[PD] Connection with sensors - new devices?

[Hans-Christoph Steiner]

On Dec 22, 2005, at 2:46 PM, Christian Klippel wrote:

> hi,
>
> Am Donnerstag 22 Dezember 2005 20:11 schrieb Sylvain:
>> Hi list,
>>
>> Just wondering, I read the thread about the topic, but I am quite
>> confused about the fact that nobody mentioned chips with TCP/IP  
>> layer. I
>> would find useful, as Pd already has pre-built objects that can listen
>> to TCP ports, to send Pd raw information about the sensors used, and
>> then do anything you want with it. It is because it is not as cheap as
>> other solutions ? Cause I find this solution pretty attractive. I am
>> actually trying to set up a little project with students using this  
>> type
>> of component to make them discover how to control sound with sensors.  
>> I
>> know that it may sound to some people like the hammer to smash the fly
>> (as we say in french), but as my lab can afford it ...
>> Any comment welcome.
>>
>> Sylvain
>>
>
> the multio will get a 10baseT extension beginning of next year, so its  
> on
> ethernet then.
>
> but im not sure about the latency of that. it may be fine when there is
> nothing else is happening on that particular netsegment, we will see.
>
> usb is pretty fast, available on almost any device nowdays, and allows  
> for
> longer distances as well.

According to spec, 10BaseT has a maximum cable length of 100m.  USB 1.0  
spec says 5m.  I've used 100BaseT on cables well over 100m, so that  
spec is quite conservative.  I've heard about people having quite a few  
troubles trying to get USB up to 20m.

Plus you can get a 300m spool of CAT5 for US$40.  Ethernet is really  
far better than USB if you want to go for long cable lengths.

.hc

________________________________________________________________________ 
____

"Information wants to be free."
                                      -Stewart Brand