Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 06 Dec 2000 09:10:18 -0000 (GMT)
From:      Duncan Barclay <dmlb@dmlb.org>
To:        Warner Losh <imp@village.org>
Cc:        mobile@FreeBSD.ORG, Wesley Morgan <morganw@chemicals.tacorp.com>, Greg Lehey <grog@lemis.com>, Christopher Yeoh <cyeoh@linuxcare.com.au>
Subject:   Re: ray committed
Message-ID:  <XFMail.001206091018.dmlb@computer.my.domain>
In-Reply-To: <200012060604.XAA76044@harmony.village.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi

On 06-Dec-00 Warner Losh wrote:
> In message <14893.53324.616401.949125@rockhopper.linuxcare.com.au>
> Christopher Yeoh writes:
>: As far as range goes, for our situation the aviators worked better
>: than even wavelans with small antennas (yes this surprised us!).  The
>: situation is fairly unusual though. We're on different floors and the
>: floor is made of reinforced concrete). Definitely no signal through
>: the floor (eg being directly above doesn't work). However, the way the
>: appartments are arranged there appears to be a path through a couple
>: of appartments, outside and then into his side window, that the
>: aviators can get a decent signal along.
> 
> OK.  I have two floors of my house.  It is mostly wood based with the
> usual duct work and such.  I was down stairs and the "base" was
> upstairs.  I'd notice that sometimes things would just work, but that
> heavy load doesn't work too well (eg a simple FTP or scp would swamp
> the link).  With the ZoomAir + Orinoco card (both with their default
> tiny anntennas) I can still get good interactive performance and 1.5mbps
> of ftp transfers.

1.5Mb/s data or about 160kB/s is about the theoretical limit for 802.11 at
2Mb/s and TCP/IP. I get 160kB/s ftp with my Libretto 50 over the raylink
driver.

> I see only slightly worse interactive performance
> during the ftp than when it isn't running.  With the aviator, the ftp
> would just about shut down the link.  Then after a while things would
> get worse and worse.  I could then take the laptop with the aviator
> upstairs to 1' of the base station (which is just an intel box with an
> aviator card) and get 0 throughput (which sounds like a driver
> problem).  The really strange part is that I can reboot both machines
> and still get horrible to zero throughput.  This points to "enemy
> action" of some sort.
> 
> I also have a wireless link with the older pre802.11 wavelan cards at
> 2.4GHz to a friends house.  Maybe that is impacting things.

This could be a problem. Do you know if these are frequency hopping or
direct sequence? If frequency hopping they are likely to be interfering.

>  It
> doesn't impact the 802.11b cards that I use.  We've also noticed
> mysterious outages with the radio link for reasons that we still don't
> understand.  We've theorized that it may be due to other users in the
> 2.4GHz band, which is also causing problems for the aviators.

How long is the link? 802.11 was designed for a couple of 100m - this is not
just Tx power/Rx sensitivity but stuff like
        resiliance to fading channels
        resiliance to mutli-path
        timing parameters (remember 150m/microsecond)
        ...

> Or that
> could just be a red herring.
> 
> I have also tried turning off the wireless link above and things
> didn't get better.

It might be direct sequence then if it's WaveLAN. This wouldn't affect the
Aviators.

I think you might have a half dead card.
 
> Warner

Duncan

---
________________________________________________________________________
Duncan Barclay  | God smiles upon the little children,
dmlb@dmlb.org   | the alcoholics, and the permanently stoned.
________________________________________________________________________


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-mobile" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.001206091018.dmlb>