Re: Inter-operability of Draft-n equipment

henry999@eircom.net (Henry) hath wroth:

>Finalisation of 802.11n is still reportedly over a year away.

Actually, the slipping schedule seems to be a continuously moving
target that always seems to remain a year or two away. The current
guess is Dec 2009 for the grand finale, and about July 2009 (my guess)
for something that can be updated to whatever is approved.
http://grouper.ieee.org/groups/802/1..._Timelines.htm

>There is
>of course no guarantee (actually, little likelihood) that any of the
>draft-n gear available now will be upgradable to the standard, when it
>comes.

Well, Dec 2009 is a bit less than 2 years away. Anything you buy
today will probably be obsolete by then. It's not just the MIMO part
that will change in the next two years. Router and wireless features
will also improve. For example, dual SSID's for "guest" WLAN's. Built
in intrusion detection.

>The conventional wisdom for using draft-n equipment at the moment is to
>get all the pieces from the same manufacturer, to make sure they work
>together as intended. But that's not always possible.

Nope. All the same chipsets from the same manufacturer. There's no
guarantee that even a single manufacturers stuff will talk to other
products made with another vendors chips.

Incidentally, the spec includes provisions for two radically different
forms of MIMO. Spacial multiplexing and beam forming are totally
incompatible.

>Have any of the computer or networking magazines, or any of the
>consumer-protection testing labs, run any studies comparing how well (or
>how poorly) 'mixed' environment networks run?

Sorta. This one is several years old:
<http://www.smallnetbuilder.com/content/view/24214/96/>
and doesn't really do any interoperability testing. Some of the
reviews on this site test performance with compatible MIMO and
ordinary wireless clients. I haven't seen any Interop style bake off
testing yet.

>Things like 'The A pcmcia
>card works with B's router but not C's access point, while D's pci
>adapter works with routers from A and C but not B', etc., etc.
>
>Given that there are only, what?, four or five or six major
>manufacturers of consumer- and SOHO-level equipment, it shouldn't be too
>hard to put the more popular models through their paces.

Yep. However, new MIMO models come and go at an alarming rate.
Product lifetimes seem to be about 6 months. By the time the tests
are organized, run, reported, and published, all the products tested
will be obsolete. However, that's not why nobody is doing such tests.
It's because many of the claims presented by MIMO proponents are
unsupportable and all too easily proven to be baloney. Nobody wants
to sponsor a comparative test that demonstrates that MIMO is about
speed, not range.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

The Wi-Fi Alliance, as usual, keep their interoperability test
procedures for Draft N under wrap.
As an aside I did find this web site which does have a pdf. of how they
did it for 11a,11b and 11g.

<http://www.iol.unh.edu/services/testing/wireless/testsuites/temp.html#802.11n%20Draft%202.0%20MAC%20STA%20Conf ormance%20Test%20Suite>

LR <lrme@privacy.net> hath wroth:
>The Wi-Fi Alliance, as usual, keep their interoperability test
>procedures for Draft N under wrap.

Well yes. The test procedures are their primary asset. If anyone can
obtain the procedures and perform their tests, then someone might
discount their rather high cost of certification (About $25K). Of
course, that won't stop someone from trying:
<http://www.azimuthsystems.com/solutions-wi-fi-certified.htm>
<http://www.azimuthsystems.com/97710dd2-94ae-44f8-9b57-4f97befcfaef/news-press-detail.htm>
There are other vendors, but as usual, I can't find them in my
bookmark mess.

Meanwhile, I'm still dealing with WEP ASCII/BINARY conversion
incompatibilities and regularly find wireless clients that can't seem
to get a DHCP assigned IP from certified wireless routers. There are
a few other big holes in Wi-Fi certification.

So, when all the products are certified and largely indistinguishable,
vendors are competing on the basis of buzzwords and unsubstantiated
claims. That attract class action suits:
<http://www.belkin.com/class_notice/classnotice.asp>
<http://www.techworld.com/mobility/news/index.cfm?NewsID=4873&inkc=0>
<http://www.buffalotech.com/files/Official%20Notice.pdf>
Everybody lies, but that's ok because nobody listens.

>As an aside I did find this web site which does have a pdf. of how they
>did it for 11a,11b and 11g.
><http://www.iol.unh.edu/services/testing/wireless/testsuites/temp.html#802.11n%20Draft%202.0%20MAC%20STA%20Conf ormance%20Test%20Suite>

Never mind interoperability. Just getting MIMO to work in the first
place is somewhat of a challenge:
<http://rfdesign.com/next_generation_wireless/broadband_technologies/radio_challenges_testing_mimo/>

Most of the vendors are assuming that MIMO will go the way of DOCSIS
certification, where there is no interoperability testing between the
DSLAM and the DOCSIS modem. Instead, there are certification tests,
performed with a standardized (CableLabs) test suite, on each device.
In theory, if they communicate reliably with the test equipment, they
should talk reliability to each other. In general, that has worked
well.
--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

"Jeff Liebermann" <jeffl@cruzio.com> wrote in message
news:n58sq39aphr4cecbui7v6ts207q6fkp3c5@4ax.com...
| LR <lrme@privacy.net> hath wroth:
| >The Wi-Fi Alliance, as usual, keep their interoperability test
| >procedures for Draft N under wrap.
|
| Well yes. The test procedures are their primary asset. If anyone can
| obtain the procedures and perform their tests, then someone might
| discount their rather high cost of certification (About $25K). Of
| course, that won't stop someone from trying:
| <http://www.azimuthsystems.com/solutions-wi-fi-certified.htm>
|
<http://www.azimuthsystems.com/97710dd2-94ae-44f8-9b57-4f97befcfaef/news-press-detail.htm>
| There are other vendors, but as usual, I can't find them in my
| bookmark mess.
|
| Meanwhile, I'm still dealing with WEP ASCII/BINARY conversion
| incompatibilities and regularly find wireless clients that can't seem
| to get a DHCP assigned IP from certified wireless routers. There are
| a few other big holes in Wi-Fi certification.
|
| So, when all the products are certified and largely indistinguishable,
| vendors are competing on the basis of buzzwords and unsubstantiated
| claims. That attract class action suits:
| <http://www.belkin.com/class_notice/classnotice.asp>
| <http://www.techworld.com/mobility/news/index.cfm?NewsID=4873&inkc=0>
| <http://www.buffalotech.com/files/Official%20Notice.pdf>
| Everybody lies, but that's ok because nobody listens.
|
| >As an aside I did find this web site which does have a pdf. of how they
| >did it for 11a,11b and 11g.
|
><http://www.iol.unh.edu/services/testing/wireless/testsuites/temp.html#802.11n%20Draft%202.0%20MAC%20STA%20Conf ormance%20Test%20Suite>
|
| Never mind interoperability. Just getting MIMO to work in the first
| place is somewhat of a challenge:
|
<http://rfdesign.com/next_generation_wireless/broadband_technologies/radio_challenges_testing_mimo/>
|
| Most of the vendors are assuming that MIMO will go the way of DOCSIS
| certification, where there is no interoperability testing between the
| DSLAM and the DOCSIS modem. Instead, there are certification tests,
| performed with a standardized (CableLabs) test suite, on each device.
| In theory, if they communicate reliably with the test equipment, they
| should talk reliability to each other. In general, that has worked
| well.


I have a native distrust of industry sponcered certification and
interoperability testing. My most recent experience was with the CTIA
certification of cell phone handsets. The process was quickly watered down
and the testing fees become a profit center for the parent organization.