|[Login] [Register] [Forgot your password?]|
| Report about presentation at the Neuss show|
More pictures of the presentation.
In a "conference room" (built out of some plastic walls which were put up, one desk, and an insufficient amount of chairs) Bill McEwen was talking about Amiga's future for nearly 45 minutes. There was a couple to listen to. So, let's rewind the dictaphones, and put together what was most imortant....
After the beamer could be woken up by the fifth trying, with plenty of drive and humor McEwen payed up (&qout; Wellcome to the Amiga sauna...."). From the start he cleared the roumor that Amiga would be in delay of payment for Gateway. Amiga owes Gateway no money. More than that, Amiga isn't in dept at all.
Further he introduced Dean Brown (DKB) as Director of Hardware once more. Mr. Brown would be responsible for creating referenz hardware for interessted third-party developers to look about. A game handheld device would allready be finished, which was in use by Amiga internally to impress interessted parties with the features of the new operating environment. Based on a StrongARM with 250 Mips, a port for a Sony Memory Stick, and a coloured display this device might figure well - and it was finished after one week of development!
About the new Vice President of Engineering McEwen showed himself mysteryous. This man might not get introduced bevore the 10th of Juli, cause he still is employed by another company - "one that is large, and is in trouble, you know...", and with this words a gesture of tearing something apart....
Further on information about the performance characteristics of the Java[TM] Virtual Machine - which is, acording to Mr. McEwen 22 times faster in handling multimedia content than any other JVM does (!)
McEwen set clear too, that he wouldn't ask the Amiga Community for patience. The fact, that this show is happening, that the audience is pressing into this (admittedly too small) room would argument enough that Amiga users are patient - over 5 years, allready. But he continues to set clear, that he and his associates are working for the same number of month to fulfill those high expectations.
As a first evidence of success Mr. McEwen took Amiga's SDK to the show, which - as he stated emphatic - still would be very incomplete, and would target experienced developers in main. The end-users will still have to wait for a seizable product for some time. E.g. there still is no sound support, streaming support or 3D graphics support included with the SDK, which will follow later (e.g. 3D support after about 3 month).
In the case of the critical theme of marketing McEwen admited, he still didn't employe a VP of Marketing - he thinks first to develope a product that would worth a respective marketing would be more important, than making generous announcements which can't get fulfilled with the best Amiga tradition.
Next he talked about the original plans to provide a deverloper's box by combining hard and software together. Because of the Amiga user's negative replies to this, they made up their minds - which required a lot of changes an additional programs (e.g. installer scripts). Now the software of the SDK was released; the HW/SW developer's box would follow after about 2 weeks.
Amiga would target "Ubiquity in Computing": mobile phones, PDAs, multimedia processor servers, a new desktop, a new game console, all of that with the same operating environment - something like that would never have been there.
Applications with this would be fully scalable. After a developer has decided were to put his applications, this one application could climb up the ",Food Chain" as high as it was wanted to. So, a game coded for a handheld would even run on a multiprocessor server. But on the other hand an application writen for a server wouldn't run on a mobile phone - or better to say - only very, very slow....
Processors allready supported by the operating environment would be PPC, M.core, X68, StrongARM, ARM, Sh/3,4,5, MIPS and a number of other CPUs, which they can't name (at this time), cause they are under NDAs with the manufacturers.
At the end Bill was talking about the JavaOne, a show that happend in San Francisco last week, and at which the Tao Group had a booth. The same demonstration shown here at the Amiga show there was running on a Sega DreamCast, on set-top-boxes, a StrongARM notebook, and a X86 computer. Only this was enough to magnetise the audience, and then they would have taken out a mobile phone on which a Pole Position clone could run, in spite of the low performance of 3,5 Mips. This would have been the same mobile phone which Scott McNeilly (Sun) had with him when he entered the stage at the CES in January, &qout;I've tolled you that once we will have Java[TM] running on such a device, once a day" - just, that it wasn't Sun's Java[TM] Virtual Machine running on that mobile phone....
And Linux wouldn't be the only operating system that would come into question as a host for the new Amiga. Other hosts would be Windows, WindowsNT, Linux, WindowsCE, QNX, and OS/9 (a market leading OS from Microware for set-top boxes). Host support for iTron, Epoc, VxWorks, and PalmOS would be in development, and close to be finished.
The cooperation between the Tao Group and Amiga Mr. McEwen illuminated in quite another way. It wasn't necessary that they would use only Tao's products. In fact they would have access to 50 developers at the Tao Group to create things needed by Amga.
Furthermore the BOOPSI libraries were portetd. This porting would be allready finished. They would now do testings and optimizations. The new scripting language (SHEEP) would also be finished during the next month - the rights for ARexx belong to a third-party manufacturer, and because of that they would develope another language. The catchword renderware was spoken, too. The new operating environment would be prepared even for this. .
And then Bill McEwen had a very special goody to offer: Once the single parts of the operating environment will be put together there will be the AmigaOne - a new multimedia desktop, developed and designed by Amiga, even if manufactured by thir-party manufacturers. McEwen didn't want to say more about that at that early stage, the way to go would still be very long.
Since it is a few minutes after midnight, and tomorrow there will be another day of the show waiting I'll finish for now. This was only the first half of the presentation. I'll report about the other half tomorrow....
McEwen continued on OEMs and ISVs - the latter being companies willing to support Amiga, but unable to "take their eyes off" what they are doing for a living right now. Eleven such ISVs had given their source code to Amiga - free of charge, free to port.
Of course, the well-known partners of the Tao-Group were mentioned: Sun, Sony, Motorola, JVC and others. McEwen recommended reading the press release from the Tao site (about the certification of their Java machine) - Amiga is in dealings with all those companies mentioned.
Then, McEwen showed pictures of his former office, and of the new "Amiga World Headquarter". Five other companies were located there, building high-end electronics for other companies for demonstratiuon purposes - with Amiga being able to use their equipment for free.
Then, Bill McEwen started demonstrating the new OS, hosted under Red Hat Linux on a notebook. He stressed that there was no hardware acceleration involved (which, as said before, will follow in August). It is difficult to put into words, but we saw lightning-fast Java demos, some 2D games, the well-known Boing demo, some free moving Boing balls you could grab and shoot inot the air with the mouse, a unicode browser, 2D filters, the transparent "Clock" window (where you could grab a ball through the clock - everything running on top of Linux, without any flicker or slow-down. Well, not true, I saw it flickering: when McEwen was starting the OS a second time, in a different window, the first one froze for a split-second, to continue as fast and fluently as before, while McEwen started the same number of demos again in the second window.
The demo was very impressive given the early stage of development, the pictures only grab a small portion of it.
McEwen was unable to elaborate on further plans, but recommended an article in Byte magazine from 1994, linked from the Amiga homepage; some of these things would redefine Distributed Computing completely.
Towards the end, McEwen answered some questions. I recorded the whole presentation on tape, but some questions couldn't be heard over the background noise, and some answers made no sense without the questions. I transscribed what I could make sense of, and left out one or two sentences I could not make sense of. As soon as I get my hands on the right adapter cable, I will make a MP3 recording of the tape (which was recorded with permission from Mr. McEwen).
F: What about the classic OS?
McEwen: We are using the pieces whereever it makes sense.
F: Will there be an emulation for it?
McEwen: We already have an emulator working. Testing is already done, we are now optimizing it so it runs very very fast. You guys know Brian King? You know where he works? He is a good man, Brian. He's helping us. So we're working to speed it up right now, it's a little slow. So we are in the process of optimizing it.
McEwen: Before somebody asks, you're asking if I'm gonna porting it to PPC. No. I'll tell you why. No matter who I've spoken with, R.J. Mical, Carl Sassenrath, Alan Havemose, everyone of them has looked at this in the past. It's an 18-month to 24-month process, and all of them agree it will only be 80% complete. Because I'd had to re-engineer the AGA chipset, I'd have to re-engineer Agnus, Denise, all those things are tied directly to the OS. In fact the hardest thing for Havemose in building the 3.1 were all those bugs that were in those chips they have to account for. So right now it would be too costly for us to make the moves necessary to capitalize on the market, we need to move forward. It doesn't help me to be everywhere. It doesn't help me to build a scaleable operating system, which is what we need to be to win. Making just another system on a chip is not enough. We need to be able to go to Sony, go to Panasonic, go to all those guys and say "pick your chip".
F: (Schlecht zu verstehen, Frage über die Hardware-Anforderungen des SDK, ob es ein spezieller Prozessor sein muß.)
McEwen: The SDK, if you are able to run Red Hat Linux you should be fine. We tested it on Red Hat and Corel. So you should be absolutely fine on that.
F: Will it also run on the Amiga?
McEwen: Not right now, no. We don't have anything on the 68000. We are hoping to work with our other friends so when you got PPC, which we know Linux can run on there, we can run on top of that. And we will be able to run native, on those, on the PPC cards. That's one of our biggest problems right now, you've got commitments from people delivering, and nothing is happening.
F: Do you plan to change the design of the interface?
McEwen: It is designed with the flexibility similar to today so you can make it the way you want. There will be defaults, so an Amiga users might want to have Workbench as their default, but for other consumers, they might to want it different. You will have defaults with the flexibility to change it the way you want.
F: (nicht zu verstehen) McEwen: If you are running Linux on a PPC card within an Amiga, alright? We have not tested it yet, however the two manufacturers have tested it and said it can run.
F: (nicht zu verstehen)
McEwen: Not with the first build, no. Part of it is a tools issue, the tools needed to utilize it. We have two companies right now in the process of making an IDE specifically for us, so that won't be required in the future. Everything is not in here (SDK), we know that, and there are certaily things to change; however, we felt it important enough to get people to understand writing for VP and some changes. Chris Hinsley, who created this, as you know, is an Amiga gaming guy; and I think you will find, as most have found as they began to play with this, it's extremly similar, in many ways as we write today. So it will be a very wuick curve, but there will be a curve. So we want it to get into peoples hands first, so they can begin working with it, utilize it, begin writing some applications, you know. Get familiar, so when we add the other components to it, you will be able to get moving much quicker.
F: (schlecht zu verstehen, eine weitere Frage zur Kompatibilität mit alten Anwendungen)
McEwen: In fact we already have an emulator that is working. It's not in this build, since we are in optimization, yes. In fact we're also looking into a hardware solution to do it.
F: Everything on the SDK is usable on an Amiga PPC?
McEwen: Nonono. What's on here is designed right now to work with Red Hat on an x86 platform or on Corel, ok? When it's completed, and you have an executive OS running, everything is fine, it can run. Again, we haven't tested it yet, according to third parties it can run on a PowerPC accelerator within an Amiga, as long as Linux is there. We haven't tested it yet, so I'm not telling you yes it can. We want to test it first. We've got two machines in the house we are doing it on.
F: (schlecht zu verstehen, eine Frage über die zu erwartende Verfügbarkeit von Software für das neue System)
McEwen: We have 131 applications already dedicated.
F: Also games?
McEwen: Most of them are games. There is a very large Indian company who does a lot of interactive games [...] they already commited to us and will be moving all their games over to Amiga. They have 63 games themselves, all of them for multiuser interactive gaming over the net.
F: (nicht zu verstehen) McEwen: Oh, oh, the multiple screens? Yeah, that's fine. What you see today from a visual perspective you still can get, ok? All the multiple layers, etcetera, that's still here, that's ok. That's what you were asking, right?
There were presentations before this one. The first, at eleven, was missed because I was still on my way to Neuss. The second one was so crowded I couldn't get inside, but I managed to record some parts of his speech, enough to catch up the fact that SUN will probably advertise the Amiga JDK to their user base. Report by: Martin Baute
All mentioned trademarkes are trademarks or registered trademarks of the respective onwner.
[News message: 10. Jun. 2000, 23:00] [Comments: 0]
[Send via e-mail] [Print version] [ASCII version]
Copyright © 1998-2022 by amiga-news.de - all rights reserved.