@ne_one,

Who cares what it's called, Call it "68K Max PC", can make it an Atari if you want maybe.

Chris

ne_one Quote:

By "Amiga" I'm assuming you mean a venture in a parallel universe excluding parties currently involved in litigation? No one controlling this platform has had the wherewithal to capitalize on the brand in a market clambering for nostalgia.

The venture wouldn't exclude all parties currently involved in litigation. The winners can participate.

The Amiga parties have become more 68k friendly recently.

Cloanto/Amiga Corporation - all Amiga products have been primarily for the 68k
Hyperion - AmigaOS is being developed for the 68k while PPC development is stalled
AmigaKit - always had good 68k support

A-Eon - primarily PPC Amiga supporter (Trevor has 68k Amigas in his collection though)

ne_one Quote:

In the last few years we've seen virtually every legacy consumer technology enjoy a renaissance. Some companies like Arcade1Up have produced millions of units selling at hundreds of dollars based on long-dismissed properties. PONG anyone?

Pong seems pretty simple compared to SpaceWars on a PDP-1 from a decade earlier.

PDP-1 - Play 1962's SpaceWar! MiSTer Countdown #21 | SmokeMonster
https://www.youtube.com/watch?v=CFMKFdkFEEE

Pong was a little cheaper. The PDP-1 cost about $120,000. Price makes a big difference as to what becomes popular.

ne_one Quote:

It's easy to slag the A500 Mini for its cost or lack of capability (neither of which is relevant) but it appears that it will be an actual product and perhaps even a successful one. Kudos to these guys for making something happen.

They have done a good job so far from a marketing and management perspective. Few businesses have taken the risk of trying to mass produce individual retro hardware products without using big appeal brand names. Their prices are highly profitable but there isn't much competition in the markets they chose. It sure is different from OS development where profitability is low and the strong competition is mostly free. The way to make OS development affordable is to bundle the OS with mass produced cheap hardware to spread out the development costs. Just $1 from every THEA500 mini sale would really add up. I wonder how much Michele will make off each THEA500 mini sale (or is it a one time set payment?). If he gets a payment per unit sold, he would know the sales volume eventually which could be useful for his own Amiga market analysis.

@matthey

I am sure the Linux developers will appreciate the 1 dollar per unit offer that you suggest.

Btw - Cloanto’s products are not mainly 68k, they are mainly targeting x86/amd64 and Windows.

Last edited by kolla on 15-Aug-2021 at 10:19 PM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@thread

In less than a week the original announcement utube has almost 60,000 views.

I was watching some mainstream utubes that only covered this as part of their overall shows.
They seem surprised somewhat and most of the utubers either weren't even born at the time or too young to have experienced the original.

#6

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

@matthey

Quote:

The price of electronics have dropped significantly since 1987. THEC64 uses an ARM SoC chip which is available in low volume for $7. The majority of the chip logic is for a 3D GPU which isn't even used. A custom 68k Amiga SoC which is much more powerful than any 68k Amiga should require less logic than this ARM chip and could be produced cheaper if mass produced. As I have stated before, the opportunity is available to create a quality low priced mass produced Amiga which competes with the Raspberry Pi and Arduino and gets the Amiga back into hardware production. I believe the Amiga has a better chance of competing in the low cost computer hardware manufacturing market than as a software only AmigaOS producer competing with free operating systems. Why use a 1 GHz ARM SoC chip to emulate a 7 MHz Amiga when a 1 GHz 68k Amiga SoC chip can be produced for the same price?

Who will pay for 1 Ghz 68090 R&D? ARM's R&D cost is spread across multiple IHVs with many million devices.

For comparison, K7 Athlon Thunderbird on 0.18 μm/180 nm process node has 600 Mhz to 1400 Mhz clock speed and 37 million transistors.

Xbox Series S has $199 USD retail price with a performance that obliterates Raspberry Pi 4 ($75 USD just for barebone mainboard with 8 GB LPDDR4).

Last edited by Hammer on 16-Aug-2021 at 05:59 PM.
Last edited by Hammer on 16-Aug-2021 at 05:57 PM.
Last edited by Hammer on 16-Aug-2021 at 05:56 PM.
Last edited by Hammer on 16-Aug-2021 at 05:52 PM.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

#6 Quote:

In less than a week the original announcement utube has almost 60,000 views.

I'm not surprised. I used to mention the Natami "MX Bringup Thread" which had nearly 800k views before the counter was reset. I believed there was enough demand to make a 68k SoC ASIC back then and started planning for it. Gunnar continued to hyper-optimize his Apollo core design and ISA for FPGA only. Gunnar still doesn't understand how much difference in competitiveness (performance/price) an ASIC could make. Some people's foresight is better than other people's hindsight.

Hammer Quote:

Who will pay for 1 Ghz 68090 R&D? ARM's R&D cost is spread across multiple IHVs with many million devices.

ARM spreads its costs out in the embedded market. I planned to do the same. One embedded business partner using high volumes (millions) of processors could be enough. I found potential embedded business partners who like the 68k. Of course my efforts were sabotaged by someone planning for a FPGA 68k core forever.

Hammer Quote:

For comparison, K7 Athlon Thunderbird on 0.18 μm/180 nm process node has 600 Mhz to 1400 Mhz clock speed and 37 million transistors. Xbox Series S has $199 USD retail price with a performance that obliterates Raspberry Pi 4 ($75 USD just for barebone mainboard with 8 GB LPDDR4).

As a first 68k core design, the target would not be max performance. The cores would likely use 10W or less for fanless designs which reduce cost and increase reliability. The fab process used would likely target transistors/cost and power efficiency. A logical but flexible target performance and price would be between the Raspberry Pi 4 and Raspberry Pi Zero.

Last edited by matthey on 16-Aug-2021 at 08:47 PM.

@matthey

So which 68k design do you wish to see in ASIC? D68000?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

kolla Quote:

So which 68k design do you wish to see in ASIC?

The best one. There are several candidates but most would need work to customize features and optimize the design and routing for an ASIC.

kolla Quote:

D68000?

The D68000 has advantages like being fully synthesizable, ASIC ready, has gone through some verification and testing, is a static synchronous design and looks like it may be complete and compatible with the 68000. Disadvantages are lack of 68020 support, likely a shallow pipeline design like the 68000, primitive MUL and shift support (poor performance like the 68000), likely not superscalar and no or minimal caches (like the 68000). The performance/MHz is likely to be poor and lack of pipelining would limit being able to clock it up. Digital Core Designs mostly makes shallow pipeline designs for deeply embedded use. Even though they claim to have 300,000,000 cores designed by them in devices, their experience with even moderately pipelined designs appears to be limited.

Quote:

Btw - Cloanto’s products are not mainly 68k, they are mainly targeting x86/amd64 and Windows.

I suppose it is an issue of definitions - but I have copied the AmigaForever contents to the HD of my A2000 and run the software from there.

@thread

First...copying brief history from another post:

Quote:

Amigakit asked "me" in the forum publcially if "THEA500" was actually the name they were going to use. (I have no idea why he asked me) Shortly thereafter, Amigakit filed an objection to the "THEA500" filing to register that trademark. This was done when the person of record was still "Angela" Andrews. The objection failed. "THEA500" was registered and changed registrant to Retro Games Ltd. Then Amigakit tried to register "A500" again... Then Retro Games Ltd. filed for an extension to consider making official objection to the "A500" trademark filing. Additional and up to date info is available that clearly shows a change in representative (lawfirm) has been made in order to somehow show "A500" is unique or can co-exist with the similarly named trademark of "THEA500". If I am wrong about any of this, I'm sure someone will correct my wrong thinking.

Today's update:

Quote:

Status:Withdrawn

Source

#6

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

@matthey

Blablabla… so not D68000 - which then?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@Nonefornow

I don’t see many Cloanto products there, do you?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@Nonefornow

You can make your own Amiga collection, to sell, you just need to download stuff from Aminet, read the license make sure its PD, GNU or MIT, or ask the developer who wrote if its ok for put on a CD. And burn CD’s, stamp your logo on it, and make some cash.

only a small % go to the developer, who wrote it.

Last edited by NutsAboutAmiga on 01-Sep-2021 at 10:06 PM.
Last edited by NutsAboutAmiga on 01-Sep-2021 at 10:06 PM.

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

@number6

This feels like amigakit realised it had not that many A500 products and likely thought they be better off trying to just protect the A1200 trademark, ie “quick let’s release some A1200 mouses to up our portfolio”

Also how come they dont sell AmigaOs 3.2?

Last edited by amigang on 02-Sep-2021 at 10:38 AM.

_________________
AmigaNG, YouTube, LeaveReality Studio

@amigang

And badges, don’t forget the official A1200 badges :)

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@matthey

Quote:

ARM spreads its costs out in the embedded market. I planned to do the same. One embedded business partner using high volumes (millions) of processors could be enough. I found potential embedded business partners who like the 68k. Of course my efforts were sabotaged by someone planning for a FPGA 68k core forever.

ARM spreads its expenses across embedded and mobile phone markets. Apple's M1 R&D has iTunes/App Store (walled garden) subsidies and multi-million units run.

So, your biz proposal is dependant on Apollo's AC68080 IP?

There's baseline TG68 FPGA or do a licensing deal with NXP/Freescale's 68060 or ColdFire 5102 (includes 68EC040 compatibility) IP. AC68080 should be renamed into AC68EC080 since it's missing MMU, hence no embedded Linux 68K.

Quote:

As a first 68k core design, the target would not be max performance. The cores would likely use 10W or less for fanless designs which reduce cost and increase reliability. The fab process used would likely target transistors/cost and power efficiency. A logical but flexible target performance and price would be between the Raspberry Pi 4 and Raspberry Pi Zero.

Raspberry Pi 4's low cost is based on Broadcom's BCM2711 SoC which is not limited to just Raspberry Pi 4.

Raspberry foundation has some UK government subsidies.

Last edited by Hammer on 02-Sep-2021 at 07:07 AM.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

@kolla

Quote:

I am sure the Linux developers will appreciate the 1 dollar per unit offer that you suggest. Btw - Cloanto’s products are not mainly 68k, they are mainly targeting x86/amd64 and Windows.

Prior to Hyperion's AmigaOS 3.1.4, Cloanto sells AmigaOS 3.X ROM. My Amiga 1200 was fitted with Cloanto's AmigaOS 3.X ROM chips.

I could use Cloanto's AmigaOS 3.X ROM chips and my existing AmigaForever 2016's AmigaOS 3.X installation on real Amiga hardware.

My Amiga 1200 with AmigaOS 3.X ROM chips was short-lived since I bought Amiga 1200 (sold as broken for parts, hence its low cost, the seller was ignorant) during 1st wave COVID-19 lockdown. My Amiga 500 was in storage prior to the 1st wave COVID-19 lockdown. Once my Amiga 500 was operational again (I gutted rev 5's chips into rev 6A motherboard + purchased ECS Agnus) and Amiga 1200 rev 1D1 arrived, further hardware purchases followed.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

@Hammer

The AC68080 DOES have an MMU, but it is not accessible from AmigaOS, nor is it openly documented. All Gunnar has said is that it is modern, nothing like the "old school" 68k MMUs, and it is already used by the Apollo Core.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@Hammer

Congrats with that one 68k product from Cloanto.
(However, anyone of us can put together kickstarts like that)

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@thread

gamepad & tank mouse both to be available separately

#6

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