That's a far reaching generalization to call that methodology anything like
systembuilder.
For a long time, I've shared a similar mentality to Barry's whereby it
irritates me to no end how juvenile and immature the AV programming
community as a whole is (I concede that there are obvious exceptions, but
for the case of this argument let's assume I'm right).
If you look at ANY other software industry and compare it to ours, the
contradictions are glaring. Why? Because we tend to cling to the silly
notion that we're "different" because we deal with A/V and archaic
proprietary RS-232 protocols and such....
The reality is that we're no different. We produce computer software. We
write code. We build graphical user interfaces. Guess what? We're not
exactly pioneers here....
I think it was Nathan in the thread about locked code/modules lately that
made a statement that brought a smile to my face and gave me hope. The
world of PC programming discovered years ago that protecting common code
that almost every one of their competitors was using wasn't making them
more competitive or profitable and that it was far more beneficial to
themselves, and the industry as a whole, to share such code amongst each
other. These days, most PC programmers rarely write anything truly
original, instead they use OOP libraries and class modules to adapt for
their purposes. The originality, profitability and ingenuity comes in
combining them in some new and more effective way that advances the
efficiency or capability of the software. It's pretty hard to get to that
point if you have to start from scratch and write everything yourself.
With a solid foundation (i.e. Andriod OS) that is developed and improved in
a open source environment, there is still a TON of room and market to build
innovative profitable products.
I watch my own team of ~10 programmers constantly each spend hours trying
to solve the same problem on their own. While I value each person's
approach and don't really have an opinion on which way is better as long as
the finished product is the same, it's terribly inefficient for the company
and themselves. What AV programmers need to realize is that we would all
make more money and do less work if we behaved more like the rest of the
programming world. Find the common problems and scenarios, solve them
generically, efficiently and in a way that is reusable in as many
situations as possible and instead spend your time and effort on the
problems and scenarios that are uncommon or adding new features that add
value.
</rant...>
toggle quoted message
Show quoted text
On Fri, Apr 13, 2012 at 12:20 PM, matt_rasmussen_2000 <mjrtoo@...>wrote:
**
So you want to write another systembuilder?
--- In Crestron@..., Barry Newton <barry@...> wrote:
You know, this really gets at programming technique which is something
Crestron has always stayed out of. They offer best practices for specific
situations sometimes, but overall architecture is absent. So like most
programmers with experience, I have my own approach I use over and over.
There are a lot of unpaid hours that went into it, but that time is
recovered in being able to produce finished programs more quickly.
I suspect most programmers eventually abandon timed logic for the most
part and begin thinking in terms of logic wave pulses and wave delays where
that sort of thing is needed. I know I spend an unhealthy amount of time
tweaking logic just to reduce the number of symbols.
I have wondered if there is interest in the community in creating a
standard architecture for program design, something like an open source
project. In my mind the focus would be creating an overall architecture
with cut-and-paste-ready logic modules with well defined standard
interfaces. There are a lot of device modules in the files section, but I
haven't seen a lot other than that, and there is no standardization at all
really.
Maybe most of you have too much invested in your own approaches to want
to share techniques, even I'm not sure I want to do that, but it won't hurt
to discuss it.
Sorry to hijack the thread.
Barry
On Apr 12, 2012, at 6:54 PM, Steve Kaudle wrote:
To some extent that's dependent on the display's in question (or more
specifically, the completeness of their control API's). Lets assume
that the displays can either be queried and/or will provide
unsolicited
feedback for the current power state (off, warming, on, or cooling).
Given that, I'd write a module that would always tell me the current
power status, and if either of the display's 'warming_fb' outputs were
high, activate the subpage.
On 4/12/2012 6:32 PM, avsystemdesign wrote:
Thanks Steve,
Then how would a person create the following condition?
Display A + B can be powered up separately, but depending on what
source has been selected they both may or may not need to be powered up at
the same time. And here is the kicker, they both have different warm up
times, and the "system is warming up" page needs to be held high during the
power up cycle.
So if A takes 10s to warm up, but has been powered up at the same
time as B (which takes 20s), the warm up page needs to be high for 20s.
--- In Crestron@..., Steve Kaudle<crestron@> wrote:
From a more abstract/best practices standpoint, you should try to
keep
the amount of timed events (steppers, delays, one-shots, etc...) you
use
to a minimum. That said, I'd bet you'd need to use a>lot< of steppers
before the burden they put on system would be perceived by the
operator. Lots of variables at play and the real world threshold will
likely vary (greatly) from system to system. Note that the previous
statement(s) should most defiantly not be translated into anything
like
'stop using timed events if you want your systems to work right'.
From a logical/program flow standpoint, I prefer to avoid timed
events
and build logic that is condition dependent. Example: I don't
necessarily want to send an input swap command to a projector based
on a
specific amount of time having passed, but rather based on real-time
conditions like the (actual) power state and current input selection.
On 4/12/2012 5:56 PM, avsystemdesign wrote:
I know there are several ways to skin a cat in Crestron world, and
I am always looking at improving and following best practices. Having said
that I have a question.
Is it bad to have 2 or 3 steppers running concurrently? Depending
on the device being used, I like to have a stepper handle the procedure.
For example, source is selected, stepper for projector 1 is
triggered, this fires up the proj and sends down the screen, whilst
indicating the system is warming up.
If a different source is selected, it might be destined for other
displays, which in turn will trigger their respective steppers.
So there might be a few steppers running at the same time.
------------------------------------
Check out the Files area for useful modules, documents, and drivers.
A contact list of Crestron dealers and programmers can be found in
the Database area.
Yahoo!
Groups Links
------------------------------------
Check out the Files area for useful modules, documents, and drivers.
A contact list of Crestron dealers and programmers can be found in
the Database area.
Yahoo!
Groups Links
------------------------------------
Check out the Files area for useful modules, documents, and drivers.
A contact list of Crestron dealers and programmers can be found in the
Database area.
Yahoo!
Groups Links
