Most of Macintosh 128K's logic board was almost an ASIC!

While digging up info on the LSI LOGIC chip used on the Apple IIe Card, I came across a book about LSI which provided some very interest historical info I had never read anywhere else. Amazing what could have been!

One day Doug came in with a copy of EE Times. It had a photo of Steve Jobs and on his desk was the Mead-Conway textbook! So I called Steve and let him know that VLSI was the leading practitioner of Mead-Conway in the entire universe (Steve likes that kind of talk). The conversation led to VLSI's very early involvement in the Macintosh computer.​
The early people at VLSI came from the same group at Xerox PARC that Apple courted for the Mac team. One time Apple and VLSI were trying to hire the same engineer and I got a call from Jobs. He reminded me how many ROMS and custom ICs Apple might buy from VLSI, and let on that he was personally interested in this particular engineer. The bidding stopped that day. We got no future draft choices, but Apple is still a very big customer of VLSI.​
DOUG FAIRBAIRN​
Our next milestone was the Bagpipe Project. It was the "lost summer of 1982." In one of our early classes, Burl Smith from Apple got excited about ASIC. He was the lead hardware designer for what was to become the Macintosh, and he convinced Apple management that they should pursue Mead-Conway technology to implement everything except the CPU and memory.​
We had a handshake agreement, and we started the design with absolutely no formal arrangements. In fact, the day Burl Smith showed up, he said that since we were doing the ASIC as a side deal, we could throw out the existing spec and add more cool features. The day we started, we were two weeks behind!​
We busted our pick for the next six months working on this chip—we were fighting power and size and everything. It had a very fast, dynamic memory control on chip and logic to control the video, the sound, the mouse, and the keyboard. We had nearly everybody at VLSI working on the project overtime.​
As the project grew, we brought in more and more people because of the business potential. But just as we got the silicon almost working, Steve Jobs canceled the project. We learned an incredible amount from the experience, but it was a low point because we had just spent our hearts and souls on the project and it got cancelled. Apple could have introduced a fully integrated Mac based on this chip.​
Silicon Destiny: The Story of Application Specific Integrated Circuits and Lsi Logic Corporation, pp197~198
 
  • Like
Reactions: YMK

retr01

Senior Tinkerer
Jun 6, 2022
2,473
1
796
113
Utah, USA
retr01.com
So, how come Jobs canceled this exciting project? :) Did he lack hindsight? Was he pressured to do so by Arthur Rock and Mike Markkula? What was the time frame? Might it have been between 1980 and 1984?

Is it game afoot? LOL. A lot of things that happened in Apple were usually "game afoot!"

Edit: Ah, around 1982.
 

JDW

Administrator
Staff member
Founder
Sep 2, 2021
1,577
1,373
113
53
Japan
youtube.com
So, how come Jobs canceled this exciting project?
Absolutely impossible to know. Andy Hertzfeld MIGHT know, but even though Andy was close to Jobs, it wasn't THAT close.

It's unfortunate Burrell Smith suffers (suffered?) from schizophrenia, as he would be the man to ask that question.

Steve was never driven by cost, so I cannot speculate if he canceled it because a custom ASIC would cost more.
 

JDW

Administrator
Staff member
Founder
Sep 2, 2021
1,577
1,373
113
53
Japan
youtube.com
More specifically, here's the exact excerpt from the link you provided...

Unfortunately there were ‘low level’ problems with the design and the part never worked properly before the project was dropped as being too expensive and risky.
There appears to be a huge disconnect between Andy's assessment ("the part never worked properly") and that of Silicon Destiny (" just as we got the silicon almost working...") The truth probably is that the decision to cancel was based solely on "risk" while they reflected on some of the problems getting the ASIC to work. The issues very well might have been fixed by the time Steve cancelled the chip. I guess Apple must have thought the multi-chip approach would be more reliable (which isn't necessarily the case because the more parts you have, the more room for failure). Putting many chips on a single VLSI chip was still a fairly new thing back in 1982, so when we consider that, the worries over risk make more sense.

The Apple IIgs (which uses the MEGA II chip) Apple IIe Card (which uses the newer Gemini chip) shows that Apple learned that the risk was worth advancing technology, and the LSI Logic Corporation was at the heart of it.
 
  • Like
Reactions: retr01

retr01

Senior Tinkerer
Jun 6, 2022
2,473
1
796
113
Utah, USA
retr01.com
Yeah. In that same paragraph, "This board has the IWM (Integrated Woz Machine), the custom floppy controller chip that Wendell Sander had designed as an improvement over Woz’s original Apple-II floppy disk controller." So, they decided to use the IWM and improve it.

Hmmm, "The ‘MacMan’ logo has been dropped." MacMan logo? Interesting. I think this one?

1664419056430.png