Re: p2c (dave gillespie's pas to c converter) & VMS pascal
- From: Baldrick <none@[127.0.0.1]>
- Date: Wed, 01 Feb 2006 11:28:39 +0000
Stanley F. Quayle wrote:
On 31 Jan 2006 at 17:48, Baldrick wrote:
I suspect it is not _real_ time but just happens to be fast enough, and that if push came to shove it would fail. Like, when 5ms must and has to be, without exception, 5ms. If windows does what Windows does best at that critical time, you're toast.
I'm not sure where the "shove" would come from. All Windows services are disabled except for those required by the product. Also, supported configurations require at least 1 processor more than the number of VAX processors being emulated. This allows "overhead" functions and network interface emulation to stay out of the VAX processors' way.
It's merely an expression (UK only), Bob Koehler makes a good point of what real time should be, and is in the real world.
How many real time operating systems are there out there?
And this is absolutely nothing to do with how many windows services are enabled, disabled, hacked, massaged, or whatever. Windows is not a real time O/S, period.
I'm trying hard to explain this in other terms. OK bit off the wall, but lets say you have this wheel thing, a round thing that rolls really. So that's your VAX.
Now, if you then emulate your VAX, but on a device which is not round, but square, so not smooth, it bump bump bumps. Therefore, no matter how smooth and round your wheel, if you are being carried by that square one, you cannot escape its effects.
I know that you are talking a 2 CPU system, but they are intrinsically linked, an axle, one round, one square, if you shut down windows, you emulator would vanish too, it really has no bearing, windows is that square wheel you cannot get far enough away from, when it absolutely must be round (real time). 2 CPUs don't count.
I hope that analogy works.
> There is a version of CHARON-VAX that uses a real-time OS under it. > Perhaps that would be more to your liking...
You mention a "real time" option that the VAX can run on. What is this operating system, is it really capable of real time itself (i.e. has it had the roundness test)? e.g. device interrupt going directly to a process, not being intercepted, handled (is it for me or the hardware I'm running in a process?), but that hardware emulation could be located anywhere in the memory space, and potentially with different latency conditions, albeit in milli or even micro seconds, but that is all the difference when it is real time.
If the emulator ran directly on the hardware, with no possible other software (operating system, call it what you may) as a managing layer then you'll have something that would pass as real time. (However I've just described a VAX haven't I?)
Real time means what it says, no artificial additives. If you have an engineer, working on something or other, and expecting that black and white is just that, no possibility of fuzzy grey edges, you could just be introducing an unknown and indeterminate into their theories.
I don't believe it matters whether some alternative is more to my liking, it's not a question of that. It is that 1 out of 1000, or more, or less, when real time is not real time, but real time +/- with potential catastrophic results. Where do you want to be today?
I worry about Boardroom Engineering.
-- Regards, Nic Clews a.k.a. Mr. Car Park Charges, CSC Computer Sciences nclews at csc dot com .
- Prev by Date: Re: Splitting serial cables ?
- Next by Date: Identify the source of a machinechk
- Previous by thread: Differences between KA690 and KA691
- Next by thread: Re: p2c (dave gillespie's pas to c converter) & VMS pascal
- Index(es):
Relevant Pages
|
