@kolla

Useful does not equal Meaningful or Significant.

If you roamed the desert for three days with nothing to eat or drink, and then upon finding you I give you 150ml of water, I’m sure it’s not useless but how much impact would it really have?

That said, Significant isn’t always useful either. The Cell BE design was significant, but to this day we debate about just how useful it was.

_________________
All the way, with 68k

@kolla

Quote:

kolla wrote: @cdimauro Quote: cdimauro wrote: @kolla Maths? Math doesn’t determine what’s useful or not.

Considered that we're talking about the AMIGA market, I would say yes: it determines what's useful or nor for THAT market.
Quote:

ECS had small yet very useful enhancements over OCS. Just because YOU don’t find them useful doesn’t mean they weren’t.

Me and MILLION of amigans. Versus YOU and A BUNCH of Unix fans.

Do you see now why Maths matters?
Quote:

Quote: Good that you've said it: YOU! NOT the VAST MAJORITY of amigans which enjoyed OCS games or applications. “which”? Were they not people? Yes ME, and MANY others -

Many? How many?!?
Quote:

or are you suggesting it was I who implemented Linux and BSDs for Amiga? No, it wasn’t me. Many of tbe names of those who did, you also find in significant software for Amiga in those days.

Which is... IRRELEVANT compared to the THOUSANDS of GAMES, APPLICATIONS and DEMOS which were released for the Amiga and that have NOT used any ECS features.

Do you see the difference now?

BTW and as I've said, Linux, NetBSD and OpenBSD were release quite AFTER ECS. Whereas ECS should have targeted the needs of the AMIGA (NOT *Unix) market OF THE TIME.
Quote:

For those enjoying OCS games and especially applications, ECS was still an improvement.

Which games?
Quote:

ECS on the A600 made it possible for me to create animations with DPaint that I could not have made on OCS, the A600 was part of a setup for a neuro-science lab, quite useful as many of these animations are still in use there.

See above: how many of those use cases compared to all other where accessing more than 512kB of Chip Mem was NOT required?

On top of that, this required ONLY the Agnus version which supported 1 or 2MB od Chip Mem. So, no ECS Denise usage.
Quote:

Quote: How many of them enjoy ECS (specific!) games or applications? You already stated that they enjoy OCS games and applications, unless these don’t work at all on ECS, what’s the problem? A lot of software was capable of making use of ECS features while still maintaining OCS compatibility - was that a BAD thing?

First of all, which ECS features?

Second, and more important, I've NOT said that using 1 or 2 MB of Chip Mem wasn't useful. That was useful, for sure. However ALL other things were NOT useful for the Amiga market of time.

Third, it would have been much better to use the transistor budget to introduce the features that I've suggested. THIS was WAY MORE IMPORTANT for the AMIGA (A-MI-GA) market, because ALL games, applications and demos could have used them AND amigans enjoyed them.

Do you see the difference now? Because I don't know how to explain you in more simple words.
Quote:

Quote: Well, if you limit the audience to you, then it's perfectly fine. Not just me, I was and still is primarily a user, not a developer - I presume the software I user was developed by actual people and that I was not the only user of these software titles.

Count you and a bunch, but this does NOT change the great picture: your super niche is NOTHING compared to the normal/natural needs of the amigans.
Quote:

Quote: However I've to reveal a secret: YOU were a rare bird. No, I wasn’t. What’s rare is that I am still here.

Yes, and you're still the ONLY one which liked the ECS features instead of the ones that I've suggested. Guess why...
Quote:

Quote: I reveal you another secret: when ECS was fully (all chips) deployed Linux didn't even exist. NetBSD neither. And OpenBSD... I just leave the answer as homework... I am fully aware of this, of course, but that wasn’t the point, was it. Your claim is that ECS was useless, but it turned out to be quite useful for those who knew how to make us of the enhancements.

Again, Mr. obvious: true! You'll find some crazy people which could have used even the two UHRES bitplane pointers added to the ECS. Nobody is stating something different here.

However you should open your mind a little bit and see what were the NATURAL needs of the Amiga market of THAT TIME. Hence, what features could have much better have made an impact of those needs for the games, applications and demos. So, to satisfy the AMIGANS and NOT the Unix lovers or some other super niche.

Understood now?
Quote:

Quote: I reveal you the last secret: my "silly" (SIC!) games were the ones which made the Amiga the platform which we know and enjoyed. WITHOUT those games you would have had no ECS, AGA, whatever. Even applications. Really, you singlehanded made everything possible with your silly games? How humble. I had not even heard of any of your games until you brought up that fighter game I can’t even remember the name of.

Eh? You were really referring ONLY to my games before? Are you serious?!?

I was talking about the OVERALL games market: the ones which have made great the Amiga platform.

Anyway, Fightin' Spirit is one of best games and very well know to the amigans, many times used as a reference of what the Amiga (OCS!) made it possible.

If you didn't know it, well. You speak for yourself, dear Unix caveman...
Quote:

Quote: Maybe you need to learn something about Amiga's history: you lived on some Un*x cave 'til now... Says the gamer cave guy. I’m quite aware of Amiga history, not sure I see what you are trying to accomplish with your rantic articles though…

That's your problem, because you clearly have no vision of what happened and what was really needed for the AMIGA (not Unix or some other fancy niche) market of the time.

Let's see if after this reply you finally got it. But I have little hopes...

@agami

If its been useful to Kolla then its clearly going to be meaningful & significant to him or anyone else that found/finds it useful. Personally I couldn't handle being restricted to less than 1mb Chip Ram (even in 1992) & also used Productivity mode for a while, in all its 4 colour glory! Sure an alternative may have been even more useful but that doesn't make ECS not useful.

Few non AGA games required 1mb chip because of all the A500 OCS machines still in use. I guess it would be fair to say for games it wasn't very meaningful or significant; although potentially useful e.g. UROPA2 is one game that needs it & I find it very significant when it comes to WHDLoad these days .

_________________
A1200+Mediator+Voodoo3+040+130mbRAM+0S3.9
A2000+Supra28mhz+9mbRAM+OS3.2.2, CD32 & WinUAE

@Tpod: 1 or 2MB of Chip Mem was a useful addition. Granted. Expected. Ultra trivial to implement (just one or two bits added to the pointers!).

However how many games required it? And used it, of course.
I mean: NOT just for caching data from the floppies to reduce inserting / rejecting them.

@thread


I always said how ECS/OCS was outdated in the 90ies how AGA came to late and how bitplane oriented GFX was the wrong way to got when it came to AGA.

But a real new GFX would have needed a chipset beeing OCS/ECS compatible including a totally new GFX mode that would be more oriented to VGA than to ECS/OCS. With VGA I meant:
8 bit chunky up to 800x600 at 60Hz (or better)
No Copper
No dual playfield
No sprites
But this would be incompatible what most Amiga keyboard/wedge users had, it would have needed a real VGA monitor and would have been reserved to real desktop computers, something most wedge computer lovers would have ignored.

This is a real problem when it comes to Amiga, we have those 'Unix is stupid, SCSI was expensive, Tiga was not my goal, i wanted a game console with built in keyboard' lovers. Who would have been glad with something like a 256 color ECS-GFX and those who really wanted to use the Amiga for work and could afford a monitor beside their TV.

C= tried to solve this problem by splitting the system into a desktop and a console version, something that made the desktop weak and the consoles to expensive.

@OneTimer1

Quote:

OneTimer1 wrote: @thread I always said how ECS/OCS was outdated in the 90ies how AGA came to late and how bitplane oriented GFX was the wrong way to got when it came to AGA. But a real new GFX would have needed a chipset beeing OCS/ECS compatible including a totally new GFX mode that would be more oriented to VGA than to ECS/OCS. With VGA I meant: 8 bit chunky up to 800x600 at 60Hz (or better) No Copper No dual playfield No sprites But this would be incompatible what most Amiga keyboard/wedge users had, it would have needed a real VGA monitor and would have been reserved to real desktop computers, something most wedge computer lovers would have ignored.

I beg to differ. I've written another article some months ago about how to implement packed/chunky on AGA (but the same applies to ECS, if it was extended to 8 bitplanes) instead of the Akiko crap: Beyond Akiko: grafica packed al minor costo usando i bitplane

It's in Italian, but it could be easily translated in any language using DeepL (which is the most accurate tool for this, AFAIK).

The same concept / mechanism could be used for implementing hi-colour and true-colour modes.
Quote:

This is a real problem when it comes to Amiga, we have those 'Unix is stupid, SCSI was expensive, Tiga was not my goal, i wanted a game console with built in keyboard' lovers. Who would have been glad with something like a 256 color ECS-GFX and those who really wanted to use the Amiga for work and could afford a monitor beside their TV. C= tried to solve this problem by splitting the system into a desktop and a console version, something that made the desktop weak and the consoles to expensive.

The two things aren't mutually exclusive: adding 256 colours AND programmable resolutions would have been possible even on the same ECS.

IIF the problem was about the transistor budget (which is an excuse: the changes are really trivial), then the two completely useless UHRES bitplanes could have been NOT implemented AND their transistors used for adding the two missing bitplanes (plus a bunch of transistors for extending the EHB to 128 and 256 colors "ala Commodore 16". And another few transistors for enabling the Dual Playfield for 7 and 8 bitplanes by selecting "16" instead of "8" as the base palette index for the second screen).

What was missing was the vision by Commodore engineers: what were the REAL needs of the Amiga market.

@cdimauro

Why didn’t you just TELL them?

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

OneTimer1 Quote:

I always said how ECS/OCS was outdated in the 90ies how AGA came to late and how bitplane oriented GFX was the wrong way to got when it came to AGA.

OCS/ECS was still acceptable for the budget gaming market in the early 1990s considering how much software it had and was still being developed. The problem is that C= didn't have a high end chipset anymore as they had one size fits all chipset and it was budget only. I have suggested having 2 chipsets where the high end gets a more expensive chipset and the low end is upgraded to the high end chipset when it becomes cheap enough.

OneTimer1 Quote:

But a real new GFX would have needed a chipset beeing OCS/ECS compatible including a totally new GFX mode that would be more oriented to VGA than to ECS/OCS. With VGA I meant: 8 bit chunky up to 800x600 at 60Hz (or better) No Copper No dual playfield No sprites But this would be incompatible what most Amiga keyboard/wedge users had, it would have needed a real VGA monitor and would have been reserved to real desktop computers, something most wedge computer lovers would have ignored.

The Amiga can have full VGA compatible gfx modes while retaining the copper, sprites and dual playfield modes. The blitter would not work as is as it is designed for planar blitting. VGA modes would have been very useful mostly because they allow to use cheap and widely available VGA monitors. The reason the Amiga didn't have better VGA modes is that the chipset didn't have enough chip memory bandwidth to display them with enough colors. ECS made the Amiga display mostly programmable which was needed but without more chip memory bandwidth it didn't help much. Chunky modes are not a problem and were planned for the AA+ chipset upgrade.

https://en.wikipedia.org/wiki/AA+_Chipset

Most FPGA Amiga hardware also has chunky RTG support and it can be displayed in layers like playfields (usually the background playfield) with planar graphics, sprites and the copper (SAGA). Memory bandwidth is no longer a problem today but C= was unwilling to raise the cost of budget Amigas back then which they planned to reduce to C64 replacement levels. There were multiple possible ways for C= to increase the chip memory bandwidth but they refused anything that increased the cost.

OneTimer1 Quote:

This is a real problem when it comes to Amiga, we have those 'Unix is stupid, SCSI was expensive, Tiga was not my goal, i wanted a game console with built in keyboard' lovers. Who would have been glad with something like a 256 color ECS-GFX and those who really wanted to use the Amiga for work and could afford a monitor beside their TV. C= tried to solve this problem by splitting the system into a desktop and a console version, something that made the desktop weak and the consoles to expensive.

My perspective is that C= should have split the chipset into a gaming and desktop version. They could have at least allowed VRAM for the desktop as the AAA chipset could use VRAM or DRAM. Double the chip memory bandwidth could have allowed higher resolutions and more colors even without more bitplane pointers or color registers (double the colors in hires and VGA modes where there are enough bit plane pointers and color registers but not enough chip memory bandwidth). VRAM may have been more compatible than the AGA method of using memory bursts which introduced new bitmap alignment requirements. Two more bitplane pointers and quadruple the color registers could allow a 256 color EHB mode in lores without too many transistors (RJ Mical supposedly claimed the Ranger chipset quadrupled the number of color registered which infers at least one more bitplane register). C= needed to aggressively explore how to increase the chip memory bandwidth, especially for the high end market. Instead, they gave the same chipset to everyone that was borderline between budget and outdated in the early 1990s. This was tolerable for the budget gaming market but was devastating to the higher margin high end market. Just after the Amiga reputation for super graphics had translated into sales, they fell flat with practically nothing new to offer. C= could have provided gfx cards with VRAM in big box Amigas but integrated gfx are cheaper and should offer better performance.

@cdimauro

Quote:

cdimauro wrote: Beyond Akiko: grafica packed al minor costo usando i bitplane[/url]

There was no Akiko in ECS/OCS, AFAIK there was never an Akiko in any of the desktop models.

And the existence of Akiko in a late AGA game console doesn't tell anything about the intentions of the developers after ECS/OCS but it might tell us more about the intentions of the management refusing a clean cut for a 256 bit GFX and their desperate search for a cheap fix.

Last edited by OneTimer1 on 29-Oct-2023 at 10:24 PM.
Last edited by OneTimer1 on 29-Oct-2023 at 10:22 PM.

@kolla

Quote:

kolla wrote: @cdimauro Why didn’t you just TELL them?

Why should I?

Who was PAID for developing the Amiga platform, me or the Commodore engineers?

Who are considered "gods" by the Amiga fanboys, me or the Commodore engineers?

So, NOT my duties!


@matthey

Quote:

matthey wrote: OneTimer1 Quote: But a real new GFX would have needed a chipset beeing OCS/ECS compatible including a totally new GFX mode that would be more oriented to VGA than to ECS/OCS. With VGA I meant: 8 bit chunky up to 800x600 at 60Hz (or better) No Copper No dual playfield No sprites But this would be incompatible what most Amiga keyboard/wedge users had, it would have needed a real VGA monitor and would have been reserved to real desktop computers, something most wedge computer lovers would have ignored. The Amiga can have full VGA compatible gfx modes while retaining the copper, sprites and dual playfield modes.

Exactly!
Quote:

The blitter would not work as is as it is designed for planar blitting.

With some changes it can work with packed graphics as well. The only hard thing to implement is its fill mode (it can be also partially emulated in hardware, anyway).

I've written some articles (in Italian, sorry) about planar vs packed, showing how to implement the latter on the Amiga. Here's the last one: https://www.appuntidigitali.it/19768/amiga-packed-16-endgame/ which reports all published articles at the beginning. As it can be seen, it covers the Blitter as well.
Quote:

VGA modes would have been very useful mostly because they allow to use cheap and widely available VGA monitors. The reason the Amiga didn't have better VGA modes is that the chipset didn't have enough chip memory bandwidth to display them with enough colors. ECS made the Amiga display mostly programmable which was needed but without more chip memory bandwidth it didn't help much.

Doubling the Chip Mem bandwidth, as I've suggested before, allows to display 640x480@16 colors without interlace, putting it on par with the equivalent VGA. Reducing the colours to 4 then it would have been possible to reach 800x600@72Hz.

All this whilst remaining on the same ECS chipset transistors budget.
Quote:

Chunky modes are not a problem and were planned for the AA+ chipset upgrade. https://en.wikipedia.org/wiki/AA+_Chipset

It looks like that the 8-bit packed/chunky mode is missing, which is a severe penalty. At least at THAT time.


@OneTimer1

Quote:

OneTimer1 wrote: @cdimauro Quote: cdimauro wrote: Beyond Akiko: grafica packed al minor costo usando i bitplane[/url] There was no Akiko in ECS/OCS, AFAIK there was never an Akiko in any of the desktop models.

Akiko was just the starting point to show how packed graphics could have been implemented on the Amiga chipset, without touching its (complex) DMA slots scheduler.
Quote:

And the existence of Akiko in a late AGA game console doesn't tell anything about the intentions of the developers after ECS/OCS but it might tell us more about the intentions of the management refusing a clean cut for a 256 bit GFX and their desperate search for a cheap fix.

Well, Akiko was an idea which came out during a lunch. Commodore's management was a disaster, but engineers weren't that better...

@OneTimer1

Quote:

Quote: OneTimer1 wrote: There was no Akiko in ECS/OCS, AFAIK there was never an Akiko in any of the desktop models. Akiko was just the starting point ...

Akiko was really not a starting point, it was at the very end and nothing the Amiga engineers would have suggested as a successor to OCS/ECS.

@cdimauro

So you chose to instead wait 30+ years and tell US instead, because… ??

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@matthey

Quote:

Most FPGA Amiga hardware also has chunky RTG support and it can be displayed in layers like playfields (usually the background playfield) with planar graphics, sprites and the copper (SAGA).

So most FPGA Amiga hardware is V4?

Also, and I am repeating myself here - the primary “function” of the A4000 systems in the “ecosystem” was to be the development system for the breadbins, hence having the same chipset was a must. For anyone needing to go beyond, there was plenty of third party options, that’s what zorro and the video slots were for. It’s my impression that CBM engineers had interest also in keeping good relations other companies making hardware for Amiga. An alternative route could have been to remove internal graphics and audio from big box systems altogther, and having them rely on additional boards, like was common on certain other platforms.

Last edited by kolla on 29-Oct-2023 at 11:12 PM.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@OneTimer1

Quote:

OneTimer1 wrote: @OneTimer1 Quote: Akiko was just the starting point ... Akiko was really not a starting point, it was at the very end and nothing the Amiga engineers would have suggested as a successor to OCS/ECS.

It was the starting point FOR THE DISCUSSION about how to implement packed/chunky graphics on an Amiga chipset (so, even on OCS).

I hope that is finally clear now.


@kolla

Quote:

kolla wrote: @cdimauro So you chose to instead wait 30+ years and tell US instead, because… ??

Because there are still many retarded which haven't got it even after so long.
Quote:

kolla wrote: @matthey Quote: Most FPGA Amiga hardware also has chunky RTG support and it can be displayed in layers like playfields (usually the background playfield) with planar graphics, sprites and the copper (SAGA). So most FPGA Amiga hardware is V4? Also, and I am repeating myself here - the primary “function” of the A4000 systems in the “ecosystem” was to be the development system for the breadbins, hence having the same chipset was a must. For anyone needing to go beyond, there was plenty of third party options, that’s what zorro and the video slots were for. It’s my impression that CBM engineers had interest also in keeping good relations other companies making hardware for Amiga. An alternative route could have been to remove internal graphics and audio from big box systems altogther, and having them rely on additional boards, like was common on certain other platforms.

Then why the AGA chipset was developed? It made non-sense, since professional already used an OCS system and then adding custom cards where they needed (with Video Toaster being THE example, in this case).

@cdimauro

You misunderstood. The 68000 CPU doesn't need anything fancy to use both edges of the clock, it does this natively. All you do is request a read from an address, this is translated into 8 phases, 4 whole clock pulses, plus wait states added by external circuitry.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@matthey

Quote:

The Amiga can have full VGA compatible gfx modes while retaining the copper, sprites and dual playfield modes. The blitter would not work as is as it is designed for planar blitting. VGA modes would have been very useful mostly because they allow to use cheap and widely available VGA monitors.

So wasting 10000(0?)s of transistors to maintain compatibility with what was a flawed dead end HW design for anything not an low clock 68000 just because C= failed to invest in SW development in the years prior?

Yeah!!

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

@cdimauro

Quote:

Then why the AGA chipset was developed? It made non-sense, since professional already used an OCS system and then adding custom cards where they needed (with Video Toaster being THE example, in this case).

Interesting choice of example - what are the minimum requirements of VT 2000 again? And why did NewTek develop VT 4000, and how … entangled… was development of AGA with development of VT 4000? I’m asking since you claim to know Amiga history to its fine details…

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@kolla

Quote:

This card can still be used in machines without the AGA style video slot, such as the A2000 and A3000 however not all features will be available. In particular some effects and wipes require AGA.

It doesn't sound like AGA was integral to the design of the VT 4000! C= probably forgot to give NewTek datasheets never mind a prototype!

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

@BigD

Even SCALA here in Norway got prototypes, A3400s half’n half ECS and AGA…

Why is backwards compatibility seen as a proof of anything? It meant VT and Amiga could be upgraded independently and based on customer needs - hardly a bad thing.

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@kolla

Yeah, in fact why upgrade to AGA machines at all when all you get is a few upgraded wipes?

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