If you want real RX sensitivity, buy something newer than either, like a good quality Atheros card and run it in 11b mode. Kicks as* on anything Prism2.x-based, and the whole Prism-thang is about to be yesterday's news now that Globespan-Virata purchased the not-yet-rotting corpse of Prism (it was far too expensive for a commodity market.)
Heck, nearly any new chipset (AMD, Broadcom, etc) with a decent LNA in front of it can come close to the claimed rx sensitivity of zcomax or Senao.
And yes, PAs run hot when they're throwing down a lot of power, and hot PAs distort the signal and worse.
The hunch about different receivers is misguided, the baseband for both the Cisco and Prism-2.5 are essentially the same.
The concern about vendors over-specing the datasheet is well-founded, however. I saw a vendor the other day claiming 268F from his "outdoor bridge" which was little more than a Ubicom + Senao 200mW combo in a cheap plastic box. As always, YMMV. :-)
Jim
ob disclaimer: My wife sells both Intersil and Atheros-based designs manufactured by Senao, and I've gone a few rounds with Tony from Demarc more than once comparing the two designs.
On Sunday, September 28, 2003, at 3:54 PM, [EMAIL PROTECTED] wrote:
<snip>roddog wrote:
O.K. So you want a bad as* card for wardriving?
What you need is a Senao/Engenious? card with 2 antenna ports on it.
These cards have a better sensitivity than just about anything out there
next to Demark
and they also have 200mW of power which is the most powerful out there.
Brian Lloyd writes next:
A couple of comments about the Senao:<Brian's notes on antenna connection info deleted below here>
As for sensitivity, the Senao card has better *published* receive sensitivity specs. It seems pretty good but I am not sure it beats the Cisco for RX sensitivity.
TX power does indeed seem to be about +23 dbm.
I agree with Brian - I've now had a chance to get a couple senao's on my
bench, and while my 2 cards are _not_ a valid statistical sample size (which
I clearly emphasized to the customer who handed them to me), I have noted
that both cards lacked the receive performance of the aironet/cisco 350
series cards. They did both deliver the +23 dBm into 50 ohms (but most all
of my cisco 350's exceed their +20 dBm spec by 1-2 dB). given all this - the
1-2 dB of extra xmit power compared to a typical cisco 350 weighed against the
loss in receive (not enough samples to say definitively, but it's alot more
than the benefit of the transmitter) doesn't justify their use in MARGINAL
links as a better alternative. They still seem to be a great all-around card
for NORMAL links, and CAN be the best bang for the buck in many situations
that don't call for the absolute best money can buy. I hope they force cisco
to make a 200 mW "360" card! It's the best part about market competition!
My hunch is that it was simpler and cheaper for senao to make a +23dBm card
with a so-so receiver than a +20dBm card with a great receiver - besides,
marketing will LOVE being able to say "twice as much POWER" - when was the
last time you heard a marketing hype "6 dB better code to noise ratio!"??
As a BIGGER note - (again - keep in mind 2 cards does not a statistical
sample make) - placing them inside a calibrated RF anechoic + temp/humidity
chamber resulted in wider variance near the high end of their specified
operating temperature (60c). The performance in the unspecified area
above 60c was substantially poorer than a comparable cisco card. But why
is 60C so important?
Once again - if your card is kept at 22 deg C inside your house, you'll most
likely never notice - but if it's outside on a 100 deg F day - inside a box
on a pole in the sun and your link margin drops below zero - it's a big deal.
NOWDAYS, EVEN THE PCMCIA SLOTS IN LAPTOPS CAN GET TO 60C! I just stuck a
micro-thin type T T/C in between my 2 pcmcia cards on the laptop I'm using
to write this - 23.2 C room temp, 56.8C in between the cards! (and this is
a PII/350 not some 3.0G PIV rocket laptop that has a 90 watt adapter!) Ever
pull out the card and notice it's hot to the touch? Senao's spec sheet stops
at 60C for their high power card. I'm just ~3C away from that temp on a 3 yr.
old laptop! When my 2.0 Ghz laptop comes back from repair, I'll duplicate
the "pcmcia slot temp" test on it. That thing is too hot for a "lap" sometimes.
I can't imagine how hot the pcmcia compartment is!
Cisco, on the other hand, rates their AIR-PCM35x series to +70C - AND THEY
MAINTAIN THER PERFORMANCE SPEC. Through the whole temperature range and
a little bit beyond.
I'm not getting these numbers out of thin air - read the spec sheets in person
http://www.cisco.com/en/US/products/hw/wireless/ps4555/ products_data_sheet09186a0080088828.html
http://www.senao.com/service%20&%20support/Spec/NL-2511CD_PLUS_Spec.pdf
(funny - senao's older cards were only rated at 50C - then 55C, then 60C )
If it seems like this bugs me - it does to a degree - It's not fair for
manufacturers to make impressive glossy ad slicks that have numbers 99.99%
of the population can never verify. Too many people make buying decisions
on bad data - then wonder why things don't work. I've heard all too often
"but the box said it would go 5 miles!" I guess it keeps customers comming
though...
Everett
-- general wireless list, a bawug thing <http://www.bawug.org/> [un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless
-- general wireless list, a bawug thing <http://www.bawug.org/> [un]subscribe: http://lists.bawug.org/mailman/listinfo/wireless
