[drats_users] WIKI & FAQ

[Dan Smith]

> Unless it's changed in the 0.3 version, the actual title in the
> first 'archive' paragraph should be "Transfers" since that's what
> needs to get clicked on in the 0.2 versions.

It hasn't changed in 0.3, but the first paragraph was actually quoting
someone else's comment.  It should be cleaned up for the wiki.

> I understand it completely.  Is it really possible to make 32 blocks
> of 4096 bytes?  I'd really be pushing my luck with those settings!

Yep, there's no reason why you couldn't.  For a socket
(i.e. ratflector) connection that would provide the best performance.
On RF, you'd likely melt your radio (or hit the timeout timer) first :)

> What about Tuning?  I don't know about Warmup, and I don't
> understand why "Force transmission delay" can have a negative
> number, or what the number does at all (seconds? ms? us?).

>From the same archive:

| There are two warm-up related values.  The goal of the "warm-up" is
| to prime the data stream with something before the real data is
| transmitted.  I initially added this to defeat the power save
| feature on the handhelds.  After a while, they go to sleep and only
| wake up every few hundred milliseconds to see if something is being
| received.  By prefixing the data you're transmitting with enough
| pad, you can wake them up before the real data comes across.
| 
| The "Warm-up timeout" setting specifies how long since the last
| transmission before deciding to transmit a warm-up block.  If you
| last transmitted a half-second ago, then the handheld listening
| hasn't gone to sleep yet, so there's not much point.  If you set
| this to, say, 5, then a warm-up block will only be transmitted if
| you haven't transmitted in the last 5 seconds.  If you set it to 0,
| it will always be transmitted before any real data (which will slow
| things down, of course).
| 
| The "Warm-up length" specifies how much data goes in the pad itself,
| and thus depends on how long you must be transmitting before the
| remote side will wake up.
| 
| Unrelated to warm-up, there is also a "force transmission delay"
| value.  This lets you tell D-RATS to *always* wait a certain number
| of seconds before transmitting anything.  No intelligence, just a
| delay.  This will, of course, slow things down a *lot*, but can be
| helpful to make sure D-RATS doesn't start transmitting too quickly
| after receiving a transmission, which can help avoid collisions
| gateway connections, I imagine.

For reference, here's the whole thing:

http://lists.danplanet.com/pipermail/drats_users/2009-February/000790.html

You might want to read the related bits of the thread to get all the
context and subsequent comments.

> In the FAQ section "What baud rate should I choose?", the ID-1 has a
> baud rate of 19200 (and the port is determined by the USB software).
> I know not many of you might use an ID-1 for low speed data, but it
> was one of the better choices for our use.

You're the first person I've heard confirm use of D-RATS on an ID-1.
Good to know it behaves the same as the other radios (it should, but
you never know).

The wiki is, well, a wiki.  Feel free to create an account and help
polish up the documentation and FAQ with your experiences.  That's the
whole idea :)

I'll try to get around to adding the tuning parameters to a new page.
If I haven't seen you (or someone else) add the baud rate thing to the
FAQ page by then, I'll do it.

> I noticed on the Ratflector web page, sending a Ping to CQCQCQ will
> find out who is online.  Does this work at all on RF?

Sure!  I use that a lot to see who I can hit when working simplex
from a new location.  D-RATS has a little logic in there to wait a
random number of seconds (between 0 and 10, IIRC) before replying.
This attempts to avoid everyone answering at the same time.  However,
sometimes you need to do a couple pings to get everyone if it's
crowded as some stations will inevitably step on each other.

Thanks!

-- 
Dan Smith
dsmith#danplanet.com, s/#/@/
www.danplanet.com
KK7DS