YouTube: Amiga as Workstation in 1986

[klx300r]

Originally Posted by aeberbach

Thom I am loving these videos, remembering the time I first sat in front of a 1-floppy 512K A1000 and tried to make a compiler work. Keep it up!

@hammer, just try to enjoy the videos...

+1

[peo]

Hi Thom,

Just found your really interesting series on historical use of the Amiga 1000 for development (and as imposed, other serious use later on). If not allowed to share the ADFs you are using in the series (none of which is available for purchase), can you at least provide some hints (md5 / tosec reference) for each part of the series what files (floppys / adfs and documentation) were used, so we curious people can follow along if we want ?

For "Part 0", the extra demo disk(s) are also interesting stuff (I probably already have these "somewhere").

The C guide was quite easy to find (it's on RCEU and other places): "Amiga C Compiler Users Reference Guide", with md5 CC00E0AFC3CAD3E23B86E3DF6009E4FC

I probably found the old Lattice C floppy you used, or an older one (1.0) without the make-c-cli script that deletes stuff from the copy of the workbench floppy.

As for Workbench/AmigaOS 1.0 or 1.1, and the narrator.device and speak-handler,you mention that Cloanto didn't include them on their disks (might or might not be so now), but when I recently installed AmigaOS 1.3 on one of my A2000s, these files were included (I'm sure it was the Cloanto version, as the startup-sequence had some modifications by them).

[Thomas Richter]

Originally Posted by tschak909

While AmigaDOS from the beginning could attach block devices from external code, there was no mount utility, so there were external programs which called exec and dos.library to present the device to the system. Mount would be added in 1.2, as would BindDrivers, the expansion library, etc, to make this easier.

The problem was actually one level earlier because there was no way to let kickstart call into code that could actually perform the mount. There is no call in exec or dos to mount something, it requires to build the data structures and add them to the doslist yourself. That became a call to expansion.library (actually) in 1.2, and to dos.library (finally) in 2.0.



Coming back to what I said, there was no boot mechanism that would execute custom code that could perform such steps, unless you somehow intercepted access to the kickstart WOM and fiddled your code in there, depending on some implementation details of the WOM. All of this arrived with autoconf in 1.2 (though romboot did not work before 1.3).
All of this had to happen as part of the startup-sequence and required access to a floppy. That's what why I'm asking...

[Thomas Richter]

Originally Posted by peo

As for Workbench/AmigaOS 1.0 or 1.1, and the narrator.device and speak-handler,you mention that Cloanto didn't include them on their disks (might or might not be so now), but when I recently installed AmigaOS 1.3 on one of my A2000s, these files were included (I'm sure it was the Cloanto version, as the startup-sequence had some modifications by them).

The issue is that the narrator and the translator are IPs of Softvoice, and the company still exists and wants to get paid for distributing their software. Yes, I asked them for including software in 3.1.4, and the answer was "$$$ or else", and therefore, it did not happen. Thus, I have some doubts on what Cloanto is doing here...

[pandy71]

Originally Posted by Thomas Richter

The issue is that the narrator and the translator are IPs of Softvoice, and the company still exists and wants to get paid for distributing their software. Yes, I asked them for including software in 3.1.4, and the answer was "$$$ or else", and therefore, it did not happen.

Isn't feasible to replace Softvoice software with some open source alternative?

[Locutus]

Originally Posted by pandy71

Isn't feasible to replace Softvoice software with some open source alternative?

Probably, but I guess you run into several problems: modern TTS codebases might not perform acceptably on a 68000 and also ensuring that you can parse the old phonemes correctly and generate something that is intelligable as the same thing is also not so simple perhaps.


I think this conversation came up before.


And all of this is compounded by how much development effort vs return this is especially once the 5 people who really want and will use such a 'back-port' could just copy narrator.device from their old floppies/whatever.

[peo]

https://www.text2speech.com/
Looks like a web page from the early 90's and their examples sounds like the AmigaOS narrator from the mid 80's..
"To date, there are over 8 million copies of the SoftVoice text-to-speech system in use world-wide, making SoftVoice, Inc. one of the largest providers of text-to-speech in the world.", so there is 8 million Amigas running WB 1.3 or 2.04

Really odd that they are so eager to hold on to their distribution rights for the software that hasn't been updated in more than 35 years.
It was impressive back then, but not that impressive 35 years later..

Anyone who have 1.3 or 2.04 have the needed files on their original disks (or on the ADFs provided by Cloanto).

[drHirudo]

Originally Posted by peo

https://www.text2speech.com/
Looks like a web page from the early 90's and their examples sounds like the AmigaOS narrator from the mid 80's..
"To date, there are over 8 million copies of the SoftVoice text-to-speech system in use world-wide, making SoftVoice, Inc. one of the largest providers of text-to-speech in the world.", so there is 8 million Amigas running WB 1.3 or 2.04

And Apple Macintoshes!
More quote from the page:

In the early '80s, SoftVoice developed text-to-speech systems for both Apple and Amiga Corporations for their soon-to-be-released machines. In fact, at its launch in 1984, the Macintosh© announced its own existence to the public using our software, the original "MacinTalk"©. The Commodore Amiga had an early SoftVoice text-to-speech system known as the "narrator.device" included in its operating system.

Five years ago, I've created a simple video with demo of the early MacSpeak software.

Some comments started popping out of nowhere that it sounded very close to a popular meme at the time. People started requesting making the speech with the meme, so I've created another video with the meme text that received over 150,000+ views and 250+ comments. Unfortunately, the meme was associated with the Apple Macintosh talk, even if it is probably the same text to speech software used for the Amiga as well.

On the topic: Early Macintosh systems were also appropriate for workstations (and costed as much as one), but Apple initially was more focused on developing their own operating system instead of UNIX clone. But they predated the Amiga UNIX with their A/UX

https://en.wikipedia.org/wiki/A/UX

A/UX is a Unix-based operating system from Apple Computer for Macintosh computers, integrated with System 7's graphical interface and application compatibility. It is Apple's first official Unix-based operating system, launched in 1988 and discontinued in 1995 with version 3.1.1.[2] A/UX requires select 68k-based Macintosh models with an FPU and a paged memory management unit (PMMU), including the Macintosh II, SE/30, Quadra, and Centris series.

The Amiga UNIX - https://en.wikipedia.org/wiki/Amiga_Unix
came later (in 1990) and was offered as alternative to AmigaOS instead of integrated with AmigaOS.

Amiga Unix (informally known as Amix) is a discontinued full port of AT&T Unix System V Release 4 operating system developed by Commodore-Amiga, Inc. in 1990 for the Amiga computer family as an alternative to AmigaOS, which shipped by default.

[Thorham]

Originally Posted by peo

"To date, there are over 8 million copies of the SoftVoice text-to-speech system in use world-wide, making SoftVoice, Inc. one of the largest providers of text-to-speech in the world.", so there is 8 million Amigas running WB 1.3 or 2.04

It's running on the peecee as a Windows DLL

[pandy71]

Originally Posted by Locutus

Probably, but I guess you run into several problems: modern TTS codebases might not perform acceptably on a 68000 and also ensuring that you can parse the old phonemes correctly and generate something that is intelligable as the same thing is also not so simple perhaps.


I think this conversation came up before.


And all of this is compounded by how much development effort vs return this is especially once the 5 people who really want and will use such a 'back-port' could just copy narrator.device from their old floppies/whatever.

Fair argumentation - most of us if not all can copy files from OS disks...

[aeberbach]

Originally Posted by drHirudo

On the topic: Early Macintosh systems were also appropriate for workstations

Apple pricing made a Sun look like a reasonable option (Mac II with A/UX: ~$9k, Sparcstation IPC: $9k) , but why go on... they were a good DTP and business tool, not so much for anyone that wanted anything else. Certainly not a weird Apple-ized Unix.

[arcanist]

I'm enjoying your videos. It's quite eye opening to see how slow Lattice C ran on a 68000.

I followed Cliff Ramshaw's book Complete Amiga C on an A500, thankfully with a hard disk. How I had the patience for that I'll never know.

[idrougge]

Originally Posted by peo

Really odd that they are so eager to hold on to their distribution rights for the software that hasn't been updated in more than 35 years.

No more odd than the fact that some companies are eager to hold on to their distribution rights for Workbench/Kickstart 1.3/3.1.

It was impressive back then, but not that impressive 35 years later..

Exactly.

[tschak909]

Working through content creation of the next part of "Amiga as a Workstation in 1986" videos.

We write a simple program to get used to the #CommodoreAmiga Intuition and Graphics libraries.

I will record the video soon. #retrocomputing

[tschak909]

In Part 4 of our "Amiga as a Workstation" series, we write a small program to verify that we understand how graphics and intuition libraries work.

We also do some quality of life enhancements such as installing MicroEMACS, and making icons for our source and binary files.

[ Show youtube player ]

#retrocomputing

00:00 Intro
01:10 Why MicroEMACS?
02:30 Installing MicroEMACS.
06:15 Starting MicroEMACS
08:20 Creating Drawers
10:35 Create a document icon from Notepad
12:45 Some more tweaks
13:50 Adding ASSIGNs
17:00 Starting on sine.c
17:38 Bringing over code/structs from one.window.c
22:56 The Intuition Manual
25:25 Altering NewWindow
29:58 Adding external global structs and our window
31:00 main(), init(); done().
36:18 K&R C == almost no prototypes
38:28 opening our window
43:22 Compile our Blank Window
44:10 Adding our grid() while compiling
45:35 Whoops, variable named wrong.
49:14 Running our blank window
50:13 Adding our grid(), for real
52:20 Compiling window + grid
52:40 Adding sine() while compiling
55:10 Running window + grid
57:16 Adding sine, and event handling & compile
1:01:48 Whoops, missing math.h
1:03:04 Drawing an icon while compiling
1:06:00 Running our finished program

[pixie]

It's been a very interesting journey, mine had begin 1.3 upwards, but only from 3 upwards that I've started to get to know more about the system. It's really interesting to see how things used to be back then, and hope to see how things have evolved over the years.
The Sine program was the cherry on top!

[CCCP alert]

I used Sun workstations in 1993-95 at university, they came with insane resolutions on massive 24 inch Sony Trinitron monitors connected via component video. Stunning machines next to the many 486s PCs in the same room.

All I did was use Netscape on them lol but a dream machine for sure.

[Gorf]

Originally Posted by pandy71

Isn't feasible to replace Softvoice software with some open source alternative?

Yes:

What is SAM?

Sam is a very small Text-To-Speech (TTS) program written in C, that runs on most popular platforms. It is an adaption to C of the speech software SAM (Software Automatic Mouth) for the Commodore C64 published in the year 1982 by Don't Ask Software (now SoftVoice, Inc.).

https://github.com/bit-hack/SAM

There is also a JS version for the web browser:

https://discordier.github.io/sam/

And here are the 68k binaries for the Amiga from yet an other port:
http://aminet.net/package/util/sys/sam

[pandy71]

@Gorf - appreciate your feedback but i don't see any added value from replacing one SoftVoice IP by another SoftVoice IP... Thomas is probably right on this - if you need TTS then copy required files from OS disks, any modern alternative will be memory and CPU hog so plain 68000 will be too slow and 512+512 probably will be also not enough... Seem there is Talkie for Arduino platforms where pre calculated LPC is used.