ACK has pretty much confirmed that their high-end Amiga system will use PA Semi chips. In an IRC session, Adam of ACK replied to the question how ACK was going to build a system faster than anything Apple ever produced on the PPC side: “Think PA Semi quad-core chips.” Adam also said ACK started talks with PA Semi about a year ago, while also promising specifications of the high-end system to be released coming Monday.
“Bah!”, I say. If they wanted to make Amiga OS 4 available to the masses they would release a version that can be loaded on old Macs. There exists a cheap and plentiful supply of such machines.
I don’t believe that the their motivation is to place Amiga OS onto absolutely cutting edge hardware. Why does Amiga OS even /need/ cutting edge hardware? It’s a small, efficient OS.
I’m sure that most Amiga who are still following Amiga OS 4 would be happy to pick up an old 500mhz PPC mac and then pay $50 or $100 dollars for a copy of Amiga OS 4.
The whole custom hardware path is built around the IP owners attempts to block piracy (if it became a problem, there would be other methods that could be used) and to build in massive margins for themselves.
Having re established the platform in such a manner, they could then pursue pie in the sky, custom hardware solutions.
Why should Amiga heads invest any more time in a company that puts its own interests so far ahead of the userbase? Oh, and the guy forgot to add the usual, “…and it’ll be available in 18 months” rubbish that Amiga users have come to know and love.
While the idea of putting Amiga on old Macs may sound appealing there are two problems with this: Will someone go on eBay and buy an old Mac to run the Amiga? A small fraction might but not that many.
Second, Amiga would have to reverse engineer the Macs to run their OS. Seems simpler to have your own hardware.
A lot of the reverse engineering has already been done in order to get Linux up and running on PPC macs.
I don’t think there would be many people who would be willing to pay for expensive custom hardware and yet wouldn’t be willing to grab an old PPC Mac from ebay to run it. At least that hardware would be capable run Classic/OSX+ a well tested and supported set of Linux dists in addition to Amiga.
I don’t see how it “easier” to custom design some hardware as opposed to porting to an existing stable, cheap and fairly well understood platform.
Second, Amiga would have to reverse engineer the Macs to run their OS. Seems simpler to have your own hardware.
Wherein lies the difference in supporting a newly put together, non existing platform as opposed to say Power Mac G5’s? Working forwards or backwards hardly make a difference to the Amiga lovers that have been hearing news like this on and off for years, now…
Actually I’d think it’d be better to actually get someplace instead of just dreaming of it!
Plus: Since it’s been possible for OpenBSD and NetBSD – not to mention Linux – to support the ‘odd’ Mac hardware, it should also be possible for the makers of the Amiga OS – if they want to…
If they wanted to make Amiga OS 4 available to the masses they would’ve written it to run on X86 or X86_64. There will never be cheap, plentiful, fast, and easy to find Amiga hardware as long as it’s tied to PPC.
I’ve always preached the combining of AmigaOS and the VIA EPIA paltform. Small, all-in-one motherboards, more than enough horsepower for AmigaOS, and best of all: they are readily available unlike expensive custom hardware. Now, that would’ve at least given the Amiga a fighting chance.
Via Epia is x86 based. If Amiga OS4 was x86 based, then I would say this discussion would be moot.
A more close suggestion would be the Pegasos or Efika boards from Genesi. But there seems to be certain discord between them that would prevent such a contract.
The reason why the OS4 does not have a fighting chance is because of the politics surrounding the parent company and the product.
Yeah, but I guess they think everyone would copy it then, but if $25.000 was all Hyperion got for it (I don’t know if it was, just a number I saw, doesn’t matter, say xxx $) just let the community buy it of and make it free.
The AmiZilla project is up to $10,373.02 (would be sad if they where never paid out, poor people who donated.), those money can’t be used to pay for AmigaOS source code and rights but if it was…
You know that windows XP can be run on 3-4 year old hardware without any problems, why does not people go buy old hardware instead of new? I think that is an stupid question just like your statement. Of course people would like to have an fast computer, no matter what OS it is running, it is not only the OS that will be running on the computer, so even if the os has an small footprint ans is blazering fast to work with, the software that will be used on the computer might require even more power.
You know that windows XP can be run on 3-4 year old hardware without any problems, why does not people go buy old hardware instead of new?
Because not everyone is sucking on the MS teets. I can install OS X on a old decrepit iMac and it works great. Just because MS bloats its OS beyond reasonable proportions does not mean all other vendors do as well.
So this nonexistent computer runs faster than a Mac? Side-by-side you can load Quark Xpress or Adobe Illustrator on each machine and the fabulous Amiga is going to be noticabley faster, be more responsive, AND the applications will be scriptable via ARexx ports!
It’s simply amazing what vaporware can do.
So this nonexistent computer runs faster than a Mac?
I bet it is infinitely faster than Mac
Because we will wait for this board forever+2 weeks.
[14:25] -ackcontrols- Theer are issues with OS4 for SMP and 64 bit at the present time, but that’s not a concern.
So they are planning to release a machine what can’t run OS4?!
The link in the logfile leads to an announcement that OS5 would have SMP support. However that wouldn’t preclude them from releasing a multi-core system that OS4 couldn’t do SMP with. It could still serve as an OS5 development machine/testbed.
Or a they might offer a potentially multicore system that would be upgradeable for OS5. But maybe you could lock a running image/instance of OS4 to a core and boost utilization that way until OS5 is ready.
[/totally_unsupported_conjecture]
Anybody able to conjecture on condensate-ware (vapor-hardware) features that could plausibly emerge from the ether for OS[45] to make use of?
Perhaps this is a redux of Be, Inc. many years later, with some important differences:
1. While the OS has existed a long time, the hardware has not been updated for awhile, according to what’s available.
2. This is a revival of an existing legacy system with major updates, not something completely done from scratch without a known compatibility goal to an existing system (this is Haiku at the moment, a bit like AmigaOS in that respect).
3. No JLG.
However, it is similar in using custom hardware designed for the OS (a good thing if you can make it work out) and focusing on PPC hardware for the main (or only) platform being supported. If things work out like they did for BeOS, perhaps they’ll migrate to using other pre-existing computers made by Apple (old PPC Macs) or Intel-based ones, or… who knows.
As to the hardware available, I honestly think that despite some people’s assertion that AmigaOS is efficient and doesn’t need much for the OS, there’s still the issue of what you do that isn’t the limit of the OS itself, assuming decent enough hardware: working with multimedia and being able to use it at a speed fast enough. h.264 (I think it is) I understand requires quite a bit of processing power to handle, and that has to come from hardware somehow, whether it is in a GPU/video card solution (very much the nature of the old Amiga), or done by the main CPU. After all, an OS without applications is just an exercise in computer science, and applications that don’t do what the users want/need due to either limitations of the OS or the hardware they’re running on tends to marginalize the value of the system, so I don’t think it is unwise for them to attempt to provide sufficiently beefy hardware to support what people want to do now, as well as somewhat future-proofing the system by allowing it to have enough room to do things not currently done. Also, it provides a known configuration to write drivers against and test, combined with the fact that beyond a certain age or date after original manufacture of various components, their prices go up quite a bit due to the law of supply and demand. If you doubt that, look for PC100 SDRAM that’s available now, and compare to what it was when it was at maximum production capacity.
Hopefully Amiga hardware, the OS and the applications exist for a long time to come, and are viable in enough numbers to grow. While I’m not an Amiga user (I did use them in the late 80’s at the equivalent of LAN parties) I would rather see a bit of diversity survive, trying out new things, etc.
“Hopefully Amiga hardware, the OS and the applications exist for a long time to come, and are viable in enough numbers to grow.”
You make some good points but I still think that getting AO4 running on hardware that is cheap, easily available and “good enough” is a better short term goal than building dream hardware to run it. Is any software available for Amiga OS that would push a fully supported G3 Mac beyond its capabilities?
Once the platform is properly established with a large enough userbase, developers can start work on the next generation of apps.
Towards that end, it’d actually be wisest to start off the Amiga PC (or whatever one wants to call the modern incarnation) from a stock motherboard configuration that’d have a relatively long production lifespan, with large enough numbers, and only do the minimal custom hardware. In other words, off the shelf components as much as possible, much like IBM did with the original PC, so the costs are cut down per machine, and they don’t have to do a lot of custom design and manufacturing. Perhaps they could work out some kind of deal with a motherboard manufacturer (Tyan, Asus, ABit, SuperMicro, whatever) to make guarantees they’ll still be making a given specific motherboard configuration with chipset for a period of time, for a small additional cost in the contract. The type of hardware this is most likely to result in are the two that tend to change the most slowly: embedded systems and servers. The most likely of those two to not mutate quickly and be manufactured for long periods of time are embedded systems. Thus, a Coldfire (or is it Crossfire? I forget what it is named for the embedded PPC) system may be the best compromise for long-term availability, but… it may not provide a fast enough processor for the more demanding consumer with multimedia hankerings.
Otherwise, supporting it on a gien computer configuration that you don’t manufacture has a the high risk of the platform evolving out from under it, and no longer being available. Depending on what the software engineering overhead is to keep up with the absolutely required drivers for a basic system is, it may actually be wisest to have a migrating supported system: for a period of 1 year, 6 months, whatever, get assurances from third party manufacturers that actually make boards to keep them in production, and be able and willing to keep up with what comes down the assembly line at the end of that period. If certain things didn’t change much (say, NIC hardware, GPU hardware, basic ports like USB) in that timeframe then the minor changes for new system board updates wouldn’t be too horrible.
Of course, the trick is getting motherboard manufacturers to agree to such deals, if there’s instability in what’s made due to advances and customer desires :/ The whole point I’m explaining is that there’s no method to assure that a lot of engineering costs don’t end up occurring somewhere to support a small userbase-used platform, relative to the number of people buying into it, whether it be in hardware design and manufacture on a semi-custom to custom level, to software development for writing and testing driver code for mutating hardware. Thus, it appears quite likely that the reality is unless they can get sufficiently large numbers of people to adopt, their costs to support their users isn’t too likely to make it easy to be viable, and yet, if the hardware that’s required is too custom, a lot of mere mortals may never consider buying a system to help provide that critical mass. Thus, I have a strong suspicion that unless there’s a lot more AmigaOS/computer fans out there with time and money to invest in this, this won’t work out in a viable manner over a period of several years for a hardware/OS manufacturer.
Crossfire is an SLK-like sport coupe/roadster by Chrysler. 😉
It’s also a multi-player role-playing game http://crossfire.real-time.com/
“Perhaps this is a redux of Be, Inc.”
Man, don’t jinx it.
IF (big if) ACK delivers and actually bring a PA Semi workstation to the market, rest assured that this is done with Linux in mind rather than the now-in-limbo OS 4.
–(From the IRC log)–
[14:25] -ackcontrols- Theer are issues with OS4 for SMP and 64 bit at the present time, but that’s not a concern.
–(End log)–
Not a concern? When we’re talking a “high-end” workstation?
I’m bettng that this piece of hardware primarily isn’t meant for the Amiga market.
There has been enough wasted time on AOS4.
I say, just like PalmOS6, ‘dump’ it. Get working on AOS5 that runs on specific intel hardware, just like apple did.
The ‘classic’ apps already run under emulation in AOS4, so why not just go the extra step ???
Even if it’s true (the Amiga world is full of empty promises) and they finally released, who is gonna buy it? People who bought Amiga One, are gonna fall again for this “Amiga Two”? And the rest of mortals…?
The closest thing to a nice Amiga hardware is the Pegasos platform, and it’s already available in many flavours (although the Efika doesn’t run MorphOS yet).
Of course, I’d be very happy if it’s released, it’s successful, it runs OS4 and Mac-on-linux ;p and Amiga Inc doesn’t mess things up.
I wanna believe in miracles…
The machine is a bit interesting, but more for the high end yet affordable PowerPC perspective that the Amiga part, which may come in handy since AmigaOS4 may or may not be available for these when released due to recent legal troubles. I’d be more likely to buy one of these as a Linux workstation than an Amiga.
I wouldn’t be surprised if Amiga, Inc. releases a Linux-based version of this board. They’ll need to recap their investments and AmigaOS is a single-user workstation so, in order to satisfy any potential server customers, they may just offer Linux as an option as well.
I have played about a bit with AROS on Intel boxes in the past but do not know much about AmigaOS to be frank BUT as a recent Mac convert with a PPC machine, with increasing numerous of Linux distros dropping offical PPC support, and not knowing what will come after Leopard from Apple itself in terms of keeping this G4 going, *I* would very much welcome the idea of making AmigaOS portable to older Macs – the recent Apple vibe I guess has already had folk like me scouring for a cheap, older Mac just to give it a go, after all: there wouldn’t need to be a second dip into eBay necessarily for such as me were the AmigaOS to become compatible.
Just a view.
Easy to make a system faster than another one just 2 years after it was released…
Easy to make a system faster than another one just 2 years after it was released…
Not that easy as you think.
Even if they will be succesfull with hardware (very little chance), the software should be upgraded to SMP support. This can lead to a number of problems in the OS itself and apps.
I agree totally with your comment regarding the next AMIGA OS 5. We definitely need to take OS 4 one step further and make it run on Intel/AMD processors. I would love to be able to install an AMIGA OS on my Mac Pro. Imagine the AMIGA OS running on two dual core Xeon processors @ 3.00 GHz – SWEET!
I started my computing life on the AMIGA and it would be so wonderful to have a modern AMIGA right now. I moved to the Mac platform after Commodore went banikrupt, and I was a Commodore AMIGA Authorised Dealer that felt quite betrayed by Commodore. Those incompetent idiots destroyed the platform, and I would be very surprised if it actually made a comeback.
Edited 2007-05-06 07:59
>I’m sure that most Amiga who are still following Amiga OS 4 would be happy to pick up an old 500mhz PPC mac and then pay $50 or $100 dollars for a copy of Amiga OS 4.
I’m pretty sure of that too… But you won’t run a company with a few hundreds of users…
But you won’t run a company with a few hundreds of users…
How about Bugatti?
Bugatti is a subsidiary of Volkswagen and not an independent company. Running a small niche company is a lot easier when you have huge rich multinational company backing you up.
Yeah, quad core PPC machine with nothing to run on it. I doubt AmigaOS 4 even if they can use it is multithreaded.
It’s all to late anyway, phase5 should have released the ABox =P
Nowadays I would much rather have had a new modern desktop OS than an updated AmigaOS, Gateways idea of using QNX back before 2000 seemed nice.
Taking Haiku and put the developer there might had worked aswell ;D
…when there is Amiga hardware, AND software to run on it.
“””
…when there is Amiga hardware, AND software to run on it.
“””
Going after Rip VanWinkle’s slot in the Guinness Book of World Records, are you? 😉
While even if the OS4 mess does get sorted out, it’s still be nice if it were modernized to take advantage of this possibly nice hardware. I guess that will have to wait until OS5 10 years from now.
Edited 2007-05-06 13:38
I’m sure I’m not the only one who cringes whenever a new Amiga machine is announced.
It’s a damn shame, too. I bloody loved Amiga OS, and kinda still do.
I loved it, but I don’t love it now. I wouldn’t want to run the same thing again, if I did I would already be running it. I would however give more or less anything to get the same experience today, but that isn’t anything AmigaOS4.0 can give me.
well if they can deliver a quad core pa-semi box, I’ll buy one for linux work. I think that would be ultra sweet.. (and imo smells more like the target audience than os4).
The whole smp/64bit “its not a problem” smell is.. a bad smell. you dont just add smp overnight. locking, race conditions. blah.
as a linux board, I’ll definitely buy one.
but until you can buy them, they remain just a smell…
If it’s true this is an interesting development. If they can cut some form of deal with PA-Semi they could become a system supplier for them, not exactly a huge market but probably better than the Amiga market.
Other PPC developers also need development machines so they could also be useful for there.
It’s never going to ever be cheap with low volumes but they need to start somewhere and this is the most modern part they could use.
I guess they can run some more system friendly coded ones? Or are there none? Can they run say TBL – tint and similair?
If not random UAE machine is more amiga to me than this will ever be anyway, what is Amiga without games and demos?
edit: Better just get a mac and get a modern thought not perfect OS (probably way more perfect than AmigaOS however) and uae.
Edited 2007-05-06 14:53
I first coveted an SMP PPC board nearly 9 years ago. At the time it was promised that this board would run Amiga OS. Though the board was quite expensive even at the time ($2500), I thought it would be an interesting piece of hardware to own. I’m glad I didn’t pursue it, because like all of the other Amiga associated hardware (with the exception of add-ons for the old systems), it has alternately been either vapourware, or extremely overpriced.
I hope that the people running ACK are really targeting the embedded ppc market, or are able to run AIX on the system, because those are the only two markets for the PPC platform (PS3 notwithstanding).
That said, as someone who has watched the Amiga revival businesses from afar for so long, it seems that everyone of these are a case study in how not to run as business.
…Gang Scheduler.
Frankly, I don’t care if they’re creating anything shining a hundred times brighter than any Apple branded thing. There’s no point in sticking to the PPC architecture after Apple itself ditched it.
No matter what Genesi or anybody behind Amiga do or think: do they want to sell PPC based thin clients? That’s just fine. Do they want to sell PPC based servers? That’s even finer. But what does that have to do with desktop OSes and desktop PC’s?
An increasing number of Amigans have been asking for a port to x86 hardware, because that’s the only way to go today. Apparently no-one is listening. The idea of someone marketing a box with a multicore PPC chip inside which AmigaOS cannot exploit fully (and perhaps doesn’t even need to perform brilliantly) makes me shudder if I think a light-weight OS like that would fly even on a VIA C7, Sempron or Celeron.
The only part of the AmigaOS 4.0 source code that is tied to the PowerPC is the 68k JIT emulation. The rest is written in C and can be recompiled when the need arises. There just isn’t enough PowerPC code written for AmigaOS yet to justify dropping the 68k emulation.
I sure hope Terrasoft Solutions jumps in and ports Yellow Dog Linux to the new hardware. I would love to have one PPC platform that supports a variety of OSes (having procrastinated buying Genesi’s ODW until it was no longer available).