The best opportunity for improving the wedge case Amigas without rising their price, was missed when Irving Gould caught a leprechaun but forgot to ask him for all his money.
Instead of investing the millions of gold coins for new developers, an improved chip company and environment friendly waste handling, he bought a private jet and hired the worst(?) managers that where chased away (not without paying them out) before their changes and actions could show some success.

Last edited by OneTimer1 on 14-Aug-2024 at 10:37 PM.

@OneTimer1: it's very easy to finger point the leader for all failures, and ignore all other contributors...

@cdimauro

That’s pretty much why leaders exist.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@kolla: when my grandpa was prisoner of war at the Stalag V-b of Villigen, it wasn't the man with Charlie Chaplin's mustaches which was beating him...

@cdimauro

Quote:

cdimauro wrote: @kolla: when my grandpa was prisoner of war at the Stalag V-b of Villigen, it wasn't the man with Charlie Chaplin's mustaches which was beating him...

Yes, but that same mustachioed leader with a Napoleon Complex was the reason there was a man at Stalag V-b to beat PoWs.

My mother grew up without her father for the same reason. None of which would be the case unless their supreme leader decreed it so.

_________________
All the way, with 68k

@kolla

Quote:

kolla wrote: @cdimauro That’s pretty much why leaders exist.

Pretty much.

_________________
All the way, with 68k

@OneTimer1

Honestly, Commodore (CBM) should be a case study at business school, for "How not to manage a technology company".

The old mantra used to be "Commodore were bad at marketing".
The more I've seen how that proverbial sausage was made, they could've hired the world's best marketers and they would've somehow mismanaged and squandered that opportunity as well.

I ultimately hold IBM accountable.
For most things gone bad in IT and not just the C= debacle.

Last edited by agami on 15-Aug-2024 at 08:21 AM.
Last edited by agami on 15-Aug-2024 at 08:20 AM.

_________________
All the way, with 68k

@agami

Quote:

agami wrote: @cdimauro Quote: cdimauro wrote: @kolla: when my grandpa was prisoner of war at the Stalag V-b of Villigen, it wasn't the man with Charlie Chaplin's mustaches which was beating him... Yes, but that same mustachioed leader with a Napoleon Complex was the reason there was a man at Stalag V-b to beat PoWs.

Sure, and others... executed his orders. I don't absolve them: they are equally culprits.
Quote:

My mother grew up without her father for the same reason. None of which would be the case unless their supreme leader decreed it so.

I'm here only because my grandpa escaped when allies were bombing Germany. He walked around 1600km 'til he reached Sicily. When he arrived he was just 40kg, and was about the die. Fortunately, he recovered...

@agami

Quote:

agami wrote: Honestly, Commodore (CBM) should be a case study at business school, for "How not to manage a technology company". The old mantra used to be "Commodore were bad at marketing".

They had a huge success with A500/A2000 systems but Irvin Gould fired the manager who organized it.

Other managers might have done important work by cutting down the costs but this could only be a temporary thing, if you don't invest you will run into troubles.

Quote:

agami wrote: The more I've seen how that proverbial sausage was made, they could've hired the world's best marketers and they would've somehow mismanaged and squandered that opportunity as well.

Most home-computer companies never got as far as C= but if you look to Apple they made something better but they made their users pay.

Quote:

agami wrote: I ultimately hold IBM accountable. For most things gone bad in IT and not just the C= debacle.

You can blame them for their bad PC development, if it had been better C= would have run into troubles earlier. But IBM themselves too lost the leadership in the desktop market, Compaq doesn't exist any more and Apple survived by switching to luxury gadgets.

agami Quote:

Honestly, Commodore (CBM) should be a case study at business school, for "How not to manage a technology company". The old mantra used to be "Commodore were bad at marketing". The more I've seen how that proverbial sausage was made, they could've hired the world's best marketers and they would've somehow mismanaged and squandered that opportunity as well.

CBM was about as bad as it gets for a large business. However, I believe the current Amiga related businesses are worse. Hyperion shenanigans exceeded Enron and are right up there with FTX (Sam Bankman-Fried was like Ben Hermans on a larger scale) while Warren Buffet and Bill Gates combined couldn't afford to keep A-Eon afloat if it was a large business with their current business ineptness. Why take the risk of stealing the Amiga IP from Amiga Inc. only to exceed them in incompetence and sabotage the Amiga perpetually? Isn't the only reason they run under the radar despite such shenanigans because they are micro businesses that don't make any money? Would anyone else but mental cases continue under such circumstances?

agami Quote:

I ultimately hold IBM accountable. For most things gone bad in IT and not just the C= debacle.

https://hackaday.io/project/190838-ibm-pc-8088-replaced-with-a-motorola-68000

IBM deserves most of the blame for the poor but fateful 8088 choice but Intel had a hand to play too. IBM was targeting a very low spec as the 68000 workstation MPU specs would compete with their too expensive mini computers. They demanded $5 for a MPU from Intel CEO Dave House.

https://archive.computerhistory.org/resources/access/text/2012/04/102658299-05-01-acc.pdf Quote:

House: And so I get word from .. God, what was that guy’s name? Uhm... He was a salesman. He worked for... Earl Whetstone who I believe was the district manager. Uhm... an Italian name. It’s going to come to me. (Paul Indaco). He (Paul Indaco) said that they’ve got this project to develop an Apple killer, they want to beat Apple. So we go down there and we make our pitch that … what are they planning on using? Well, they plan to use the Z80. Well, this is a big loss. So we go down there with our pitch, which is you can’t beat an established 8-bit machine with another 8-bit machine. You gotta go to 16-bits. You-you gotta use the 16-bit machine, and it’s the only way you’re going to beat an entrenched 8-bit machine. So then they opened up their decision-making process, and they look at the 68000 and the Z8000, and the 8086 and they said well, clearly the 68000 or Z8000 are the better of the three, but we can’t meet our cost numbers with any of those three approaches. Now, we had a chip set- in those days, your I/O chips- that was a parallel I/O, which in 8-bit was 8255, a serial I/O, which was an 8251 with a UART interface, a timer counter, which is the 8253, a DMA controller, 8257, and an interrupt controller, an 8259. We had produced those and… guess what? AMD and a bunch of other people wound up second sourcing them, and so with competition and the semiconductor learning curve, the price had gone down low. We had...when we designed the 8086, some of my engineers came to me with an idea which was... it would be very easy to take the bus of the 8086, and fold it over on itself, so that it did its 16-bit accesses two bytes at a time. In so doing, they could make the bus compatible with the 8085 bus, so they could use the 8085 peripherals (8251, 8253, 8255, 8257 & 8259). (It was really a back up plan in case our new 8086 peripheral chips were late.) Now meanwhile, we had a bunch of 16-bit peripherals we were developing, and if course, they were all new, and they weren’t all quite done yet. We had to price ‘em much higher, etc. but I’d been beaten down on price on the 8251, 53, 55, 57 and 59, by competition who was second sourcing the product, and they were very economical. They just happened to fit the wrong bus. So the design engineers they said, it won’t take much to do this (the 8088 bus). I said, I don’t think there’s much use for it. I mean, you’re really hog tying the 16-bit machine, making it do two bus cycles, but go ahead and put it in. We just won’t bond it out. We won’t test it. So they put it in the design. It was there. We didn’t test it. So I’m trying to get this design win, and I’m going to lose on architecture to the 68K or the Z8000, so I said, where’s that 8-bit bus version (the 8088)? So I had them go pull out the design and bond up some chips and before they had even done that, we went and sold it to IBM. Hendrie: Oh, my House: We said (to IBM), okay, here’s the deal. You have a 16-bit machine, 16-bit addresses, 8-bit peripheral chips, low cost, multi-sourced...one of the things they were big on was multi-sourcing. They later made us second source to AMD, I guess it was at that time, they made us second source the 8088 to AMD. AMD, so (later) we sourced the 286 to AMD. By the time we got to the 386, we were able to stave it off. But the process went like this. It (the IBM PC) used the 8088 and because they’ve been working with us with the 8080 and 8085 earlier, particularly the 8085 earlier…and then…286 being the second generation. But with the 8086, they wanted a second source and we kind of had to cave on that and let them manufacture it. So we licensed them to do it internally. We made the internal pitch that their operation be our second source on the 386. And so we sold ‘em on the 8088 based on the peripheral chips, the low cost peripheral chips, and that’s how we got our- our design win. Hendrie: Oh, really. House: Yeah. Hendrie: Yeah, Okay House: So it was based on- you get 16-bits compatible. All your software in the future… you can do the 16-bit bus version later when the cost comes down but I had to commit to get the price of the 8088, the 8-bit version, I had to get it- promised to get it - down to 5 bucks in two years. Had to commit to a price, gotta get it down to 5 bucks in two years. So obviously, once we had that going, the IBM PC down - announced in August of 1981, we had to get ‘em off the 8088 and onto something we could make some money on. Hendrie: Yeah. Because the… House: ‘Cause I’d committed to get the price to 5 bucks to win the design. Hendrie: This was the loss leader. House: That’s right. Hendrie: Okay House: So the IBM PC is announced, and it’s, it starts taking off and it exceeds everybody’s expectations, including ours, and like wow. I mean, turns out to be the defining moment really for Intel, and Intel published this publication on the (Intel) 25th anniversary, the 25 identifying events of the first 25 years- or defining events of the first 25 years. In the middle of the book is the IBM PC design win. You’ll see Earl Whetstone’s and my picture there, the people credited with getting that design win. ‘Cause the sales force was really in there every day and I was directing it from a strategy standpoint. We had a number of other people that were obviously involved up and down [the organization]. It was a team effort, like everything at Intel. But the defining moment is when- when we got that design, and then now they’re developing the thing at the…It was really interesting the way they did it. They would not...they were so secretive. The story is true that they had rented this space in a shopping center of an old, I think it was a super market that had gone out of business, and they had set up all this office space in there and designed it there. When they would have a problem, we’d send in our field application engineer (Paul Indaco), or somebody from the factory (usually Dane Elliot), and they would have a black curtain up and the machine would be behind the curtain and the development system being on the other side of the curtain. We were able to- we could run our development system, and we could interface with it, but we could never see it (the PC). They never let us see the machine.

The 8088 was a piece of shit that didn't make much sense like the later 68008. The 8-bit data bus 8088 and 68008 have about the same number of transistors and the same number of pins as the full 16-bit 8086 and 68000 meaning there wasn't any cost savings and decreased value. Dave House would not sell the 8086 to IBM for $5 but did sell the crippled 8088 that cost approximately the same to produce to them for $5 (the 68000 became as cheap as the 68008 due to economies of scale for the better value MPU and 8-bit data operation was added to later 68000 variants so the illogical 68008 could be discontinued). Dave House agreed that the 8088 was a loss leader (sold below cost) and started 8088 production so surely he gambled that the crippled 8088 would need to be upgraded. The 68000 was obviously the vastly superior MPU but it used more than twice the transistors, a large package and more pins so it was at a cost disadvantage to match just on price. I expect IBM would have made adjustments based on the 68000 value if the deal with Intel fell through but the 68000 was going to have a difficult time with IBM looking at price, 2nd sources producers, peripheral chips and existing software. The Z8000 may have still been an option with IBM's requirements. As Dave House said, the choice of 8088 is one of the defining moments in computer history. The 68k was still in the game as Motorola made a similar deal with Apple and the Amiga should have been a dominant force taking market share due to the superior ISA and chipset but due to incompetence CBM ended up being the most forgettable of the big 3 personal computer businesses. Then Motorola threw their baby out with the bath water and bet the farm on IBM's PPC. The rest is history.

Last edited by matthey on 16-Aug-2024 at 06:34 AM.

@cdimauro

Commodore The Final Years book has identified the persons involved in greater detail.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@matthey

68008 was released in 1982 which is too late for IBM's original PC.

For 68020, Commodore spent R&D resources on its custom MMU before Motorola released 68851. This caused Commodore to a double R&D spending on the A2620 card.

Timing matters.

-----

Last edited by Hammer on 16-Aug-2024 at 01:41 AM.
Last edited by Hammer on 16-Aug-2024 at 01:23 AM.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@OneTimer1

Quote:

You can blame them for their bad PC development, if it had been better C= would have run into troubles earlier. But IBM themselves too lost the leadership in the desktop market, Compaq doesn't exist any more and Apple survived by switching to luxury gadgets.

Apple's luxury gadgets are backed by entertainment distribution iTunes and good GUI design i.e. digital era's Walkman.

https://www.youtube.com/watch?v=eywi0h_Y5_U
Ballmer Laughs at iPhone.

Steve Jobs has been focusing on "easy to use" GUI since the original MacOS.

Windows phone's GUI design sucks. It's good that Google Android has removed Windows phone crap UI from the market.
-----

In 2002, Compaq signed a merger agreement with Hewlett-Packard where each Compaq share would be exchanged for 0.6325 of a Hewlett-Packard share.

Compaq's pre-merger ticker symbol was CPQ. This was combined with Hewlett-Packard's ticker symbol (HWP) to create the current ticker symbol (HPQ). HP(HPQ) later split into two companies i.e. HP Inc (HPQ) and Hewlett Packard Enterprise Company (HPE).

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Hammer

Quote:

Hammer wrote: @cdimauro Commodore The Final Years book has identified the persons involved in greater detail.

Which is good, but I'm also identifying some of them.
Quote:

Hammer wrote: @matthey 68008 was released in 1982 which is too late for IBM's original PC. Timing matters.

Not only that, from what was reported: MONEY matter.

Intel was selling its 8088 at loss for IBM's $5 price for the CPU.

What other company could provide a 16-bit processor at that price at that time?
Quote:

https://www.youtube.com/watch?v=eywi0h_Y5_U Ballmer Laughs at iPhone. Steve Jobs has been focusing on "easy to use" GUI since the original MacOS. Windows phone's GUI design sucks. It's good that Google Android has removed Windows phone crap UI from the market.

Windows Phone's GUI was very easy to use, very fast & reactive, and A PLEASURE to program.

I missed it a lot, both as user and developer.

Android is not only a mammoth using Linux as its kernel & (part of) ecosystem, but its architects provided a crappy platform for its development and don't even know the OOP basics...

@cdimauro

Quote:

What other company could provide a 16-bit processor at that price at that time?

Maybe MOS? A 16 bit 6502 chip was a possibility, had it not been for Tramiel and his cost-cutting campaign.

_________________

WolfToTheMoon Quote:

Maybe MOS? A 16 bit 6502 chip was a possibility, had it not been for Tramiel and his cost-cutting campaign.

There was no 6502 family 16-bit CPU available early enough. However, the 8-bit 6509 is capable of addressing up to 1 megabyte of RAM via bank switching and may have been available early enough. CBM used the 6509 in the successor to the CBM Pet called the CBM-II released in 1982.

in the https://en.wikipedia.org/wiki/Commodore_CBM-II

The cheaper C64 replaced it but the CBM-II with 6509 offered several advantages including the 1MiB addressing and planned 1-3 MHz operation (only up to 2MHz for CBM-II but 6509 data sheet shows 3MHz). Addressing more than 64kiB is clumsier than the 8088 but IBM was ignoring the ideal way to address more than 64kiB with a large flat address space and making decisions based more on price. Jay Miner did not make the same compromise in relentlessly fighting for the 68000. Jay was smarter than IBM but IBM was smarter than CBM so advantage lost. Jay made the right choice and now the Amiga lives with the 68k or dies without. Some of us know it would be wrong to compromise on anything less just like Jay Miner knew it would be wrong to compromise on anything less.

Last edited by matthey on 16-Aug-2024 at 06:02 PM.

@cdimauro

Yeah, I thought the Windows Phone tiles looked clean and I liked the way you could choose one big icon or four small ones to git the same space!

_________________
"Art challenges technology. Technology inspires the art."
John Lasseter, Co-Founder of Pixar Animation Studios

@matthey

Quote:

There was no 6502 family 16-bit CPU available early enough

That is not MOS's fault. When Tramiel took over it didn't take long for him to cut most R&D and a lot of the original team that developed the 6502 family left. Both 16 and 32 bit versions of the 650x were planned by MOS.

_________________

WolfToTheMoon Quote:

That is not MOS's fault. When Tramiel took over it didn't take long for him to cut most R&D and a lot of the original team that developed the 6502 family left. Both 16 and 32 bit versions of the 650x were planned by MOS.

Are you sure about a 32-bit 6502 family CPU that early? A 32-bit 6502 family CPU doesn't make much sense that early because of few registers requiring excessive memory traffic meaning there would be little advantage without 32-bit wide memory which was expensive. With many GP registers, a 16-bit wide data bus and memory offered some performance improvement which is what the 68000 used.

year | MPU | GP regs | reg bits
1979 MC68000 16 32 https://en.wikipedia.org/wiki/Motorola_68000
1980 Bellmac-32 9-12 32 https://en.wikipedia.org/wiki/Bellmac_32
1982 NS32016 8 32 https://en.wikipedia.org/wiki/NS32000#32016
1984 NS32032 8 32 https://en.wikipedia.org/wiki/NS32000#32032
1984 MC68020 16 32 https://en.wikipedia.org/wiki/Motorola_68020

The 32-bit registers not only allowed 32 bit address pointers and a large flat address space but the GP registers reduced data memory traffic for a 16-bit data bus. The 68k ISA has the most GP registers but the unusual data and address register split. The Bellmac-32 ISAs above started with 16 orthogonal registers but rigid register designations and control registers using the orthogonal register encoding bits limited their GP use. The 6502 ISA has 4 non-GP registers and lacks encoding space to add more without an incompatible change to the encoding. Most instructions access memory but few registers caused an increase in instruction counts and code size as well. A 16-bit 6502 family CPU makes sense to retain compatibility with some enhancements but beyond that I would be looking for a better ISA. Operations with 32-bit wide data were difficult at this time. The Bell Labs Bellmac 32 was the first MPU to support single cycle 32 bit ALU operations and even had a 32 bit barrel shifter all operating in CMOS which the NMOS 68000 lacked. The superior 68000 ISA and the right balance of features made the 68000 practical and a standout despite Motorola's poor fab tech vs Bell Labs pioneering fab tech at the time. Too bad Motorola and Bell Labs did not team up sooner as their successors Freescale and Lucent (with Infineon) teamed up to create the StarCore DSP.

Last edited by matthey on 16-Aug-2024 at 11:08 PM.
Last edited by matthey on 16-Aug-2024 at 09:32 PM.

@Hammer

Quote:

Hammer wrote: Apple's luxury gadgets are backed by entertainment distribution iTunes and good GUI design i.e. digital era's Walkman. Steve Jobs has been focusing on "easy to use" GUI since the original MacOS. Windows phone's GUI design sucks.

Software, APIs, System Infrastructure are mandatory for professional PC usage, something that was neglected by Commodore.

But Apple went some steps farther with a media shop called iTunes, they became a professional player in the content industry. If C= knew how to get out some money on the games played on their Amigas, they may have survived.

But like I posted before, the high prices on Macs made it easier for Apple.

Quote:

Hammer wrote: Windows phone's GUI design sucks.

I was told it it was good for phone, it had good responsiveness using less resources than iOS or Android but they sucked on App stores and interoperability. Sharing Word or Outlook data with Android phones was easier than on a Windows Phone.

But crippling Windows9 GUI for Windows Phone didn't help, I may have made it harder to convince PC Users.