@utri007

Quote:

what we actually know about this device? Where is said that it is ARM, not a FPGA?

It's probably similar to the C64 Mini. See @amigang's post:

https://amigaworld.net/modules/newbb/viewtopic.php?topic_id=44253&forum=2&start=40&viewmode=flat&order=0#843950

Last edited by bison on 30-Dec-2021 at 09:26 PM.

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

@utri007

There's an interview with Paul Andrews from Retro Games Ltd in the latest issue of BlitterWolf.

Does the A500 Mini run using recreated classic hardware or through emulation?
A. Emulation, the cost of doing it all in hardware just is not practical for mass market mini consoles sadly..

Emulation makes sense for the consumer market not only on price but it's also more flexible allowing you to have a fancy menu systems and reconfigured controls to suit playing with a joypad.

@Rob



I wonder what image smoothing will be, higher refesh rates?

_________________
AmigaNG, YouTube, LeaveReality Studio

@amigang

Educated guess, anti-aliasing for upscaling.

Not sure this has been mentioned, but there were new firmwares released for the existing THEC64, THEC64 Mini and THEVIC20...

https://retrogames.biz/thec64/support/upgrade-thec64

most striking is perhaps...

Quote:

[All models] Adds support for THEMOUSE [All models] Adds support for THEGAMEPAD

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@Hammer

Quote:

Amiga ROMs and Workbench: © 1985–1993 Cloanto Corporation. All Rights Reserved. Furnished under license from Cloanto Corporation. Cloanto is a trademark owned by Cloanto Corporation throughout the world and registered in the United States and internationally. WHDLoad is a trademark of WHD Llc.

I certainly disagree with this. And I had seen Hyperion do the same. Neither Hyperion nor Cloanto existed in 1985 and I don't think they have any right to stick 1985 against their name. Perhaps I misunderstand copyright law. But the Amiga ROMs and Workbench were certainly not copyright Cloanto in 1985!

@amigang

Quote:

I really wish they did partner with the Pi foundation to make this, it be so much cooler and offer much more support, I guess a arm board is a arm board at the end of the day.

Would Amiga be of any relevance to the Pi foundation? Then again I don't know much about the foundation. So I don't know if they would be interested in retro projects like this.

However, I'd like to see Hyperion team up with the Pi foundation, and get that OS4 ported to the Pi! I keep reading about people who say OS4 should be ported to the Pi. Which somehow is considered as the next Sam, X1000 and X5000 beater! Or maybe, it's the next Tabor beater.

Last edited by Mobileconnect on 03-Jan-2022 at 11:08 AM.

_________________

And AmigaKit has been supporting the Amiga community since 1994, because they bought amiga.org :)

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@agami

Quote:

agami wrote: Pre-ordered. From JB Hi-Fi

JB! You've done it again! All that's missing is a video of them smashing an A500Mini through a TV set.

EB has them as well.

Funny, after all this time, the Hyperion guys spend all this time on something AmigaOS native you won't see in any shops. Then someone puts together an Amiga emulator in an Amiga case with input devices. And suddenly it's on pre-order in normal bulk and game shops.

@Hypex

Quote:

Would Amiga be of any relevance to the Pi foundation? Then again I don't know much about the foundation. So I don't know if they would be interested in retro projects like this.

Eben Upton the guy behind the PI, is a big fan of Amiga,
https://www.youtube.com/watch?v=argxsReuWHw&ab_channel=TheCentreforComputingHistory

the whole idea of the original Pi is going back to making low cost computers and gets kids into programming again and he admits the Pi400 idea is largely based on Amiga.

I just feel it would be a much better partnership and product if Pi had build say a Pi500 in maybe slightly more Amiga themed, they could of had 3 SD cards provided, 1 the PI OS, 2 a Pre-setup theme Amiga Game build just like the A500 mini setup will be and 3 Amikit, because I feel it show how how a serious Amiga setup can work on a computer like this, mostly likely could of been done at the same or slightly higher costs, consider the Pi400 is only £70 and the A500 mini is £120.

Also Pi system have been very popular with the Retro community and likely sold a lot to just this market, what I feel is holding it back a bit is a fully licensed one that can run games out the box as even though it a fairly simple processes people are still put off having to download and burn Retro images and mess with setting, the mass market wants something that just Works. This could be it and get more kids/parents interested in the platform using the Amiga brand and its games.

_________________
AmigaNG, YouTube, LeaveReality Studio

amigang Quote:

Eben Upton the guy behind the PI, is a big fan of Amiga, https://www.youtube.com/watch?v=argxsReuWHw&ab_channel=TheCentreforComputingHistory the whole idea of the original Pi is going back to making low cost computers and gets kids into programming again and he admits the Pi400 idea is largely based on Amiga.

RIP Amiga. The Raspberry Pi is the philosophical successor to the Amiga.

amigang Quote:

I just feel it would be a much better partnership and product if Pi had build say a Pi500 in maybe slightly more Amiga themed, they could of had 3 SD cards provided, 1 the PI OS, 2 a Pre-setup theme Amiga Game build just like the A500 mini setup will be and 3 Amikit, because I feel it show how how a serious Amiga setup can work on a computer like this, mostly likely could of been done at the same or slightly higher costs, consider the Pi400 is only £70 and the A500 mini is £120.

There is nothing "serious" about emulation, especially when emulating both the CPU and chipset. It is the cheapest, most inefficient and least competitive way to use hardware. With Amiga emulation:

Archimedes > Amiga
RISC OS > AmigaOS
ARM > 68k

amigang Quote:

Also Pi system have been very popular with the Retro community and likely sold a lot to just this market, what I feel is holding it back a bit is a fully licensed one that can run games out the box as even though it a fairly simple processes people are still put off having to download and burn Retro images and mess with setting, the mass market wants something that just Works. This could be it and get more kids/parents interested in the platform using the Amiga brand and its games.

Cheap retro hardware which is inferior to the original is a good way to kill the retro resurgence instead of using it to bring the Amiga back.

@matthey

Quote:

There is nothing "serious" about emulation, especially when emulating both the CPU and chipset. It is the cheapest, most inefficient and least competitive way to use hardware.

I think you worry too much about the "inefficient and least competitive way to use hardware" too much, would I like a native port of AmigaOS and a custom Pi hardware made for Amiga, Yes of course! But i live in the real world, where the AmigaOS rights are still locked in court rooms, the cost and time take to port ARM likly would be to high. So we got to work with what we got.

What have we got, well a really well developed emulator that is only getting better, hell in the latest Amiberry GPIO pins can be used to emulate the LED lights of Amigas.

We got a growing retro market, where as I said before 90% of customer for the mini console market are not stopping to ask, hemmm, how much Ram is inside Mini Nes? How fast is the CPU inside Mini C64, they are not asking these question, they are looking at the shiny box, case and games and going oh I remember that, i like to play with that again.

Specs are getting less and less important for the masses, you ask around most people wont know how fast / ram / core etc are inside there Phones, they just know, well iphone 16 will be faster than 15.

All Im arguing for is this project they could of maybe really surprised people by what an old Amiga can do, you know the people that are not into computers, may not even know what a Pi is, it could of done so much more, really surprised them for not that much more work or money, I feel its a slight missed opportunity.

_________________
AmigaNG, YouTube, LeaveReality Studio

amigang Quote:

I think you worry too much about the "inefficient and least competitive way to use hardware" too much, would I like a native port of AmigaOS and a custom Pi hardware made for Amiga, Yes of course! But i live in the real world, where the AmigaOS rights are still locked in court rooms, the cost and time take to port ARM likely would be to high. So we got to work with what we got.

I don't worry. Either do Amiga cheap and build the Raspberry Pi platform or do it right and rebuild the Amiga platform. The Raspberry Pi Foundation is doing it right which is why I say it is the philosophical successor to the Amiga, the personal computer for the masses. Doing a platform on the cheap affected the Amiga as Atari not upgrading their hardware and making cheap games was a major contributor to the video game crash in 1983. Jay Miner left Atari because they wouldn't let him make newer more powerful hardware based around the 68000. His decision to target an affordable personal computer allowed the Amiga to survive the 1983 collapse caused by Atari arrogance. Marketing executives at Atari had bragged that they could "shit in a box and sell it" and that is what they did to collapse the new video game market. The retro market is building today and there is an opportunity to build on it with quality products or "shit in a box and sell it" destroying it. I believe THEA500 Mini is closer to a "shit in a box and sell it" product while Raspberry Pi products build up a market with competitive quality hardware that outperforms expectations but they mostly miss the retro market. I see an opportunity for the Amiga if wanting to build the Amiga market back while riding the retro wave.

The most important Amiga court case should be settled soon and I believe there is a good chance that Amiga Corporation will be successful at reassembling the Amiga IP. Hyperion isn't really needed for retro products anyway although it would be undesirable to have multiple 68k AmigaOS variants like there are for the PPC AmigaOS. All that is necessary to weaken the competition is to make successful products and standards which are significantly more popular than the competition.

amigang Quote:

What have we got, well a really well developed emulator that is only getting better, hell in the latest Amiberry GPIO pins can be used to emulate the LED lights of Amigas.

The nostalgia of blinking LEDs is all that is needed for a successful retro product?

amigang Quote:

We got a growing retro market, where as I said before 90% of customer for the mini console market are not stopping to ask, hemmm, how much Ram is inside Mini Nes? How fast is the CPU inside Mini C64, they are not asking these question, they are looking at the shiny box, case and games and going oh I remember that, i like to play with that again.

The shiny box is important for nostalgia but I believe quality and value are important too. Do we want a product that will bring back nostalgic memories before they throw the inferior to original product in the closet or one that the customer thinks this is really cool and what else can I do with this much improved Amiga product?

amigang Quote:

Specs are getting less and less important for the masses, you ask around most people wont know how fast / ram / core etc are inside there Phones, they just know, well iphone 16 will be faster than 15.

Apple built up a reputation of quality products offering good value. Customers trust that the hardware inside is not cheap but incrementally improving and competitive. Based on this philosophy, Apple became the first publicly traded company to reach a valuation of $3 trillion recently.

https://www.nytimes.com/2022/01/03/technology/apple-3-trillion-market-value.html

amigang Quote:

All Im arguing for is this project they could of maybe really surprised people by what an old Amiga can do, you know the people that are not into computers, may not even know what a Pi is, it could of done so much more, really surprised them for not that much more work or money, I feel its a slight missed opportunity.

I understand your perspective but presenting a product with 1/3 at best of the performance of a low end product may not fulfill the expectations of customers. While the price to try a Raspberry Pi with Amiga emulator would be low, I see it more as enhancing the Raspberry Pi than helping the Amiga platform. UAE is an excellent software emulation product offering more performance on affordable x86-64 hardware but it has failed to revive the Amiga market. Emulation is cheap, adequate for a one and done dying product like the Amiga. Is that the legacy of the Amiga we want?

Last edited by matthey on 04-Jan-2022 at 01:21 AM.

@matthey

most people are not going to be aware this is a emulated product, I really don’t think for the masses that what people are thinking. Nor are they thinking wow this can only do 1/3 of the true power of the board inside this mini computer, most people are think what Amiga games can it play. That’s it, end of the thought process .
Again all I’m purposing is there was a opportunity here to really surprise people what a current classic Amiga can do, even just in emulation form, most will remember the OS as just a blue and grey desktop, I doubt outside the Amiga community many are aware how much more Amiga can do, even on 1/3 of performance.

I Think you and me both agree the A500 mini could be better. I get it your thinking make custom chip do it right really show off what Amiga can do on low cost hardware, i get it, it would Be nice. Im just saying even with the way Retro Game ltd is doing it, it could be a lot better, how do I know this, largely due to the Pi builds that are out there today, that are only getting better, even if it’s just silly things like Amiga led being emulated.

Quote:

I see it more as enhancing the Raspberry Pi than helping the Amiga platform

I think it would befit both platform and if anything I think help Amiga more if Amiga was promoted by them. Pi Amiga builds are already getting quite a lot of attention thanks to pimiga & Amikit builds, plus even the classic Amiga community are befitting from things like the Pistorm that already got coverage in its pi mag and retweed by Pi official Twitter account its got us on the radar of a growing community that is what we need in the Amiga world, more projects that gets the rest of the world to notice that Amiga is still around.

Last edited by amigang on 04-Jan-2022 at 08:03 AM.

_________________
AmigaNG, YouTube, LeaveReality Studio

@matthey

If a new 68k ASIC is done, wouldn't it just have problems with compatibility when it comes to retro, at least for gaming. I presume that games that are tied to CPU speed would have problems with this especially for console/arcade 68k machines. Many of those don't just have 68k as well, each of them have varying coprocessors. How would this be handled?

As for embedded, since IoT is the big thing, wouldn't AmigaOS have problems with regards to security? And if modern stuff like USB stack, Network stack, etc. are added, would it still have the small footprint it once has?

@AmigaNoob

Well no, Amiga games do not busy loop x cycles to create delays, like old MSDOS games do, instead they sync to screen refresh, typical 50hz PAL, 60hz NTSC, but most game only do 25fps in double buffering, as long don’t change refresh rate you should be fine.

How amiga games busy loops is constantly read the beam y register, or constantly read control registers, for state changes, like joysticks and mouse or keyboard, you name it, interrupt vectors are limited on Amiga., so you clock up CPU without any bad effect, it’s the chipset you can’t clock up.

Last edited by NutsAboutAmiga on 04-Jan-2022 at 05:35 PM.
Last edited by NutsAboutAmiga on 04-Jan-2022 at 05:30 PM.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

AmigaNoob Quote:

If a new 68k ASIC is done, wouldn't it just have problems with compatibility when it comes to retro, at least for gaming. I presume that games that are tied to CPU speed would have problems with this especially for console/arcade 68k machines. Many of those don't just have 68k as well, each of them have varying coprocessors. How would this be handled?

Good questions. Most later 68k CPUs, Amiga accelerators and Amiga custom chips were willing to sacrifice some compatibility for performance. Performance was more difficult to achieve in those early days so more sacrifices were made. Also compatibility was not as important as a retro focused computer. Today, it is possible to have a large increase in performance without sacrificing as much compatibility. Some of the design choices were available back then but were not used for various reasons.

Higher CPU core clock speeds will increase incompatibility with some games. Amiga games are less likely to have problems because the timing is often based off of screen refresh. There are poorly written games which use CPU timing loops though. Turning off caches is one way to slow the performance. The Apollo core also has a turtle (slow) mode which I suggested (Gunnar disliked the alternative suggestion of swapping to a TG68 core when better compatibility was wanted). It is possible to allow a programmable variable clock speed with a little additional cost. This allows to change the clock speed in fractional intervals based on the oscillator. This is often employed today with a temperature sensor for core power management. The 68060 and 68040V were fully static designs which allowed to reduce the clock frequency from max all the way down to zero but no Amiga accelerator designs that I'm aware of took advantage of this to increase compatibility. It would be interesting to see if CPU core clock speeds could be adjusted without rebooting as most timing comes from Amiga chipset screen refresh or CIA timing which can be separately clocked. Multiple CPU cores could allow one core to be low clocked in a compatibility mode while the AmigaOS runs at a higher clock speed for emulation control and general purpose use (basically an AMP setup with the possibility for simultaneous dual AmigaOSs on different cores). Even games which are affected by the CPU core clock speed should allow some clock increase and decrease if programmable which retro game fans may have fun experimenting with (game freeze by zero clocked core and memory examination using another core could be quite useful). I believe flexible hardware clock management would be a good investment.

I don't believe coprocessors are much of a problem unless they are using caches. The Amiga copper and blitter do not use caches (but possibly prefetch?). The Natami project attempted to add caches to the Amiga coprocessors but failed. Certainly configuring a 68k core like original would work with chipsets like the original. There should be room for clock speed increases without increasing CPU core caches if that is a problem. Bus snooping can take care of self modifying code with more compatibility than the original 68020+ processors as the Apollo core demonstrates. The best candidate 68k systems to enhance are the Amiga, Atari ST, 68k Mac and X68000 as they all supported 68k CPU upgrades. It would be convenient and flexible to have some FPGA capabilities for chipsets which are more efficient to simulate than a CPU. Many chipsets and simpler 8 bit CPU cores are already available for FPGA too.

AmigaNoob Quote:

As for embedded, since IoT is the big thing, wouldn't AmigaOS have problems with regards to security? And if modern stuff like USB stack, Network stack, etc. are added, would it still have the small footprint it once has?

Security expectations for low cost embedded hardware are often reduced. A network stack and trusted software with security updates would be able to keep most threats out. The Amiga also has virus checkers. One big problem is naive users allowing threats in. Security and SMP are two of the most difficult issues to solve for AmigaOS. There are some cases where the AmigaOS as it exists now would be inadequate for some uses. A compatible enough 68k MMU could allow small footprint 68k Unix like OSs to run as well. The 68k owned that market for awhile and there is still some support. There is even some embedded support without a MMU.

Maintaining the small footprint is a challenge but should be possible. Development tools for the 68k need some love and programs would benefit from recompiling for the target hardware. At least the vbcc compiler is getting an update and looks to be an improvement but we need affordable hardware to really motivate development. The new AmigaOS is only compiled for the 68000 where compiling for the 68020 could provide 5%-10% better code density (most of the AmigaOS is code). I believe simple enhancements to the ISA and ABI could provide another 5-10% code density improvement but that would likely be for a 64 bit mode although I believe it could operate at the same time as 32 bit code (separate 68000, 68020 and 68k64 modes would provide good compatibility although the 68000 may be better off simulated in FPGA). There is more improvement possible from optimizing as I have demonstrated by reducing some professional C programs by nearly 50%. For example, I was able to reduce the Genesis program (Networking stack) which came with AmigaOS 3.9 by 15% from mostly dead code removal as I recall. New low end hardware would still likely have 256MiB to 1GiB of memory as we would like to improve on the Amiga standard even as higher end 68k hardware would make the AmigaOS less attractive for embedded use. The 68k can offer competitive 32 bit code density to ARM Thumb 2 mode and should be able to have better 64 bit code density than any common architecture. It would require cooperation and discipline to avoid the bloat and maximize the advantage though.

@matthey

You know things are pretty bad when you want swap to tg68 for compatibility.

On actual tg68 systems we swap to Fx68k for compatibility.

And for THEA500, nobody cares about OS, the entire front-end carousel is a linux program and amiga games are launched using a dedicated uae config with whdload to avoid floppy image from even being visible.

And please elaborate on the Genesis “stack”, from what I recall, it was merely a GUI frontend to a legally disputed AmiTCP4 installment.

Last edited by kolla on 06-Jan-2022 at 04:45 AM.
Last edited by kolla on 06-Jan-2022 at 04:29 AM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

kolla Quote:

You know things are pretty bad when you want swap to tg68 for compatibility. On actual tg68 systems we swap to Fx68k for compatibility.

The Apollo core had more bugs and less compatibility back then. The hardware has a FPGA so I supported allowing to switch to a more compatible FPGA core as desired. There is no disgrace for the Apollo core project in that and it increases value instead of decreasing it. Gunnar and I were not always on the same page despite using some of my innovative ideas that he liked.

kolla Quote:

And for THEA500, nobody cares about OS, the entire front-end carousel is a linux program and amiga games are launched using a dedicated uae config with whdload to avoid floppy image from even being visible.

Lame. I expect most customers would rather have a more powerful, compatible and versatile Amiga, especially for the price.

kolla Quote:

And please elaborate on the Genesis “stack”, from what I recall, it was merely a GUI frontend to a legally disputed AmiTCP4 installment.

Genesis is the main program used to launch the AMITCP stack. You could probably call it part of the frontend along with the libraries in the libs directory. The stack kernel is in the kernel directory. It has room for improvement too but it has lots of entry points making it difficult to improve. The Genesis GUI is nice but it is far from optimal as is the AMITCP kernel, especially if wanting a small memory footprint. I guess most people only care about the eye candy unless they are low on memory. Pretty much every 68k web browser user could use more memory though. There is a more efficient competitor without the eye candy but it probably doesn't sell because Amiga users only want an eye candy frontend and don't care about what is underneath.