Software companies and small markets

[mikeymike]

In the past, the Amiga has been beset with the issue of how to get Amiga users to upgrade their hardware in order to use new software.  Occasionally a software maker would pull this feat off with an exceptional game and the upgrade not being too expensive, but certainly one particular issue was Amiga users upgrading to have hard disks in their machines.  This would have seriously helped the Amiga market in its later days when support began to ebb.

Ok, this isn't a historical piece I'm writing.  I'm wondering how this situation might affect the future Amiga hardware/software markets, assuming that A1/OS4 gets a good start, and what lessons can be learnt.

Does the history really help with working out a decent strategy in the future?  To a certain extent no.  Particularly regarding hard disks as they're now very cheap.

Assuming that x is a relatively new hardware product:

The first problem, for the users, is that they don't know whether upgrading to x is worth it and what for, and whether there will be continuing software support that won't run out before the user feels they've got their "money's worth" from the hardware.  All of this comes under "do they really need it?".

My solution (gifted with hindsight of course, always useful), would be to have a consortium of software makers, as well as a couple of hardware makers.  If a particular piece of hardware (such as an accelerator) would be particularly beneficial for them to take advantage of, then they all commence (or upgrade) projects to take advantage of the hardware, and try to aim the releases relatively close together, announcing way in advance their hardware requirements and featuresets of the new software.  And also the hardware manufacturer offering a discount price for a while, or a pack of products altogether for users to buy.

I've been thinking how hardware upgrade paths and the new PPC Amiga market might evolve.  The easiest way is to identify where the PPC market lags behind currently (that would be: raw processor speed and FSB speed.  It will be: hardware features that become standard on x86 mobos, firewire? USB2, gigabit ethernet? serial ATA, SCSI for super high-end mobos? PCI-X?).

(I haven't included AGP3.0 as PCI-X appears to be destined to replace AGP, which I find surprising but not unbelievable, but anyway)

Then it needs to be worked out what markets are going to particularly need "new Amiga" users to upgrade:

 - Games is the obvious one.  CPU, FSB, possibly serial ATA are the most important factors then (or in the future)

 - connectivity to external devices like video cameras, scanners (USB2? Firewire? SCSI?)

 - in a few years I guess getting *new* (as in cutting edge) hardware is going to be strictly PCI-X, which means that OS support for PCI-X needs to at least start being considered now.

I don't think Amiga users are going to get into the habit of upgrading because they want to [primarily], while PPC-related hardware remains over-expensive and underpowered, so it is the software makers that are going to have to really sell this kind of thing to the Amiga community, rather than just hoping they'll upgrade at a time convenient for the software makers.

[mikeymike]

What, no takers? Have I hit the nail on the head, or so obvious it didn't need saying, or what?

I thought I'd chuck it up (with a few edits) as an article on the amiga section of my website as well :-)

[Yogi27]

I think part of the problem was the custom slots and such.
Our hardware was always specialized and hard to upgrade.
I think with the AmigaOne we have gotten away from that
to some extent.  An example would be, I love the Amiga
keyboard.  But, now I can get a wireless keyboard if
I want one.  The same with the mouse.  Not to mention
expansion cards that had to be Zorro compatible.  I
personally think this might help our situation some
with user upgrading.  I can to the computer store
by my house and get upgrades (assuming there is
drivers for it.), instead of having to order it from
Europe or somewhere far away.

Yogi27

[mikeymike]

But hardware manufacturers (in order for them to stick around commercially) are particularly going to want you to upgrade your A1 more often then when something goes permanently wrong, the question is, how do they do that?  Wintel does it by producing a resource-hogging OS, we don't want it done like that with AmigaOS, yet we want the same kind of power/value for money as x86 people get.

Well, quite a significant percentage of people want that, as we want gaming to come back to the Amiga.

[Wolfe]

(I haven't included AGP3.0 as PCI-X appears to be destined to replace AGP, which I find surprising but not unbelievable, but anyway)

Since everyone is now using AGP, switching would make all the smaller players play catch-up.  Picture the AOne after making a bunch of Mobo for their small market witch isn't going to sell fast anyway try to R&D a new Mobo when the first ones haven't had a real impact yet.  Making it hard for any new competition to ge a foothold on any turf!  :rtfm: Sun Tzu

 - Games is the obvious one.  CPU, FSB, possibly serial ATA are the most important factors then (or in the future)

Definitely!  Unfotunately I don't think any new CPU/OS combo entering this market is going to make much impact due to "Lack Of Vision" :-(

Why:  To much looking back and not enough looking forward.

The computers of the 21st century - Feature Rich, Smaller, Low Power, Low Heat, efficient and  easy to use.  :-D  That doesn't mean slow.

But if you follow the history of the PC - it's headed the other direction.  Empire's only last so long and then bust.  The next best idea rules and it is generally different from its predecessor.  :-D  History!

I don't think Amiga users are going to get into the habit of upgrading because they want to [primarily], while PPC-related hardware remains over-expensive and underpowered, so it is the software makers that are going to have to really sell this kind of thing to the Amiga community, rather than just hoping they'll upgrade at a time convenient for the software makers.

Lack of vision once again.  

Amiga One Mobo should have been an ITX form factor with all the Bells, Whistles & Goodies on board.  

Amiga OS 4 should have been written to run on it and not the back-ward tech.    Too many complications in developing it - "still not ready for prime time"?

Why?  Because AOne would then have been able to be a "Set-Top-Box", "Small Office Server", stacked for "Full Server", "Game Console", and the obvious everyday personal computer.  Not to mention dozens of other uses attracting developers from near and far.   :-(

LACK OF VISION

Thinking 90's insteat of 2010.

SUN TZU

[mdwh2]

mikeymike wrote:
But hardware manufacturers (in order for them to stick around commercially) are particularly going to want you to upgrade your A1 more often then when something goes permanently wrong, the question is, how do they do that?  Wintel does it by producing a resource-hogging OS

I think it was true that the requirements of Windows encouraged people to upgrade during the 90s, but I don't think it's true these days. Any old CPU, hard drive, graphics card and so on is more than enough for Windows - people upgrade these for other reasons (eg, games, mp3s). The biggest remaining constraint for Windows is probably RAM, but this is getting pretty cheap too.

So I'd hope that people are encouraged to upgrade by having good software which takes advantage of new hardware, rather than it simply being wasteful and bloated.

[mikeymike]

Lack of vision once again.
Amiga One Mobo should have been an ITX form factor with all the Bells, Whistles & Goodies on board.

Why?  Mini-ITX is primarily for a small physical footprint, not performance, not expandability.  Haven't Amiga users seen enough Amigas with poor upgradability potential?

IMO:

I don't think most Amiga users want a "set-top box", "small business server", "full server" or "game console :-)

The year 2010 will not see everyone using "appliances" rather than general purpose computers.

[mikeymike]

Another factor that "encourages" people to upgrade is better hardware support in Windows.  The next new factor that MS want to introduce is that people will have to upgrade in order to do anything "clever" on the Internet, with MS's visions of "secure content", DRM, etc.  MS want to own the server market, and if they manage that, the entire process of client to server across the internet will become proprietary.

[Doobrey]

mikeymike wrote:
But hardware manufacturers (in order for them to stick around commercially) are particularly going to want you to upgrade your A1 more often then when something goes permanently wrong, the question is, how do they do that

 But the thing is, there won`t be many companies producing A1 specific add-ons.
 Apart from Jens and his PCI flipper cards, everything else will be bog standard parts that you can pick up from PC World etc...
 I think it`s more a question of what the  remaining companies like Elbox etc can do to stay in the Amiga market once people stop spending money on the classic amigas.
 

[mdwh2]

Doobrey wrote:
 I think it`s more a question of what the  remaining companies like Elbox etc can do to stay in the Amiga market once people stop spending money on the classic amigas.
 

I think there's still some room for Amiga-specific development. For example, possibly accelerator cards for faster PPCs (I forget what sort of CPU upgrades the AmigaOne or Pegasos use - whether it's a socketed CPU, or something else). Or alternatively hardware specifically designed with AmigaOS in mind - for example, have a look at http://www.villagetronic.com/ - whilst they seemed to have dropped support for the Amiga, they seem to be happily selling graphics cards for the Mac, even though it uses standard PCI cards. Presumably they differentiate their products from standard PC ones by adding things like software, drivers and support.

[yoodoo]

Alan's comments at AmiWest about new mobo designs were very enlightening. Basically, he believes there is no point changing the design every few months as it will shrink the potential market (people hanging on for the next design) and prevent economies of scale.

New graphics cards, USB2, Firewire, Serial ATA etc could all be achieved using PCI cards + drivers. Perhaps a 3 year mobo cycle would be decent, with yearly updates on processsor cards?

There are only 2 designs needed: desktop ATX (which we have) and small footprint all-in-one jobby for embedded stuff/ lifestyle computer, perhaps shoehorned into a laptop size thing if it fits.

Personally, I'm very happy with my (outdated) A1G3 which runs at 733Mhz. Cant wait to put OS4 on it.

[yoodoo]

Historically, Amiga users are quite happy to pay for hardware at seemingly high prices, but perceive less value in software (hence all that piracy).

The key for software vendors will be to maximise sales by hitting AOS, MorphOS,AROS and Amithlon too if possible. Someone might actually cough up the cash then.

The key for the OS developers is to attempt to provide APIs that are sufficiently compatible to make the software developers job easy while still differentiating their products. Compare the different Linux distros - they all run (pretty much) the same software but have different audiences.