• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

What if Woz got what he wanted for a 8MHz version of Apple IIGS

Gorgonops

Moderator
Staff member
Sadly too, thinking that a better and faster processor could have made any of the Apple II variants into a contender that could compete and outclass emerging technologies from other manufacturers is rather like thinking if you simply put a bigger engine in a Trabant, you could compete with Ferraris at Le Mans.

Of course when you're talking about the Apple IIgs you can't overlook the fact that the original II architecture was so tightly integrated and non-extensible that backwards compatibility had to be implemented essentially by building the whole Trabant into your new racecar; they share the same engine but to run older software you have to throw a lever that lifts the racing slicks off the ground so the original 13" donut wheels can cart you slowly around town through a copy of the original transmission. (Wacky column shifter and all, and of course the engine is rev limited in this mode so you don't immediately turn all your gears into sawdust.)

I've never used a IIgs with an accelerator, I guess, but I think it is worth reiterating that at the 2.8mhz the IIgs actually shipped at its performance seems very much in-line with the 8mhz-ish Turbo XT clones that were becoming very cheap entry level options by 1986. The IIgs has a nicer sound chip, but most of the "best" games for a IIgs were also available in versions that ran just fine on a Tandy 1000 which could be had significantly cheaper, especially if you were wanting a hard disk with that. (And they were probably available on the Tandy 1000 first, because so many more of them were sold.) Even if we argue that an 8mhz IIgs is more comparable to a 286 class machine (which is debatable, especially if you don't overlook some of the severe bandwidth limitations embedded in the IIgs architecture, well, keep in mind that Compaq introduced the first 80386-based PC the same month the IIgs came out. The whole point of the Macintosh II was to compete with that. The IIgs is completely out of that league.
 

Unknown_K

Well-known member
The only plus side of a IIgs compared to a Tandy 1000 or early 286 is ram expansion to 8MB but nothing really uses that much RAM on the IIgs. An accelerator, SCSI card, and max RAM make GS/OS run well but there isn't much software to take advantage of it. Back in the day those upgrades would have cost a bunch of cash (not cheap now either).
 

Unknown_K

Well-known member
I like the original Tandy 1000A and the 1000HX I own. For playing early DOS games from floppy that support Tandy sound and video they rock.
 

Gorgonops

Moderator
Staff member
Tandy 1000SX DIE! The one thing it wouldn't do, that or support a parallel port for a dongle. DIE! PC Barely Compatible POS. 🤐

Pshaw, the Tandy 1000 rules. Of course I mostly say that because of all the utterly pointless work I did to hotrod the most ridiculous versions of them beyond recognition.

The only plus side of a IIgs compared to a Tandy 1000 or early 286 is ram expansion to 8MB but nothing really uses that much RAM on the IIgs. An accelerator, SCSI card, and max RAM make GS/OS run well but there isn't much software to take advantage of it. Back in the day those upgrades would have cost a bunch of cash (not cheap now either).

Any 286 with a standard ISA slot can go to 15+ MB Extended (plus up to 32MB EMS Expanded) RAM with the right memory card, but let's not be ridiculous. A Tandy 1000 is indeed stuck hard at 640k unless you count EMS RAM... which frankly would be a legit option if you really wanted to run real-mode software roughly comparable to GS/OS. (I have 1MB of homemade EMS in my Tandy 1000HX and it works with Windows 2.x and GEOS software; that combined with conventional RAM gives it half again more memory as a ROM1 IIgs with a fully populated Apple RAM card.)

Perhaps it's worth noting that Apple themselves only sold the one 1MB capacity RAM card for the IIgs, which meant you needed to go a third party if you wanted more than 1.25MB (or 2MB in a ROM3) of your IIgs, which seems a little telling. Back in the day I did desktop publishing using GEM based software on a Turbo XT using just conventional RAM and a swap file, and I'd rate the user experience very much in the same ballpark as a WYSIWYG GS/OS program. Which is to say, yeah, kinda sucked, but you could get by. A Mac Plus with 4MB of RAM would be better than either, not because its CPU is meaningfully different but because CGA resolution *sucks* for any kind of GUI.

(Chuck a VGA card in the Turbo XT and run it in monochrome 640x480 mode to keep the RAM/CPU usage down and it's not a contest anymore, it's a better environment than the IIgs for productivity software. Or heck, maybe even go with Hercules Monochrome; it's not square pixel either but it's close to the same resolution as a Lisa, also a huge improvement.)
 

Cory5412

Daring Pioneer of the Future
Staff member
- 1990 - 68020 LC/IIe Card combo continues support of massive 8bit K-12 software base and IIgs is canned.

To add here, the IIgs was on sale until December 1992. The Apple IIe was on sale until November 1993.

However, in reality, the primary reasons to buy them can be distilled to:
  • You're a K12 school and you can buy three or four Apple IIes for each LC+IIEcard, and your software is predominantly 8-bit Apple II stuff anyway.
  • You're an Apple II enthusiast and want a new one even though it's objectively worse than most other available computers.
  • You need a computer and your budget is that the computer needs to cost like $600 or less, and work with your existing TV as a display.
  • You're using some type of Apple II software and need/want to continue using it longer but your original machine has developed problems and/or you want the logistical benefits of a newer machine, e.g. bigger floppy, smaller case, whatever.
I saw Apple IIs in K-12 (as a student) a few years after that and I have to presume they actually lasted way longer than I think they did, mostly because they were reasonably reliable little taskboxes, and because lots of K-12 schools treated each new computer purchase from basically 1970 to 2000 as a capacity expansion, which sometimes meant you had a school where there was a Windows domain and Windows computers, Services for Mac was enabled and there was a lab or two of System 7/8/9 Macs, and then there was yet another lab or fleet of Apple IIs doing things like Oregon Trail and Munchers.


Had it not for Apple II continued success, Apple would have gone under long time ago.

Between this and another comment about pricing -- one of the things I think is genuinely most impressive about Apple is that, in reality, they've been incredibly long lasting compared to "everyone else from that era who wasn't IBM or HP."

I think Apple's ability to hop platforms like that, and parlay the financial success of the Apple II into an ability to try out a handful of next-generation platforms is exactly what ensured their success long term, and, in turn, has ensured the continued success of the Mac as a platform, through several transitions.

Apple continues to parlay the success of existing products into their continued success as a corporation, although the way that manifests in the last 20 years is more additive than the way it manifested in the 1980s. For example, newfound success at the Mac gave Apple the power they needed to start the iPod which in turn bolstered Mac sales and both of them let the iPhone come into existence, so-on and so-forth.

It's always interesting to think about what a particular alt-history could look like. "Apple builds a seprate 68516 16-bit computer unencumbered by 8-bit Apple II compatibility" is one I do like, but it ultimately ends up looking like a completely new platform that, due to its unrelatedness to the Apple II, can last slightly further into the '90s as a low-cost alternative to the Mac, and probably ends up with several of the same problems the existing real-world Apple IIgs has, because it would be squeezed in between the existing "mostly good enough" 8-bit platform and the Obviously The Future Mac.

It's tough to tell where the Mac, and perhaps all of Apple, would be without DTP. Nominally, the Mac was meant as an information appliance. I can see Apple shopping it out to IBM/DEC/HP et al as a front-end application/processing platform for data stored in minicomputers and mainframes, or perhaps Apple trying to see if minicomputers and mainframes is the next place they themselves should go.
 

AndyO

Well-known member
Of course when you're talking about the Apple IIgs you can't overlook the fact that the original II architecture was so tightly integrated and non-extensible that backwards compatibility had to be implemented essentially by building the whole Trabant into your new racecar; they share the same engine but to run older software you have to throw a lever that lifts the racing slicks off the ground so the original 13" donut wheels can cart you slowly around town through a copy of the original transmission. (Wacky column shifter and all, and of course the engine is rev limited in this mode so you don't immediately turn all your gears into sawdust.)

Oh, yes, that's the analogy I was looking for!!

I certainly recognize the nostalgia factor in the line of Apple II systems, and even the enthusiasm for re-litigating the whole 'Jobs used the Apple II to fund the Mac' debate too. But the reality is that for all Woz's undoubted genius, there was a limit to how far that technology could actually go, and how much longer Apple could have sold it when set against the rapid evolution of systems from everyone else.

It seems to me that based on the @Cory5412 commentary above, the II series lasted remarkably well even if for ever-narrowing reasons. Trying to stretch it even further doesn't seem like it would have been the great idea one might have hoped, and would likely not have made Apple much in the way of financial reward for doing it.

It's tough to tell where the Mac, and perhaps all of Apple, would be without DTP.

I doubt it would be anywhere. The market for Macs was very limited until the DTP revolution put design, layout and publishing tools in the hands of people who saw those, and the Mac's WYSIWYG interface as a natural springboard. Up until then, the Macs I knew of or saw in use were almost entirely niche-market toys for relatively affluent word processor users who might also dabble a bit in graphics doodling.

Certainly in my area of work, there was a huge amount of pressure on Mac users to 'conform' based on the relatively lower cost of IBM/compatibles which had a much wider software base. Introduction of DTP, fueled greatly by the arrival of the LaserWriter, gave the Mac a real-world purpose that justified it as an expense in comparison to a PC. Without that, I doubt the user base would have notably expanded.

Admittedly, that's from my perspective in the UK, where Apple was not even remotely as widely represented in the education sector (where I worked) as it was in the US. I must say that as one of the fairly early pioneers of DTP, when I was called on to teach staff and clients about it and what it could do, I could see how it fueled a lot of Mac-centric system purchases, even peeling users away from PCs.

Up until then, it seemed to me that Mac users were a rather isolated and scattered group - not a common recipe for success!
 

Gorgonops

Moderator
Staff member
It's always interesting to think about what a particular alt-history could look like. "Apple builds a seprate 68516 16-bit computer unencumbered by 8-bit Apple II compatibility" is one I do like, but it ultimately ends up looking like a completely new platform that, due to its unrelatedness to the Apple II, can last slightly further into the '90s as a low-cost alternative to the Mac, and probably ends up with several of the same problems the existing real-world Apple IIgs has, because it would be squeezed in between the existing "mostly good enough" 8-bit platform and the Obviously The Future Mac.

It's hard to imagine a world where building a "brand new" 65C816-based computer from whole cloth that didn't include some level of backwards compatibility would ever have made a lot of sense. Frankly the one selling point the chip has is its 6502-emulation mode; because of the architecture weirdness that came along with that it's a far more awkward chip to program (especially for high-level language compilers) than the 68000 and by the mid-80's the 68000 had come down in price enough compared to it's 1979 introduction that a 65C816-based computer wouldn't really be able to undercut a 68000 in parts cost either. The closing paragraphs of this Byte Magazine review of the IIgs makes some specific observations about how the 68000 and, for that matter, the 8086, outclass the 65816 when it comes to programming capabilities. On its own it just doesn't offer that much; I think even the Super Nintendo, the other well-known thing that employed it, used it mostly because Nintendo designed the system as a direct outgrowth of the original NES, not because it was "better" than other options.

If Apple had been willing to sell products at the same kind of margins as, say, Atari and Commodore (which already had the Atari ST and the Amiga, respectively, on the market for more than a year before the IIgs was introduced) they could have easily flooded the market with the equivalent of a Macintosh 512ke or Mac Plus for well under a thousand bucks in 1986. The closest equivalent to a Mac 512k, the Atari 520ST bundled with the 640x400 monochrome monitor, could be had for less than $600 the day the IIgs was introduced at $999 without even a floppy drive included. (And less than half what a Mac 512k was going for.) In some alternate universe if Apple's highest priority had been to spike the market share of the Macintosh they certainly might have been able to do simply by cutting the price, there was no need to come up with some "third platform" to sell more computers. The risk there would have been, of course, that until the Mac II came out there wouldn't have been anything to "upsell" customers to; the death of the Lisa made that an obvious problem.

(And dirt-cheap Macs also probably would have significantly undercut Apple's cachet as a "premium" brand, which frankly was the only thing keeping them afloat: in a sane world by 1985 there's no way an Apple II should have cost just as much as a PC clone with four times the memory, but miraculously people were still willing to pay that for one. They were probably smart milking that for everything it was worth, but it also results in this schizophrenic marketing structure where the backward-looking products cost as much or more than the forward looking ones to make.)

There was another recent thread asking why the Macintosh never got "real" multitasking, where the answers pointed out the original sin of the Mac's birth: it was pretty much designed as a one-off appliance product with no real allowance for it to ever organically evolve and grow. It took three years for Apple to undo enough of the damage from those original decisions to bring the Macintosh II to market with color support. (And a few more years to iron out remaining issues like 32 bit cleanliness, etc.) In some alternative universe it would have clicked that this whole "appliance" concept was stupid and self-limiting sometime in late 1983 and the work would have happened to instead make what they were building the foundation for future growth. In this world the original Macintosh would have shipped with 512k (the way Apple jacked up the price $1000 at the last minute there's no reason why they couldn't have done it) and less work put into it to minimize code size, and work would have immediately proceeded on color and other extensions. Assuming all went well Apple probably could have introduced simultaneously in late 1986 the 68020-powered Macintosh II for the top end and a 68000-powered color-capable machine roughly equivalent to 1990's Macintosh LC as a midrange option with a list price starting around $2000, with the basic black-and-white 512k/1MB toaster dropping to a street price around a thousand bucks. (And probably eventually phased out in favor of configurations of the midrange machine paired with a mono monitor.) A color 68000 machine is not a crazy idea; there are documents suggesting the LC was originally targeting that CPU but was changed to the 68020 because Color Quickdraw ended up with some code dependencies on the later CPUs. And it wouldn’t have cost much more than the IIgs to build, especially if it’d used a 512x384 monitor instead of 640x480.

(Apple might have actually been able to pull this off in our universe if Apple hadn’t been left in such a shambles in 1985, with Jobs’ departure, the failure of the Lisa, etc.)

This scenario of a late-1986 "LC equivalent" as their ‘new‘ platform makes way more sense than the IIgs ever did as anything more than a sop to keep the Apple II‘s hard-core fans hooked for one more round. It'd still be more expensive than its 68000 competitors (a combination of the "Apple Tax" and the reasonable observation that a non-innterlaced color 384-480 line monitor would probably end up retailing for a couple hundred bucks more than an NTSC frequency one), but it would immediately have the software base of the original Toaster plus an growth path to the fully 32-bit Macintosh II and onward. The "old" software it ran would also be mouse/GUI based so you wouldn't have that jarring disconnect between the "native" and "emulated" modes you get with the IIgs, and, heck, if someone had already wasted the time coming up with the MEGAII they could have thrown together the equivalent of the IIe compatibility card so schools could have started migrating off the old platform five years ahead of when they actually did. Look at the prices PC compatibles were going for in that 1986 Byte Magazine and ask what looks like a better value, the IIgs we got for almost the same price as similarly equipped 286 AT clone, or a 12-16mhz LC-like machine able to run the whole existing library of Mac software plus new color programs for a few hundred bucks more? I know my answer.
 

Unknown_K

Well-known member
Apple got lucky with DTP and with early video editing where Apple hardware ended up being the standard which software companies wrote for and which printing companies then used for actual printing. The hardware attracted the software which grew the hardware base which is why Apple coasted a long time before almost going bankrupt.
 

NJRoadfan

Well-known member
For those interested in the development of the Apple IIgs, I highly recommend reading "The Apple IIgs Book". It goes into some of the development background of the machine. IMHO it kinda romanticizes some of the development choices but still a good read.

CPU Speed
The IIgs was never going to ship with a 8Mhz CPU in 1986. WDC had problems getting samples running much faster then 3Mhz at the time. Even by 1989 they had trouble with higher clocked CPUs as the Transwarp GS shipped clocked at around 7Mhz. AE had to "patch" the CPU using GAL logic in order to fix bugs. Later production cards and the ZipGSX ran at 8Mhz. I don't know much about the 90s development of the 65C816, but the CPU was finally stable (bugs fixed) when the modern CMOS versions arrived in the mid 90s. They were "rated" for 14Mhz, but usually ran at 20Mhz without a problem. The 14Mhz rating was due to limitations of WDC's testing equipment. This is the revision found in the CMD SuperCPU upgrades for the C64.

Platform Limitations
The biggest one is the 1Mhz bus limit. Any thing that hits the I/O space will throttle down the system to 1Mhz, including video. The VGC's frame buffer is stored in "slow" RAM. The system was designed as an 8-bit Apple II and extended to 16-bit using a 16-bit CPU and some glue logic called the "Fast Processor Interface". Anything on the CPU side (pretty much memory) ran at 2.8Mhz. The max theoretical transfer speed over the slot bus is 1MB/sec using tricks like DMA. Even a lowly IBM PC beats this with its 4.77Mhz bus speed.

The 65c832
WDC had put out a datasheet outlining the "what if?" of a 32-bit ISA. It didn't get much further then that. Long term it would have been hard to justify this. The only folks clamoring for it would be Apple IIgs users with a large base of GS/OS and Toolbox friendly software. It's the only way Apple could have made a clean break from the wacky 8-bit Apple II architecture. The path Apple chose with the LC IIe Card was the correct one in many ways (although it would have been neat if it was a IIgs on a card). The only way a "modern" Apple II could exist is the same way things were done on the PC, a hardware based virtual machine to run old software on a modern OS. Apple wasn't going to do this with the Macintosh hanging around!

Apple's Multimedia Intentions
This one is quite baffling. It involves the Apple II Video Overlay Card, a product that really shouldn't have existed (unlike, say the Ethernet card that was canned at the last minute). The board itself was VERY expensive to design and sell. It used a ton of VLSI chips and the most modern of manufacturing methods for 1990. Going by the promotional materials, Apple was clearly trying to target the educational group and check off the "see we have a genlock too!" box against the Amiga and Atari STe. Well, except those machines came with all that built in.

The end goal was likely to base the Apple IIgs as a heart of an interactive LaserDisc system. Hypercard would power the application, the VOC would provide UI overlay, while a serial controlled LaserDisc player provided the video and pretty menus. All for a ridiculous price that no school would pay for. The only other thing the VOC was good for is adding Apple IIgs SHR video modes to the IIe, or 400 line (interlaced!) output to a IIgs. Otherwise, pretty useless unless you want to watch TV on your AppleColorRGB monitor.

My high school did have an Apple IIgs controlling a LaserDisc player for interactive educational work. It didn't use the VOC, but something way more unexpected. The machine booted into AppleWorks and used Time-Out! macros to control the LaserDisc player!

Apple's "Investment" in the IIgs
In hindsight, it was remarkable the machine lasted as long as it did. It took Apple TWO years to release a proper operating system for the machine. It shipped with the Apple II DeskTop (MouseDesk) as a GUI for the first year or so. It later got a IIgs native Finder in 1987. All this was based on the ProDOS 16 core, which was basically the 8-bit ProDOS extended with the IIgs Toolbox. Disk access was DOG SLOW.

When GS/OS showed up in 1988, the native 16-bit disk code was much faster and they finally added a disk cache. Apple poured a ton of money into the OS on this machine for some reason. It many ways it was more advanced than classic MacOS (software loadable file system translators among other things). I guess that's what happens when you learn from your mistakes. Reading period Usenet posts, its clear that people were very unhappy with how long it took Apple to release all this stuff.

The same folks were also very unhappy when the ROM 3 showed up. It was pretty clear the machine needed more CPU horse power and a built-in SCSI controller, neither of which it got. The only thing the ROM 3 solved was the lack of RAM. A base IIgs with 256k (less actually available for programs) is pretty useless and only a handful of native software can run with that. Almost all games needed at least 512k, which Apple later made standard by including a RAM card.

To end things, Macintosh users should be very grateful to the Apple II line. It was Apple's cash cow that subsidized the Macintosh for at least the first 3 years of it's life..... maybe longer.
 

Gorgonops

Moderator
Staff member
CPU Speed
The IIgs was never going to ship with a 8Mhz CPU in 1986. WDC had problems getting samples running much faster then 3Mhz at the time. Even by 1989 they had trouble with higher clocked CPUs as the Transwarp GS shipped clocked at around 7Mhz. AE had to "patch" the CPU using GAL logic in order to fix bugs.

Thank you for the reference on this issue. I was pretty sure this was the case that the popular myth that the 2.8mhz shipping speed was all a grand conspiracy to "cripple" the machine in favor of the Macintosh was overblown, but didn't see a reference immediately that said definitively what CPU grades were actually available (and reliable) in 1986. Considering development of the CPU was kicked off in 1982 and it didn't really get reliable at double-digit speeds until the mid-1990's I think that pretty much says it all.

(After all, Intel went from the 286 to the Pentium Pro in roughly the same timeframe...)

The system was designed as an 8-bit Apple II and extended to 16-bit using a 16-bit CPU and some glue logic called the "Fast Processor Interface". Anything on the CPU side (pretty much memory) ran at 2.8Mhz. The max theoretical transfer speed over the slot bus is 1MB/sec using tricks like DMA. Even a lowly IBM PC beats this with its 4.77Mhz bus speed.

Honestly I think it's a little hard to argue that *any* of the IIgs is actually "16 bit" outside of the CPU registers. (Unless I'm forgetting that the VGC has 16 bit RAM access or something; the Tandy 1000 has that, ironically enough.) The "fast" part of the IIgs isn't any "wider" than the slow part, the CPU only has 8 data bus pins and there was never any roadmap to expand that. (Even the vaporware "32 bit" 65832 kept an 8-bit data bus!) This makes it, ironically, the opposite of the 8088, which was an externally cut-down version of a fully 16 bit wide predecessor.

There's actually a whole crazy rabbit hole you can go down here. People sometimes argue that the Motorola 6809 used in the humble Radio Shack Color Computer should have counted as a 16 bit CPU. I personally think the case for that is a little overblown but its architectural similarity to the 65816's "16 bit" extensions are eerily similar (compare the available registers), it has the same width data bus, and it may actually be better at natively supporting position-independent code. The only thing it's really "missing" is built-in paging registers to allow greater-than-64K memory addressing internally, but that's not a necessary qualifier for a 16 bit computer. The DEC PDP-11 is generally considered the definitive 16 bit computer and its original architecture only allowed for 64K addressing. A couple patches that eventually expanded that to 22 bits happened over the years, but if you measure "bit-ness" by, say, how far you can do a relative jump then the last PDP-11 is the same as the 6809, 16 bits.

I'm willing to be kind and call the 65816 a 16 bit CPU "architecturally" but there's a part of me that thinks it kind of deserves an asterisk. If this fight really went to the floor I'd say the case for it being a "real" 16 bit CPU, by whatever you arbitrary define "real" to mean, might be significantly weaker than the 68000's claim of being a "real" 32 bit CPU, vs. a 16 bit CPU with 32 bit aspirations.
 
Last edited:

Ajaxermd

Well-known member
Oh, I really hope not!

I think the point you're making regarding the potential for performance in the IIGS is an interesting one, but I don't think it was an opportunity missed.

To my mind, the real issue here is that the Apple II line was destined to come to an end anyway, if for no reason other than that in designing for backwards compatibility, the eventual product would turn out to be overly constrained in comparison to new technologies from other manufacturer/vendors.

The Apple II range of systems were excellent in their day, but they came to an end because they had to, else Apple were painting themselves into a corner, and in the context of the rapid development of systems and turnover of businesses of the time, may well not have survived. As such, what Woz wanted was likely and almost invariably extremely clever, but far less good business sense. The Apple II range survived and sold based on the vast amount of software already in circulation, and as long as it remained backwards compatible, new software would have largely been that too. Even though Jobs could have done with a kick up the backside with a major dose of maturity, he was right that the time had come for something new in a new direction.

I have a IIc, in fact two of them, but even by the time that launched, it was creaking under the millstone of compatibility with much earlier models in the range.
DOS machines continued to evolve into Windows. The IIGS was really a new platform that could have evolved in parallel with the Mac. However, I can see why that would be confusing for Apple brand marketing.
 

AndyO

Well-known member
DOS machines continued to evolve into Windows. The IIGS was really a new platform that could have evolved in parallel with the Mac. However, I can see why that would be confusing for Apple brand marketing.

DOS machines did indeed evolve into Windows systems, and there was a reasonably articulate flow to that, which was not least the result on building on the earlier platform, rather than the constraints of it.

There's no doubt the IIGS could have been a new platform that broke away from the prior II family, and perhaps it should have been, but a parallel evolution to the Mac doesn't seem like it would have worked, simply because Apple would mostly be in competition for customers with themselves. And of course the IIGS conflicted with the ideal Jobs had of a closed 'appliance'. I think the processor was amongst the least of the IIGS's problems!
 

tetsujin

New member
I am always curious about this question... It frankly irks me that the claims about the IIgs, that it could have been so wonderful if only Apple hadn't held it back, are spread so freely and accepted so unquestioningly. I love the Apple II but it was a really old platform by the time the IIgs came out. With early home computer platforms being so deeply rooted in bare-metal programming, they were architecturally pretty inflexible. PCs grew past their 8088 roots but it took years of awkward half-measures and various OSes before the platform as a whole finally moved to a 32-bit OS and began to leave the old limitations behind.

It seems to me that regardless of the CPU speed, the system is bottlenecked by large portions of the system running at 1MHz and an overall architecture that's not designed to help the system overcome that bottleneck in any meaningful way. (i.e. no coprocessors or blitters, no hardware display scrolling or sprites - I think it couldn't even page-flip video memory in "super-hi-res") so graphics performance especially was very limited by that 1MHz bottleneck. It had RAM shadowing, at least, and this led to techniques like PEI-slamming to update the display from a double-buffer in fast RAM.. But it was a hell of a bottleneck.

Granted I can only say this with the benefit of hindsight, but Apple pulled off an amazing feat with the Macintosh: they leveraged their prior success with the Apple II and made a non-PC platform that could survive the 1990s. It's hard to imagine they could have managed two such platforms.
 

NJRoadfan

Well-known member
Apple was able to maintain two platforms for 9 years because..... one was paying for the other. The Macintosh was not a big money maker in its early days. The Apple II line paid the bills until the late-80s and kept the company awash in R&D funds. Apple also developed "family" peripherals to be shared among the platforms too. Remember that ADB was introduced on the IIgs and later adopted on the Macintosh.

Apple II users weren't asking for a lot. They wanted a revision that came with goodies like SCSI and 1.44MB floppy support out of the box. A little bit of speed would have helped too. Even with the 1Mhz bus limitation, an accelerated IIgs running at 7Mhz was quite a bit more snappy running GS/OS.
 

tetsujin

New member
Apple was able to maintain two platforms for 9 years because..... one was paying for the other. The Macintosh was not a big money maker in its early days. The Apple II line paid the bills until the late-80s and kept the company awash in R&D funds.
Well, yes! Macintosh was entirely built on the back of the Apple II. This is what allowed Apple to create the new platform, and sustain it long enough for it to become successful. IMO that was the way to go - if you have a successful platform, make money from it for as long as you can - but recognize that platform isn't necessarily going to be viable forever, and you need to invest the money you make from that platform to get to what's next. Apple had already tried and failed this twice before the Macintosh, of course...

Further developing the Apple II line wasn't going to do much for them because its biggest assets (by 1986) were its large established install base, and its established presence in the schools - all of which meant there was value in keeping the platform pretty much as it was. It was a great platform, but the things it had going for it, like the tightly optimized 5.25" disk controller or the clever implementation of color graphics - were all rooted in 1970s tech. It wouldn't have been impossible to bring the Apple II platform up to date a bit, but there's only so far they could take that (the CPU couldn't really keep up with new peripherals like DD 3.5" drives without a buffered disk controller, and the Apple II only had so many expansion slots, and a lot of these would be dedicated to specific peripherals, even as those functions became more integrated - things like the 80 column card, mouse card, printer card, etc.) So with IIgs you'd have a bunch of empty slots, but to actually use a card in one you'd have to disable a piece of functionality that was assigned to that slot... Another sign IMO that the Apple II platform was more or less already pushed to its limit.

Aiming to make the IIgs into a competitor for late-80s home computing platforms like Amiga and Atari ST was potentially a more forward-looking decision than simply updating the Apple II a bit - but even beyond the CPU limitation they really didn't turn out a machine that was suited to that role. Could they have done so? I don't know, maybe... Certainly it was possible but I don't think it was something Apple was at all prepared to do at that time. IMO basing this late-80s home computer on the Apple II mostly just held them back. But with the Macintosh already on the market (and still finding its footing) it wouldn't have made sense for them to introduce a new, different platform. The only decisions that made sense for introducing a new Apple home computer in 1986 IMO were to make it as part of the Apple II line, or part of the Mac line. Going with the Mac line at that point would have been a little odd as there hadn't been a color Mac yet - to make the first color Mac a "low cost" (i.e. under-$2000) machine with (presumably) a 15kHz display - it would have been pretty weird, and might have thrown a wrench in their plans for the Macintosh II. But it wouldn't be impossible if you look at what the ST and Amiga accomplished. Making it part of the Apple II line makes a certain kind of sense but it came with all this baggage... The 1MHz bus and the fairly rigidly-defined slot assignments, the inertial mass of 8 years or so of software back-catalogue potentially impeding development for the new hardware, and the 8-bit CPU.

I don't think "SCSI out of the box" really would have made sense for an Apple II. From my perspective anyway I feel like if you were still on an 8-bit machine in 1987, you probably weren't getting a hard drive or CD-ROM any time soon. (That said, a friend of mine did have a hard drive for his Commodore 64, for his BBS... And I guess another friend of mine probably had a hard drive for his IIgs-based BBS...) SCSI cards did exist as add-ons for the IIe and IIgs, of course, so you could disable your printer port or something and add one. :)

1.44MB floppy support would have been nice for sure. Though the Apple II had already gotten an updated disk controller to handle the 800K drives (integrated in IIgs and IIc+, add-on for IIe, incorporated into the drives themselves for other models) - by 1988 (when the 1.44 drives were first supported on Mac) I'm not sure it would have been worth updating the Apple II disk controller again... I'm not sure. From my perspective, in 1989 or so an 800K drive was still pretty sweet. Though of course more disk capacity is always nice.
 

NJRoadfan

Well-known member
Apple had prototyped a machine with onboard SCSI, built-in disk drive, along with 30-pin SIMM based memory expansion. There are plenty of photos online of the "Mark Twain" ROM 04 machines. The platform had support for Superdrives eventually with the release of the FDHD controller.
 

tetsujin

New member
Apple had prototyped a machine with onboard SCSI, built-in disk drive, along with 30-pin SIMM based memory expansion. There are plenty of photos online of the "Mark Twain" ROM 04 machines. The platform had support for Superdrives eventually with the release of the FDHD controller.

That's nice. That would have been around 1990 or something, right? I think it probably was the right choice for them not to put resources into actually releasing it... I don't think the IIgs was doing all that well at that point, and even with pretty significant upgrades by the early 1990s I don't see how they could have turned that around.

I didn't know the Apple II got HD floppy disk support, though. That's cool, but maybe a bit pointless IMO.
 
Last edited:
Top