@Karlos

Quote:

Karlos wrote: If you are a major player you can afford to write off a couple of million in your R&D path to commercial silicon. Except you also have the expertise and resources to avoid it most of the same time anyway. If enough enthusiast backing and technical expertise were assembled together to design and build the MH68100 ASIC and it was found to be broken, you've burned all the fuel for your moonshot already. You can't just write it off and fix it because you aren't a multi billion chip design/manufacturing outfit.

With TSMC's 40 ns fab, RPi has spent money on wafer starts for their semi-custom RP2040 SoC designs.

None of the Amiga-related companies are comparable to RPI's company strength level.

RPI in 2023 is a $266 million revenue company. https://investors.raspberrypi.com/

Last edited by Hammer on 16-Nov-2024 at 09:16 AM.

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

@Hammer

How many Amigas of each kind had been sold at that time did not matter, how much were still in use by people considering pricey updates was what matters.

The vast majority of A500 had already been stored in attics or gone to the landfill at that time so not much of a market.

The BPPC/BV combo is actually a good argument for something else as it was pretty much unusable in a desktop case.

I did buy a Blizz2060 on launch and adjusting for inflation it would have been about 2000€ with just 8MB.

Add RTG and a HD and you not also increase pricing but you are also running into space and power constraints in an A500. These were real even with the updates people really did back than, the 030 card that collided with the Kick-switcher, the FF that just didn't play nice with the 2MB-Chip addon. External CD and HD that had their own PSU needed to be switched on (and of) in a specific order to avoid killing the HW.

All could at that time be solved by spending pocket money on a used A2000.

But yeah all these companies with years of prior experience were just to stupid to target the huge market of people still running a 2 floppy 1MB A500 in 96!!

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

@Kronos

Quote:

How many Amigas of each kind had been sold at that time did not matter, how much were still in use by people considering pricey updates was what matters. The vast majority of A500 had already been stored in attics or gone to the landfill at that time so not much of a market.

A500 didn't receive 68060, PPC and RTG upgrades from the likes of Phase 5. The major reasons for the A500 being less used is the lack of RTG upgrades.

Smaller install base OCS/ECS A1500/A2000/A3000 has CGX RTG upgrades.

I pulled out my broken A500 rev5 from storage and repaired it when I heard about Vampire V2 which is before I purchased risky "for parts" sale A1200 rev 1D1 during COVID19 lockdown. Vampire V2 has RTG for the A500.

I sold my A3000 to partly fund my Pentium 166 PC build during 1996, but I kept the A500. I needed the Pentium PC when I entered university in 1997.

Before COVID19 lockdown, I purchased Wicher 508i as a temporary accelerator card (with IDE controller) when I waited in the queue for Vampire V2. I canceled my Vampire V2 queue when I heard about the original PiStorm. WHDLoad games are matured and I started re-use my retro A500.

WHDload games are the major reason for many users reactivating retro A500 with Fast RAM and IDE controller upgrades.

Hyperion also release AmigaOS 3.1.4 (Sep 2018) and 3.2.x (May 2021) 16-bit Kickstart ROM for A500/A600/A1500/A2000 and 32-bit Kickstart ROMs for A1200/A3000/A4000.

Dread/Grind running on stock A500.

There are major factors why many users are reactivating their A500s for retro Amiga gaming scene.

https://www.lemonamiga.com/games/list.php?list_year=1995&list_year_option=equal&lineoffset=0
Amiga 1995 game releases for AGA and OCS/ECS.

From 1994 to 1995, Amiga games were being released for existing OCS/ECS and AGA install base.

Examples
1. Gloom Deluxe and Nemac IV runs on 68020/68030 CPU-accelerated OCS/ECS Amigas.
2. Amiga's Mortal Kombat 2 OCS/ECS was released in 1994 by Acclaim Entertainment.
3. Shadow Fighter OCS/ECS/AGA was released in 1994 by Gremlin Interactive.
4. Colonization OCS/ECS/AGA was released in 1995 by MicroProse Software.
5. Skidmarks ECS/AGA was released in 1995 by Acid Software.

From 1994 and beyond, the Amiga didn't have PC's big title name texture-mapped 3D games, but Amiga had game ports from the PC when it's possible.

Amiga's 68060-50 and CGX RTG experience started from Phase 5's 1995 releases for Amiga models with smaller install base.

Later ACA500 (A1200 accelerator adapter for A500), Vampire V2, Firebird V4 and the original PiStorm were released and targeted the larger install base A500.

Warp 560, Vampire V2, Firebird V4 and PiStorm upgrades enables the A500 majority to re-join the fast 68K with RTG or AGA Amigas ecosystem.

Last edited by Hammer on 17-Nov-2024 at 10:59 PM.
Last edited by Hammer on 17-Nov-2024 at 10:54 PM.
Last edited by Hammer on 17-Nov-2024 at 10:38 PM.
Last edited by Hammer on 17-Nov-2024 at 10:27 PM.
Last edited by Hammer on 17-Nov-2024 at 10:25 PM.
Last edited by Hammer on 17-Nov-2024 at 10:19 PM.

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

Usually, Hammer's longer posts often seem to be spurious but I think I agree with the sentiment that faster CPU expansions for A500 don't make that much sense without RTG. Without putting too fine a point on it, how much software is there that requires processing power in excess of a fast 68030 (let alone 040 and above) that is fine on OCS/ECS display modes?

Higher resulotion and faster CPUs go hand in hand for productivity. A local RTG solution makes even more sense in an OCS/ECS machine having 16-bit legacy buses.

It is interesting to contemplate if or how the landscape may have been different if the A500 did get some more attention back in the day.

_________________
Doing stupid things for fun...

@Hammer

OEM's don't make product choices based purely on hardware capability. They also use demographics, and the truth is that the vast majority of budget-oriented buyers do so because of their limited budget. Meaning that despite there being an install base of 2M+, the actual target market for an expansion costing as much or more than than the original budget product, is much smaller.

What are the odds that a person who purchases an airline ticket during a flash sale, discounted to 30% of its regular list price, would take up the option of spending 4x that much to be bumped up to business class? The airlines' data shows: not very good.

And it's not just a matter of recouping the upfront development costs. It's that they also have to expand their support and maintenance capabilities for additional hardware. And they can be significantly more for non-integrated products (A500) as compared to integrated products (A2000).

_________________
All the way, with 68k

@Karlos

Quote:

Usually, Hammer's longer posts often seem to be spurious but I think I agree with the sentiment that faster CPU expansions for A500 don't make that much sense without RTG. Without putting too fine a point on it, how much software is there that requires processing power in excess of a fast 68030 (let alone 040 and above) that is fine on OCS/ECS display modes?

For new hardware performance, it's the chicken or the egg issue. This is why Japanese game console vendors have exclusive games.

Microsoft has licensed the Doom IP for Doom 95 release as the Windows 95 gaming's "tip of the spear" while IBM didn't care about native OS/2 gaming. Microsoft was showing previews of Windows Chicago with Doom 95 previews in PC magazines.

For 3rd party Amiga-related add-on vendors, the targeted Amiga models are the small install base which has economies of scale issues.

Post-Commodore and before Vampire V2 / PiStorm, no 3rd party add-on Amiga related vendors has taken over RISC hardware upgrade path for the A500 majority. There was a small timing window for 3rd party add-on Amiga vendor to take over the Amiga hardware evolution direction.

Without RTG upgrade path, A500 with Progressive 68040-33 Mhz accelerator is a dead end. The same problem for all A500 with GVP 68030 @ 40 Mhz accelerators who can't join the "next-gen" 256 color gaming. Meanwhile, most 32-bit PCs with i386DX-25, i386DX-33 or Am386-40 CPUs can be upgraded with faster SVGA clone cards and join the 256 color gaming.


Quote:

@Karlos Higher resulotion and faster CPUs go hand in hand for productivity. A local RTG solution makes even more sense in an OCS/ECS machine having 16-bit legacy buses.

For 68060 and CGX RTG, Phase 5 focused on smaller base of A1200, A1500/A2000, A3000 and A4000.

A1500/A2000's Zorro II slots are 16-bit like on A500's Zorro I edge connector.

For Doom class games, fast VGA clone like ET4000AX ISA would do the job. It's good enough PC ISA, it's good enough for Zorro II and Zorro I.

Amiga's core audience is games e.g. retro theA500mini's bundled games software. The Amiga is not a Mac.

A1200/CD32 wasn't able to supplant the A500 as the Amiga standard.

Quote:

@Karlos It is interesting to contemplate if or how the landscape may have been different if the A500 did get some more attention back in the day.

At the end of the day, Phase 5 needs to find a niche market with big enough economies of scale or they'll go bust.

Raspberry Pi has found a niche market with big enough economies of scale.

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

@agami

Quote:

OEM's don't make product choices based purely on hardware capability. They also use demographics, and the truth is that the vast majority of budget-oriented buyers do so because of their limited budget. Meaning that despite there being an install base of 2M+, the actual target market for an expansion costing as much or more than than the original budget product, is much smaller.

The main reason for A600's existence is to kill GVP. Commodore management's Mehdi Ali hates GVP.

Commodore doesn't want a strong 3rd party Amiga addon vendor like GVP.

By 1991, GVP earned over $20 million in revenue.
By 1992, GVP earned over $35 million in revenue.

From Micheal Dell's Direct from Dell story, Dell started from 3rd party add-on business for the IBM PC.

From Commodore Final Yeas by Brian Bagnall

Quote:

One of GVP’s most profitable and popular peripherals relied on the Amiga 500 and 2000’s CPU expansion capabilities. “It's all made possible because the A500 and the A2000 had the CPU expansion slot,� explains Bucas. “That implies that you can literally disable the CPU on the motherboard and plug in another board. .... In 1992, GVP released a CPU accelerator called the A530 Turbo. Amazingly, it boosted the A500’s 7 MHz speed up to 40 MHz using a Motorola 68EC030 processor. “In A500 we had a box that you plug in the side slot because it also had the CPU expansion board,� says Bucas. “Bottom line, that's why GVP could do all those CPU accelerator products which were super successful. We made tonnes and tonnes of money on those.� Mehdi Ali and Bill Sydnes continued looking for ways to hurt GVP’s dominance of the Amiga peripherals aftermarket. “Basically a number of people at Commodore felt that GVP was growing too large,� recalls Bucas. “So finally Commodore management decided that, ‘Listen, on the so called A300 (which eventually was A600), we're going to make no expandability. We're going to screw GVP.’ That was the mantra inside. ‘We want to make sure they can make no money on this.’� The A600 would allow no expansion, other than devices that could plug into the PCMCIA port. “The bottom line is, they took away the expansion port,� says Bucas.

Last edited by Hammer on 18-Nov-2024 at 03:58 AM.
Last edited by Hammer on 18-Nov-2024 at 03:52 AM.
Last edited by Hammer on 18-Nov-2024 at 03:44 AM.
Last edited by Hammer on 18-Nov-2024 at 03:41 AM.
Last edited by Hammer on 18-Nov-2024 at 03:35 AM.

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

@Hammer

Say you buy an 68030 for your amiga 500, is there any reason thereafter for you to upgrade to an Amiga 2000?

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

@pixie

How is that an upgrade? They're just different form factors with different ways of putting in expansions! The A2000 will be neater as you add more cards and hard drives but beyond that I'm not sure!

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

@BigD

A GVP provided you with an HD and a CPU + memory, some even with a bridge board pc... I know that A2000 could provide you with expansion slots, but all of that isn't exactly free, so if you had an A500 you could have an upgrade path that would perhaps fit 90% of your needs and for a cheaper price than buying a new machine, also add a CPU card memory and hd. Graphics board would cost you an arm and a leg back then. I see it as an upgrade.

My A1200 with a blizzard accelerator plus memory was for me more than enough. I would never pay for A4000 because it didn't suit my needs vs price. Afterwards I even got RTG. I was more than satisfied.

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

If you were an A500 owner and you had a lot of peripherals you didn't want to replace *and* money was no object...

https://amiga.resource.cx/exp/progressive540

I remember this being reviewed in Amiga Shopper magazine back in the day. It was an absolute beast.

I always thought it was a curious expansion though. So much CPU power but not much else. A DMA SCSI interface for fast access to storage would've been a real value add, but it was already insanely expensive as I recall.

_________________
Doing stupid things for fun...

@Thread

Main Problem with 060 Cards where the price, you could get a 030/50 Card for 300€ a 040/25 a 040/20 Card was in the same price range but with a 060/50 CPU it was 350€ more. A500 users where known to be short on money and the transfer speed to ChipRAM was less than on a 030, there for people who where interested in games and didn't had a ZII/ZIII GFX card should not opt for those CPUs.

That's one of the reasons why I call this cards somehow 'not really suitable for Amiga', the bottleneck with ChipRAM access was to big, a whole new Amiga design would have been needed taking advantage of the speed possible with 040/060 CPUs.

@OneTimer1

Quote:

That's one of the reasons why I call this cards somehow 'not really suitable for Amiga', the bottleneck with ChipRAM access was to big, a whole new Amiga design would have been needed taking advantage of the speed possible with 040/060 CPUs.

The QuickPak A4000T licensee manufacturers at least produced a new accelerator card 4060T design for the A4000T that supported EDO Ram. The only other cards that I now supported this advancement in faster Ram design were the SX32 boards from DCE and possibly the Phase 5 Cyberstorm Mk3/PPC cards? Commodore never pushed advancements and would have never have got around to bundling 060s with Amigas even if they had survived! DEC Alpha chips with Hombre would have taken over before that!

Last edited by BigD on 18-Nov-2024 at 03:35 PM.

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

@Karlos

Quote:

A DMA SCSI interface for fast access to storage would've been a real value add, but it was already insanely expensive as I recall.

Not just price, but a (good) SCSI chip of that vintage would have also meant extra power and heat. For what? Your not gonna put SCSI drive into the case and there was no clean way to route an external plug (at which point you could have just bought a sidecar controller).


Even earlier before 040/060 RTG and so on you could buy A500 version of FFs, SCSI/IDE/OMTI HD controllers and 030 cards.
All at just slightly higher prices and all clearly selling in far smaller numbers than the A2000 counterparts.

Sometimes common sense just isn't that common....

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

@pixie

Quote:

pixie wrote: @Hammer Say you buy an 68030 for your amiga 500, is there any reason thereafter for you to upgrade to an Amiga 2000?

DKB Wildfire

_________________
B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC

@kolla

https://amiga.resource.cx/exp/wildfire
Ok, but from I saw that was launched in 1996... I was thinking by the time first accelerators were done for a500 until a600 was released. Because I guess the issue from commodore was that those accelerators were seen as lost Amiga 2000 sales.

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

@pixie

Quote:

Say you buy an 68030 for your amiga 500, is there any reason thereafter for you to upgrade to an Amiga 2000?

There are professional reasons to purchase an A2000 e.g. Video Toaster or Opal cards, extra SCSI and 'etc'.

This is different from improving texture-mapped 3D gaming performance use case.

I don't purchase Matrox broadcast products for my gaming/blender/Unreal Engine 5 PC.
I don't use SAS hardware RAID card on my gaming PC which is different from my other Xeon server's use case. I do have 28 cores Xeon acting as a file server with 10Gbit NIC cards.

The use case for Amiga 2000's multi-symmetric expansion slots mirrors the modern Xeon/Epyc workstations with multiple symmetric PEG (PCie 16X) slots. A desktop and gaming PCs only has one active PEG (PCIe 16X) slot and it would be divided into two PCIe 8X slots.

Amiga 500 has one active Zorro I edge connector, but the mass-produced wedge Amigas didn't evolve. Amiga 500's one active Zorro I edge connector concept is just a step away from mainstream AMD B650 or Intel B660's single PEG slot concept.

By 1997, AGP was released a single high-performance slot for gaming PCs while workstations had multiple symmetric high-performance PCI-X slots.

486 desktop and gaming PC has asymmetric VLB and ISA slots. The cheaper 486 boards have very few VLB slots.

The evolved Amiga 500 concept for expandable 3D gaming Amiga would be a single high-performance Zorro III slot with one lower-performance Zorro II slot or one local CPU slot.

Commodore didn't evolve the A500 concept into a 3D gaming Amiga, the basic idea for low A500 vs high A2000 is okay.

Last edited by Hammer on 19-Nov-2024 at 12:34 AM.
Last edited by Hammer on 18-Nov-2024 at 10:41 PM.
Last edited by Hammer on 18-Nov-2024 at 10:35 PM.
Last edited by Hammer on 18-Nov-2024 at 10:27 PM.

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

@OneTimer1

Quote:

OneTimer1 wrote: @Thread Main Problem with 060 Cards where the price, you could get a 030/50 Card for 300€ a 040/25 a 040/20 Card was in the same price range but with a 060/50 CPU it was 350€ more. A500 users where known to be short on money and the transfer speed to ChipRAM was less than on a 030, there for people who where interested in games and didn't had a ZII/ZIII GFX card should not opt for those CPUs. That's one of the reasons why I call this cards somehow 'not really suitable for Amiga', the bottleneck with ChipRAM access was to big, a whole new Amiga design would have been needed taking advantage of the speed possible with 040/060 CPUs.

Reminder, A500's 16-bit Zorro 1 edge connector and A2000's 16-bit Zorro 2 slot are similar.

A2000 is just A500 with Buster switch chip and Zorro 2 slots.

A500 with 030 @ 40Mhz or PPS 040-28 accelerators couldn't join the 256 color gaming, hence the majority of A500 users have purchased gaming PCs.

There's a small install base for A1200, A2000 (16-bit), A3000 and A4000.

During 1992, there's major supply issue with A1200's 44,000 units while A600 has "1 million" units production.

If AGA sales numbers are treated as worldwide,
44,000 (the UK has 30,000 during its launch),
100,000 (AF50, Sep 1993),
170,000 (AF56, Feb 1994),
166,000 (CD32, Commodore US president, Jan 1994),
7,500 (Germany's A4000/030),
3,800 (Germany's A4000/040),
Total: 491,300 AGA units, mostly in 1993 year.

There's another 65,000 CD32 boards for Commodore Canada's A2200-1 and A2200-2 partnership with Amitech that were locked up in the Philippines warehouse.

For profitability with existing debt servicing, CD32's production target goal was 400,000 units (cite: Commodore - The Final Years).

For the 1993 year, Commodore must produce, ship and sell about 681,300 AGA machines level.

Last edited by Hammer on 19-Nov-2024 at 01:37 AM.
Last edited by Hammer on 19-Nov-2024 at 01:17 AM.
Last edited by Hammer on 19-Nov-2024 at 01:06 AM.
Last edited by Hammer on 19-Nov-2024 at 12:55 AM.

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

@Hammer

Quote:

Hammer wrote: There are professional reasons to purchase an A2000 e.g. Video Toaster or Opal cards, extra SCSI and 'etc'.

Not just 'professional' reasons. A base model A2000 with single floppy drive wasn't that much more than an A500 if you were thinking of expanding it. Several of my friends had one, and they were were only using it for hobby stuff (playing games, chatting online etc.). In 1989 a B2000 with RGB color monitor sold for £875 in the UK, while an A500 with the same specs sold for £598. So you paid an extra £277 to get a much more expandable machine.

Quote:

This is different from improving texture-mapped 3D gaming performance use case.

During the Amiga's commercial lifetime texture-mapped 3D was simply a matter of more mips, which was good for other 'use cases' too. Only problem is most Amiga owners were barely able to scrape up enough to buy an A500 with modulator, and didn't expect games to cost more than the price of a blank disk. This meant there wasn't a big enough market to justify targeting higher spec machines. By the time texture-mapped 3D became desirable enough to incentivize fans into upgrading (ie. Doom in 1994) the Amiga was already dead.

Quote:

Amiga 500 has one active Zorro I edge connector, but the mass-produced wedge Amigas didn't evolve. Amiga 500's one active Zorro I edge connector concept is just a step away from mainstream AMD B650 or Intel B660's single PEG slot concept.

There were many PCs before then that had few or no free slots too, particularly around the time of the A500+ and A1200 when slimline PCs were popular.

Quote:

486 desktop and gaming PC has asymmetric VLB and ISA slots. The cheaper 486 boards have very few VLB slots.

Yep. Usually both slots were used up by the I/O card and graphics card. Machines with 3 slots often couldn't actually drive them all reliably. VLB was a bad design mechanically too. My supplier hot-glued the cards in to stop them from popping out during shipping.

Quote:

The evolved Amiga 500 concept for expandable 3D gaming Amiga would be a single high-performance Zorro III slot with one lower-performance Zorro II slot or one local CPU slot.

Not necessary. Motherboard I/O is plenty fast enough for 3d games. The problem is rendering speed, and lack of colors if you really care about that. The answer is the same as for consoles of that era - a 3D graphics chip on a card with its own video output (come to think of it that's what PCs did too, with the Voodoo card). But once again the market just wasn't big enough to justify it. By the time 3d games were a big deal fans were switching to PCs anyway - and Commodore was already gone.

But would fans have stayed with the Amiga if it was more expandable? No because they were also buying PCs to get Microsoft Word etc. The time of diversity and innovation was over. Consumers settled on the platform with 90% market share for good reason - who wants choice when it just leads to incompatibility?

Quote:

Commodore didn't evolve the A500 concept into a 3D gaming Amiga

They were certainly thinking about it. But before 1993 3D (texture-mapped) games were a novelty, and even after that it was just another genre. Myst, which has pre-rendered static graphics, was ported to the PC in 1994. It was the best-selling computer game in the United States for over 4 years. Myst was ported to the Amiga in 1997, and only needed an AGA (or RTG) machine with 030, 8MB RAM and a CD-ROM drive.

However me and my friends were more into strategy games like The Settlers and Dune II, and simulators like A10 Tank Killer. An A1200 with fast RAM and a hard drive was fine for these 'use cases'. I did get hooked on Tomb Raider in the late 90's, but all the other 3D games I tried on the PlayStation were meh. The PlayStation's rendering wasn't that great either. Textures got very distorted when viewed close up, and there were often disturbing gaps between the polygons. It got away it mainly because players of the time weren't expecting much.

But what about today? Who has a yearning to play modern 3D games on their Amiga? Not me. This morning I spent 6 happy hours programming in 68k assembler, and the A1200's 50MHz 030 did a fine job of it as always. I even hopped on the web for a few minutes to do some research and download stuff (faster than powering on the PC and waiting several minutes for it to boot up!). Plus I had my collection of MOD files continuously playing on the big stereo speakers to keep me in the mood. Why would I want more than this?

Last edited by bhabbott on 19-Nov-2024 at 07:28 AM.

@BigD

Quote:

BigD wrote: Quote: That's one of the reasons why I call this cards somehow 'not really suitable for Amiga', the bottleneck with ChipRAM access was to big, a whole new Amiga design would have been needed taking advantage of the speed possible with 040/060 CPUs. The QuickPak A4000T licensee manufacturers at least produced a new accelerator card 4060T design for the A4000T that supported EDO Ram.

On a really well integrated system you wouldn't had the need to put the RAMs on the accelerator board, the system would have been fast enough to hold them on the main board and the read/write performance on ChipRAM should have been higher than on ZIII.

Putting it on the CPU board is just a sign of poor motherboard design.