128k Mac - What were apple thinking!

[gavo]

Well, after getting the old 128k I have into a state where I could use it, I've spent the last couple of nights going through some of my old floppies that have survived my travels. My thoughts are:

1. What an amazing little machine this really is

2. What were apple thinking only releasing it with only 128k

I guess ultimately releasing it with 128k didnt have a long term impact on Apple or the Mac itself (and its success), but I cant help thinking that the uptake would have been greater if the original had a little more RAM. Its amazing how little will run on this beast - even more amazing how much that will!

The other amazing thing (to me anyway), is that this 20-odd year old machine will quite happily read my 20-odd year old floppies - most without a missing a beat! I mean, come on, I have 5 year old PC floppies that wont read in my 5 year old (or newer) PC that have been treated with the same respect that my Mac stuff was. This stuff was just built I reckon.

I'm also stunned over how relevant some of the UI stuff is, even compared to OS'es of today. Sure, not as wiz-bang as XP or OS-X, but very usable none the less, and it doesnt feel to me like a vintage OS at all (maybe thats because I'm vintage tho!). Still, I would probably say the same about ProDOS on the apple II - I guess (in my view) Apple just knew what it was all about.

Its funny, I have used emulators in the recent past to relive some of my early Mac fun, but I have to say, emulators just dont compare to the real experience. An emulator wont give you the sound of that musical internal 400k drive as it seeks, the feel of the big square mouse in your hand or the clunky thunk as you bang those keys. Dont get me wrong, I think in the end, the way these things will be remembered will be through emulators, but for a hobbyist - if you can get your hands on the real thing - thats where the enjoyment is

Its all a bit nostalgic I guess, but damn, that was a smokin machine and I'm having a blast playing with it again.

Just thought I'd share

 

[bluekatt]

there were no bigger ram modules availble at a reasonable price at the time

why do you think the moment that the ram prices dropped apple released the 512k mac the same year ?

the mac developers had to perform miracles to get a full working gui in 128 k and miracles they did perform

i am not at al enamoured by classic mac os but it is amazing what they managed to do with just 128 k at the time

 

[Mac128]

there were no bigger ram modules availble at a reasonable price at the time why do you think the moment that the ram prices dropped apple released the 512k mac the same year ?

You forgot greed. Apple typically enjoyed a 4x markup. Divide by 4 and you'll see Apple only paid about $600 per unit to make. Had they not been so greedy, they might well have put 256K or even 512K in there understanding it would be better in the long run. But despite what Jobs said about starting a revolution, his real motivation has always been making as much money as possible for as little investment as possible and making the consumer think they've just gotten a deal. There are well known stories about how he exploited Woz (i.e. working the man to death saying he would split $700 when in reality Jobs was being paid $3,000), so there is no reason to think anything has changed.

But there really is more to it than that. Apple already had designed the 128K board to be a 512K board. All they needed to do was socket the RAM and add a multiplexer socket. Then any dealer could add the additional RAM if a customer could afford it, costing Apple very little in terms of manufacturing. Much is made of Job's insistence that the Mac was a closed box and he didn't want anybody inside them, even technicians – if people wanted an upgrade they would have to buy a new box. Well that's just dumb. Why then did it take Apple 8 months to introduce the 512K? They weren't just waiting for a RAM price drop, since the thing already existed and people were screaming for it at any price.

I think we all tend to forget about the Lisa when we start debating Apple's missteps with the Macintosh in those days. But in January 1984 they had a $7,000 albatross hanging around their neck. Nobody was buying it. When the Mac was introduced, they dropped the price on the Lisa and offered technology the Mac didn't have and Job's said he didn't want. If you wanted more RAM, you HAD to buy a Lisa. I'm sure Apple thought the success of the Mac was going to help save Lisa. But it was immediately evident, people wanted the Mac not the Lisa. So MacWorks showed up in April to help turn it into a Mac and Apple bit the bullet in September releasing the more affordable 512K which really nailed the lid on Lisa's coffin, even though Lisa benefited by the price drop in RAM as well. Once repositioned in January 1985 as a much cheaper, optional 1MB MacXL with a hard drive, Lisa finally began to sell well. Perhaps too well, because now the Mac wasn't, so they simply "forgot" to re-order parts and that was the end of Lisa.

That, I think is the true story of Apple's original 128K.

As for software, there was tons of it! Just looking at the software offerings from a late 1984 Mac Buyer's Guide, I would say 90% of it was 128K compatible and there's almost 50 pages worth. It just isn't easy to find today.

But then again, this was just a rhetorical question, wasn't it.

 

[tomlee59]

Indeed, old-fashioned greed is the primary reason that so little RAM was put into the machine. First, Jobs' minimalism drove the team toward 128K, and then Sculley's pencil-pushing decreed a $2500+ retail price. The engineering team wanted to include, at the very least, an easy way to increase RAM to 512K, but Jobs refused. When the team tried to hide expansion capabilities in the guise of a bus-connected "diagnostic port," Jobs saw through the ruse and shut that down.

Even at 1984 prices, 512K of RAM would've been entirely within reason, and still would have allowed Apple to enjoy handsome margins.

 

[Aoresteen]

Apple could have released the 512k instead of the 128K - they had 512k macines in Feb of 1984 (I met a developer who had been seeded one). The cost would have pushed the retail price over $3,000. Instead of the RAM they bundled the ImageWriter printer with it. $2500 plus $250 tax made the purchase $2750 - I bougt mine from GlenElllen Computer in IL.

The developement team did make the 128K board upgradeable by putting decode logic traces on the 128K mb. I know 'cause I cut the 128k chips off my June 1984 128k mb and did my own 512k upgrade. Yes it was a PITA to clean out all those holes and solder sockets onto the mb but it was doeable. [] ]'> I still have the coumputer but it sports a 6MB MacRescue board now.

I've thought about retuning it to it's orginal configuration but since I had to turn in the 64K roms when I did the 800k drive upgrade I just bought a virgin 128K a few years ago instead.

 

[tomlee59]

Apple could have released the 512k instead of the 128K - they had 512k macines in Feb of 1984 (I met a developer who had been seeded one). The cost would have pushed the retail price over $3,000.

Again, only because of greed. They wanted margins that were well above industry norms. Sculley's calibration was that $2500 was about the upper limit on retail price, so their margin targets forced the use of smaller amounts of RAM. Without changing the retail price target, 512K would still have allowed Apple to enjoy extremely generous margins.

 

[tomlee59]

Oh -- and the upgrade-via-traces method? That was not on the original rev of logic boards. The original 128K boards had no means of upgrading without fairly major surgery.

 

[Aoresteen]

Oh -- and the upgrade-via-traces method? That was not on the original rev of logic boards. The original 128K boards had no means of upgrading without fairly major surgery.

I don't have the date of mfg of my 128k board or revision but I do not recall there being any difference as it relates to upgrability of the 128k boards. For the decode logic I used a Bit9 card from Ken Wilhelm; AFAICR it worked with all versions of the 128k mb. When I said traces I was refering to the decode logic for the extra RAM.

I used the Jan 1985 Dr. Dobb's Journal article "Fatten Your Mac" by Thomas Lafleur and Susan Raab to do my upgrade. As I'm deployed in Iraq I can't look at my copy and Google didn't find a copy. Perhaps someone can check the article to see if there were any issues with upgrading different versions of the 128k motherboard. I just don't recall that there were any. Maybe since my 128K was purchased in June of 1984 it was a later version motherboard and I didn't have any issues upgrading it. Hell, it was 22 1/2 years ago that I clipped the chips off of the motherboard!

 

[tomlee59]

The very first revs of the boards indeed had no provision for an extra address decoder chip (I know this because I performed upgrades for several of my friends in school). Upgrading those required adding an ugly little kludge board to accommodate the extra chip. Mounting it in a reliable way that still left adequate clearance for the logic board to slide into the chassis was a bit of a mechanical engineering challenge. Thankfully, Apple revved the boards very soon afterwards with space for a decoder chip. On those, the upgrade was much simpler (and looked nicer). Then you were ready to swap out the RAM chips themselves.

 

[Mac128]

Oh -- and the upgrade-via-traces method? That was not on the original rev of logic boards. The original 128K boards had no means of upgrading without fairly major surgery.

Tom, I have often wondered about what prompted the redesign of the 128K to the hybrid 512K (which is what I assume you are referencing) and the only answer I could come up with was the hybrid is nicer looking, which Jobs would have demanded.

The reality is, the 128K board is actually a 512K board designed to handle the extra RAM from the beginning. I have one of the early 128ks with what I call the "Larry Pina method" upgrade and it has a fairly elegant implementation of the mini-mux board at location E3 (someday I will post pics). In fact, if the pins were a little shorter or they didn't put the IC in a socket, it would not be any higher than the socketed ROM chips. It actually looks so good that if I didn't know it wasn't supposed to be there, I would think it was OEM. That said, Apple themselves must have implemented the exact same kludge in order for Jobs to do the Macintosh intro demo in January '84. FYI, the upgraded looks nicer on that hybrid board, but for my money involves a lot more work.

My point is that greed combined with the threat of competition against the Lisa which had been reduced to 512K in its stock configuration, prevented Apple from configuring the board properly for a 512K implementation from the beginning. Even so, that Mux board could have been popped into the Mac as soon as at it became clear from the first review, everyone wanted more RAM and eventually, crafty reverse-engineers or leaks from Apple allowed the public to take matters into their own hands. So Apple ended up losing anyway. As Aoresteen correctly points out, the very first boards had a way to implement 512K RAM (however inelegantly), Apple just chose not to and we can only speculate at the reasons why they waited as long as they did.

The real question is what did it cost them to make a can of Pepsi before Sculley left?

 

[tomlee59]

Well, I think there's some confusion about definitions. That the 128's logic board had a way to implement 512K RAM is certainly true, but that's a statement about the determination and ingenuity of hackers, not the foresight of Apple. With only a little hyperbole, an equivalent statement would be "the 128 had a way to implement overclocking." All that is true, but it is true despite Apple's intent, not because of it. That's the distinction I am making.

Aoresteen's comments indicate that he hasn't seen the original version of the 128 logic board. There were at least two that I'm aware of, with the original lacking a space for the decoder chip, and a later one with. Neither was ever intended to be upgraded. The later version was in anticipation of shipping a 512K Mac. Both 128K versions had soldered-in DRAMs as a consequence, requiring nontrivial surgery to perform the upgrade.

And we actually don't have to speculate about why things were done this way. It is a matter of historical record that Jobs viewed the Mac as a closed system. Part of this was in reaction to the messiness of the Apple II's open architecture. The other was Jobs' wish to move computing from the world of geeks to "ordinary folk." As a result, the compact Macs would not allow even RAM upgrades until the Plus, and would not accept expansion cards until the SE (and then, only in limited form), after Jobs had left.

 

[Mac128]

Well, I think there's some confusion about definitions. That the 128's logic board had a way to implement 512K RAM is certainly true, but that's a statement about the determination and ingenuity of hackers, not the foresight of Apple. ... All that is true, but it is true despite Apple's intent, not because of it. That's the distinction I am making.
Aoresteen's comments indicate that he hasn't seen the original version of the 128 logic board. There were at least two that I'm aware of, with the original lacking a space for the decoder chip, and a later one with. Neither was ever intended to be upgraded. The later version was in anticipation of shipping a 512K Mac.

Tom, you never cease to expand my thinking on the original Macintosh. Just when I think I understand what was going on in those days, you come along and smash it all to hell. LOL

Just so we are all thinking along the same lines, these are the two boards I am referring to:

1) the original 128K board http://www.digibarn.com/collections/parts/mac-prototype-boards/1983-medium.jpg

2) the 128K/512K hybrid board http://picasaweb.google.com/radius226/CompactMacintoshDisassembly/photo#5126677476008105266

(2 alt – not as good:
)

These are the only two 128K boards I know of. The one of mine I mention has been upgraded to 512K by means of the addition of a mux board is the 1st one. The mux board is installed at location E3 which as pictured seems to be an intentionally empty location on the logicboard (especially when compared to the rest of it) with conspicuously empty pads. The later board (#2) had an empty spot at G13, which in this picture shows it filled with the mux chip making this the 512K (the 128K would have a nice clean spacious spot in which to insert the chip).

Based on articles like this and others from Folklore: http://www.folklore.org/StoryView.py?project=Macintosh&story=Macintosh_Prototypes.txt&topic=Prototypes&sortOrder=Sort%20by%20Date&detail=medium

I have always thought that the 128K board was always intended to be upgraded to 512K and that the connection point at E3 was how Apple originally intended it to happen, but Jobs blocked it. I made a huge leap in logic assuming that the secret to upgrade it leaked out or was discovered.

You seem to be indicating that Apple never intended that empty spot at E3 to ever be used for RAM expansion, but it was just happenstance that the necessary traces happened to converge there allowing a mux board to be inserted and expand the RAM. Indeed a "hack", not something the engineers even knew was there, much less designed? I KNOW Apple never intended these boards to ever be upgraded, my question is whether the engineers intended that original design (#1) to be used for a commercial 512K board and were shot down, or if it was just a fluke?

In the latter case, this Folklore article http://www.folklore.org/StoryView.py?project=Macintosh&story=Intro_Demo.txt&topic=The%20Launch&sortOrder=Sort%20by%20Date&detail=medium

suggests that Apple had back in "the lab" a 512K logicboard that would have resembled logicboard #2 pictured above and the one they later put into production. I had always assumed that the 512K board they used for the shareholders meeting was modified exactly like the one I have, i.e. an original 128K with a mux board at E3 which generally resembles this one:

http://68kmla.org/forums/viewtopic.php?t=4298

Though as I have said, mine looks like OEM Apple. I will take some pictures soon to complete this discussion.

Assuming we are talking about the same boards, then I can see Jobs gloating about the correctness of his vision while the Mac was selling gangbusters until June of '84. Even though from the first reviews in January, everyone realized it needed more RAM, something his engineers had been insisting since 82, it wouldn't have been until the sales dropped dramatically that Jobs would have lost his ability to cling to his 128K ideals and acquiesce to the demands to release a 512K model. It is unimaginable to me that Apple would have really let him get away with that line of thinking, particularly when they stood to make substantially more money from releasing a more powerful version. Jobs is on record saying he didn't want the 128K to be expandable, but also that if people wanted more RAM they could buy and upgrade instead. So he wasn't against the upgrade. That's why I can only surmise the Lisa played some part in the timing of the decision as to when to release the 512K upgrade. It could be too that Jobs was hoping the RAM prices would drop so that he could simply offer the 512K Mac upgrade and just discontinue the 128K Mac, for the same price. After all, "Macintosh" is so much more elegant a name than adding on all those pesky model numbers.

 

[Scott Baret]

Is it true that Jobs used to use a Fat Mac before they were available during his presentations?

 

[Mac128]

Is it true that Jobs used to use a Fat Mac before they were available during his presentations?

yes. the folklore article goes into detail about it.

 

[Aoresteen]

Lots of good stuff here! A worth while project would be to identify (with photos) the evelution & versions of the 128k motherboard. Once we have that established we could then discuss what drove the engineering changes.

Then when we speak about "Revison 2 motherboard" eveyone know exactly what we are talking about.

My Bit9 card is installed at E3 as Mac128 showed. The E3 area is what I was refering to when I mentioned "traces" on the motherboard.

Due to the net nanny here I could not load the 2nd motherboard picture that Mac128 posted.

It would be interesting if there was an early 128k mb that did not have the pads & traces at E3. I've looked at 5 or 6 128k motherboards (not a big sample by any means) and everyone of them have the E3 traces present.

It would be nice to have the serial numbers of the computers that the 128k motherborads came out of so we could establish the aproximate timeline for the various 128k motherboards.

 

[tomlee59]

Tom, you never cease to expand my thinking on the original Macintosh. Just when I think I understand what was going on in those days, you come along and smash it all to hell. LOL

I think your understanding is quite deep and solid. It may very well be that we are actually talking about the same thing(s), but just interpreting/labeling them differently. The debate between Jobs and his team about the size of the RAM complement certainly resulted in the spinning of several versions of the logic board. It would not be a stretch to imagine that the 128K logic board that ultimately shipped could have resulted from stripping the decoder chip out of a 512K prototype, leaving a nice space for a hacker to re-introduce the decoder chip. But a comparison of the original 128K board and the hybrid 128/512 board shows a more extensive rearrangement (thanks for linking to those two pics -- my memory is not very trustworthy; you can see that the boards differ somewhat more than one would expect if the only operation they performed was chip removal). That doesn't preclude your theory, of course, but it is clear that the engineers took extra steps to distance this board from a 512K-capable one. Undoubtedly, Jobs played a role here!

As to whether frustrated Apple engineers leaked information about how to perform the upgrade, I have no idea. I can say, though, that the mod is fairly trivial, and would not have required any detailed knowledge of the Mac architecture to devise (particularly in that era of simple page-mode DRAMs). Doing so was undoubtedly made even more trivial by BMUG's successful effort at generating a logic board schematic by the end of 1984 (!), and circulating it widely.

 

[Mac128]

It would be interesting if there was an early 128k mb that did not have the pads & traces at E3. ...

It would be nice to have the serial numbers of the computers that the 128k motherborads came out of so we could establish the aproximate timeline for the various 128k motherboards.

I have a very early 128K board from a March '84 Mac. It has the traces. Given that the final Digibarn production model shows the traces, it is unlikely there was a version in-between, particularly because the P/Ns are the same. AFAIK, the next revision occurred with the introduction of the 512K, based on various unchanged boards I've seen from 128Ks between March and September '84. The board in the 512K, of which I have one from November '84, is identical to my 512Ke Mac from November '86.

So there were only two boards most likely.

In any event, that's what http://www.mac128.com/m0001 is for: research. When you get home, make sure to register yours.

 

[Temetka]

This thread is awesome. I was never aware of some of the history of the 128k and 512k boards. Or the fact that Jobs was against an upgrade. Greed certainly does make sense in terms of how pricing was determined.

Now I have to go read more of those older articles. I always thought that I had a good understanding of Mac history and hardware, but along comes this thread colors me corrected.

 

[tomlee59]

Given your interest in the early Macs, there are a couple of books that you might enjoy. One is Levy's "Insanely Great." Reading about the final push to release the system disks to production had me feeling exhausted in sympathy. And of course, there's Andy Hertzfeld's book (based on the stories he accumulated at folklore.org). If you haven't read Levy's book, or looked at Hertzfeld's site, have a gander. I'm sure you'll enjoy them greatly!