VirtualAcorn boasts 3000 users | Drobe.co.uk archives

VirtualAcorn boasts 3000 users

By Chris Williams. Published: 19th Aug 2005, 17:13:27 | Permalink | Printable

You win a gold star

Emulation software developer VirtualAcorn revealed today that it has bagged its 3000th registered user. The maker of VirtualRiscPC and the discontinued VirtualA5000 is also celebrating its 4th birthday, according to an announcement this afternoon.

The size of VA's customer base is usually a closely guarded secret, fuelling the debate on whether or not sales of emulated hardware eclipse the number of native hardware products shifted. Although, it's known that, at least according to VirtualAcorn, VA5000 was retired after selling "thousands of copies" and it's likely that these sales contributed to today's figure of 3000.

To put this number into context, as far as we're aware, around 4000 copies of RISC OS 4 on ROM, over 2000 PS2MouseMinis and over 1000 NET100 ethernet cards have been shifted over the course of the past 5 years. These sales figures, whilst often seen as gossip or confidential secrets, are useful in gauging the size of the RISC OS market.

To celebrate their birthday, VirtualAcorn are running a special offer on VirtualRiscPC-SE from August 22nd to 26th.

Update at 19:27 19/8/2005
VirtualAcorn has added that the 3000 number is not the total number of products that they have sold - it's the number of registered users in their database.

"The figure includes all VirtualRPC-SE and VirtualRPC-Adjust's sold by us directly, but it doesn't include MicroDigital Alpha's," said VA's Aaron Timbrell.

"The figure also includes some VirtualA5000 sales, although in many cases the same user also has a VirtualRPC, and would only appear once in the database. Also dealer sales of VirtualA5000 aren't included and neither are show sales, unless the customer sent us their details and asked to be out on the database."

Links

VirtualAcorn website

Previous: ArtWorks 2.4 in summer special offer
Next: Firefox pledgers hold onto their cash

Discussion

Viewing threaded comments | View comments unthreaded, listed by date | Skip to the end

Given the current Worm problem striking Windows - the operating system Virtual Acorn/VARPC uses- the timing of this announcement probably could have been better - IMHO.

But be that as it may, yes VA/VARPC is an impressive piece of software. I'll hold up my hands and say I (did) buy a copy of VA5000 (but not until after an RO4 purchase), and both done at a time when Imago vs Omega was going on and it was not clear how the hardware front would resolve itself. I also had to use a PC to work on files from work - so I felt if I was being railroaded into buying a Windows license I might as well have some RISC OS stuff run on it too).

Now that RISC OS *has* viable alternatives available in the form of a mature platform in the form of Iyonix which is now my preferred platform (or the soon to arrive A9) I would find it more difficult to make the same justification - especially considering that the 98SE that is on that *old* PC would need to be upgraded to Windows XP just in order to run VARPC - and let's face it I don't want to pay Bill Gates any more money to run RISC OS - thank you very much

The real question is the VARPC success at the expense of killing the RISC OS hardware platform and tying our future use to that of Windows - and that only time will tell

AMS on 19/8/05 6:59PM
[ Reply | Permalink | Report ]

VRPC does not tie the future use of RISC OS to Windows, since there's nothing to stop VRPC being ported to an alternate host OS, provided there is a commercial justification for such a port.

ninja on 19/8/05 7:22PM
[ Reply | Permalink | Report ]

VARPC certainly has a home, and I cannot understand people who say it is robbing RISC OS hardware manufacturers of sales.

I for one would definitely NOT have bought new RISC OS hardware, as I need a PC, so it makes sense to use VARPC. And anyway, the PC I have can run VARPC as fast as an IYONIX (yes, I have tested both), with much faster disk access, and it is portable.

VARPC is, if anything, widening RISC OS use in the world. I can see a day when RISC OS hardware ceases, and we all go the way of the PC, it's just a question of which OS you run on it.

demondb on 19/8/05 7:22PM
[ Reply | Permalink | Report ]

Given that VRPC on a modern PC at least has one concrete advantage over new native hardware (speed), I'd say it will potentially have had to have led to lost sales for native hardware manufacturers. Although I'd concede that probably the vast majority of sales come from users who'd either stick with their RPC until future hardware became cheaper, or who would have entirely abandoned the platform.

Also, the existence of a faster way to run RISC OS will also be a driver to the native hardware manufacturers to provide faster hardware. Until VRPC came out, many were happy to stick with their trusty old RPCs, so there was *no* market for faster, newer ways to run the OS.

That said, to deny that VRPC could compete with native hardware in any circumstance is shortsighted.

ninja on 19/8/05 7:30PM
[ Reply | Permalink | Report ]

What are the limitations? I'm very tempted to buy when the Mac version's available.

freder on 19/8/05 7:59PM
[ Reply | Permalink | Report ]

demondb said >"VARPC certainly has a home, and I cannot understand people who say it is robbing RISC OS hardware manufacturers of sales. "

You can't extrapolate from your case and suggest that VARPC has had *no* effect on native hardware sales. If (as the numbers suggest) more copies of VARPC/VA have sold than *actual* hardware there *must* have been some people who would have opted for VARPC over actual RISC OS hardware - particularly given the "spin" put on by some (I stress *not* by Virtual Acorn themselves) such as describing machines running VARPC as being "hybrid computers" - when in actual fact what was being discussed was a plain ordinary PC running VARPC on top of Windows.

ninja> Yes *if* VARPC *is* ported to another operating system then yes RISC OS won't be dependant alone on Windows. I would suspect though that while money can be made hawking VARPC on Windows then that's where it's going to stay. Do bear in mind no matter *what* OS VARPC moves to it makes *RISC OS* dependant on THAT OS and VARPC - RISC OS becomes *dependant* and therefore I would argue *vulnerable*.

demondb said >"VARPC is, if anything, widening RISC OS use in the world. I can see a day when RISC OS hardware ceases, and we all go the way of the PC, it's just a question of which OS you run on it."

So how many people use VARPC who never used RISC OS on actual RISC OS hardware then ? I'd suggest very few in short it's shuffling the deckchairs on the Titanic time....

I would agree that I can see a day too when RISC OS hardware ceases, and VARPC may well have helped in its own way to acchieve that, thing is what makes you think once we get to that point that you'll have a *choice* over which OS you use ?

Currently Intel/Microsoft/Phoenix (of the BIOS fame) are drawing up a next generation BIOS - given the credentials of the first two organisations there I'd be unwilling to wager that they'd be too keen on allowing people free choice over what OS they use.... and if you think Apple are any better what's this about Digital Rights Management (DRM) being built into the x86 version ?????

Nah, our best hope is to keep improving the ARM hardware and keeping RISC OS running "for real" rather than pushing it into a space where ultimately it will not survive.....

AMS on 19/8/05 8:04PM
[ Reply | Permalink | Report ]

AMS: If by 'vulnerable' you mean vulnerable to worms, viruses and the likes, then I agree. But that's just something you have to put into the comparison of speed and price versus efficient CPUs and not being vulnerable to nasties.

If you mean vulnerable to an attack by DRM and not being allowed to install VRPC on the host OS without paying for some additional license check, thereby killing the use of RISC OS on that host OS, and, just assuming we'd stopped using native hardware by then, stopping all of us using RISC OS, then I'd disagree. The existence of VRPC makes it much *easier* to port RISC OS over to other hardware and OSes. And if all host OSes were to become so offensive to creators of small software then I'm pretty sure hardware and a host OS to run on it would magically become a lot more popular all of a sudden.

VRPC has made the use of RISC OS less vulnerable to the whims of a few select companies, while making it more vulnerable to worms and viruses.

ninja on 19/8/05 8:31PM
[ Reply | Permalink | Report ]

ninja> "If by 'vulnerable' you mean vulnerable to worms, viruses and the likes.."

Yes, there is that. It then boils down to if the risk is worth the benefits percieved by the user. As I am a naturally cautious sort I prefer not to take the risk - naturally though not all people take that view. If people wish to use VARPC then I suggest that they *must* take it upon themselves to ensure that their machines have their PC Antivirus updated regularly - and Windows updates done as well - so that the risks are as low as can be acchieved.

ninja > "VRPC has made the use of RISC OS less vulnerable to the whims of a few select companies, while making it more vulnerable to worms and viruses."

So it then boils down to which does the greater damage ? I'd also point out that Virtual Acorn are (like ROL, Castle, Ad6 et al) just another company and will no doubt be looking out for *their* interests while these coincide with that of the RISC OS using public *grand* if they should ever differ then potentially we're all stuffed. I fear the natural outcome (suggested by demondb) that eventually RISC OS hardware would cease and then rather than having our eggs in two baskets (that of ROL/Ad6 and Castle) it would be in one basket (VA/Microsoft).

VRPC for all it's potentially for having RISC OS running on disparate platforms has failed to cause this meaningfully to materialise - and even if it did now with the Mac going to go x86 one can only wonder what effect that would have on VARPC trying to run on a hardware platform that effectively ceased to exist - unless of course emulating ARM on an emulated PowerPC running on an x86 sounds like a sensible proposal to you

AMS on 19/8/05 8:52PM
[ Reply | Permalink | Report ]

With regard to the size of our customer base being a closely guarded secret. Every single product we supply mail order to end users has a customer number on the address label. These numbers start at 2 (1 is reserved for us) and carry on from there. So anyone who has ever purchased directly from us by mail order could easily tell how many registered customers we had at the time their order was processed just by looking at their customer number on the label.

VirtualAcorn on 19/8/05 8:54PM
[ Reply | Permalink | Report ]

In reply to AMS:

"So how many people use VARPC who never used RISC OS on actual RISC OS hardware then ? I'd suggest very few in short it's shuffling the deckchairs on the Titanic time.... "

I could word this another way. How many are using VARPC would have left the RISC OS platform instead of buying new hardware. I admit that I am one of them. It seems short sighted to insist the everyone should buy native hardware. What I meant earlier when I said it was a matter of which OS it runs on, should have read which computer it runs on, because as we all know, Windows runs on a multitude of different PC's.

And you said it, however you shuffle the deckchairs, you're still on the Titanic.

demondb on 19/8/05 9:09PM
[ Reply | Permalink | Report ]

freder: "What are the limitations?" None, as far as I can tell

ams: "You can't extrapolate from your case and suggest that VARPC has had no effect on native hardware sales." You're right - it could have increased hardware sales by tempting ex-users back into the market.

"So how many people use VARPC who never used RISC OS on actual RISC OS hardware then ? I'd suggest very few in short it's shuffling the deckchairs on the Titanic time...." In my opinion, RISC OS is not in a state where we can really have the luxury of attracting brand new users. However, there are a /vast/ number of ex-users out there and if VA temps some back then that's great

Adam

adamr on 19/8/05 9:29PM
[ Reply | Permalink | Report ]

demondb>"It seems short sighted to insist the everyone should buy native hardware."

In fact I am *insisting* on *nothing*. I am simply stating the obvious - IMHO VARPC damages the sale of *REAL* RISC OS hardware. It's then down to people to decide if putting Microsoft in charge of RISC OS's future is a good idea. At least with *actual* RISC OS hardware about that risk is limited - if RISC OS hardware dissapears then you *are* dependant fully on Virtual Acorn and Microsoft for RISC OS's future.

demondb> "What I meant earlier when I said it was a matter of which OS it runs on, should have read which computer it runs on, because as we all know, Windows runs on a multitude of different PC's. "

And that matters because ? If at the end of the day Windows (on all those different PC's) can't run VARPC you're stuffed - be it a Dell, an Acer, a Siemens-Nixdorf, be it based on a P-IV/AMD64/Semperon etc., makes no difference if Windows just *won't* run VARPC. Or if Windows changes in such a way that it is *uneconomic* for Virtual Acorn to update VARPC to keep it running on newer PC hardware. Yes I know that is conjecture and may not necessarily happen - but a multitude of PC hardware running ONE OS (Windows) is still just running ONE OS (Windows) and leaves you dependant on Microsoft and having different badges on the PC Box or a different brand of PC processor does *not* get around that.

Ask the former users of DR-DOS they know all about this sort of thing

AMS on 19/8/05 9:32PM
[ Reply | Permalink | Report ]

adamr>"it [VARPC] could have increased hardware sales by tempting ex-users back into the market. "

A singularly inefficient way of doing it. Yep, you buy a PC with Windows XP to buy a further piece of software (VARPC) to run RISC OS for you and then you're so overcome with joy you buy an Iyonix ? Is this really likely ? Really ?

I would suspect a person with an old chugging RPC sees VARPC running on a cheap PC - it's faster than their old box - the package costs less than an Iyonix - YIPPEE they buy the PC. Technically I suppose they're still using RISC OS - but the only new hardware bought was a PC. Then at some point they say - nah, why boot VARPC at all - save the time - just run Word/Excel/Powerpoint - game set and match Microsoft.

AMS on 19/8/05 9:54PM
[ Reply | Permalink | Report ]

AMS: "Is this really likely ? Really ?" I don't see why not - ever heard of "try before you buy"? In any case, my comment was rhetorical - the point I was trying to get across is that it is far from clear exactly what impact VA has had on native hardware sales.

"Then at some point they say - nah, why boot VARPC at all - save the time - just run Word/Excel/Powerpoint - game set and match Microsoft." I don't understand your point. Surely if using MSOffice is best for some individual user then that's what they should do, by definition?

In any case, all this talk of hardware leaves me cold. The software is what interests me - and it seems likely, to me, that VA has benefited the RISC OS software scene.

Adam

adamr on 19/8/05 10:47PM
[ Reply | Permalink | Report ]

Happy Birthday to Virtual Acorn.

Revin Kevin on 20/8/05 2:33AM
[ Reply | Permalink | Report ]

I don't think the effect of VA on sales of realAcorn hardware is as simple as some people make out. Sure there will be some people who bought VA and chose not to buy new the realAcorn hardware that they could have bought. Mainly these will be people who might have bought an Iyonix. However, I guess for many of them buying realAcorn wasn't a proposition because they needed a portable machne (which remains non-existent) or because they also needed a realWindows machine because of the limitations of RISC OS.

What strikes me now is that the people who bought VA because realAcorn wasn't a proposition may now be changing their minds with the advent of a cheaper realAcorn machine. So in these cases VA has kept them using RISC OS when they could easily have gone over completely to another OS - and now they might well be buying new real Acorn hardware. If this is the case, and I suspect there is something in it, then VA will have helped sales of realAcorn by keeping them in the game.

Of course, no-one really knows for certain!

SkyPilot on 20/8/05 8:54AM
[ Reply | Permalink | Report ]

as a VA user (MD Alpha, actually), I bought it to get a laptop - there being no native RISC OS alternative [despite at least two attempts by firms to make one]. VA suffers from having an underlying Windows OS. Virus checking with McAfee makes VA almost unuseable because of speed, and was probably responsible for a complete crash resulting in the need to reinstall. I'm now using AVG Free, as suggested by VA, but it needs to go through a checking routine when turning the machine on. It takes 3 or 4 minutes after turning on before I can use RISC OS. There is no alternative, but I wish there was.

charles on 20/8/05 9:08AM
[ Reply | Permalink | Report ]

They should've given a copy of Star Fighter to their most recent customer

Hertzsprung on 20/8/05 12:55PM
[ Reply | Permalink | Report ]

The split between VRPC users using laptops and those using it on desktop machines would be very interesting to know, but I doubt there's any reliable way of finding out.

If the emulator can do everything the native hardware can, and faster, and cheaper, then even if there are some other issues going that way then frankly the native hardware side is in a bad state anyway - without VRPC to highlight that would just be burying heads in the sand.

SimonC on 20/8/05 1:42PM
[ Reply | Permalink | Report ]

Damned impressive sales figures I'd say. I expect the majority of that lot is schools who have been forced to switch to PC's, or people who want a RO laptop.

I would definitely pay 70ukp for a Linux version, although that's not done yet - I heard a Mac version will be done first? I don't use Windows enough these days to justify buying the Windows version, I guess I could run it under VMWare.

Come to think of it I sold my RiscPC's with all my RISC OS software anyway, might still have my RedSquirrel images somewhere....

I'd love a VirtualAcorn that allowed you to use any OS like the RedSquirrel Dev version

simo on 20/8/05 2:36PM
[ Reply | Permalink | Report ]

simo: A Linux port has been shown working, but support for it was too demanding... Too bad, eh

Perhaps you can try the vanilla version out under Wine.

What if AMS will be right? What if VA users will prefer Windows / Mac OS X software above RO software, if it will outperform native machines in all respects? What if dealers rather sell a jolly VA bundle, than supporting native machines and find out their customers gradually stop using RO software? Thankfully Castle would never let that happen! They would rather stop the license, than see their beloved platform become cyborged, won't they? Sure, they need us, remember?

hEgelia on 20/8/05 4:19PM
[ Reply | Permalink | Report ]

The emulated operating system is only "vulnerable" in the hands of foolish Windows users. I have had spyware problems, but no viruses, trojans et al on any of my Windows systems in a great number of years.

Frankly native RISC OS hardware is still very expensive given the relatively small market and limited sales. I could pay £1000 for an Iyonix or I could spend that and build up two extremely decent PCs and have money spare for a couple of VRPC licenses.

Trapper on 20/8/05 8:32PM
[ Reply | Permalink | Report ]

I've always believed in just using the best tool for the job. I use RISC OS, Windows and Linux, well don't use RISC OS anymore, but that's besides the point. I use Windows for DVD creation, and Linux for pretty much anything else.

I had to make some letterhead, a logo and some compliment slips the other day and started on OpenOffice.org, then got annoyed as there are no decent vector graphics programs for Linux really, so thought OK, I'll switch to Windows and use Office2003 and Illustrator, then got fed up of the crappy export options (needed decent PDF and Word97 really).

If I'd used Artworks and Publisher+ (maybe even Techwriter Pro+) I could have done it in about an hour instead of the day or so it took (eventually in a mixture of OO.org Draw exporting as WMF into Word and printing to Acrobat).

So for me, in that instance, RISC OS would have been the best tool, but then for internet and programming Linux kills RISC OS/Windows, and for DVD creation and video editting Windows kills the others. Glad I don't have a Mac to confuse the issue even more ;-)

So what I'm trying to say is that I'd be very surprised if many people would just switch to using Windows after using VirtualAcorn.

Oh and hEglelia, RedSquirrel doesn't work nicely under WINE (mouse issues at least) but works fine under VMWare, as you're basically running a complete Windows install.

Some interesting screenshots for ya though: [link]

simo on 20/8/05 8:36PM
[ Reply | Permalink | Report ]

simo:

so why the hell didn't you use a RISC OS computer for your task?

& isn't VMWare still *very* expensive these days?

ROHC on 21/8/05 11:57AM
[ Reply | Permalink | Report ]

Skypilot said >"So in these cases VA has kept them using RISC OS when they could easily have gone over completely to another OS - and now they might well be buying new real Acorn hardware. If this is the case, and I suspect there is something in it, then VA will have helped sales of realAcorn by keeping them in the game. "

That's a bit like saying all the people waiting for the Omega to appear kept people running RISC OS - so therefore Omega was a good thing. While (perhaps) in a few instances that was the case in the long run damage occurred (the long overdue Omega, loss of confidence and eventually MD enticing ROL to have a release for VARPC on the MD Alpha - which I would argue further damaged "real" RISC OS hardware sales).

I've had a good long think about all of this and *perhaps* things are *not* as bad as I thought. Perhaps the "lower" VARPC-SE price is down to the fact that with the lower priced Iyonix (and soon to follow the A9Home retail version) available people were doing the maths and saying - "the PC/VARPC option is *not* the cheaper option anymore" and that there *are* in fact "real" ARM hardware alternatives - in short VA is having to lower their price to compete against a resurgent ARM hardware platform. If that were the motivation I'd be somewhat less worried.

Trapper said > "The emulated operating system is only "vulnerable" in the hands of foolish Windows users. "

Yes quite, but remember many people who buy VARPC may *not* have used windows before (they are RISC OS users in the main). They may not know that the average time for an unprotected Windows PC to become infected on the internet is 20 minutes, they may not realise you need *broadband* to ensure that you *can* download Windows and/or AV patches *quickly enough* to ensure your machine is protected *before* the viruses strike.

Simply using VARPC on Windows does not somehow render them immune to viruses or the need to maintain the Windows patch level currency or AV updates. People using VARPC *ARE* windows users and must forget the RISC OS world - if you want to run RISC OS under Windows you have to do so under *WINDOWS* rules.

adamr said > "I don't understand your point. Surely if using MSOffice is best for some individual user then that's what they should do, by definition?"

I did *not* say they couldn't nor did I indicate that Word was preferable for anything. What I meant was that people would in time get used to Windows and through sheer convenience just load and run the available Windows apps rather than loading VARPC (delay) loading techwriter (or whatever further delay), access file from HostPC etc.,. Windows takes long enough to boot without adding the VARPC delay. Over time RISC OS usage would drop - if you like VARPC is a means to allow RISC OS users to get used to windows - a sort of stepping stone to become a Windows user. Microsoft did such a thing for WordPerfect users - they provided key stroke compatibility in Word so that WP users could get used to and finally switch to Word - which is of course what they did.

This infers *nothing* about the appropriateness or otherwise of Windows Software over RISC OS software - all it says is that people like convenience - if you can do something with one or two fewer mouse clicks and fewer delays people will generally opt for the more convenient. I am not *blaming them* for that either - its just a natural consequence of trying to graft one OS on top of another - generally the base OS (Windows in this case) is the one that will be used and whatever apps are associated with that.

AMS on 21/8/05 12:05PM
[ Reply | Permalink | Report ]

AMS: "generally the base OS (Windows in this case) is the one that will be used and whatever apps are associated with that"

Only if the two are equally good at every job. The whole point of using VRPC is that people want to do certain things in RISC OS rather than Windows.

You are suggesting that the benefits of using RISC OS are so tiny that they are outweighed by the 20 second delay, once or twice a day, as VRPC loads up. If that's the case then, frankly, no amount of cheap dedicated hardware (or whatever you are proposing) is going to keep people with RISC OS!

Adam

adamr on 21/8/05 4:02PM
[ Reply | Permalink | Report ]

Re. vulnerability, I would suggest that the Wintel platform is less likely to die than the real ARM hardware. On that basis, ARM hardware draws RISC OS users away from the "safe" option.

Wintel/DRM licensing will always be cheaper than ARM motherboard development. So ARM hardware also draws users away from the lower-cost, higher-performance option.

Now I am not saying that is all a good thing, but it does seem to be an valid angle to consider.

What users want is to be able to choose the best hardware, then choose the best desktop OS regardless of the hardware, then choose the best apps regardless of either.

RISC OS should aim to be the OS of choice on as many platforms as possible: roll on VA Mac and Linux. And it should make migrating apps across as easy as possible - roll on the Unix Porting Project too!

steelpillow on 21/8/05 7:29PM
[ Reply | Permalink | Report ]

Maybe migrating RISC OS to the x86 platform would be the best possible solution in the long run. Not the cheapest though I'd guess.

Gulli on 21/8/05 7:54PM
[ Reply | Permalink | Report ]

gulli:That would be great, I think. However, it's been pointed out before that it simply is not viable.

adamr on 21/8/05 10:09PM
[ Reply | Permalink | Report ]

adamr: No, probably not viable but perhaps a move to an existing kernel like BSD could be a way. Not that it would be much more viable

Gulli on 21/8/05 11:01PM
[ Reply | Permalink | Report ]

This whole thing about Virtual Acorn versus native hardware is a good way to kill a bit of time on pure speculation, which is fine but ultimately doesn't achieve much of anything other than opinion exchange.

I'm waiting for the Mac OSX version of Virtual Acorn for a couple of reasons: we run both Windows XP and OSX in my work, and of those two I prefer OSX because the interface is easier to work with and it doesn't crash as much as XP (which admittedly is miles better than earlier versions). There isn't the engineering and business software available for RISC OS that we use here (let alone dealing with the easier stuff like MS Office files that have to be interchanged with clients: I've tried the RISC OS apps that do, as well, by the way), so there's no easily workable option for me, but to have non-RISC OS hardware.

My old RiscPC recently started misbehaving, and here at the bottom of the earth (a paradox, I know) there isn't much joy in looking to service old equipment. I have a backup machine, but that requires some setup time that I don't have at the moment. I like a couple of key RISC OS apps and the way in which the OS works, so I haven't abandoned it, but I need something for day-to-day work. My logic leads me to Virtual Acorn on my Mac PowerBook.

As for new native hardware, I'm interested in a less pricey Iyonix (I've seen the new options, thanks) or A9 Home (when it's available in final form).

So I'll probably be equipped with both in a short period of time, depending on how quickly the A9 gets to market, if only for comparison purposes. I know that's maintaining the current user base, and not expanding it, but you can't expand if you're losing users too quickly, either.

RobA on 22/08/05 05:24AM
[ Reply | Permalink | Report ]

Good arguments Rob. I think your case is perhaps a common one among users. It surely is a time, and has been for some years, for RiscPC users to upgrade/replace their machines. If VA for both (commercial) mainstream platforms is an option, it could at least keep several users as RO users.

As to Gulli's idea about porting RO to x86, I think this has been raised quite a few times now. I tend to agree to certain benefits it poses, but perhaps the disadvantages (especially time/money involved) outweigh them. One point to consider is that the xScale is probably the most advanced and attractive ARM design available for RO machines and in this respect we depend on Intel's agenda. One could say RO is hard-coded to the ARM architecture and porting it to x86 would involve an enormous amount of time and effort, just consider what RO4-Select by itself is taking to be fully 32-bitted. However, eventually it may still be needed for RO to survive for the desktop, but that certainly depends on a lot of factors which currently do not yet seem to be relevant.

hEgelia on 22/08/05 10:16AM
[ Reply | Permalink | Report ]

Porting RISC OS to x86 offers no real benefits over VirtualRiscPC, and a huge list of disadvantages for reasons reguarly discussed. In short, it's a bad idea.

mrchocky on 22/08/05 11:07AM
[ Reply | Permalink | Report ]

What are the advantages of the ARM based systems? None that I can think of. It's merely the complete impractability of such work that prevents it (surely a built-in emulation for backward compatibility would be feasible, if such an unfeasbile port existed).

And "merely" is, of course, giving slightly the wrong impression. It's the only significant barrier, but it is a very, very large one.

SimonC on 22/08/05 11:45AM
[ Reply | Permalink | Report ]

Even if not a single sale of native hardware has been lost due to VirtualA5000/VirtualRPC, there is the unfortunate fact that it has been percieved by some to be have an adverse impact on native hardware by some Software Publishers, peripheral Hardware manufactures and I expect native computer manufacturers. It is impossible to quantify the negative effect :-(

We here at CJE/4D have only ever supplied Laptop Computers with VirtualRPC because we want to encourage native hardware: Iyonix and A9home.

CJE on 22/08/05 12:14AM
[ Reply | Permalink | Report ]

Advantages of ARM based systems: RISC OS runs on them. They use less power. They're quieter and more reliable due to no CPU fans. They're easier to program in machine code. They're not evolved from 25 year old designs and therefore not (as) full of hacks. They're simpler and therefore more understandable. They don't have a BIOS

mavhc on 22/08/05 6:09PM
[ Reply | Permalink | Report ]

mrchocky: Would you not consider running the OS natively on hardware that's available off the shelf, costs much less and is many times more powerful a benefit? Plus having hundreds of millions of potential customers that don't need to buy different hardware or an emulator to run the OS seems to me a real benefit for the platform. Running the OS on an emulator in my opinion undermines the system's credibility, what other operating systems are run on an emulator? Outdated legacy operating systems for the most part. People will NOT convert (maybe a handful) to using an OS that requires an emulator to run - they will continue to use the host OS and possibly use the emulated OS for very specific tasks until a similar application has been developed for the host OS.

mavch: "RISC OS runs on them" - hmm, RISC OS actually runs on x86 and now PowerPC as well - albeit via emulation but people claim that it's actually faster than the "real thing" "They use less power" - Does that really matter? Not in laptops as ARM based laptops aren't exactly filling stores and homes! As has been pointed out many times, video cards and hard discs are the main culprits in modern computers - not the processor. Plus you can use a Pentium Mobile processor that's not so far above the XScale in power consumption that people actually care. "They're quieter and more reliable due to no CPU fans." - a definite and huge advantage. "They're easier to program in machine code." - Does this really make the slightest difference any more? Does anyone outside RISC OS program in machine code for desktop machines (other than OS of course)? "They're not evolved from 25 year old designs and therefore not (as) full of hacks." - But still the 25 year old design (which is actually closer to 30 years) is faster, cheaper and more readily available so what is the actual benefit here? "They're simpler and therefore more understandable." - Who actually cares? If the system works and is cheaper I can't see people giving a hoot. "They don't have a BIOS" - Just the CMOS instead. Not the same, I know but no system is perfect and the ARM based hardware is no gods gift to people.

Gulli on 23/08/05 00:00AM
[ Reply | Permalink | Report ]

Gulli:

I certainly agree with you that running RISC OS on an emulator undermines its credibility, making it look like a dead OS. But to me this is one reason why it's so important that the hardware market is kept alive.

Converting to native x86 won't solve this problem, or necessarily allow more people to run RISC OS: in fact, even if it were ported, it would still probably make more sense for most people to run it under emulation. Running it on top of Windows shifts all of the hard work of maintaining hardware compatibility onto Microsoft. There are so many variants of x86 hardware that it would be impossible for an x86 RISC OS to accommodate them all. Most people would probably find that it simply wouldn't run on their x86 machine.

One of the only reasons RISC OS can be made to work so well on the hardware that's available, is that there is so little variety in the hardware that it will run on.

Regarding your point that people don't care whether or not the machine code is easier to program, I agree that this may be true for the majority of people. However, I for one remain with RISC OS in part because of the ease and control you get programming the ARM chip directly. An emulator wouldn't suit me for this without the real hardware also being available.

Now this may not be what most people want, but I'd have thought the people most likely to care about this kind of thing are developers, and they are people that the platform will need if it's to remain viable.

By the way, I'm not knocking the emulator, which is clearly very useful. I just think the hardware market is totally essential to keeping the platform alive.

flypig on 23/08/05 01:08AM
[ Reply | Permalink | Report ]

Well said, Gulli.

jouster on 23/08/05 03:22AM
[ Reply | Permalink | Report ]

I've done this argument again and again. The point remains the same - there are huge disadvantages in making a "native" x86 RISC OS. You can start with the huge range of drivers that would be required. There are many more reasons, that I won't repeat, since I've already gone over them at least 5 times on drobe alone in the past.

mrchocky on 23/08/05 08:22AM
[ Reply | Permalink | Report ]

I also bought a Virtual A5000 just for getting the opportunity to see RISC OS (almost) in reality. Well, a key value for keeping an alterative OS alive is to have a userbase where people buy software. So it seems pretty possible that those 3000 who have registered (I'm not registered btw) could be such a "market". Because of it's age the A5000 may not be the right choice for this (VA5000 is more nostalgia), but I think the current projects are. If an emulator helps to hold users and developers then this is also good for any other RISC OS product (as long as third party software appears/remains).

The Emulator itself is really fine - I would guess that RISC OS is one of the best (or even the best) emulated OS in our days. I was really impressed by some RISC OS features and by the consistent way of usability. I wrote a small review about VA5000 in the Techhnoids Magazine. My conclusion is that you can even be productive with this emulator and that's an important fact, too. Of course you'll always see and feel that it is not a real RISC OS Computer but VA is damn close. After all I sold the VA5000 - I now have a Mac which not runs the software. Maybe I buy a more advanced Emulator for MacOS X.

mbreiter on 23/08/05 08:52AM
[ Reply | Permalink | Report ]

"Huge range of drivers." That's surely only an issue if you plan to make something useable on a huge range of hardware. Just because you're switching the processor you use doesn't mean you have to go for a disc loaded OS that has to cover everything, you could still sell a ROM based version that is tied to the specifics of the box its in, and would still be faster and cheaper than anything we've got now.

ARM is just a legacy we're stuck with, that's holding RISC OS back. It might be unfeasible to do anything about that, but any argument that says it's undesirable sounds more like wishful thinking, unless you've got pretty specialist requirements.

SimonC on 23/08/05 10:14AM
[ Reply | Permalink | Report ]

In reply to SimonC: Compared to RISC OS hardware you need a huge range of drivers just to support one model of machine from one manufacture for any period of time due to the constant changes of components.

But ignore drivers, assume you've ported the RISC OS operating system to x86, simulate the situation by deleting every single RISC OS application you own that isn't pure BASIC. Now wait for people to port applications across like they did for 32bit RISC OS on the Iyonix. Wait some more, and more, and more. You'll be waiting until the far side of never, as I don't see any RISC OS developers wanting to completely re-write millions of lines of nice ARM assembler, in to the stinking shambles of the x86 ISA. So you'll still be running all application under emulation, as and the RISC OS is very light weight you wont have gained more than a 5%-10% performance improvement for a huge amount of work and lots of compatibility issues.

But getting back to Virtual Acorn. I'm happy to buy a copy to take RISC OS with me where I have to use a PC, which is at work. But theres no way I'll ever replace native hardware with anything that has to run Windows underneath - why invite that sort of constant hassle and frustration in to your home?

druck on 23/08/05 11:01AM
[ Reply | Permalink | Report ]

It's clear that current ARM chip manufacturers have gone for low power consumption and cheapness rather than high power, but the ongoing development of PDAs and mobile phones, and their expansion into more power-hungry areas such as video will presumably spur the development of faster ARM chips and close the Mhz gap with x86 chips. After all, the latter have been stuck at around 3.2-3.4 Ghz for some time now. Another possibility would be the development of a multi-processor version of RO, but I presume this would be a mammoth undertaking, and would also require special application versions to run.

bucksboy on 23/08/05 12:52AM
[ Reply | Permalink | Report ]

I think it's more or less a strategic thing (like George says). If there would be a more powerful computer manufacturer then there would surely be more powerful than powersaving ARM CPUs. Or you need a chipfactory and construct your own powerful CPU (like it is done with any ARM CPU) - but then you would need a bunch of money, too. No money, no fun - sad but the truth.

mbreiter on 23/08/05 1:13PM
[ Reply | Permalink | Report ]

mrchocky: If you don't wish to take part in an "argument" why do you keep sticking your nose in it? Why do you bother commenting at all?

Gulli on 23/08/05 1:21PM
[ Reply | Permalink | Report ]

flypig: I asked wether or not people outside the RISC OS world actually use machine code programming on desktop computers for other than OS development and drivers because I know programmers have to use assembly language on RISC OS machines for speed optimisation. Even the vast majority of games developed on PCs don't have any machine code bits.

Just because you're stuck with it on ARM hardware doesn't mean you're stuck with it on other hardware.

Gulli on 23/08/05 1:25PM
[ Reply | Permalink | Report ]

Gulli: is such rude language really neccessary? On the flipside, if you want to pursue this line of argument, why can't you research what's been said on it in the past?

SimonC: and when the hardware changes every 3 months? Yes, you can in theory find workarounds for these issues, but that's just one of many.

mrchocky on 23/08/05 2:45PM
[ Reply | Permalink | Report ]

Why should the hardware change every few months, any more than it does with current RISC OS machines? That's a problem if you take an existing machine and try to put RISC OS on it, but not if you're building your own design with whatever components you choose. Then it may not be any cheaper, but it would be faster.

druck: Would it not be possible (albeit more complicated, since it's a completely different architecture) to do an Aemulor type solution to run existing software? That wouldn't provide much gain to begin with, admittedly, but new software would be a lot faster, and hopefully quite a bit of existing would just need recompiling. The Iyonix showed that there is a will to put the work in if faster hardware presents itself, although I admit that that was an order of magnitude or more less complicated.

If ARMs catch up with other processors, then great, let's stick with it, but if the performance gap keeps getting wider they will eventually be the millstone around the neck that drags RISC OS beneath the water. I believe people like Druck and Peter Naulls when they say it would be completely unfeasible, they know far more about it than I do (and have done a lot to keep it going), but it leaves me very, very worried. If it doesn't concern you then ask why the RISC OS market is far smaller than it was a few years ago.

SimonC on 23/08/05 3:50PM
[ Reply | Permalink | Report ]

Gulli:

Sure, you're right. There are very few reasons to use assembler on a desktop machine. This is pretty much the case even on RISC OS, where there are not really many occasions you'd need to use machine code in a program and the speed increase isn't generally that great anyway. So I wouldn't really say that this is something you're stuck with on RISC OS any more either.

I just wanted to point out that for some people (e.g. me) the access to ARM code development is a specific reason for sticking with RISC OS, so I wouldn't want it to change.

flypig on 23/08/05 5:29PM
[ Reply | Permalink | Report ]

Perhaps Mavhc's taking the mick a bit, but just in case:

"They use less power."

In the same way that a toy glider uses less power than a fighter jet. It's just that you sometimes only need a glider, other times you need a jet or somewhere in between.

"They're easier to program in machine code."

It's just a matter of taste. X86 isn't so bad, you just spend 20 minutes accepting it and getting over it. The segmentation is a headache, but it's not a show stopper.

"They're not evolved from 25 year old designs and therefore not (as) full of hacks."

In your opinion. Unless you know the ins and outs of the recent Pentium designs.

"They're simpler and therefore more understandable."

This takes us back to the glider vs. jet fighter argument.

"They don't have a BIOS"

What the hell is that supposed to mean? A BIOS is typically a simple, ROM based program that initialises a machine and then finds the start of the operating system, or a bootloader for the OS. ARM based systems have them (for example, my StrongARM Netwinder) just as much as any other system. For RISC OS, the 'BIOS' and the OS are rolled into one because it's all supplied in ROM. Someone else mentioned CMOS, which unfortunately also makes no sense. CMOS refers to a type of memory that can be kept persistent by a battery back up. It's not exclusive to a BIOS.

As it happens, Intel are announcing some new 'low power' cores, which I'm about to write up. In short (and in my opinion), ARM is best suited for embedded and 'fatter-than-ARM' (for example, Intel) is best suited for non-embedded. If you want to use RISC OS optimally on a non-embedded platform, you need a processor that's not aimed at the embedded world. Feel free to correct me, this is just my take on things.

diomus on 23/08/05 5:40PM
[ Reply | Permalink | Report ]

A thought that has just occured to me is that although current ARMs being aimed at embedded devices means we don't get nice fast ones that would make wonderful RISC OS machines, if they weren't any good at embedded would we even have RISC OS any more? If ARMs hadn't found such a niche it was very suited to the development of them might have stopped years ago, in direct competition to the x86.

SimonC on 23/08/05 6:08PM
[ Reply | Permalink | Report ]

mrchocky: I'm sorry if you felt my comment was rude, it surely wasn't ment to be. Just got a bit irritated at getting a condescending remark about everything I said being a load of rubbish without any willingness to explain it any further.

I'm not about to do a complete search of any comments you have made about the alleged complete lack of benefits of moving RISC OS to the x86 platform. If you insist on making comments like you did the least you can do is to show the courtesy of providing a link to at least one of your previous comments or anything that supports your opinion. Don't just jump into a discussion saying that you've said things a million times somewhere else - this is your opinion, support it somehow or stop telling people they're full of it.

Gulli on 23/08/05 6:09PM
[ Reply | Permalink | Report ]

diomus: I was the one that named the CMOS, it was actually meant to be ironic but I guess that failed.

Gulli on 23/08/05 6:11PM
[ Reply | Permalink | Report ]

Rather than spending millions of pounds porting RISC OS to x86, wouldn't it make more sense to get it running on a Crusoe laptop?

[link]

That way you could have a truly 'hybrid' computer, with the processor switching between an x86 and an ARM as needed.

(I'm not really serious, but to me this seems more likely to happen than RISC OS making the wholesale move to x86).

flypig on 23/08/05 6:49PM
[ Reply | Permalink | Report ]

1.25GHz Xscales announced.

They use less power: Per MIPS

25 Year old designs: I was referring more to the whole PC. By move to x86 do people want RISC OS to run on a standard PC motherboard?

If I need to blow up somewhere I'd have to spend years learning to fly a jet, for most uses a glider is better, especially if you're also interesting in how it works

BIOS: They don't have a seperate system for bootup, which can't be controlled from the OS.

> As has been pointed out many times, video cards and hard discs are the main culprits in modern computers - not the processor.

Those people are wrong.

> Plus you can use a Pentium Mobile processor that's not so far above the XScale in power consumption that people actually care.

Fanless P-Ms? excellent. URL?

> "They're easier to program in machine code." - Does this really make the slightest difference any more? > Does anyone outside RISC OS program in machine code for desktop machines (other than OS of course)?

Does to people interesting in how computers work, and people optimising code, eg for 3d graphics engines, simulations etc.

> "They're not evolved from 25 year old designs and therefore not (as) full of hacks." - But still the 25 year old design (which is actually closer to 30 years) is faster, cheaper and more readily available so what is the actual benefit here?

Simplicity, elegance. Fixability.

> "They're simpler and therefore more understandable." - Who actually cares? If the system works and is cheaper I can't see people giving a hoot.

Depends which people.

Some people drive cars from A to B, some people like to get under the bonet.

mavhc on 24/08/05 02:01AM
[ Reply | Permalink | Report ]

I'm a past user of RISC-OS, but haven't used it for some years (since my A5000 was moved to the basement to make room for a Wintel box my wife had to use for work). I intend buying a VPC for my laptop now with the special offer partly to run some of the old programs (!Draw and !Elite among them), but I also intend to do some experimental compiler work. For this use, the underlying CPU is quite important -- I don't fancy targeting x86 code -- so RISC OS on native x86 would have little appeal to me. But I guess this particular point would matter to quite few users.

torbenm on 24/08/05 2:06PM
[ Reply | Permalink | Report ]

mavhc: "Fanless P-Ms? excellent. URL? "

Sorry, but I never mentioned fanless P-Ms nor would I even fantasize about such a thing and I've seen no one do here. You've probably misunderstood something along the way.

Gulli on 24/08/05 5:12PM
[ Reply | Permalink | Report ]

> You've probably misunderstood something along the way Nah, it's just my Xtreme Sarcasm, to the max! <fx>air guitar</fx>

VRPC must be profitiable for ROL though, say 60ukps/copy, pure profit, no need to spend money on ROMs. So 3000*60=180000gbps

mavhc on 25/08/05 00:59AM
[ Reply | Permalink | Report ]

Mavhc> As much as I'd like to believe ROL made 180000 I doubt they got that much. Many of the copies sold may have been VA5000 - which as it was based on RO3.1 which *magically* became available (from where we know not) - and about which ROL objected - I doubt from the noise at the time that ROL got a penny from that version.

The later versions of VA (VARPC-SE and so on) *did* include ligit ROL Licensed version of RISC OS so *would* have contributed towards ROL's balance sheet. It would be helpful to know what *proportion* of the VA users used VA5000 compared to those using the later (ROL licensed) variants - perhaps the much vaunted Drobe survey will reveal all (drum roll - cue Chris Williams (opens envelope) and the winner is....

)

AMS on 25/08/05 7:12PM
[ Reply | Permalink | Report ]

...Microsoft

mavhc on 25/08/05 10:19PM
[ Reply | Permalink | Report ]

Gulli: "complete search" - but it looks like you haven't made any search at all. On the contrary, I've gone to quite a lot of effort to explain this issue in the past, it's to you to put some effort in also. I've certainly not said anyone was "full of it".

Alas, this long thread shows this probably isn't about to happen, and the wild ideas will continue without any kind of foundation. Fortunately, druck has outlined some of the serious problems in making it happen. And I've written the whole thing up, so it's there once and for all, and we can go onto more productive use of our time, and avoid the 11th iteration of this discussion.

[link]

mrchocky on 26/08/05 1:34PM
[ Reply | Permalink | Report ]

mrchocky: Sorry, been a bit busy lately and haven't had the time to check out what you posted and currently all I can get is a "Bandwith exceeded" message - will check again on Thursday or Friday.

Gulli on 30/08/05 5:27PM
[ Reply | Permalink | Report ]

Please login before posting a comment. Use the form on the right to do so or create a free account.

Search the archives

Today's featured article

Click right on with RISC OS
Martin Hansen runs through a few useful desktop tricks you may not be aware of

16 comments, latest by Anon on 26/3/09 1:01PM. Published: 22 Mar 2008

Random article

New XScale supports PCI-Express
800Mhz IOP80332

16 comments, latest by NoMercy on 2/8/04 12:59PM. Published: 30 Jul 2004

Useful links

News and media:
Iconbar • MyRISCOS • ArcSite • RISCOScode • ANS • C.S.A.Announce • Archive • Qercus • RiscWorld • Drag'n'Drop • GAG-News

Top developers:
RISCOS Ltd • RISC OS Open • MW Software • R-Comp • Advantage Six • VirtualAcorn

Dealers:
CJE Micros • APDL • Castle • a4 • X-Ample • Liquid Silicon • Webmonster

Usergroups:
WROCC • RONE • NKACC • IRUG • SASAUG • ROUGOL • RONWUG • MUG • WAUG • GAG • RISCOS.be

Useful:
RISCOS.org.uk • RISCOS.org • RISCOS.info • Filebase • Chris Why's Acorn/RISC OS collection • NetSurf

Non-RISC OS:
The Register • The Inquirer • Apple Insider • BBC News • Sky News • Google News • xkcd • diodesign

Â© 1999-2009 The Drobe Team. Some rights reserved, click here for more information
Powered by MiniDrobeCMS, based on J4U | Statistics
"I see Drobe'll suck up any old s*** as news?"
Page generated in 0.1597 seconds.