“Launched in 2005 for developers to beta test, the A9home went on general sale in time for the Wakefield show in May this year. The machine is still missing a few features, but seeing as it is mostly complete, it’s high time a review was published. Having ditched his aging RiscPC for an A9home, Paul Stewart reveals his first impressions with AdvantageSix’s ARM9-powered computer and its 32bit build of RISC OS 4.”
The RISC architecture is interesting from a computer science standpoint. But, the operating system seems to really show its age when compared to “alternative” operating systems.
It seems to me that by using the open source community, either by opening up the source (linux) or using some of the technology as a basis (OSX) they would be able to advance to the “modern age” of multimedia operating system without all of these awkward growing pains that we see now.
I’m not discrediting the project, but at this point in the development game shouldn’t they be open for more option to stay commercially viable.
This is only my American Opinion, perhaps there is a greater following in the UK?
It sounds like you’re confusing RISC processor architecture with the RISC OS operating system. I might be getting the wrong end of the stick – I’ve got a cold today so I’m not quite as bright as usual.
“RISC OS” is the name that Acorn gave their operating system that followed Arthur – RISC OS started at version 2.0. It was never a truly modern OS design – it lacked protected memory, and adopted a cooperative multi-tasking system, rather than pre-emptive. Acorn chose to call it RISC OS because they wanted to, not because of how the OS was designed/architected.
RISC OS runs exclusively on ARM chips, which are a RISC processor design. ARM chips however run many different operating systems, and are probably these days the most common processor architecture, owing to their near ubiquitous presence in mobile phones, PDAs, hand-held Nintendos, as well as appearing in many routers, none of which run RISC OS.
The A9home was built with the RISC OS market in mind, which is a small market. I’m sure you could run a Linux distro on it if you wanted to, although that is not the market AdvantageSix are currently targeting.
it has quite a large following in the UK.
also, according to recent news, castle is starting to opensource the OS. right now its just some of the modules afaik. not the kernel thought…
Browser: Mozilla/4.0 (compatible; MSIE 6.0; Windows 95; PalmSource; Blazer 3.0) 16;160×160
Might be worth considering for some.
I love the tiny and minimalist design, and the amazingly low power consumption, but I can’t see it selling to many people who aren’t already nuts about RISC OS. I’m surprised that they’re even trying to market this as a home computer, rather than purely for specialist industrial use.
At just over £600 (roughly $1,150) it’s not a cheap computer. Especially when you consider that it only has a 40Gb hard drive and 128Mb RAM, doesn’t support USB2, lacks bundled applications (apart from the ones included with RISC OS), and the price doesn’t include an optical drive. I can’t see many people buying one as a secondary computer just out of curiosity about the system.
Definitely an interesting addition to the limited selection of computers that can natively run RISC OS, but how many home users will be interested? Even if they want a tiny desktop and are happy to run an “alternative OS”, I can’t see the A9home competing with a Core Duo Mac Mini for £400.
In my opinion they should consider turning this design into a laptop (I know making a portable computer isn’t quite that simple). There hasn’t been a RISC OS laptop since the early 90s and I know a lot of RISC OS users would love to have one, rather than having to run an emulator on Wintel hardware.
With the hardware’s low power consumption battery life would be great and it should run silently. I think people accept a higher price when it comes to portable computing, especially if it had some clear advantages over the opposition. Does anyone else think it could find a niche in the portable market?
Yes. I have thought, for a long time, that RISCOS would be an ideal OS for a clamshell PDA.
It works on so many levels:
It would stimulate the development of some niche software (mileage calculators, map displayers, etc).
It provides justification to the user – it is hard for the user of other alt. OSes to justify spending 20ukp on a word processor when he can use a free WP program on his PC. But, 20 pounds for a really good word processor for his PDA, sounds like a more reasonable proposition.
Same goes for the platform as whole – it’s hard to justify so much money for a PC that does so little. However, a portable machine in addition to the main PC seems more justifiable.
All in all, I’m somewhat cynical about some of the RISCOS hardware development of recent years. Why do they NEED to develop a custom platform and then port RISCOS to that platform? Why couldn’t they simply port RISCOS to one of the existing ARM based platforms? Existing ARM implementations include PDAs, phones and ARM development boards. These development boards range from keyring sized, low-power units all the way up to standard ATX motherboards. What digital logic exists on these expensive custom devices solely for for RISCOS support?
In summary, I am hoping that the shared source initiative amounts to something that can allow the platform to progress in a manner which provides good value to the customer.
Speaking of clamshell, the Nintendo DS runs on an ARM9 , if I’m not mistaken. How hard would it be to hack RISCOS on to the DS?
If you had the source code to RISCOS, I don’t think it would be too difficult.
The original Archimedes/RISC PC machines had custom hardware for IO (IOMD) and video/audio (VIDC). The more modern machines don’t have any of this custom hardware.
TBH, the graphics hardware was never very elaborate anyway. It just presented you with a frame buffer + 1 hardware sprite for the pointer. The screen modes were setup via direct register poking or through OS routines.
As far as I know, the 32bit-clean versions of RISCOS don’t rely on legacy hardware features apart from the ARM chip itself. In the 32bit versions, the code has already gone from being tied to the 26bit ARMs to being 32bit clean so, I imagine that the code itself must be fairly clean as all of it must have been examined and recompiled to run on newer hardware.
I agree, the DS would be an OK platform to host RISCOS and there is a Linux port. Personally, I’d prefer a clamshell with a keyboard though.
Edited 2006-11-06 16:16
Is this for sale anywhere in the States?
Well I’m sure CJE micros http://www.cjemicros.co.uk (who are the selling agents for the machine) would ship to the states.
You are right. I emailed CJE and they said there are no US dealers, but they could ship me one for approx 40-50ukp.
We see the occasional story about RISC OS and ARM powered PCs here on OSNews, and I’m interested since it’s a totally different architecture. But it seems odd that there isn’t enough interest to have even a single importer.
I think Castle ship worldwide.
http://www.iyonix.com
Making the A9 into a laptop:
The Physical architecture of the machine won’t allow this.
The processor & memory are on a SO-DIMM that plugs into a bigger I/O board.
The biggest issue with laptops in general is that they’re all physically different!
This means that there is a major issue with the case.
Unfortunately the few thousand RISC OS laptops that would be sold makes it prohibitavely expensive to do.
Porting RISC OS to other hardware:
The problem here is drivers for the hardware.
CPU might be OK but, for instance, can we get access to the driver needed for the GPU?
RISC OS5 has a HAL and is being progressively open sourced so this problem might decrease.
Linux on the A9:
Has been done but the support for this product is for RISC OS.
Other uses for the A9:
In car MP3 players!
It may be for a very small niche market but its still a funky little bit of kit! 🙂