View Single Post
Old 07 June 2017, 00:01   #99
Join Date: Jan 2010
Location: Kansas
Posts: 1,284
Originally Posted by kolla View Post
Because other retro-computer communities are aware that they are retro-computer communities. Amigaland is ridden with people who think they can revive the Amiga as a commercial platform, attract hordes of programmers etc.
The Amiga (AmigaOS and 68k) was so far ahead of its time that it could still be viable as a commercial platform today. No, I'm not talking about the high end desktop or server markets here. I'm talking about the much bigger embedded, small electronic device, hobbyist and retro markets. The AmigaOS already has many features which are desirable for an (embedded) RTOS (small footprint, modular, scalable, multitasking, exokernel like features) while most other retro OSs have big limitations. We saw the potential of the hobbyist/educational/embedded markets with the Raspberry Pi selling over 12.5 million units (surpassing C= 64 sales). Many of these sales have been used for embedded purposes but it is just scratching the surface of the embedded market. Express Logic has a RTOS called ThreadX with claims of "Over 5.4 Billion Deployments".

The Amiga only sold about 5 million units and the Commodore 64 12.5 million to put the 5.4 billion number in perspective (1080 times Amiga sales and 432 times C= 64 and Raspberry Pi sales). ThreadX only has about 4% of the RTOS market with percentages below from 2015.

FreeRTOS 22%
Micrium (uC/OS-II) 19%
inhouse/custom 17%
Ubuntu 14%
Android 14%
Debian (Linux) 13%
Micrium (uC/OS-III) 8%
Keil (RTX) 8%
M$ Windows Embedded 7 earlier 7%
Wind River (VxWorks) 6%
M$ Windows 7 Compact earlier 6%
Freescale MQX 5%
Angstrom (Linux) 4%
QNX 4%
Red Hat (Linux) 4%
Express Logic (ThreadX) 4%
Green Hills (INTEGRITY) 3%
Analog Devices (VDK) 3%
Mentor Graphics (Net) 2%
eCos 2%

Of course the AmigaOS would need too much work and could never compete in that market right? See eCos down at the bottom of the list above? It used to have a larger market share and was developed by a company called Cygnus Solutions which merged with Red Hat which ceased support for eCos in 2002 (thus the drop in market share). There was a book written by Cygnus founder Michael Tiemann called "Open Sources: Voices from the Open Source Revolution" which is worth reading on the vision and profitability of this unique business making money off open sources using economies of scale while providing valuable services to the community (they made GCC the quality product it is today for example).

Development and support costs are expensive but the AmigaOS could share costs by supporting multiple applications with a flexible OS. They probably would need to be more open about sources to be successful in the RTOS market though. Are A-Eon/Hyperion tech savvy and smart enough to go after a market in the billion range for a 1% share using a product which is already adequate for some embedded uses or would they rather go bankrupt wasting their money on lawyers and developing for a few thousand Amiga PPC users?

Originally Posted by kolla View Post
Really? Look at the Apollo Core, for example. It is not the users that are to blame for all the craziness surrounding that project.
Right. It seems to follow a certain developer and his "yes" henchmen ever since the Natami days. Is is sad that the "Vampire incompatibility discussion Thread" was closed primarily due to personal attacks by Apollo Core developers and supporters.

Originally Posted by kolla View Post
I would say it is really a lot about developers going in different directions to what the users really want.

Considering the legacy software that exists and the extremely slow pace of both OS development as well as software releases, what should the ultimate Amiga be like?
Us developers certainly have different visions. I would move into embedded seeking to gain a few percent of the embedded OS market share and compete against ARM's Thumb 2 (an enhanced 68k CPU could significantly beat Thumb 2 in code density) in the mid performance CPU market while ARM is away trying to compete with AArch64 against x86_64 and the lofty Apollo Core in the high performance CPU market. An embedded focus is similar to the original lean and efficient Amiga philosophy and should have synergies which could improve optimization levels, reliability and affordability of the AmigaOS and Amiga hardware.
matthey is offline  
Page generated in 0.05843 seconds with 9 queries