Poster | Thread |
Nonefornow
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 4-Aug-2021 1:07:07
| | [ #21 ] |
|
|
|
Regular Member |
Joined: 29-Jul-2013 Posts: 339
From: Greater Los Angeles Area | | |
|
| |
Status: Offline |
|
|
matthey
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 4-Aug-2021 6:26:55
| | [ #22 ] |
|
|
|
Super Member |
Joined: 14-Mar-2007 Posts: 1968
From: Kansas | | |
|
| @Nonefornow Bil is talented but practically admits the Commodore LCD showed more promise than the C128 in the video you posted. Most C128 customers just used the C64 compatible mode as C128 mode was not enough faster to matter and lacked software. Most business customers were buying 8088 or 8086 IBM PC compatible hardware by 1985 so the highest performance processor (Z80) in the C128 was rarely used. Who wants three slow modes when they could have higher performance hardware?
Example Byte Sieve benchmarks (lower is better)
6502 1MHz 13.9s (MOS/CSG technology was low end by 1985) Z80 4MHz 6.8s 6809 1MHz 5.1s 8088 5MHz 4.0s 8086 8MHz 1.9s Z8000 5.5MHz 1.1s (original CPU CBM planned to use with C128 VDC in the C900) 68000 8MHz .49s
The Amiga with 68000 was released in the same year as the C128 and had much better performance than the C128 with 8502 (6502@2MHz with Frankenstein limitations) and z80 (limited to about 2MHz due to Frankenstein limitations). Perhaps the acquisition of the Amiga ended the C900 project which may have been released in late 1985. Although the Z8000 was a capable 16 bit processor, it used segmented memory like the 808x. The 68000 was better performance, had a better ISA and used a flat memory map. The 68020 was released in 1984 which was much higher performance than the 68000 but it took CBM several more years to put it in an Amiga. Cost was a factor too but CBM seemed oblivious about performance and performance/price. The C128 cost almost as much to manufacture as the Amiga 500 when it was introduced.
Last edited by matthey on 04-Aug-2021 at 06:56 AM. Last edited by matthey on 04-Aug-2021 at 06:52 AM.
|
|
Status: Offline |
|
|
RobertB
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 23-Aug-2021 20:53:55
| | [ #23 ] |
|
|
|
Super Member |
Joined: 16-Jun-2006 Posts: 1482
From: Visalia, California | | |
|
| I’ve just scanned the book, and I discovered it to be surprisingly enjoyable and easy to read! No stuffy pronouncements! Just an informative story of the days that Bil Herd was with Commodore. After listening to Bil speak at various engagements and in various videos, I found that the book filled in areas that were not covered in the past. And there's even a page on the Amiga!
Below is a page listing of what the book emphasizes (not including the handy-dandy glossary at the book’s end):
Jack Tramiel – p. 12-15
TED computers (116, C16, Plus/4, 264, 364) – p. 20-49, 63-64, 72-74, 79-90, 93-94, 97-108, 116-119, 127
VIC-20 – p. 92
Commodore LCD – p. 119-126, 229, 254-255
C128 – p. 126-254, 257, 267-269
Amiga – p. 257
Truly, Robert Bernardo Fresno Commodore User Group – http://www.dickestel.com/fcug.htm Southern California Commodore & Amiga Network – http://www.portcommodore.com/sccan Nov. 6-7 Commodore Los Angeles Super Show 2021 – http://www.portcommodore.com/class |
|
Status: Offline |
|
|
RobertB
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 10-Sep-2021 21:29:09
| | [ #24 ] |
|
|
|
Super Member |
Joined: 16-Jun-2006 Posts: 1482
From: Visalia, California | | |
|
| |
Status: Offline |
|
|
Nonefornow
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 11-Sep-2021 17:43:54
| | [ #25 ] |
|
|
|
Regular Member |
Joined: 29-Jul-2013 Posts: 339
From: Greater Los Angeles Area | | |
|
| @matthey
In the book there is a discussion regarding the usage of the Z80 processor and why it went into the C128.
Also Bil acknowledges the fact that the CBM team was aware that the C128 was going to be the last 8-bit computer to be made since the Amiga was being developed. However it does comment about price/ performance and the team effort to contain costs in making the C128.
I like to note that while there is one C128 model (with a couple of variants) and while it was produced for only about 4 years, the number of units sold is about the same as all the Amiga computer models.
|
|
Status: Offline |
|
|
BigD
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 11-Sep-2021 18:19:43
| | [ #26 ] |
|
|
|
Elite Member |
Joined: 11-Aug-2005 Posts: 7307
From: UK | | |
|
| @Nonefornow
Well it was cheaper and compatible with the best selling home computer of all time so yes it was going to sell. It was still a dead end technologically and marketing the Amiga properly while cost reducing the C64 was ALL that C= should have been focussing on. A roadmap for developing the Amiga as an ongoing platform with proper R&D with associated pay-cut for executives would have been great
Problem solvers like Bill Herd would have been invaluable in making the A300 (A600) hit its proper price point while taking a stand against feature creep! Whether he’d have been an advocate for the Ranger chip set or A3000+ with DSP, I don’t know. At least he’d have told management when they were dropping the ball. Last edited by BigD on 11-Sep-2021 at 06:21 PM.
_________________ "Art challenges technology. Technology inspires the art." John Lasseter, Co-Founder of Pixar Animation Studios |
|
Status: Offline |
|
|
matthey
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 12-Sep-2021 9:31:06
| | [ #27 ] |
|
|
|
Super Member |
Joined: 14-Mar-2007 Posts: 1968
From: Kansas | | |
|
| Nonefornow Quote:
In the book there is a discussion regarding the usage of the Z80 processor and why it went into the C128.
|
The 68000 was powerful enough it could emulate a 6502, Z80 or 8088 processors with acceptable performance for C64 emulation or most business use of that time. Multiple weak processors is not as good as one powerful processor. The most difficult C64 hardware to emulate was likely the SID sound chip. It would have been interesting if SID could have been incorporated in the Amiga Paula chip which does not have synthesis capabilities. They were both NMOS chips but maybe SID would have added too many transistors back when the Amiga was introduced. Today, SIDs could be added into an FPGA for cheap or ASIC for practically free. Actually, we talked about what capabilities SID could provide when I was part of the Apollo Team. There are several impressive FPGA SID projects.
Nonefornow Quote:
Also Bil acknowledges the fact that the CBM team was aware that the C128 was going to be the last 8-bit computer to be made since the Amiga was being developed. However it does comment about price/ performance and the team effort to contain costs in making the C128.
I like to note that while there is one C128 model (with a couple of variants) and while it was produced for only about 4 years, the number of units sold is about the same as all the Amiga computer models.
|
C128 sales were high enough to be a successful product but likely reduced Amiga sales. An Amiga with an optimized C64 emulator when introduced would have been seen as the upgrade path for the C64, would have provided the software compatibility needed and would have been much more powerful than the dead end C128. Perhaps the C128 development effort could have gone to cost reducing the Amiga sooner. The C128 and cost reduced Amiga 500 ended up looking somewhat similar.
Last edited by matthey on 12-Sep-2021 at 09:32 AM.
|
|
Status: Offline |
|
|
Nonefornow
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 13-Sep-2021 22:12:09
| | [ #28 ] |
|
|
|
Regular Member |
Joined: 29-Jul-2013 Posts: 339
From: Greater Los Angeles Area | | |
|
| @matthey
Well... The C128 was the last 8bit computer to sell in the million of units, but also it was the last computer developed with the objective to be compatible with the C64. And in that sense it was still a product of the legacy and company culture set up by Tramiel.
The Amiga (A1000) which C= purchased in Aug, or so, of 1984 is the development of a totally culturally different company. When released in 1985 both computers were by Commodore but they were developed under different philosophies, different teams, and in different locations. I think that when Amiga Corporation designed the OCS, they were not thinking about C64 compatibility or emulation.
Given the relatively low number of sales of the A1000, it is safe to assume that the R&D money for the A500 and A2000 came from the sales of C64 and C128.
At that point, Commodore probably thought that the 8-bit computers were on their way out and as such there was no need to have another C64 compatible computer.
|
|
Status: Offline |
|
|
matthey
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 14-Sep-2021 1:45:01
| | [ #29 ] |
|
|
|
Super Member |
Joined: 14-Mar-2007 Posts: 1968
From: Kansas | | |
|
| Nonefornow Quote:
Well... The C128 was the last 8bit computer to sell in the million of units, but also it was the last computer developed with the objective to be compatible with the C64. And in that sense it was still a product of the legacy and company culture set up by Tramiel.
|
The C128 was *not* the last 8 bit computer to sell in the million of units and not even the last 6502 or Z80 based computers. The definition of a "computer" is broad and would include devices like the Atari Lynx (WDC 65SC02: 3 million sold circa 1989), Nintendo Game Boy (custom Z80: 118 million sold circa 1989), Sega Game Gear (Z80: 10 million sold circa 1990) as well as embedded computers (hundreds of millions of units a year for the 6502 based architecture). The 6502 based NES sales were likely in the millions after the C128 was discontinued. The Apple IIgs and Super Nintendo were 6502 based but would generally be considered 16 bit computers. Even saying the C128 was the last 8 bit "personal computer" to sell millions of units brings in the question of how a "PC" is defined (no consensus) and how an 8 bit computer is defined (8 bit data bus and 8 bit ALU usually). In my opinion, the C128 sales indicate a moderately successful product due to a highly successful predecessor.
Nonefornow Quote:
The Amiga (A1000) which C= purchased in Aug, or so, of 1984 is the development of a totally culturally different company. When released in 1985 both computers were by Commodore but they were developed under different philosophies, different teams, and in different locations. I think that when Amiga Corporation designed the OCS, they were not thinking about C64 compatibility or emulation.
Given the relatively low number of sales of the A1000, it is safe to assume that the R&D money for the A500 and A2000 came from the sales of C64 and C128.
At that point, Commodore probably thought that the 8-bit computers were on their way out and as such there was no need to have another C64 compatible computer.
|
The C128 for CBM was like the Apple IIgs for Apple which both had higher end 68000 based computers they were trying to transition to. The Apple IIgs was a better computer than the C128 but also cost much more reducing its popularity. Both the C128 and IIgs were expected to be the last of the 6502 based computers. Their success was all about compatibility which was largely ignored in the newer 68000 based systems. The importance of software compatibility to technology, especially CPU architectures, became more obvious later for PC compatibles.
The sales of the C64 no doubt provided R&D money for the Amiga as it remained popular for several years after the Amiga came out. The C128 was late enough that it required R&D money of its own and competed with the C64 and Amiga. The C128 made some customers happy but I feel like most would have been better off with a more expensive but much more powerful Amiga or cheaper C64. My impression is that CBM would have been better off assigning the C128 team to help cost reduce the Amiga and improve C64 compatibility in the Amiga. There was little respect given to C64 compatibility and much given to 808x PC compatibility for the Amiga. CBM needed its loyal C64 customers to upgrade to the Amiga but didn't give them much reason to with their software becoming obsolete.
|
|
Status: Offline |
|
|
Lou
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 14-Sep-2021 13:16:22
| | [ #30 ] |
|
|
|
Elite Member |
Joined: 2-Nov-2004 Posts: 4169
From: Rhode Island | | |
|
| The problem with the C128 was that it's 80 column VDC chip was slow and difficult to program. This means 40-column graphics were the way to go but if you are going to do that you might as well just be a C64...
Interestingly it had it's own 80col video memory like 'modern systems'. But the interface to it left much to be desired. Registers had to be set 1 at a time using 2 or 3 cpu commands, IIRC...and you could set up a burst-copy of up to 255 bytes from normal ram to video ram. It lacked SPRITES completely. Basically had no features. Had a weird mode where you could increase the color cells in an 8x8 character to 1x2 so that the top-half of the block was 2 colors (foreground and background) and the bottom-half was another pair [IIRC] and using interlace mode it could simulate up to 512 colors but again it was difficult and slow.
I always had a "start-up" boot disk in my C128D where I would load the first program on disk that would turn on 80 column interlace mode and 80x50 character display. I had a Magnavox monitor and there were some timing settings in the registers of the 80 column chip that made the 80 column interlace display really sharp on that monitor.
I agree that the CPU itself shouldn't have only been 2Mhz. 4Mhz would have been great ... Lack of 80 column mode sprites doomed that mode.. Last edited by Lou on 14-Sep-2021 at 01:24 PM.
|
|
Status: Offline |
|
|
Nonefornow
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 0:23:06
| | [ #31 ] |
|
|
|
Regular Member |
Joined: 29-Jul-2013 Posts: 339
From: Greater Los Angeles Area | | |
|
| @matthey Your definition of computer does not seem to agree with the generally accepted engineering definition.
|
|
Status: Offline |
|
|
matthey
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 0:59:47
| | [ #32 ] |
|
|
|
Super Member |
Joined: 14-Mar-2007 Posts: 1968
From: Kansas | | |
|
| Nonefornow Quote:
Your definition of computer does not seem to agree with the generally accepted engineering definition.
|
Don't believe everything you read. Misinformation (and disinformation) today seems to be rampant.
The Merriam-Webster definition of computer is, "a programmable usually electronic device that can store, retrieve, and process data". Wikipedia has a similar definition for computer and includes embedded devices.
https://en.wikipedia.org/wiki/Computer Quote:
A broad range of industrial and consumer products use computers as control systems. Simple special-purpose devices like microwave ovens and remote controls are included, as are factory devices like industrial robots and computer-aided design, as well as general-purpose devices like personal computers and mobile devices like smartphones. Computers power the Internet, which links hundreds of millions of other computers and users.
|
https://en.wikipedia.org/wiki/Computer
Last edited by matthey on 15-Sep-2021 at 01:01 AM.
|
|
Status: Offline |
|
|
Hammer
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 1:49:35
| | [ #33 ] |
|
|
|
Elite Member |
Joined: 9-Mar-2003 Posts: 5246
From: Australia | | |
|
| @matthey
Quote:
matthey wrote: @Nonefornow Bil is talented but practically admits the Commodore LCD showed more promise than the C128 in the video you posted. Most C128 customers just used the C64 compatible mode as C128 mode was not enough faster to matter and lacked software. Most business customers were buying 8088 or 8086 IBM PC compatible hardware by 1985 so the highest performance processor (Z80) in the C128 was rarely used. Who wants three slow modes when they could have higher performance hardware?
Example Byte Sieve benchmarks (lower is better)
6502 1MHz 13.9s (MOS/CSG technology was low end by 1985) Z80 4MHz 6.8s 6809 1MHz 5.1s 8088 5MHz 4.0s 8086 8MHz 1.9s Z8000 5.5MHz 1.1s (original CPU CBM planned to use with C128 VDC in the C900) 68000 8MHz .49s
The Amiga with 68000 was released in the same year as the C128 and had much better performance than the C128 with 8502 (6502@2MHz with Frankenstein limitations) and z80 (limited to about 2MHz due to Frankenstein limitations). Perhaps the acquisition of the Amiga ended the C900 project which may have been released in late 1985. Although the Z8000 was a capable 16 bit processor, it used segmented memory like the 808x. The 68000 was better performance, had a better ISA and used a flat memory map. The 68020 was released in 1984 which was much higher performance than the 68000 but it took CBM several more years to put it in an Amiga. Cost was a factor too but CBM seemed oblivious about performance and performance/price. The C128 cost almost as much to manufacture as the Amiga 500 when it was introduced.
|
Intel 80286 was released in 1982.
Intel 80386 (integrated MMU, 32-bit ALU, 32-bit bus, 32-bit memory address space) was released in 1985. Compaq release Deskpro 386 in 1986.
Microsoft Xenix was the dominant Unix OS variant for business. All 386 has integrated MMU.
http://bitsavers.trailing-edge.com/components/intel/80186/210826_iAPX_186_286_Benchmark_Report_Oct82.pdf In the February, 1981 edition of IEEE Micro Magazine (pp. 23-41), H. T. Nagle and V. P. Nelson at Auburn University presented the results of some work they had been doing using l6-bit microprocessors in digital filters. They discussed methods and algorithms for various types of digital filters, and presented benchmark data for a particular instance of an eighth-order cascaded filter. In the article, performance measurements were run on the 8086, the Motorola 68000, the Texas Instruments 9900, the Zilog Z8000, and the Fairchild 9445 (a bipolar machine), all running (presumably) with no wait states. All programs for this benchmar:.~~re_written in assembly language.
Total sample time for DIGITAL FILTER BENCHMARKS
80286 @ 8 Mhz: 139 usec
80186 @ 8 Mhz: 273 usec
68000 @ 8 Mhz: 327 usec
Z8000 @ 4 Mhz: 584 usec
8086 @ 5 Mhz: 857 usec
9900 @ 3.3 Mhz: 1000 usec
_________________ Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68) Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68) |
|
Status: Offline |
|
|
Hammer
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 2:20:12
| | [ #34 ] |
|
|
|
Elite Member |
Joined: 9-Mar-2003 Posts: 5246
From: Australia | | |
|
| @BigD
Amiga ECS = aging gaming hardware with high" business" resolution with 4 colors. Baseline PC VGA has 640x480p 16 colors.
C128 = aging gaming hardware with high" business" resolution with monochrome.
The company mentality is still the same for C128 and Amiga ECS.
PS; My Dad unexpectedly traded in our Amiga 500 (Rev 6, 1 MB Chip Ram) for ex-fleet Amiga 3000/030 in early 1992. Amiga 3000's flicker fixer was a workaround for ECS's overscan 736×567 interlace with 16 colors. Too bad, Amiga 3000 didn't have an Amiga Ranger chipset when 32-bit Chip RAM has at least twice the memory bandwidth over Amiga 500's 16-bit Chip RAM.
In late 1992, my Dad has purchased 386DX-33 with a fast ET4000 SVGA chipset to replace the ex-fleet IBM Model 55SX (386SX-16 16-bit bus with x387 FPU, slow VGA). PC is used for "business" applications that sometimes double duty as a Doom gaming machine. 386DX-33 with ET4000 SVGA lasted until 1996 when I traded our Amiga 3000 into a Pentium 150+S3 Trio 64U clone PC. The Amiga 3000 was running MacOS to access MS Office Mac.
I was using the Amiga has presentation, CAD drafting, and visual artwork machine for school work besides games. in 1990, my school friend abandons a broken Amiga 500 Rev 5 to me since He already has an Amiga 1000. My Dad's work friends have Amiga 2000s.
The broken Amiga 500 Rev 5 was the basis for my classic Amiga usage when reused its chips to re-populate Rev 6 motherboard from Germany.
Last edited by Hammer on 15-Sep-2021 at 02:23 AM. Last edited by Hammer on 15-Sep-2021 at 02:20 AM.
_________________ Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68) Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68) |
|
Status: Offline |
|
|
Hammer
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 2:28:31
| | [ #35 ] |
|
|
|
Elite Member |
Joined: 9-Mar-2003 Posts: 5246
From: Australia | | |
|
| @matthey
Quote:
matthey wrote: Nonefornow Quote:
Your definition of computer does not seem to agree with the generally accepted engineering definition.
|
Don't believe everything you read. Misinformation (and disinformation) today seems to be rampant.
The Merriam-Webster definition of computer is, "a programmable usually electronic device that can store, retrieve, and process data". Wikipedia has a similar definition for computer and includes embedded devices.
https://en.wikipedia.org/wiki/Computer Quote:
A broad range of industrial and consumer products use computers as control systems. Simple special-purpose devices like microwave ovens and remote controls are included, as are factory devices like industrial robots and computer-aided design, as well as general-purpose devices like personal computers and mobile devices like smartphones. Computers power the Internet, which links hundreds of millions of other computers and users.
|
https://en.wikipedia.org/wiki/Computer
|
https://en.wikipedia.org/wiki/Desktop_computer A desktop computer is a personal computer designed for regular use at a single location on or near a desk due to its size and power requirements. The most common configuration has a case that houses the power supply, motherboard (a printed circuit board with a microprocessor as the central processing unit, memory, bus, certain peripherals and other electronic components), disk storage (usually one or more hard disk drives, solid state drives, optical disc drives, and in early models a floppy disk drive); a keyboard and mouse for input; and a computer monitor, speakers, and, often, a printer for output.
The context for C128 "personal computer" was a "desktop computer" and it was the last 8-bit desktop computer from Commodore, not including the late 1990 C65 with its upgraded 256 color display with 4096 color palette chipset. C65 prototype was working. Commodore's R&D resources were split between C65 and AAA/AA chipsets.
Game consoles are specifically designed for games.
Last edited by Hammer on 15-Sep-2021 at 02:36 AM. Last edited by Hammer on 15-Sep-2021 at 02:34 AM. Last edited by Hammer on 15-Sep-2021 at 02:32 AM. Last edited by Hammer on 15-Sep-2021 at 02:30 AM.
_________________ Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68) Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68) |
|
Status: Offline |
|
|
Hammer
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 2:51:50
| | [ #36 ] |
|
|
|
Elite Member |
Joined: 9-Mar-2003 Posts: 5246
From: Australia | | |
|
| @matthey
Quote:
matthey wrote: Nonefornow Quote:
In the book there is a discussion regarding the usage of the Z80 processor and why it went into the C128.
|
The 68000 was powerful enough it could emulate a 6502, Z80 or 8088 processors with acceptable performance for C64 emulation or most business use of that time. Multiple weak processors is not as good as one powerful processor. The most difficult C64 hardware to emulate was likely the SID sound chip. It would have been interesting if SID could have been incorporated in the Amiga Paula chip which does not have synthesis capabilities. They were both NMOS chips but maybe SID would have added too many transistors back when the Amiga was introduced. Today, SIDs could be added into an FPGA for cheap or ASIC for practically free. Actually, we talked about what capabilities SID could provide when I was part of the Apollo Team. There are several impressive FPGA SID projects.
Nonefornow Quote:
Also Bil acknowledges the fact that the CBM team was aware that the C128 was going to be the last 8-bit computer to be made since the Amiga was being developed. However it does comment about price/ performance and the team effort to contain costs in making the C128.
I like to note that while there is one C128 model (with a couple of variants) and while it was produced for only about 4 years, the number of units sold is about the same as all the Amiga computer models.
|
C128 sales were high enough to be a successful product but likely reduced Amiga sales. An Amiga with an optimized C64 emulator when introduced would have been seen as the upgrade path for the C64, would have provided the software compatibility needed and would have been much more powerful than the dead end C128. Perhaps the C128 development effort could have gone to cost reducing the Amiga sooner. The C128 and cost reduced Amiga 500 ended up looking somewhat similar.
|
C65 has Amiga-style planar (bit plane) graphics for its 256 colors mode with a 4096 color palette and Amiga 500 style 16-bit memory bus. C65 has backward compatibility with C64.
C65 R&D effort could have brought AA/AGA's 256 colors (8 bit planes) capability forward in 1990.
_________________ Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68) Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68) |
|
Status: Offline |
|
|
matthey
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 20:30:13
| | [ #37 ] |
|
|
|
Super Member |
Joined: 14-Mar-2007 Posts: 1968
From: Kansas | | |
|
| Hammer Quote:
Intel 80286 was released in 1982.
Intel 80386 (integrated MMU, 32-bit ALU, 32-bit bus, 32-bit memory address space) was released in 1985. Compaq release Deskpro 386 in 1986.
|
Intel was usually faster to market with their x86 CPUs while the 68k provided performance improvements when the equivalent CPU was released. Most people preferred the 68k ISA.
Hammer Quote:
The context for C128 "personal computer" was a "desktop computer" and it was the last 8-bit desktop computer from Commodore, not including the late 1990 C65 with its upgraded 256 color display with 4096 color palette chipset. C65 prototype was working. Commodore's R&D resources were split between C65 and AAA/AA chipsets.
|
Where was "personal" or "desktop" computer mentioned?
Hammer Quote:
C65 has Amiga-style planar (bit plane) graphics for its 256 colors mode with a 4096 color palette and Amiga 500 style 16-bit memory bus. C65 has backward compatibility with C64.
C65 R&D effort could have brought AA/AGA's 256 colors (8 bit planes) capability forward in 1990.
|
While sharing R&D costs is generally advantageous, I feel like all the resources of the C65 project (and C128 project) should have gone toward improving the Amiga. New Amigas should have had at least a 68020 by 1990 which should be adequate to emulate a C64. I believe there were some pretty good Amiga C64 emulators by that time. The Amiga was losing competitiveness and CBM was more interested in upgrading the C64 instead of the Amiga.
Last edited by matthey on 15-Sep-2021 at 08:32 PM.
|
|
Status: Offline |
|
|
Nonefornow
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 15-Sep-2021 23:59:53
| | [ #38 ] |
|
|
|
Regular Member |
Joined: 29-Jul-2013 Posts: 339
From: Greater Los Angeles Area | | |
|
| @matthey
I do not know if there is any official accounting of the R&D money spent. But it seems to me that, as I mentioned above, because the Amiga1000 project was actually bought by Commodore it did not compete with R&D money used by the C128.
After being released in 1985, it does not look as if there was much more $ spent for the development of the C128. The only other significant variant was the C128DCR (which was a cost reducing project)
All the various upgrades to the C64 (C64C, board revisions, C64GS, etc.) were also really cost reducing initiatives, not new R&D.
In the same 6 years time frame (1985-1990) Commodore released at least 7 Amiga models, depending on how you count the variants:
A1000 - A500 - A2000 - A2500 - A2500UX - A1500 - A3000.
My point being that overall I do not think the 8-bit computers took much money away from Amiga R&D. Actually, because of the successful continuous sales they were subsidizing the R&D of the Amiga.
Last edited by Nonefornow on 16-Sep-2021 at 12:02 AM.
|
|
Status: Offline |
|
|
Hammer
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 16-Sep-2021 0:12:45
| | [ #39 ] |
|
|
|
Elite Member |
Joined: 9-Mar-2003 Posts: 5246
From: Australia | | |
|
| @matthey
Quote:
Where was "personal" or "desktop" computer mentioned? |
Semantics debate is useless and a waste of time. You mentioned other 65xx based devices that are not desktop computers like C128 i.e. they are game consoles.
_________________ Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68) Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68) |
|
Status: Offline |
|
|
matthey
| |
Re: Back Into the Storm by Bil Herd, Margaret Morabito Posted on 16-Sep-2021 2:12:38
| | [ #40 ] |
|
|
|
Super Member |
Joined: 14-Mar-2007 Posts: 1968
From: Kansas | | |
|
| Nonefornow Quote:
I do not know if there is any official accounting of the R&D money spent. But it seems to me that, as I mentioned above, because the Amiga1000 project was actually bought by Commodore it did not compete with R&D money used by the C128.
After being released in 1985, it does not look as if there was much more $ spent for the development of the C128. The only other significant variant was the C128DCR (which was a cost reducing project)
|
The C128 project is at least an understandable mistake as the original Amiga team was working on the Amiga 1000 under loose CBM control. CBM was likely hedging their bets with the C128 in case something went wrong with Amiga development. The C128 R&D still had a negative impact on CBM's financials which was likely difficult to recover as it appears the C128 had a low profit margin. I'm not sure how the profit margin compared to the C64 and Amiga 1000 but rumors at least had the C64C and Amiga 500 having a higher profit margin than the C128.
https://archive.org/details/1987-02-compute-magazine/page/n25/mode/2up?view=theater
CBM still tried to cost reduce the C128 instead of cancelling it and other new 8 bit projects well ahead of the Amiga 500 release so there would be time to clear C128 inventories. CBM made other similar mistakes like continuing to make PC clones after they cost more to make than they could buy them for and producing the Amiga 600 despite the failed efforts to adequately cost reduce it causing inventory that wouldn't move because it competed with the superior Amiga 1200 which ended up in short supply.
Nonefornow Quote:
All the various upgrades to the C64 (C64C, board revisions, C64GS, etc.) were also really cost reducing initiatives, not new R&D.
|
The C64 is where CBM was making most of their money and it made sense to cost reduce it. The big problem was that the C64 was difficult to upgrade. The cheap 6502 architecture was a big part of what made the C64 successful but it was also a big part of why it didn't make sense to upgrade it, especially with the acquisition of the much superior 68k Amiga. It looks to me like the C128 Frankenstein monster was unnecessary and ill advised which should have become clear as profit margins deteriorated and the value was reduced by lower cost Amigas and the decline of the Z80 in the business market. Some of this is based on rumors so I really don't know with any certainty but that is my feeling.
Nonefornow Quote:
In the same 6 years time frame (1985-1990) Commodore released at least 7 Amiga models, depending on how you count the variants:
A1000 - A500 - A2000 - A2500 - A2500UX - A1500 - A3000.
My point being that overall I do not think the 8-bit computers took much money away from Amiga R&D. Actually, because of the successful continuous sales they were subsidizing the R&D of the Amiga.
|
In my opinion, CBM didn't do a particularly good job of product management with the Amiga either. They would show signs of moving in the right direction at times and then return to typical CBM mistakes or new management would reset R&D pipelines. Yes, the C64 funded the inefficient Amiga launch. Despite the C128 sales numbers, I don't expect it was too successful after paying for R&D, after considering declining profit margins and after considering that many of the C128 purchases would have been Amiga or C64 purchases instead where there was a higher profit margin. If you have any information to the contrary from Bil's book then feel free to post it.
Hammer Quote:
Semantics debate is useless and a waste of time. You mentioned other 65xx based devices that are not desktop computers like C128 i.e. they are game consoles.
|
Sure, we know the author is probably talking about 8 bit personal or desktop computers but he should have been more careful in his wording.
Last edited by matthey on 16-Sep-2021 at 03:28 PM.
|
|
Status: Offline |
|
|