kolla Quote:

So you’re saying you and Matthey cannot even agree on an imaginary SoC… and you wonder why others cannot join efforts on real projects?

No. I believe Agami and I are on the same page. Agami gave reasonable enough base cost estimates for a quality SoC ASIC and board production which is difficult considering all the variables. The goal of an ASIC is to integrate as much functionality as possible into a SoC as this maximizes cost reductions and simplifies the parts supply chain and SBC. It is reasonable to suggest that hardware I/O functionality could initially be left out of the SoC and provided by commodity I/O chips due to time or financial constraints. Modern hardware I/O is not simple and easy. It would be good to find a hardware I/O expert like Joe Decuir.

https://en.wikipedia.org/wiki/Joseph_C._Decuir#Engineering_standards Quote:

Decuir made substantial technical and editorial contributions to wired and wireless communications engineering standards, including: o ITU-T V-series modem standards: V.8, V.8bis, V.32bis V.34, V.80, V.90, V.250, V.251 and V.253 o ITU-T T-series facsimile standards: T.30, T.31 and T.32 o European Telecommunications Standard Institute (ETSI): ETS 300 642: AT Command Set for GSM Mobile Equipment (ME) o Universal Serial Bus (USB) standards: USB 1.1 & 2.0, Interchip USB 1.0 and High Speed Interchip USB 1.0 o USB Communications Device Class, Wireless Mobile Communications and Network Control Model o European Computer Manufacturers Association (ECMA International): ECMA-368, High Rate Ultra Wideband MAC and PHY Standard o Wireless USB v1.1 o Bluetooth Core Specifications: v3.0, v4.0, v4.1 and v4.2 o Bluetooth Profiles and Services: RESTful API, Internet Protocol Support v1.0, HTTP Proxy Services v1.0

Joe is getting old but appears to be mentally sharp and may be interested in helping. He is already familiar with the Amiga hardware and has the experience of the original Amiga development team. Other ex-Amiga developers have helped or offered help with modern Amiga hardware projects like Davie Haynie. Bring back some of these guys and break out the old Amiga schematics and this turns into a very cool modernized retro project which would draw interest and publicity from around the world and could result in more financial and developer support.

kolla Quote:

Here’s your chance, both of you - figure out together here in the open how this imaginary SoC of yours will work, which hardware to use, what features the 68k CPU must have, what bus technologies to use, which operating systems and software you want it to run, and how that can accomplished in terms of drivers etc.

We can only look at goals and costs generally. We don't know our resources and budget. Without a budget and green light on a project we aren't getting quotes and have minimal interaction with potential developers and business partners. We have some technology and business understanding but we aren't hardware guys who we need input and advice from. Cost benefit analysis hasn't been performed. An IP lawyer hasn't reviewed the plans. This is not a hard project but more of a brain storming proposal and even if it did get a green light would require additional exploration. AmigaNG hardware R&D is currently about assembling commodity hardware which is trailing the technology leaders and falling behind. Agami and I are just thinking outside the box and exploring business ideas which are difficult but reasonable and I believe even compelling.

Last edited by matthey on 08-May-2022 at 07:31 PM.
Last edited by matthey on 08-May-2022 at 06:43 PM.
Last edited by matthey on 08-May-2022 at 06:38 PM.

@matthey

So you are all in on the concept of a 68k SoC with Linux as primary target OS, like Agima suggests.

Well, that’s news worthy. Awesome.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

kolla Quote:

So you are all in on the concept of a 68k SoC with Linux as primary target OS, like Agami suggests. Well, that’s news worthy. Awesome.

I would like to support having a 68k MMU and open enough hardware to allow Linux. Choice is good, extra sales for Linux improves economies of scale and hardware testing under Linux would likely benefit development. I'm not sure how popular 68k hardware would be for Linux with all the ARM competition though. The AmigaOS is quite nice on low end low resource hardware and has software including many games without requiring recompiling. I would hope there would be other alternative OSs including retro ones like for the Atari ST, x68000, Mac and NeXTcube/NeXTstation. Some of the hardware features may be missing but emulation should be simplified at least as Mac emulation was on the Amiga. A MMU would allow remapping memory locations for emulation. The AmigaOS could likely gain optional partial memory protection for new software and/or a new next generation AmigaOS could provide process isolation and memory protection.

@kolla

Quote:

kolla wrote: @agami So you’re saying you and Matthey cannot even agree on an imaginary SoC… and you wonder why others cannot join efforts on real projects?

First of all, @matthey and myself are not yet partners and this is not yet a fundend business venture, but we are very much philosophically aligned.
Quote:

as matthey wrote: No. I believe Agami and I are on the same page.

We might have slightly different ideas on execution, but ultimately our goal is the same.


Quote:

kolla goes on to write: Here’s your chance, both of you - figure out together here in the open how this imaginary SoC of yours will work, which hardware to use, what features the 68k CPU must have, what bus technologies to use, which operating systems and software you want it to run, and how that can accomplished in terms of drivers etc.

The details of the above can take quite a bit of time, but it wouldn't take too long for @matthey and myself to have a bullet-point list. Which of course will only be a wish list of a target SoC/board + software environment, saying nothing of the intermediary steps, nor duration or effort, until funding is available to allow for a more detailed design and plan.
Quote:

as matthey wrote: Without a budget and green light on a project we aren't getting quotes and have minimal interaction with potential developers and business partners. We have some technology and business understanding but we aren't hardware guys who we need input and advice from...

Last edited by agami on 09-May-2022 at 01:24 AM.

_________________
All the way, with 68k

Looks like a first firmware update for A500 mini out,

firmware update includes –
• Adds ADF (floppy) file support
• Adds Playlist support
• Adds Hot-Crop
• Adds Keyboard control of the user interface
• Allows the Virtual Keyboard to be accessed at all times
• Allows controllers to be physically swapped during play
• Swaps our THEJOYSTICK fire buttons
• Fixes the Virtual Keyboard Cursor-Right button

Pleased to see ADF support add, good on retro game adding that official, odd it wasn't their from the start, but still no HDF support and iso support would be nice.

there are reports this disables host run and might make some hacks not work, so i maybe hold off for more testing
https://retrogames.biz/support/thea500-mini/manuals/update-1-1-0/

_________________
AmigaNG, YouTube, LeaveReality Studio

@amigang

It doesn't seem to play nice with Pandory, the most interesting emulation soft mod only released a couple of days ago itself! Hold off for a bit IMHO.

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

So how’s it going with the mini?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@kolla

Pros
We have 4-player adapter game support and 2-player mouse game support through Pandory.

The 1.1.1 firmware update adds multi-disk ADF support and corrects the button mapping on THE JOYSTICK (but PLEASE see Cons before updating).

CD32 games plus CD music support works well now (see the Discord group).

Cons
The 1.1.1 update takes away Pandory compatibility (this can be mitigated with the PCUAAE but only if set up BEFORE updating the stock firmware).

The 1 Amp power supply recommended does not seem sufficient for stable updating of firmware. 2 Amp power supplies are recommended to avoid bricking your console during updates!

AMiniMiga Deluxe Paint IV & V implementations has some major bugs which limits its use for animation.

Some demos run slow on the RGL Carousel as does Jim Power.

Summary
A great bit of kit but you need a 2 Amp power supply, PCUAAE, the firmware 1.1.1 update installed on a memory stick AFTER installing PCUAAE then Pandory and AMiniMiga in addition to the standard RGL Carousel to get the most out of the system IMHO. And even then Deluxe Paint is buggy at the moment under AMiniMiga (the only way to run it)!

P.S. I haven't installed PCUAAE or firmware update 1.1.1. myself because I use a 1 Amp power supply and am scared of bricking the machine! It still works great and I use a separate memory stick for use with my THE JOYSTICK with different mapped controls configs for ALL my sideloaded games (set up manually) so that I can use it with all my favourite games either with a pad or joystick (depending on preference on the day) without upgrading the firmware. I have also softmodded it to play single disk ADF games with the 1.0 firmware. It works for me.

Last edited by BigD on 25-Jun-2022 at 10:37 PM.
Last edited by BigD on 25-Jun-2022 at 10:36 PM.
Last edited by BigD on 25-Jun-2022 at 10:30 PM.

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

@BigD

Quote:

The 1 Amp power supply recommended does not seem sufficient for stable updating of firmware. 2 Amp power supplies are recommended to avoid bricking your console during updates!

Thanks for that tip, I did wonder why mine bricked after I did the update and was struggling to find any info about that online. Thankfully Amazon arranged for a replacement which was delivered the next day.

Based on my very positive experiences with a C64 Maxi, I'd be very tempted to buy a full size version of the A500 mini as it desperately needs a keyboard and some extra USB ports for proper use especially with the AminiMiga package.

_________________
Amigaone X5000/20 4GB Radeon RX 550 Polaris 12 AmigaOS4.1 Final Edition Update 1
Amiga 1200 Workbench 3.1.4
Amiga CD32

@thread

Somewhat related. Evercade utube discussion.
Includes talk of the relationship with RetroGames Ltd., licensing and sublicensing, C64 coming to Evercade, and a few comments about Amiga that might be connected.

https://www.youtube.com/watch?v=HAJq5ggkTj8&ab_channel=BaconIceCreamProductions

About 5 and 1/2 minutes in, but perhaps more is worth a listen.

#6

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

@thread

A bit more on the topic in an article today:

Here

Scroll down for Amiga mention.

#6

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

Quote:

Based on my very positive experiences with a C64 Maxi, I'd be very tempted to buy a full size version of the A500 mini as it desperately needs a keyboard and some extra USB ports for proper use especially with the AminiMiga package.

I really wonder how likely it is that they're going to make an A500 Maxi. The original A500 had such a large footprint compared to the original C64, so a lot more plastic is going to be needed to make the exterior case. That also translates into a much larger retail box that occupies more retail space. That's going to drive up the cost.

And the original A500 never matched the original C64's popularity, which we all know is still the best-selling personal computer of all time. That might mean that an A500 Maxi would never sell as well as the C64 Maxi.

Given the above two points, would Retro Games refrain from taking a chance on making an A500 Maxi?

I personally wouldn't mind an A600 Maxi, with its much smaller form factor that would translate into a lower cost. But the original A600 was nowhere near as popular as the original A500, so I don't think many people would be interested in buying an A600 Maxi.

It's been over three months since anyone posted on this thread. Is everyone still enjoying their A500 Mini?

I really, REALLY want one, but I refuse to pay the price that it current retails for. Here in Canada, I'd have to pay over CAD $200 to buy one, which is twice the price that the C64 Mini cost when it was first released.

I'm hoping that the A500 Mini eventually comes way down in price like the C64 Mini did. But I've got an awful feeling that won't happen, as it seems like the massive price reduction on the C64 Mini was due to too many units flooding the market. I get the impression that Retro Games didn't make as many A500 Mini units to saturate the market the same way.

Last edited by MarcioD on 18-Oct-2022 at 05:18 AM.

@matthey

Quote:

Joe talks about creating a cheap easy to use computer for retro creative use. The Birth of the Commodore Amiga - Interview with Engineer Joe Decuir 4K UHD https://youtu.be/9TWuTKJNZIk?t=1075 Too bad that is a Raspberry Pi instead of the Amiga.

Raspberry Pi is surfing the ARM CPU wave.

PiStorm (Broadcom's ARM-based SoC) and TF1200 Buffee (TI's ARM-based SoC) projects enable the classic Amiga hardware to also surf ARM CPU waves.

ARM's R&D CPU cost is spread across many vendors.

Quote:

Sure. CPU cores are good at handling sequential data and the chipset is parallel logic. FPGAs are good at parallel logic. GPUs are good at parallel logic too but often have considerable latency.

AMD's RDNA has reduced latency e.g. GCN issued one instruction per wave once every 4 cycles while RDNA issues instructions every cycle.

On GCN, vs RDNA in Wave32 or “backward compatible” Wave64.

GCN Vega takes 12 clock cycles to complete the instruction on a GCN SIMD.

NAVI (RDNA) in Wave32 (optimized code) completes it in 7 clock cycles. In backward compatible (optimized for GCN Wave64) mode, NAVI completes it in 8 clock cycles.

RDNA 2 has an extra clock cycle latency due to higher clock speed optimizations.

Instructions are executed in the pipeline. RDNA 2 IGP on Ryzen 7000's TSMC's 5 nm process tech can reach 3 GHz OC.

RDNA 3 on TSMC's 5 nm process tech is said to reach 4 GHz. At high clock speeds, RDNA's scalar instructions would be fast.


Quote:

Hardware compatibility is easier to test today in FPGA as Amiga compatible FPGA devices demonstrate. Logic simulation and verification is easier than ever on computers. ASIC creation has become a cheap commodity service. IP is readily available for licensing from eDRAM to GPUs. An Amiga on a single chip can be produced for about $1 today and a whole computer for about $5. Compare all this to the hand laid out logic, wire wrapped prototypes, lack of computer simulation and the immense cost of the development and the Amiga custom chips themselves. Joe Decuir talks about some of it in the video I posted above. The Birth of the Commodore Amiga - Interview with Engineer Joe Decuir 4K UHD https://www.youtube.com/watch?v=9TWuTKJNZIk Everybody can do software, emulation and virtual hardware today though. It's just fine for Amiga Nowhere.

Unlike Emu68/PiStorm, "Amiga Anywhere" doesn't run classic Amiga 68K legacy and it's useless for most retro Amiga users.

Last edited by Hammer on 18-Oct-2022 at 06:18 AM.

_________________
Ryzen 9 7900X, DDR5-6000 32 GB RAM, GeForce RTX 4080
Amiga 1200 (rev 1D1, KS 3.2, TF1260, 68060 @ 63 Mhz, 128 MB)
Amiga 500 (rev 6A, KS 3.2, PiStorm/RPi3a/Emu68)

@MarcioD

Quote:

MarcioD wrote: It's been over three months since anyone posted on this thread. Is everyone still enjoying their A500 Mini? I really, REALLY want one, but I refuse to pay the price that it current retails for. Here in Canada, I'd have to pay over CAD $200 to buy one, which is twice the price that the C64 Mini cost when it was first released. I'm hoping that the A500 Mini eventually comes way down in price like the C64 Mini did. But I've got an awful feeling that won't happen, as it seems like the massive price reduction on the C64 Mini was due to too many units flooding the market. I get the impression that Retro Games didn't make as many A500 Mini units to saturate the market the same way.

Here it's still $199 AUD
A500 mini on Amazon.com.au

I have a feeling that RetroGames did not flood the supply chain like they might have done with the C64 mini, so you might have to wait for a while to see that discount on the A500 mini. If there are any left after the holiday season, then you could see price reductions.

How's the enjoying going so far? I myself have not even taken it out of the box yet. Haven't had the time.

The A500 maxi does not need to be the actual size of an original A500.
C64 maxi is different as an actual C64 uses a compact keyboard.

I too would prefer the smaller form-factor of the A600. Perhaps they could create their own A500 compact form-factor which looks like an A500, but narrower to only house a compact (60%) keyboard.
The A500 Compact. The A500 Mini+. The A500 Mini Pro. The A500 II.

Either way, if they do end up making an A500 maxi, I will buy one. And probably have it in the box until the year-end break to get around to using it.

Last edited by agami on 18-Oct-2022 at 07:35 AM.

_________________
All the way, with 68k

Hammer Quote:

Raspberry Pi is surfing the ARM CPU wave. PiStorm (Broadcom's ARM-based SoC) and TF1200 Buffee (TI's ARM-based SoC) projects enable the classic Amiga hardware to also surf ARM CPU waves. ARM's R&D CPU cost is spread across many vendors.

Is emulation on a smart phone SoC for a desktop like hobby computer the way to surf the ARM CPU wave for the Amiga? It's probably good enough for Acorn Archimedes RISC OS fans but the more popular and powerful 68k Amiga should be hosting the less popular and weaker CPU platforms. 68k CPU R&D and production costs could be spread across multiple customers including embedded customers as well.

agami Quote:

Here it's still $199 AUD A500 mini on Amazon.com.au I have a feeling that RetroGames did not flood the supply chain like they might have done with the C64 mini, so you might have to wait for a while to see that discount on the A500 mini. If there are any left after the holiday season, then you could see price reductions.

With supply/logistics bottlenecks and energy/transportation inflation, it may be an expensive Christmas this year. Retro Games Ltd. may not be able to ramp up production of THEA500 Mini like they want to so any planned discount may be skipped for awhile (if they have enough stock of SoCs which reduces BoM they could be ok too). Within the next few years, I expect a major turnaround and electronics to become cheaper than they were before as extra chip foundry capacity comes online. R&D now with production later is the way to take advantage not that Amiga businesses have been smart for a long time.

agami Quote:

How's the enjoying going so far? I myself have not even taken it out of the box yet. Haven't had the time. The A500 maxi does not need to be the actual size of an original A500. C64 maxi is different as an actual C64 uses a compact keyboard. I too would prefer the smaller form-factor of the A600. Perhaps they could create their own A500 compact form-factor which looks like an A500, but narrower to only house a compact (60%) keyboard. The A500 Compact. The A500 Mini+. The A500 Mini Pro. The A500 II. Either way, if they do end up making an A500 maxi, I will buy one. And probably have it in the box until the year-end break to get around to using it.

The Amiga 600 was cute but a flop that contributed to its demise. It's not worth it to delete the numeric keypad which is annoying. The Amiga 1200 was popular and has the keyboard. It was not as popular as the Amiga 500 but it is more practical, expandable and sexier. Many Amiga 500 users knew about the Amiga 1200 and AGA but didn't upgrade as the Amiga was falling behind. The experience can be improved over a sluggish 68020+AGA that CBM offered. The idea should be to make the 68k Amiga like CBM should have made it and they didn't use emulation.

@number6

Quote:

A stick with the RetroGames Ltd. trademarked name on it (as well as the stylized Amiga "A"), conveys a seriously bad message regarding distribution.

Story yesterday from Reuters:

Etsy shares fall after research firm says platform showcases fake goods

#6

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