Nikon D600 a compromise but ok

On Thu, 20 Sep 2012 17:50:08 -0800, (Floyd L. Davidson)
wrote:
: Alan Browne wrote:
: On 2012.09.20 02:06 , nospam wrote:
:
: they can skip it in the low end models, since the target market
: probably doesn't care about those features.
:
: I disagree. There are likely a lot of people at the "lower end" who
: would enjoy the GPS benefit of tagging and the WiFi benefit of quick
: uploading. That would be worth a small premium in those models.
:
: Indeed the first cameras to sport GPS widely were P&S.
:
: I tend to agree with that concept. More over, I don't
: think professionals will find WIFI as useful as others
: expect.
:
: It's something that works just fine if there are only
: one or two being used. But just imagine a PJ showing up
: at a press conference expecting rapid transfer of
: images... and discovering that there are 47 other PJ's
: with wireless enabled. Oppps.
:
: Only 3 of the channels in the 2.4Ghz band can be used at
: the same time, and all it takes is one fool choosing a
: channel 3, as an example, to wipe out the entire lower
: half and limit it to only two channels at a time. With
: 30 or 40 users standing in line for packet time, data
: transfers will be just exceedingly slow! A dialup modem
: from a laptop would be way faster!

It's been several years since I've been heavily involved in setting up Wi-Fi
networks, but my recollection is that the client rarely gets to choose. If the
access point is broadcasting on Channel n, that's what gets used.

BTW, while it's true that there are only three channels that don't bleed over
onto each other, it's an exaggeration to say that only three channels can be
used at the same time.

Bob

Robert Coe wrote:
On Thu, 20 Sep 2012 17:50:08 -0800, (Floyd L. Davidson)
wrote:
: Alan Browne wrote:
: On 2012.09.20 02:06 , nospam wrote:
:
: they can skip it in the low end models, since the target market
: probably doesn't care about those features.
:
: I disagree. There are likely a lot of people at the "lower end" who
: would enjoy the GPS benefit of tagging and the WiFi benefit of quick
: uploading. That would be worth a small premium in those models.
:
: Indeed the first cameras to sport GPS widely were P&S.
:
: I tend to agree with that concept. More over, I don't
: think professionals will find WIFI as useful as others
: expect.
:
: It's something that works just fine if there are only
: one or two being used. But just imagine a PJ showing up
: at a press conference expecting rapid transfer of
: images... and discovering that there are 47 other PJ's
: with wireless enabled. Oppps.
:
: Only 3 of the channels in the 2.4Ghz band can be used at
: the same time, and all it takes is one fool choosing a
: channel 3, as an example, to wipe out the entire lower
: half and limit it to only two channels at a time. With
: 30 or 40 users standing in line for packet time, data
: transfers will be just exceedingly slow! A dialup modem
: from a laptop would be way faster!

It's been several years since I've been heavily involved in setting up Wi-Fi
networks, but my recollection is that the client rarely gets to choose. If the
access point is broadcasting on Channel n, that's what gets used.

The client of course uses the channel configured at
chosen AP.

But the client does choose which AP, and in most cases
where a camera is used with wireless the AP will in fact
be specifically set up by the photographer. In a
situation where there are multiple pros who might find
wireless useful, it wouldn't make sense for all of them
to connect to a single AP.

BTW, while it's true that there are only three channels that don't bleed over
onto each other, it's an exaggeration to say that only three channels can be
used at the same time.

Only three channels with high signal strength can be
used at one time. A weaker signal will not interfere at
all, but it also will not be usable itself.

The point is that if multiple cameras all equipped with
wireless are set up in a line, the data transfer rates
can at most equal the total rate that three of them will
have. If they are all on one channel, the total rate
will be the same as if only one camera is there. (And
will necessarily be true if they all connect to a single
AP.)

If the connections are spread over multiple AP's, each
on different channels, the total transfer rate will
never exceed the transfer rate that only 3 cameras can
achieve. That can be accomplished by choosing channels
1, 6 and 11 (or 12). If a forth AP is set for channel
3, then it will necessarily share the bandwidth with
units on channels 1 and 6 equally, and only 1 of those
plus the unit on channel 11 can operate simultaneously.

An example would be a press conference... where there
might be far more than 3 or 4 photographers, but the
data rate from using wireless will be limited to 3 at
any one time.

--
Floyd L. Davidson http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)

On 2012.09.22 07:16 , Robert Coe wrote:

It's been several years since I've been heavily involved in setting up Wi-Fi
networks, but my recollection is that the client rarely gets to choose. If the
access point is broadcasting on Channel n, that's what gets used.

The access point chooses. That can be automatic or set to a specific
channel.

BTW, while it's true that there are only three channels that don't bleed over
onto each other, it's an exaggeration to say that only three channels can be
used at the same time.

I notice that in the homes around here there are about 8 WiFi
stations and they all use one of 3 channels (1,6, 11). So I set my
station to the middle of 3 unused channels (9).

--
"There were, unfortunately, no great principles on which parties
were divided – politics became a mere struggle for office."
-Sir John A. Macdonald

Alan Browne wrote:
On 2012.09.22 07:16 , Robert Coe wrote:

It's been several years since I've been heavily involved in setting up Wi-Fi
networks, but my recollection is that the client rarely gets to choose. If the
access point is broadcasting on Channel n, that's what gets used.

The access point chooses. That can be automatic or set to a specific
channel.

The AP is not "automatic". The client can be automatic,
by using a name rather than a channel.

BTW, while it's true that there are only three channels that don't bleed over
onto each other, it's an exaggeration to say that only three channels can be
used at the same time.

I notice that in the homes around here there are about 8 WiFi
stations and they all use one of 3 channels (1,6, 11). So I set my
station to the middle of 3 unused channels (9).

The others apparently know what they are doing.

The channel separation in the 2.4GHz band is 5 Mhz, but
the actual bandwidth used by different authorized
classes of service is often much greater. The WIFI
802.11g standard uses a 22 Mhz channel with 5 Mhz
separation. Thus only 3 of the WIFI 802.11G 2.4GHz
channels available in the US can be used at once on a
non-interfering basis.

A unit assigned to Channel 6 will interfere with
any unit assigned to Channels from 2 to 9. The upper
limit of bandwidth needed for a unit centered on Channel
1 is using the bandwidth of Channel 3. The lower
limit of bandwith needed for a unit centered on Channel 6
is using the bandwidth of Channel 4.

Here is a chart showing which channels are used by a WIFI
unit assigned to a specific channel:

Assigned / Used Assigned / Used
1 (-1) to 3 7 5 to 9
2 (0) to 4 8 6 to 10
3 1 to 5 9 7 to 11
4 2 to 6 10 8 to 12
5 3 to 7 11 9 to 13
6 4 to 8 12 10 to 14

Here's an ASCI graphic (that will only look right if you
use a monospaced font):

-15 -10 -5 0 +5 +10 +15 MHz
3 4 5 6 7 8 9 Channel
| | | | | | |

\__22 Mhz Used Bandwidth__/ 802.11G Bandwidth
for Channel 6

\__22 Mhz Used Bandwidth__/ BW Channel 7

\__22 Mhz Used Bandwidth__/ BW, Ch 8

\__22 Mhz Used Bandwidth__/ BW, Ch 9

As you can see, the upper 7 Mhz of the bandwidth
required for an AP assigned a center frequency of
Channel 6 is the same as the lower 7 Mhz bandwidth used
by an AP assigned to a center frequency of Channel 9.
To get the 802.11G 5 Mhz spacing between channels the closest
assignment above Channel 6 is at Channel 11.

Extend the diagram and you'll find that Channel 9 is
also sharing significant bandwidth with channels 10, 11
and barely with channel 12. Hence the use of channel 9
interferes with both channel 6 and channel 11. And they
interfere with you on channel 9.

When you are not using your unit, a single neighbor on
channel 6 can get full time use of the bandwidth. At
the same time another neighbor on channel 11 can get
full time use of the bandwidth. (And if there are two
units using each channel, the two share each channel
equally.)

But when you crank up and put traffic on channel 9 you
share bandwidth with channel 6, and each of you gets
half as much time. The same is true of channel 11. So
if *either* of those channels is actually in use you will
not get full time use of the bandwidth. If both are in
use you will at most get about 1/3 of the shared time,
instead of 1/2 that you'd get if you were using either
channel 6 or 11.

--
Floyd L. Davidson http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)

Floyd L. Davidson wrote:

[WiFi]

It's something that works just fine if there are only
one or two being used. But just imagine a PJ showing up
at a press conference expecting rapid transfer of
images... and discovering that there are 47 other PJ's
with wireless enabled. Oppps.

Only 3 of the channels in the 2.4Ghz band can be used at
the same time, and all it takes is one fool choosing a
channel 3, as an example, to wipe out the entire lower
half and limit it to only two channels at a time. With
30 or 40 users standing in line for packet time, data
transfers will be just exceedingly slow! A dialup modem
from a laptop would be way faster!

A dialup modem manages an upload of at most 33.6 kBit/s (for a
56k modem).
A WiFi (SOTA 2003) manages between 6 and 54 MBit/s on a
channel.
6 MBit/s = 6144 kBit/s
6144 kBit/s / 48 PJs = 128 kBit/s
(assuming there is only one single channel available and
it's at it's slowest speed because the modem is far away
behind thick walls. I.e. worst case)
128 kBit/s = 3.8 * 33.6 kBit/s

= A dialup modem at it's best would be about 4 times slower than
WiFi at it's very worst --- even worse than the scenario
Floyd painted.

Not that the truth or reality would have any chance against
a pointless Floyd rant filled with speculation and missing
basic knowledge.

Since the press conference will have enough base stations at close
distances to the PJs, we can assume 2-3 channels at 54 MBit/s,
giving 48 PJs (wow, must be a huge press conference!) 3,3 MBit/s.
(which is quite more upload per PJ than most households have,
even on fast connections). A 22 MPix JPEG will take ~15 seconds,
but you'll probably reduce the size anyway, and get 3 or 4
seconds/JPEG.

-Wolfgang

Alan Browne wrote:
On 2012.09.22 07:16 , Robert Coe wrote:

It's been several years since I've been heavily involved in setting up Wi-Fi
networks, but my recollection is that the client rarely gets to choose. If the
access point is broadcasting on Channel n, that's what gets used.

The access point chooses. That can be automatic or set to a specific
channel.

BTW, while it's true that there are only three channels that don't bleed over
onto each other, it's an exaggeration to say that only three channels can be
used at the same time.

I notice that in the homes around here there are about 8 WiFi
stations and they all use one of 3 channels (1,6, 11). So I set my
station to the middle of 3 unused channels (9).

Clever, then your signal interferes (and is interfered by)
channels 6 and 11. There's a reason they only use 1, 6 and 11
.... as only these are fully decoupled between each other.

-Wolfgang

Wolfgang Weisselberg wrote:
Floyd L. Davidson wrote:
It's something that works just fine if there are only
one or two being used. But just imagine a PJ showing up
at a press conference expecting rapid transfer of
images... and discovering that there are 47 other PJ's
with wireless enabled. Oppps.

Only 3 of the channels in the 2.4Ghz band can be used at
the same time, and all it takes is one fool choosing a
channel 3, as an example, to wipe out the entire lower
half and limit it to only two channels at a time. With
30 or 40 users standing in line for packet time, data
transfers will be just exceedingly slow! A dialup modem
from a laptop would be way faster!

A dialup modem manages an upload of at most 33.6 kBit/s (for a
56k modem).
A WiFi (SOTA 2003) manages between 6 and 54 MBit/s on a
channel.
6 MBit/s = 6144 kBit/s
6144 kBit/s / 48 PJs = 128 kBit/s
(assuming there is only one single channel available and
it's at it's slowest speed because the modem is far away
behind thick walls. I.e. worst case)
128 kBit/s = 3.8 * 33.6 kBit/s

= A dialup modem at it's best would be about 4 times slower than
WiFi at it's very worst --- even worse than the scenario
Floyd painted.

Not true.

Not that the truth or reality would have any chance against
a pointless Floyd rant filled with speculation and missing
basic knowledge.

Apparently I hit it just about on the head with the 48
PJs. If the bit rate works out to 4x, then it's
probably just about the right number. There is *never*
going to be a smooth transition between any two of the
48 clients. Instead there will virtually always be
contention, and instead of being 4 times faster than a
dialup, it would probably be about 10 times slower... at
best!

Since the press conference will have enough base stations at close
distances to the PJs, we can assume 2-3 channels at 54 MBit/s,
giving 48 PJs (wow, must be a huge press conference!) 3,3 MBit/s.
(which is quite more upload per PJ than most households have,
even on fast connections). A 22 MPix JPEG will take ~15 seconds,
but you'll probably reduce the size anyway, and get 3 or 4
seconds/JPEG.

Never tried any of this stuff, have you! :-)

--
Floyd L. Davidson http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)

Floyd L. Davidson wrote:
Wolfgang Weisselberg wrote:
Floyd L. Davidson wrote:

It's something that works just fine if there are only
one or two being used. But just imagine a PJ showing up
at a press conference expecting rapid transfer of
images... and discovering that there are 47 other PJ's
with wireless enabled. Oppps.

Only 3 of the channels in the 2.4Ghz band can be used at
the same time, and all it takes is one fool choosing a
channel 3, as an example, to wipe out the entire lower
half and limit it to only two channels at a time. With
30 or 40 users standing in line for packet time, data
transfers will be just exceedingly slow! A dialup modem
from a laptop would be way faster!

A dialup modem manages an upload of at most 33.6 kBit/s (for a
56k modem).
A WiFi (SOTA 2003) manages between 6 and 54 MBit/s on a
channel.
6 MBit/s = 6144 kBit/s
6144 kBit/s / 48 PJs = 128 kBit/s
(assuming there is only one single channel available and
it's at it's slowest speed because the modem is far away
behind thick walls. I.e. worst case)
128 kBit/s = 3.8 * 33.6 kBit/s

= A dialup modem at it's best would be about 4 times slower than
WiFi at it's very worst --- even worse than the scenario
Floyd painted.

Not true.

In which case you would be able to show us the math and the
circumstances, under which that would not be true.

But you're just handwaving ... very wildly.


Not that the truth or reality would have any chance against
a pointless Floyd rant filled with speculation and missing
basic knowledge.

Apparently I hit it just about on the head with the 48
PJs.

So ... which press conference was that?
Or did you pull that number from your ass?

If the bit rate works out to 4x, then it's
probably just about the right number. There is *never*
going to be a smooth transition between any two of the
48 clients.

Obvious. None of the cameras are interested in talking with
another camera. And that being true there will never be any
kind of transaction, smooth or otherwise, between any 2 of the
48 clients.

Instead there will virtually always be
contention, and instead of being 4 times faster than a
dialup, it would probably be about 10 times slower... at
best!

So 3 channels at 54 MBit/s can't keep up with 48 clients which
sporadically try to pass data through them to the outside.

I see.

I wonder how they managed with 10 MBit ethernet and a bunch
of computers on the same cable ...

Since the press conference will have enough base stations at close
distances to the PJs, we can assume 2-3 channels at 54 MBit/s,
giving 48 PJs (wow, must be a huge press conference!) 3,3 MBit/s.
(which is quite more upload per PJ than most households have,
even on fast connections). A 22 MPix JPEG will take ~15 seconds,
but you'll probably reduce the size anyway, and get 3 or 4
seconds/JPEG.

Never tried any of this stuff, have you! :-)

No, I've never been to your mythical press conference with 48
PJs all using WiFi, as one of the PJs, with a network set up as
incompetently as you describe --- probably because *you* don't
get to set up wireless networks at something more important than
a cleaning brigade for washing the walls behind the mirrors.

-Wolfgang

PS: Unless you can provide proof, it's EOD.

In article , Floyd L. Davidson
wrote:

It's been several years since I've been heavily involved in setting up Wi-Fi
networks, but my recollection is that the client rarely gets to choose. If
the access point is broadcasting on Channel n, that's what gets used.

The access point chooses. That can be automatic or set to a specific
channel.

The AP is not "automatic".

some are.

nospam wrote:
In article , Floyd L. Davidson
wrote:

It's been several years since I've been heavily involved in setting up Wi-Fi
networks, but my recollection is that the client rarely gets to choose. If
the access point is broadcasting on Channel n, that's what gets used.

The access point chooses. That can be automatic or set to a specific
channel.

The AP is not "automatic".

some are.

Define "automatic"...

--
Floyd L. Davidson http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)