Kronos Quote:

—1,5 mllion Mini-Segas All build on ultra cheap ARM SoCs that have plenty of headroom for emulation and convenience features around it. 99.999% of buyers don’t care and wouldn’t even know the difference between SW, EMU, HW EMU (FPGA) or an ASIC

Why are Analogue console remakes selling out quickly when they are often more expensive than the Minis and original hardware at the end of production? Why are people buying MiSTer setups for $200-$600 USD? With so few MiSTer boards being sold, how could a new MiSTer main board clone be produced?

https://www.timeextension.com/news/2024/05/the-cost-of-owning-a-mister-fpga-is-about-to-come-down-dramatically

Why are Vampire V4 standalones and accelerators selling for 535+ Euros when a Pi-Storm is higher performance? Why are there so may new Amiga 68060 accelerator boards and so few rev 6 68060s available when emulation is good enough for 99.999% of retro users?

Kronos Quote:

Market might also already be saturated at this point

Saturation is possible at the current value but the bigger problem may be retro fans turned off by crap emulation hardware, limited hardware and poor ARM value.

Kronos Quote:

68k ASIC for that market would require substantial upfront investments to actually develop it and a SW stack making it an as nice retro gaming experience as the existing ARM based ones.

It depends if you want to invest in ARM like the A600GS and x86-64 like WinUAE while 68k software dies, especially development tools like compilers, or invest in the 68k.

Kronos Quote:

Even if you could sell 7 figures still not enough economics of scale when competing SoCs that can be ordered and payed in 6 or even 5 figures per batch for under 1$.

I don't know if those cheap ARM Cortex-A53 SoCs are quite $1 USD. Maybe $3-$7 USD? There are ARM chips that are under $1 USD like the RP2040 but it would be difficult to emulate the original 68k Amiga while the successor RP2350 probably could barely for about $1 and with a lot of software work.

https://en.wikipedia.org/wiki/RP2040
https://en.wikipedia.org/wiki/RP2350

An ASIC 68060&AA+ SoC would use fewer transistors than the RP2040 which sells in quantity for $0.80 and which has sold in the millions despite being less compatible with the Acorn Archimedes predecessor than the Apollo Core is with any Amiga. The biggest problem for the Apollo Core is the following.

https://www.quora.com/How-competitive-are-FPGAs-compared-to-ASICs-and-GPUs Quote:

Efficiency comparison to ASIC Compared to hardwired ASIC; Same logic on FPGA typically roughly: o Can run at about 1/10th of the clock rate o Needs 10 times more die area o At same clock rate consumes about 10 times more power. So, hardwired ASIC can give about 100 times better performance per die area, or about 10 times better performance and 10 times better energy efficiency than FPGA, or anything in between depending on the clock rate.

So where the AC68080@100MHz SoC SBC uses 2 Watts, an AC68080@1GHz SoC SBC may use 2 Watts and has many times the value. Actually, transistors are cheap and it is relatively easy to beef up the caches so it may have much more than 10 times the performance but use more than 2 watts for $1-$3 (a 68060@1GHz may see a 40 times performance boost with more modern caches and it has low power features that may reduce the power lower than the AC68080 on semi-modern silicon). A real 68k SoC needs a fraction of the memory too as the V4SA only uses 1GiB of memory where the A600GS has 4GiB of memory to give something like 1GiB to the Amiga wasting 75% of the memory on emulation or about $15 USD of memory. A crazy good ASIC 68k SoC is possible for the price savings of the memory. The A600GS can't hit 68060@1GHz performance on semi-modern silicon, it will at times have many time the latency, load-to-use stalls kill performance without instruction scheduling as used by emulation and some JIT emulation will not be compatible. There is no elegance here. This is ARM development and not 68k development. As many issues as I have with Apollo core development, at least it is 68k development.

Last edited by matthey on 03-Nov-2024 at 11:39 AM.
Last edited by matthey on 03-Nov-2024 at 01:54 AM.
Last edited by matthey on 03-Nov-2024 at 01:52 AM.
Last edited by matthey on 03-Nov-2024 at 01:50 AM.

@matthey

Quote:

RGL approached Jeri Ellsworth about a 68k ASIC

And she told them that it was a ridiculous idea?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@matthey

Quote:

Why are Vampire V4 standalones and accelerators selling for 535+ Euros

Are they though?

Quote:

Why are there so many new Amiga 68060 accelerator boards and so few rev 6 68060s available when emulation is good enough for 99.999% of retro users?

Because that’s the “weight class� for AGA demo compos.

Bringing a PiStorm (or V4 for that matter) to a demo compo is like showing up at GoKart competition with a Formula 2 car.

Also, 68060 is a must if you wish to work with alternative operating systems.

Last edited by kolla on 03-Nov-2024 at 09:45 AM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

matthey Quote:

RGL approached Jeri Ellsworth about a 68k ASIC

kolla Quote:

And she told them that it was a ridiculous idea?

No. Jeri said she was busy doing AR/VR.

https://metro.co.uk/2022/04/08/a500-mini-amiga-console-interview-thats-our-passion-for-commodore-16427365/ Quote:

GC: How difficult was this to do on a technical level, because I’m sure I’ve heard that the Amiga is actually quite hard to emulate. Have you got the original chip in there or is this all emulated? DM: I’ll turn the clock back again to 2004. We managed to rebuild the entire chipset of a C64 on one chip. And when I say we, I mean there’s a genius called Jeri Ellsworth who is a lone woman who lives out in Portland, Oregon who’s an absolute genius. And I spent quite a lot of time with her. We redesigned how we wanted the chip to work, and she got it running on one chip; that is utterly unique for one person to be able to do, the job she did is normally the job of an entire fabrication factory. We weren’t able to replicate that for the Amiga, Jeri’s off doing other marvellous things in AR and VR. We looked at doing it again but to create a bespoke solution for it would’ve been prohibitively expensive, so it is an emulation.

She said in another video that the Amiga chipset was straight forward and easier to reverse engineer than the C64 even before an Amiga engineer slipped her the Amiga schematics.

The Amiga on a Chip Project - Too bad it was canceled
https://www.youtube.com/watch?v=5uaDzF99a80 Jeri Quote:

In fact it was much easier to reverse engineer the Amiga, even having the schematics that helped a lot, but the Amiga was a far more straight forward design than the original C64 because they didn't have to do as many naughty tricks to make things work.

matthey Quote:

Why are Vampire V4 standalones and accelerators selling for 535+ Euros

kolla Quote:

Are they though?

Thomas Hirsch is really going to town designing boards if they aren't. There are 5 different V4 accelerators and 2 V4 standalone products in production. There is a more expensive system than the V4 standalone.

Vampire Standalone Maximum + V4 Ultimate Expansion Board + ITX Case
€ 1249.00 Excluding VAT and € 1486.31 Including VAT
http://shoq.apollo-computer.com/product_info.php?products_id=28

It competes with the A1222+ for low value with a high price. They are still using the "Vampire" name even though Majsta first used it developing low priced FPGA hardware with the intention of driving down prices from what he saw as profiteering in the Amiga market. Sorry Majsta, Gunnar chose low production high margin being the business expert he is, as well as expert in everything else. There was a lot more interest in the Natami when Thomas Hirsch was in charge with a more open development than the one man's vision development. There is a thread on the apollo-core.com forum that has 362,002 views but it required years where the Natami "MX Bringup Thread" had about twice as many views in months. It was funny that Trevor in his What is an Amiga in 2024 speech at AmiWest mentioned Majsta and Renee Cousins, two other "team" casualties of Gunnar along with me, but forgot to mention Gunnar.

matthey Quote:

Why are there so many new Amiga 68060 accelerator boards and so few rev 6 68060s available when emulation is good enough for 99.999% of retro users?

kolla Quote:

Because that’s the “weight class� for AGA demo compos. Bringing a PiStorm (or V4 for that matter) to a demo compo is like showing up at GoKart competition with a Formula 2 car. Also, 68060 is a must if you wish to work with alternative operating systems.

AGA demos and 68k alternative OSs sound much too niche for the 68060 demand. There are 68060 accelerators with rev 6 68060s pushing $1000 USD. Some retro people want quality, compatibility, performance and features. The ACV4 is almost a spiritual successor but Gunnar ignored the community of Amiga developers too many times and now it is just another branch of a large divided 68k market that is smaller than it should be because of low value hardware. And no MMU for you because it is a performance limitation in FPGA which is what Gunnar has planned, developed and optimized for.

Last edited by matthey on 03-Nov-2024 at 05:03 PM.
Last edited by matthey on 03-Nov-2024 at 01:52 PM.

@matthey

Quote:

matthey wrote: Why are Analogue console remakes selling out quickly when they are often more expensive than the Minis and original hardware at the end of production? Why are people buying MiSTer setups for $200-$600 USD? With so few MiSTer boards being sold, how could a new MiSTer main board clone be produced?

*shrug* without sales numbers that isn't even a pointless set of data...

Again, if you want to make a super 68k you would need to invest big $$$ to come up with something basic and you would than produce so many to bring the cost down to the point where your are competitive in markets that could swallow such numbers.

Quote:

An ASIC 68060&AA+ SoC would use fewer transistors than the RP2040 which sells in quantity for $0.80 and which has sold in the millions despite being less compatible with the Acorn Archimedes predecessor than the Apollo Core is with any Amiga.

99.999999999999999% of current ARM costumers don't care if and to what extend that chip is compatible with some obscure homecomputer that failed in the UK and just didn't exist elsewhere.
Lets say you undercut that 80ct RP2040 with your low transistor count 68k+AA ASIC by 10ct. How much of that 70ct would be your margin and how much would you need to sell to finally break even.
And why should anybody (outside the retro Amiga community) buy such a SoC when they are plenty other options at that price point? Options that are prooven and tested, options that come with a full tool chain both for SW and HW design?

The numbers will just never match up when most of the plausible projects are ISA agnostic and there are plenty option form NOS Z80, ATMEGA, ARM, RISC_V and so that just can be bought of shelf at any given price point.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

Kronos Quote:

*shrug* without sales numbers that isn't even a pointless set of data... Again, if you want to make a super 68k you would need to invest big $$$ to come up with something basic and you would than produce so many to bring the cost down to the point where your are competitive in markets that could swallow such numbers.

MiSTer retro sales are very difficult to estimate. The main board is a FPGA development board used for the embedded market too. The 68k retro market in general is difficult to estimate for many reasons.

We are talking about low cost relatively simple hardware, what are considered EOL cores and cheap hardware to produce. A few million US dollars goes a long way. The ARM based Ouya microconsole raised $8,596,474 USD on Kickstarter in 2012 (it raised $1 million USD in 8 hours).

https://www.kickstarter.com/projects/ouya/ouya-a-new-kind-of-video-game-console/posts/376464

I would prefer not to use crowd sourcing early if at all but I believe this would be enough investment to bring such a product to market. It's not that much money any more. What is big $$$ and do you consider a 10 year old Kickstarter crowd funding campaign for an ARM microconsole with minimal available software big $$$?

Kronos Quote:

99.999999999999999% of current ARM costumers don't care if and to what extend that chip is compatible with some obscure homecomputer that failed in the UK and just didn't exist elsewhere. Lets say you undercut that 80ct RP2040 with your low transistor count 68k+AA ASIC by 10ct. How much of that 70ct would be your margin and how much would you need to sell to finally break even. And why should anybody (outside the retro Amiga community) buy such a SoC when they are plenty other options at that price point? Options that are proven and tested, options that come with a full tool chain both for SW and HW design? The numbers will just never match up when most of the plausible projects are ISA agnostic and there are plenty option form NOS Z80, ATMEGA, ARM, RISC_V and so that just can be bought of shelf at any given price point.

Why are there new SoCs introduced into the market all the time? Why was the newcomer to commodity fabless semiconductor development, RPi Foundation, able to sell millions of ASIC SoCs going into at least 30 hardware devices?

The 68k has software development tool chains. They are aging poorly with lack of updates due to emulation but they are probably as mature, if not more mature, than RISC-V developer toolchains which have been holding back hardware like the Visionfive 2 SBC.

https://www.kickstarter.com/projects/starfive/visionfive-2

The Kickstarter didn't do nearly as well but there was already lackluster RISC-V hardware available and this was just a cheaper SBC. Also, RISC-V doesn't have nearly the support of the 68k or the huge game library of software.

@matthey

Quote:

matthey wrote: Why was the newcomer to commodity fabless semiconductor development, RPi Foundation, able to sell millions of ASIC SoCs going into at least 30 hardware devices?

Are we talking about the rPI foundation that just took a cheap off the shelf put still current SoC and made a simple board around it? The one that rinsed and repeated that process for years selling millions and millions of devices until finally licensing an existing current core to have their own (semi) custom SoC made?

Quote:

The 68k has software development tool chains. They are aging poorly with lack of updates due to emulation but they are probably as mature,

They are outdated, just like the whole ISA, no 64Bit no multicore no nothing.

Just mediocre low end stuff that offers no benefits while it's support is mostly due no one haven't bothered yet to remove the long obsolete code from the projects.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@matthey

"Vampire Standalone Maximum + V4 Ultimate Expansion Board + ITX Case
€ 1249.00 Excluding VAT and € 1486.31 Including VAT"

you are here taking the most expensive solution availbable. You can also just buy Standalone V4, it is not "cheap" in todays terms but you should compare it with current other hardware options like turbocards because V4 is a major solution now (like PiStorm) and not compare it to buying a PC. All this is part of the retro platform, no current 68k hardware can even far distant compete with modern hardware. But that is not needed in my view and not expected by the customers.

Regarding Natami project, Thomas Hirsch not communicated himself so the talk about Natami had nothing to do with the real project. So I do not know how far it really was. Never saw a working system.

@kronos

yes 68k hardware can not compete with modern hardware. But who expects that? BTW combining and configuring modern desktops (in amiga terms scalos and magellan) and lots of 68k components can make it feel much more modern. I think my own aros 68k distribution, currently working intensive on a new version, is offering a lot more than even common modern platforms by default. Of course it must run on better hardware or emulation and you must live with the known shortcomings like no modern browser. But it is a playground for fun, not a competing platform to Windows/MacOS/Linux today. 3.X is very old of course, looks and feels old too and missing most of the modern components you expect today. It will become more interesting if a amiga desktop and a real amigaos theme is available on linux, then you can create something with the look & feel of amiga combined with more up to date software, drivers and a modern OS as base. But that is future now.

Last edited by OlafS25 on 04-Nov-2024 at 04:41 PM.
Last edited by OlafS25 on 04-Nov-2024 at 10:08 AM.
Last edited by OlafS25 on 04-Nov-2024 at 10:03 AM.
Last edited by OlafS25 on 04-Nov-2024 at 09:56 AM.

@OlafS25

Quote:

yes 68k hardware can not compete with modern hardware. But who expects that?

Dunno, people who also believe in untold millions just itching to get their hands on a "true" 68k for ....... well don't ask me.......

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@matthey

Quote:

There was a lot more interest in the Natami when Thomas Hirsch was in charge with a more open development than the one man's vision development. There is a thread on the apollo-core.com forum that has 362,002 views but it required years where the Natami "MX Bringup Thread" had about twice as many views in months. It was funny that Trevor in his What is an Amiga in 2024 speech at AmiWest mentioned Majsta and Renee Cousins, two other "team" casualties of Gunnar along with me, but forgot to mention Gunnar.

I know! I heard a great quote that, "there is no limit to what you can achieve as long as you don't mind who takes the credit." I think that applies here. Sometimes people think it is more important to retain control at all costs and that leads to stagnation.

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

@matthey

Quote:

A 68k SoC ASIC with both performance and compatibility like RPi hardware would kill the AmigaNOne.

The AmigaOne is killing itself. People that wanted AmigaOS put it back in via AMiNIMiga on THEA500 Mini. Those that didn't...well didn't!

For everyone else who want the AmiStore and a more involved productivity usage involving constantly updating for the chance to one day match THEA500 Mini and its plethora of softmods, there is the A600GS.

It's all covered its just Amigans are extremely picky and would rather buy an original Atari ST ("the 'auld enemy") then recommend that their friends buy THEA500 Mini! We're a strange bunch!

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

@BigD

The A600GS does not need any "soft mods" and is supplied with AmiBench and applications out of the box. All the software on the A600GS is legally licenced unlike the third party modfications to THEA500 Mini such as AGS and AMiNIMiga.

It is surely better to recommend to new members to our community to buy and use a legal product which is actively developed and supported?

Furthermore, the revenue from the A600GS gets reinvested back into Amiga projects and not extracted from the Amiga market.

Last edited by amigakit on 04-Nov-2024 at 05:19 PM.
Last edited by amigakit on 04-Nov-2024 at 05:18 PM.
Last edited by amigakit on 04-Nov-2024 at 05:17 PM.
Last edited by amigakit on 04-Nov-2024 at 05:16 PM.

_________________
Amiga Kit Amiga Store
Links: www.amigakit.com | New Products | A600GS

@amigakit

I really doubt ANY of my ex-Amigan friends will become aware of any product out there in the mainstream other than THEA500 Mini or possibly the Maxi when it becomes available.

I await their questions about the Mini and no I won't confuse them by bringing up the A600GS as an alternative in conversation. Your product, in my opinion, is for people that have given up on sourcing a 'real' A1200 on eBay, and instead "settle" for an A600GS instead. I hope it brings people who fit that profile or who instead just buy it as an impulse buy at an Amiga show, much joy!

No hard feelings, but considering lots of A500 owners pirated software as late as 1995 while amazing games like Worms were still being released for 9 year old OCS machines which deserved our support, I really don't think having a moralising attitude towards AMiNIMiga on THEA500 Mini or CaffeineOS on the Pistorm is particularly helpful! Know your audience! Maybe this is the first time they have paid for ANY Amiga software legally by buying THEA500 Mini bundle! I personally use software that I already own and add new software that I buy over USB or SD card on the Pistorm and Mini. You don't get to label everyone as a pirate who doesn't sign up to your competing product!

I hope that you reassess whether to stock the A600? Maxi for customers of yours that would like it, allowing you to skim at bit off the top rather than Amazon. I fear you're more likely to be gearing up for a trademark dispute because you used "A600" first! I hope that I am wrong!

Last edited by BigD on 04-Nov-2024 at 05:55 PM.
Last edited by BigD on 04-Nov-2024 at 05:50 PM.
Last edited by BigD on 04-Nov-2024 at 05:47 PM.

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

Kronos Quote:

Are we talking about the rPI foundation that just took a cheap off the shelf put still current SoC and made a simple board around it? The one that rinsed and repeated that process for years selling millions and millions of devices until finally licensing an existing current core to have their own (semi) custom SoC made?

Yea, the RPi Foundation that started with 5 investors targeting 1000 SBCs primarily for education with a single core scalar CPU less advanced than a 68060 into a market saturated by more powerful ARM products but creating a successful business that has sold over 50 million RPis, has become a major embedded market player setting de facto embedded form factor standards, created a RPi Holdings business worth 628 million GBP with investments from Sony and ARM Holdings and is now doing it's own fabless semiconductor designs which are also selling millions of ASIC SoC chips. They should have listened to grumpy old man Kronos talk them out of that mistake.

Market Cap: 628.78M GBP (814,710,246 USD)
https://www.google.com/finance/quote/RPI:LON

RPi CEO Eben Upton did all this starting with the original RPi in 2012 while A-Eon Technology founder Trevor Dickinson released the first AmigaOne X1000 also in 2012. Eben 50,000,000 to Trevor 5,000 but Trevor "has had a long and successful business career"?

https://www.a-eon.com/?page=about_full Quote:

About A-EON A-EON Technology Ltd is headquartered in Cardiff, UK and was established to help fund hardware and software development for the benefit of the whole Amiga community. ... Key People Trevor Dickinson: Trevor has had a long and successful business career, and as a Business Angel, has investments in a diverse range of companies covering many industries. However, his hobby and passion is the Amiga computer.

Is that kind of like Mehdi Ali's major operational turnaround of Commodore International?

https://stoneridgepe.com/files/stone-ridge-partners-brochure.pdf Quote:

PRINCIPALS MEHDI ALI Mehdi has been a principal of the firm since its inception in 1996. Mehdi’s background includes more than twenty years of operating experience. His prior experience includes serving as the President of Commodore International, where he accomplished a major operational turnaround. Mehdi served as a Vice President at General Motors Corporation, where he was instrumental in improving the performance of a number of GM’s European and Latin American operations. Mehdi was also a Vice President at PepsiCo Inc., where he headed-up a major restructuring which led to the divestiture of all non-core businesses. Mehdi was a Managing Director at Dillon, Read & Co. Inc., where he headed the firm’s restructuring business and performed turnarounds for several clients. Mehdi has a Bachelor’s and a Master’s degree from Yale University.

Did Medhi's "major operational turnaround" benefit the "Amiga community" as much as Trevor's PPC AmigaNOne for the classes though? Is it better to go bankrupt and step aside or perpetually subsidize noncompetitive hardware for the classes while blocking any real attempts for Amiga relevance? Is the "Amiga community" the masses that bought good value affordable Amigas or the AmigaNOne classes?

Kronos Quote:

They are outdated, just like the whole ISA, no 64Bit no multicore no nothing. Just mediocre low end stuff that offers no benefits while it's support is mostly due no one haven't bothered yet to remove the long obsolete code from the projects.

The 68k has supported multiprocessing since the 1980s.

A Lock-Free Multiprocessor OS Kernel
https://cs.uwaterloo.ca/~brecht/servers/readings-new/massalin91lockfree.pdf

Multi-core and 64-bit were not priorities even for the workstation market the 68000 created at that time. They are actually less of a problem to support for the 68k than they were for 808x/x86. The AC68080 added partial 64-bit support and had at least experimental multi-threading while multi-core is as simple as copy and paste when the transistor budget allows it.

OlafS25 Quote:

"Vampire Standalone Maximum + V4 Ultimate Expansion Board + ITX Case € 1249.00 Excluding VAT and € 1486.31 Including VAT" you are here taking the most expensive solution available. You can also just buy Standalone V4, it is not "cheap" in todays terms but you should compare it with current other hardware options like turbocards because V4 is a major solution now (like PiStorm) and not compare it to buying a PC. All this is part of the retro platform, no current 68k hardware can even far distant compete with modern hardware. But that is not needed in my view and not expected by the customers.

I posted the normal V4SA price first in post #361 of this thread. It offers better value than the high end "Ultimate" system which just has more expansion with the same base hardware kind of like the high end big box Amigas did not offer as much value with the same chipset as the low end Amigas like the Amiga 500, 1200 and CD32. Some people need the expansion, although less often today, and the added features and convenience of a high end system are worth the higher margin for some customers. I have no problem with this which is a good way to market the systems. My point was that even this high end "Ultimate" system is likely selling enough to create the product along with all the V4 accelerator variations despite the very limited value of using a FPGA CPU.

OlafS25 Quote:

Regarding Natami project, Thomas Hirsch not communicated himself so the talk about Natami had nothing to do with the real project. So I do not know how far it really was. Never saw a working system.

The Natami forum was open with free speech and transparent development. This is not the case at all for the AC forum and development where many people have been banned from the forum, developers have left and development remains secretive. This is Gunnar! Jens wanted to open the N68050 core source and I suspect Thomas made a deal for SAGA to be open sourced as part of allowing Gunnar to use it as announced years ago. Even back in the Natami days, there were rumors about Gunnar being domineering and disruptive in ways. Thomas didn't want to lead and he was focused on his work with perhaps the mistake of making the Natami public too soon. He gained a lot of help but also had pressure from many directions which is likely one of the reasons why he abandoned the project. There was also the big problem of the lack of a competitive CPU which was only being partially addressed with a FPGA 68k core. Financing was never addressed as he didn't want to accept money which would have increased the pressure to develop faster. I expect crowd funding would have broken records for the Natami but development needed to be closer to if not ready for production. He probably should have accepted donations which I also expect would have been substantial and should have reduced stress. Natami hardware was at an advanced and mostly operating state. Thomas is extremely talented.



Gunnar has some knowledge, skills and positive personality traits but his domineering ways unfortunately overshadow the more professional and talented Thomas and Jens. Gunnar sabotages himself and others too often. I doubt his AC toy project will ever go anywhere under his leadership. Even if an AC68080 ASIC SoC was created, I expect it would be a cheap FPGA to ASIC conversion rather than a more professional ASIC 68k SoC the 68k Amiga needs.

Last edited by matthey on 04-Nov-2024 at 07:01 PM.
Last edited by matthey on 04-Nov-2024 at 06:50 PM.

@matthey

Quote:

Yea, the RPi Foundation that started with 5 investors targeting 1000 SBCs primarily for education with a single core scalar CPU less advanced than a 68060 into a market

Sowhat?

They took an off the shelf part -> no such part available for 68k --> spend a fortune just to get to their starting line

It may have been "less advanced" but it was surely faster and better supported than a 68060 -> you still don't have any option as hunting down NOS 680x0 or ColdFire chips

That was like 15 years ago and that chips was minimal viable for the still somewhat untapped "tinkering market" --> you come around with 30+ year old designs and the performance to match it to do exactly what?

So yeah rPI is a good example on how to kickstart a platform in recent time making a perfect showcase on why anything 68k could not follow it now or 15 years ago.

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

@matthey

Wouldn't it make sense in a first phase improve upon software using cheap hardware like a standalone RPI and afterwards a ASIC 68k?

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

pixie Quote:

Wouldn't it make sense in a first phase improve upon software using cheap hardware like a standalone RPI and afterwards a ASIC 68k?

Your suggestion may seem logical but no, I don't think using emulation is a good way to improve software. Emulation doesn't spur software development and markets in the same way as hardware. THEA500 Mini and UAE haven't created large healthy software markets for the Amiga despite many tens of thousands of users. Development support like improving a 68k compiler backend or adding a 68k instruction scheduler is not worthwhile for emulation of 68k hardware which relies on inefficient emulation for performance and wastes more memory and caches than would ever be gained by improvements. Even with FPGA 68k cores with more consistent timing, compiler developers ignore these targets because FPGA cores may change and FPGA hardware is practically guaranteed to be low production, noncompetitive and unprofessional hardware. JIT emulation can have compatibility issues, undefined behavior and timing that can get out of sink, especially on low end host hardware. Try reporting a WinUAE JIT emulation bug and see how that goes. Most x86-64 hardware is high performance with more consistency than ARM hardware where low end hardware can really struggle or lose half of performance to load-to-use stalls. Emulation is EOL with customers using emulation to prolong their software investment but are less likely to invest in new software for a dying platform. Embedded market opportunities are completely lost which could improve hardware economies of scale and enlarge the hardware base. Software before the hardware may sound good in theory but it is putting the carriage before the horse.

@matthey

Quote:

Emulation is EOL with customers using emulation to prolong their software investment but are less likely to invest in new software for a dying platform.

And yet I have only run my new game Ceccenoid on my Mini thus far. I plan to buy and run Metro Siege on the Mini too! I have yet to do any coding on my real Amiga but the code runs fine under emulation. The A600GS is going sell software on its AmiStore on an ARM emulator!! Crazy!

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

@matthey

So you would envision that had the500 mini had a 68k ASIC it would be any different? Because the user base was expanded already, what is missing for new software to show up?

How come there's so many minis sold and no one taping that potential user base, emulated, or not. I have a mini, being emulated is the least of my concerns, if it has no new software is dead already, what difference makes a real 68k processor?

Where's Trevor to tap into this new wave of users instead of the attachment he has to a dead platform of a few hundreds

_________________
Indigo 3D Lounge, my second home.
The Illusion of Choice | Am*ga

@matthey

You wrote:
Quote:

An ASIC 68k SoC can be ridiculously cheap to produce

And then, the article you use to support your claim says...
Quote:

We weren’t able to replicate that for the Amiga, Jeri’s off doing other marvellous things in AR and VR. We looked at doing it again but to create a bespoke solution for it would’ve been prohibitively expensive, so it is an emulation.

So... it was a ridiculous idea.

Quote:

irrelevant blabla about Amiga chipset

Why are you babbling so much about the chipset - it has been reimplemented a handful of times now - do you plan to have Amiga chipset on your ASIC 68k SoC which can be ridiculously cheap to produce?

Quote:

matthey Quote: Why are Vampire V4 standalones and accelerators selling for 535+ Euros kolla Quote: Are they though? Thomas Hirsch is really going to town designing boards if they aren't. There are 5 different V4 accelerators

Actually, just three - for A500, A600 and A1200.

The one for A1000 is the A500 one with a CPU socket relocation board, and the A2000 one is the A500 one with a simple adapter for the A2000 CPU expansion socket - there are quite a few of these around, and the AC ones can even be used with PiStorm.

On the other hand, where is the Draken?

Also - how does it compare to PiStorm?
Original PiStorm (68000 socket, A500, A1000, A2000, CDTV - with or without CPU socket relocation boards)
PiStorm 600, like original PiStorm, but reworked to piggy-back on the A600 SMD 68000 CPU.
PiStorm32 - PiStorm for the 32bit A1200 (also works with CD32, but not "comfortably")
PiStorm16 - "Real Soon Now", new PiStorm for A600, reworked with knowledge and experience gained over the years
Expect a new PiStorm16 based variant for A1000/A500/A2000/CDTV as well.

Quote:

and 2 V4 standalone products in production.

Isn't the only difference 512MB of RAM?

Anyways, you've not answered the question - are they _selling_?

How many V4 users do you know?

Quote:

AGA demos and 68k alternative OSs sound much too niche for the 68060 demand.

Exactly - it is very niche. But that's really where it's at now.

What are the features of your fantasy cheap 68k SoC?

Quote:

There are 68060 accelerators with rev 6 68060s pushing $1000 USD. Some retro people want quality, compatibility, performance and features.

Yes, compatibility with the "high end" AGA demos, for the most part.


Quote:

The ACV4 is almost a spiritual successor

It isn't - classic Amiga was a rather open architecture, the ACV4 is the opposite.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC