted the same.   But Apple also wrote their own OS, and it was on ROM from the start.   Atari was shopping for a GUI from third parties and were at their mercy for memory requirements.

 

 

  • Like 1
Link to comment
Share on other sites

Quote

 

According to that page 130ST was presented at CES Winter 1985 with 192Kb ROM and 128kB RAM

https://www.islandnet.com/~kpolsson/comphist/comp1985.htm

"At the Winter CES, Atari introduces the Atari 130ST computer. It features 128 kB RAM, 192 kB ROM including Digital Research's GEM operating system, 640x400 monochrome or 320x200 16-color graphics from a palette of 512 colors, 32 kB screen RAM, MIDI interface, and mouse. Price is US$399."

 

 

Yes, I think this is taken from the official info you could read in most of the magazines and articles from the time. Even from those who covered the event. But I really give more credit to Landon Dyer who was there working in the port. Nobody from the press neither the visitors could know those computers really ran with 256 KB of RAM and with TOS booted from disk. It was all made so fast!

Link to comment
Share on other sites

I see here lot of contradictional claims, and some really not realistic ones.

Like price of $399 for computer with MC68000 in 1985. As I know price was around 1000 $ in beginning. And even, if it was with only 128 KB RAM, could not be so low.  Maybe there was really 192 KB ROM in that prototype, with unfinished TOS. Still, it was pretty much useless with so little RAM, because OS, GEM need much more than some 8-bit computer.  Remember disappointment of testers when first 520 ST with TOS on disk arrived on market ? When DR Basic was loaded, there was only some 10-15 KB free RAM left.  I think that better would be to delay launch, or at least to ship finished ROM TOS for early buyers for free.

19 hours ago, zzip said:

Well the management wanted a 16-bit Macintosh killer,   the rest is just details :)

 

The Mac had a 128K model, and Atari wanted the same.   But Apple also wrote their own OS, and it was on ROM from the start.   Atari was shopping for a GUI from third parties and were at their mercy for memory requirements.

 

 

According to Landon Dyer, who moved to Apple, their code was written much more efficient considering memory usage.

GUI was not from third party - it was done by DRI, and it was practically port of GEM for PC. Somehow I tend to think that DRI programmers did not care for things like low RAM/ROM usage. Surely, it is better when all it goes in own company. Tramiel should recruit some experienced programmer people - but that's lot of money ..  In case of C64 it was similar - they bought MS Basic .

 

So, CP/M was without folders - that would be serious mistake to launch computer intended for serious work with such OS.

Yes, some general GEMDOS C sources are available, and I saw them earlier.  What would be nice to see are concrete TOS version sources.

And I wonder, how Apple did it ?  Maybe OS was done in ASM, big part.

 

Cyprian wrote: "yep, that was really short-term thinking to place ROM area at $FC0000. They should use $Exxxxx and bigger ROM from the beginning. "

Yes, should put it lower already at start. But bigger ROM - that would cost a lot. They used 32 KB ROMs because 128 KB ones were too expensive then. And I guess early plan was 4, and not 6 of them. But as we know, SW is what determines things since about 1983 .

Link to comment
Share on other sites

6 hours ago, ParanoidLittleMan said:

I see here lot of contradictional claims, and some really not realistic ones.

Like price of $399 for computer with MC68000 in 1985. As I know price was around 1000 $ in beginning.

520ST originally sold for $799.   That was a bundle with monitor and external floppy.   I don't think they could save $400 by going to 128K,  but they could maybe unbundle it too.   Jack was alway reaching for some crazy low retail price, so a $399 target doesn't surprise me.

 

6 hours ago, ParanoidLittleMan said:

According to Landon Dyer, who moved to Apple, their code was written much more efficient considering memory usage.

GUI was not from third party - it was done by DRI, and it was practically port of GEM for PC. Somehow I tend to think that DRI programmers did not care for things like low RAM/ROM usage. Surely, it is better when all it goes in own company. Tramiel should recruit some experienced programmer people - but that's lot of money ..  In case of C64 it was similar - they bought MS Basic .

In this case "third party" means it was not written by Atari.   They bought an existing UI,  although they did heavily customize it.    GEM was written to be cross-platform, therefore written in C which does not produce as efficient code as assembler.  (I don't know if Mac used assembler)

 

The MacOS fit into 128K ROM.  GEM was more bare-bones feature wise and rougher around the edges than Mac OS, especially the version of GEM in TOS 1.0.  So there was a lot that could have been optimized better for sure.  

 

Maybe it's for the best that Jack didn't get his 128K model.   Because then that would have been the baseline that developers coded for.  :)

Edited by zzip
  • Like 1
Link to comment
Share on other sites

The Mac had the System and Finder files that were loaded from disk on boot - I assume the 128K rom (64K originally) had the barebones OS and the disk files additional pieces.

 

IIRC in the US - 1040ST w/Mono $999, w/color $1199.  First "Megabyte" computer under $1000.

Link to comment
Share on other sites

12 minutes ago, Goochman said:

The Mac had the System and Finder files that were loaded from disk on boot - I assume the 128K rom (64K originally) had the barebones OS and the disk files additional pieces.

Ok, but even still   128K mac with 64K ROM   is 192K  available for OS (and some must be left for user-space code, housekeeping and screen RAM).

 

It's hard to imagine that a 128K ST even with a proper 192K TOS ROM installed would have enough memory to boot itself.   Mac OS still seems much more efficient here.

 

16 minutes ago, Goochman said:

IIRC in the US - 1040ST w/Mono $999, w/color $1199.  First "Megabyte" computer under $1000.

Correct.   The 1040ST came out a few months after the 520 for $200 more.

Link to comment
Share on other sites

Some very interesting contributions all that I read here. Thanks!

 

Just a little appreciation:

 

Quote

I think that better would be to delay launch, or at least to ship finished ROM TOS for early buyers for free.

As far as I know, the finished ROM TOS were effectively shipped or delivered through dealers to early buyers once they were completed. This fact was announced by Atari in those days and has been confirmed by some of those early buyer in some forum.

 

Returning to the original question on this topic, the proto-TOS Boot Disk, if we compare the content of the floppy disk they used during Winter’85 CES for the ST prototypes (the same disk the authors of “The First Atari ST book” received along the prototype to test the machine), with the content of the original TOS Boot Disk delivered to early costumers, there are some coincidences: both contained the DESK1.ACC and DESK2.ACC. Same names but different content in those files I guess. This is the GEM Menu diagram for the ST prototypes:

 

Proto-TOS_GEM_Menu_Diagram_red.thumb.jpg.d5687beaaab65d50dc39a4fab25adf0e.jpg

 

Differences in the GEM Menu between the prototypes and TOS 1.00 (using TOS Boot Disk):

 

  • “Desk” dropdown menu in prototypes: Desktop Info / Breakout / Calculator
  • “Desk” dropdown menu in TOS 1.00 (with Boot Disk containing DESK1.ACC and DESK2.ACC): Desktop Info / VT52 Emulator / Control Panel / Set RS232 Config. / Install Printer

It’s funny to see the Desktop Info from the prototypes: Copyright 1984 Digital Research Inc. / Copyright 1985 Atari Corp.

 

Gem_Desktop_130ST1.thumb.jpg.1a51e98bdbcdcf056da2861d87f09233.jpg

 

  • “File” dropdown menu in prototypes: To output… / Quit. These two options were dropped in TOS 1.00
  • “Options” dropdown menu in TOS 1.00: Save Desktop / Print Screen. These two options were added here

Regarding the rest of files and folders from the floppy disk used for the prototypes:

Quote

 

Three folders: AFOLDER / GEMDOS / SRC_DIR

Seven files: DESK1.ACC / DESK2.ACC / DESKHI.IMG / DESKLO.IMG / DESKTOP.SRC / GEM.KOM / GEM.SRC

 

Any clue about its possible content? Looking at them, can we deduct this floppy disk was used as a boot disk?

Link to comment
Share on other sites

I based that "to ship finished ROM TOS for early buyers for free" because there was lot of STs with only 16 KB boot ROM even years later. Maybe it was some sellers who 'forgot' to inform people about that ?

Anyway, in adverts all it seems so great, 'revolutionary' (if I had a $ for every case seeing that word in advert ...), simple ...

According to pic of desktop and file names, it was quick conversion of DR DOS and GEM for PC to Atari 68K.  5.25 inch floppy icons, GEM.KOM - probably executable file extension (from COM). AFOLDER - AUTO folder ?

  • Like 2
Link to comment
Share on other sites

53 minutes ago, ParanoidLittleMan said:

there was lot of STs with only 16 KB boot ROM even years later.

Yes, that’s also true for sure. So we can affirm here that those finished ROM TOS were delivered at least to some buyers, definitively not all of them.

 

And what about DESKHI.IMG and DESKLO.IMG? What could that “high” and “low” distinction have meant?

Link to comment
Share on other sites

Was the Apple Lisa OS on ROM from the start? 

 

I think the price of DRAMs was hugely expensive when they were working on the ST, by the time of the 520STM in Spring 1986 (launched with 1040STF at the Atari Show) the price of RAM had dropped so much and showed no signs of levelling off that JT abandoned the 130ST idea (128k RAM, GEM in ROM 130ST), and they stuck with 512kb base spec and knew the 512k DRAM requiring machines would just get cheaper and cheaper so unbundled the lot and added the modulator for composite and RF output for 399 520STM....perhaps Jack thought the C64's biggest pull in 1982/83 was the 'huge' 64k RAM for the price? I bought a 520STM in spring 1986.

 

As I understand it the 260ST is just a rebadged 520ST with TOS on disk...a bit like how a 256k Amiga 1000 was actually 256k+192K protected RAM for OS on disk to be loaded into (survives crashes and resets).

 

There was also the 260STD (changed to STF later lol) and that had the disk drive on the left hand side so it wasn't a 256k 520STFM rebadged as those have disk drive on the right hand side like all STF/STE/Falcon models. No idea about what the RAM/ROM situation was on those.

 

Something I can find zero info on today, but read about in the weekly magazines of the mid 1980s, was the Atari ST console variant which was meant to have games on cartridge, standard ST spec cartridge port and standard 520ST specs with 128k RAM and no keyboard and to be sold for £199. Atari did have a habit of putting out news items for things they hadn't actually started engineering or designing to gauge public reaction like their Amstrad PCW8256 rival they also said they would be putting out by 1986 (not even a prototype picture!!). Of course if you look at Ghosts n Goblins and Commando on the ST they are good enough for 16bit console games technically.

 

I have a 260ST in the box somewhere. I did have an original TOS Boot disk but haven't seen that for over a decade (I have a lot of stuff!!) but used to see 1985 520ST models in Laskys all the time with that iconic boot disk request screen.

Link to comment
Share on other sites

I think that '260 ST' cases were just used for 520 STs, and never heard that anyone bought ST with only 256 KB RAM. There were diverse ideas, plans, but many of it appeared not that good, and situation with component prices, especially DRAM changed all time. I guess that Tramiel managed to get very good price by some higher quantity order, and that decided faith of machines with less than 512 KB RAM - dropped further development.

 

Lisa OS ? No way. Apple for sure did not sale it. Probably was just codename used by DRI programmers.

 

In case of home computer equipped with floppy drive, cartridges with SW are not really useful. Even if ROM prices dropped, capacity should be much bigger, so total price would stay high. C64, Atari 400 ... were with slow tape loading normally, so some like 32 KB cartridge with game was very handy - instant start instead minutes.

Other problem is time needed to code well and efficient with new, pretty much different CPU. ST was not equipped with HW scroll, HW sprites and like. Explanation was that CPU is enough fast to do it in SW.  And yes, that was actually true. Only that it needed couple years that programmers achieve it.

There are games with really smooth and fast scroll for bare STs. And to add, 512 KB RAM helped in it a lot (preshifted sprites, background need plenty of space) .

  • Like 2
Link to comment
Share on other sites

Might well be early ST GEM versions indeed ran on top of CP/M-68k, but I'm pretty certain the video linked above does *not* show this version. I would strongly assume this is already an (early) GEMDOS version.

 

As stated, the video clearly shows some GEM folders. CP/M-68k never had folders as it was a - more or less - direct translate from Z80/8080 CP/M that didn't support the subdirectory paradigm (but the "USER" concept instead that allowed to subdivide a disk in 16 different logical views). Of course it might be the folders in the video are just a fake, but why should they have shown them if they didn't already know they will support it?

  • Like 1
Link to comment
Share on other sites

On 3/13/2021 at 11:13 AM, mfro said:

Might well be early ST GEM versions indeed ran on top of CP/M-68k, but I'm pretty certain the video linked above does *not* show this version. I would strongly assume this is already an (early) GEMDOS version.

 

As stated, the video clearly shows some GEM folders. CP/M-68k never had folders as it was a - more or less - direct translate from Z80/8080 CP/M that didn't support the subdirectory paradigm (but the "USER" concept instead that allowed to subdivide a disk in 16 different logical views). Of course it might be the folders in the video are just a fake, but why should they have shown them if they didn't already know they will support it?

That’s a good point. Obviously they couldn't have used a folder system on a pure CP/M-68K, and the more I investigate the more I tend to think also this proto-TOS was really based on an early GEMDOS implementation. But it's neccesary to point out here that, as Landon Dyer said in his blog (https://dadhacker-125488.ingress-alpha.easywp.com/the-atari-st-part-2/), GEMDOS was still been written at the time. And the final decision to definitively abandon CP/M-68K was made in late January, only after continuing to work on GEMDOS for several more weeks and finally convincing Leonard Tramiel of its virtues.

 

The shortcomings of using a language like CP/M-68K as the basis for a GUI like GEM were very apparent. Quoting Landon Dyer:

Quote

CP/M (in any of its variants) didn’t really do a whole lot. There was a simple flat file system. There was some character-at-a-time console output (useless on a computer with a graphical interface). And CP/M could load and programs. That was about it. (By modern standards it was missing: A hierarchical file system with directories, networking, memory management, processes and process scheduling, a notion of time, synchronization and locking primitives, a driver architecture, graphics, fonts, character sets . . .

They had been working for months on the GEM port to a 68K-based system (without even having the final HW) and they had been using CP/M-68K as the base for a long time, so they were squeezing it and twisting it heavily to try to get GEM working on it. But during that time they at DRI also started to work in parallel to write a new operating system for the 68000 and, as Landon says

Quote

when we heard that someone at DRI had been doing something much better, even though it was still unfinished, we unofficially jumped at it.

A few days before CES they had already received the workable prototypes of the ST and were able to start implementing the OS directly on them. But exactly which version for the CES show? Landon states that those computers were running CP/M-68K, but, yeah, who knows? it's significant to note also that one of the folders of that possible boot disk that we see in the video and in The First Atari ST Book includes a folder named GEMDOS. And the presence of folders really indicates it was impossible than that Atari GEM version was running on top of a pure CP/M-68K.

 

Regarding this topic I've found another interesting article (http://cd.textfiles.com/crawlycrypt1/program/books/progem/gemdos.15) written in 1986 by Tim Oren, who was also working on the DRI side in the port of GEM, which is very enlightening:

Quote

 

PROFESSIONAL GEM  by Tim Oren
Column #15 - Coping with GEMDOS


           [...]

 

A BIT OF HISTORY. There has been a good deal of confusion in the Atari press and among developers over what GEMDOS is, and how it relates to TOS and CP/M-68K. It's important to clear this up, so you can get a true picture of what GEMDOS is intended to do. The best way is to tell the story of GEMDOS' origins, which I can do, because I was there.

           

As most developers are aware, GEM was first implemented on the IBM PC. PC GEM performed two functions. The first was a windowed graphics extension to the PC environment. The second was a visual shell, the Desktop, which ran on top of the  existing operating system, PC-DOS. When work started on moving GEM to the ST, there were two big problems. First, no STs actually existed. Second, there was no operating system on the 68000 with which GEM and the Desktop could run. Unix was too large, and CP/M-68K lacked a number of capabilities, such as hierarchical files, which were needed to support GEM.

 

Work on porting the graphics parts of GEM to the 68000 had to start immediately to meet schedules. Therefore, CP/M-68K running on Apple Lisa's was used to get this part of the project off the ground. Naturally, the Alcyon C compiler and other tools which were native to this environment were used. In parallel, an effort was begun to write a new operating system for the 68000, which would ultimately become the ST's file system. It was designed to be a close clone of PC-DOS,  since it would perform the same functions for GEM in the new environment. At this point, the term TOS was introduced. TOS really meant "the operating system,  whatever it may be, that will run on the ST", since not even the specifications, let alone the code, were complete at that time.

 

The  first engineer to work on "TOS" at Digital Research was Jason  Loveman. This name leaked to the press, and in some distorted fashion generated a rumor about "Jason DOS", which was still just the same unfinished project. As "TOS" became more solid,  the developer's tools were ported to the new environment one by one, and the GEM programming moved with them. CP/M-68K was completely abandoned, though the old manuals for C and the tools lived on and are still found in the Atari developer's kit.

 

All of this work had been done on Lisas or Compupro systems fitted with 68000 boards. At this point, workable ST prototypes became available. An implementation of "TOS" for the target machine was begun, even before the basic operating system was fully completed.

 

The other intent for the new operating system was to be a base for GEM on other 68000 systems as well as the ST. Because of this, Digital Research named it GEMDOS when it was finally complete, thus providing the final bit of nomenclature. "TOS" as now found in the ST is in fact a particular implementation of generic GEMDOS, including the ST specific BIOS.

 

 

The only way to completely get rid of doubts is getting the image of that floppy disk and the possible ROM content of those prototypes.

 

  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...