@matthey

Quote:

matthey wrote: kolla Quote: You keep on and on about MMU-less usage and point to projects and products that are using Linux - which is NOT AmigaOS. I have probably used uCLinux a heck lot more than you have (even on 68k - once upon a time I attempted tweaking OpenWRT for uCLinux on 68k), and yes it is quite capable, mostly because it is so similar to “real” Linux, and can (mostly) use same device drivers and interfaces, and that’s very much the point. Unlike AmigaOS. kolla Quote: The goal of μCLinux was NOT primarily to just have any embedded OS for non-MMU environments, the goal was specifically to have LINUX running on non-MMU environments. Unix like software ported to the AmigaOS without using a MMU is broken according to you.

No, I wrote that porting unix like software to AmigaOS - regardless of MMU or not - has always been a compromise - most obvious is that AmigaOS IS NOT POSIX, but even if you somehow overcome that by using ixemul, you will still struggle porting to AmigaOS due to other limitations, such as the VERY OLD IP stacks (even Roadshow is old IPv4-only BSD 4.4, tops)

Quote:

kolla Quote: The Amiga IP stacks have always been compromises between what’s practical and what’s possible to achieve with all the limitations that the OS comes with, there was never any “complete” stack, always lots of features lacking and seemingly strange limitations. ... Yes there are, even simple things like the lack of fork() creates headaches already, never mind more complex "modern" stuff that we take for granted, like process isolation, thread isolation etc that is widely used in IP software for the sake of security.

Yes, all this very true.

Also very true for μCLinux, with the HUGE difference in that μCLinux is still Linux and have the huge advantage of both being mostly POSIX compliant AND supporting common Linux interfaces and device drivers.

Quote:

Hacked up no MMU Linux

Please - show some respect - it is not "hacked up", it is more than 20 years of engineering to make sure as much as possible of the kernel and userland remains compatible, with and without NOMMU.

Quote:

with the same limitations as the AmigaOS is all about Linux compatibility though?

Not at ALL "the same limitations" - you can stretch it and talk about "the same limitations as ixemul.library", but even that is not a fair comparison. More fitting would have been to talk about the limitations of AmiNIX, but that never came around, so rather pointless.

Quote:

What do basic Linux functions like fork and mmap do without a MMU? Are you contradicting yourself or just trolling again?

No - μCLinux usecases are also limited, but not at all as limited as AmigaOS.

Quote:

kolla Quote: Do you understand the difference? LINUX was the GOAL! On non-MMU hardware! Jeff would never have even considered AmigaOS, as it was NOT THE GOAL! blablabla... The AmigaOS would still not be an easy sale for Jeff as open hardware and software are important to him. This is the big advantage of Linux.

Well, duh!

Quote:

The advantages of embedded Linux over proprietary embedded operating systems include multiple suppliers for software, development and support; no royalties or licensing fees; a stable kernel; the ability to read, modify and redistribute the source code. The technical disadvantages include a comparatively large memory footprint (kernel and root filesystem); complexities of user mode and kernel mode memory access, and a complex device drivers framework.

Duh, again.

Quote:

If the 68k Amiga was as cheap as the Raspberry Pi then it would also be used some in embedded markets, including with the AmigaOS.

Yes
* if 68k Amiga was as cheap as Raspberry Pi
* if AmigaOS had native support for modern input and output
* if AmigaOS had a native modern IP stack
* if AmigaOS had support for modern, commenly used languages
and several more "ifs".

Last edited by kolla on 21-Jul-2021 at 06:50 PM.
Last edited by kolla on 21-Jul-2021 at 06:48 PM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@kolla

Ifs and buts all seems a bit pointless since even RISC OS beat us to the punch! RISC OS developers got their act together and survived in style! The Vampire and CoffinOS/ApolloOS/AmigaOS 3.2 is all we have and it's expensive and confusing for newbies. Even I thought RISC OS looked fun from Dan's video:

https://www.youtube.com/watch?v=sneYGad3j5I

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

geen_naam Quote:

No, that's the one talking about a market and competition in the context of amigaos. Connected people see it for what it is. A hobby. Nothing more and nothing less. Because it will never be more than a hobby. Hobbies don't need to make sense and are allowed to be expensive.

Oh, I see. The Amiga is a joke and Trevor is creating a few expensive hobbyist Amiga toys for a tax deduction. He might as well cancel Tabor and create a POWER9 Amiga for himself then. A true elitist would want a unique one of a kind most powerful Amiga ever for his hobby collection.

Lou Quote:

I mean we are talking about the guy that sees Gunnar marketing the 080 as having 3DNOW but still believes SAGA is something other than the cpu running code meant for gpu+audio chips...

After real-time capabilities, DSP is one of the most needed embedded capabilities.

Real-time capabilities 54%
Digital signal processing 51%
Network capabilities 49%
Analog signal processing 46%
Wireless capabilities 42%
Battery powered 34%
Rugged design 31%
GUI 26%
AI 15%
GPU 9%

A SIMD unit is generally considered to be more powerful than a MAC unit. However, AMMX does not do floating point operations and does half or 1/4 the operations per instruction of more modern SIMD units in chips costing a fraction of the FPGA cost necessary to hold the Apollo core. Most potential embedded customers would either use a commodity processor or create an ASIC where more DSP is required but the Apollo core is optimized for an FPGA and inappropriate for an ASIC. Maybe Gunnar is not marketing for the embedded market but then most 68k Amiga users would probably rather have a better FPU and hardware Blitter than an old SIMD unit.

kolla Quote:

No, I wrote that porting unix like software to AmigaOS - regardless of MMU or not - has always been a compromise - most obvious is that AmigaOS IS NOT POSIX, but even if you somehow overcome that by using ixemul, you will still struggle porting to AmigaOS due to other limitations, such as the VERY OLD IP stacks (even Roadshow is old IPv4-only BSD 4.4, tops)

You made it sound like the AmigaOS is handicapped and that networking support can't be supported properly. In fact, Linux and POSIX are more handicapped than the AmigaOS without a MMU on small footprint systems.

https://www.embedded.com/posix-in-the-age-of-iot-computing/ Quote:

System Limitations. The challenges of POSIX and embedded Linux systems continue to be the real-time options, large memory size requirements, RAM based execution and “Size Weight and Power” or SWaP. The real-time options in Linux have been addressed many ways. From hypervisor integration for RTOS co-existance to optimized algorithms progress has been made. Interrupt latency and jitter are still unpredictable and SRAM based execution is required, but powerful multi-core solutions help overcome limitations; but with multi-core SWaP increases. Sizes of most embedded Linux systems exceed many Mbytes and most require 16MB minimum. In comparison, many MCU IoT products are delivered in less than 300KB. The cost and power required for these extra resources is significant even with energy harvesting and improving battery performance. RAM-based execution is also a concern. Because the operating system must be taken from storage and moved to RAM before initialization, boot times are significant. An often used workaround is to keep an image in flash and boot the old image which improves system boot time by eliminating start up initialization. In comparison, MCUs run directly from flash eliminating the OS loading, and may optionally boot an existing image, creating a zero boot time system.

This is coming from pro-POSIX literature. The AmigaOS has none of theses limitations of embedded Linux.

Quote:

Example #2: Wearable device analysis Consider the case of building a new wearable clothing device. One option is to implement an Android wearable. This requires an MPU and the smallest option is something similar to an Arm Cortex A9. In this case we get the following results: BOM Cost – $40 (volume dependent) Power Consumption – very high Physical Size – large Weight – heavy (large battery) An alternative implementation would use the Unison OS and an Arm Cortex M3 class machine. BOM Cost – $20 (volume dependent) Power Consumption – low Physical Size – interim size using smt medium, using BGA – tiny Weight – light (small battery)

The 68k AmigaOS has more functionality than the minimalist Unison OS. Using a 1MiB flashrom, the Amiga could deliver a full featured and standard Amiga environment that can operate with less than 1MiB of memory. We can even have gfx and a GUI with less than 1MiB of memory. The 68k has similar code density with Thumb2 used by the Cortex M3 CPU and I believe it can have better performance/MHz and better code density.

kolla Quote:

Also very true for μCLinux, with the HUGE difference in that μCLinux is still Linux and have the huge advantage of both being mostly POSIX compliant AND supporting common Linux interfaces and device drivers.

Monolithic kernel drivers are usually complex and not portable. Even eliminating the MMU support and moving to a better RTOS kernel can mean a rewrite. The best RTOS kernels are microkernels like the AmigaOS uses. No MMU Linux uses a small subset of the POSIX standard and pays a high cost for this small advantage (see Linux System Limitations above). Some POSIX compatible OSs have done a better job of resource management and delivering real time capabilities.

kolla Quote:

Please - show some respect - it is not "hacked up", it is more than 20 years of engineering to make sure as much as possible of the kernel and userland remains compatible, with and without NOMMU.

Linux without a MMU is pretty hacked up. It still has most of the disadvantages of Linux while losing a lot of POSIX compatibility and still having less than impressive real time capabilities.

OS worst response timings with MMU
OS Response Time (1/Max frequency), Latency, Latency Jitter
Win XP 200μs, 848μs, 700μs
Win CE 20μs, 99μs, 88.8μs
Linux 13.89μs, 98μs, 77.6μs
QNX Neutrino 20μs, 35.2μs, 32μs
VxWorks 3.85μs, 13.4μs, 10.4μs

OS worst response timings without MMU or with partial MMU usage
OS Response Time (1/Max frequency), Latency, Latency Jitter
RTLinux 5μs, 11.4μs, 7.01μs
Micrium μC/OS-II 1.92μs, 3.2μs, 2.32μs

Data from "A Real Time Operating Systems (RTOS) Comparison". All tests were run on the same hardware.

Why anyone would use Linux when there are better embedded OS choices I don't know. Windows and Linux using MMUs also crashed during tests with frequent interrupts. VxWorks has impressive performance with an MMU although we can still see significant overhead of MMU usage. It's too bad they don't have results for FreeRTOS, Android, Haiku and more no MMU OSs for comparison. Windows XP has a "real time" priority but the OS is a joke. Sadly, the most used embedded OSs are the following.

Embedded Linux 31%
FreeRTOS 27%
In house/custom 16%
Debian Linux 15%
Ubuntu 15%
Android 14%
Windows 10 12%
TI RTOS 9%
Micrium μC 9%
Other 6% or less

Free and easy are good I guess. It gives a lot of amateur embedded developers jobs. Most just use Linux and ARM because they are popular for embedded use (Matthew effect). Linux and ARM were originally mediocre at best choices for embedded use.

kolla Quote:

Yes * if 68k Amiga was as cheap as Raspberry Pi

The question is not if Amiga hardware could be as cheap as the Raspberry Pi but rather how powerful it could be in that price range.

kolla Quote:

* if AmigaOS had native support for modern input and output

Modern I/O already mostly exists in FPGA hardware and it is logical to assume that anything missing could be added.

kolla Quote:

* if AmigaOS had a native modern IP stack

Don't forget WiFi, Bluetooth and USB support. Amiga driver development has been more focused on advanced GPU development than basic I/O support.

kolla Quote:

* if AmigaOS had support for modern, commenly used languages and several more "ifs".

For embedded use, C is still the most important language and the Amiga most supported language.

C 49%
C++ 26%
Python 8%
C# 3%
assembly 3%

ARexx would be cool on a cheap embedded Amiga too. It was certainly utilized well in the embedded Toaster days.

"LOL", as the youngsters would say!
We have elicited another ejaculation of digits and percentages.

Oh comrades in Amiga, I am reasonably certain that our good fried Matthey wet his pants typing this

Quote:

OS worst response timings with MMU OS Response Time (1/Max frequency), Latency, Latency Jitter Win XP 200μs, 848μs, 700μs Win CE 20μs, 99μs, 88.8μs Linux 13.89μs, 98μs, 77.6μs QNX Neutrino 20μs, 35.2μs, 32μs VxWorks 3.85μs, 13.4μs, 10.4μs

/MEGA!

_________________
I HAVE ABS OF STEEL
--
CAN YOU SEE ME? CAN YOU HEAR ME? OK FOR WORK

@matthey

I suggest you start a company or something and do things right, just to prove your own points. Gunnar _DID_ try to cater to the embedded market, but noone was interested.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@BigD

Super off-topic, but yes, RISC OS is fun to play around with on RPi. I and six other people think OS3.1 should be ported to RPi.

_________________
"Unix is supposed to fix that." -- Jay Miner

kolla Quote:

I suggest you start a company or something and do things right, just to prove your own points.

The Amiga IP still needs to be put back together and become usable. Amiga businesses need to cooperate and work together. Mass production is necessary for competitive hardware. Investment would be needed to improve the Amiga technology and software. The AmigaOS and Amiga hardware should be more open and lower cost. Another business in the mix isn't going to solve these problems.

The Amiga has some serious hurdles to solve but still has some highly valued features which could be leveraged, especially in the embedded space. A low priced product strong in some areas and weak in others is often easier to sell than a flawless product with a high price. The idea that the Amiga has no modern value, with all of the technology even becoming a liability, after neglect for all these year is just not true. I would much rather use what is left of the 68k Amiga technology than create and try to market a new OS and CPU architecture.

kolla Quote:

Gunnar _DID_ try to cater to the embedded market, but noone was interested.

Gunnar wanted the Apollo core to appeal to the embedded market but he also wanted to compete with desktop features. There are compromises here. Embedded markets need small cores that fit in low cost FPGAs or powerful cores with more modern features that can be turned into ASICs. The Apollo core misses the mark on both targets and has an identity crisis. A 32/64 bit 68k core with 20-30% better code density than AArch64 and RISCV64IMC would likely be appealing to the embedded market but adding registers everywhere like the Apollo core is a poor choice for real time capabilities.

There was interest in the embedded market. Gunnar, like Hyperion, expects customers to see the superiority of their product and come to them. I approached targeted potential embedded business partners and customers and asked what they needed. I found embedded interest from multiple sources that could have made mass production possible and professionally helped with producing an ASIC. I also wanted to bring in a well respected professional processor chief architect but that was pointless as Gunnar was unwilling to allow anyone to review the code. I found most potential 68k customers want a professionally developed product and that 68k compatibility is important. Gunnar ignored me and what they wanted continuing right along doing what he wanted to do. Yes, nobody is knocking on his door but opportunity already knocked and nobody was home.

Last edited by matthey on 22-Jul-2021 at 04:36 PM.

@bison

Quote:

bison wrote: @BigD Super off-topic, but yes, RISC OS is fun to play around with on RPi. I and six other people think OS3.1 should be ported to RPi.

Just 6?

I'm waiting with bated breadth for Dr. Michael Shultz to finish his AROS-64 native port...though I think he's gone off on an emulation-layer tangent...

Last edited by Lou on 22-Jul-2021 at 08:16 PM.

@matthey

Right, so just like I wrote, without going via long winded anecdotes… noone was interested.

For embedded, let’s make it more interesting - AROS, it has everything you want from AmigaOS, plus IP stack, USB stack etc. and as far as I can recall, the license is more permissive BSD type…and it doesn’t rely on neither m68k nor Amiga chipset - why not use AROS? Speed? That is only really a problem because of amiga native chipset, since AROS natively use RTG (CGfX3) for all graphics output - including VESA and OCS/ECS/AGA. Any modern 68k embedded board would surely NOT use old Amiga chipset for graphics (if graphics at all), so it wouldn’t be an issue.

Last edited by kolla on 23-Jul-2021 at 12:42 AM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@thread

Today both the checkmark and the boing ball have been officially published as trademarks, opening the opposition period.

See page 4 of this thread for direct links.

#6

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@thread

Hyperion's response to Evert Carton's testimony (German)

#6

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@number6

Maybe it's Google's translation, but the wording of that submission by Hyperion's lawyer hardly sounds like it is based in any facts, and is more of just a hopeful opinion of Hyperion to have the damaging testimony of Every Carton thrown out of court.

As Evert stated, his lack of knowledge about the Amiga led him to extensive discussions with Ben Hermans, so he knew very well what Ben wanted, and what Hyperion as a company received with the 2009 agreement, and that it was clearly NOT the right to market and further develop AmigaOS3.1 for the profit of Hyperion. Their only intentions were to have access to the code to develop AmigaOS4 and higher version numbers with a focus of getting rid of the PPC only restriction of the previous agreement. That plus wanting the core code of AmigaOS3.1 to possibly use as a base for some embedded devices, but that too would be AmigaOS4 for embedded devices, based on the core principals of AmigaOS3.1, not AmigaOS3.1.4, or 3.2 for Classic Amiga hardware.

Anyone want to buy an A-Eon X1000 with less than 100 hours of use on it? The only reason I would keep it, is if Hyperion loses this case and some one other than Ben Hermans is in control of AmigaOS4 development.

Last edited by amigadave on 29-Jul-2021 at 07:03 PM.
Last edited by amigadave on 29-Jul-2021 at 07:01 PM.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@amigadave

Not sure what you mean by google translation. The Amiganews article?

the filing itself is in english

#6

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@number6

I clicked on your link of the German version, as that was the only link in your previous message. If the English version was available, without having to sign up for anything, that would have been easier to read, as the entire text exceeded Google's character limit in their translator.

I haven't been able to read some of the other court documents, because they were behind some kind of pay wall, or some other technicality that I did not have the patience to figure out. I'm just interested in the final outcome, not the process that we have to wait through to see that result.

Thanks for all the reporting about this topic that you do. I would not have the patience to sift through all of it.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@amigadave

Quote:

as the entire text exceeded Google's character limit in their translator.

You can simply paste the URL to the page you want to translate, and it will provide you a link to a full translation. No need to manually copy and paste the entire page into the text form.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

"Mr. De Groote was famous in the Amiga scene... was involved with Amiga before joining Hyperion" - famous, huh?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@number6

Now the news-item is also available in English.

_________________
amiga-news.de

@Senex

It's like ping-pong.

Now court listener shows an Amiga Parties' reply to response to motion. Pacer should have it tomorrow.

#6

Last edited by number6 on 30-Jul-2021 at 09:47 PM.

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@thread

Just repeating this in the thread it really belongs in:
Amiga (uspto) trademark registration to Amiga Corporation

I realize people will keep posting "cloanto". So I guess I'll have a full time job correcting them.

again the recent timeline in summary and a 2nd time with supporting legal links

And once again I'll post my open invitation to those others who claimed ownership of the Amiga mark to the CEO of Intellivision...
Please step forward here and reveal yourself, so we can all have a good laugh. We could use one.

#6

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@thread

Amico Utubers are on the case

I guess this is 1 way for the "multiple" companies claiming ownership of Amiga to get some PR.

#6

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*