68k details

[roondar]

Originally Posted by Megol

Intel iRMX was released in 1980. Then we have other released soon after the IBM PC release: Xenix and QNX.

Was actually thinking of Xenix but that released later though on a non-IBM system.

There were a lot of multitasking systems and extenders for PC systems including DESQview (and later DESQview X), Topview, as well as DOS compatible systems with multitasking (intended for multi-users actually) that a quick search says was Concurrent DOS. I remember at least another system but can't remember the name, searching doesn't help either.

I've checked the iRMX page on Wikipedia and this is pretty impressive. Released in 1980 for the 8088 and 8086.. The features are really cool:

Originally Posted by iRMX Wikipedia

iRMX supports multiple processes (known as jobs in RMX parlance) and multiple threads are supported within each process (task). In addition, interrupt handlers and threads exist to run in response to hardware interrupts. Thus, iRMX is a multi-processing, multi-threaded, pre-emptive, real-time operating system (RTOS).

Thanks for the info, I like knowing that kind of stuff

[Bruce Abbott]

Originally Posted by plasmab

Is there an auto code beatifier/formatter for 68k.

68k code doesn't need beatifying - it's naturally beatiful!

[plasmab]

Originally Posted by Bruce Abbott

68k code doesn't need beatifying - it's naturally beatiful!

Sigh. No software professionals here i guess.

[roondar]

Originally Posted by plasmab

Sigh. No software professionals here i guess.

Plenty of them I think. I personally don't need a 68k beautifier though, assembly language is not that hard to format to my liking

[pipper]

The question for an m68k assembly beautifier is not far fetched. In particular if you start of with an old big project, it is helpful to first go through and reformat and unify all the code to your personal liking. This helps reading and understanding the code. I recently started looking into the Alienbreed 3D II source and I wish I had some automated tools to unify all the mess So far SED worked out ok, but a ‘real’ dedicated tool would be nicer.

[plasmab]

Originally Posted by pipper

The question for an m68k assembly beautifier is not far fetched. In particular if you start of with an old big project, it is helpful to first go through and reformat and unify all the code to your personal liking. This helps reading and understanding the code. I recently started looking into the Alienbreed 3D II source and I wish I had some automated tools to unify all the mess So far SED worked out ok, but a ‘real’ dedicated tool would be nicer.

Ok I really didn’t want to derail things. I wanted consistent tabs, capitalisation, spaces after commas, exactly 2 lines between subroutines, casing on labels. The sort of stuff that gets enforced at professional level.

I was reworking someone else’s code base and wanted to get some consistency. Perfectly content with no as a answer. But all code can be beautified.

[Megol]

Originally Posted by meynaf

It may look delusional and still be true.

But it isn't true. I posted just some examples, there are more.
I also posted some examples of other systems that had proper multitasking very late (68k based Macintosh) or never (ARM based RiscOS). There are more.

If "ludicrous" and "delusional" are you only arguments, beat it.

They weren't.

But even if they were would those be better than your kind of arguing? You often use even shorter, unsubstantiated yes/no as an answer. You often just refuse to give supporting data, you argue about things that are obvious you know nothing about. So how about you follow your own advice?

[roondar]

Originally Posted by plasmab

Ok I really didn’t want to derail things. I wanted consistent tabs, capitalisation, spaces after commas, exactly 2 lines between subroutines, casing on labels. The sort of stuff that gets enforced at professional level.

I was reworking someone else’s code base and wanted to get some consistency. Perfectly content with no as a answer. But all code can be beautified.

This will be my last post on this subject, I promise.

My problem with code beauty standards is that I've found them to be more a representation of how whoever wrote them likes the code to be and what is 'hip' at the time it was written, rather than a rational case for what makes sense and why this particular way is the best option. This is compounded by how one persons readable is another's mess.

That said, I would agree that outside of hobby projects it is a very good idea to all write code conforming to the same layout standard. Which does not require a beautifier

[meynaf]

Originally Posted by Megol

But it isn't true. I posted just some examples, there are more.
I also posted some examples of other systems that had proper multitasking very late (68k based Macintosh) or never (ARM based RiscOS). There are more.

If these so great multitasking systems were so cool, why did they just fade out ? How to check them now ? Because it's written on wikipedia then it is necessarily true ?
It's easy to say they had all these great features. But that's in no manner a proof it was really done properly, or at least, that it was easy enough to do.
Now what will you find next ? Systems with 8086 which had full memory protection ?

But anyway - it's just irrelevant.
What i'm basically saying is that it's a PITA to do with x86 - not that it can not be done at all. Same story as writing whole asm programs.

Originally Posted by Megol

They weren't.

But even if they were would those be better than your kind of arguing?

My kind of arguing is right. Yours is wrong. What you're actually doing is called a straw man fallacy.
Proof : have i said anywhere that it was impossible to multitask on old x86 ?
I just wanted to point it had to be clumsy, inefficient.
But no. You want me to mean it was just impossible and believe you're countered my arguments, where you clearly have not.

Originally Posted by Megol

You often use even shorter, unsubstantiated yes/no as an answer.

That's just not true. But ppl have the tendency to write the kind of sentence like yours above, simply because they have nothing to backup their claims and they badly want me to be wrong.
At least I don't extract a part of what the other says, attempt to refute it, and pretend i've refuted the whole.

Originally Posted by Megol

You often just refuse to give supporting data,

Wrong again. I give supporting data, and you just don't. If asked, i can give code, and you don't.

Originally Posted by Megol

you argue about things that are obvious you know nothing about.

What a typical, useless personal attack.
If i were a moderator i'd emit warnings/bans for this kind of sentences.

Originally Posted by Megol

So how about you follow your own advice?

What advice ?

[meynaf]

Originally Posted by roondar

That said, I would agree that outside of hobby projects it is a very good idea to all write code conforming to the same layout standard. Which does not require a beautifier

Well, I wouldn't. I like the ability to know who in the project wrote the code by just looking at how it's formatted. This way you can go straight to the right person without having to ask everyone.
In addition, reformatting code while reading it lets me know immediately the parts i have deciphered and the parts i have not.
And it is wrong idea that conforming everyone to the same layout enhances readability - it does so only for the one who decided of said layout !

[plasmab]

Originally Posted by roondar

This will be my last post on this subject, I promise.

My problem with code beauty standards is that I've found them to be more a representation of how whoever wrote them likes the code to be and what is 'hip' at the time it was written, rather than a rational case for what makes sense and why this particular way is the best option. This is compounded by how one persons readable is another's mess.

That said, I would agree that outside of hobby projects it is a very good idea to all write code conforming to the same layout standard. Which does not require a beautifier

That’s why you have beatification. To change from standard A to B. The standard itself is broadly irrelevant. Consistency is all I look for.

EDIT: I really should have posted this line of enquiry in its own thread. Sorry

Remember when you go for a job interview your potential employer can lookup your published hobby code and probably won’t judge it with hobby eyes.

[StingRay]

Originally Posted by plasmab

Remember when you go for a job interview your potential employer can lookup your published hobby code and probably won’t judge it with hobby eyes.

Hobby code is hobby code and should not be judged for any job interview! If you publish your hobby code using your real name so any employer can look it up that's your very own fault anyway.

[plasmab]

Originally Posted by StingRay

Hobby code is hobby code and should not be judged for any job interview! If you publish your hobby code using your real name so any employer can look it up that's your very own fault anyway.

While this is true, if you want to keep the copyright on code you need to put your name on it. First thing I do when I get a CV is look for their code on github.

[roondar]

Originally Posted by plasmab

Remember when you go for a job interview your potential employer can lookup your published hobby code and probably won’t judge it with hobby eyes.

Well, if a company can't see the difference between doing something for fun and doing something professionally and how this might lead to more lax standards, then frankly I'd never want to work there to begin with. I'd rather work for a company that shows it understands how the real world works

Secondly, if they do look up my code and judge my abilities and merits based on whether or not I hold to some internal coding standard they've decided on using (instead of realising I might have my own preferences and reasons for using the style I use), then frankly they're being really short sighted. And since there is no shortage of demand for skilled people - well, it's their loss.

While this is true if you want to keep the copyright on code you need to put your name on it. First thing I do when I get a CV is look for their code on github.

Is this before or after you've checked their Facebook to make sure they didn't do something silly ten years back?

In all seriousness, I'd never do that without prior approval from the candidate in question. I'd consider doing so an act of bad faith (not the looking up, the not telling you're doing so). Now - if they don't want me to look it up that will probably lead to me asking why, but that's a different story.

But doing it without telling them... No, that's just not acceptable in my view.

[plasmab]

Originally Posted by roondar

But doing it without telling them... No, that's just not acceptable in my view.

Happens. Always will. I’ve never not interviewed anyone on the basis of this because it might be mistaken identity. Much of the time candidates point you at their hobby code.

You look at the code to see if it’s mostly consistent or a dogs breakfast. Also to see if there is really good patterns used that I would want championed in my team. Complete unit test coverage is something that would make interview someone I might otherwise not have. It’s not all negative.

I don’t see why you are getting so upset. If you’re a pro make your hobby code to a good standard.

EDIT: Anyways its free advice. Take or leave. I cant prove companies do this. Nor do I care either way.

[nogginthenog]

Originally Posted by plasmab

While this is true, if you want to keep the copyright on code you need to put your name on it. First thing I do when I get a CV is look for their code on github.

Funny you should mention that. I interviewed a guy today and his github code was awful. OK, it's a junior support/development role but still.
He didn't know what XML or JSON was either.

[roondar]

Originally Posted by plasmab

I don’t see why you are getting so upset.

Fair enough, I'll try to explain my point of view in a calmer way.

It's a combination of two things. The first is that I'm (and this might not be surprising ) really quite big on privacy. Basically, I just don't feel that it's OK to look people up on the internet and see what they have or have not done (essentially I'd disagree with the statement that putting something on the internet is equivalent to making yourself a public figure). I won't get into the reasons as this stuff is far enough off topic as is, but that's part one of it.

If you’re a pro make your hobby code to a good standard.

Now, part two is more about what you write here.

I personally do not agree that this is how it should be. You may disagree, but to me a hobby is "for Sh-ts 'n' Giggles" and shouldn't really be taken seriously by the outside world unless either the person involved makes it clear it is meant to be taken seriously, or there are other good reasons to assume the person involved wasn't just having some fun. I just really don't like the idea that you can only ever do things the "proper" way if you happen to be a pro. I have my time off, I just want to have some fun and not worry about coding standard of the week #12.

And I say all this, while I'm an absolute stickler about proper indentation, proper white space use, etc

Anyway, that might explain it better. I'm not really trying to change people's minds here.

[plasmab]

I get all the arguments. The rights and wrongs etc. But the real world is pretty harsh. BTW it is even more important to get your published code in order if you are a contractor. Essentially employment law does not exist for you

[Thorham]

Originally Posted by litwr

68000 two stacks was theoretically wrong

No, it's not. What happens when an interrupt occurs when you only have one stack? The interrupt writes to the user stack. Why is this a problem? Because it can write data past the end of the stack causing program or data corruption. Next, the interrupt handler has to change the stack pointer in order not to increase the risk of corruption.

Having a user stack and a supervisor stack keeps two completely different worlds separated from each other witch is a great advantage.

Now explain why having two stacks is not good, because I really want to hear that.

[Bruce Abbott]

Originally Posted by plasmab

all code can be beautified.

Sure, but can it be beatified?

One problem with using a modern web browser is that you rely on it to correct spelling mistakes. Now that I am using IBrowse on the Amiga I have to be a lot more careful.

Originally Posted by pipper

The question for an m68k assembly beautifier is not far fetched. In particular if you start of with an old big project, it is helpful to first go through and reformat and unify all the code to your personal liking. This helps reading and understanding the code.

Yes, and I like to do it manually because while I am reformatting the code I am also getting familiar with it.

I recently got back into programming on the Amiga. 20 years ago I thought my coding style was pretty good, but on review it looks quite messy. One bad habit I carried over from earlier machines was setting the assembler to case-insensitive, so some of my variables had different capitalization in different places. Turned case-sensitivity on for compatibility with OS 3.1 includes, and all my code broke! The good news is that since going through it to fix the case sensitivity issue it doesn't look so foreign anymore...