mailto: blog -at- heyrick -dot- eu
Getting a heater going
I dug up the old oil-filled radiator thing to get it running again. Not because I felt I needed it, so much as it was a useful thing to finish drying damp clothes, as the outdoor drying line in the hangar is good but there's always a little bit of dampness that remains this time of the year.
It was, actually, rather worse. I've already been over it with the vacuum. But there's a lot more inside to tidy up.
Rick + Screwdriver = Viscera!
Do you see the heating element of the radiator part? Do you see it looks damp? This will become important later. I dried it off, and I'm guessing it's some sort of condensation, but...
After a lot of cleaning, time to reassemble and fire it up.
It was pretty obvious that the fan heater would spit out some smoke given that it's not been used since forever. There was decades of dust in there. I think I cleared most of it out, but not all. So I decided to run it for a few minutes.
That picture above is after about ten minutes. I had to open the doors each end and get some airflow through here due to all the smoke.
And not a peep from the smoke detector. I think it needs a new battery, though it does scream at me if I push its button. <shrug>
I switched to the radiator and... after about twenty seconds - CLUNK!
Clunk? What clunks in a radiator?
It wasn't. It was the RCCB tripping out. I reset it and... CLUNK! It was then that I remembered that the Pi was on (and off). A quick check with DiscKnight and everything is good. Phew!
I brought through an extension lead from where the washing machine plugs in and ran it from that. If the sensitive trip switch can't cope, then I'll just run it through the insensitive one. It's okay, I'm not a special little snowflake, I grew up in the eighties, I can handle insensitive. ☺
Anyway, a few hours later and I'd imagine whatever was causing the problem has evaporated. As for my clothes? Well...
Nice and dry.
The future of RISC OS?
There seem to be two main ideas being discussed on the ROOL forums. The first, and the one with possibilities, is to create an emulation solution to run on 64 bit machines. It will, effectively, emulate an environment in which 32 bit RISC OS and its 32 bit applications can run.
To my mind, this is the only sensible solution as it permits one to go forward with the minimum of disturbances. Obviously the initial target of this is likely to be something "sort of RiscPC like" as the drivers and such already exist for it, with a future project to start to revise parts of the OS to be able to talk directly to the emulator rather than believing it's a certain type of machine (and in this manner could, for example, ultimately paste data directly into the frame buffer without it needing to be translated on the fly).
The alternative is to rewrite RISC OS to run on 64 bit hardware, most likely rewriting it in C. This, I feel, is doomed to failure for two reasons. The ARM64 is sufficiently different that retaining the existing API would be next to impossible, and all of those legacy apps may only work on silicon that supports a 32 bit user mode world. A fair few SoCs do so at the moment, but the writing is on the wall for that. Soon chips will be 64 bit exclusively. Which means all that 32 bit RISC OS stuff? Won't work. Won't even get close to sort of maybe working. Those who think these are just little road bumbs...probably don't understand the scale of the differences.
Sveinung, on the ROOL forum, put the problem into perspective when he said: It's said to take around 20 man-years to do a full 64-bit conversion, in C, total wages at around a million UKP/year.
Of course, I called him on that. Very very few programmers in the world make €600/hr. And absolutely none of them work in the RISC OS world. As for 20 man years? Possible, but given the accuracy of the salary, I think we'd need a second opinion here.
Please note that while I check this page every so often, I am not able to control what users write; therefore I disclaim all liability for unpleasant and/or infringing and/or defamatory material. Undesired content will be removed as soon as it is noticed. By leaving a comment, you agree not to post material that is illegal or in bad taste, and you should be aware that the time and your IP address are both recorded, should it be necessary to find out who you are. Oh, and don't bother trying to inline HTML. I'm not that stupid! ☺ ADDING COMMENTS DOES NOT WORK IF READING TRANSLATED VERSIONS.
You can now follow comment additions with the comment RSS feed. This is distinct from the b.log RSS feed, so you can subscribe to one or both as you wish.
|David Pilling, 6th November 2023, 02:18|
Heaters usually come with a sticker "do not cover". Seemingly LIDL sell clothes racks with inbuilt heating.
But I came for 64 bit RISC OS... read the mind of ROOL:
I am surprised they're taking it seriously, I thought they would stick with 32 bits. I think if there is a project launched things might get done. At the moment there seems to be a situation of highly motivated individuals doing their own things.
I'm not clear if a 64 bit RISC OS would propagate all the 32 bit RISC OS problems, like no threads/multiple cores use.
|jgh, 6th November 2023, 14:33|
Yes, I had a hand in writing an ARM64 assembler, and it is so different from ARM32 that it really should have a different name. It's not just ARM with 64 bits, it's like the difference between 6502 and ARM32. A completely different processor. (At the instruction coding level, more like 68000 - ie, a complete mess.)
|Stuart Swales, 7th November 2023, 16:09|
"I am surprised they're taking it seriously"
You and me both.
"all the 32 bit RISC OS problems, like no threads/multiple cores"
Stuff that could be being worked on right now, to get the most out of our existing hardware.
And who cares about ARM64 per se? 'Cos it starts with ARM? So what.
|David Pilling, 7th November 2023, 23:19|
'who cares about ARM64 per se?'
Quite right, once you take the next step you don't care about hardware.
I've never looked at the ARM64 instruction set, some people say that it is "ARM32 without the conditional execution", the ROOL document talks about mapping the SWIs.
Talking about not caring about the hardware, they struggle moving Windows to new hardware, like ARM. What is the Mac story, at what point did they make that OS independent of hardware.
20 years ago, I ported some of my Windows versions of RISC OS stuff to 64 bit Windows. I did not do Ovation Pro, I could see 32 bit fields in the file format looking at me.
So you could have 64 bit ARM applications that only understand 32 bit addresses.
I have an idea to see if I can compile Ovation Pro for Windows on ARM Windows - what will it be 32 bit or 64 bit.
|jgh, 8th November 2023, 20:34|
"ARM32 without the conditional execution"
Good god, no way! Andybody saying that has clearly not actually looked at the instruction set. This is as far as I got in writing an ARM64 assembler, it shows how much of a mess the instruction set is: mdfs.net/Software/BBCBasic/SDL/bbasmb_arm_64.c
There's so much where you start to encode orthogonally, and then suddenly it goes all off piste and full of exceptions.
(Felicity? Marte? Find out!)
- I survived - really, How the videos were made, SimpleSeq v0.19. (2023/12/02)
- I survived!, Afterwards, Fast food from a machine. (2023/11/28)
- SimpleSeq v0.18. (2023/11/26)
- Winter is coming, Christmas cards, Torygraph purchase, A foreigner's guide to mainstream British newspapers, Redmi Note 12 Pro low light photos, The creeping tentacles, Domestic stuff. (2023/11/25)
- Inside a USB DVD player. (2023/11/23)
List all b.log entries
Return to the site index
PS: Don't try to be clever.
It's a simple substring match.
Last read at 14:36 on 2023/12/04.
© 2023 Rick Murray
This web page is licenced for your personal, private, non-commercial use only. No automated processing by advertising systems is permitted.
RIPA notice: No consent is given for interception of page transmission.