Should apple have split into a Woz ethos company and a Jobs ethos company after Apple ii

[B3_B3_B3]

Should apple have split into a Woz ethos company and a Jobs ethos company after Apple ii?.....

The Apple ii's more open architecture appeals to me more than the Macintosh (Jobs Era?); and given the Apple III was seemingly nobbled by the insistence on no fan (sounds like a Jobs-ian idea) against Wozs better judgement , and the IIgs seemingly had its processor speed nobbled to avoid competing with the Mac, perhaps a split into two separate companies (AppleW and AppleJ or Apple and Pear...?)

[paulb]

perhaps a split into two separate companies (AppleW and AppleJ or Apple and Pear...?)

It is possible that Apple needed Apple II series revenues to shore up the bottom line, not entirely unlike Commodore who needed revenues from the C64 to keep the company going despite the introduction of the Amiga. Over time, the Apple II series will have been a steady revenue earner in markets where it was entrenched, like education, not entirely unlike Acorn's 8-bit range which also lasted into the 1990s. I would say that Acorn's revenues largely transitioned to the 32-bit range with the introduction of the A3000, however, whereas Apple's revenues may have shifted more decisively to the Macintosh.

Splitting Apple into two could have endangered the Macintosh part of the company in periods where revenues from the Mac could have been better, and it seems that this led to the conflict that saw Jobs leave. With Jobs gone, there would have been quite a bit less of the dogmatic Jobs ethos around in the company, and indeed the Mac started to encourage things like expansion and - gasp! - colour. So, there was actually a convergence of these two schools of thought in the absence of both of the founders.

The effect of a split on the Apple II series business is worth some consideration, I suppose. The IIgs seems to have been an attempt to update the line with something somewhat more credible in comparison to potential competitors like the ST and Amiga, these having the potential to siphon off sales in specialist roles in education - music and art/design respectively - where they were also competitors to Acorn's range in the UK. Maybe this business unit could fund itself from ongoing education sales and didn't need cross subsidisation from Mac revenues (and may not have got any such thing, either).

Might there have been opportunities for an upwards expansion of the 8-bit models, bringing them into competition with the Mac? Back in the day, much was made about Apple's rumoured but apparently genuine ARM-based Möbius prototype, perhaps by people who thought that the Archimedes only needed to be a faster Beeb with better graphics and who thought that Apple missed a trick delivering something similar. But like Acorn, they would have needed to provide a more sophisticated graphical environment to appeal to customers. The IIgs appears to borrow technology from the Mac to provide a graphical environment, and so such a separate company would have either needed to procure such technology from its sibling or develop its own.

[flibble]

Should apple have split into a Woz ethos company and a Jobs ethos company after Apple ii?.....

Yes,
wait ...
No!
wait ...
maybe?

[B3_B3_B3]

.....

Apparently the II was a useful cash cow? for a while as Mac was introduced, but if the IIgs was being nobbled speedwise that would seem to indicate the end of that 'need', plus the gs slow speed was mentioned unfavorably in reviews, and it was the last Apple II, which nobbling its speed might explain....

Interesting that an ARM was considered for a II descendant.

[paulb]

Apparently the II was a useful cash cow? for a while as Mac was introduced, but if the IIgs was being nobbled speedwise that would seem to indicate the end of that 'need', plus the gs slow speed was mentioned unfavorably in reviews, and it was the last Apple II, which nobbling its speed might explain....

There were reported problems with getting 65816 parts at higher frequencies. I think the IIgs fell within Jean-Louis Gassée's remit, and indeed the following article covers a dispute between Gassée and Bill Mensch (of the Western Design Center) about such 65816 availability issues, along with a treatment of Apple's strategy with the II series:

"C'est la Guerre", inCider, December 1989.

More on the dispute here:

"Chip Shots", The Apple IIGS Buyer's Guide, Winter 1990.

I know the 65816 has its enthusiasts, but there will have been few compelling reasons to use it towards the end of the 1980s.

Interesting that an ARM was considered for a II descendant.

Apple had all sorts of side projects going on.

[BigEd]

Nice finds in the archives, as ever, Paul!

Quoted for exposure:

The highlight of the keynote, however, was a shouting match between Gassée and Bill Mensch, designer of the GS' 65816 microprocessor. Mensch took exception to a comment Gassée made about a lack of available chips for a faster GS. Mensch reached the floor microphone and said he had chips Apple could use. Gassée responded quickly and emotionally that he was surprised to hear Mensch make that claim, considering the number of problems Apple had had dealing with Mensch and Western Design. Mensch tried to keep the argument going, but the ever-alert auditorium sound crew turned off the mike.

Not to be denied an audience with Gassée, Mensch tracked him down off stage, and the two nearly came to blows. I'm sure many jokes about the incident are circulating, but the best line I've heard came from my wife, who said, "You mean you went to a fight and a keynote address broke out?" Yes, and let's hope Gassée continues to think the II is worth fighting for.

Apparently. Gassée was trying to wrap up his speech when he said something about how Apple was unable to speed up the IIGS hardware because it couldn't find enough high-speed 65816 chips.

At this point. Bill Mensch pounced on one of the open microphones, held up some little packages and said something to the effect Thai, "I'm the designer of the 65816 and I have here three 12 MHz 65816s, If you order them we can ship." Mensch claimed that he had been trying to tell Apple, without success, that the high-speed chips were available. Meanwhile, Gassée said he hadn't heard from Mensch since die IIGS was first introduced.

Chips being as they arc, the existence of three ultrafast renditions in Bill Mensch 's

pocket doesn't necessarily mean that his company, Western Design Center, can manufacture the silicon in sufficient quantity to meet Apple's needs. Even if it could, Apple may have other reasons for not using the faster chip, including reliability issues and price.

I think I vaguely understood that Apple didn't want the IIɢs to be fast enough to knock any shine off the Mac, but that could be wrong. Wikipedia cites this article:

As for processor speed, the choice of a maximum speed of 2.8 MHz which was made at the time of design corresponded as much to a commercial choice as to a technical choice. From a technical point of view, the processor could easily go up to 16 MHz but the supplier, due to lack of suitable testing means, only certifying it up to 4 MHz, forced Apple to take a lower base frequency for the processor. to this certification. On the commercial side, the real problem was that at Apple, the new Apple IIGS risked competing, with a higher processor speed, with the Macintosh which, at the time, did not yet have a color screen and whose graphics resolution and especially the sound was far from equaling that of the Apple IIGS. This maximum speed of 2.8 MHz remained, throughout the commercial life of the Apple IIGS, a real shortcoming for the machine; which largely explains the success of accelerator cards which were offered by suppliers other than Apple (ZipGS and Transwarp GS for example).

(machine translation)

[paulb]

I think I vaguely understood that Apple didn't want the IIɢs to be fast enough to knock any shine off the Mac, but that could be wrong.

Maybe the initiator of this thread shares such sentiments. Here's another article conveying similar information to the French language article:

"10-The Apple IIGS"

The French article (which I have to take my time over due to my lack of French practice and its use of "literary" verb tenses) seems to suggest that the positioning of the machine was an issue as well. The pricing was quite high compared to competitors like the ST and Amiga 500, and the IIGS was perceived as an intermediate product for people moving up the Apple product range.

That said, it can be easy to overlook that they tried to endow it with quite a bit of Mac-like software technology, and that does raise the question of whether it could have attracted developers and made it a viable platform in its own right. I don't know enough about the platform's limitations to say. Obviously, the Mac became gradually more sophisticated over time, gaining cooperative multitasking with MultiFinder (like the way the RISC OS desktop works), although classic Mac OS never really took full advantage of the 68000 chipset.

As for 65816 speeds, some of the commentary seems very familiar, and I must have looked into this a while ago. There genuinely were problems with performance, and the TransWarp GS accelerator card was unable to provide parts faster than 8MHz, which is admittedly a speed-up from 2.8MHz.

When dealing with the 6502 and its relatives, people can be quite insistent that it delivers more performance per cycle than other processor families, which means that people do all sorts of weird comparisons and make all sorts of weird claims. Wozniak himself claimed the following about the 65816:

It should be available soon in an 8-MHz version that will beat the pants off a 68000 in most applications, and in graphics applications it comes pretty close. Some of the Macintosh people might disagree with me, but there are ways around most of the problems they see. An 8-MHz 65816 is about equivalent to a 16-MHz 68000 in speed, and a 16-MHz 68000 doesn’t exist.

So then people start claiming that the Commander X16 with its 8MHz 65C02 is as fast as the Amiga, which is only going to upset 68000 enthusiasts and get everyone going on about instructions per second again, never mind that MIPS figures were benchmark figures not literal instructions per second measurements. It should also be noted that Motorola were already producing the 68020 by 1985.

With regard to graphics applications, I am curious whether Wozniak would concede the following noted by Federico Faggin:

Intel, by the way, also had a segmented architecture, so it was not unique in that sense, but it was not the best choice. Motorola, with the 68000, chose a linear addressing and that was much more effective, because in those days the graphics applications were beginning to appear and they were beginning to be important, so we were basically not effective with our architecture in that space, and that was a major mistake.

More specifically:

Well, yeah, basically if you want to process an image, you have artificial boundaries that are created by these 64k segments. They are way too small segments. They should be much larger segments.

Maybe the 65816 allows linear access to the 24-bit address space. If I wanted to spend all day looking into this, I suppose I might have my answer!

I think the IIGS is an enticing system for those with a fondness for the II series, but even if they had got the 65816 running at 16MHz, where would it go from there? I suppose Acorn navigated many of Apple's headaches by introducing a new platform that substantially maintained compatibility whose machines were priced at a level that didn't really leave much room for the existing product range to grow upwards, even if the pricing was initially prohibitive for much of the existing product range's users.

[BigEd]

Indeed the '816 is quite handy at linear 24 bit addressing for data. Although the indexed addressing only has 16 bit offsets, so it depends how you program, maybe that's not good enough. As for programs, it seems they can't branch over a 64k boundary.

[Coeus]

With regard to graphics applications, I am curious whether Wozniak would concede the following noted by Federico Faggin:

Was Federico still at Intel at the time? At some point he left to design the Z80.

I have commented before that the design of the 8086 seemed to miss where the PC revolution was going, i.e. that on a personal computer people were much more interested in running bigger programs on larger data rather than running multiple copies of the same programs they had used on CP/M. But then the design of the 8086 must have pre-dated the IBM PC and maybe some of the home computers too.

Debating the virtues of the various processors those can sometimes seem like comparing apples and oranges. I have been looking at some of the Z80 BIOS code for Acorn CP/M recently and I thought how clumsy it is to have to load HL with an address just to be able to fetch a non-immediate byte from memory while the 6502 can do it in one instruction, and it is, but then when you want to loop over data, especially more than 256 bytes, the Z80 approach may be better. I am sure it all makes it harder to get a sensible comparison even with benchmarks.

On 65816 performance, I wonder what the real situation was there. I believe it was common to make ICs and then test them to see how fast they could be clocked and still work reliably and probably still is. That a few would go fast doesn't mean high volumes of fast chips could be made. Yet it seems Apple may also have had a convenient scapegoat.

On GUIs, people used IBM PCs running DOS. The machine probably had richer display options compared to many CP/M machines, and the fact that there was less diversity because of the cloning would have helped but still, most of the time, there was an interface quite like CP/M.

It is also worth noting that Apple did not invent the WIMP use interface. Neither did Microsoft. Both these companies saw this new idea at Xerox's Palo Alto research centre as, presumably, did people from MIT who went on to develop the X window system as well as some Unix workstation manufacturers who, initially at least, had competing windowing environments.

Back to Apple`s product line-up and the hypothetical split, where does the LISA fit into this?

Oh, and with regard to Apple and ARM, would we call their real success the iPhone, or maybe the iPod? This wasn't the first thing they did with ARM. I seem to remember something called the Newton? I think it is often the case when a product is disruptive to consider that it has "come from nowhere" but I bet that the truth is there has been a much longer period of development and maybe some failures along the way.

[paulb]

With regard to graphics applications, I am curious whether Wozniak would concede the following noted by Federico Faggin:

Was Federico still at Intel at the time? At some point he left to design the Z80.

Yes, this was a reflection on the Z8000 and how Intel had managed to make a success of the 8086, largely thanks to IBM, I suppose.

I have commented before that the design of the 8086 seemed to miss where the PC revolution was going, i.e. that on a personal computer people were much more interested in running bigger programs on larger data rather than running multiple copies of the same programs they had used on CP/M. But then the design of the 8086 must have pre-dated the IBM PC and maybe some of the home computers too.

Indeed. There were systems at the turn of the 1980s that ran multiple programs and supported multiple users, like the RAIR Block Box that got a few column inches in PCW at the time. One might have argued that for many users, just being able to run CP/M programs more cheaply and conveniently would have sufficed. But as we have discussed before, once it became apparent that bitmapped graphics would drive interest in computers, 8- and 16-bit architectures were always going to fall short.

It is also worth noting that Apple did not invent the WIMP use interface. Neither did Microsoft. Both these companies saw this new idea at Xerox's Palo Alto research centre as, presumably, did people from MIT who went on to develop the X window system as well as some Unix workstation manufacturers who, initially at least, had competing windowing environments.

Yes, the genesis of the graphical desktop has a long history with PARC being influential, and numerous players making their own contributions, typically neglected by people pushing the simple narrative as if it helps them elbow their way onto the trophy winning team's hall of fame photograph.

Back to Apple`s product line-up and the hypothetical split, where does the LISA fit into this?

Well, Jobs had been ejected from that project and thus probably did his level best to get it cancelled in favour of the Mac, despite it being a more ambitious and PARC-like endeavour. I'm not sure whether it would have made a return with access to better hardware, though.

Oh, and with regard to Apple and ARM, would we call their real success the iPhone, or maybe the iPod? This wasn't the first thing they did with ARM. I seem to remember something called the Newton? I think it is often the case when a product is disruptive to consider that it has "come from nowhere" but I bet that the truth is there has been a much longer period of development and maybe some failures along the way.

Yes, there's a lot of history involved. For some background on the Newton, the architecture choices, ARM adoption and StrongARM, take a look at this interview:

"Oral History of Allen Baum"

[B3_B3_B3]

I think I vaguely understood that Apple didn't want the IIɢs to be fast enough to knock any shine off the Mac, but that could be wrong.

Maybe the initiator of this thread shares such sentiments
......

He does-ish: All I read seemed to suggest that hence my notion that the Woz (Apple II) and the Jobs style (Mac) sides could split into 2 companies rather than Woz's IIgs project get permanently nobbled with a slow processor.

Also, were Apple not big enough then to help Mensch get a fab that could reliably make/test faster 65816s.



NB 16 bit internal architecture microprocessors seemed underwhelming to me, if abandoning 8bits, might as well jump straight to a nice (for the programmer) 32 bit internal programming model*, a la 68000/68008 and whatever bus width is affordable. *I wonder why Woz never wrote a Sweet32 for the 6502.

[paulb]

He does-ish: All I read seemed to suggest that hence my notion that the Woz (Apple II) and the Jobs style (Mac) sides could split into 2 companies rather than Woz's IIgs project get permanently nobbled with a slow processor.

Here's a better reference to that Möbius project:

"The ARM Processor or The RISC for the Rest of Us"

In the last interview I referenced, I think the interviewee mixes the ARM up with the Am29000 in recalling a project to make a RISC-based accelerator for running Mac software, which would be easy to do given that the Am29000 was used quite a bit for accelerator cards and, later on, in LaserWriter products.

Also, were Apple not big enough then to help Mensch get a fab that could reliably make/test faster 65816s.

It was noted in one of the sources I found that Sanyo helped WDC sort out their issues. I imagine that this kind of thing happened quite a bit when processor designs struggled to meet their performance targets.

[helpful]

NB 16 bit internal architecture microprocessors seemed underwhelming to me, if abandoning 8bits, might as well jump straight to a nice (for the programmer) 32 bit internal programming model

Luckily Sophie and Steve clearly thought so too, and we got ARM