25-Apr-2024 02:05 GMT.		 UNDER CONSTRUCTION
Anonymous, there are 20 items in your selection
[Rant] PEGASOS'ANN.lu
Posted on 24-May-2001 21:04 GMT by Christian Kemp20 comments
View flat
View list
.jon writes: I am a wondering why the PEGASOS is 'pronounced' as 'modular'. It is not. It is just a standard MicroATX board. Now, if you call PCI slots 'modular' then it might be okay for you. For me it isn't. When the PEGASOS was announced I was happy about the word 'modular'.
Because, so I hoped, we would get the long awaited modular system.

What we get is an (admittedly very nice) DUAL PPC MoBo (and I will get one if I do not find an alternative). I have been searching for a MoBo that has Firewire in addition to IRDA and Ethernet and all the other standard plugs onboard. I did not find one in neither CPU world, something was missing, either it had Firewire onboard but was missing the LAN or similar) so I was really lucky when I heard of the PEGASOS.

But: What is the userbase for such a board ? It is quite universal, true. But, regarding the fact that it has IEEE1394 (FireWire) onboard I would target it in first place to Video people, who, in addition, will benefit of dual CPU power. Plans I have are somewhat similar. I'd personally define the PEGASOS as a multimedia mobo.

Today I read about the release of the final IEEE1394b specification, which, at its best, means FireWire at four times the current max-speed. It will require new cables. New plugs. But remain compatible to the old signals in a special compatibility mode. First products are awaited in late summer/Q4.

So, when I buy a PEGASOS system ('cause, as it seams, you will not be able to buy the mobo solo (according to the interview with bplan on Amiga Future, it was only named together with "licensed system" and "pre-built system") I will buy an outdated Firewire I/O. Sure, not much outdated, just a little. It is not of a big concern, since Firewire itself is quite new. But where is the modularity ? Exactly this would be a case for modularity. If it would be modular, the Firewire could be replaced easily. Can it ? I guess not. Sure, I might be in error. Then I appologize. If not, I hope this will be read or told by/to someone responsible. Maybe the final Pegasos will include already the new standard ?

PCI/AGP is just not what I would name 'modular'. Then every and any PC is modular. Why name it than as a special feature ?

And I hope for a bit more fanatsy in future PPC mobos, by whoever they might be designed. World is changing... PPC is a bit 'special'. It is not an everyday's CPU. So I hope the underlying hardware will be special as well.

Since I am at it: My ideal PPC mobo would look like the PEGASOS.

  • 2 more PCI slots
  • IEEE1394b (even better: make the IO ports some module)
  • modular DSP option ?!
  • (modular) onboard U2WSCSI (option)
  • wake-on-lan/modem
  • special port for output of system diagnositcs and perhaps other data to LCD

Why not put all the extras on PCI ? Well, the PPC can be suficient at low speeds with less cooling. If I get a real modular mobo, that has anything onboard, I might even leave the PCI cards out and get a perfect multimedia/homebox. Noiseless, powerfull.

P.S.Christian, what about another PREVIEW button after the first preview ?
PEGASOS' : Comment 1 of 20ANN.lu
Posted by Joe "Floid" Kanowitz on 24-May-2001 22:00 GMT
This is half a wild guess, but I've always assumed it referred to their chipset's extensibility for multiprocessor situations.
PEGASOS' : Comment 2 of 20ANN.lu
Posted by Anonymous on 24-May-2001 22:00 GMT
Ever thought about taking the somewhat advanced step of ASKING bPLAN instead of ranting here?
I guess that would have been too sensible, though, eh? :-(
PEGASOS' : Comment 3 of 20ANN.lu
Posted by Fran on 24-May-2001 22:00 GMT
<Since I am at it: My ideal PPC mobo would look like the PEGASOS.
<
<2 more PCI slots
<IEEE1394b (even better: make the IO ports some module)
<modular DSP option ?!
<(modular) onboard U2WSCSI (option)
<wake-on-lan/modem
<special port for output of system diagnositcs and perhaps other data to LCD
Hey!!! why not AMD hypertransport ports, amiga should be revolutionary to survive, well the OS is but why old standarts, why not make a MoBo with the future standarts so the Amiga will be again a dream powerfull machine.May be is better make a cheap line of computers with current AmigaONE specs and make an profesional version
The professional MoBo should have:(my dream config)
-1 AGP x4 (or more)
-4 hypertransport spansion slots
-USB2 and firewire B
-ethernet 10/100
-modem for digital and analogic connections
-UWSCSI3
-PC266 or better memory
-A CPU modularity slot like A4000; and may be
a spansion slot daughterboard like a4000 so
you can chosee a daugtherboard with pci or
hypertransport or wathever thing that anybody
invent in future.So you don´t have to change
motherboard to use newer tecnologies
PEGASOS' : Comment 4 of 20ANN.lu
Posted by Fran on 24-May-2001 22:00 GMT
<Since I am at it: My ideal PPC mobo would look like the PEGASOS.
<
<2 more PCI slots
<IEEE1394b (even better: make the IO ports some module)
<modular DSP option ?!
<(modular) onboard U2WSCSI (option)
<wake-on-lan/modem
<special port for output of system diagnositcs and perhaps other data to LCD
Hey!!! why not AMD hypertransport ports, amiga should be revolutionary to survive, well the OS is but why old standarts, why not make a MoBo with the future standarts so the Amiga will be again a dream powerfull machine.May be is better make a cheap line of computers with current AmigaONE specs and make an profesional version
The professional MoBo should have:(my dream config)
-1 AGP x4 (or more)
-4 hypertransport spansion slots
-USB2 and firewire B
-ethernet 10/100
-modem for digital and analogic connections
-UWSCSI3
-PC266 or better memory
-A CPU modularity slot like A4000; and may be
a spansion slot daughterboard like a4000 so
you can chosee a daugtherboard with pci or
hypertransport or wathever thing that anybody
invent in future.So you don´t have to change
motherboard to use newer tecnologies
PEGASOS' : Comment 5 of 20ANN.lu
Posted by David Scheibler on 24-May-2001 22:00 GMT
You can find a new interview with bplan here:
http://www.amigafuture.de/eng/eng_aktion/eng_interviewbplan.html
He speaks about the price and OS4 compatibility and also he says that
currently Linux and a MorphOS alpha version are running on Pegasos.
PEGASOS' : Comment 6 of 20ANN.lu
Posted by Amifan on 24-May-2001 22:00 GMT
In reply to Comment 5 (David Scheibler):
So david, the Pegasos might be compatible with AmigaOS4.0 afterall, it's a wise descision of bPlan. But I'll like to know what CPU i'll get for just 1000 EURO
PEGASOS' : Comment 7 of 20ANN.lu
Posted by David Scheibler on 24-May-2001 22:00 GMT
In reply to Comment 6 (Amifan):
As you can read it's up to Amiga if they support the Pegasos. I don't
know if Amiga Inc. will develop for Pegasos with the consequence that
maybe the AmigaOne Board by Amiga's "number one partner" Eyetech
won't sell good.
I don't know what CPU you will get (maybe G3 333Mhz?) but concerning
the other hardware I think that you will get more RAM and HDD as
announced in the interview (64MB and 10GB) in Q3/01.
PEGASOS' : Comment 8 of 20ANN.lu
Posted by Anonymous on 24-May-2001 22:00 GMT
In reply to Comment 2 (Anonymous):
I think what .jon wrote was legit. I've asked some of the same questions without a response from bPlan. Sometimes it takes a waking up from some people to realize that specific situations could be a problem.
PEGASOS' : Comment 9 of 20ANN.lu
Posted by Ralph Schmidt on 24-May-2001 22:00 GMT
First some background about PCI...
A PCI bus can handle *10* loads.
A load means (1 load for a socket (PCI) and 1 for the pci device), so
a pci *card* takes 2 loads. With 1 load by the pci bridge you have
9 loads left4 slots max if you keep to the specs.
Conclusion...you can put more pci devices on the motherboard than into
a socket system.
Furthermore the first bplan machine will use superio chips which handle
the major IOs in one chip...(USB;Ethernet,AC97,Serial,Parallel,Floppy,
keyboard and so on is part of that chip.
Firewire is another pci chip on the motherboard, so 7 loads are left
for 3 33Mhz PCI slots.
Furthermore the ppc bus chipset bplan uses has a 2nd pci bus which
handles 1 66mhz slot and one *real* AGP-2x slot with sidebanding and
fast writes.
The PCI/AGP controller handles writecombining which means it puts
single beats(32bit-64bit) writes into a cacheline to *burst* to pci
and agp units.
That`s a must to get performance on these buses as the ppc itself
can`t do that itself and that`s only possible with real ASIC buscontrollers
for complexity reasons of the needed buffering.
Furthermore it would also be possible to add a pci to pci bridge onto
the 66mhz pci bus to expand the slots if necessary for special motherboards
designs.
So..now about IEEE1394b(the new Fireware). At the moment companies work
on the *first* implementation of chips for this *new* standard.
You shouldn`t expect such chips on end user cards/mbs before the end of
the year anyway and it takes probably another year until real consumer
products will make use of so you can connect them.
Furthermore it also exceeds PCI 33Mhz busbandwidth by a factor of 2x.
But if some day such firewire pci cards are available you can surely
put them into a pci socket and if some driver supports it you can
use it.
It doesn`t matter if firewire is already on the motherboard or not.
About DSPs...we really looked into some *special* audio solution last
year but there weren`t any audio dsps with a pci socket with *documentation*.
o The Motorola audio dsps have no pci bus...
o The Cyrrus audio dsps are windows only for example which means you don`t
get docs for them and have to live with the windows drivers.
Quite pointless i would say....
o wake on lan/modem is possible with the current chipset...just a matter
of sw support.
o a special port for system diagnostics would also be possible but well..that`s
developer stuff and they use serial output anyway.
Adding LCD would require additional costs for a not mission critical LCD
(I also had such an idea some years ago for the abox design:-)..but well)
It would also require special case support for the lcd.
About Hypertransport and AGP4x....yep..nice features but if there are no
PPC buschipsets yet which support this it`s quite pointless to "demand" it
as that`s *impossible* under the circumstances.
AGP 4x is beyond SDRAM bandwidth anyway and DDR only makes sense when the
PPC can use the bandwidth...
I hope that clears up some misperceptions people may have...
PEGASOS' : Comment 10 of 20ANN.lu
Posted by Joe "Floid" Kanowitz on 25-May-2001 22:00 GMT
In reply to Comment 9 (Ralph Schmidt):
First impression: "Hey, you're like, that MorphOS guy, aren't you?" [This is what happens when I hang out in the 'real' world for too long...]
>First some background about PCI...
>A PCI bus can handle *10* loads.
>A load means (1 load for a socket (PCI) and 1 for the pci device), so
>a pci *card* takes 2 loads. With 1 load by the pci bridge you have
>9 loads left4 slots max if you keep to the specs.
That's interesting, since that limitation doesn't seem to have held in x86-land... http://www.epox.com/html/english/products/motherboard/ep-8kta3.htm
This FreeBSD dmesg output seems to give a hint as to how that board rigs it:
pcib0: <Host to PCI bridge> on motherboard
pci0: <PCI bus> on pcib0
agp0: <VIA Generic host to PCI bridge> mem 0xd0000000-0xd1ffffff at device 0.0 o
n pci0
pcib1: <PCI to PCI bridge (vendor=1106 device=8305)> at device 1.0 on pci0
pci1: <PCI bus> on pcib1
...but even so, the rest of the output shows that every device in the system lives on pci0, integrated or not; pci1 is just the AGP implementation!
Few Wintel users will occupy all those slots, but I imagine someone tested that setup somewhere along the line... Inneresting.
>Conclusion...you can put more pci devices on the motherboard than into
>a socket system.
The above doesn't negate that point, even if it reveals something interesting about the state of PCI implementations in x86-commodityland :)
Y'know, a point to raise is that the uATX formfactor limits the number of slots anyway, and there are a number of good reasons for bPlan wanting to start with a uATX board...
> [...]
>About DSPs...we really looked into some *special* audio solution last
>year but there weren`t any audio dsps with a pci socket with *documentation*.
>o The Motorola audio dsps have no pci bus...
>o The Cyrrus audio dsps are windows only for example which means you don`t
>get docs for them and have to live with the windows drivers.
>Quite pointless i would say....
Cyrrus == Cirrus Logic? That's also interesting to hear; back in the OS/2 scene, I used to run into Timur from Crystal Audio (Cirrus's sound division, at least at that point...) and they took great pride in keeping the specs available... Imagine times have changed on that front...
>o wake on lan/modem is possible with the current chipset...just a matter
>of sw support.
>o a special port for system diagnostics would also be possible but well..that`s
>developer stuff and they use serial output anyway.
>Adding LCD would require additional costs for a not mission critical LCD
>(I also had such an idea some years ago for the abox design:-)..but well)
>It would also require special case support for the lcd.
Time to give away another patentable idea- USB port extender that pops in a 5.25" bay and includes a USB-controlled LCD. (Or better, use a vacuum fluorescent display if you're putting it in a nice-looking aluminum case :))
There's no sense in making that sort of stuff standard, especially on a piece of hardware that'll probably get less press than the Mac Cube... Blinkenlights are cool, but you don't see a lot of people with BeBoxes.
>About Hypertransport and AGP4x....yep..nice features but if there are no
>PPC buschipsets yet which support this it`s quite pointless to "demand" it
>as that`s *impossible* under the circumstances.
>AGP 4x is beyond SDRAM bandwidth anyway and DDR only makes sense when the
>PPC can use the bandwidth...
What's the deal with PPC bus speeds? Enlighten the poor x86 guys like me? :)
(Alternatively, wouldn't there be something- albeit something relatively lame, to be gained from widening the bandwidth so AGP can sneak in without 'stealing' bandwidth from the processor?)
I imagine the DDR scene for PPC is still emerging, though, and PC133 does have the advantage of being slightly cheaper than air at the moment..
PEGASOS' : Comment 11 of 20ANN.lu
Posted by Ralph Schmidt on 25-May-2001 22:00 GMT
In reply to Comment 10 (Joe "Floid" Kanowitz):
Yes..i`m the morphos "guy"..sort of:-)
I`ve looked at the VIA datasheet and it contains this
"PCI-toPCI bridge configuration on the 66Mhz PCI Bus", so
the additional 3 PCI Slots are on a bridge.
Yep..cirrus logic..i always mix them up with the ex x86 cloner cyrrus:-)
Yes..another reason is the formfactor as mini atx gives us a lot
possibilities.
About busspeeds...
As you probably know DDRRam boards for the PentiumIII have no real effect
instead of the AMD(expecially with the new SIS DDR chipset).
The reason is that the PIII memorybus only runs with 100/133mhz and not
with 200-266Mhz like AMD, so it can`t make any use of the DDR 2x clock.
Sure you could give AGP the additional bandwidth but well...up to now
i haven`t seen significant speed differences between 2x/4x in real
usage so it`s possible that there's also an interface bottleneck in
the gfxcards.
Besides that the primary reason 1997 for AGP was the usage
of the dma system to fake a UMA system where textures are kept in
the system memory but that`s too slow for gfxcards anyway so the only
data you usally send over a lot are the vector datasets and these
are processed with the cpu first which has to write them back to
the ram anyway before they can be send...chicken&egg situation:-)
I don`t really expect DDR support in Macs before there aren`t new
PPCs which have a new system bus and IBM's roadmap shows some
entries there.
PEGASOS' : Comment 12 of 20ANN.lu
Posted by .jon on 25-May-2001 22:00 GMT
In reply to Comment 2 (Anonymous):
"ever thought about asking bplan...?"
Yes, I did. I wrote them one or two emails. No reply.
Besides: If I announce something as being modular I should state <em>what</em> is modular.
BIOS in Flashrom ? Stuff on "modular" PCI cards ?
.jon
PEGASOS' : Comment 13 of 20ANN.lu
Posted by .jon on 25-May-2001 22:00 GMT
In reply to Comment 5 (David Scheibler):
"You can find a new interview with bplan here:
http://www.amigafuture.de/eng/eng_aktion/eng_interviewbplan.html
He speaks about the price and OS4 compatibility and also he says that
currently Linux and a MorphOS alpha version are running on Pegasos. "
Have you guys realized how carefull they are in choosing their words when the word AmigaOS appears ?
No direct confirmation. "It is up to Amiga Inc." (okay, has some validity)
But they answered to this question only after it was asked a second time.
In another interview they said no word about AmigaOS at all, if I remember correctly.
In a mail from fleecy, I /remember/ (again) reading:
Amiga Inc. wrote to bPlan but did not get any reply....
Yes, they have a Zico/Amiga license....
To be a bit harsh: MorphOS does not like AmigaOS4 (ever seen Vaporware mentioning AmigaOS >3.1 ?)
Amiga Inc is not happy about MorphOS.
MorphOS is tied tightley to Pegasos. This all is okay and their own business.
I plan to run it on LinuxPPC and for sure will purchase a copy of MorphOS. But I would not be too optimistic about bplan's support for AmigaOS4+. They don't really seem after it.
All these are wild guesses (maybe not so wild ?) and my personal sightings. I will be very happy if I get prooven wrong.
PEGASOS' : Comment 14 of 20ANN.lu
Posted by .jon on 25-May-2001 22:00 GMT
In reply to Comment 9 (Ralph Schmidt):
"...the cyrrus DSPs are Windows only..."
Hmmm...Cirrus Logic (AKA Crytal Audio) does a lot of chips. They produce some of the best AD/DA codecs for HiFi upto High End.
There is an Amiga Audio card using at least ADDA of Crystal.
Investigating someinfo about the cheap Crystal chip on my pc's soundcard I ran into a lload of PDF docs on their side.
Ralph is completly right when he mentions FirewireB not being available until late of this year/start of next year to the consumer. I mentioned that as well.
But, my point was not availability but modularity and I felt that such things would have fitted into a concept of modular design.
Especially since three PCI slots are full very quickly. Sound + (MPEG Decoder or TV/Video Card) + SCSI Controller (no way around for serious backup solutions, with todays huge disk sizes)3 occupied PCI. A FirewireB card would need a 4th slot, a DCE microserver sounds like a *lot* of fun to me, better having two of them than noen....add that as well.
I am not an average user. I have special demands. This is why I am seeking for a special solution. If I'd like a more standard solution I'd go dual Athlon later this year (it still might happen). Since I am seeking for special solutions I allow myself to express my feelings about the candidates for "special" setups.
That's all.
.jon
PEGASOS' : Comment 15 of 20ANN.lu
Posted by .jon on 25-May-2001 22:00 GMT
In reply to Comment 8 (Anonymous):
"Sometimes it takes a waking up from some people to realize that
specific situations could be a problem."
This is how I feel. The Amiga scene is "hyping" strongly atm.
Zaurus - It gets dealed by several naive people like a "new Amiga".
"Content Delivery" - this is very thin ice. There is no fixation involved.
I would not be surprised if I'd hear somewhere that Sharp has abandoned the 'kiddy' Amiga market.
AmigaOS5/MorphOS. I want both. I know that some of the people behind and
involved into MorphOS are the same who brought very good quality apps
to the Amiga.
But a split market will take strentgh away from any side.
Personally I believe into Amiga Inc on the long run. If they do what I think
they want to do they might become some bigger players over the next years.
They'd need a lot of funding for this but then, who knows...
MorphOS /might/ never leave its state as Shareware OS for Ex-Amiga users.
I whish Ralph and his buddies that I proove wrong ;-)
80iesHomeComputer, 00ies = HomeServer ?
An ideal candidate would be PEGASOS, _much_ more than AmigaONE by Eyetech. We have to wait and see how things evolve.
.jon
PEGASOS' : Comment 16 of 20ANN.lu
Posted by Joe "Floid" Kanowitz on 25-May-2001 22:00 GMT
In reply to Comment 15 (.jon):
With regard to Ralph Schmidt's response... general agreement, and a minor 'ouch' on the PPC bus speed front... Obviously, the whole chipset scene is utterly different from Wintel, and I suppose I do get to run my Thunderbird 850 at 133MHz DDR ("266MHz") because the officially-133MHz Athlons exist (there'd be no KT133A without them)...
Now, to the issue at hand-
>Zaurus - It gets dealed by several naive people like a "new Amiga".
>"Content Delivery" - this is very thin ice. There is no fixation involved.
>I would not be surprised if I'd hear somewhere that Sharp has abandoned the >'kiddy' Amiga market.
Amiga Inc. has a tough mission; Bill and Fleecy want to support the legacy crowd while also making steaming piles of cash. :) I really liked the sound of the DE, as initially announced, although it was always obvious that it'd be hard to pull off. So, between the lack of real cooperation from Tao (and Tao's apparently found protected mode to be enough of a headache to avoid for a long time, now) and the whines from some of the community, it gets officially niched to make room for OS4.
The thing is, the DE really is marketable; it's what commercial developers always wanted from Java. (Hopefully, it'll be up to speed before regular Java implementations fully catch up to it.) It's an Elate 'distribution' at heart, but when they finally get the APIs in place, it's the difference between running... say, FreeBSD 3.5 and OS X. [Not quite so, but effectively so for the user- and if Amiga Inc. can become Tao's #1 client, they'll really acheive the 'partnership' status they hope they have, and Tao will be friendlier when it comes to extending their system for Amiga... Apple was able to 'evolve' much faster, since they could just take the BSD codebase and spit out Darwin.]
Think about this- services like i-Mode really *are* popular in Japan; American and European users don't get to play with all the weird games and applets (i-Mode makes a good example, because it's proprietary to NTT DoCoMo) because nobody's running an i-Mode service in those regions.
If the DE gets established on Zaurus, it not only enters a niche where it's welcome, but all that software ("content" works here; the first wave of popular DE programs will probably be equivalent to Macromedia Flash-based games and such) becomes accessible to, say, Palm/WinCE/Psion users elsewhere, and to anyone with Windows/Linux/BSD/whichever who feels like fiddling with it on their desktop.
Eventually, the system will be mature (or Amiga will go bankrupt ;)), it'll become a no-brainer for the development of bigger games/entertainment projects (count the number of ports that have to be made per year in the games market, versus those for applications software- a game has to run on PSX, PSX2, Dreamcast, XBox, Gamecube, GBA?, Windows, MacOS (now two flavors), and Linux to cover the full market- at least applications developers only have to worry about the *computer* platforms, not the consoles too..)
It does seem a bit specious, but you have to remember that Jay Miner always said the motive was to bring better gaming to the masses- making the Amiga into a computer over a console let people write it off on their taxes :) 'Kiddy' stuff can be big business.
>80iesHomeComputer, 00ies = HomeServer ?
I'd say it's more like- 1980sPersonal Computer; 1990s = Internet; naughts = Personal Network.
Even normal people are wiring their homes for ethernet (watch any home-improvement show, at least on US television)... Eventually, Bluetooth will come out of vaporware, cellular/PCS/3G data networks will stop sucking (and that doesn't stop some people from using them now)... In the '90s, it was impressive to connect to the global network; now it's about people taking their own data, and their own systems, and making them interconnect in a useful manner. People were doing this in the 90s, but it was mainly a business thing- just as computers themselves were, before Commodore and Apple showed up to the party.
Home servers are just one (optional) part of it all. After all, we can multitask- my personal network, such as it stands, uses only 2 servers- one is on the other side of the country, hosting a website, and the other is a grungy 486 serving DHCP and getting rigged up for some routing. Everything else is a peer. The existence of a server is not, in itself, the sexy part- it's what the server can enable.
Woah. That's a long rant. No more ANNing for 24 hours, I swear :)
PEGASOS' : Comment 17 of 20ANN.lu
Posted by sutro on 26-May-2001 22:00 GMT
>Today I read about the release of the final IEEE1394b specification, which, at its
> best, means FireWire at four times the current max-speed.
So, you believe that AGP 4x is 4 times the speed of AGP 1x too ?
PEGASOS' : Comment 18 of 20ANN.lu
Posted by Anonymous on 26-May-2001 22:00 GMT
In reply to Comment 15 (.jon):
"I know that some of the people behind and
involved into MorphOS are the same who brought very good quality apps
to the Amiga."
What??
PEGASOS' : Comment 19 of 20ANN.lu
Posted by Casey R Williams on 26-May-2001 22:00 GMT
In reply to Comment 18 (Anonymous):
Here where I am, it's almost 3 in the morning on a Saturday night... If you know what that means, then be forgiving... First off, Ralph, it's nice to see you in a thread in which people are not nailing you to the cross (so to speak)... You've done alot of hard work for this community and even though many don't realise it yet, you're an Amiga legend. Secondly, I wanna know why people don't get what DE is about! I'll admit having a bit of a problem seeing where Amiga and Tao are distinct entities- Tao seem like they don't need or have a plan for Amiga but perhaps that's part of the plan I don't see. What I do get is this... Tao's technology is small enough to be installed on an individual application basis. People act as if there has to be this huge Amiga logo all over the place, as if otherwise all is for naught. You just don't get it. What Amiga are working on is more like OpenGL than an OS. If an application needs it, it will either install it or locate it. That it happens to be an OS is beside the point, as most every desktop machine will still be running it hosted. Or should I say that apparently that's the only way to run it. This is great, as like OpenGL ubiquity will breed familiarity and in time maybe no one will care that the OS is hosted, much as many Windows users don't give a damn that Win has it's MSDOS heritage (something that I thought was horrible until I actually got a Win machine-the one I'm using this very moment). So basically we're talking about this trivial bit of code that can included with any application to allow it to run on some 14 different CPU based platforms. But, in order to insure that the majority of apps are able to run on portable hw, they don't want to have a desktop machine just yet. This is great as it means that Amiga OS will once again move forward (or at least laterally in worst case situations ;) and, and this seems to be the stumbling block for reasons I can't fathom, since the new "OS" is hosted, it shouold be able to run on MorphOS. Now is it Amiga (who shouldn't worry about competitors with their desktop OS if they truly believe in Tao's technology) or MorphOS/bPlan (who have no reason to fear either Amiga product, unless they're afraid Amiga will try to dictate the direction things go) who are being hard-headed? Ralph, Bill, either of you celebrate a birthday recently?
PEGASOS' : Comment 20 of 20ANN.lu
Posted by Anonymous on 26-Feb-2004 01:07 GMT
Bah!
Anonymous, there are 20 items in your selection
Back to Top