What is Real Time Operating System?

On 26 Mar 2004 22:35:19 -0800, sfbong@tm.net.my (Allen) wrote:
>I saw the word "RTOS" appeared many times in this group but have no
>idea what it is.  So could the experts here give me a brief idea what
>is it in non-technical terms?
This means that it reacts fast enough on your process to be able to
control it "really in time". So if something is realtime depends on
your process - software combination.

In practice, by real-time systems are meant that react fast. So a
windows systems that makes your program halt for 2 seconds is not
real-time.

>Which OSes in the market are qualified as RTOS?
There are a lot, see other newsgroup messages.

It is just like the word "embedded".

I usually work with embedded systems that conatin a small processor,
some electronics and firmware to control machines etc. That is my view
on "embedded".

But I also have a customer that has huge servosystems that they
control with PC's with Windows NT (RealTime..), vision software,
windows based robot software, and they call such a system "embedded".
Probably because it "is embedded in the machine".

Pieter

On Sat, 27 Mar 2004 12:36:23 GMT, CBFalconer <cbfalconer@yahoo.com>
wrote:

[...re: "What is an RTOS?"...
>
>A system where you can make a guarantee of the maximum interval
>between requesting an action and performing that action.

I mostly agree with this...

>
>In other words it can't continuously put it off because something
>more important came up.

But I strongly disagree with this.  As least the way it's written.

In fact, I'd say one of the primary differences between an RTOS and a
Desktop OS is that the RTOS *can* indefinitely put something off
because something more important came up (but not because something
*less* important came up).

A Desktop OS has "fairness" as a major design goal -- don't starve any
philosopher.  An RTOS replaces "fairness" with "correctness" -- just
because the printer driver hasn't run in the last ten seconds doesn't
mean it gets to run before the actuator control task is finished
making sure there's no collision with the robot arm...

Regards,

                               -=Dave
-- 
Change is inevitable, progress is not.

Dave Hansen wrote:
> CBFalconer <cbfalconer@yahoo.com> wrote:
> 
> [...re: "What is an RTOS?"...
>>
>> A system where you can make a guarantee of the maximum interval
>> between requesting an action and performing that action.
> 
> I mostly agree with this...
> 
>>
>> In other words it can't continuously put it off because something
>> more important came up.
> 
> But I strongly disagree with this.  As least the way it's written.
> 
> In fact, I'd say one of the primary differences between an RTOS
> and a Desktop OS is that the RTOS *can* indefinitely put something
> off because something more important came up (but not because
> something *less* important came up).
> 
> A Desktop OS has "fairness" as a major design goal -- don't starve
> any philosopher.  An RTOS replaces "fairness" with "correctness"
> -- just because the printer driver hasn't run in the last ten
> seconds doesn't mean it gets to run before the actuator control
> task is finished making sure there's no collision with the robot
> arm...

Well, let's reword it then :-)  "Because it fabricated something
more important to deal with" maybe.  Obviously all bets are off if
it is given tasks beyond its ability, such as exceeding maximum
throughput.   But we have to be able to say that if task a appears
no more than every A millisecs, and task b every B millisecs, etc.
then task c will get handled within CC millisecs of a c request. 
And so forth for each and every task.  What delay constitutes real
time can vary with the task, but must be predicatable.

At any rate, I wanted a really short summary :-)

-- 
fix (vb.): 1. to paper over, obscure, hide from public view; 2.
to work around, in a way that produces unintended consequences
that are worse than the original problem.  Usage: "Windows ME
fixes many of the shortcomings of Windows 98 SE". - Hutchison

> 
> The license must have changed a lot since then -- now it's a couple grand.

I hope i read my email wrong, but i was told by someone that i have to
pay $2500 PER UNIT (!?!?) sold. I hope he meant that i pay a one-time
$2500 fee for that product, and then am allowed to sell as many as i
want to. Otherwise, there is no way i can price my product at $300.
I'd be making a $300-$2500 = --(-$2200) loss per product. oh well.

Mike V. <valemike@yahoo.com> says...

>> The license must have changed a lot since then -- now it's a couple grand.
>
>I hope i read my email wrong, but i was told by someone that i have to
>pay $2500 PER UNIT (!?!?) sold. I hope he meant that i pay a one-time
>$2500 fee for that product, and then am allowed to sell as many as i
>want to. Otherwise, there is no way i can price my product at $300.
>I'd be making a $300-$2500 = --(-$2200) loss per product. oh well.

Yes, but you make it up by volume... <grin>



-- 
Guy Macon, Electronics Engineer & Project Manager for hire. 
Remember Doc Brown from the _Back to the Future_ movies? Do you 
have an "impossible" engineering project that only someone like 
Doc Brown can solve?  My resume is at http://www.guymacon.com/ 

 said...
> 
> Mike V. <valemike@yahoo.com> says...
> 
> >> The license must have changed a lot since then -- now it's a couple grand.
> >
> >I hope i read my email wrong, but i was told by someone that i have to
> >pay $2500 PER UNIT (!?!?) sold. I hope he meant that i pay a one-time
> >$2500 fee for that product, and then am allowed to sell as many as i
> >want to. Otherwise, there is no way i can price my product at $300.
> >I'd be making a $300-$2500 = --(-$2200) loss per product. oh well.
> 
> Yes, but you make it up by volume... <grin>

That was the same concept for The Change Bank, wasn't it?



Casey

In article <8188616d.0404041458.16bd20aa@posting.google.com>, Mike V. wrote:

>> The license must have changed a lot since then -- now it's a couple grand.
> 
> I hope i read my email wrong, but i was told by someone that i
> have to pay $2500 PER UNIT (!?!?) sold. I hope he meant that i
> pay a one-time $2500 fee for that product, and then am allowed
> to sell as many as i want to.

Exactly.  IIRC, there's also a site-license available (for an
unlimited number of products all developed at a particular
location).

-- 
Grant Edwards                   grante             Yow!  My Aunt MAUREEN was
                                  at               a military advisor to IKE &
                               visi.com            TINA TURNER!!

Wait,so it is a one-time fee PER PRODUCT designed? Or i keep paying
that fee for every unit I ship?


Guy Macon <http://www.guymacon.com> wrote in message news:<a4qdnZgR2pocL-3dRVn-sA@speakeasy.net>...
> Mike V. <valemike@yahoo.com> says...
> 
> >> The license must have changed a lot since then -- now it's a couple grand.
> >
> >I hope i read my email wrong, but i was told by someone that i have to
> >pay $2500 PER UNIT (!?!?) sold. I hope he meant that i pay a one-time
> >$2500 fee for that product, and then am allowed to sell as many as i
> >want to. Otherwise, there is no way i can price my product at $300.
> >I'd be making a $300-$2500 = --(-$2200) loss per product. oh well.
> 
> Yes, but you make it up by volume... <grin>

On 2004-04-05, Mike V. <valemike@yahoo.com> wrote:

> Wait,so it is a one-time fee PER PRODUCT designed?

Yes.

> Or i keep paying that fee for every unit I ship?

No.

>>>> The license must have changed a lot since then -- now it's a couple grand.
>>>
>>>I hope i read my email wrong, but i was told by someone that i
>>>have to pay $2500 PER UNIT (!?!?) sold.

Who told you that?

-- 
Grant Edwards                   grante             Yow!  I predict that by
                                  at               1993 everyone will live in
                               visi.com            and around LAS VEGAS and
                                                   wear BEATLE HAIRCUTS!