Slashdot Mirror
Amiga Inc. Files Multiprocessing Patent
Pappy writes " It looks as if Amiga Inc's development wing has filed a patent involving a very unique Multiprocessing scheme, in which groups of processors are thrown into 'Clusters', as well as coming up with a interesting Bus-Arbitration scheme for Multiple Processor groups. Check out the patent online. " Given this, the political movements with Linux, and the recent silence, I'm inclined to think of the Tom Waits song: "What's he doing in there?"
I was a longtime Amiga user as well. I'm going to keep an eye on what they do because I'd like to see them succeed, but I'm not going to put unrealistic hopes on them just because they're Amiga. Who is left from the original design team? Is the Amiga name the only thing left from the old days? They've been out of the running for so long that they very nearly have to start from scratch. There really isn't that much that is salvagable from the old architecture (hardware and software). What made it so great then is now being done as well as or better on current systems; multitasking, graphics coprocessors, video production (NewTek makes an NT Toaster now), etc. Sure, the OS was small and efficient, but if you want to compete with what's out there now you'd have to add a quite a bit to it's size. (They're going Linux anyway so that's not even an issue.)
So, assuming they're going to put out a A2000/4000 type system, what will be so revolutionary that everyone will drop what they have now and buy Amigas again?
The OS? - They've got a lot planned for networking and multimedia, but again, they're using Linux. Not that Linux isn't up to being extended that way but if Amigas are the only systems that can use these features they aren't going see widespread use. Developers will want a larger base than that. Linux was chosen for the same reasons, already available apps. They can roll those features back to Linux but then what an Amiga any better than any other Linux box?
Hardware? - They're going to use the PCI bus, Apple did it, others did it, hardware compatibility is just as important as software compatibility. If they use the Transmeta CPU, great, but it's also an unknown quantity, who knows how well it will really perform. I also can't see Transmeta only selling to Amiga.
Graphics? - ATI is doing the chipset and they'll never come up with anything remotely comparable to what NVIDIA announced for the GeForce (NV10) or what's being promised for the Playstation II. Its pointless to have a dedicated chip these days, this year's top-of-the-line chip is next year's clearance bargain and some other company is on top.
Desktop video? - Again, you can get the Video Toaster for NT, there are other systems on the market for the PC as well.
The computer market really isn't geared to 'unique' systems anymore. Compatibility is the name of the game now. The PC market is so large and moves so fast now because there are thousands of companies making advancements to the platform, not just one. The PC is also a lot more advanced now than it was when Amiga was around, not just in performance but in architecture as well. PCI takes away a big chunk of the ISA issues and brings it a lot closer to what the Amiga design was like.
I wish Amiga luck, but they've got a LOOONG uphill battle ahead of them.
A lyric in the song is "What's he building in there?" Hemos didn't claim he was stating a song title. Thus no further correction was needed (besides changing "doing" to "building").
No, but they might after reviewing the relevant patents and patent applications.
have you ever listened to "nighthawks at the diner"? it has both some great music and some of the funniest stories you've ever heard. i love it ;)
/. k.d /. earth trickle - Monkeys vs. Robots Films
do you know where to get any bootlegs? like of that show w/ Zappa?
16 shells from a 30 ought 6... oh yeah! (not on nighthawks...)
Uh, you know, I read through all the previous comments and did not find *one* anti-patents post. What, just because Amiga filed one means that patents are suddenly okay?!? Or do you all just use them as a sword against companies that you don't care for??? Tom
The way the system works, patent filings are confidential unless and until the patent is granted.
Amiga is probably filing lots of patents, but you won't hear about them unless they're granted and that takes several years.
You need to understand that the whole slashdot community is incredibly hypocritical. Only they are doubly hypocritical in that they are actually hypocritical about hipocracy. Sheesh, a bunch of anonymous and unvoted people 'moderating' things they dont like down, and any mindless pro linux frothing of the mouth up. Quite pathetic really. (PS i'm a linux user at home and work)
I truely pity the poor, pathetic specimens clinging to this anciant and archaic technology.
Not all of us rabidly spout Amiga advocacy to anyone unfortunate enough to be within shouting distance. Not all of us are kidding ourselves that the current Amiga technology is anything other than "archaic". Not all of us care that the Amiga Classic is, to all intents and purposes, "dead".
We use Amigas because we like them. They're not the latest and greatest. They have clock speeds measured in tens of megahertz, not hundreds. Their OS is a little creaky and a little unstable. So what? They're fun, and charming, and friendly, and a little eccentric. Just like their users :-)
We're not doing you any harm. So why do you, and others like you, insist on calling us names?
"absolutely" unique post there.
Seriously...
They're not worth the effort. I use Linux, but even I don't buy the QNX fallout. And I loved Amigas in the old days.
The message on the other side of this sig is false.
The Amiga may be vapour ware, only time will tell. But since Amiga's already run on PowerPC chips your concept of the Amiga state of the art is wildly inaccurate. Not that you should let reality get in the way of your posturing.
There are a lot of valid points that can be made about the viability of the Amiga in the long run. Similar cases on differing points can be made for the BeOS, Windows NT, Windows 98, Windows 2000, MacOS, MacOS X, FreeBSD and Linux. If you'd have mentioned the reluctance of development houses to port software to anything but Windows you'd have a point. If you'd have mentioned the high costs of manufacturing a new niche player in the hardware arena you'd have a point. However, your position can best be summarized by:
Have you ever had a relative who laid on his deathbed for years and years and refused to just keel over and die?
Those are braindead posturers. Boy I wish they'd just roll over and die already...
this was filed in 1997
they where working so nah to all who said they have done nothing
its intresting because it describes what I think to be a clustering tec that intel used to build ACSI RED ?
with P Pro this would work I think
anybody into scrossbars ca they tell me ??
john
a poor student @ bournemouth uni in the UK (a deltic so please dont moan about spelling but the content)
bwahahaha.. slashdot's poor researching blown out of the water by an obvious expert (and now I get flamed because this guy probably isn't an expert and I havn't put any effort into researching whether he is or isn't.. but I don't run a news service people...) Not that I think it is my place to say but I really do think that slashdot needs to hire some IT/Journalism multiple degree graduates.
How we know is more important than what we know.
Linus' visa.. we've explained this.
How we know is more important than what we know.
Hehe.. it's amusing flamebait though.. I mean, just obscurity is appreciable.. am I the only one who finds really obscure references funny.. and then there's the linking.. that futile attempt to suttlely introduce it into the conversation. Such depth.
How we know is more important than what we know.
BEOWULF get it right already =P
Just because you disagree doesn't make it offtopic or flamebait.
Gateway (or a whollyowned subsidary of GW2k) bought out Advanced Logic Research at around this
time (circa 1997). I wouldn't be suprised if these are patent rights carried over from that purchase to solidify their IP rights on this bus/ cpu clustering technology. I do recall that ALR was the first company to actually break the 4 cpu barrier for servers at this time with their unique Tri6 cpu module boards. You can see it here
http://www.ciaincorp.com/images/tri6x3.jpg.
Gateway after their purchase of ALR used them in their Gateway ALR 9000 servers, for awhile. At least in the first 9000 series servers did, but it looks like they replaced them in the later 9000
series with just plain old 4 way Xeon servers. I think that perhaps the Tri6 cpu module boards where PPro specific so that could be the reason.
can't we just agree that we want to beowolf cluster everything that can process including the cat and the bacteria growing in my fridge (we can do it.. I've seen journal articles about bacteria processing networks). Lets just connect everything to everything and have one big cluster (big cluster, hmm.. that's very close to an oxymoron - oops.. I can see the one word flame now, can you?) and everyone gets to share it and run anything they want and we all frown upon people who exploit the network for their own use without considering others.
ok.. now to diverge from the topic vastly, seeing we are talking about distributed computing, can someone answer this question about distributed.net. How do they know when they have found the right answer? Like, I can understand utilising everyone's pc to do the decryption with a random key and everyone submitting the decrypted garble that they get, but how do you tell the real plaintext from any old junk that the decryption process spits out because you gave it the wrong key. We assume that there are only english words (or at least a few) in the plain text so we could grep the lot with a dictionary but wouldn't this processing take as long (or as close to as long) as the decryption process. What I'm saying is, if the plaintext is random garbage can you still brute force search the key space? Or is there a checksum added to the plaintext, if so, isn't that really stupid?
ok.. off topic mode *off*.
How we know is more important than what we know.
I heard the same joke with a 12-inch pianist.
"The Amiga may be vapour ware, only time will tell"
...
That's funny, I guess we can say that about ALL vapourware, right?
If I may quote Monty Python:
Praline: Look matey (picks up Amiga) this computer wouldn't go voom if I put four thousand volts through it. It's bleeding demised.
Shopkeeper: It's not. It's pining.
Praline: It's not pining, it's passed on. This computer is no more. It has ceased to be. It's expired and gone to meet its maker. This is a late computer. It's a stiff. Bereft of life, it rests in peace. If you hadn't nailed it to the perch, it would be pushing up the daisies. It's rung down the curtain and joined the choir invisible. This is an ex-computer!
Shopkeeper: Well I'd better replace it then.
Praline:(to camera)If you want to get anything done in this country you've got to complain till you're blue in the mouth.
Shopkeeper: Sorry guv', we're right out a Amigas.
Praline: I see. I see. I get the picture.
Shopkeeper: I've got a Pentium III.
Praline: Does it run AmigaOS?
Shopkeeper: Not really, no.
Praline: Well, it's scarcely a replacement, then is it?
J.
That's a pretty standard delay for a US patent.
I once was a LOOOOONG time Amiga user. I was there before the comp.sys.amiga split. I was there for the release of BLAZEMONGER! And I hung on, fighting the good fight well past the point of reasonable behaviour.
I converted to Linux in '97, and haven't looked back. Now I have a couple of old, tricked out Amigas (one that runs Linux, even!) gathering dust.
I suppose I should sell them on EBay. I might make enough to buy lunch today.
The Amiga is dead. Dead dead dead dead dead. Anything else is pure delusion.
Hell, it was sad that I held on as long as I did. I truely pity the poor, pathetic specimens clinging to this anciant and archaic technology. You truly are missing out guys.
Not worth it.
Speaking of Anonymous people... Nah.
The message on the other side of this sig is false.
Pappy makes a common error in writing "very unique". Few qualifiers are appropriate for "unique", since it has a kind of "binary" meaning. It means "the only one" or maybe "one of a kind". It does NOT mean "unusual", which is often a more appropriate word to use than "unique", unless you are one of those lying salesman types.
Heh, the computer that submitted the patent must have been an Amiga for it to take this long.
This sounds almost like an extension of the original Amiga design. The original Amigas used dedicated ASICs tied together to increase performance and lower the cost. By dedicating sound, video, I/O to each chip, the old Amigas ran very fast on realtively slow hardware. This may be an extension of that to tie together a group of processors on the same board to do clustering within a single machine. This is probably similar to the Amiga designs that included a 68060 and dual 68040 processors.
It's even more history because the scheme doesn't work with the PIIs/celereons being sold today (because if a chip has N request pins then you can put N-1 in a cluster - with their 2 pins you get 2 clusters of 1 cpu which is the same as 2 CPUs so it's kinda pointless) it would work on xeons but their a bit too upmarket
..a new Linux-based OS
...multiprocessing (using some magical new processor)
...Transmeta MAY have something to do with the design
...It's not Windows
All these things generally tweak all you geeks' hotspots..so who cares if its an Amiga or an XYZ? It looks like its gonna be groovy.
http://uutiset.amiga.tm/
I must agree that the patent really sounds like something that Amiga Inc might benefit from, but there is one thing that's bothering me.
Does/did Amiga Inc really have anything going in North Sioux City, SD? And who are those inventors really?
/ Mårten Björkman - Celebrandil / Phenomena
Wow!
That bum on the street was right. Oh god. Well at least we'll get to find out what transmeta has up their sleeves ;)
> I'm inclined to think of the Tom Waits song:
;).
> "What's he doing in there?"
I belive it's "What's he _building_ in there?".
Sorry, just had to point that out, as it seems
that everyone has to correct at least one
error on every Slashdot post
--- witty signature
Have you ever had a relative who laid on his
deathbed for years and years and refused to just keel over and die?
That's Amiga.
So now they've filed a patent on hardware they're never going to sell. Add that to the OS revision for a machine that hasn't been built in about ten years. In a world where 500 MHz machines are becoming commonplace, how long can people limp along on machines designed around a 68040?
Amiga did cool stuff. But boy I wish they'd just roll over and die already...
-F
This patent was filed in 1997, and Amiga Inc's plans have about-faced a number of times since then, along with management changes. It seems unlikely that whatever they are working on has roots going back two years, so I wouldn't try to infer much about their current plans from this old patent.
---
Have a Sloppy day!
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
...because he is the man. the earth died screaming... /. k.d. /. earthtrickle - Monkeys vs. Robots Films
Check out the claims of the patent. Especially claim number three, which reads: "the CPUs are all Intel® Pentium® Pro processors.".
Amiga won't get my money if they go Intel, that's for sure. Then it's phase5 g4 box with qnx and linux for me.
Collas, are you listening?
This patent is very, very old.
The main opinion in the Amiga-community
is, that Amiga Inc. filed the patent for
another company, also bought by Gateway2000
in 1997. See comp.sys.amiga.misc for details.
Joerg Dietrich
There seem to be a few problems with this patent as far as prior art. For example, the clustering idea has been around since at least the 80's (the Carnegie Mellon CM* project, for example). Maybe they had a different scheme for arbitration, though. Many of the more advanced PCI architectures might qualify, however. There are compact PCI boards that have more than 1 processor. If a system had more than one of these cards in a system, it could fall under this patent (PCI arbiters sometimes use round robin schemes as mentioned in the patent).
Overall, I'd say it's interesting, but hardly novel, but I do look forward to seeing mainstream devices with this architecture.
Interesting page.. although, I'll admit that I didn't read it.. but, I found one part particularly interesting (it was actually a footnote, but who's looking?):
Operating systems, even Windows (which hides the fact from its users as thoroughly as possible), are actually collections of components, rather than undivided unities. Most of what an operating system does (manage file systems, control process execution, etc.) can be abstracted from the actual details of the computer hardware on which the operating system runs. Only a small inner core of the system must actually deal with the eccentric peculiarities of particular hardware. Once the operating system is written in a general language such as C, only that inner core, known in the trade as the kernel, will be highly specific to a particular computer architecture.
Of course.. that's because it mentioned Windows, and their futile efforts.. mwahaha!
Insert mind here.
tsia
Napster-to-go says "Fill and refill your compatible MP3 player", which is a lie. It's not MP3. It's WMA with DRM.
Even a cursory look at the phase5 and qnx sites tells a lot about the future of the amiga aftermarket.
G4 accelerators, QNX neutrino, possibility of a G4 ATX mobo, replete with the best in german engineering.
I suspect the "silence" is Amiga's way of keeping their options open after the sloppy way they dumped QNX for Linux a few months back.
QNX site
Phase5 site
Collas won't care anymore cos he's resigned.