The Personal Computer Division White Papers

[sometimes99er]

On 1/25/2020 at 3:14 AM, Toucan said:

• "TI BASIC Interpreter System Documentation"....WRITTEN BY Robert B. Greenberg of Microsoft (8/24/1978).

Holy Grail. Guess some people will find it hard to accept any level of MS involvement.

 

[+dhe]

Many of the larger libraries will have book scanners that you can use, that scan from above and also auto correct for page curvature.

[Toucan]

I'll try scanning it on my flatbed scanner and see how that goes. Thanks for the offers of help. If it turns out to be too much of a task, I'll let you know what I decide.

[fabrice montupet]

On 1/26/2020 at 11:58 PM, Toucan said:

Awesome! I would love to see the Joysticks in color as well.

I had some free time last night, so here is what I made:

I could do better but I had to go to sleep a little :-)

 

[ti99iuc]

ahah, Fabrice, fantastic images you are creating! lovely!

[fabrice montupet]

Thank you Ciro ?

Maybe we'll get color information about this IR Joystick, if yes I will add them to the picture.

[Airshack]

On 1/28/2020 at 2:31 PM, sometimes99er said:

Holy Grail. Guess some people will find it hard to accept any level of MS involvement.

 

I'd really like to know the answer with certainty. Does this settle matters? Curious as to why this ends the debate?

[kl99]

6 hours ago, Airshack said:

I'd really like to know the answer with certainty. Does this settle matters? Curious as to why this ends the debate?

My guess is that this documentation was created for a machine not ending up the 99/4. The approach was withdrawn, otherwise we would have seen references to it in all other internal TI Basic Specifications. There were many computer projects being specced and started, which competed with each other. I did go through the TI Basic Interpreter Design Specification again, and I simply can't imagine having Microsoft coming up with a Basic Interpreter fitting nicely into a GPL interpreter which interacts into the Monitor System. So my guess is that this is a non GPL based TI Basic, maybe for the 9985, maybe for the Z80 or meant for a business machine or a 990 mini. We know for a fact people working in bouse on the 99/4 Basic Interpreter.

 

@Toucan can you make some photos of the first pages, alternate making a video going page by page through the document. Shouldn't take more than 15 min to browse through 100 pages and get it to youtube.

[fabrice montupet]

Looking at the photocopy of the IR Joysticks again and again, I thing that they look more like this:

 

Edited January 30, 2020 by fabrice montupet

[+acadiel]

I always wondered why they called it a “handheld unit”. Kinda makes sense now. Thanks for tracking these folks down, Bryan!

[Toucan]

See first post. I added the entire scanned document there in PDF form. Ooops. Guess I can't edit the first post. Anyway, here's the PDF attached to this message. bIt's 117 pages. Enjoy! Can't wait to see what other tidbits people dig up and share.

 

White Papers.pdf

 

One of the things you'll notice is how TI was big on the whole security system thing. Must have thought it a big selling point.

Edited February 1, 2020 by Toucan

[fabrice montupet]

Thank you very much Brian!

[ti99iuc]

wow! very nice scans ! Thanks Toucan!

[kl99]

thank you very much. reading now...

[+Ksarul]

Paragraph 3.4.1 on page 87 answers one of the known speculatory ideas on the origin of certain BASIC features: it was designed to meet the ANSI Minimal BASIC standard (with extensions) and the TI Standard BASIC (for upward compatibility). This explains why the full set of TI-990 BASIC tokens were defined, even though not all of them were used.

[Toucan]

1 hour ago, Ksarul said:

Paragraph 3.4.1 on page 87 answers one of the known speculatory ideas on the origin of certain BASIC features: it was designed to meet the ANSI Minimal BASIC standard (with extensions) and the TI Standard BASIC (for upward compatibility). This explains why the full set of TI-990 BASIC tokens were defined, even though not all of them were used.

Cool bit in there as well about the bubble memory devices as something that could be coming to the TI for storage. I read a lot of articles from the late 70s about TI being a big player in the bubble memory market, and people were wondering if the 99/4 was going to have that. Kind of neat since it was solid state with no moving parts.

Edited February 2, 2020 by Toucan

[senior_falcon]

Does anyone know anything about the specs for the 9985? I see it would have been clocked at 5MHz or maybe bumped down to 4MHz. Clearly a faster clock would have helped. Did the instructions take the same number of clock cycles as the 9900 or did they require fewer clock cycles like the 9995?

[Toucan]

32 minutes ago, senior_falcon said:

Does anyone know anything about the specs for the 9985? I see it would have been clocked at 5MHz or maybe bumped down to 4MHz. Clearly a faster clock would have helped. Did the instructions take the same number of clock cycles as the 9900 or did they require fewer clock cycles like the 9995?

I might be wrong on this, but I almost get the feeling after reading that if the 9985 would have worked out like intended, it would have been better in the 99/4 than the 9900? Meaning the 99/4 would have been a faster, cheaper, and better performing system. Is this true or am I wrong on that?

Edited February 2, 2020 by Toucan

[+acadiel]

Kind of interesting that they wanted a separate RAM expansion connector (separate from the I/O), and wanted to attach it to the VDP.  I’m guessing they decided later on just to incorporate the 16K altogether to the VDP.  
 

I wonder what the show stopper was for the IR remote.  Also, if anyone has pictures of the vestigial IR cutout from inside a 99/4, or speaker, or headphone jack, or volume slider.  Or better yet, motherboard placement on early 99/4 units for any of this stuff. 

[+OLD CS1]

I found it interesting they planed for the remote controllers to be "dockable" with an overlay for them to act as a keyboard.  They planned the IR keyboard before IBM aced it with the PCjr.

[kl99]

Thanks again for preserving this document.

Here are some notes so far.

 

Document Date

The White Paper has no date as such, or at least I didn't discover one yet.

It refers an Appendix A from July 24th, 1978, so it is at least as young as this.

The White Paper refers 08/31/78 as Date for a Figure (pdf page 61), indicating the Document is younger than this.

The White Paper Page 20 (pdf page 38) refers to a future Q1/79 date, indicating the Document is older than this.

So the date of the document must be somewhere in Q3/Q4 1978.

 

TI-99

so far I only went through the main document. There was no mentioning of the name TI-99.

 

GROM

The GROM is a main part of the TI Home Computer, already in the White Paper from 1978. I didn't see a reference to GPL in the main document.

On Page 11 (pdf page 18) GROM is set as TI software strategy.

 

TI Basic

TI Basic is already the name established for the BASIC Interpreter of the Home Computer in this White Paper from 1978.

 

UI Limits

When using Module Software for the TI I often wondered why the developers forces the user to switch between Keyboard and Joysticks input when staying in one game. The remote keyboard would have had eliminated this issue.

 

Software Module Library

On Page 18 (pdf page 28) a Software Module Library is referenced. My assumption is that the Review Module Library is the Software counterpart to exploit this peripheral.

 

Yet another White Paper

On Page 1 (pdf page 4) refers a White Paper from May 6th, 1978, describing the complete Personal Computer product line.

In contrast to that the CB Wilson White Paper is dedicated about the Home Computer product line.

 

Home Computer BASIC

On Page 15 (pdf page 22) it is said that the SR-60 and the SR-72 are programmable in supersets of the Home Computer Basic.

 

THE BASIC TUTOR

On Page 15 (pdf page 22) a GROM for The Basic Tutor is referenced. Afaik this cartridge was never released.

 

No Joystick Port

Figure 7.5 (pdf page 34) shows no Joystick port. I mean the digitial one that ended up being there in the production model.

 

64K RAM Expansion

Figure 7.6 (pdf page 35) shows a planned product RAM Expansion with 64K.

 

Product Specification Home Computer System

Appendix A - Product Specification Home Computer System - is actually a document we are aware of, at least in a more recent revision.
This shows the revision of July 24th, 1978. The Cyc contains a revision of the same document from May 29th, 1979.
The later revision already refers to the Home Computer as 99/3 and 99/4.

 

[Toucan]

1 hour ago, kl99 said:

Thanks again for preserving this document.

Here are some notes so far.

 

Document Date

The White Paper has no date as such, or at least I didn't discover one yet.

It refers an Appendix A from July 24th, 1978, so it is at least as young as this.

The White Paper refers 08/31/78 as Date for a Figure (pdf page 61), indicating the Document is younger than this.

The White Paper Page 20 (pdf page 38) refers to a future Q1/79 date, indicating the Document is older than this.

So the date of the document must be somewhere in Q3/Q4 1978.

 

TI-99

so far I only went through the main document. There was no mentioning of the name TI-99.

 

GROM

The GROM is a main part of the TI Home Computer, already in the White Paper from 1978. I didn't see a reference to GPL in the main document.

On Page 11 (pdf page 18) GROM is set as TI software strategy.

 

TI Basic

TI Basic is already the name established for the BASIC Interpreter of the Home Computer in this White Paper from 1978.

 

UI Limits

When using Module Software for the TI I often wondered why the developers forces the user to switch between Keyboard and Joysticks input when staying in one game. The remote keyboard would have had eliminated this issue.

 

Software Module Library

On Page 18 (pdf page 28) a Software Module Library is referenced. My assumption is that the Review Module Library is the Software counterpart to exploit this peripheral.

 

Yet another White Paper

On Page 1 (pdf page 4) refers a White Paper from May 6th, 1978, describing the complete Personal Computer product line.

In contrast to that the CB Wilson White Paper is dedicated about the Home Computer product line.

 

Home Computer BASIC

On Page 15 (pdf page 22) it is said that the SR-60 and the SR-72 are programmable in supersets of the Home Computer Basic.

 

THE BASIC TUTOR

On Page 15 (pdf page 22) a GROM for The Basic Tutor is referenced. Afaik this cartridge was never released.

 

No Joystick Port

Figure 7.5 (pdf page 34) shows no Joystick port. I mean the digitial one that ended up being there in the production model.

 

64K RAM Expansion

Figure 7.6 (pdf page 35) shows a planned product RAM Expansion with 64K.

 

Product Specification Home Computer System

Appendix A - Product Specification Home Computer System - is actually a document we are aware of, at least in a more recent revision.
This shows the revision of July 24th, 1978. The Cyc contains a revision of the same document from May 29th, 1979.
The later revision already refers to the Home Computer as 99/3 and 99/4.

 

Great observations. My gut instinct is that this is from September 1978, as they had a quarterly review in September of 1977, as mentioned in the 9/9/1977 drawing I have that says "Quarterly Financial Review".

Edited February 2, 2020 by Toucan

[+OLD CS1]

@kl99 GPL appears to be referenced as "H.C. Graphics Language" (Starting at PDF page 79.)

[senior_falcon]

According to the white papers, Plan B was to use the TMS9980 if the 9985 did not work out. According to some information found on the web:

"The TMS9980 is fully object-code compatible with the TMS 9900 microprocessors, albeit it's approximately 33% slower than the TMS9900 at the same frequency. The slower execution speed of the TMS9980 is exclusively due to its more narrow, 8-bit external data bus."

This performance is about the same as the more expensive 9900 running on the 8 bit bus. Anyone know why the decision to use the 9900 instead of the 9980 was made?

[Omega-TI]

19 minutes ago, senior_falcon said:

According to the white papers, Plan B was to use the TMS9980 if the 9985 did not work out. According to some information found on the web:

"The TMS9980 is fully object-code compatible with the TMS 9900 microprocessors, albeit it's approximately 33% slower than the TMS9900 at the same frequency. The slower execution speed of the TMS9980 is exclusively due to its more narrow, 8-bit external data bus."

This performance is about the same as the more expensive 9900 running on the 8 bit bus. Anyone know why the decision to use the 9900 instead of the 9980 was made?

 

That's a good question.  Usually it comes down to availability in the supply chain, and cost per unit which affects over all profitability.  IIRC correctly the 9900 was more expensive than what was originally planned, so I wonder it it was an availability issue?  

 

I know the "Spec Warz" of the early days were meant to impress people to drive sales, but I simply cannot bring myself to believe marketing had that big of a role in the process of shoehorning a 9900 in there.

 

But, with advertising like what you see below, it's gotta make you wonder...