Click Here
home features news forums classifieds faqs links search
6071 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)
Login

Nickname

Password

Lost Password?

Don't have an account yet?
Register now!

Support Amigaworld.net
Your support is needed and is appreciated as Amigaworld.net is primarily dependent upon the support of its users.
Donate

Menu
Main sections
» Home
» Features
» News
» Forums
» Classifieds
» Links
» Downloads
Extras
» OS4 Zone
» IRC Network
» AmigaWorld Radio
» Newsfeed
» Top Members
» Amiga Dealers
Information
» About Us
» FAQs
» Advertise
» Polls
» Terms of Service
» Search

IRC Channel
Server: irc.amigaworld.net
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
19 crawler(s) on-line.
 166 guest(s) on-line.
 0 member(s) on-line.



You are an anonymous user.
Register Now!
 BigD:  5 mins ago
 matthey:  19 mins ago
 AndreasM:  49 mins ago
 kolla:  50 mins ago
 zipper:  57 mins ago
 OlafS25:  1 hr 21 mins ago
 Swisso:  1 hr 25 mins ago
 amigakit:  2 hrs ago
 amigang:  2 hrs 33 mins ago
 clint:  2 hrs 58 mins ago

/  Forum Index
   /  Amiga General Chat
      /  How good or bad was the AGA chipset in 1992/1993.
Register To Post

Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 Next Page )
Poll : How good or bad was the AGA chipset in 1992/1993.
10p Excellent (Best at 2D/3D, colors, and resolution, frame rate etc.)
5p Good / better than most computer.
0p Barely hanging in there.
-5p Below average / slow but usable
-10p useless / horrible
 
PosterThread
Hammer 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 0:16:18
#421 ]
Elite Member
Joined: 9-Mar-2003
Posts: 5286
From: Australia

@cdimauro

Quote:

Let's see when they will be used. Actually the palette isn't that colorful, despite the claim:

Screen mode changed to EHB allowing perfect sprite coloring

It's certainly not perfect, looking at the SNES version: https://www.youtube.com/watch?v=GPP_RcMdqW4

It's close enough considering the OCS/ECS's limitations.

For Amiga's core European market and for apple-to-apple comparison, SNES arrived in 1992 just like Amiga AGA.

Quote:

Have you checked on WinUAE? Because emulation isn't cycle-accurate on everything which isn't a 68000 with OCS/ECS.


I have run it on WinUAE's "A500 Plus" with 2 MB Chip RAM and A500 Rev6a with 1 MB Chip RAM + 1 MB Fast RAM via the "Cycle-exact CPU emulation speed" and Chipset "Cycle-Exact (Full)"

None WinUAE configuration improves the US Gold's SF2 frame rate like this SF2 tech demo.

When compared to WinUAE's A500 Rev6a with 1 MB Chip RAM + 1 MB Fast RAM, WinUAE's "A500 Plus" with 2 MB Chip RAM shows frame rate degradation.

SF2 tech demo is very smooth on a faster CPU.




Last edited by Hammer on 23-Oct-2022 at 01:55 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)

 Status: Offline
Profile     Report this post  
Hammer 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 1:51:00
#422 ]
Elite Member
Joined: 9-Mar-2003
Posts: 5286
From: Australia

@bhabbott

Quote:

bhabbott wrote:
Amiga fans. When the Amiga can do something better than a PC, they always minimize it.

PC ISA slot is also dependent on bus clock speed.

From Doom benchmarks, the results from ET4000AX ISA with fast CPU beats AGA with fast the CPU, but AGA's results are not bad.

PC CPU doesn't perform math computation with the ISA slot i.e. PC CPU performs math computation with system memory, onboard L2 cache, and L1 cache before throwing the framebuffer to ISA slotted VGA card.


Quote:

Perhaps Commodore should have just stuck a 7MHz 68000 in there, then we wouldn't have Amiga fans whining about 'crippled' CPU performance.

68000 doesn't have 68020's barrel shifter which performs one-cycle shifts of any amount of data.

When Amiga Blitter hogs the Chip RAM bus, 68020's barrel shifter can't operate, hence A1200 couldn't maximize its functional resource. Game AI logic runs with CPU. 68020's barrel shifter can be used as a secondary software Blitter.


Quote:

Nobody complained about the 386SX being crippled by having a 16 bit bus and no cache RAM.

The benefits of the 1990s PC clone market are many alternative PC vendors who offer better specs, better bang for the buck, and run existing software faster.


Quote:

Instead they praised it for bringing 386 enhanced mode to low-end PCs. Similarly, Commodore brought true 32 bit computing to low-end Amigas. But that wasn't good enough for Amiga fans, who are never satisfied and always want more.

All PC 386 CPUs include MMU while Motorola sells MMU as a premium market segmentation.

Commodore's fortunes were largely tied with Motorola's 68K road map. Many 68K platforms died when Motorola EOL'ed 68K as a high-performance CPU, and there's NO "second source AMD insurance" for the 68K CPU market to continue the 68K high-performance CPU legacy. Apollo-Core didn't exist in the early 1990s.

NXP/Freescale still sells non-Altivec with 64-bit SIMD offerings instead of offering Altivec "double pumped" with 64-bit hardware implementation.

Motorola has exited the CPU design business and Freescale has been purchased by NXP.

"Second source AMD insurance" worked when Intel pulled the "Motorola PowerPC" switcheroo with Itanium adventure.

Quote:

To realize the 020's full potential in the A1200 you simply have to add a RAM board in the trapdoor slot, just like A500 users did.

A500's trap door memory bus is connected to custom chips, hence the term "Slow RAM". There are no performance benefits from "Slow RAM".

A1200's trap door memory bus is 32-bit Fast RAM, but it's missing the RAM modules.


Quote:

For more advanced computing you will be needing that extra RAM anyway, so the fact that you need it for maximum CPU speed isn't a big deal. By not putting that RAM on the motherboard, the price was reduced and compatibility increased, and the options for expansion dramatically improved. That meant people could buy an AGA Amiga at the lowest possible price, but with unlimited expansion capability if they wanted it.

Commodore made A1200 like C64/C128 early 1990s computer instead of restoring the Amiga as a very competitive $599-to-$699 USD early 1990s computer.

After requests from large 3rd party software companies that later jumped ship to Sony's Play Station, David Pleasance pushed for an accelerated A1200 bundled deal.

According to David Pleasance, Mehdi Ali told large 3rd party software leadership who requested accelerated A1200 SKU to f_ckoff in front of their faces.

Soon after, EA abandons the Amiga for 3DO and Psygnosis abandons the Amiga for Sony's Playstation.

You're not factoring in the economies of scale instead of Amiga being stuck with non-cost-effective and relatively expensive Phase 5-like pricing solutions!

Mehdi Ali's behavior angered large 3rd party software vendors.

It's too bad the Amiga wasn't the "Made in The UK" 400,000-per-month production Raspberry Pi

PS: I have purchased Raspberry Pi CM4 and PiTray mini which is known to work with PiStorm.

Last edited by Hammer on 23-Oct-2022 at 02:13 AM.
Last edited by Hammer on 23-Oct-2022 at 02:07 AM.
Last edited by Hammer on 23-Oct-2022 at 02:04 AM.
Last edited by Hammer on 23-Oct-2022 at 02:00 AM.
Last edited by Hammer on 23-Oct-2022 at 01:53 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)

 Status: Offline
Profile     Report this post  
cdimauro 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 7:04:53
#423 ]
Elite Member
Joined: 29-Oct-2012
Posts: 3650
From: Germany

@bhabbott

Quote:

bhabbott wrote:
@Kronos

Quote:

Kronos wrote:

So the 020 could access it's local RAM faster them a 386/486 going through ISA bus.

Shock horror!!!

Amiga fans. When the Amiga can do something better than a PC, they always minimize it.

Bruce, he was making fun of you...
Quote:
Quote:
Now tell us how fast could these x86 CPUs access the RAM on the motherboard?

Why should I? Do your own research.

Oh, here you don't post links. Guess why...
Quote:
Quote:
-> AGA crippled CPU performance -> AGA was a bad idea (but maybe still the best C= could have had at that time).

Perhaps Commodore should have just stuck a 7MHz 68000 in there, then we wouldn't have Amiga fans whining about 'crippled' CPU performance.

And here you immediately started the delirium telling completely absurd things that nobody ever thought about.

Which is also 100% pure blind Amigan victim complex: you miss nothing...
Quote:
Nobody complained about the 386SX being crippled by having a 16 bit bus

Well, it was BY DESIGN and it allowed to sell cheap processors on... 1988 (one year after the full/DX 80386).

Motorola hadn't a "SX" on its processors line: there's no "68020SX". Which would have allowed to create cheaper systems.

You had to wait the 1992 for a cheaper Amiga: when it was already too late.
Quote:
and no cache RAM.

OK, and? But it had a fully-fledged PMMU and hardware tasks switching in hardware.
Quote:
Instead they praised it for bringing 386 enhanced mode to low-end PCs.

For very good reasons.
Quote:
Similarly, Commodore brought true 32 bit computing to low-end Amigas.

When? On 1992.
Quote:
But that wasn't good enough for Amiga fans, who are never satisfied and always want more.

Of course. Did you see the market at the time, or were still isolated on your cave?
Quote:
To realize the 020's full potential in the A1200 you simply have to add a RAM board in the trapdoor slot, just like A500 users did.

Plainly wrong.

First, you had to split your CPU and Blitter routines to operate in parallel, and that wasn't easy or possible all the times.

Second, the Blitter (Agnus/Alice, in reality) could give you only 1 memory slot every 4, so the CPU cannot operate at its full potential, since he always misses one slot.
Quote:
For more advanced computing you will be needing that extra RAM anyway, so the fact that you need it for maximum CPU speed isn't a big deal.

Computing is a completely different thing.

At the time Amiga = low cost game machine.
Quote:
By not putting that RAM on the motherboard, the price was reduced and compatibility increased, and the options for expansion dramatically improved. That meant people could buy an AGA Amiga at the lowest possible price, but with unlimited expansion capability if they wanted it.

Which isn't the case, since usually they were used for games.
Quote:
Many users were quite satisfied with the stock machine, which was at least twice as capable as an A500+ in CPU performance, memory, and graphics features.

As I've said, it was good for what it gave for that price.
Quote:
Install a 4MB RAM board and it was at least 4 times better. Put an accelerator card in it and the possibilities are limitless.

How much both costed?
Quote:
But of course none of that was good enough for Amiga fans. It had to have the processing and graphics power of a fast 486DX, along with 16 bit audio (with 32 voice synth), an HD floppy drive, DMA hard drive, and all the same games that were coming out on the PC - at a price well below the cheapest 386SX. And even then they would find something to complain about (...it won't run my entire collection of pirated A500 games!).

And here starts again the delirium + the pathetic Amigan victim complex...
Quote:
The A1200 wasn't the most perfect home computer possible, but what was? None of them.

Who said something different?
Quote:
Could you have done better? I doubt it.

This is a logical fallacy: you're (desperately) trying to change the discussion.

BTW, someone else (The Creator) already provided an enhanced platform some years before.
Quote:
We are lucky that we got what we did.

Correct.
Quote:
Just look at other 'famous' home computers of the day to see how bad it could have been.

Another logical fallacy: this doesn't hide the situation with Commodore's machines.
Quote:
"Oh but look at the PC!", you say enviously. Commodore made those too. In my shop you could have had a 386SX-16 with 2MB RAM, 80MB hard drive and VGA monitor for the same price as an A1200 with 2MB RAM, 40MB hard drive and 1084 monitor. No sound card or speakers in the PC, but hey - they're a cheap upgrade, right?

First, sound cards at the time were cheap.

Second, you had double capacity for the hard drive.

Third, a VGA monitor was usually multisynch and allowed higher frequencies / resolutions.

Forth, the VGA card provided additional, separate, and usually higher performance RAM.

So, you're comparing oranges with apples. As usual.
Quote:
And you can upgrade the RAM on the motherboard to 16MB too with the right SIMMs -

Without requiring an expansion board...
Quote:
but not the CPU, which is forever stuck in 16 bits.

Wrong: you can change the CPU with higher frequency ones. Up to 40Mhz for the 386SX. Which give a consistent performance boost.

Yes, it still had the 16-bit data bus.

Now tell me how I could have replaced the 68EC020 on my Amiga 1200 with one with higher frequencies.
Quote:
But you don't care about that - it can run Microsoft Windows 3.1, and Doom! (in a postage stamp sized window).

Well, at least they could run, eh! And not only them...

 Status: Offline
Profile     Report this post  
cdimauro 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 7:26:50
#424 ]
Elite Member
Joined: 29-Oct-2012
Posts: 3650
From: Germany

@Hammer

Quote:

Hammer wrote:
@cdimauro

Quote:

Let's see when they will be used. Actually the palette isn't that colorful, despite the claim:

Screen mode changed to EHB allowing perfect sprite coloring

It's certainly not perfect, looking at the SNES version: https://www.youtube.com/watch?v=GPP_RcMdqW4

It's close enough considering the OCS/ECS's limitations.

It can get closer using the EHB mode.
Quote:
For Amiga's core European market and for apple-to-apple comparison, SNES arrived in 1992 just like Amiga AGA.

On 1992 my team started Perpetual Craze / Fightin' Spirit for an Amiga OCS/ECS with a 512kB expansion, using EHB. So, I know the situation and limits of the platforms.
Quote:
Quote:

Have you checked on WinUAE? Because emulation isn't cycle-accurate on everything which isn't a 68000 with OCS/ECS.


I have run it on WinUAE's "A500 Plus" with 2 MB Chip RAM and A500 Rev6a with 1 MB Chip RAM + 1 MB Fast RAM via the "Cycle-exact CPU emulation speed" and Chipset "Cycle-Exact (Full)"

None WinUAE configuration improves the US Gold's SF2 frame rate like this SF2 tech demo.

When compared to WinUAE's A500 Rev6a with 1 MB Chip RAM + 1 MB Fast RAM, WinUAE's "A500 Plus" with 2 MB Chip RAM shows frame rate degradation.

OK
Quote:
SF2 tech demo is very smooth on a faster CPU.

Which wasn't available at the time.

To me the reference was always Amiga OCS/ECS + 512kB ram or the Amiga 1200.

 Status: Offline
Profile     Report this post  
Kronos 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 8:34:30
#425 ]
Elite Member
Joined: 8-Mar-2003
Posts: 2562
From: Unknown

@bhabbott

Quote:

bhabbott wrote:

Quote:
-> AGA crippled CPU performance -> AGA was a bad idea (but maybe still the best C= could have had at that time).

Perhaps Commodore should have just stuck a 7MHz 68000 in there, then we wouldn't have Amiga fans whining about 'crippled' CPU performance.



Your almost there.

C= should have made AGA run at 14MHz and ship it with a 14MHz 68000 in a proper A500+ for XMas 1989 (or at bargain prices in 1990).

Would have made perfect sense.


But the fact remains that even a "crippled" 386SX had more compute and more RAM bandwidth then the 020 on a stock A1200.

Sure you can't replace the CPU, but why would you you? A BabyAT board for the 386DX or 486 with the CPU was cheaper than a 030 or 040 accel at any given point in time making it yet another point where the Amiga lost to the PC.

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

 Status: Offline
Profile     Report this post  
BigD 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 10:54:10
#426 ]
Elite Member
Joined: 11-Aug-2005
Posts: 7323
From: UK

@Kronos

Sure ĄGA in 1990 would have been great. But ECS was pointless as all games were targeted at 1.3 KickStart Roms and 1MB Ram expansions which were possible under OCS!

C= always wanted to charge a premium for new technology so A3000 with AGA first was the best case in reality with 1200 in early 1992 with Fast Ram and CD1200 launched at same time as CD32. No A4000 desktop launch required in an AGA A3000+ world but A4000T in 1993 with CD-ROM, SCSI etc would have been great. Maybe A4000 later rebadged the A5000 with AAA or Hombre would have taken the fight to the competition?

In reality the Pet Jet, Irving Gould's excessive salary, no investment in CSG/MOS and the growing PC focus of C= management scuppered all of it. We were lucky to get what we did out if C= IMHO.

Last edited by BigD on 23-Oct-2022 at 11:18 AM.
Last edited by BigD on 23-Oct-2022 at 11:16 AM.
Last edited by BigD on 23-Oct-2022 at 11:00 AM.
Last edited by BigD on 23-Oct-2022 at 10:56 AM.

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

 Status: Offline
Profile     Report this post  
Bosanac 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 11:20:23
#427 ]
Regular Member
Joined: 10-May-2022
Posts: 255
From: Unknown

@Kronos

A guy I went to college with had some clip on Cyrix (I think) 486 on top of his 386sx.

It was slow as hell but a bit faster than without it.

 Status: Offline
Profile     Report this post  
BigD 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 11:30:16
#428 ]
Elite Member
Joined: 11-Aug-2005
Posts: 7323
From: UK

@Bosanac

Same as all those clip on A600 CPU upgrades.

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

 Status: Offline
Profile     Report this post  
Kronos 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 12:27:42
#429 ]
Elite Member
Joined: 8-Mar-2003
Posts: 2562
From: Unknown

@Bosanac

The 486SLC was nothing worth the name "486" but just a really fast 386SX.

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

 Status: Offline
Profile     Report this post  
Hypex 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 12:36:46
#430 ]
Elite Member
Joined: 6-May-2007
Posts: 11215
From: Greensborough, Australia

@cdimauro

Quote:
Indeed. But the main problem is that SuperHires using 8 bitplanes is utterly slow: it leaves only a few memory slots free for doing something other than displaying the screen.


I've read talk on forums over a number of years that HAM mode was slower because it was more taxing to display it on the hardware. However, HAM is a hardware mode. More taxing for software if it calculates it in realtime. But, I didn't see any evidence of HAM6 being more taxing than 6 bitplanes, or HAM8 being more taxing than 8 bitplanes for that matter. Which was implied by what I read.

Quote:
Yes, but the quality is much lower then.


That would be, though if a 24 bit AGA base palette could be set, then hamming RGB could be 8 bit total. A static high nibble and variable low nibble for RGB value would still be limiting. For direct RGB only HAM6 would limit to 4 bit regardless.

Quote:
Indeed. And if you work without fast mem then you end-up having only a few free memory slots available to do this process.


Yes, back buffer rendering makes sense with fastest ram. With only graphics memory, especially in planar it seems somewhat convoluted to render as packed into it, then read it out and render as native to it.

 Status: Offline
Profile     Report this post  
cdimauro 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 13:08:24
#431 ]
Elite Member
Joined: 29-Oct-2012
Posts: 3650
From: Germany

@Hypex

Quote:

Hypex wrote:
@cdimauro

Quote:
Indeed. But the main problem is that SuperHires using 8 bitplanes is utterly slow: it leaves only a few memory slots free for doing something other than displaying the screen.


I've read talk on forums over a number of years that HAM mode was slower because it was more taxing to display it on the hardware. However, HAM is a hardware mode. More taxing for software if it calculates it in realtime. But, I didn't see any evidence of HAM6 being more taxing than 6 bitplanes, or HAM8 being more taxing than 8 bitplanes for that matter. Which was implied by what I read.

In fact the problem is not about the number of bitplanes: HAM = 6 bitplanes = EHB or DPs and HAM8 = 8 bitplanes = normal display mode or 4-bit DPs. So, the workload is exactly the same.

However such number of bitplanes in SuperHires consumes ALL slots on the active display, leaving only the ones for the horizontal and vertical blanking. Then the CPU has very limited access to the Chip-mem for doing something.
Quote:
Quote:
Indeed. And if you work without fast mem then you end-up having only a few free memory slots available to do this process.


Yes, back buffer rendering makes sense with fastest ram. With only graphics memory, especially in planar it seems somewhat convoluted to render as packed into it, then read it out and render as native to it.

Absolutely. Especially in the above condition.

 Status: Offline
Profile     Report this post  
Hypex 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 13:21:33
#432 ]
Elite Member
Joined: 6-May-2007
Posts: 11215
From: Greensborough, Australia

@Karlos

Quote:
I reckon I could knock up a simple enough program to test the idea. It would just read a source RGB image and generate one that's 4x wider and repeats every row 4x, applying the logic described.


Actually this could be simple enough. It doesn't even need to render it on screen. Just generate it as a HAM IFF file. IFF library may help. On second thoughts hardware direct avoids an API.

 Status: Offline
Profile     Report this post  
Karlos 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 23-Oct-2022 14:23:26
#433 ]
Elite Member
Joined: 24-Aug-2003
Posts: 4404
From: As-sassin-aaate! As-sassin-aaate! Ooh! We forgot the ammunition!

@Hypex

I was thinking even simpler. Just export a new RGB image at 4x the resolution to see how bad it is.

It can also be resampled back to the original size and the error per pixel determined.

_________________
Doing stupid things for fun...

 Status: Offline
Profile     Report this post  
Hammer 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 24-Oct-2022 6:49:07
#434 ]
Elite Member
Joined: 9-Mar-2003
Posts: 5286
From: Australia

@cdimauro

Quote:

On 1992 my team started Perpetual Craze / Fightin' Spirit for an Amiga OCS/ECS with a 512kB expansion, using EHB. So, I know the situation and limits of the platforms.

FYI, I'm already aware of Fightin' Spirit's EHB mode, but I dislike its artwork, but the gameplay controls are better than Elfmania's.


Fightin' Spirit's 2D engine's performance is okay but lacks Elfmania's tech demo features.

Fightin' Spirit's 2D engine could probably support Street Fighter Zero 2's toon artwork style or Mortal Kombat 2's artwork style.

64-color EHB mode for Mortal Kombat 2 would be nicer.

https://www.youtube.com/watch?v=lAcknPOgzTM
Simultaneous colors comparison of Mortal Kombat II for Sega Genesis, SNES, and vs 32X

Amiga AGA's 7-bitplane (simultaneous 128 colors) mode would have covered SNES's simultaneous 116-137 color range average.

Commodore didn't have 1st party game developers for creating good game 2D game engines for licensing, hence each 3rd party developer reinvents "the wheel" and the Amiga platform lacks game quality consistency.

Microsoft's Xbox Team has learned the lessons from the success of Japanese gaming platforms and the failures of western gaming platforms.

PC hardware companies like NVIDIA have large-scale software developer relations teams, technical software example libraries, and excellent SDK. Nvidia's software Gameworks initiative is not a paperweight.

I support Scorpion Engine for the Amiga since indie game developers don't need to reinvent "the wheel" and focus on providing artwork content.

Quality software sells hardware.




Last edited by Hammer on 24-Oct-2022 at 07:38 AM.
Last edited by Hammer on 24-Oct-2022 at 06:56 AM.
Last edited by Hammer on 24-Oct-2022 at 06:50 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)

 Status: Offline
Profile     Report this post  
bhabbott 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 24-Oct-2022 8:20:49
#435 ]
Regular Member
Joined: 6-Jun-2018
Posts: 336
From: Aotearoa

@BigD

Quote:

BigD wrote:

Sure ĄGA in 1990 would have been great. But ECS was pointless as all games were targeted at 1.3 KickStart Roms and 1MB Ram expansions which were possible under OCS

ECS wasn't developed for games.

Quote:
C= always wanted to charge a premium for new technology so A3000 with AGA first was the best case in reality with 1200 in early 1992

So the VIC-20, C64 and A1000 were all 'premium' priced? (A1000 US$1,285, IBM PCjr US$1,269, Tandy 1000 US$1,200).

You may may be right about the A3000 being the first to sport AGA if it was produced in 1990, but you can bet they would have released a low-end model with AGA shortly afterwards. However the A3000 was a change of tack for Commodore, who previously had concentrated on the low-end consumer market. AGA brought them back to that path. Producing AGA in 1990 would indicate a desire to continue concentrating on the low end like they did with the VIC-20 and C64 etc., so the A3000 might not have happened.

Quote:
...with Fast Ram and CD1200 launched at same time as CD32. No A4000 desktop launch required in an AGA A3000+ world but A4000T in 1993 with CD-ROM, SCSI etc would have been great. Maybe A4000 later rebadged the A5000 with AAA or Hombre would have taken the fight to the competition?

So you wanted Commodore to produce machines they would have to charge a premium for?

The A3000 was an expensive mistake. Instead of letting the engineers go wild with a developer's dream machine, they should have done what they did with the VIC-20 - enhancing its gaming capabilities while keeping the familiar user interface and low price. But in 1990 the engineers were deluding themselves into thinking the Amiga could compete head-to-head with PCs (if only they could make it more 'professional'). This was an understandable delusion considering the technological state of PCs at the time - but ignored the realities of the marketplace.

Business customers had already gone all-in on IBM in 1981, and that was never going to change. What would change is that with such overwhelming support the PC was bound to overtake any other architecture, even if had to reinvent itself several times over to do it. Before 1990 Windows was a joke, and the typical PC was stuck in single-tasking DOS with clunky high memory management and direct hardware access for performance. But this soon changed. Windows 3.1 introduced 386 enhanced mode which was a portent of things to come. By 1995 the PC had overtaken the Amiga in every respect except efficiency (which didn't matter when PC hardware was rapidly becoming more powerful and cheaper). A few years later it switched to NT (now called XP) keeping the familiar interface but fully virtualizing the hardware. Games now went totally through the OS, so hardware was irrelevant.

On the Amiga we agonized about what hardware we should have in it. On the PC you didn't have to worry. Graphics chipset not up to the task? Just put in a new video card. Sound not acceptable? Plug in a better sound card. Don't worry about hardware differences, the OS will take care of it. So long as your game or app will work with the version of Windows you have, you're done!

Quote:
In reality the Pet Jet, Irving Gould's excessive salary, no investment in CSG/MOS and the growing PC focus of C= management scuppered all of it. We were lucky to get what we did out if C= IMHO.

In reality Commodore's sole reason for existence was to make money for its owners. That we got a lasting legacy out of it is pure serendipity. Gould invested a lot of money in CBM and deserved to get a return on it. In 1991 he was 70 years old and CEO of a billion dollar multinational company. Why shouldn't he get something back from the company he saved several times?

But Amiga fans would insist that he pour all the profits back into the business in order to ensure the Amiga's survival. Then they would present a laundry list of absolutely essential features the Amiga must have, hire a few hundred 'top' engineers to make it a reality, then wonder why it couldn't be made for the price they wanted and bombed in the marketplace. A few years later Commodore would still be bankrupt, and its owners destitute. Amiga fans would then point fingers everywhere but at themselves.

 Status: Offline
Profile     Report this post  
bhabbott 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 24-Oct-2022 9:42:47
#436 ]
Regular Member
Joined: 6-Jun-2018
Posts: 336
From: Aotearoa

@Hammer

Quote:

Hammer wrote:

Amiga AGA's 7-bitplane (simultaneous 128 colors) mode would have covered SNES's simultaneous 116-137 color range average.

Who cares how many colors the SNES had? Each platform's version should stand on its own merits.

There are some ZX Spectrum games that I still enjoy playing despite the machine's severe limitations. These are games that were coded to those limitations. And guess what? Enhancing them with 128 colors and fancy parallax scrolling etc. wouldn't make them any more playable. In fact often it's the opposite, with the flashy graphics distracting the player.

Quote:
Commodore didn't have 1st party game developers for creating good game 2D game engines for licensing, hence each 3rd party developer reinvents "the wheel" and the Amiga platform lacks game quality consistency.

What you call 'quality consistency' I call blandness. One of the great things about less capable platforms was the various tricks programmers tried to visualize their ideas, which often resulted in unique graphical effects. Some worked better than others for sure, but each had its own charm.

Commodore wasn't in the business of producing games. They made the hardware for us to do great things with. That is why the Amiga has such a large and varied game library. But they didn't just put the machines out there and leave us to it. They published detailed hardware and software documentation, and provided support for anyone who needed it. Anyone could become a registered developer for a small fee, and get advanced information, tips on how use the hardware and OS etc., and answers to questions. What they didn't do - in contrast to game console companies - was lock you into onerous development and marketing agreements with strict 'quality' standards.

Quote:
Microsoft's Xbox Team has learned the lessons from the success of Japanese gaming platforms and the failures of western gaming platforms.

The 'failures' of western gaming platforms?

Quote:
PC hardware companies like NVIDIA have large-scale software developer relations teams, technical software example libraries, and excellent SDK. Nvidia's software Gameworks initiative is not a paperweight.

With revenue of US$27 billion this year, and over 22,000 employees, they can afford it. But game development these days isn't for amateurs. Call of Duty: Modern Warfare II cost an estimated US$300 million.

Quote:
I support Scorpion Engine for the Amiga since indie game developers don't need to reinvent "the wheel" and focus on providing artwork content.

Yet another 'paint by numbers' game designer?

I hadn't heard about that one until now. But...

Quote:
The Scorpion Engine is a fast and flexible game creation kit for Amiga, offering a modern Windows IDE for development (The editor itself is not open source...

If I am going to design a game for the AmIga, I want to do it on an Amiga. And I want to code it myself too. Source code is available for existing games which has just about everything I would need if I don't want to 'reinvent the wheel'. But inventing wheels is fun. Part of the reason I would want to make a game is to learn how to do that.

As for producing 'new' games, what would I want to do that hasn't been done already? I can think of a few things, but they probably wouldn't fit the mold of this game engine.

 Status: Offline
Profile     Report this post  
cdimauro 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 25-Oct-2022 5:29:09
#437 ]
Elite Member
Joined: 29-Oct-2012
Posts: 3650
From: Germany

@Hammer

Quote:

Hammer wrote:
@cdimauro

Quote:

On 1992 my team started Perpetual Craze / Fightin' Spirit for an Amiga OCS/ECS with a 512kB expansion, using EHB. So, I know the situation and limits of the platforms.

FYI, I'm already aware of Fightin' Spirit's EHB mode, but I dislike its artwork, but the gameplay controls are better than Elfmania's.

I can say nothing about personal tastes. As I've said, Fightin' Spirit was strongly inspired by Art of Fighting for Neo Geo, which is quite evident.
Quote:
Fightin' Spirit's 2D engine's performance is okay but lacks Elfmania's tech demo features.

Do you know how much I like demos, right?

Jokes aside, the two games have completely different goals and consequently completely different engines that cannot be "merged".

Fightin' Spirit was created with the precise goal to "port" as much of possible from Art of Fighting to an Amiga OCS/ECS. So, to give big backgrounds, huge characters, a lot of animations, "energy bubbles", special moves, and especially to be very colorful.

I think that we succeed on the goal (considering the very limited hardware) and that it was enough for us.

Nothing else from Elfmania could be implemented (besides the parallax scroll on 1MB Chip-Mem machines, that I've already developed) simply because there are not enough performance resources.
Quote:
Fightin' Spirit's 2D engine could probably support Street Fighter Zero 2's toon artwork style or Mortal Kombat 2's artwork style.

64-color EHB mode for Mortal Kombat 2 would be nicer.

It's certainly possible: our engine was born specifically for these kind of games and to move a lot of graphic on the screen.

However time is over...
Quote:
https://www.youtube.com/watch?v=lAcknPOgzTM
Simultaneous colors comparison of Mortal Kombat II for Sega Genesis, SNES, and vs 32X

Amiga AGA's 7-bitplane (simultaneous 128 colors) mode would have covered SNES's simultaneous 116-137 color range average.

Yes. Fighitin' Spirit AGA was using 128 colors (AFAIR).
Quote:
Commodore didn't have 1st party game developers for creating good game 2D game engines for licensing, hence each 3rd party developer reinvents "the wheel" and the Amiga platform lacks game quality consistency.

It wasn't Commodore fault. At the time there was computer vendor which was acting like that.

BTW, this usually applies to console vendors and not to computer vendors.
Quote:
Microsoft's Xbox Team has learned the lessons from the success of Japanese gaming platforms and the failures of western gaming platforms.

XBox is a console...
Quote:
PC hardware companies like NVIDIA have large-scale software developer relations teams, technical software example libraries, and excellent SDK. Nvidia's software Gameworks initiative is not a paperweight.

nVidia tackles everything which is GPU-related.
Quote:
I support Scorpion Engine for the Amiga since indie game developers don't need to reinvent "the wheel" and focus on providing artwork content.

Quality software sells hardware.

But it's too late know: Amiga as a platform is dead since very long time.

 Status: Offline
Profile     Report this post  
Hammer 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 25-Oct-2022 5:45:33
#438 ]
Elite Member
Joined: 9-Mar-2003
Posts: 5286
From: Australia

@bhabbott

Quote:

Who cares how many colors the SNES had? Each platform's version should stand on its own merits.

There are some ZX Spectrum games that I still enjoy playing despite the machine's severe limitations. These are games that were coded to those limitations. And guess what? Enhancing them with 128 colors and fancy parallax scrolling etc. wouldn't make them any more playable. In fact often it's the opposite, with the flashy graphics distracting the player.

Refer to this topic "How good or bad was the AGA chipset in 1992/1993".

On A500 with PiStorm/Emu68, I run "quake.gcc-2.95.3.030" (from Coffin R60) in HAM6 mode at 320x200 resolution, and the frame rate is like Doom EHB mode with a fast CPU i.e. the frame rate is smoother when compared to slide show IBM VGA ISA with Athlon XP 2200+ (1800 Mhz).

Quake HAM6 mode's appearance is a close approximation of RTG or VGA versions.

Based on Quake RTG "quake.gcc-2.95.3.030" results, PiStorm/Emu68/RPI 3a+ has about Pentium II 266/300 results.

Vampire RTG results are about half the frame rate from PiStorm/Emu68/RPI 3a+ RTG.

The modern low-cost "ZX Spectrum" is the Raspberry Pi. Raspberry Pi 400 is a keyboard computer with a Raspberry Pi GPIO port.


Quote:

What you call 'quality consistency' I call blandness.

That's blindness.


Quote:

One of the great things about less capable platforms was the various tricks programmers tried to visualize their ideas, which often resulted in unique graphical effects. Some worked better than others for sure, but each had its own charm.

Commodore wasn't in the business of producing games.

For Commodore's main revenue source, Amiga's use case is predominantly gaming. Amiga's Video Toaster market is tiny when compared to PC's large-scale business office market.

Commodore wouldn't survive just from Amiga Video Toaster's unit sales.

Quote:

They made the hardware for us to do great things with. That is why the Amiga has such a large and varied game library. But they didn't just put the machines out there and leave us to it. They published detailed hardware and software documentation, and provided support for anyone who needed it. Anyone could become a registered developer for a small fee, and get advanced information, tips on how use the hardware and OS etc., and answers to questions.

Not good enough.

One of Nintendo's main criteria for selecting NVIDIA is developer support, not just commodity shippers like Qualcomm.

Both Microsoft and Sony have large enough 1st party developers to self-support their respective AMD-based game hardware platforms.

With Intel ARC and incoming Battlemage, Intel's game developer support is improving.

Quote:

What they didn't do - in contrast to game console companies - was lock you into onerous development and marketing agreements with strict 'quality' standards.

1st party exclusive games enabled Nintendo to survive the weak Wii U (~9 million units) and late N64 releases.

Quote:

The 'failures' of western gaming platforms?

Not including general purpose platforms e.g. Windows, the Xbox platform is the only western gaming remaining.

Quote:

With revenue of US$27 billion this year, and over 22,000 employees, they can afford it. But game development these days isn't for amateurs. Call of Duty: Modern Warfare II cost an estimated US$300 million.

Genshin Impact had an initial budget of around $100 million USD and was built on Unity3D 2017 engine. It costs $200 million per year to produce new content and keep everything running on schedule. $100 million in 2017 is about $53 million in 1990.

On January 18, 2022, Microsoft announced that it would be acquiring Activision Blizzard for $68.7 billion in an all-cash deal.

With the Activision Blizzard buyout deal, it would make Microsoft the third-largest gaming company in the world and the largest headquartered in the Americas, behind Chinese company Tencent and the Japanese conglomerate Sony.
------

NVIDIA's "The Way It's Meant To Be Played" initiative started in the year 2002 era.
NVIDIA's revenue in 2002 is about $1.93 or $2.97 billion in 2022.

Commodore's 1991 $1 billion revenue is about $2.19 billion in 2022.

GeForce FX's inferior hardware design resulted in revenue of $1.81 billion in 2003. GeForce 6000 series replaced the failed GeForce FX in 2004.

NVIDIA 's Gameworks initiative replaced "The Way It's Meant To Be Played" initiative.

ATI's Radeon 9700/9800's performance superiority has been minimized by NVIDIA's "The Way It's Meant To Be Played" initiative.

Within 14 months of the FX5800's release, GeForce FX's VILW design was replaced by the SIMD-based GeForce 6800.


Last edited by Hammer on 25-Oct-2022 at 07:00 AM.
Last edited by Hammer on 25-Oct-2022 at 06:13 AM.
Last edited by Hammer on 25-Oct-2022 at 06:10 AM.
Last edited by Hammer on 25-Oct-2022 at 06:02 AM.
Last edited by Hammer on 25-Oct-2022 at 06:01 AM.
Last edited by Hammer on 25-Oct-2022 at 06:00 AM.
Last edited by Hammer on 25-Oct-2022 at 05:52 AM.
Last edited by Hammer on 25-Oct-2022 at 05:51 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)

 Status: Offline
Profile     Report this post  
cdimauro 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 25-Oct-2022 5:54:04
#439 ]
Elite Member
Joined: 29-Oct-2012
Posts: 3650
From: Germany

@bhabbott

Quote:

bhabbott wrote:

Business customers had already gone all-in on IBM in 1981

In your parallel universe.

I'm really impressed by your fervid imagination and the incredible things that you invent to sustain your distorted propaganda.
Quote:
Windows 3.1 introduced 386 enhanced mode which was a portent of things to come.

It was already introduced on Windows 3.0. However Windows 2.0 already supported / used the 80386.
Quote:
By 1995 the PC had overtaken the Amiga in every respect

Time is passing very slowly on your cave...
Quote:
except efficiency

?!?
Quote:
(which didn't matter when PC hardware was rapidly becoming more powerful and cheaper). A few years later it switched to NT (now called XP)

It was Windows 2000.
Quote:
keeping the familiar interface but fully virtualizing the hardware. Games now went totally through the OS, so hardware was irrelevant.

Not "irrelevant": it wasn't possible to directly access the hardware. For very good reasons.
Quote:
On the Amiga we agonized about what hardware we should have in it. On the PC you didn't have to worry. Graphics chipset not up to the task? Just put in a new video card. Sound not acceptable? Plug in a better sound card. Don't worry about hardware differences, the OS will take care of it. So long as your game or app will work with the version of Windows you have, you're done!

Which is simply awesome!
Quote:
But Amiga fans would insist that he pour all the profits back into the business in order to ensure the Amiga's survival. Then they would present a laundry list of absolutely essential features the Amiga must have, hire a few hundred 'top' engineers to make it a reality, then wonder why it couldn't be made for the price they wanted and bombed in the marketplace. A few years later Commodore would still be bankrupt, and its owners destitute. Amiga fans would then point fingers everywhere but at themselves.

And here restarts your delirium....
Quote:

bhabbott wrote:
@Hammer

Quote:

Hammer wrote:

Amiga AGA's 7-bitplane (simultaneous 128 colors) mode would have covered SNES's simultaneous 116-137 color range average.

Who cares how many colors the SNES had? Each platform's version should stand on its own merits.

There are some ZX Spectrum games that I still enjoy playing despite the machine's severe limitations. These are games that were coded to those limitations. And guess what? Enhancing them with 128 colors and fancy parallax scrolling etc. wouldn't make them any more playable. In fact often it's the opposite, with the flashy graphics distracting the player.

Specify: in YOUR case.

Games can take advantage of better resources while providing the same mechanics or even enhancing them.

Only on your parallel universe the opposite happens.
Quote:
Quote:
Commodore didn't have 1st party game developers for creating good game 2D game engines for licensing, hence each 3rd party developer reinvents "the wheel" and the Amiga platform lacks game quality consistency.

What you call 'quality consistency' I call blandness. One of the great things about less capable platforms was the various tricks programmers tried to visualize their ideas, which often resulted in unique graphical effects. Some worked better than others for sure, but each had its own charm.

And here you talk because you developed no game at all, so you're daydreaming about something where you've no clue at all.

In the real world game developers like to have hardware which is very easy to program to achieve their goals. And the same applies consequently to graphic and music artists.

Special effects and strange tricks are demos' dominion, which are a completely different thing.
Quote:
Quote:
PC hardware companies like NVIDIA have large-scale software developer relations teams, technical software example libraries, and excellent SDK. Nvidia's software Gameworks initiative is not a paperweight.

With revenue of US$27 billion this year, and over 22,000 employees, they can afford it.

nVidia already did those programs and supported developers since AGES. That's the reason why NOW it's so big.

As usual, you talk of things that you have no clue at all.
Quote:
But game development these days isn't for amateurs. Call of Duty: Modern Warfare II cost an estimated US$300 million.

True, and? It sells much more than that each time.
Quote:
Quote:
The Scorpion Engine is a fast and flexible game creation kit for Amiga, offering a modern Windows IDE for development (The editor itself is not open source...

If I am going to design a game for the AmIga, I want to do it on an Amiga.

And here it's evident that you have no idea of what does it mean developing a game.

So, you prefer to torture yourself instead of taking advantage of any better development tool.

I've a word for this: masochism.
Quote:
And I want to code it myself too. Source code is available for existing games which has just about everything I would need if I don't want to 'reinvent the wheel'.

Those games are much more advanced that it was possible on the most widespread Commodore machines. Enjoy porting them...
Quote:
But inventing wheels is fun. Part of the reason I would want to make a game is to learn how to do that.

Makes sense: since you retired you've a lot of time. So, you can waste it as you wish.

 Status: Offline
Profile     Report this post  
Hammer 
Re: How good or bad was the AGA chipset in 1992/1993.
Posted on 25-Oct-2022 6:48:42
#440 ]
Elite Member
Joined: 9-Mar-2003
Posts: 5286
From: Australia

@cdimauro
Quote:

I can say nothing about personal tastes. As I've said, Fightin' Spirit was strongly inspired by Art of Fighting for Neo Geo, which is quite evident.

Fightin' Spirit's player and NPC object can get lost in background artwork.

Quote:

Do you know how much I like demos, right?

Jokes aside, the two games have completely different goals and consequently completely different engines that cannot be "merged".

Fightin' Spirit was created with the precise goal to "port" as much of possible from Art of Fighting to an Amiga OCS/ECS. So, to give big backgrounds, huge characters, a lot of animations, "energy bubbles", special moves, and especially to be very colorful.

I think that we succeed on the goal (considering the very limited hardware) and that it was enough for us.

Nothing else from Elfmania could be implemented (besides the parallax scroll on 1MB Chip-Mem machines, that I've already developed) simply because there are not enough performance resources.

I have seen console gamers criticizing Fightin' Spirit's lacking parallax scrollers.

SNES's SF2 player objects were reduced in size.

The major criticism of Elfmania is the control scheme, hence the "tech demo" label for the game.

Quote:

It wasn't Commodore fault. At the time there was computer vendor which was acting like that.

BTW, this usually applies to console vendors and not to computer vendors.

These days, Intel has gaming teams that manage Intel's PC gaming aspects.

For example, Intel assisted World of Tank's developers in implementing real-time BVH raytracing in software on PC CPUs via Intel Embree integration.
https://www.youtube.com/watch?v=-w7wUs30OXk

_________________
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)

 Status: Offline
Profile     Report this post  
Goto page ( Previous Page 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 Next Page )

[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 Amigaworld.net.
Amigaworld.net was originally founded by David Doyle