ppcamiga1 the gift that keep on giving!

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

@ppcamiga1

Quote:

ppcamiga1 wrote: DOOM on 14 MHz 68020 with FAST RAM is playable in low detail. on amiga with gfx card that has chunky pixel. It was AGA that made Commodore bankrupt in 1994.

For playing Doom, A1200 with Fast RAM is like my IBM PS/2 Model 55SX (386SX-16 with 16-bit front side bus, 5 MB System RAM, on-board IBM VGA) level.

A1200's AGA can still scale to 50 fps 320x200p 256 colors Quake or Star Wars Dark Forces
with increased CPU power.

The major problem is mass production with CPU accelerated Amiga.

A1200 with 32-bit Fast RAM is comparable to 386SX-16-based PCs.

Stock A1200 couldn't run Doom due to 2 MB RAM.

_________________
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)

@agami

Quote:

agami wrote: @Hammer Quote: Hammer wrote: @ppcamiga1 Chunky graphics alone wouldn't solve 68EC020 @ 14 Mhz being gimped by Chip RAM. In his mind it would've. I assume it goes a little something like this: - Commodore's AGA does 640x480 at 256 colours with native chunky, instead of what we got - John Carmack continues dev of Doom on 68k and releases it first on AGA Amiga's - This produces a boon for sales of 030 accelerator cards for A1200's, and the demand for 030 A4000s drives the price down to 486 PC levels - CD32 + Doom bundle becomes the No. 1 holiday season gift request - 1994 bankruptcy avoided So in another way it's not the chunky pixels, it's John Carmack's December 1993 Doom that is the "Kingmaker" in the computer gaming saga. I suppose in his simple mind it wasn't Steve Jobs or the NeXTSTEP-based OS X which ultimately saved Apple. It was John Carmack announcing Doom III on Mac OS X at MacWorld Tokyo Expo 2001.

Amiga's 1st and 3rd party 68030 CPU solution wasn't price vs performance competitive.

For Q4 1993, Commodore didn't match Apple's under $1000 Quadra 605 with 68LC040 @ 25 Mhz offerings and 486SX-25 PC clones.

Last edited by Hammer on 10-Dec-2023 at 02:32 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)

@cdimauro
Quote:

It was nice to see, but the frame rate wasn't smooth.

It's good enough for 640x480 RTS and isometric projection RPG DirectDraw games, but OEM S3 Trio 64UV+ PCI was cheap in 1996.

My H2 1993 Gateway 486SX PC examples have VLB SVGA cards.

For A4000/030 or A1200/030 @ 50Mhz money, Gateway's 486SX-33 wins cost vs performance. Apple's Quadra 605 68LC040 CPU power vs cost is competitive.

The general marketing in 1993 from PC clones and Apple has 4th gen CPUs e.g. 486SX or 68LC040.

A4000/040's asking price is a ripoff i.e. it's either a Pentium or RISC workstation CPU. Amiga 4000/Video Toaster 4000 market was tiny.

Quote:

Those weren't available on 1990: don't mention them, they are out of context.

My point is AGA is sufficient for displaying rendered frame buffers at 320x200p 50 fps.
3D composition is a different topic.

Example https://www.youtube.com/watch?v=b0wBXnDOcoc

The CPU should be greater than 68030 @ 32Mhz. Better with Apple-style Quadra 605's 68LC040 CPU mass production.

Alternative, Amiga AA3000+ was spec'ed with 68040 killer AT&D DSP3210 on Ramsey 32-bit Fast RAM bus.

https://www.landley.net/history/mirror/commodore/haynie.html
Dave Haynie is aware that the Amiga AA3000+ needs a math compute power boost.

DSP3210 supports modern PC gaming's FP32 data format and single-cycle instructions for up to 33 MFLOPS.

My PC from 1992-to-1994 was 386DX-33 with ET4000AX. It started showing its age in 1995 and against Sony's PSX, hence I planned for the Pentium PC build in 1996 i.e. CPU+motherboard+RAM+PCI SVGA card upgrade.

PSX's western market release was in Q4 1995 and both Sony PSX and gaming (Pentium) PCs need to be 1995 to 1996 hardware level. Commodore would need Amiga Hombre in 1995.

Quote:

AGA would have been already too little too late even on 1990 with its crappy Blitter: that's the major bottleneck of the Amiga platform, since ECS times.

AGA Blitter's importance is less when there's sufficient CPU power.

Fixed function Blitter's purpose is to shift bottleneck from CPU to memory bandwidth since 68000 CPU doesn't have RISC 1 instructions per clock performance.

68020/68030 has a hardware barrel shifter, but Chip RAM is the bottleneck.

PC VLB or PCI graphics chipsets can brute force this 2D problem on top of DirectDraw 2D acceleration.

In most cases from 1990 to 1995, PC DOS games needed CPU power and a fast VGA framebuffer.

Quote:

2D games REQUIRED a faster Blitter for moving more stuff and/or handling more colours.

It depends on the installed CPU power.

Fast Blitter coupled with TI486DLC 40Mhz CPU helps with a single aspect.

StarCraft was released in 1998 and I have an infamous Celeron 300A (overclocked to 450 Mhz with 100 Mhz front side bus) gaming PC build. My university was upgrading towards Pentium II and Celeron workstations.

Quote:

This was never the case because Commodore engineers had no idea at all about what was really needed for Amiga games...

Commodore International CEO insulting major 3rd party developers wasn't good.

Last edited by Hammer on 10-Dec-2023 at 11:20 AM.
Last edited by Hammer on 10-Dec-2023 at 11:09 AM.
Last edited by Hammer on 10-Dec-2023 at 11:06 AM.
Last edited by Hammer on 10-Dec-2023 at 11:01 AM.
Last edited by Hammer on 10-Dec-2023 at 10:46 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)

@Hammer

Quote:

Hammer wrote: @cdimauro Quote: It was nice to see, but the frame rate wasn't smooth. It's good enough for 640x480 RTS and isometric projection RPG DirectDraw games, but OEM S3 Trio 64UV+ PCI was cheap in 1996. My H2 1993 Gateway 486SX PC examples have VLB SVGA cards. For A4000/030 or A1200/030 @ 50Mhz money, Gateway's 486SX-33 wins cost vs performance. Apple's Quadra 605 68LC040 CPU power vs cost is competitive. The general marketing in 1993 from PC clones and Apple has 4th gen CPUs e.g. 486SX or 68LC040. A4000/040's asking price is a ripoff i.e. it's either a Pentium or RISC workstation CPU. Amiga 4000/Video Toaster 4000 market was tiny.

Again, the context was 1990...
Quote:

Quote: Those weren't available on 1990: don't mention them, they are out of context. My point is AGA

But the context was/is ECS & 1990.
Quote:

is sufficient for displaying rendered frame buffers at 320x200p 50 fps. 3D composition is a different topic. Example https://www.youtube.com/watch?v=b0wBXnDOcoc

It doesn't look running at 50FPS.
Quote:

The CPU should be greater than 68030 @ 32Mhz. Better with Apple-style Quadra 605's 68LC040 CPU mass production. Alternative, Amiga AA3000+ was spec'ed with 68040 killer AT&D DSP3210 on Ramsey 32-bit Fast RAM bus. https://www.landley.net/history/mirror/commodore/haynie.html Dave Haynie is aware that the Amiga AA3000+ needs a math compute power boost. DSP3210 supports modern PC gaming's FP32 data format and single-cycle instructions for up to 33 MFLOPS. My PC from 1992-to-1994 was 386DX-33 with ET4000AX. It started showing its age in 1995 and against Sony's PSX, hence I planned for the Pentium PC build in 1996 i.e. CPU+motherboard+RAM+PCI SVGA card upgrade. PSX's western market release was in Q4 1995 and both Sony PSX and gaming (Pentium) PCs need to be 1995 to 1996 hardware level. Commodore would need Amiga Hombre in 1995.

Again, see above: the context was/is 1990.
Quote:

Quote: AGA would have been already too little too late even on 1990 with its crappy Blitter: that's the major bottleneck of the Amiga platform, since ECS times. AGA Blitter's importance is less when there's sufficient CPU power. Fixed function Blitter's purpose is to shift bottleneck from CPU to memory bandwidth since 68000 CPU doesn't have RISC 1 instructions per clock performance. 68020/68030 has a hardware barrel shifter, but Chip RAM is the bottleneck.

In fact you cannot resolve all chipset issues / bottlenecks by just using a faster CPU.
Quote:

PC VLB or PCI graphics chipsets can brute force this 2D problem on top of DirectDraw 2D acceleration.

VLB, PCI and DirectDraw are out of context (1990).
Quote:

In most cases from 1990 to 1995, PC DOS games needed CPU power and a fast VGA framebuffer.

Correct.
Quote:

Quote: 2D games REQUIRED a faster Blitter for moving more stuff and/or handling more colours. It depends on the installed CPU power. Fast Blitter coupled with TI486DLC 40Mhz CPU helps with a single aspect.

Again, the context: ECS & 1990. Low-end machines -> 68000.
Quote:

StarCraft was released in 1998 and I have an infamous Celeron 300A (overclocked to 450 Mhz with 100 Mhz front side bus) gaming PC build. My university was upgrading towards Pentium II and Celeron workstations.

Again, out of context.
Quote:

This was never the case because Commodore engineers had no idea at all about what was really needed for Amiga games...

Commodore International CEO insulting major 3rd party developers wasn't good.[/quote]
Managers and engineers had their own sins.

@Hammer

Quote:

Hammer wrote: @agami Amiga's 1st and 3rd party 68030 CPU solution wasn't price vs performance competitive. For Q4 1993, Commodore didn't match Apple's under $1000 Quadra 605 with 68LC040 @ 25 Mhz offerings and 486SX-25 PC clones.

In our reality, yes.

But in his version of the alternate reality, where AGA comes with native chunky gfx capable of 8-bit 640 x 480 instead of what we got, both of your above points would be sufficiently different to avoid the 1994 bankruptcy.

Last edited by agami on 10-Dec-2023 at 10:09 PM.

_________________
All the way, with 68k

@cdimauro

Quote:

Again, the context was 1990...

FYI, Intel 486 and ET4000AX existed in 1989 which led to the 1993 situation.

IBM's 256 display colors existed with the 1985 IBM PGA with a 4096 color palette.
IBM recycled their 256-color display intellectual property for 1987 VGA. IBM improved PGA into VGA and created the 8514 foundation for XGA.

SVGA foundations during 1993 can be found in 1987's IBM's VGA and 8514. PC cloners have cost-reduced IBM's 1987 "Amiga 1000" moments and improved upon them i.e. faster frame buffer performance.

Commodore's Amiga 1000 OCS provided a 4096 color palette with 32 colors, 64 colors EHB, and compressed pixel 4096 colors HAM mode that is lower priced than IBM PGA.

The PC has multiple "Amiga 1000" moments.

For AGA to be released in 1990, the completion work should be completed in 1989.

AA3000+ was completed in Feb 1991 and already has ECS time wasting from Commodore management. Commodore management wasted time during the A3000's development.

Both the 1987 Amiga Ranger (128 colors display, 4096 color palette, high-performance VRAM) and A3000 have 2 MB Chip RAM. There was an R&D stall around 1987.

1987 Amiga Ranger was supposed to compete against Sharp X68000 and Apple Macintosh II.

Apple's 1987 Macintosh II.
Graphics: The Macintosh II includes a graphics card that supports a true-color 16.7 million color palette[14] and was available in two configurations: 4-bit and 8-bit. The 4-bit model supports 16 colors on a 640×480 display and 256 colors (8-bit video) on a 512×384 display, which means that VRAM was 256 KB. The 8-bit model supports 256-color video on a 640×480 display, which means that VRAM was 512 KB in size. With an optional RAM upgrade (requiring 120 ns DIP chips), the 4-bit version supports 640×480 in 8-bit color.[15] The video card does not include hardware acceleration of drawing operations.

AGA paper spec display capability rivals the 1987 Macintosh II.

Commodore's failure was set around 1987-1989 when new chipset R&D stalled.

The What IF change is removing the R&D stall in 1987.

Last edited by Hammer on 11-Dec-2023 at 04:00 AM.
Last edited by Hammer on 11-Dec-2023 at 03:56 AM.
Last edited by Hammer on 11-Dec-2023 at 03:52 AM.
Last edited by Hammer on 11-Dec-2023 at 03:48 AM.
Last edited by Hammer on 11-Dec-2023 at 03:45 AM.
Last edited by Hammer on 11-Dec-2023 at 03:33 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)

@agami

Quote:

agami wrote: @Hammer Quote: Hammer wrote: @agami Amiga's 1st and 3rd party 68030 CPU solution wasn't price vs performance competitive. For Q4 1993, Commodore didn't match Apple's under $1000 Quadra 605 with 68LC040 @ 25 Mhz offerings and 486SX-25 PC clones. In our reality, yes. But in his version of the alternate reality, where AGA comes with native chunky gfx capable of 8-bit 640 x 480 instead of what we got, both of your above points would be sufficiently different to avoid the 1994 bankruptcy.

Avoid 1994 bankruptcy perhaps, but not much more. 3rd quarter 1993 would be too late. The Quadra 605 is a red herring. Firstly it was useless without the companion monitor, which wasn't cheap. And then it only ran Mac software. That model only sold for 1 year - I wonder why?

Chunky graphics in 640x480 256 colors is not much use by itself. You also need a fast CPU and faster blitter etc. This is no simple upgrade to OCS, and the machine would be expensive. But people wouldn't buy it at the same or higher price as a PC, which makes competing head-to-head with PCs very difficult when clone makers were churning out stuff cheaper than any name-brand manufacturer could. The only reason Commodore could make the A500 so cheap was having their own chipset in a minimalist design.

640x480 8 bit was overkill. 320x200/256 8 bit was the sweet spot for games - AGA would have been OK and perfect with chunky pixels at that resolution. But it needed to work at 15kHz so people could use their TV or 1084 monitor. It still wouldn't be much use without a faster CPU though. And why do you need it anyway? Only to make porting PC games easier. That means the Amiga would still be playing catchup while not being IBM compatible.

For the Amiga to stay relevant it had to continue carving a niche between consoles and PCs. The A1200 did that - it was just a little too late.

Hindsight is always 20/20 as they say, but we can speculate on what might have been if Commdore's priorities were different. They tried to compete against the PC with high-end Amigas and their own PCs. It didn't turn out well. Unfortunately the entire PC industry suffered a downturn in 1990/91, just at the wrong time for Commodore. In the long term the PC would win, but 1991 was the Amiga 500's best year!

Had Commodore pushed out the AGA chipst in 1990/91, in a 'low-end' machine to attract hobbyists and gamers who couldn't afford a PC, it might have revitalized the Amiga gaming scene in time to prevent a 1994 bankruptcy. And had they avoided disasters like the A3000 and CDTV they might have had enough resources to do it. But there were different factions in Commodore pulling in different directions. Engineers had their pet projects, and a disdain for management that caused them to not design what the market wanted.

I recently purchased the book "Commodore the Final Years" by Brian Bagnall. He interviewed many Commodore insiders to get a more detailed and accurate picture of what went on at Commodore and why. Here are some extracts from the chapter 'The Final Takeover 1991':-
Quote:

"When Bill Sydnes had been hired, one of his tasks was to figure out why the Amiga chipset development was always falling behind schedule. Dave Haynie felt the reason was simple, "... they weren't funding it at the level they should heve been". However, with 15 people working 100% on the AAA chipset alone, it seemed like Commodore had been willing to invest a lot into the project... One of the key problems of the AAA chipset was the breadth of the spec, which required it to work with a variety of disparate computer systems... the high-end Unix workstation using VRAM, high-end multimedia Amiga computers using fast DRAM, and low-cost Amiga 500 level systems using DRAM. It was also designed to work with Reduced Instruction Set Computing (RISC) processors, 16-bit and 32-bit Motorola 680xx chips. It was designed for backward compatibility... It was also designed to produce 8-bit, 16-bit and 32-bit... chunky modes. Sydnes sought proposals from his engineers for the short, medium, and long term... the medium term proposal, written primarily by Lew Eggebrecht, was... an Enhanced Amiga Architecture to replace AAA. Eggebrecht's proposal was meet with disdain. Hepler, Redfaild, Hynie and Davis all wrote responses largely deriding the naivety of the author... Regardless, they would continue trying to fix the problematic development of the Pandora and AAA chipsets. AA First Prototypes Back in September 1989, George Robbins proposed an intermediate level Amiga chipset, called Pandora, to bridge the gap until AAA appeared... consisting of a graphics chipset called Lisa and an improved Agnus called Alice, was supposed to be ready in early 1990.... six chip engineers had been working full time under the project leader, Bob Raible. As work progressed on the graphics chip, Lisa, it soon became clear that the timeline was too ambitious. Jeff Porter set a more realistic expectation of late 1990. CSG produced the first prototype chips by late November 1990 and testing began... Dave Haynie managed to boot up his A3000 Plus with Amiga OS and AA chipset in February 1991. In an interview with Amiga Computing magazine, Irving Gould revealed, "As a matter of fact, there is a new chip set for the Amiga that we've been working on now, I guess, for almost two years and should be ready this fall". The reason Gould was unsure of the release date was that the engineers were considering some improvements to the chipset. By March 27, when the tested AA chipsets were ready the list of stable features was impressive... On Thursday, March 28 at 9 AM, Jeff Porter, Lew Eggebrecht, Greg Berlin Dave Haynie, George Robbins and Hedly Davis met to decide what they wanted to do with the AA chipset spec.... This change to the spec would cause the chipset to slip behind schedule and Porter estimated that AA systems such as the A1000 plus would be pushed back to April 1992.

The reality is that despite having several engineers on the job, it took a long time to get AGA out. Had the AA guys been working on it too they could have released it in 1991, before A500 sales waned. But the engineers wanted their high-end machines so it was not to be. Even an April 1992 launch would been good (A1200 instead of A600) but that wasn't to be either. The extreme disappointment of Amiga fans of seeing the A600 when Gould promised so much more was too much for many.

That's what happens when everyone wants their pet 'high-end' features and won't settle for something more basic to sell now. The more complex and high performance the design, the harder it is to turn it into reality. We can dream about what AGA could have had, but I doubt that any of us have the skills to actually do it (especially with the technology of that time).

Last edited by bhabbott on 11-Dec-2023 at 04:58 AM.

@Hammer

Quote:

Hammer wrote: @cdimauro Quote: Again, the context was 1990... FYI, Intel 486 and ET4000AX existed in 1989 which led to the 1993 situation.

Nobody had the crystal ball to see on 1989 and 1990 what could have happened on 1993.

Please stop this incorrect representation of the facts, because it's historically inaccurate.
Quote:

IBM's 256 display colors existed with the 1985 IBM PGA with a 4096 color palette. IBM recycled their 256-color display intellectual property for 1987 VGA. IBM improved PGA into VGA and created the 8514 foundation for XGA. SVGA foundations during 1993 can be found in 1987's IBM's VGA and 8514. PC cloners have cost-reduced IBM's 1987 "Amiga 1000" moments and improved upon them i.e. faster frame buffer performance. Commodore's Amiga 1000 OCS provided a 4096 color palette with 32 colors, 64 colors EHB, and compressed pixel 4096 colors HAM mode that is lower priced than IBM PGA. The PC has multiple "Amiga 1000" moments. For AGA to be released in 1990, the completion work should be completed in 1989. AA3000+ was completed in Feb 1991 and already has ECS time wasting from Commodore management. Commodore management wasted time during the A3000's development. Both the 1987 Amiga Ranger (128 colors display, 4096 color palette, high-performance VRAM) and A3000 have 2 MB Chip RAM. There was an R&D stall around 1987. 1987 Amiga Ranger was supposed to compete against Sharp X68000 and Apple Macintosh II. Apple's 1987 Macintosh II. Graphics: The Macintosh II includes a graphics card that supports a true-color 16.7 million color palette[14] and was available in two configurations: 4-bit and 8-bit. The 4-bit model supports 16 colors on a 640×480 display and 256 colors (8-bit video) on a 512×384 display, which means that VRAM was 256 KB. The 8-bit model supports 256-color video on a 640×480 display, which means that VRAM was 512 KB in size. With an optional RAM upgrade (requiring 120 ns DIP chips), the 4-bit version supports 640×480 in 8-bit color.[15] The video card does not include hardware acceleration of drawing operations. AGA paper spec display capability rivals the 1987 Macintosh II. Commodore's failure was set around 1987-1989 when new chipset R&D stalled. The What IF change is removing the R&D stall in 1987.

That's totally irrelevant regarding the discussion / context.

FACT is: Amiga was wide spread due to its LOW-END, GAMING machines.

That should have been the PRIMARY focus of Commodore (managers AND engineers).

You can ALSO think about providing MID/HIGH-end systems IF YOU CAN, but the focus should have been the above low-end AND, specifically, the gaming market.

For this reason, a 486 with a phenomenal SVGA, a Mac-II and even the Sharp X68000 are completely irrelevant as a reference for such market.

This is particularly true looking also at the console: they all were low-cost and 2D games were their target. Yes, there were also polygonal 3D games, but they were not their focus.

Is it clear now? So, let's stick to THIS context.

@bhabbott: AGA was too little even on 1990.

Damn, do you understand how important was/is the Blitter on the Amiga, or not?!?

@cdimauro

Quote:

FACT is: Amiga was wide spread due to its LOW-END, GAMING machines. That should have been the PRIMARY focus of Commodore (managers AND engineers). You can ALSO think about providing MID/HIGH-end systems IF YOU CAN, but the focus should have been the above low-end AND, specifically, the gaming market. For this reason, a 486 with a phenomenal SVGA, a Mac-II and even the Sharp X68000 are completely irrelevant as a reference for such market.

While I get what you're aiming at, there's some strength in having an halo product, like nvidia does with their *090 tier line. Not to say that commodore had it in mind though, just saying...

Last edited by pixie on 11-Dec-2023 at 06:18 AM.

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

doom works well on atari falcon little faster than amiga 1200.
so doom in low detail is possible on amiga 1200 with fast ram.
commodore bankrupt because aga has not chunky pixels.

@pixie

Quote:

pixie wrote: @cdimauro Quote: FACT is: Amiga was wide spread due to its LOW-END, GAMING machines. That should have been the PRIMARY focus of Commodore (managers AND engineers). You can ALSO think about providing MID/HIGH-end systems IF YOU CAN, but the focus should have been the above low-end AND, specifically, the gaming market. For this reason, a 486 with a phenomenal SVGA, a Mac-II and even the Sharp X68000 are completely irrelevant as a reference for such market. While I get what you're aiming at, there's some strength in having an halo product, like nvidia does with their *090 tier line. Not to say that commodore had it in mind though, just saying...

np with that, as long as you don't miss the primary target / market.

@bhabbott

Quote:

640x480 8 bit was overkill. 320x200/256 8 bit was the sweet spot for games.

Nah.. it was to slow to use 640x480 8 bit mode, have you ever used PAL productivity mode… horrible slow.. 640x480 interlaced mode was half bad, but flickering was too annoying. Amiga was target at TV not a normal monitor, so had use a scan doubler to even use a decent monitor. Easy to forget that with amount of Rasbary PI’s you put in your Amiga Computers today, when you have overcome all kinds of limitation. One for Denice, and for CPU, It’s amazing that need 4 cores for simple computer like that.

Last edited by NutsAboutAmiga on 11-Dec-2023 at 08:00 PM.

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

@NutsAboutAmiga

I always thought productivity mode to be as slow as interlaced, after all they have around the same pixels,but I never tested it on a real Amiga, in WinUAE even 1280x1024 can be made to be bearable.
Quote:

Easy to forget that with amount of Rasbary PI’s you put in your Amiga Computers today, when you have overcome all kinds of limitation. One for Denice, and for CPU, It’s amazing that need 4 cores for simple computer like that.

When you say it need x for y, it's not like it emulates 68000 1:1 ratio now is it, it goes far and beyond.. why do you say it needs 4 cores, and for Denise??

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

@NutsAboutAmiga

Regarding Pistorm, the information I have is that:
Quote:

On the pistorm, the 2nd core is used as "housekeeper" does interrupt sync stuff, Ctrl+A+A handling... The 3rd core is used for debugging, the 4th is currently free.

One might there's no much going on besides the first one.

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

@cdimauro

Quote:

FACT is: Amiga was wide spread due to its LOW-END, GAMING machines. That should have been the PRIMARY focus of Commodore (managers AND engineers).

AFAIK it was, but it’s also part of the story that the game thing mostly was a European thing (I would even say mostly a UK thing) and not just for Amiga.

Quote:

You can ALSO think about providing MID/HIGH-end systems IF YOU CAN, but the focus should have been the above low-end AND, specifically, the gaming market.

In period of 1985 to 1994, what would be your system to develop Amiga games (and for that matter, other software) on?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@kolla

Quote:

kolla wrote: @cdimauro Quote: FACT is: Amiga was wide spread due to its LOW-END, GAMING machines. That should have been the PRIMARY focus of Commodore (managers AND engineers). AFAIK it was, but it’s also part of the story that the game thing mostly was a European thing (I would even say mostly a UK thing) and not just for Amiga.

Indeed (and it was also a success in Italy).
Quote:

Quote: You can ALSO think about providing MID/HIGH-end systems IF YOU CAN, but the focus should have been the above low-end AND, specifically, the gaming market. In period of 1985 to 1994, what would be your system to develop Amiga games (and for that matter, other software) on?

1985-1987: Amiga 1000
1987-1992: Amiga 500
1992-1994: Amiga 1200

Which is obvious, but... that's it.

@cdimauro

Quote:

Nobody had the crystal ball to see on 1989 and 1990 what could have happened on 1993. Please stop this incorrect representation of the facts, because it's historically inaccurate.

That's a false narrative. You didn't factor in Moore's law as the constant drumbeat pace.

PC VGA and PC chipset cloners followed Moore's law with highly integrated ASICs.

For the 1990 product, the A3000's chip integration and high chip count are inferior.

https://vintageapple.org/pcworld/pdf/PC_World_9011_November_1990.pdf
Page 354 of 364 for the US market
November 1990.

Full 32-bit 386 CPU @ 20Mhz PC with Phoenix BIOS, 1 MB RAM, PSU, 101 Keyboard, 1.44 FDD, MS-DOS 3.3, IDE controller, and IDE Seagate
The base system asking price is $999.

VGA card with 48 MB HDD, extra $850
SVGA card with 20 MB HDD extra $709
SVGA card with 48 MB HDD extra $1019

A low-cost VGA color monitor can be obtained.

Direct competition is the main driver for pushing PC's lower asking prices, hence the direction is set for the 1993 situation.

https://archive.org/details/eu_BYTE-1990-05_OCR/page/n209/mode/1up
Commodore Amiga 3000 review by Byte Magazine, May 1990

The reviewer correctly noted that the lack of native 8 and 24-bit graphics modes is a concern and that Commodore needed to address this to stay competitive.


Commodore's asking price for A3000 @ 16Mhz 68030 is $3495 and 25 Mhz version for $3995.
The mentioned Amiga 3000 has 2 MB memory (1 MB Fast, 1 MB Chip) and 50 MB SCSI HDD.

On the subject lack of native 8 and 24-bit graphics modes, Commodore's Amiga Product Manager Keith Masavage commented Commodore has both short-term and long-term solutions are the works. (read Byte Magazine's review).

Recall that the AA3000+ AGA chipset was completed in Feb 1991. Commodore management wasted time on ECS-based "A1000+ Jr".
----
Gaming PC 1990 Wing Commander VGA was the killer game app. My Dad noticed certain PC games in 1990 and made plans for a later 386DX-33 PC clone build in early 1992.

In 1992, My Dad obtained our ex-corporate A3000/030 (25Mhz, KS 2.04, 1MB Fast, 1 MB Chip, Amiga Vision) from corporate contacts which is not a standard retail path. A500 Rev 6 (1MB RAM) was sold at a similar time to help fund the ex-corporate A3000/030 purchase.

The signs for 1993 already exist in the 1990's PC price decline direction.

Parts of Commodore were already aware of the 256-color VGA problem in 1990! Commodore sales teams did NOT purchase "A1000 Jr".

The Australian market is largely attached to the US market.

Last edited by Hammer on 12-Dec-2023 at 01:48 AM.
Last edited by Hammer on 12-Dec-2023 at 01:47 AM.
Last edited by Hammer on 12-Dec-2023 at 01:45 AM.
Last edited by Hammer on 12-Dec-2023 at 01:44 AM.
Last edited by Hammer on 12-Dec-2023 at 01:37 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)

@bhabbott

Quote:

bhabbott wrote: @agami Avoid 1994 bankruptcy perhaps, but not much more.

Oh I agree.
Any action in the realm of our reality which would've resulted in the avoidance of the Commodore bankruptcy in 1994 would ultimately only be kicking the proverbial can just a little down the road.
But try and explain that to a person who lives in their own constructed reality.

Everything you've written is very reasonable. I was just trying to illustrate how @ppcamiga1's mind works with his:
10 Phase 1 - Collect underpants
20 Phase 2 - GOTO 30
30 Phase 3 - Profit
hindsight view of the world.

_________________
All the way, with 68k