Your support is needed and is appreciated as Amigaworld.net is primarily dependent upon the support of its users.
|
|
|
|
Poster | Thread | number6
|  |
Re: The "Let's Buy Commodore" Project Posted on 14-Jun-2025 16:50:43
| | [ #41 ] |
| |
 |
Elite Member  |
Joined: 25-Mar-2005 Posts: 11855
From: In the village | | |
|
| @OneTimer1
That's where the AW thread about this and Nostalia Nerd's video comes into play.
Company exists. Company stricken off. Rinse and repeat.
3 times.
#6
Added: note 2nd name same mark. As usual you have to through names and addresses etc. to get there.
Here
Carlo became new CEO some time back, but the way things change... Last edited by number6 on 14-Jun-2025 at 04:57 PM.
_________________ This posting, in its entirety, represents solely the perspective of the author. *Secrecy has served us so well* |
| Status: Offline |
| | amigang
|  |
Re: The "Let's Buy Commodore" Project Posted on 14-Jun-2025 22:21:20
| | [ #42 ] |
| |
 |
Elite Member  |
Joined: 12-Jan-2005 Posts: 2140
From: Cheshire, England | | |
|
| To me it’s a shame the brand name means so much,
But it does.
I mean I would like to think if you have a good enough project people will buy it, vampire kinda proved this and weather it called “Vampire v4” or “Amiga 1200+” for example, if it had the exactly the same spec/features just branding, would it made a difference to sales?
But then I can’t help but think about The A500 mini vs Raspberry Pi 400+AmikitXE, the pi400 has better/faster hardware, is a full computer, keyboard, sd card, wifi, blu tooth, the emulation setup and Amiga system that Amiberry/Amikit provides is to me far better and is easy to add the same games and was cheaper.
But thanks to Amiga branding, the licensing of some old games and I will give them credit a very good mini replica of a500 case and mouse. Oh and easy of use, it’s all setup and ready go (which can be a big factor for many of the casuals) retro games ltd are able to sell Raspberry pi3 level hardware, with less features at a higher price than pi400.
So my hope that a simple branding wouldn’t effect the level of success a project has is on shaky ground.
_________________ AmigaNG, YouTube, LeaveReality Studio |
| Status: Offline |
| | matthey
|  |
Re: The "Let's Buy Commodore" Project Posted on 14-Jun-2025 23:26:45
| | [ #43 ] |
| |
 |
Elite Member  |
Joined: 14-Mar-2007 Posts: 2738
From: Kansas | | |
|
| amigang Quote:
To me it’s a shame the brand name means so much,
But it does.
I mean I would like to think if you have a good enough project people will buy it, vampire kinda proved this and weather it called “Vampire v4” or “Amiga 1200+” for example, if it had the exactly the same spec/features just branding, would it made a difference to sales?
|
If the V4SA used the Amiga brand, used the AmigaOS and used a retro Amiga case, it likely would have sold in the tens of thousands. It has too low of FPGA CPU performance reducing the value for mass production to lower the price. An AC68080 ASIC is needed to vastly increase the CPU performance but it is one man's toy rather than a developer consensus for the way forward so that is not likely to happen.
amigang Quote:
But then I can’t help but think about The A500 mini vs Raspberry Pi 400+AmikitXE, the pi400 has better/faster hardware, is a full computer, keyboard, sd card, wifi, blu tooth, the emulation setup and Amiga system that Amiberry/Amikit provides is to me far better and is easy to add the same games and was cheaper.
But thanks to Amiga branding, the licensing of some old games and I will give them credit a very good mini replica of a500 case and mouse. Oh and easy of use, it’s all setup and ready go (which can be a big factor for many of the casuals) retro games ltd are able to sell Raspberry pi3 level hardware, with less features at a higher price than pi400.
So my hope that a simple branding wouldn’t effect the level of success a project has is on shaky ground.
|
THEA500 Mini did not have Amiga or Commodore branding though. It did not even have the AmigaOS as Hyperion claims and challenges ownership of all versions of it. It was mostly sold through nostalgia including the box, case, mouse and CD32 pad which offer some value as the USB controllders can be used elsewhere. The sales channels also played a big part in the success with it available practically everywhere compared to a device like the A600GS. RGL are pros and make the IP squatters look like amateurs.
The RGL A1200 Maxi will be more expensive but may sell as well as THEA500 Mini if it has the AmigaOS, uses a FPGA for better accuracy, and has a quality keyboard. Amiga branding is still possible which would likely increase sales more as an "Amiga 1200 Maxi" or similar but could tarnish the Amiga brand in some circumstances. A 68k SoC ASIC with FPGA chipset could likely bring the price down to near THEA500 Mini price and approaching the RPi500 price while vastly improving value by providing 68060@1-2GHz Amiga performance and more modern I/O everyone wants. It is the whole package value that matters and some things have synergies together. Even with everything including a 68k SoC ASIC, the device likely would have less performance than a RPi500 but it would likely be lower power/heat, could have a higher quality A1200 keyboard with numeric keypad, could have a small FPGA for chipsets and could have access to a large library of retro 68k games without emulation. A more powerful ray tracing capable GPU than the RPi VideoCore IV GPU could be included and add to the appeal do to the Amiga early ray tracing heritage.
|
| Status: Offline |
| | BigD
|  |
Re: The "Let's Buy Commodore" Project Posted on 14-Jun-2025 23:30:23
| | [ #44 ] |
| |
 |
Elite Member  |
Joined: 11-Aug-2005 Posts: 7551
From: UK | | |
|
| @amigang
Yep, branding may lead to a more consumer friendly product rather than out and out faster machine! The fact that THEA500 Mini is fast enough for most use cases is worth considering too! Pinball Dreams is highly optimised and comparable to a MiSTer whereas I've seen an A600GS stutter with it! Horses for courses! Do we need C= branding. Ans: No, but someone will pay for the privilege of having it! You might as well buy the Nintendo Seal of Approval badge and slap that on everything! _________________ "Art challenges technology. Technology inspires the art." John Lasseter, Co-Founder of Pixar Animation Studios |
| Status: Offline |
| | matthey
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 0:02:14
| | [ #45 ] |
| |
 |
Elite Member  |
Joined: 14-Mar-2007 Posts: 2738
From: Kansas | | |
|
| BigD Quote:
Yep, branding may lead to a more consumer friendly product rather than out and out faster machine! The fact that THEA500 Mini is fast enough for most use cases is worth considering too! Pinball Dreams is highly optimised and comparable to a MiSTer whereas I've seen an A600GS stutter with it! Horses for courses! Do we need C= branding. Ans: No, but someone will pay for the privilege of having it! You might as well buy the Nintendo Seal of Approval badge and slap that on everything! 
|
The "Commodore quality seal of approval" exists and is on the My Retro Computer Ltd Commodore 64x box.
A new PC from Commodore: The C64x is Finally Here! (1:14 showing the Commodore seal) https://youtu.be/JG1TdfezJzo?t=74
The box also says "Commodore branding" which Peri points out later. So cringe with pointing out the seal and branding. RGL uses eye candy instead. RGL eye candy with official branding may have synergies but they may need to improve their hardware to be more authentic and accurate first so they do not tarnish the branding.
|
| Status: Offline |
| | kolla
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 0:30:25
| | [ #46 ] |
| |
 |
Elite Member  |
Joined: 20-Aug-2003 Posts: 3473
From: Trondheim, Norway | | |
|
| @matthey
Quote:
A 68k SoC ASIC with FPGA chipset |
That’s the original Minimig.
Quote:
vastly improving value by providing 68060@1-2GHz Amiga performance |
How would that suddenly be possible?
Quote:
and more modern I/O everyone wants |
Using which northbridge solution? Hot glued to the 68k asic?Last edited by kolla on 15-Jun-2025 at 08:20 AM.
_________________ B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC |
| Status: Offline |
| | bhabbott
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 2:32:36
| | [ #47 ] |
| |
 |
Cult Member  |
Joined: 6-Jun-2018 Posts: 551
From: Aotearoa | | |
|
| @matthey
Quote:
matthey wrote:
An AC68080 ASIC is needed to vastly increase the CPU performance |
Why?
Quote:
it is one man's toy rather than a developer consensus for the way forward |
'Way forward' to what? |
| Status: Offline |
| | number6
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 3:36:37
| | [ #48 ] |
| |
 |
Elite Member  |
Joined: 25-Mar-2005 Posts: 11855
From: In the village | | |
|
| @OneTimer1
I suggest you keep following the comment section of the video. Peri is dealing with some blowback (no surprise) He's written a rather long response along with many other replies.
I don't think it's proper to cut and paste his comment section, so just poke around a bit.
There is one comment that is very interesting that leads me to believe he has an understanding of provenance. Actually he has a doc link under the video covering that as well.
Commodore IP Ownership Provenance
#6
Quote:
Between you and me, we took out a second mortgage |
I gotta ask. How is that between he and the one asking for more info if everyone can read the comments? hehLast edited by number6 on 15-Jun-2025 at 03:44 AM.
_________________ This posting, in its entirety, represents solely the perspective of the author. *Secrecy has served us so well* |
| Status: Offline |
| | matthey
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 4:07:42
| | [ #49 ] |
| |
 |
Elite Member  |
Joined: 14-Mar-2007 Posts: 2738
From: Kansas | | |
|
| kolla Quote:
Quote:
A 68k SoC ASIC with FPGA chipset |
That’s the original Minimig.
|
I thought the original MiniMig used a 68000 CPU. I do recall someone was looking at using a DragonBall MCU which has some SoC like features but they were primitive and old by that time.
https://en.wikipedia.org/wiki/Freescale_DragonBall
The 68000 came out in 1979 and DragonBall in 1995 also with a 68000 core only clocked higher.
kolla Quote:
Quote:
vastly improving value by providing 68060@1-2GHz Amiga performance |
How works that suddenly be possible?
|
About 30 years of silicon improvements should do it.
kolla Quote:
Quote:
and more modern I/O everyone wants |
Using which northbridge solution? Hot glued to the 68k asic?
|
Everything on one SoC ASIC die.
bhabbott Quote:
Quote:
An AC68080 ASIC is needed to vastly increase the CPU performance |
Why?
|
FPGA CPU cores and Amiga 68k virtual machines are nowhere near competitive and 68k fans want performance and compatibility. Do you think the higher CPU performance and cheaper PiStorm have reduced Vamp/AC sales? Is value important?
bhabbott Quote:
Quote:
it is one man's toy rather than a developer consensus for the way forward |
'Way forward' to what?
|
Forward to the 68k and Amiga survival as opposed to extinction.
Last edited by matthey on 15-Jun-2025 at 04:08 AM.
|
| Status: Offline |
| | number6
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 4:09:57
| | [ #50 ] |
| |
 |
Elite Member  |
Joined: 25-Mar-2005 Posts: 11855
From: In the village | | |
|
| @number6
Provenance doc:
Quote:
In 2014, Polabe Holding N.V. (an investment company based in Luxembourg) |
uspto:
Quote:
Owner Name: Polabe Holding N.V. Owner Address: Caya Dr. J.E.M. Arends 18-A Oranjestad-West ARUBA |
hmm.....I'm guessing he didn't go to work on a daily basis given the distance.
#6
_________________ This posting, in its entirety, represents solely the perspective of the author. *Secrecy has served us so well* |
| Status: Offline |
| | number6
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 4:20:24
| | [ #51 ] |
| |
 |
Elite Member  |
Joined: 25-Mar-2005 Posts: 11855
From: In the village | | |
|
| @number6
Provenance doc:
Quote:
Following legal disputes between C=Holdings B.V. and Asiarim concerning a loan that was not repaid, on 16 December 2013, the United States District Court for the Southern District of New York adjudged and confirmed that C=Holdings B.V. was indeed the sole and beneficial owner of the Commodore brand and trademarks |
ok. That's true but...I gotta ask:
Why did they list their website as Tulip.com eleven years after Tulip was supposedly out of the picture?
Source
Just curious.
#6
Edit: bah. corrected 8 to 11 since my post was 2016 and Tulip was supposedly sold in 2005
Last edited by number6 on 15-Jun-2025 at 05:18 AM.
_________________ This posting, in its entirety, represents solely the perspective of the author. *Secrecy has served us so well* |
| Status: Offline |
| | matthey
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 4:49:23
| | [ #52 ] |
| |
 |
Elite Member  |
Joined: 14-Mar-2007 Posts: 2738
From: Kansas | | |
|
| #6 Quote:
I suggest you keep following the comment section of the video. Peri is dealing with some blowback (no surprise) He's written a rather long response along with many other replies.
I don't think it's proper to cut and paste his comment section, so just poke around a bit.
There is one comment that is very interesting that leads me to believe he has an understanding of provenance. Actually he has a doc link under the video covering that as well.
Commodore IP Ownership Provenance
|
He has a better understanding of Commodore than Ben Hermans who wrote the 2009 Amiga Inc v Hyperion VOF settlement agreement licensing "AmigaOS 3.1" ... "originally developed, owned and marketed by Commodore Business Machines (CBM) for their Amiga line of computers in 1994."
https://docs.justia.com/cases/federal/district-courts/washington/wawdce/2:2007cv00631/143245/147/1.html Quote:
"Software" means AmigaOS 3.1, which is the Operating System (including without limitation it's Software Architecture as described in the Documentation) originally developed, owned and marketed by Commodore Business Machines (CBM) for their Amiga line of computers in 1994.
|
https://docs.google.com/document/d/1w_jZe-8KpVY6IlvKjT1ptXJV7_2Pes88_90tOIrSueM/edit?tab=t.0 Quote:
Commodore Trademark Ownership Transfers (1994–2021)
Summary Timeline
1958–1976: Commodore Business Machines 1976–1994: Commodore International Corporation 1995–1997: Escom A.G. ...
|
Hyperion licensed "AmigaOS 3.1" from "CBM" developed in "1994" and now they claim ownership of any AmigaOS from any Commodore business at any time. Hyperion is like playing the board game Clue with someone who claims they won on their first very wrong guess. Sometimes a judge has to teach lessons in how to play fair to the clueless.
|
| Status: Offline |
| | bhabbott
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 5:48:56
| | [ #53 ] |
| |
 |
Cult Member  |
Joined: 6-Jun-2018 Posts: 551
From: Aotearoa | | |
|
| @DiscreetFX
Quote:
DiscreetFX wrote: I always thought the Commodore name was the weakest link for the Amiga. |
I was not a Commodore fan, but they did breath life into the Amiga. I don't think anyone else would have done better.
Quote:
People thought it was like a C64 which is wasn't. |
The 'developer edition' of the A1000, distributed before the official release, was branded with the Commodore 'chicken lips' logo and name. In the production version this was replaced with the iconic rainbow tick and AMIGA name, distancing it from Commodore's other products. This was welcome not just for being more aesthetic, but because it differentiated the Amiga from the C64. You'd have to be pretty ignorant to think it was 'like the C64' even without reading the specs. Just looking at it was enough to tell you that this was no C64.
However Commodore then released the C128D with similar styling to the A1000, and the A500 with a shape similar to the C128. The AMIGA name remained (embossed in large letters) on the A500, but the rainbow tick was replaced with Commodore's 'chicken lips' logo in muted brown. The box artwork also showed the familiar red/blue 'chicken lips' and Commodore name. So by 1987 Commodore was branding the Amiga as their own.
It may seem irrational, but this is one of the reasons I bought an A1000 in 1987 rather than an A500. Aesthetics matter.
OTOH, in some ways the A500 was 'like' a C128 (and therefore a C64). Both had hardware scrolling, sprites, and 'realistic' sound. Both had 80 column text with crisp RGB output. Both were stuffed full of unique Commodore custom chips to perform their magic. The C128 was rightly seen as a poor man's alternative to the A500, and the connection between them (being designed by Commodore) was clear. Obviously they weren't similar enough to share software etc., but few machines were back in those days so it wasn't a big deal.
The problem with the Amiga was that it was too good. Other home computers had limitations that kept them firmly in the 'toy' class, but the Amiga's powerful 16/32-bit CPU, open-ended expansion, sophisticated preemptive multitasking GUI OS and hardware accelerated graphics and sound suggested that it was capable of far more. Fans imagined it matching or exceeding the capabilities of the PC, Mac and even high-end workstations. They then became bitterly disappointed when this didn't happen.
Sadly, the same fans who couldn't understand why the Amiga wasn't able to fulfill their dreams of world domination in the 90's still don't understand why it won't do the same today. The hardware isn't the problem. Branding isn't the problem. The problem is it's Amiga - not PC, not Mac, not Raspberry Pi. It can't play in those markets because it's not compatible, and if it was compatible it wouldn't be Amiga.
For the Amiga to be a success it has to find it's own niche that attracts users. Attaching the Commodore brand to it won't help much. It wasn't needed in 1985 and it isn't needed now. Using other logos like the Boing Ball etc. won't help either, as these have already tarnished it and/or created brand confusion. What it should do is reclaim the rainbow tick and AMIGA name.
Quote:
I was glad when the Amiga was separated from Commodore. In the end Commodore became the Amiga's greatest liability instead of an asset. Commodore never seemed to know what to do with the Amiga or its unique strengths. Also, they allowed plenty of time for Apple and Microsoft to catch up. Commodore should have let the innovations keep coming and coming so no-one could catch up. |
I wasn't glad, because I knew it was the end. Nobody would care for the Amiga as well as Commodore did. Could they have done better? Sure. But the real world isn't perfect. Every Commodore employee was trying to do good, and they did do good.
It's true, most people didn't know what to make of the Amiga - including the top-shelf advertising firm Commodore paid millions to promote it. Magazine reviewers prattled on about lack of IBM compatibility and other irrelevancies. The engineers soon lost the plot too, trying to turn it into a high-end workstation. But users quickly learned what it was really about. Luckily a few people in Commodore did understand - which why we got the A500, A1200, A2000 and A4000.
Had more people in Commodore understood what the Amiga was about we might have gotten even more models that 'hit the spot', but those four alone were a pretty good effort. We couldn't realistically expect more. Unfortunately many Amiga fans are not realistic. Commodore couldn't 'keep the innovations keep coming and coming so no-one could catch up'. To start with the Amiga wasn't ahead in all areas, and changing that wouldn't be easy. Secondly the areas it was ahead in were bound to be copied and/or improved upon as technology improved. Expecting it to stay so far ahead that 'no-one could catch up' was unrealistic and unreasonable.
The truth - that many fans refuse to face - is that the Amiga always had a limited lifespan. In this respect it was no different to all the other home computers, most of whom expired before the Amiga did. We didn't cry when the ZX Spectrum, Amstrad CPC, Apple II, MSX, C64 etc. died, even if we were avid fans. We simply accepted that nothing lasts forever and moved on to the Next Big Thing. But the Amiga was different. It was so good that we could imagine it being the last computer system we would ever need. Of course it wasn't - because the computer industry had already settled on a different architecture to go all-in on.
|
| Status: Offline |
| | Hammer
 |  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 7:43:23
| | [ #54 ] |
| |
 |
Elite Member  |
Joined: 9-Mar-2003 Posts: 6482
From: Australia | | |
|
| @bhabbott
Quote:
The 'developer edition' of the A1000, distributed before the official release, was branded with the Commodore 'chicken lips' logo and name. In the production version this was replaced with the iconic rainbow tick and AMIGA name, distancing it from Commodore's other products. This was welcome not just for being more aesthetic, but because it differentiated the Amiga from the C64. You'd have to be pretty ignorant to think it was 'like the C64' even without reading the specs. Just looking at it was enough to tell you that this was no C64.
|
A1000 was manufactured by Sanyo (Panasonic) VCR factory, not by Commodore's in-house manufacturing. The Sanyo (Panasonic) VCR factory was expensive. CSG fabricated A1000's custom chips.
A500 was manufactured by Commodore's Hong Kong and German factories.
Commodore's system engineering group applied cost reduction measures for A500 e.g. designed the Gary chip with Toshiba instead of CSG.
In 1986, Commodore's system engineering group already had a view that CSG's work reputation was slow._________________ Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68) Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68) Ryzen 9 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB |
| Status: Offline |
| | Hammer
 |  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 8:04:49
| | [ #55 ] |
| |
 |
Elite Member  |
Joined: 9-Mar-2003 Posts: 6482
From: Australia | | |
|
| @cdimauro
Quote:
cdimauro wrote: @matthey: even the Commodore 64 deserves much better.
The C65 was created by engineers that had no idea of the software which was already running on the C64, and created a quite incompatible product. In fact, I want to see the existing turbo tapes/disks running on it: plenty of the software was releases using them, which can't run on the C65!
If the MEGA65 works like the C65, then the situation will be the same.
A new, modernized C64 should be built with 100% compatibility at its foundations. On top of this a lot of things can be added. So, similar to the C128, but without the alien Z80 and the super-crappy VDC. |
C65 was a stealth R&D rebellion against Henri Rubin's bet the farm on ECS with "read my lips, no new chips". C65 project allowed the CSG LSI group to learn about Amiga's bitplane graphics architecture with 8 bitplane improvement. This is important after Commodore management kicked out Amiga's original graphics engineers.
In 1989, CSG's C65 surprise on corporate politics forced the system integration group to quickly start AGA (Pandora) R&D with a go-ahead permission from Henri Rubin.
With a go-ahead permission from Henri Rubin, the same CSG LSI group designed the AA Lisa chip with 8 bitplanes and ported some improvements from AAA, e.g. 16.7 million color palette.
A single engineer was allocated to debug C65, ECS Agnus B, and AAA Andrea, which created an R&D bottleneck. Commodore HR and Herni Rubin would rather hire PC system integration and AMIX software engineers (from Coherent) instead of ASIC engineers.
From Commodore - The Final Years
CeBIT 1988 As usual, Commodore was the highlight of CeBIT when the doors opened on March 16, 1988. CEO Irving Gould, along with US president Max Toy and COO Henri Rubin in tow, gave a press conference in front of 500 attendees and journalists.
Gould announced the positive financial results from the previous year. The bright spot was that 40% of the revenues were generated by Amiga products. The Amiga 500 and Amiga 2000 were truly taking over revenues from the C64. Gould was also remarkably open, or perhaps undisciplined, and revealed all the upcoming products, including the redesigned PC line, the A2286 bridgeboard, the A2500AT, the A2500UX, the A2620 accelerator, the A2300 Genlock for the Amiga 2000, the Transputer, and even the Amiga 3000 (much to the chagrin of Commodore’s engineers).
Commodore Germany’s emphasis on PC-compatible products became understandable when Gould revealed Commodore was the leader in the West German microcomputer industry, with a market share of over 50%. In the business marketplace, Commodore ranked second, only behind IBM itself.
Commodore Germany’s leadership, including Harald Speyer and Winfried Hoffmann, were ecstatic about the new PC40-III with an ATI graphics card. In fact, a demo running on the PC40 put the noninterlaced scan converter running on the Amiga 500 to shame.
(Rubin made Porter turn off the A500 demo in response.) The new PC would also include a Commodore-made 1352 mouse.
Although there was much to be excited for at the show, overall the German Commodore employees were upset that there were no new Amiga computers. In 1987, Porter had intended to have early samples of the new chipset ready by now, and incorporated into an Amiga 3000, but CSG had just missed the deadline for the chips by a few weeks.
Make ATI Great by Commodore. Commodore gave ATI's 1st major OEM contract._________________ Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68) Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68) Ryzen 9 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB |
| Status: Offline |
| | OneTimer1
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 10:35:18
| | [ #56 ] |
| |
 |
Super Member  |
Joined: 3-Aug-2015 Posts: 1249
From: Germany | | |
|
| @matthey
Quote:
matthey wrote:
kolla Quote:
[quote] vastly improving value by providing 68060@1-2GHz Amiga performance
|
How works that suddenly be possible?
|
About 30 years of silicon improvements should do it. [/quote]
Motorola/Freescale/NXP dropped the 68k development for a reason, no one will be interested in starting it again.
The tiny amount of Amiga/Atari/SinclairQL fanatics are no economical viable market to start it again, no matter if improvements could be inserted easily.
And for people designing new products it is less risky to get an existing ARM board, develop something with a tiny Linux under it and put it on a cheap < 25$ board and sell it in the thousands.
There are reasons why C64 retro consoles are using ARM boards instead of ASICs now. Last edited by OneTimer1 on 15-Jun-2025 at 10:40 AM.
|
| Status: Offline |
| | minator
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 16:59:07
| | [ #57 ] |
| |
 |
Super Member  |
Joined: 23-Mar-2004 Posts: 1034
From: Cambridge | | |
|
| @OneTimer1
Other nice to have, I don't see the logic of doing an ASIC either.
Quote:
Quote:
About 30 years of silicon improvements should do it. |
|
But without the equivalent 30 years of architectural improvement. So, it'll run slow and hot.
Without out of order execution, modern prefetch or branch prediction it'll be constantly stalling. Without the power saving tech, you wont be able to save power while the processor is stalled, and it'll stay hot. This on top of CISC processors being inherently inefficient because of the big decoders.
Quote:
The tiny amount of Amiga/Atari/SinclairQL fanatics are no economical viable market to start it again, no matter if improvements could be inserted easily. |
I think there'd be some market, but it'll be pretty small. It might be more interesting with a more modern design, but that'd be so expensive there really would be no market for it.
_________________ Whyzzat? |
| Status: Offline |
| | matthey
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 17:49:22
| | [ #58 ] |
| |
 |
Elite Member  |
Joined: 14-Mar-2007 Posts: 2738
From: Kansas | | |
|
| OneTimer1 Quote:
Motorola/Freescale/NXP dropped the 68k development for a reason, no one will be interested in starting it again.
|
Sure, there was "reason". IBM went to Apple and convinced them to adopt PPC. Apple then went to Motorola and convinced them to be the 2nd source supplier of PPC. Motorola leadership decided it was better to have some desktop presence with PPC rather than none so they signed onto the AIM Alliance which was a completely political decision. Their organically developed 68k and 88k were abandoned and their developers betrayed without technical consideration resulting in some of them leaving and beginning a long decline at Motorola. At the same clock speed, The 8-stage 68060 had similar performance to the 4-stage PPC601 with double the caches but the 68060 should have allowed a 50%-100% higher clock speed with the deeper pipeline and there was a 68060+ "proposed within Motorola" boosting performance by "20%-30% independent of clock frequency", likely with double the caches planned (see article below). The 68060@66MHz was already in testing and announced but I have never even seen a picture or price sheet with a real and full 68060@66MHz so political decisions were followed by product sabotage of technically superior products.
Motorola Introduces Heir to 68000 Line https://websrv.cecs.uci.edu/~papers/mpr/MPR/ARTICLES/080502.pdf Quote:
Production quantities of the 66-MHz 68060 will be available late in 4Q94, according to the company; no price has been announced.
|
Motorola leadership should have let 68060 development proceed on its own based on sales as it was successful in the embedded market and would have been more successful with higher clock speeds available increasing performance and price efficiency (performance/$). If Steve Jobs could see that the 68k allowed higher performance and clocked CPUs for the desktop and lower power CPUs for laptops, he may have returned to the 68k rather than x86. Third party Mac vendors using 68060 accelerators could have forced Apple to abandon shallow pipeline PPC if Apple could not sabotage 68060 compatibility (earlier versions of the 68k Mac OS were 68060 compatible while later versions around when PPC was released were not). A return to the 68k for Apple would have been better for Motorola than the move to x86 but it would have been very embarrassing for the Motorola leadership and their political instead of technical decision making. At least Intel left the door open to return to x86 when they made their Itantic mistake so it did not sink the ship like Motorola's smaller bet. The Itanium Solutions Alliance, another political alliance, invested $10 billion in Itanium development. AMD used more technical and practical decision making to develop the much cheaper x86-64 ISA/ABI and CPUs. Is it clear that alliances sometimes make political decisions and have conflicts of interests?
OneTimer1 Quote:
The tiny amount of Amiga/Atari/SinclairQL fanatics are no economical viable market to start it again, no matter if improvements could be inserted easily.
|
I believe the most popular retro platforms/systems that allow or could allow a 68060 are the following.
1. Amiga 68k and CD32 console 2. X68000 3. Atari 68k 4. 68k workstations and OSs (Linux/Unix/BSD/OS-9, NeXT, 68k workstations) 5. Mac 68k 6. SinclairQL
Including a 68000 core with fine grained control of the clock speed could allow compatibility with the following systems in what I believe would be the order of popularity.
1. Genesis/Mega Drive 2. Neo Geo 3. TRS-80 Model 16(B)/Tandy 6000
A cycle exact full static 68000 core design, which exists, allows changes to the clock speed including full clock stop with the best possible compatibility and may aid in Action Replay like features.
https://en.wikipedia.org/wiki/Action_Replay
The 68060 also has a full static core design but cycle exact operation is no longer important for systems that support it from the first list above.
OneTimer1 Quote:
And for people designing new products it is less risky to get an existing ARM board, develop something with a tiny Linux under it and put it on a cheap < 25$ board and sell it in the thousands.
There are reasons why C64 retro consoles are using ARM boards instead of ASICs now.
|
ARM is easy and relatively cheap but their Cortex-A abandonment of 32-bit will likely make small SBCs more expensive and/or less standard. ARM does not have a cost advantage for the production of small SBCs. Their royalties give a cost advantage to competitors who do not pay royalties for cores. There is a much larger value increase for a 68k ASIC than a C64 ASIC. Even the CD32 console allows a 68060@100MHz with good compatibility which means a 68060@1-2GHz is likely compatible too. A 40 times performance increase over a 68060@50MHz is likely possible with a modernized in-order 68060 that is the equivalent of ARM Cortex-A in-order cores as used on cheap fanless SBCs.
|
| Status: Offline |
| | Kronos
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 19:32:20
| | [ #59 ] |
| |
 |
Elite Member  |
Joined: 8-Mar-2003 Posts: 2766
From: Unknown | | |
|
| @matthey
>Motorola leadership should have let 68060 development proceed on its own > based on sales as it was successful in the embedded market and would >have been more successful with higher clock speeds
Embedded rarely paid top $ and if it did it lacked the volume. Moto simply lacked the means to compete against Intel on their own which would have lead to their demise even if 060 had beaten the Pentiums. So they made a bet on a cooperation, which also failed.
>I believe the most popular retro platforms/systems that allow or >could allow a 68060 are the following. > >1. Amiga 68k and CD32 console >2. X68000 >3. Atari 68k >4. 68k workstations and OSs (Linux/Unix/BSD/OS-9, NeXT, 68k workstations) >5. Mac 68k >6. SinclairQL
All of these were dead even before the 060 came along, most of them fully died with a SW stack that could be well served with an 030 (in some cases even a plain 68k).
What remains of these today is mostly retro fans wanting to play old games. Games that will stumble on any CPU but what was shipped 40 years ago. Hence they are better served with SW defined CPUs.
> 1. Genesis/Mega Drive > 2. Neo Geo > 3. TRS-80 Model 16(B)/Tandy 6000 > >A cycle exact full static 68000 core design, which exists,...
In the form of millions of actual 68000 CPU and SW emulation.
>that is the equivalent of ARM Cortex-A in-order cores as used on >cheap fanless SBCs.
CHEAP is the word here, if the have 1 core and under 4GB the SoC will be itself 1$ with full support in modern compilers an open source OSes.
The problem is as always you do not widely overstate the size of the potential market but you also mix up several quite different markets where the only ones that have decent volume are the ones that leave no marging for the needed R&D costs both for SW and HW. _________________ - We don't need good ideas, we haven't run out on bad ones yet - blame Canada |
| Status: Offline |
| | number6
|  |
Re: The "Let's Buy Commodore" Project Posted on 15-Jun-2025 21:04:40
| | [ #60 ] |
| |
 |
Elite Member  |
Joined: 25-Mar-2005 Posts: 11855
From: In the village | | |
|
| @thread
FAQ added
#6 _________________ This posting, in its entirety, represents solely the perspective of the author. *Secrecy has served us so well* |
| Status: Offline |
| |
|
|
|
[ home ][ about us ][ privacy ]
[ forums ][ classifieds ]
[ links ][ news archive ]
[ link to us ][ user account ]
|