* Protected system files with a password required to bypass said protection. This isn't a full "multi-user" system by any stretch, just one password for a special case of operations. Protection is a hash table stored in ROM and masks all IO operations and is thus transparent to the actual filesystem.

* Separation of "dangerous" and "not-dangerous" Preferences, and protect settings and ENV files using the same mechanism as above.

* Enforcer enabled on all MMU machines to cleanly and safely shut down bad software and/or provide debug hooks when debugging is available.

* A proper firewall. Keeping with the Amiga mentality, this should be a "router on a chip" that also provides networking offload, etc, thus, not taking from the main CPU.

Thierry Atheist wrote:

What was the DOWNFALL of their attempted re-entry into the mainstream computing scene was DISGUSTINGLY expensive hardware, and NOTHING ELSE but THAT!!!!!

Absolutely agree. Power PC was the wrong choice.

Thierry Atheist wrote:

When Hyeprion came on the scene, they were a decidedly SOFTWARE company. They had to go with a CPU that didn't exist --->advanced 680x0 CPUs, or PPC, where there was a history established, or cold start on 80x86 CPUs where endianness was a BIG issue too.   They chose PPC. I was 100% opposed to that, but as there was NOTHING ELSE AVAILABLE, I nonetheless, supported the idea.

Have to stop you here. No other alternative? You want a big-endian processor that's dirt cheap -- ARM. By 2001, ARM had a 76.8% hold on the 32-bit RISC processor market. They weren't as fast as the PowerPC's of the time (about 200MHz vs 800MHz), but the cost savings alone would have been unimaginable, and the CPU deficit should have been offloaded with other custom chips (or even additional ARM cores running dedicated ROMs -- like one that just handles all low-level IO, and one that handles Networking and firewall), etc.

It would have also solved all the technical problems of the PowerPC and within a few years, ARM would catch up and beat PowerPC. Not that Eyetech or Hyperion couldn't have licensed the core from ARM and made their own improvements (e.g., Digital).

Even **IF** ARM was not considered, the ColdFire dates back to 1994 and we had the V4 architecture by 2000's. The V5 was cancelled when Freescale saw the writing on the wall -- everyone was adopting ARM because it was stupidly cheap, flexible and very low power (which is important in the gradual shift to mobile computing). Although, ARM would have still been the better choice.

Thierry Atheist wrote:

Amiga Inc. and Eyetech and Hyperion should have realised that the BULK (80% +) of AMIGA users, what they REALLY wanted was OLD AMIGA made ANEW.

Assuming they had to start development a couple of years before their release, say 1998, there wasn't a 68000 option other than the ColdFire. The V4 was available under NDA, so if they were planning for a release in 2000, that would have worked out okay. It isn't a true 68K processor, but I imagine it would have been a thousand times easier to port the OS to ColdFire than it would to PowerPC or ARM. And while "emulating" the missing 68K features was a huge hit on performance, it's less so than emulating the 68K outright on a PowerPC or ARM. As long as critical core OS functions were recompiled to not use those instructions, things like memory copy, list handling, intuition, etc., would all be native and run eight times as fast as legacy 68K code.

As with the ARM comment above, they could then offload "work" to other custom chips -- a custom Ethernet CPU would be a big one. This way, the slower core CPU speed has less real impact on performance. I would have also integrated something akin to a PowerVR on board along with a simple SVGA and DAC (prob. Cirrus Logic) chips and have built-in support for these two chips. If people want something better in the AGP slot, then it's their choice to deal with slower boot times, etc.

Perhaps if there were more people buying them, Motorola/Freescale/NXP wouldn't have left them die on the vine.

It is exhausting reading all of that!! I quit. When every time I see post by you I will simply just ignore it. I am just drained with all capitalization and exclamation mark and shouting!

Yes I was posting on Amiga.Org. I am not banned from there.

You're sitting a "1". Got a "somewhere else" that I was banned from?

I fully agree, but how then they will squeeze some more money from users.
Some people still lives very good selling residues of Amiga

So there it is, Steve, evidence that it's the equivalent of you in the video I posted a link to. Since that's your desire, you are obsessed with continuing your anti-hyperion commentaries. Pathetic.

motorola totally failed when they lost Apple, because they just weren't capable of keeping up with intel. They thought RISC was the answer, but it was really their prices that did them in.

If there are to be real updates to OS3.1 they are going to come from the Thor/Olaf duo working with Hyperion. They have the source code updated and compiling. A lot of work went into that. I don't see any other party with the capability.

If something comes out of the Hyperion's 3.1+ development that benefits Amiga users in general then that works for me. If beyond that something of benefit comes to the Vampire product line then that is great as well.

Let's just see how this plays out.

And yes, IT IS a MARVEL that so few people involved COULD make an OS with a guideline, but effectively from scratch!!!

Hyperion couldn't even afford to hire most of the coders FULL TIME! What has been delivered IS A MARVEL!

I don't know what you've achieved in your life, but I hope someone who knows spits on it too.

If they sell enough of these, it will more than likely lead to them making a 68080 enhanced version.

So, you gotta buy them.

1.  The Mac had already undertaken this transition, so there was already a proven way to interface PPC to a 68K system.

2.  ARM chips weren't actually all that powerful by 1997.  A typical ARM of 1997 was the ARM710T or 720T;  36-60 MIPS rating.  By comparison, a PowerPC 603 133-200 mhz = 188-283 MIPS.  The competition: a 200 MHz Pentium MMX was ~ 276-375 MIPS.

3.  ARM would continue to fall behind -- ARM's own computers stopped selling years previously and the mobile market wasn't hot yet so ARM was forced to reduce R&D heavily to avoid bankruptcy, while x86 investment (and PPC) was accelerating. ARM 920T from the Year 2000 would finally hit 200 MIPs, while x86 was pushing ~ 9x that throughput.

4.  The IBM-Apple-Motorola Alliance was pouring tons of money into PPC development circa 1995-1998, meaning there was confidence PPC would continue to grow and scale. 

PowerPC was probably also a better choice at the time than the MIPS path that Commodore wanted to go with Hombre.. 

Now, to figure out whether to get a Vampire V4 for my Amiga 500 (which is unknown condition) or wait for the 1200 version..

Despite protests to the contrary, having Hyperion as an active participant in the Vampire realm is definitely a good thing - regardless of what you think of the company or their products.

- claps - bravo! I agree 100% with Steve Ferrell!!!

I thought there was something weird in his posts O_O

Its everyone list of disapointment since OS4 would be quite useful if they kept the promises. Add the TW browser.

Its also a shame every hardware has to go over beta OS phase and drivers waiting. And is unfair to users who always kind of prepaid the OS development.