@bhabbott

Quote:

ECS was new for the A3000. Amber was part of the graphics chipset in the A3000.

You just contradicted Commodore engineers i.e. Mike Sinz and Dave Haynie.

Amber was also found in the A2320 add-on card. This is not unique to the A3000.

Quote:

If the A3000 was commanded to have 'no new [graphics] chips' then it would have been OCS.

You just contradicted Commodore engineers i.e. Mike Sinz and Dave Haynie.

1. Except for the 2MB address capability, A500 Rev 6A's ECS Agnus A has all the improvements from A3000's ECS Agnus A and B.

2. A500 Rev 6A has the jumpers for 2MB Chip RAM configuration. The prototype A500 Rev 6A would have tested the 2MB Chip RAM capability.

From https://www.amibay.com/threads/2megs-chipram-on-rev-6a-amiga-500-motherboard.89654/
This has the jumper config information for 2MB Chip RAM for A500 Rev6A.

Jumpers JP3, JP5, JP2, and JP7A need to be changed.
Insert ECS Agnus AB or B
Remove four DRAM chips.
Install four 1 MEG DRAM chips.

This is A500 Rev6A's reserve 2MB Chip RAM capability that is inherent on this PCB model.

You created an argument based on product releases, not on the production prototype.

Quote:

The truth is, the 'no new chips' directive referred to getting the design finalized without further delays, which there would have been if they had added more features. So what do you want - ECS, or nothing? AAA was in development for years and they couldn't get it working properly. CSG engineers doubted that even AGA would be possible using the process they had, let alone the extra stuff people here are proposing.

AAA has been in development since 1989. http://www.bambi-amiga.co.uk/amigahistory/leweggebrecht.html

Only 1 year of serious development for AAA. AA was branched from it e.g. 16 million colors palette.

Timeline
1984 = 5 bitplane display for 32 colors. Amiga Lorraine development prototype shown in CES 1984.

1985 = 6 bitplane EHB display for 64 colors. Some original NTSC A1000 is missing 64 color EHB mode.

1987= 7 bitplane for 128 colors and 2 MB Chip VRAM. R&D was canceled and the proven original Amiga team was dismantled.

There's an R&D disruption.

Commodore management selected C900's path i.e. low colors with high "business" resolution.

1989, the moonshot AAA R&D started. Only 1 year of serious development for AAA. AA was a panic move. A500 Rev 6A (ECS Agnus, ECS 2MB Chip RAM capable motherboard) was released.

(Based on the original Amiga's team's chipset evolution pace, the 8-bitplane chipset should been completed by this point).

1990, C65 team's 256 colors chipset was operational in Dec 1990. This is designed for MOS 65xx efficient memory access.

1991, AA3000+'s 256 colors AGA chipset was operational (bootable state) in Feb 1991.

C65 was canceled in Q1 1991. Wasted R&D resources. A300 R&D replaced C65.

Commodore Germany demanded a hard disk-capable Amiga model during A300's R&D. (Commodore the Inside Story). IDE mandate was born e.g. Gayle R&D which also impacted Budgie (Buster/Ramsey).

"Wasted more than 6 months on ECS A1000Jr" adventure. Budgie and Gayle chips don't exist in this model.

AGA's Buster/Ramsey (leads to Budgie) has to be modified for Gayle's IDE/PCMCIA link.

1992, Gayle IDE-equipped A600 (A300 project) was released in March 1992. A500 was discontinued. A600 sales flop and tanked Commodore's revenue. The A600 looks like a toy when compared to the A500.

A600's IDE and PCMCIA are meaningless for kick-the-OS Amiga games.

1993, the key original Amiga team-led 3DO product was released in Q4 1993. CD32 was released in 1993.

3DO is true color (24-bit) and high color (16-bit) capable and quadrilateral 3D accelerator @ 25Mhz. Commodore's infamous 25 Mhz speed issue still exists with the Madam graphics chip. https://3dodev.com/_media/documentation/manuals/panasonic_fz-1_-_technical_guide_us.pdf

---
AA3000+ was later cloned by the Amiga community.
C65 was later cloned by the C64 community.

Quote:

Mike Sinz can call the A3000 design 'radical' if he wants, but in reality it wasn't that different. Zorro-III was basically just Zorro-II with a multiplexed bus to get 32 bit.

The Baster chip evolved into the Super Baster chip. PC had the 32-bit EISA and 32-bit MCA.

Quote:

Super DMAC was just a DMA controller.

16-bit DMAC provides a DMA interface.

32-bit Super DMAC provides a DMA interface.

BusSpeedTest for A3000's motherboard memory.
read = 13.4 MB/s
write = 7.9 MB/s

Quote:

RAMsey was just a DRAM controller. 25MHz 030 had already been used on the A2630 card (which also used ZIP RAM). So in what way was it a 'radical departure'?

A2630 does NOT have Ramsey and Super DMAC (32-bit DMA controller) chips. https://amiga.resource.cx/photos/photo2.pl?id=a2630&pg=3&res=hi&lang=en

A2091 and A590 addons had 16-bit DMAC.

A3000's Fast RAM has both DIP and ZIP sockets.

Quote:

It's not just the CPU you have to consider. A 25MHz 020 is a waste when the rest of the system only runs at 14MHz. Some have suggested overclocking to 28MHz for synchronous operation, but this is a no-no. if for some reason they were flaky, Motorola would say "Your fault, not ours. No warranty!".

Reminder, Ramsey can support 25 Mhz 68EC020/68020/68EC030/68030.


Quote:

The context is clear. They stuffed the A3000 full of expensive new chips, including the 2MB ECS Agnus and Amber.

2 MB ECS Agnus wasn't expensive when the A500 Rev 8A motherboard was used in batches of disguised 1991 A500s.

A3000/AA3000's Fat Gary doesn't have Gayle's PCMCIA and IDE.

Amber is not needed for A500-type cost reduction.

Quote:

They had visions of it being used as a Unix workstation, but this was not the machine the market needed. They already had the A2000 (which the Video Toaster actually fitted into!), and the A2630 already had a 32 bit expansion socket onboard which they could have added a SCSI controller and graphics card to (just like GVP did with their G-Force 030). Or they could have gone 'radical' by using the bridgeboard slots as a 32 bit extension connector like VL bus did. Instead they gave us a crippled Zorro-III with less then 1/10th of the theoretical bandwidth, and a flaky DMA controller.

Built-in DMA SCSI on the Ramsey 32-bit bus worked on A3000 and AA3000+.

Bridgeboard ISA slots weren't wired with the Amiga side.

Quote:

On the contrary, they weren't milking it enough. They stopped production of the best-selling A500 too early because they thought the A1200 would be ready by then.

You have forgotten Commodore Germany demanded a hard disk-capable Amiga model during A300's R&D.

Read Commodore the Inside Story, The Untold Tale of a Computer Giant by David John Pleasance.


Quote:

They did everything they could to keep the price down so customers would get a good deal - rather than being 'milked' like the PC industry was doing, forcing people to upgrade every 2 years in order to have a powerful enough machine to run the latest games etc.

Are you saying Quake or Doom should run on a stock A1200-level hardware?

Are you saying Lightwave 3D should run on stock A4000/040 and ignore higher performance competition?

Is "A1200" your "640K is enough for everybody"?

"Forcing people to upgrade every 2 years" is nonsense. For students with a smaller budget, a gaming PC only needs to match its major game console counterpart.

Highest resolution, highest detail settings, and highest frame rate are just for "PC Master Race" club.

Quote:

I dispute that. But the A500 was generally bought in preference to a game console because it was a lot more. Parents bought one because they thought it would educate the kids - which in many cases it did.

Which parents in which country?

From https://www.intel.fr/content/dam/doc/report/history-1994-annual-report.pdf
Intel reported the following
1. In 1994's fourth quarter, Pentium unit sales accounted for 23 percent of Intel's desktop processor volume.
2. Millions of Pentiums were shipped.
3. During Q4 1993 and 1994, a typical PC purchase was a computer featuring the Intel 486 chip.
4. Net 1994 revenue reached $11.5 billion.
5. Net 1993 revenue reached $8.7 billion.
6. Growing demand and production for Intel 486 resulted in a sharp decline in sales for Intel 386 from 1992 to 1993.
7. Sales of the Intel 486 family comprised the majority of Intel's revenue during 1992, 1993, and 1994.
8. Intel reached its 6 to 7 million Pentiums shipped goal during 1994. This is only 23 percent unit volume.
9. More than half of total sales are from the USA.

By the end of 1994, Intel's Pentium PC install base crushed the entire Amiga install base of 4 to 5 million units!

Quote:

Hobbyists bought them because they wanted to do more than play mindless arcade games.

Can your Hobbyist target market sustain Commodore?


Quote:

Those who were only interested in games bought it for the huge library of 'free' software available (total cost of ownership much cheaper than an SNES!).

They purchased "Doom" PC and pirated Doom and Doom 2 with more than 15 million estimated copies. PC Doom and Doom 2 have 4 million unit sales.

https://www.theguardian.com/technology/blog/2007/jul/18/winningthrough
Winning through piracy -- Microsoft in China

https://www.linkedin.com/pulse/fascinating-reason-why-bill-gates-allowed-pirated-spread-rajvanshi
Reason Why Bill Gates Allowed Pirated Windows to Spread in India.


https://www.latimes.com/archives/la-xpm-2006-apr-09-fi-micropiracy9-story.html
Bill Gates:
“Although about 3 million computers get sold every year in China, people don’t pay for the software. Someday they will, though,� Gates told an audience at the University of Washington. “And as long as they’re going to steal it, we want them to steal ours. They’ll get sort of addicted, and then we’ll somehow figure out how to collect sometime in the next decade.�


Did you think PC gamers are different from Amiga gamers?


Fact: SNES reached a 60 million install base and there are game rentals. SNES's strong 2D game experience is a major factor.

Last edited by Hammer on 22-Apr-2024 at 02:41 PM.
Last edited by Hammer on 22-Apr-2024 at 02:27 PM.
Last edited by Hammer on 22-Apr-2024 at 02:19 PM.
Last edited by Hammer on 22-Apr-2024 at 02:13 PM.
Last edited by Hammer on 22-Apr-2024 at 02:06 PM.
Last edited by Hammer on 22-Apr-2024 at 01:54 PM.
Last edited by Hammer on 22-Apr-2024 at 01:52 PM.
Last edited by Hammer on 22-Apr-2024 at 01:51 PM.
Last edited by Hammer on 22-Apr-2024 at 01:44 PM.
Last edited by Hammer on 22-Apr-2024 at 01:41 PM.
Last edited by Hammer on 22-Apr-2024 at 01:39 PM.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7950X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

@Hammer

Quote:

Hammer wrote: @bhabbott Quote: ECS was new for the A3000. Amber was part of the graphics chipset in the A3000. You just contradicted Commodore engineers i.e. Mike Sinz and Dave Haynie.

That doesn't make me wrong. Or Mike Sinz either for that matter.

According to Brian Bagnall in 'Commodore the Final Years', the A3000 was conceived in 1986. He says:-
Quote:

When the chipset entered redesign, they missed debuting the A3000 at the 1988 CeBIT. The chipset belatedly went into production in May 1988. Finally, in late 1988, it looked like CSG would deliver chip samples so the engineers could begin designing the A3000 prototype. ECS Chip Samples Although the new chipset had been called the Hi-Res chipset for years, marketing redubbed it ECS when it went into production in 1988... Finally, in September 1988, the fist chip sample came out... The QA group uncovered problems with the sprites and genlock transparency, among other issues.

So when design of the A3000 prototype began they already had the ECS chipset, albeit not in a fully working state. I guess that's what Mike Sinz meant by "the chipset did not change so that part was not "new"". Nevertheless it was a new chipset compared to current models (A1000/500/2000).

When the A3000 was designed there was no chipset more advanced than ECS anywhere near to being produced. The A3000 was already late, and they couldn't wait (possibly for years) for AA or AAA to materialize. Ranger was just ECS with VRAM and (possibly) a 1024x1024 monochrome screen mode that would require a special monitor (the ECS chipset actually has registers in it that appear to be for supporting VRAM, but AFAIK this functionality has never been tested).

Quote:

Amber was also found in the A2320 add-on card. This is not unique to the A3000.

Amber was designed specifically for the A3000. After the A3000 was released they designed a plugin flicker fixer card for the A2000 that used the same chip.

Commodore also allowed 3rd parties to sell the A3000's 2MB 8372B Agnus as an upgrade for the A500 and A2000. This required a daughterboard to get 2MB ChipRAM. At the same time they developed the 8272A, which was a drop-in replacement for the 8371 that converted 512k ChipRAM + 512k SlowRAM to 1MB ChipRAM. This 1MB Agnus was used in the CDTV, and A500/A2000 revision 6 motherboards.

Quote:

You created an argument based on product releases, not on the production prototype.

Wrong, see above. ECS was a new chipset first used in the A3000 prototype.

Quote:

AAA has been in development since 1989. http://www.bambi-amiga.co.uk/amigahistory/leweggebrecht.html

But in 1989 when the A3000 was designed, AAA was nowhere near ready for use. The initial specification for the 4 chip set was completed in February 1989. The date set for for full production was September 1, 1990. However given the length of time it typically took for Commodore to produce fully working chips this schedule was quite optimistic, even with 8 engineers designing it. Furthermore it required a 1.2 micron process, and CSG was already having problems at 2 microns.

The A3000 was released in June 1990. Even in the most optimistic scenario it couldn't have had AAA until at least September 1990, and realistically not before 1991. By that time they already had a partly working AA chipset. If they had concentrated on the lower-spec AA instead and forgotten about the A3000, CDTV and C65 we could have had AGA machines in late 1990 or early 1991. But the engineers had their sights aimed higher - too high.

Quote:

1987= 7 bitplane for 128 colors and 2 MB Chip VRAM. R&D was canceled and the proven original Amiga team was dismantled.

This is not true. R&D was not cancelled. In fact the R&D budget and staffing was increased. Jay Miner later denied that 'Ranger' (which was really just ECS with VRAM) had 7 bitplanes or 128 colors. He said it was 'finished', but that was probably just the specification. No chip design or silicon for this 'Ranger' chipset has ever surfaced.

Quote:

1989, the moonshot AAA R&D started. Only 1 year of serious development for AAA. AA was a panic move.

AA was proposed by George Robbins on September 19, 1989, when it became clear that the AAA timeline would not be met. The C65 was another factor. If they released that then low-end Amigas would also need 256 colors to better it. But there was no 'panic' until late 1991 when the machine Gould told us would be out that fall failed to materialize.

Quote:

3DO is true color (24-bit) and high color (16-bit) capable and quadrilateral 3D accelerator @ 25Mhz. Commodore's infamous 25 Mhz speed issue still exists with the Madam graphics chip

.
3DO was an over-priced proprietary game console. Not at all interesting to those of us who wanted a computer, and irrelevant to the Amiga.

Quote:

AA3000+ was later cloned by the Amiga community. C65 was later cloned by the C64 community.

No, they weren't. A reproduction of the A3000+ motherboard has been created. It requires original Amiga chips to complete. The Mega65 emulates a C65 in its FPGA, with extra capabilities that the C65 never had. The motherboard also includes various features that the C65 didn't have too. The case is a reproduction of the C65, but the insides are very different.

But even if the A3000+ and C65 really were cloned, doing it 30 years later is totally irrelevant to this discussion - unless you think Commodore should have built a time machine and pinched technology from the future to use in the Amiga.

@bhabbott
Quote:

According to Dr. Peter Kittel, who worked for Commodore Germany until the end, a total of 5,292,200 Amigas were sold world wide. In Germany a total of 1,081,000 A500s were sold, compared to 284,300 C128s

There was a great investigation piece from Ahoy which put's it around 4.91M

Nobody Knows How Many Amigas Commodore Sold
Ahoy

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