OS4 rollout schedule

[HyperionMP]

That about sums it up, yes.

I honestly cannot see what is so terrible about this.

We don't want to be involved in shipping end-user copies to dealers for several reasons:

1. This would put a strain on us as we would need to put into place an infrastructure for shipping, warehousing and distribution which goes beyond what we have now for selling our games.

Obviously a manufacturer already has this infrastructure in place.

2. We would end up at the losing end of all the piracy. At least with the OEM scheme we'd be reasonably sure that every board sold for the explicit purpose of running OS 4, would actually have a paid for copy of OS 4.

[HyperionMP]

The answer is obvious from 1) the fact that Eyetech also sells boards in the Linux market 2) my post above.

Only customers that explicitly want OS 4 will need to pay for it.

[HyperionMP]

Why bundle?

For the reasons outlined above: to cut down on piracy and to reduce distribution related expenses. Again, we do not have an infrastructure in place to handle distrubution on this scale. You cannot compare an OS release for several hardware platforms with an AmigaOS game of which maybe a few hundred get sold.

The hardware certification as you call it is also necessary to ensure proper cooperation and disclosure of the hardware manufacturer.

How does the firmware operate especially when it contains proprietary extensions? We also require chipset documentation of all onboard chipsets etc.

[HyperionMP]

>Ben, you have lost me now. Taking this statement >alone, it appears that Hyperion will give the OS to >the mainboard distributors for a bundled sale, and >take royalties for sold units? That would be a >perfectly normal scheme. Until now, I was under >the impression that Hyperion plans a certified->hardware licensing scheme (on top of selling the >OS and taking royalties).

No. That was never the intention.

It must be recognised however that such a scheme also means that we need to consider carefully whether or not a platform is actually worth supporting or not from a purely economic standpoint unless the producer is willing to guarantee a certain number of sales.

Currently, with the Pegasos I, I cannot see it. It is a discontinued product of which only 600 units were built, quite a few of which are still in the dealer-channel or were handed out for free.

Many of these Pegasos owners might not at all be intrested in OS 4 and the upper limit for sales is 600.

People point to the Cyberstorm PPC version.

The Cyberstorm PPC version of OS 4 was developed for several reasons.

1) it provides a platform for development with all the custom chipsets still in place. Results are obtained faster this way. Proof of concept is important due to the pervasive but understandable skepticism.

2) a lot of people paid good money for this hardware and these people are our customers: they bought games like Shogo, Heretic 2 etc. We want to support them.

3) there is no point in replacing a small PPC market (Cyberstorm PPC and BlizzardPPC) with an even smaller market of more powerful machines (A1).
This will not benefit software developers wishing to target OS 4.

[HyperionMP]

>What is the "minimum number of sold copies" >required for a platform to have OS4?

It is very hard to answer this question in abstracto.

This would depend on the amount of work that would need to go into developing a version of OS 4 for a specific hardware platform.

When the used chipsets are significantly different, this would entail rewriting the device drivers again: USB support, IDE, ethernet, floppy etc. and even simple things like RTC support.

Now obviously the hardware producer could do this inhouse to some extend thus cutting down on our development cost.

It would obviously also depend on the price we would charge per unit but it is clear that you cannot ask for 300 USD per copy otherwise you won't find anybody willing to buy it.

[HyperionMP]

You are obviously not familiar with the cuurent state of AROS.

One of the main problems is that they are shooting for 3.1 compatibility whilst we were aiming for 3.9.

Moreover, some of the AROS work is not entirely compatible with OS 3.1. DOS and Intuition come to mind.

[HyperionMP]

What I meant in my previous post, was that the 68k ASM stuff that you had to rewrite in C, could have been taken from the AROS code to be used as a base and modified until it suited you.
As matt points out above, maybe you don't want to be bound by the AROS licence for the parts you could have used for free, which is fair enough.

---

Believe me, we looked at that initially, especially for DOS.

Gary Peake introduced us to Aaron and we were on friendly terms.

The license wasn't so much of an issue but rather the fact that subtle but sometimes far-reaching incompatibilities existed and still exist to this date.

[HyperionMP]

Stormtoopers were clad in white armor.

Besides, if the analogy holds, you know is going to prevail.

[HyperionMP]

Can you maybe start another thread then?

Quite frankly, I've lost track.

[HyperionMP]

Stormtoopers were clad in white armor.

Besides, if the analogy holds, you know is going to prevail.