> side.  That is a LOT of data traversing the wire between the radio
> hardware and your computer arriving in the correct sequence in near
> real-time. When the FLEX-5000 was in the engineering/development
stage,
> a wide variety of data bus technologies where evaluated and only
> Firewire was able to meet all of the stringent requirements.
> 

I think we've debated this one or twice before.

Let's do the math!

192K samples per second * 24 bits per sample = 4.7Mbps

12 channels * 4.7Mbps = 56.4Mbps

(I admit that I don't understand how we get 12 channels... but that's
what Tim said, so let's go with it)

Is my math wrong?  If not, the max output of an SDR-5000 be WELL within
the range of 100Mb Ethernet.

For reference: I regularly observe sustained transfers in excess of
70Mbps over my 100Mbps Ethernet between systems, using a commodity DLink
switch from Best Buy.

And, for fun, note that on USB V2 320Mbps of actual bulk data throughput
is routinely observed.  This is actual data rate, and does not include
USB SOF and other overheads.

> Ethernet was NEVER designed to be a multi-channel anything since it is
> CDMA so those inherent limitations have to be reliably over come
before
> it can be a viable replacement technology.

I'm not sure what this means.  1394 doesn't send multiple channels of
data simultaneously (it interleaves transfers, just like Ethernet).  But
this is a specious argument in any case, because we're sending discrete
digital data, and sending such data at a sufficiently high rate is
equivalent to being continuous.  And interleaving multiple packets of
this discrete data, given sufficient capacity, is exactly equivalent to
"multi-channel" capability.

> When the FLEX-5000 was in the engineering/development stage,
> a wide variety of data bus technologies where evaluated and only
> Firewire was able to meet all of the stringent requirements.

I don't know what the additional factors might have been, but it is
simply not correct to state that 1394 is the only medium that can handle
this data rate.

Peter
K1PGV


_______________________________________________
FlexRadio Systems Mailing List
FlexRadio@flex-radio.biz
http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz
Archives: http://www.mail-archive.com/flexradio%40flex-radio.biz/
Knowledge Base: http://kc.flex-radio.com/  Homepage: http://www.flex-radio.com/

Reply via email to