Click Here
home features news forums classifieds faqs links search
6068 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)
Login

Nickname

Password

Lost Password?

Don't have an account yet?
Register now!

Support Amigaworld.net
Your support is needed and is appreciated as Amigaworld.net is primarily dependent upon the support of its users.
Donate

Menu
Main sections
» Home
» Features
» News
» Forums
» Classifieds
» Links
» Downloads
Extras
» OS4 Zone
» IRC Network
» AmigaWorld Radio
» Newsfeed
» Top Members
» Amiga Dealers
Information
» About Us
» FAQs
» Advertise
» Polls
» Terms of Service
» Search

IRC Channel
Server: irc.amigaworld.net
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
27 crawler(s) on-line.
 19 guest(s) on-line.
 0 member(s) on-line.



You are an anonymous user.
Register Now!
 dframeli:  23 mins ago
 eliyahu:  30 mins ago
 matthey:  33 mins ago
 MEGA_RJ_MICAL:  33 mins ago
 an_overeducated_idiot:  34 mins ago
 NutsAboutAmiga:  1 hr 7 mins ago
 Karlos:  1 hr 9 mins ago
 Spectre660:  1 hr 17 mins ago
 Rob:  2 hrs 23 mins ago
 zipper:  3 hrs 4 mins ago

/  Forum Index
   /  Amiga General Chat
      /  Irving Gould's House
Register To Post

Goto page ( Previous Page 1 | 2 | 3 )
PosterThread
Nonefornow 
Re: Irving Gould's House
Posted on 14-May-2022 1:03:34
#41 ]
Regular Member
Joined: 29-Jul-2013
Posts: 308
From: Greater Los Angeles Area

@BigD

I do not completely disagree with your assessment.

But that would seem to indicate that the blame falls more Irving and Medhi than on Jack.

 Status: Offline
Profile     Report this post  
matthey 
Re: Irving Gould's House
Posted on 14-May-2022 4:41:27
#42 ]
Super Member
Joined: 14-Mar-2007
Posts: 1603
From: Kansas

It is difficult to assign blame as a percentage but things could have played out much different at critical points. Poor leadership made big differences in the early computer landscape.

Chuck Peddle, Bill Mensch and other Motorola 6800 designers left Motorola because of poor leadership. This resulted in the revolutionary low cost 6502 CPU from MOS Technology instead of Motorola. Motorola may have been a bigger name in CPUs but it is possible they would have created a 16 bit version of the 6502 instead of trying to jump ahead of the competition with a more aggressive 16/32 bit 68000 design. Maybe they would have had room for both the 6502 for the low end and 68000 for the high end in which case they would have been an even bigger name in the embedded market.

Bill Mensch Quote:

The legendary 65xx brand microprocessors with both 8-bit and 8/16-bit ISA's keep cranking out the unit volumes in ASIC and standard microcontroller forms supplied by WDC and WDC's licensees. Annual volumes in the hundreds (100's) of millions of units keep adding in a significant way to the estimated shipped volumes of five (5) to ten (10) billion units. With 200MHz+ 8-bit W65C02S and 100MHz+ 8/16-bit W65C816S processors coming on line in ASIC and FPGA forms, we see these annual volumes continuing for a long, long time.


https://www.westerndesigncenter.com/

The 68000 is still in production too even though no new 68k cores are developed, licensed and marketed for embedded use like WDC still does with the 6502 family resulting in hundreds of millions of cores sold annually. It's also possible that the revolutionary fab technology developed by the ex-TI team at MOS Technology would have resulted in a different architecture than the 6502 becoming popular though the minimalist 6502 design and TI fab improvements had synergies that revolutionized the CPU market.

Bill Mensch could have easily ended up at CBM as well. All he had to do was stay at MOS Technologies when it was purchased by CBM. Jack Tramiel had Bill reverse engineer Japanese chips and design clones. Jack offered Bill more than Chuck Peddle was payed at CBM in front of Chuck. The head of the first company Bill talked to to design the chips Jack wanted didn't want to have anything to do with the deal because it involved Jack so Bill created his own company WDC. Bill came out poorly on the deal because the chips were not produced as Jack wanted them as leverage in negotiating with Japanese companies. Bill went on to create 6502 family CMOS and 16 bit designs the latter of which were used in the Apple IIgs and SNES. Bill also claims to be responsible for the fabless semiconductor development idea.

Oral History of William David "Bill" Mensch Jr.
https://www.youtube.com/watch?v=Ne1ApyqSvm0

This could have all been Motorola or CBM with better leadership.

Larry Kaplan, Jay Miner and other valuable Atari employees left Atari because of poor leadership. This resulted in the Amiga, Mindset and Activision which could have been next generation hardware and more software from Atari. Jay had some good ideas for Atari computers including making them more compatible with the game machines since they were similar hardware and making more advanced and flexible hardware which would have helped the company to survive the cyclical video game market and the 1983 video game crash. It was the gaming PC that grew out of the crash which benefited CBM, especially the C64. The Amiga was a little too expensive and lacked software to take advantage until the Amiga 500. CBM did not value Jay Miner and the Amiga Corporation employees who they let go and cost them as Amiga development languished. This was post Jack who tried to buy the Amiga and Mindset before buying Atari and developing his own Atari ST. If Jack had a better reputation and made a reasonable offer for Amiga Corporation then they probably wouldn't have been opposed to an acquisition. Mindset, which also created advanced hardware by ex-Atari engineers, tried to go it alone and only survived about a year.

https://en.wikipedia.org/wiki/Mindset_(computer)

Mindset was a PC compatible with only about 80% compatibility which seemed to matter more in 1984 than the advanced hardware. The Amiga had more advanced hardware but less compatibility with a PC compatible or the C64 predecessor. Of the non-compatible advanced hardware introduced in the mid-'80s, the best hardware Amiga failed slowly, the most compatible hardware Mindset failed quickly, the cheapest hardware Atari ST failed slowly and the worst value hardware the Macintosh survived and Apple became one of the most valuable businesses in the world. CBM tried to buy Apple and if they had it likely would have failed due to poor leadership. CBM had good cards but discarded the winning cards. Apple had poor cards, played them well and got lucky.

 Status: Offline
Profile     Report this post  
bhabbott 
Re: Irving Gould's House
Posted on 14-May-2022 6:08:39
#43 ]
Regular Member
Joined: 6-Jun-2018
Posts: 151
From: Aotearoa

@Nonefornow

Quote:

Nonefornow wrote:

...the AtariST (after Jack bought Atari ) was released in 1985 and discontinued in 1993 and sold something like 2.1million unit.

The Amiga 1000 (After Commodore bought Amiga Corporation) was also released in 1985 and was discontinued in 1987 and sold less than 100,000 units.
Not fair, you are comparing a single Amiga model to the entire ST range. The original Atari ST had TOS on disk (using up most of the computer's RAM when loaded) and an external single-sided 360k floppy drive. Perhaps 100,000 were sold before being supplanted by later more advanced models.

Quote:
I believe the A1000 to be clearly superior to the AmigaST, but that's not what the market wanted at that time.
The A1000 wasn't supposed to be a mass market computer. Commodore didn't need huge A1000 sales because it had the C128, which sold 5.7 million units between 1985 and 1989. What they did need for the Amiga was greater familiarity and more software, which came as a result of developers and early adopters having machines to play with. In the next 2 years a lot was learned, so when the A500 and A2000 were introduced in 1987 sales rapidly increased.

But what were the alternatives? Instead of creating an advanced multitasking graphical OS for the Amiga, they could have just thrown in something like CP/M and GEM or even Microsoft BASIC. That probably would have gotten them more initial sales, but dramatically reduced the Amiga's overall appeal and eventual lifespan. That's what Jack would have done to it.

 Status: Offline
Profile     Report this post  
bhabbott 
Re: Irving Gould's House
Posted on 14-May-2022 9:05:33
#44 ]
Regular Member
Joined: 6-Jun-2018
Posts: 151
From: Aotearoa

@matthey

Quote:

matthey wrote:
It is difficult to assign blame as a percentage but things could have played out much different at critical points. Poor leadership made big differences in the early computer landscape.

Chuck Peddle, Bill Mensch and other Motorola 6800 designers left Motorola because of poor leadership. This resulted in the revolutionary low cost 6502 CPU from MOS Technology instead of Motorola. Motorola may have been a bigger name in CPUs but it is possible they would have created a 16 bit version of the 6502
If you mean like the 65C816, no thanks.

Chuck Peddle may have been a genius, but the 6502 was really just a hack that wasn't scalable to 68k level. Paddle's foray into the desktop computer market didn't turn out too well either, ending abruptly in the same year Commodore bought the Amiga - probably because his creation used an 8088 CPU but wasn't IBM compatible (facepalm).

Poor leadership abounded in the computer industry. Intel was one of the worst. Several engineers left to produce the legendary Z80, to which the most important 'feature' they added was changing the instruction mnemonics to actually memorable names. But they later lost their way trying to produce a 16 bit successor. Meanwhile Intel wasted years trying to develop a 32 bit CPU, and only got lucky with the 8088 because IBM wanted something cheap to counter the Apple II. That saved Intel, but ultimately led to IBM making a loss of over 8 billion dollars in 1992.

Trying to determine how things could have played out different at critical points is like calculating the effect of a butterfly flapping it wings. The only thing we can say for sure is that things would have been different - and quite possibly not in a good way. It could well be that Commodore's missteps were a necessary ingredient for what success the Amiga did have.

Quote:
instead of trying to jump ahead of the competition with a more aggressive 16/32 bit 68000 design.
...and that would been a shame, because the competition were making much worse 16/32 bit designs.

Quote:
Maybe they would have had room for both the 6502 for the low end and 68000 for the high end in which case they would have been an even bigger name in the embedded market.
The 68HC11 was good to work with and sold very well. I doubt the 6502 would have brought any meaningful advantage to the embedded market for Motorola.

Quote:
The legendary 65xx brand microprocessors with both 8-bit and 8/16-bit ISA's keep cranking out the unit volumes in ASIC and standard microcontroller forms supplied by WDC and WDC's licensees. Annual volumes in the hundreds (100's) of millions of units keep adding in a significant way to the estimated shipped volumes of five (5) to ten (10) billion units. With 200MHz+ 8-bit W65C02S and 100MHz+ 8/16-bit W65C816S processors coming on line in ASIC and FPGA forms, we see these annual volumes continuing for a long, long time.
Curious. All those embedded 6502s must be flying under the RADAR. I've read numerous reports on global MCU sales and WDC never gets a mention. But hey, billions of 8051 and 8 bit PIC derivatives are still being used in smart cards and USB sticks etc. too. This doesn't seem to have any relevance to Gould's house or the Amiga.

Quote:
Mindset was a PC compatible with only about 80% compatibility which seemed to matter more in 1984 than the advanced hardware.
Yep. Anything less than 100% IBM compatibility was the kiss of death.

Quote:
The Amiga had more advanced hardware but less compatibility with a PC compatible or the C64 predecessor. Of the non-compatible advanced hardware introduced in the mid-'80s, the best hardware Amiga failed slowly, the most compatible hardware Mindset failed quickly, the cheapest hardware Atari ST failed slowly and the worst value hardware the Macintosh survived and Apple became one of the most valuable businesses in the world.
Yes, the Amiga, ST and Macintosh all survived in part because they didn't pretend to be compatible. Apple made a point of it with 'think different', a brilliant marketing slogan that called a liability an asset. The ST had 'power without the price'. The Amiga didn't need a slogan - you just had to see it an action to know you had to have one.

But most people don't want to think differently, and low price or wow factor only works when the competition doesn't have it. Apple eventually hit on the winning formula - appealing to snobbery. Deliberately not compatible, not cheap, and not exciting!

Quote:
CBM tried to buy Apple and if they had it likely would have failed due to poor leadership. CBM had good cards but discarded the winning cards. Apple had poor cards, played them well and got lucky.
Commodore had some good cards, but some bad ones too. Perhaps they didn't play them as well as they could, but they stayed in the game longer than many other players.

Apple had good cards, but discarded them and put poor cards in the deck. Then they bluffed their way out of the bad hands they dealt themselves. Steve Jobs made Apple what it is today, after he almost destroyed it with poor decisions and bad leadership. Apple's success was due to a combination of genius, bravado and luck. It could just as easily have been its downfall.

The Dark Truth behind Apple's Biggest Lie

 Status: Offline
Profile     Report this post  
Hypex 
Re: Irving Gould's House
Posted on 14-May-2022 14:44:00
#45 ]
Elite Member
Joined: 6-May-2007
Posts: 10704
From: Greensborough, Australia

@bhabbott

Quote:
Which is perfectly understandable. Just imagine a bunch of customers demanding a refund because their new computer can't run PC software!


LOL. It should be obvious. It wasn't a PC!

Quote:
'Realistically' the P166 was introduced in 1996, so I think your date is a little off. However it is true that CPU speeds were continuously going up. To keep up the average owner would have to buy a new PC every 2 years or so, which the industry loved!


You're right. A family member bought a Windows '95 PC in 1996 and it had a 166Mhz CPU.

Around a similar time I had bought an A1200. And I was told I bought the wrong machine. I bought a games machine. I didn't have Road Rash 3d or Freddi Fish! LOL!

Quote:
The A1200 sold for £249 and didn't need a monitor or hard drive. An expensive PC couldn't compete with that!


It did to use as a practical computer. A computer that needed to be plugged into a TV and could only boot off floppy was a remnant of the 80's. By the 90's this home computer setup was obsolete and so old school it couldn't even be used for school. So yes it needed a monitor and a HD. I bought my A1200 with HD. But I also needed a monitor I didn't have.

Quote:
But you needed a PC to play the latest games etc., so the fact that the A1200 was much cheaper was irrelevant. You could justify the price of an expensive PC, but not an expensive Amiga because it wasn't IBM compatible.


It was never IBM compatible! Except peripheral devices that had to be switched into IBM mode. However at the time some companies still produced a DOS version or an Amiga version. Until they dropped DOS then both went.

Quote:
Where interoperability is concerned 'there can only be one'. When IBM introduced the PC in 1981 it wasn't immediately obvious that it would be 'the one', but the industry's response soon showed that it was. Eventually it would beat all rivals and leave no room for alternatives. Blaming Gould for the Amiga's demise ignores the fact that it was doomed anyway - by a flood of PC clones and consumers who didn't want a choice.


I actually wonder where this IBM PC compatible industry standard came from. In the mid 80's it wasn't around for long. How did they get it established as the only computer that should be in existence?

Did consumers have a choice? You walk into a computer shop and all they sell is PCs. Only alternate "indy" computer shops sold other machines on the market.

But, just because the Amiga days were numbered doesn't mean you just give up. If they predicted the future which they could around 1983 they wouldn't have created the Amiga at all because the PC would eventually catch up and there was no point. By the same token there was no point Apple releasing the Mac, because of PC, and eventually they dumped both the Mac hardware and OS and turned it into a PC. But, my point is, competition is life, and they could have engineered the Amiga to be a decent machine on it's own and establish a market. Sure, one day, the time would come when it's taken as far as it can and would end up like a Mac. But Apple kept at it, despite being over shadowed by the PC, and they are still around.

Quote:
Even a company with huge resources and stellar reputation would struggle under the same circumstances. Sony invented the home video recorder with Betacam, but protecting it with several patents and licensing the design to other manufacturers didn't prevent JVC from creating an inferior format that flooded the market with slightly cheaper clones. In the end there could only be one. Eventually video stores were forced to choose one format or the other, and VHS won out because more people had them. As a result the 'better' format failed, despite Sony's enormous clout.


The unofficial story is VHS was cheaper to copy porn videos with. So porn killed Beta. They preferred size over quality.

Quote:
Gould could have done everything 'right', sacrificing his own fortunes for the good of the Amiga community, and it wouldn't have helped - the Amiga would still have failed. The vitriol that Amiga fans direct towards Gould is misplaced. Commodore was - like any company - built for the sole purpose of making money for its owners. Any legacy coming out of that is a bonus, and we should be thankful that despite its problems Commodore managed to produce something of beauty that we can still cherish today.


I think investing money in a moving platform to encourage growth would be better. Rather than investing in bitrot. The engineers were already doing R&D. Like AAA and A3000 AA+. Jay had the Ranger design. It was left to stagnate. Of course the OS needed a bit of work and ended up being too sloppy so not the best for any pro work.

Quote:
The A1200 is a wonderful machine and I love it. That this model might not have existed had Commodore not followed the path it did with Gould's input makes it even more valuable. It could have been so much worse.


I loved it as well. But at the time I didn't know much else. And now in retrospect it looks rather flawed with competing machines. It left the Amiga in a better state than the A500. But a lot of Amiga people left after the A500 and now the Amiga scene gives more respect to the A500.

Last edited by Hypex on 14-May-2022 at 02:47 PM.

 Status: Offline
Profile     Report this post  
matthey 
Re: Irving Gould's House
Posted on 14-May-2022 22:10:58
#46 ]
Super Member
Joined: 14-Mar-2007
Posts: 1603
From: Kansas

bhabbott Quote:

If you mean like the 65C816, no thanks.

Chuck Peddle may have been a genius, but the 6502 was really just a hack that wasn't scalable to 68k level. Paddle's foray into the desktop computer market didn't turn out too well either, ending abruptly in the same year Commodore bought the Amiga - probably because his creation used an 8088 CPU but wasn't IBM compatible (facepalm).


I do not agree that the 6502 is a hack but the minimalist design leaves little room for scalability or much of anything else. The minimal design was important early for mass produced CPUs as there was barely enough space for a CPU on a single chip. This shifted later where more features, better code density and better performance became more important and why the 68000 was fortunately chosen for the Amiga. Development of the Victor 9000/Sirius 1 started before the IBM PC was released and it was released just a few months after in 1981. Making it fully compatible would have likely delayed release. There were many partial IBM PC compatible computers early on though it became more detrimental as the competition became more compatible. The Victor 9000/Sirius 1 was in many ways ahead of its time but the technology advantage didn't create a winner as with the later Mindset and Amiga.

bhabbott Quote:

Poor leadership abounded in the computer industry. Intel was one of the worst. Several engineers left to produce the legendary Z80, to which the most important 'feature' they added was changing the instruction mnemonics to actually memorable names. But they later lost their way trying to produce a 16 bit successor. Meanwhile Intel wasted years trying to develop a 32 bit CPU, and only got lucky with the 8088 because IBM wanted something cheap to counter the Apple II. That saved Intel, but ultimately led to IBM making a loss of over 8 billion dollars in 1992.


Intel had issues and was behind the technology curve for awhile. They remained conservative and persevered though unlike Motorola who threw away the 68k baby with the bathwater. The 8086 was a good CPU with good code density especially for programs using text like business programs. It also had good performance. The 8088 was a weak cheapened variation but at least it was mostly compatible with the 8086. The 68k was better in most ways but bigger in many ways.

bhabbott Quote:

Trying to determine how things could have played out different at critical points is like calculating the effect of a butterfly flapping it wings. The only thing we can say for sure is that things would have been different - and quite possibly not in a good way. It could well be that Commodore's missteps were a necessary ingredient for what success the Amiga did have.


Or a feather drifting in the wind?

Forrest GumpQuote:

Lieutenant Dan got me invested in some kind of fruit company. So then I got a call from him, saying we don't have to worry about money no more. And I said, that's good! One less thing.


Forrest Gump's $100k in Apple stock is now worth almost $49 billion
https://www.imore.com/forrest-gumps-100k-apple-stock-now-worth-almost-49-billion

Is there an invisible hand behind the wind?

bhabbott Quote:

...and that would been a shame, because the competition were making much worse 16/32 bit designs.


As I recall, there wasn't much 16/32 bit microcomputer CPU competition when the 68000 came out in 1979. Minicomputers supporting 32 bit hadn't been out long like the VAX which had only been released in 1977. There was a vision by leadership at Motorola to replace the time share minicomputer with more convenient personal computer workstations which the 68000 achieved. Most of the other microcomputer competition was moving to 8/16 bit core designs instead of 16/32 designs. This is why I say the 68000 jumped ahead of the competition in technology and revolutionized the PC market. Too bad the IBM leadership couldn't see it coming but maybe they were too worried about the 68000 being too powerful and threatening their minicomputer market which it did even though the PC revolution couldn't be stopped by using a weak 8088 in the IBM PC.

bhabbott Quote:

The 68HC11 was good to work with and sold very well. I doubt the 6502 would have brought any meaningful advantage to the embedded market for Motorola.


Motorola made some nice 6800 family CPUs for the embedded market which were nicer than the 6502 to work with. They were practically in a different category though.

6502 3510 transistors
6800 4100 transistors
Z80 8500 transistors
6809 9000 transistors
8086 29000 transistors
68000 68000 transistors

The 6809 was one of the nicest to work with 8 bit computers but had trouble competing. In 1980 a 6809 in single-unit quantities was $37 compared to $9 for a Zilog Z80 and $6 for a 6502. If price and power are more important then the 6502 had a significant advantage over the 6800 family. If performance is more important then there were better choices as well.

BYTE Sieve in seconds (lower is better)
6502 1 MHz 13.9
Z80 4 MHz 6.8
6809 2 MHz 5.1
8086 8 MHz 1.9
68000 8 MHz 0.49

https://en.wikipedia.org/wiki/Motorola_6809

bhabbott Quote:

Curious. All those embedded 6502s must be flying under the RADAR. I've read numerous reports on global MCU sales and WDC never gets a mention. But hey, billions of 8051 and 8 bit PIC derivatives are still being used in smart cards and USB sticks etc. too. This doesn't seem to have any relevance to Gould's house or the Amiga.


I expect most of the 6502 family cores are used under license in minimalist SoCs. The 6502 has something like 10%-15% of the transistors of a minimal ARM core which should provide a modest power savings as long as the code is small (the 6502 does not have good code density). The embedded market is huge and 6502 units of hundreds of millions annually is not leading for the embedded market.

ARM 26.8 billion units annually
ARC 1.75 billion units annually
PC 275 million units annually

842 Chips Per Second: 6.7 Billion Arm-Based Chips Produced in Q4 2020
https://www.tomshardware.com/news/arm-6-7-billion-chips-per-quarter

Annual 6502 units are closer to PC market units which is relatively small compared to total embedded market volumes. That would explain why the 6502 is rarely mentioned for embedded markets. The 68k family may still sell millions of units annually too even though it has been decades since it was the top seller in the 32 bit embedded market.

bhabbott Quote:

Yep. Anything less than 100% IBM compatibility was the kiss of death.


Especially as late as 1984 when Mindset computers were introduced. Compatibility won out over technology advantages. It took about another decade but PC clones eventually had the technology advantages and compatibility. AmigaNGs also tried to push the Amiga forward with less than 100% Amiga compatibility and THEA500 Mini sales have made it obvious that full compatibility is important.

Jay Miner Quote:

In addition, all present Amiga software is compatible with all Amigas, big and small, old and new. I want to repeat that because I think that is one of the most important features of the Amiga. All present Amiga software is compatible with all Amigas, big and small, old and new. Those two features expandability and compatibility are I think unique in the personal computer industry and in the game industry and a big advantage in both of those fields yet it isn't even advertised.


The History of the Commodore Amiga - Rare Jay Miner Speech AmiExpo 1990
https://youtu.be/n-MqC35aWrQ?t=1026

The great computer visionaries keep repeating themselves but are ignored. It's incredibly difficult to build a new platform which is why it is better to build on an existing one. Minimalist computers are still useful just like minimalist CPUs. Actually, the 68k Amiga wasted a few transistors for flexibility but in the overall scheme it is minimalist hardware with a minimal footprint.

bhabbott Quote:

Yes, the Amiga, ST and Macintosh all survived in part because they didn't pretend to be compatible. Apple made a point of it with 'think different', a brilliant marketing slogan that called a liability an asset. The ST had 'power without the price'. The Amiga didn't need a slogan - you just had to see it an action to know you had to have one.

But most people don't want to think differently, and low price or wow factor only works when the competition doesn't have it. Apple eventually hit on the winning formula - appealing to snobbery. Deliberately not compatible, not cheap, and not exciting!


lol

bhabbott Quote:

Commodore had some good cards, but some bad ones too. Perhaps they didn't play them as well as they could, but they stayed in the game longer than many other players.

Apple had good cards, but discarded them and put poor cards in the deck. Then they bluffed their way out of the bad hands they dealt themselves. Steve Jobs made Apple what it is today, after he almost destroyed it with poor decisions and bad leadership. Apple's success was due to a combination of genius, bravado and luck. It could just as easily have been its downfall.

The Dark Truth behind Apple's Biggest Lie


Apple liked to reshuffle the deck all right. Steve Jobs had some good leadership traits and bad ones. His overall vision was close enough even if the individual steps were sometimes awkward. The Macintosh was anything but revolutionary hardware wise with the 68000 being about the only thing good about it even though it was released nearly 5 years earlier. His GUI and software fared better often being copied though it needed to be redone. The Amiga hardware blew it away and the Amiga OS was better though less polished. The lack of software was an Achilles heel for most 68k based systems at launch. When we started to get good 68k software, the 68k was abandoned to inferior x86 and RISC hardware.

Last edited by matthey on 14-May-2022 at 10:33 PM.
Last edited by matthey on 14-May-2022 at 10:24 PM.

 Status: Offline
Profile     Report this post  
bhabbott 
Re: Irving Gould's House
Posted on 15-May-2022 9:51:52
#47 ]
Regular Member
Joined: 6-Jun-2018
Posts: 151
From: Aotearoa

@matthey

Quote:

matthey wrote:
Intel had issues and was behind the technology curve for awhile. They remained conservative and persevered though unlike Motorola who threw away the 68k baby with the bathwater.
Yes, Motorola made the same mistake that Intel did, not understanding that the market had changed from wanting innovation to compatibility. Intel quickly corrected course, but Motorola didn't because they believed the hype about RISC (which IBM was also pushing).

Quote:
The 8086 was a good CPU with good code density especially for programs using text like business programs. It also had good performance. The 8088 was a weak cheapened variation but at least it was mostly compatible with the 8086.
The 8086 and 8088 were virtually identical except for the different data bus size. Code density was helped by using single byte opcodes for 16 bit register operations, and two byte opcodes for the less used 8 bit operations. This is opposite to typical 8 bit CPUs with 16 bit extensions.

The other thing that helped was separating the code fetch unit from the execution unit. This acted like a cache, often allowing the 8088 to run at close to the same speed as the 8086 because fewer bus cycles went unused while the CPU was busy executing instructions (in comparison the 6502 and 6809 both waste a lot of bus cycles while doing internal operations, making them slower than they could be). OTOH it also meant that the 8086 wasn't that much faster than the 8088, which made it less popular because the extra expense of implementing a 16 bit bus was hardly worth it.

68k code is also pretty dense. When comparing identical programs compiled for 68k Amiga and x86 AROS, the 68k version is often significantly smaller. Hand-crafted 68k assembler can also be more compact and faster than x86.

Quote:
Forrest Gump's $100k in Apple stock is now worth almost $49 billion
https://www.imore.com/forrest-gumps-100k-apple-stock-now-worth-almost-49-billion

Is there an invisible hand behind the wind?
Yes, and its name is Bill Gates. He decided that Apple was worth saving from oblivion.

Quote:
As I recall, there wasn't much 16/32 bit microcomputer CPU competition when the 68000 came out in 1979.
There were other 16 and 32 bit designs, but they were either flawed, buggy, late or stupidly expensive.

Here's an interesting story about how the TMS9900 and 68000 lost out to the 8088 in the IBM PC:-

The Inside Story of Texas Instruments’ Biggest Blunder: The TMS9900Quote:
In October 1978, six months after the announcement of the Intel 8086, I moved to TI’s MOS Division and became the manager for microprocessors. By this time, everyone at the company, and many people outside the company, knew that TI’s 16-bit microprocessor strategy wasn’t working. Compounding that problem was the division’s largely unsuccessful attempt to develop a compatible 16-bit microcontroller, called the TMS9940, which was in its fifth or sixth re-spin by the time I arrived...

Shortly after I arrived in Houston, I was told that I would need to give a presentation on the TMS9900 to a group from IBM that was working on a very secret project that required a 16-bit microprocessor. The group came from a rather unusual location for IBM: Boca Raton, Florida...

Selection of a 16-bit microprocessor by the IBM team couldn’t have been much of a debate. The Motorola 68K, as it was later known, was undoubtedly the hands-down winner. It had the largest logical address space, which was even more important than the minimum 16-bit internal architecture. It was also easily expandable to a full-fledged 32-bit architecture. And, most important, the 68K was a “Big Endian,” unlike the other contenders...

So why aren’t we all using 68K-based computers today?

The answer comes back to being first to market. Intel’s 8088 may have been imperfect but at least it was ready, whereas the Motorola 68K was not. And IBM’s thorough component qualification process required that a manufacturer offer up thousands of “production released” samples of any new part so that IBM could perform life tests and other characterizations. IBM had hundreds of engineers doing quality assurance, but component qualifications take time. In the first half of 1978, Intel already had production-released samples of the 8088. By the end of 1978, Motorola’s 68K was still not quite ready for production release.


Note the time frame. IBM needed production quantities for testing in 1978, but didn't release the PC until 1981. This despite the fact that the PC used only 'industry standard' parts and a very conventional non-critical design. And people accused Commodore of being slow getting the Amiga out...

Quote:
Most of the other microcomputer competition was moving to 8/16 bit core designs instead of 16/32 designs. This is why I say the 68000 jumped ahead of the competition in technology and revolutionized the PC market.
...or would have, if was it little earlier to market.

Quote:
Too bad the IBM leadership couldn't see it coming but maybe they were too worried about the 68000 being too powerful and threatening their minicomputer market
I've heard that too but it doesn't seem to be the case. IBM management gave the PC team full control over what they put in it, with no strings attached apart from quality control.

The truth is, Motorola had an excellent design but just couldn't keep up with Intel.

Quote:
6502 3510 transistors
6800 4100 transistors
Z80 8500 transistors
6809 9000 transistors
8086 29000 transistors
68000 68000 transistors
... and here we can see why. More transistors = lower yields and more expansive chips that are harder to make. Whoever has the best process wins. Motorola was always behind Intel in that respect.

Quote:
The 6809 was one of the nicest to work with 8 bit computers but had trouble competing. In 1980 a 6809 in single-unit quantities was $37 compared to $9 for a Zilog Z80 and $6 for a 6502.

CPU price isn't that important, since there is only one of them in the computer. How easily the CPU integrates with the rest of the machine is often more important.

In 1980 I built a computer of my own design using an MC6800 and MC6847 video display generator. I figured out how to use alternate halves of the bus cycle without contention, using a simple synchronization circuit. I also used the new HM6116 2k static RAM chips that others eschewed because they were more expensive, but in the end were cheaper because they needed fewer support chips. A year later I bought a 6809 bundled with programmer's manual for NZ$25, but never used it because I got a ZX81 and got hooked on the Z80. This was a pity because the 6809 has a very nice instruction set, but I didn't have the time or skill to create a sophisticated OS for it.

Some day I would like to reproduce my design, but with a 6809 in place of the 6800. I lost the plans long ago, but It only used a few TTL 'glue' chips so shouldn't be hard to 'remember' the circuit. Today I can design a professional looking PCB on a PC and have it made in China for a few bucks, so building it should be a piece of cake. The most expensive part will be the keyboard!

Quote:
Annual 6502 units are closer to PC market units which is relatively small compared to total embedded market volumes. That would explain why the 6502 is rarely mentioned for embedded markets. The 68k family may still sell millions of units annually too even though it has been decades since it was the top seller in the 32 bit embedded market.
I've been working in the embedded market for decades and never heard of anyone using 6502. Perhaps some companies are using the IP in FPGAs and ASICs, but again this seems to be 'under the RADAR'.


Quote:
Jay Miner Quote:

In addition, all present Amiga software is compatible with all Amigas, big and small, old and new. I want to repeat that because I think that is one of the most important features of the Amiga. All present Amiga software is compatible with all Amigas, big and small, old and new. Those two features expandability and compatibility are I think unique in the personal computer industry and in the game industry and a big advantage in both of those fields yet it isn't even advertised.


The History of the Commodore Amiga - Rare Jay Miner Speech AmiExpo 1990
https://youtu.be/n-MqC35aWrQ?t=1026

The great computer visionaries keep repeating themselves but are ignored. It's incredibly difficult to build a new platform which is why it is better to build on an existing one.

True, but at some point you have to make a clean break in order to progress. The break between 8 bit and 16 bit home computers was difficult, but necessary. 8 bit machines that tried to build much beyond the 64k limit failed because they were too complicated and couldn't achieve the desired performance.

It was easier with the PC due to its loosely coupled subsystems. However maintaining compatibility became harder as the technology advanced. A modern PC has a lot of stuff in it for legacy support, but still isn't 100% compatible. The only reason this isn't an issue is that (almost) nobody wants to run old hardware and software on a new PC. Even in the early days it was a problem. remember all those clones with a turbo button to reduce the CPU speed? The video cards with switches on the back to select different video modes? Cards that had to be 'poked' to get them to run in 16 bit mode? Then the PCI bus came out and many cards dropped full ISA compatibility, using software emulation to make up for it (which didn't always work).

We are fortunate that Jay Miner and Co. designed the Amiga to be expandable without losing compatibility. However it gets harder as the power of the machine is increased. At some point one wonders why it is necessary. You want more power to run fancy 3D games and the latest web browsers? Why not just use a PC? (they do it out of the box you know).

I have an A600 with Vampire V2 and it's cool to run a 1024x768 screen, run IBrowse in glorious 24 bit color, and play Duke Nukem in 640x480. But I still prefer to use my A1200 with 50MHz 030 hooked up to my TV in composite with an 8 color Workbench. Why? By not asking too much of it I get snappy operation. My old eyes easily can see the big bold text and graphics without a magnifying glass. Paula sounds great on my stereo system with big speakers (don't need 16 bit 44kHz audio because my high frequency hearing is shot). There are a ton of great games that don't need an 80MHz 060 and RTG, and IBrowse works fast enough to be useful (download speeds of 100~200kB/s are fine when you are only pulling in a few MB). I was happy with it in 1995, and I'm still happy with it today. It will never get old for me.

Quote:
The lack of software was an Achilles heel for most 68k based systems at launch. When we started to get good 68k software, the 68k was abandoned to inferior x86 and RISC hardware.

I remember the early days when good software for the Amiga was rare and expensive. This changed later though. We now have more games than I could play in a lifetime, lots of excellent apps, programming languages and utilities, thousands of songs to listen to, a ton of PD stuff with more coming out all the time, as well as the ability to port stuff from other platforms. And we don't need hundreds of unreliable floppies to store it all on, or be worried about original copy-protected disks going bad. There has never been a better time to own an Amiga than right now!

Last edited by bhabbott on 15-May-2022 at 09:58 AM.
Last edited by bhabbott on 15-May-2022 at 09:57 AM.

 Status: Offline
Profile     Report this post  
QBit 
Re: Irving Gould's House
Posted on 15-May-2022 10:32:17
#48 ]
Regular Member
Joined: 15-Jun-2018
Posts: 217
From: Unknown

@all

In the Middle Ages Galileo Galilei got executed for telling an advanced World View.

Now the Pope admits Evolution is true

Now there are Flat Earthers trying to convince everybody Earth is flat.

Now any PC and any Mac is a Multimedia Machine!

What was the first Multi Media Computer?

 Status: Offline
Profile     Report this post  
Nonefornow 
Re: Irving Gould's House
Posted on 15-May-2022 18:26:10
#49 ]
Regular Member
Joined: 29-Jul-2013
Posts: 308
From: Greater Los Angeles Area

@bhabbott

Quote:
The A1000 wasn't supposed to be a mass market computer. Commodore didn't need huge A1000 sales because it had the C128, which sold 5.7 million units between 1985 and 1989.


Margaret Morabito - technical manager of RUN magazine, who still writes about computers and the c128 - in her introductory article about the A1000 listed the following 3 market targets for the A1000:

First - Business market
Next - Educational market
Next - Professional artists and musicians.

To note that the gaming market was not listed. And possibly because she knew that in 1985 the C64 and the C128 were the machines that CBM intended for that market.

Maybe CBM did not need huge A1000 sales, but certainly wanted them.

The original article appears on Amiga World - Premiere Issue Sep / Oct 1985.

 Status: Offline
Profile     Report this post  
Turrican3 
Re: Irving Gould's House
Posted on 16-May-2022 11:35:37
#50 ]
Regular Member
Joined: 20-Jun-2003
Posts: 368
From: Italy

@Nonefornow

Quote:
To note that the gaming market was not listed

That shouldn't come as a surprise, I mean how could you compete with a sub-200$ Nintendo NES (or even your own 8 bit line, as you said) with a 1000+ dollars workstation, albeit hugely more powerful?

There's a reason Amiga sales started to become significant only with the cost-reduced version, aka the A500.

Last edited by Turrican3 on 16-May-2022 at 11:36 AM.

 Status: Offline
Profile     Report this post  
QBit 
Re: Irving Gould's House
Posted on 16-May-2022 11:59:12
#51 ]
Regular Member
Joined: 15-Jun-2018
Posts: 217
From: Unknown

@bhabbott

Quote:
Nonefornow wrote:

...the AtariST (after Jack bought Atari ) was released in 1985 and discontinued in 1993 and sold something like 2.1million unit.

The Amiga 1000 (After Commodore bought Amiga Corporation) was also released in 1985 and was discontinued in 1987 and sold less than 100,000 units.
Not fair, you are comparing a single Amiga model to the entire ST range. The original Atari ST had TOS on disk (using up most of the computer's RAM when loaded) and an external single-sided 360k floppy drive. Perhaps 100,000 were sold before being supplanted by later more advanced models.

Quote:
I believe the A1000 to be clearly superior to the AmigaST, but that's not what the market wanted at that time.

KICKTOS: Kickstart disk for the Amiga 1000 with Atari TOS
https://www.youtube.com/watch?v=K4ta4OYhsEY

 Status: Offline
Profile     Report this post  
matthey 
Re: Irving Gould's House
Posted on 17-May-2022 6:26:53
#52 ]
Super Member
Joined: 14-Mar-2007
Posts: 1603
From: Kansas

bhabbott Quote:

The 8086 and 8088 were virtually identical except for the different data bus size. Code density was helped by using single byte opcodes for 16 bit register operations, and two byte opcodes for the less used 8 bit operations. This is opposite to typical 8 bit CPUs with 16 bit extensions.

The other thing that helped was separating the code fetch unit from the execution unit. This acted like a cache, often allowing the 8088 to run at close to the same speed as the 8086 because fewer bus cycles went unused while the CPU was busy executing instructions (in comparison the 6502 and 6809 both waste a lot of bus cycles while doing internal operations, making them slower than they could be). OTOH it also meant that the 8086 wasn't that much faster than the 8088, which made it less popular because the extra expense of implementing a 16 bit bus was hardly worth it.

68k code is also pretty dense. When comparing identical programs compiled for 68k Amiga and x86 AROS, the 68k version is often significantly smaller. Hand-crafted 68k assembler can also be more compact and faster than x86.


808x is the closest to being on par with the 68k in code density. The 808x often has better code density when using small datatypes in tiny programs while the 68k wins with larger datatypes and larger programs. The 68k has a much more flexible ISA with more GP registers that is easier to target with a compiler than the 808x though. I believe the 68k has better code density than the Z80, x86, PDP-11, VAX, ARM Thumb2, ARM Thumb and SuperH as well. Prefetching data was the predecessor to pipelining and when a 16 bit instruction could be prefetched then optimizing for a datatype size was no longer important as 8, 16, 32 and 64 bit could be supported nearly equally.

bhabbott Quote:

Note the time frame. IBM needed production quantities for testing in 1978, but didn't release the PC until 1981. This despite the fact that the PC used only 'industry standard' parts and a very conventional non-critical design. And people accused Commodore of being slow getting the Amiga out...


IBM is still slow to adapt despite some of the best employees and technology in the computer industry. They have been too conservative at times. The choice of 8088 for the IBM PC was an example whether it was for testing and qualification reasons, availability and supply reasons or the existing business relationship with Intel. In 3 years, they could have produced their own CPU design but they probably weren't convinced of the viability of the PC market.

bhabbott Quote:

The truth is, Motorola had an excellent design but just couldn't keep up with Intel.

Quote:
6502 3510 transistors
6800 4100 transistors
Z80 8500 transistors
6809 9000 transistors
8086 29000 transistors
68000 68000 transistors
... and here we can see why. More transistors = lower yields and more expansive chips that are harder to make. Whoever has the best process wins. Motorola was always behind Intel in that respect.


The 68000 was in a different class compared to the 808x. While the 808x was used by many small business computers for text based programs, the 68k was the CPU of choice for PC workstations with bitmap capabilities and more memory. The 68000 competed more with the PDP-11 and VAX minicomputers. Intel didn't have a competitor for the 68000 released in 1979 until the 80286 was released in 1982. Even then, the 68000 only had 68000 transistors while the 80286 had 134,000 transistors. The equivalent 68k CPUs had the transistor advantage over x86 CPUs which increased yields, lowered production costs and reduced power. The 68k was even supporting twice as many GP registers while x86 was supporting old 808x segmented memory compatibility.

bhabbott Quote:

We are fortunate that Jay Miner and Co. designed the Amiga to be expandable without losing compatibility. However it gets harder as the power of the machine is increased. At some point one wonders why it is necessary. You want more power to run fancy 3D games and the latest web browsers? Why not just use a PC? (they do it out of the box you know).

I have an A600 with Vampire V2 and it's cool to run a 1024x768 screen, run IBrowse in glorious 24 bit color, and play Duke Nukem in 640x480. But I still prefer to use my A1200 with 50MHz 030 hooked up to my TV in composite with an 8 color Workbench. Why? By not asking too much of it I get snappy operation. My old eyes easily can see the big bold text and graphics without a magnifying glass. Paula sounds great on my stereo system with big speakers (don't need 16 bit 44kHz audio because my high frequency hearing is shot). There are a ton of great games that don't need an 80MHz 060 and RTG, and IBrowse works fast enough to be useful (download speeds of 100~200kB/s are fine when you are only pulling in a few MB). I was happy with it in 1995, and I'm still happy with it today. It will never get old for me.


Why use a PC compatible when a powerful 68k Amiga could be produced for a similar cost to the Raspberry Pi?

bhabbott Quote:

I remember the early days when good software for the Amiga was rare and expensive. This changed later though. We now have more games than I could play in a lifetime, lots of excellent apps, programming languages and utilities, thousands of songs to listen to, a ton of PD stuff with more coming out all the time, as well as the ability to port stuff from other platforms. And we don't need hundreds of unreliable floppies to store it all on, or be worried about original copy-protected disks going bad. There has never been a better time to own an Amiga than right now!


The problem is that the old Amiga hardware is expensive and failing. Also, the old Amigas don't have modern I/O capabilities. Amiga nostalgia is alive but the old hardware is not enough to sustain it. Most Amiga users end up using an Amiga emulator which is not a development target so there is limited development. THEA500 mini revives the nostalgia but can't sustain it with emulation either.

Turrican3 Quote:

That shouldn't come as a surprise, I mean how could you compete with a sub-200$ Nintendo NES (or even your own 8 bit line, as you said) with a 1000+ dollars workstation, albeit hugely more powerful?

There's a reason Amiga sales started to become significant only with the cost-reduced version, aka the A500.


Rumor is that the Amiga 500 had a higher profit margin than the C128 which makes sense. CBM could have improved Amiga integration as the years went by to save cost. The Amiga engineers who went to 3DO reduced the custom chip count to one chip in about 1 year while CBM was only planning to reduce the AA+ custom chip count from 3 to 2 chips after nearly 10 years. CBM was also considering licensing the 68k from Motorola for a single SoC with custom chips but management spent too much time exploring x86 and RISCier Amiga replacements.

Last edited by matthey on 17-May-2022 at 06:28 AM.

 Status: Offline
Profile     Report this post  
bhabbott 
Re: Irving Gould's House
Posted on 17-May-2022 10:24:46
#53 ]
Regular Member
Joined: 6-Jun-2018
Posts: 151
From: Aotearoa

Quote:

Nonefornow wrote:

Margaret Morabito - technical manager of RUN magazine, who still writes about computers and the c128 - in her introductory article about the A1000 listed the following 3 market targets for the A1000:

First - Business market
Next - Educational market
Next - Professional artists and musicians.

To note that the gaming market was not listed. And possibly because she knew that in 1985 the C64 and the C128 were the machines that CBM intended for that market.

Possibly. Or perhaps she didn't fully appreciate the Amiga's advantages for games? To developers it was obvious that the Amiga would take gaming to the next level. All that talk about art, animation, video and music only meant one thing to us - games that would knock your socks off!

As for the business market, Commodore and others tried to pitch their lesser 8 bit home computers for business too, but it didn't fool anyone. The Amiga was certainly much better in this regard, but the essentials were still missing - a hard drive, high resolution non-interlace display, and of course most importantly - IBM compatibility. I don't know where Margaret Morabito got the idea that the Amiga's first target would be the business market, as it obviously wasn't doing that. Perhaps being the technical manager of RUN magazine is a clue to her cluelessness.

Quote:
Maybe CBM did not need huge A1000 sales, but certainly wanted them.

Sure it wanted them, but the A1000 simply wasn't ready for the mass market. I bought one in 1987. It came with KS1.1 and an upgrade to the new KS1.2 which the A500 had. KS 1.1 was clunky and horribly unstable - I managed to lock up the Workbench in less than a minute - and this was the second version released to A1000 users :(. I loved the styling of the A1000, but it was clearly a 'concept' model with several design issues (that were fixed in the A500 and A2000).

Quote:
The original article appears on Amiga World - Premiere Issue Sep / Oct 1985.

Ah, the memories! Reading that premiere issue again 37 years later reminded me of the excessive hype and outrageous claims that were common in the computer industry back then. The entire issue read like an extended advert for the Amiga. The adverts for vapourware were also 'interesting'.

I threw away my collection of Amiga World magazines when I moved house 6 years ago, and now I'm regretting it (how was I to know the Amiga would came roaring back to life as a 'retro' computer!).

 Status: Offline
Profile     Report this post  
bhabbott 
Re: Irving Gould's House
Posted on 17-May-2022 12:25:32
#54 ]
Regular Member
Joined: 6-Jun-2018
Posts: 151
From: Aotearoa

@matthey

Quote:

matthey wrote:

Why use a PC compatible when a powerful 68k Amiga could be produced for a similar cost to the Raspberry Pi?

Because I have more powerful PCs that cost me nothing and run the software I want to use. And I'll be honest - Windows XP on a semi-modern PC is great. I have no desire to make my Amigas take over its role (or vice versa).

Quote:
The problem is that the old Amiga hardware is expensive and failing.

Yes, this is a problem. But it is being solved. We have replacement motherboards, cases, keyboards, and custom chips. The increasing rareness and price of original hardware is providing more incentive to repair and maintain them. I also suspect that a lot of people are hoarding spares, and I know of two people locally who have Amigas in their cupboards that they want to get going again 'some day'.

Quote:
Also, the old Amigas don't have modern I/O capabilities. Amiga nostalgia is alive but the old hardware is not enough to sustain it.

Exactly what 'modern' I/O capabilities do Amigas need that aren't available?

I actually have less need for I/O on my Amigas today than I did back in the 90's. My A1200 and A600 are networked via second-hand PCMCIA Ethernet cards that I bought for $10 each on TradeMe. CF Cards and PCMCIA or IDE adapters are cheap, so no need for a SCSI interface. In the rare case where I need to print or scan something I use the PC.

Quote:
Most Amiga users end up using an Amiga emulator which is not a development target so there is limited development.

Actually you are wrong there - Amiga emulators are now a development target. A500 emulation is so accurate now that most software can be developed using an emulator knowing that it will work on the real thing too. I have done the same thing on other retro computers like the Amstrad CPC and Mattel Aquarius, as debugging is much easier to do in the emulator.


Quote:
THEA500 mini revives the nostalgia but can't sustain it with emulation either.

The THEA500 mini is reaching mainstream markets that real Amiga hardware can't today. It's cheap enough that people will buy one just to try it out, and hackable enough to achieve great things. I predict it will cause a huge resurgence of interest in the Amiga. I can see other models coming out too (CD32 mini, anyone?).

Quote:
The Amiga engineers who went to 3DO reduced the custom chip count to one chip in about 1 year

3DO
, whats that? Quote:
Despite having a highly promoted launch (including being named Time magazine's "1993 Product of the Year") and being a host of cutting-edge technologies, the 3DO's high price and an oversaturated console market prevented the system from achieving success

Seems squashing the hardware into a single chip didn't help.

CBM couldn't do it because their chip foundry wasn't set up to do anything larger than 84 pin PLCC. But what they could make they were able to make very cheaply. That's one reason they were able to make good profit on the A500. Another was low cost through-hole 2 layer PCBs. The A500 PCB was large, but cheap to make and assemble with the equipment and manpower available at the time. When they made the first much smaller surface mount A600 motherboards in Ireland the cost was higher. It's not just the custom chips that determine the price.

Quote:
CBM was only planning to reduce the AA+ custom chip count from 3 to 2 chips after nearly 10 years.

So? Some other home computers also got enhanced customs chips with higher integration, but it didn't save them from extinction. Amstrad's CPC range was so successful that it got Alan Sugar a knighthood, but the CPC Plus range was a poor seller despite having greatly enhanced features. The 5 year gap between it and the original CPC series may have been the reason it failed.

If Commodore had managed to squeeze the AGA chipset into two packages, would it have helped? The A1200 was already cheap, so I doubt knocking a few more bucks off the price wouldn't have done much. Ironically the biggest impediment to A1200 sales was actually that Commodore couldn't make them fast enough!

Quote:
CBM was also considering licensing the 68k from Motorola for a single SoC with custom chips but management spent too much time exploring x86 and RISCier Amiga replacements.

Which would be a pity because such a highly integrated custom chip would be less hackable and less interesting, as well as harder to repair. This is the way PCs went, to the point where motherboards became virtually unrepairable (I remember fixing a rather unique Philips PC that we couldn't get a replacement chip for. I scavenged a replacement chip off a new OEM motherboard that had the same chipset, destroying it in the process. What a waste!).


 Status: Offline
Profile     Report this post  
matthey 
Re: Irving Gould's House
Posted on 17-May-2022 20:58:21
#55 ]
Super Member
Joined: 14-Mar-2007
Posts: 1603
From: Kansas

bhabbott Quote:

Because I have more powerful PCs that cost me nothing and run the software I want to use. And I'll be honest - Windows XP on a semi-modern PC is great. I have no desire to make my Amigas take over its role (or vice versa).


Windows XP is very old and few 3rd party programs and hardware support it anymore. Even Windows 7 is losing support and some 3rd party programs are requiring a newer version of Windows often for no obvious reason. When enough users have moved on, the older versions of Windows will be as orphaned as the AmigaOS using out of date 3rd party programs. Actually, the AmigaOS is more supported at this point than Windows XP as there have been updates more recently. Microsoft wants you to upgrade to their bigger and more closed Windows versions.

bhabbott Quote:

Yes, this is a problem. But it is being solved. We have replacement motherboards, cases, keyboards, and custom chips. The increasing rareness and price of original hardware is providing more incentive to repair and maintain them. I also suspect that a lot of people are hoarding spares, and I know of two people locally who have Amigas in their cupboards that they want to get going again 'some day'.


The new replacement hardware is more expensive that the original Amiga hardware inflated prices. Yes, some people have spare Amigas to use as donor parts computers for the ones that still work. People are storing Amigas because it is too expensive to repair and maintain them. A few Amiga users will be left with valuable Amiga antiques but this is not the way to grow the Amiga user base. The PPC Amiga like hardware for the classes instead of the masses has the same problem.

bhabbott Quote:

Exactly what 'modern' I/O capabilities do Amigas need that aren't available?

I actually have less need for I/O on my Amigas today than I did back in the 90's. My A1200 and A600 are networked via second-hand PCMCIA Ethernet cards that I bought for $10 each on TradeMe. CF Cards and PCMCIA or IDE adapters are cheap, so no need for a SCSI interface. In the rare case where I need to print or scan something I use the PC.


Most monitors and tvs today have digital video inputs which the Amiga did not have. HDMI is also the most common way to hook up sound which is convenient. Amiga IDE and PCMCIA transfers are slow and SCSI adapters that avoid a bottleneck are not cheap. Ethernet and Wi-Fi hardware are not cheap except for PCMCIA and network performance is usually poor. USB is the most common general purpose I/O today yet USB hardware is among the rarest of Amiga hardware. Many Amiga computers are completely isolated without networking support or USB thumb flash drive support. Amiga users often give up and throw the computers in storage as I/O solutions are too expensive and then leave the Amiga community or switch to emulation which has modern I/O support.

bhabbott Quote:

Actually you are wrong there - Amiga emulators are now a development target. A500 emulation is so accurate now that most software can be developed using an emulator knowing that it will work on the real thing too.


No. The Amiga 500 is the development target. Code is optimized for a real 68000 CPU. Also, the Amiga 500 is the emulation target.

Compiler->real hardware target
Emulation->real hardware target

You can say that Amiga emulation like UAE is for an Amiga virtual machine which it is but no compiler generates code for that hardware target as it varies depending on the host hardware and OS. No Amiga compiler generates code for a FPGA CPU target. All Amiga development is for a real Amiga hardware target no matter how old and antiquated it is. A significant amount of Amiga code, including the AmigaOS 3 updates, is still compiled for a 68000 target which is a 40 year old 16 bit non-pipelined CPU with no caches even though it is often executed on a 68020+ pipelined full 32 bit CPU with caches even on FPGA CPU cores.

bhabbott Quote:

The THEA500 mini is reaching mainstream markets that real Amiga hardware can't today. It's cheap enough that people will buy one just to try it out, and hackable enough to achieve great things. I predict it will cause a huge resurgence of interest in the Amiga. I can see other models coming out too (CD32 mini, anyone?).


THEA500 Mini is so handicapped that it has to be hacked just to get to Workbench but at least its hackable.

bhabbott Quote:


3DO
, whats that? Quote:
Despite having a highly promoted launch (including being named Time magazine's "1993 Product of the Year") and being a host of cutting-edge technologies, the 3DO's high price and an oversaturated console market prevented the system from achieving success

Seems squashing the hardware into a single chip didn't help.


Like the CDTV, the high price of the CD-ROM drive was a problem. Jeff Porter figured out how to bring the price down to $15 with an audio CD mechanism for the CD32.

Jeff Porter Quote:

I got called to big guy's office in New York City. He said Jeff we need to cost reduce the CDTV. I said ok. They were paying $400 just for the CD-ROM drive in this thing. I was like holy crap that is a lot of money. So I went to Japan and I talked to Sony. I said, you guys invented the CD-ROM. I'm guessing you have a cheap one somewhere. And they said here's all the CD-ROMs we have. Blah, blah, blah. I want the motorized one where you push the button and the CD comes out. Oh, that's in the audio product line. I said ok, show me the audio product line. That one looks good, that's the one I want. Oh, that's not for data. Dude, it's a CD, there's data on it already. We can use that. Everybody gave me the uhh and I said watch me. Watch me. So I went to a factory in Japan that had built all our floppy drives and I said I need a little help. You guys speak Japanese and I don't. Work with Sony and suck their brains dry. You know, I'm going to make an IDE interface for a CD-ROM drive. No one had ever done that before. With an audio mechanism and a double speed CD-ROM drive with predictive track seeking and all this other kind of stuff. We had some pretty cool software and firmware to make the CD-ROM drive work. And uh...$15. It went from $400 to $15. Unfortunately, by the time the cost reduced CDTV got around, Commodore had kind of screwed up with the Christmas inventories building things.


https://youtu.be/9_dAZ5R6yNI?t=1385

If Dave Needle and RJ Mical had taken Jeff Porter with them, things may have been different. A $385 cheaper 3DO would have been a game changer. Instead, it was the later CD32 which outsold all CD-ROM based systems in the UK including the Sega Mega CD. Price is very important as Jeff also demonstrated with his work to cost reduce the Amiga resulting in the Amiga 500.

bhabbott Quote:

CBM couldn't do it because their chip foundry wasn't set up to do anything larger than 84 pin PLCC. But what they could make they were able to make very cheaply. That's one reason they were able to make good profit on the A500. Another was low cost through-hole 2 layer PCBs. The A500 PCB was large, but cheap to make and assemble with the equipment and manpower available at the time. When they made the first much smaller surface mount A600 motherboards in Ireland the cost was higher. It's not just the custom chips that determine the price.


The AGA Lisa was produced by HP. I expect the custom chips would have benefited from improved processes. OCS used a 5000nm NMOS process which I don't know was ever improved other than the AGA Lisa (HCMOS 1500nm?). Jay Miner complained in the '80s about CBM not upgrading to CMOS. The 1979 68000 used a 2500nm process and the later HCMOS version circa 1986 was 800nm-1200nm. The slow chip memory with dinosaur age chip process made for slow custom chips. Integrating into one chip would have reduced the total number of pins which should have reduced cost. The board sizes could have been reduced and the power reduction, which is significant with CMOS, would have reduced the cost of power supplies. I see your point about using in-house fabbing but at some point a 5000nm NMOS process becomes ridiculous like it did in the '90s. The state of the art 5nm process today is roughly 1/1000 the size and would use something like 1/1,000,000 the chip area?

bhabbott Quote:

So? Some other home computers also got enhanced customs chips with higher integration, but it didn't save them from extinction. Amstrad's CPC range was so successful that it got Alan Sugar a knighthood, but the CPC Plus range was a poor seller despite having greatly enhanced features. The 5 year gap between it and the original CPC series may have been the reason it failed.

If Commodore had managed to squeeze the AGA chipset into two packages, would it have helped? The A1200 was already cheap, so I doubt knocking a few more bucks off the price wouldn't have done much. Ironically the biggest impediment to A1200 sales was actually that Commodore couldn't make them fast enough!


Lowering the price is a good thing but sufficient production is necessary to leverage the advantage. The savings could have been used to upgrade the CPU or add fast memory providing more value too. A 68030@28MHz instead of 68020@14MHz may have been affordable with a small savings. With fast memory and the chunky of AA+, this likely would have allowed playable fps games like Doom.

bhabbott Quote:

Which would be a pity because such a highly integrated custom chip would be less hackable and less interesting, as well as harder to repair. This is the way PCs went, to the point where motherboards became virtually unrepairable (I remember fixing a rather unique Philips PC that we couldn't get a replacement chip for. I scavenged a replacement chip off a new OEM motherboard that had the same chipset, destroying it in the process. What a waste!).


Make the boards so cheap you toss them when something goes bad and grab another. The Raspberry Pi boards are so cheap that it is not worth diagnosing problems. This is a huge advantage instead of a "pity".

Last edited by matthey on 17-May-2022 at 09:05 PM.

 Status: Offline
Profile     Report this post  
bhabbott 
Re: Irving Gould's House
Posted on 17-May-2022 23:30:00
#56 ]
Regular Member
Joined: 6-Jun-2018
Posts: 151
From: Aotearoa

@matthey

Quote:

matthey wrote:
bhabbott Quote:

Because I have more powerful PCs that cost me nothing and run the software I want to use. And I'll be honest - Windows XP on a semi-modern PC is great. I have no desire to make my Amigas take over its role (or vice versa).


Windows XP is very old and few 3rd party programs and hardware support it anymore.

I said "the software I want to use". They are all installed on my PC already, so obviously it supports them. And several of them are no longer available or the latest versions have onerous licensing restrictions. I'm sticking with XP because it works, and 'upgrading' to Windows 10 would cause a lot of pain. I am not the only one with this issue - the US military still uses XP for similar reasons.

"Ahah!' you say, "That proves my point. You are using yet another outdated incompatible OS!". True, but it's vastly less outdated and incompatible than Amiga OS. And let's face it, this user himself is also outdated and incompatible. I hate the modern look and feel of Windows 10, and don't have the patience or energy to learn new apps. Who knows how long I will even be in this world for, let alone capable of keeping up with the latest trends.

But if I did want to run XP I can buy an ex-corporate PC for NZ$45, less than the price of a Raspberry Pi here.

Quote:
The new replacement hardware is more expensive that the original Amiga hardware inflated prices.

You sure about that? I can get a new A500 motherboard for much less than people are trying to sell used ones for that are probably full of bad chips (got one of those, still haven't managed to get it running even after replacing all 16 RAM chips). I can buy a PCB and keyswitches to make my own keyboard much cheaper than originals typically sell for now. Ridiculous prices are being asked for old Amiga stuff now, and people are still buying it!

In any market, as prices go up the incentive to produce cheaper copies increases. We are seeing that right now with Amiga stuff, which can be made cheaper now with the manufacturing facilities that are available to us.

Quote:
Most monitors and tvs today have digital video inputs which the Amiga did not have. HDMI is also the most common way to hook up sound which is convenient. Amiga IDE and PCMCIA transfers are slow and SCSI adapters that avoid a bottleneck are not cheap. Ethernet and Wi-Fi hardware are not cheap except for PCMCIA and network performance is usually poor.

Apparently new TVs are tending to come with HDMI only now. Doesn't worry me because older TVs with composite, S-Video, SCART or Component and VGA are being sold for next to nothing or given away for free. I have so many of them - as well as VGA monitors - that I don't know what to do with them all.

But if you do have a modern TV with only HDMI that you want to use, adapters are available to connect your Amiga to them. The good ones are not cheap, but they work with a variety of retro machines so are worth it if that's what you really want. Or if you want to accelerate your Amiga then you can do several upgrades at once with a Vampire.

Wi-Fi cards that work on the Amiga are cheap enough. I get ~200kB/s on my network, which is often faster than the sites I download from. I aren't stupidly trying to view online movies or download gigabytes of data into my Amiga, so the speed I am getting is fine.

Nobody uses SCSI anymore. IDE is plenty fast enough for anything you could reasonably expect a 30 year old machine to do. Why is that Amiga fans are never satisfied?

Quote:
USB is the most common general purpose I/O today yet USB hardware is among the rarest of Amiga hardware.

Why do you think it's rare? The reason is simple, lack of demand. If Amiga owners really wanted USB ports that badly then plenty of solutions would be available. But they don't. Instead, dedicated peripherals with USB capability have been developed. These don't suffer the problems associated with running a USB stack.

Quote:
Many Amiga computers are completely isolated without networking support or USB thumb flash drive support. Amiga users often give up and throw the computers in storage as I/O solutions are too expensive
This simply isn't true. If people are doing that it's because they are ignorant, can't be bothered or don't want to spend any money on it - in which case keeping their machines in storage is probably a good thing (I know two people who are doing that right now).

Quote:
and then leave the Amiga community or switch to emulation which has modern I/O support.

If people don't want to muck around with actual hardware and find emulation more convenient that's fine. But don't put the blame on lack of modern I/O support. It's not that hard to get solutions for common needs and they aren't expensive.

Quote:
No. The Amiga 500 is the development target. Code is optimized for a real 68000 CPU. Also, the Amiga 500 is the emulation target.

Actually it's not that hard to find examples of code that was written specifically for the emulator, and which may never have been tested on the real thing. In some cases the authors don't even pretend that the ultimate target will be a real Amiga.

Quote:
Compiler->real hardware target
Emulation->real hardware target

And when the code is written for an emulator configuration that no real Amiga ever had or could have?

Quote:
No Amiga compiler generates code for a FPGA CPU target.

So a compiler that generates 68080 specific code is not an Amiga compiler? Cute.

Quote:
Like the CDTV, the high price of the CD-ROM drive was a problem.

Like I said, squashing the custom chips into a single package didn't help.

Quote:
The AGA Lisa was produced by HP.

Yes, and they did that specifically because Commodore's process couldn't do it (too many pins on the chip). I bet that cost them more. To go to one package they would have to outsource the whole thing, which would have been even more expensive. But MOS Tech was for the chopper eventually anyway, as they had an environmental time bomb that was about to explode.

Quote:
I expect the custom chips would have benefited from improved processes.
Oh for sure, but that would mean huge investment in a new fab - for a platform that was going to die soon anyway. The only practical solution was outsourcing.

Quote:
Lowering the price is a good thing but sufficient production is necessary to leverage the advantage. The savings could have been used to upgrade the CPU or add fast memory providing more value too. A 68030@28MHz instead of 68020@14MHz may have been affordable with a small savings. With fast memory and the chunky of AA+, this likely would have allowed playable fps games like Doom.

Oh God, not that again! Doom didn't exist when the AGA chipset was being developed, and its importance couldn't have been predicted. You needed a very expensive PC to run Doom properly, and yet PC users did the upgrade thing because spending that money was worth it for a PC - not an Amiga. John Carmack knew this, which is why he refused to produce an Amiga version.

But the A4000 included a 25MHz 68040 standard - plenty fast enough - and sold for the same price as a name-brand 486 PC with similar specs. The A1200 had an expansion bus with limitless possibilities. Doom runs quite well on a 50MHz 030, not so well at 28MHz. If Commodore had put a 50MHz 030 and 16MB RAM in the A1200 it would have raised the price dramatically, up close to the price of a 386-SX PC. We all know what that would mean... no A1200 sales!

Amiga fans don't seem to appreciate that absent certain cost savings Commodore got from having their own facilities, the hardware will cost about the same no matter what platform it is designed for. And when that happens, the choice will be a PC because it's IBM compatible.

Quote:
Make the boards so cheap you toss them when something goes bad and grab another. The Raspberry Pi boards are so cheap that it is not worth diagnosing problems. This is a huge advantage instead of a "pity".

Correction - the Pi is expensive enough that having it go bad and not be repairable might put someone off the platform. It's one reason I am hesitant to buy one. I would rather play with some old retro computer that isn't so delicate.

 Status: Offline
Profile     Report this post  
Goto page ( Previous Page 1 | 2 | 3 )

[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 Amigaworld.net.
Amigaworld.net was originally founded by David Doyle