I finally got around to exploring the CLS preflash sequence in a bit more depth. The following pair of images and explanation are my latest effort.

Two D2X cameras were used, separated by about ten feet in a darkened room. The subject D2X was on a tripod with an SB800 mounted in the hot shoe. Beside it was a light stand with three more SB800s clamped to the column and aligned with the on-camera unit. This camera was set for 1/30 sec shutter time and rear curtain synch for the SB800 speedlights, to provide an interval between the preflash and image capture sequences. The shutter delay was set for ten seconds.

The second D2X (capture camera) was on a tripod with the a 14-24 mm F/2.8 lens. Its shutter time was set to three seconds. To capture the images, I pressed the shutter on the subject camera, starting the ten second timer. I walked to the capture camera and watched the red timing light on the subject camera. When it turned steady, I pressed the shutter on the capture camera and started rotating it through a ninety degree or so arc. With a few practice tries, I could capture the entire preflash and image capture pulse sequence in the center of the capture camera frame. You will notice slight timing variations in the images as I cannot guarantee that my rotational velocity was constant.

Top Image

This image shows all four SB800s in the CLS TTL mode at 0.0 exposure compensation. Under column "M" (master) the on-camera SB800 emits a series of pulses and calculates an exposure compensation value for itself. Then, under column "A" (group A) it exchanges a series of pulses with the Group A SB800 and sends it an exposure compensation value. This process is then repeated for groups B and C. During this interval, the shutter remains closed as indicated at the bottom of the chart.

The shutter opens at the left edge of the "All" column. Just prior to the end of the shutter interval, the on-camera master emits a low power triggering pulse. A short interval later, all four units fire in sync for image capture. Then the shutter closes.


Bottom Image

Two changes were made in this sequence. The on-camera master was set to OFF (---) and the Group B SB800 was set to manual mode at the M1/2 power level.

Under the left-most column "B", the on-camera master emits a pulse that delivers the M1/2 power level to the Group B SB800. This is a one way transmission, as there will be no reading of reflected light and exposure compensation calculation. In other tests, it appears that the Manual mode power levels are always sent prior to any TTL exchange and exposure compensation calculations.

Following this, the Group "A" and "C" preflash sequences are exchanged as in the Top Image.

AT the left edge of the "All" column, the shutter opens. Just prior to the end of the shutter interval, the on-camera master emits a low power triggering pulse. A short interval later, the Group "A", "B" and "C" SB800s fire in sync for image capture. Notice that the on-camera master SB800 does not fire in synch with the other units, as it has been set to OFF (---). Then the shutter closes.


Summary Comments

Without a four channel recording oscilloscope, I am unable to decode the preflash pulses in any greater depth. Earlier work with Nikonian Alson in Holland disclosed that some form of amplitude/frequency modulation coding is apparently employed for the preflash sequence, which must contain channel, group and exposure compensation data at a minimum. My current, ancient oscilloscope is a single channel, non-recording model. My NAS keeps overpowering my motivation to acquire such an instrument.

I suspect, but cannot prove, that the CLS technology makes some kind of negative exposure compensation calculation for multiple remote units to allow for overlapping coverage of the subject. Only Nikon knows, and they never reveal things like this.

Hope this helps take a bit of mystery out of the CLS preflash sequence.

Comments and questions welcome!

More later ... perhaps.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Attachment #1, (jpg file)
Attachment #2, (jpg file)

Hi Hal,
Wow, what great images and such tremendous research you did, into the CLS sequence of flashing ! I wish I had your understanding of this topic so I could fully use my flashes like you do !! I do have the new SB900 flash and I have ordered the new book by Mike Hagen(due out shortly) to help with it, so there is hope yet !
I really enjoyed reading the description of your test and seeing your images shot during the sequence, and much appreciate your input into understanding how to use these nice flashes better !!
Debra

Visit My Nikonian Gallery: http://www.nikonians-images.com/galleries/showgallery.php?cat=500&ppuser=94067
My Flickr Gallery: http://www.flickr.com/photos/dgillilan/

Debra:

How very nice to hear from you again ... long time, no chat!

Thank you for your kind words.

The SB900 is a very nice unit and I would like to see some of your work with it. What are you shooting these days?

Remember, speedlights multiply in the dark.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hello Hal and all Speedlight users !
"Speedlights multiply in the dark" - how true !! That is exactly what worries me. Seriously, I have used three speedlights together on some macro shots a while back (this was before the SB900). The images turned out alright, but clearly I need more experience and understanding with these multiple CLS flash abilities. I now own a D700 which I truly love for its true wide ability. I no longer have to think "1.5 times focal length". I still use the D2X and D2HS so the DX/FX auto feature of the SB900 will be very handy for me and others with both DX and FX bodies are also certainly pleased with this feature on the SB900.
I will study your posting here, more this weekend. I perhaps can offer some postings soon, as I learn to use the SB900 with my SB800 and the juniors, SB28 and SB22(which I use with the SU4 radio slave transmitter things that allow them to work wirelessly).
Thanks again for all the work on this post, I plan to study it and learn from it and I know it will help me, and others.
Debra

Visit My Nikonian Gallery: http://www.nikonians-images.com/galleries/showgallery.php?cat=500&ppuser=94067
My Flickr Gallery: http://www.flickr.com/photos/dgillilan/

Hal,

I suspect, but cannot prove, that the CLS technology makes some kind of negative exposure compensation calculation for multiple remote units to allow for overlapping coverage of the subject.

I had hoped so too but think I disproved this in: CLS i-TTL Power Settings Independent of Group

Regards,
Bill

Visit me at My site

Bill:

Thanks for reminding me of our earlier discussion ... Senior moment.

I guess I'm still hoping that the CLS system is smarter than it really is. If NAS would subside for a while, I would acquire the four channel recording oscilloscope I've talked about and decode the preflash and triggering pulses to see just what is conveyed. It would be nice to probe the signals sent to the flash tube, instead of trying to read the modulated optical pulses emitted by it. Somewhere around here I have a high voltage probe that I used with my ancient scope. Even better, read the date stream ahead of the high voltage section. Both of these of course require taking an SB800 apart and probing around. With SB800s no longer available, I'm a bit reluctant to do this. So much to do and so little time!

FYI: I am assembling a herd of a dozen or more pocket wizards, mixed between the Plus II and Multimax models for use with the speedlight herd. I'm planning some long shutter time night shots of moving vehicles with timed intervals between the firing of the remote groups. The Multimax PWs give me four remote groups (A, B, C, and D) plus the on-camera local (L), instead of the three plus one provided by CLS.

I'm awaiting delivery of the rest of the PW herd and look forward to getting started.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal,

I occasionally think about an oscilloscope too.
I'm sure you would put one to better use.

Great work as always.

Regards,
Bill

Visit me at My site

Amazing stuff, Hal!
_________________________________
Neil
Nikonians Team
My Gallery

Neil:

Thanks you for your comments. They are appreciated.

I believe I have a pretty good handle on the preflash and trigger pulse sequences, both TTL and manual. The remaining mystery is the detailed content of the pulses. We know that, at a minimum, they must contain channel, group, and power level data. Beyond that, it is still a question.

Why am I doing this? I really haven't a clue. Once the rest of my herd of PWs arrives, I will likely drop this for a while.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal, it's not very often that I come across a piece of work as elegant as this on discussion forums. Thank you for taking the time to crack the mystery of CLS.

I'm still amazed at the effectiveness of Nikon's flash technology in such a wide variety of shooting conditions. Flash management has really moved light years ahead (pun intended) in the last 10 years!

All the best,

Ted

Ted Zimmerman

My Website

Visit my Nikonians gallery.

Ted:

Many thanks for your very kind words.

The Nikon CLS technology continues to be a mystery to many Nikonians. I started building my current herd of speedlights several years ago, when the SB800 appeared. I now have a dozen of them, plus at least one of every other current production Nikon speedlight.

Once I decoded the SB800 manual and learned my way around CLS, I discovered the confusion surrounding the mysterious preflash sequence, and started looking for a way to explain it graphically.

The idea for this sequence capture technique came to me at 3:00 am one morning a year or so ago. I got up, set it up and captured the first set of images, and went back to bed. Since then, I've wanted to return to it and see how the manual option worked. Hence, the pair of images above.

I conduct sporadic speedlight workshops in the Phoenix area and will be making a second, annual guest appearance at Winston Hall's Nikonian speedlight workshop in Phoenix on February 14th.

The rest of my herd of pocket wizards (the MultiMax units) should arrive next week and I can get into the challenge of exploring PW and Nikon speedlight configurations. Soooo many configuration options!

Thanks again for your comments.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal,

I have been looking at your CLS sequence images, but there are a couple of things I can't understand for TTL.

When does the Commander send the power setting to each remote? I expected to see a longer burst with that data right before the main fire pulse, but it is not there. Maybe the power setting is sent right after the monitor preflash? But if that were the case, it would be sent before the system knew how many groups were out there, so three could not be any adjustment for multiple overlapping groups.

Ideas?

Russ
Russell MacDonald Photography
Nikon CLS Practical Guide

Russ,

But if that were the case, it would be sent before the system knew how many groups were out there, so there could not be any adjustment for multiple overlapping groups.

I think it has been established that there is no adjustment for overlapping groups.


Bill

Visit me at My site

>Russ,
>
>But if that were the case, it would be sent
>before the system knew how many groups were out there, so
>there could not be any adjustment for multiple overlapping
>groups.
>
>I think it has been established that there is no adjustment
>for overlapping groups.
>
>
>Bill

OK. I didn't realize that that had been established.

So, when in the sequence is the power setting sent to the Remotes?

Russ
Russell MacDonald Photography
Nikon CLS Practical Guide

Russ,

It looks like right away, on a Group by Group basis.
I'm surprised too.


Bill

Visit me at My site

>Russ,
>
>It looks like right away, on a Group by Group basis.
>I'm surprised too.
>
>
>Bill
>

That's what I thought, and I'm very surprised! I had always thought that the power setting was sent to each group right before the main fire pulse.

That would explain why there is no consideration of overlapping flashes. There can't be with the power setting being sent to each group right after the monitor preflash. Each one is treated individually with no regard for the others!

Russ
Russell MacDonald Photography
Nikon CLS Practical Guide

Wow. Glad I popped in here again. Great teaching as usual, Hal.

Tim
We think because we have words, not the other way around. Madeleine L'Engle

My Gallery

Colleagues:

As much as we would like to think that CLS does compensate for multiple, possibly overlapping remote groups, I am convinced it does not occur.

Consider: On-camera master contributing to the image capture illumination plus three remote groups, A, B and C.

1) CLS does not know how many speedlights are included in each remote group. The on-camera master is obviously just one.

2) CLS cannot know if all speedlights are illuminating the same subject or different subjects.

3) The system knows how many remote groups have been selected prior to shutter opening. That information is sitting there in the on-camera SB900/800 or SU800.

4) It is not clear if CLS know the type of remotes responding to preflash pulses. They could be SB900, SB800, SB600, SB-R200 units. I suspect the answer is no.

5) The exact content of the apparent multiple pulses by the master and each remote group remains a mystery. From the all-TTL image (top) I captured, it appears that the on-camera master emits three brief pulses and the remotes emit two. We know that the on-camera pulses must contain channel number, group code and exposure compensation value at a minimum.

6) With reflected light metering, CLS has no idea of the reflective nature of the one or possibly multiple subjects. Hence, it tries to make everything a neutral gray, group by group, based on what is reflected.

7) Trying to guess at overlapping illumination and subject reflectivity would likely lead to chaos. Given the above, I suspect that metering each remote group independently and leaving exposure compensation adjustments up to the photographer is a wise decision. After all, it is quite simple to do this.

8) Until I get my hands on a four-channel recording oscilloscope (not soon, given ongoing NAS priorities) this is as far as I can go.

9) I suspect the reason for delivering the exposure compensation settings prior to shutter opening is to avoid adding unacceptable delay during shutter open intervals, particularly at the max sync speeds (e.g., 1/250, 1/500). Assuming my rotational velocity is relatively constant and scaling the above images, the preflash sequence occupies approximately twice as much time as the shutter open interval of 1/30 sec, making it 1/15 second. Assuming that it takes one-half of this time to transmit the exposure compensation data to the remotes, this would be about 1/30 sec (0.033 sec = 33 milliseconds). When shooting at 1/250 sec (0.004 sec = 4 milliseconds) with on-camera master and three remote groups firing, there clearly is not enough time (4 milliseconds) to deliver 33 milliseconds of data.

10) While circuit times are very fast, they are not instantaneous. While the remote speedlights appear to be firing at relatively low power during the preflash sequence, they may need a bit of time to recycle in case they will be firing at full power during the shutter open, image capture interval.

11) I presented the above two images at 16 by 20 inches on easels at Winston Hall's Nikonian speedlight workshop here in the Phoenix area last weekend and we had a great time discussing them.

Thanks to all for your comments and discussion.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal (et al.), thank you for helping me understand Nikon's flash system. Fortunately for me, it usually just works . This reply got me thinking, please tell me if I'm wrong. Flash exposure is determined by the amount of light that hits the sensors in the flash exposure area. So any light that is reflected back to that area is used to determine exposure. So, for example, if the flash exposure area is on the subject's cheek, and both group A and B reflect off that cheek, wouldn't the "base" power be determined by A+B? And the actual power of each flash determined by the relative FEC?

I think I made that clear as mud. Basically, I think I'm asking "Won't CLS compensate for overlap if the overlap is in the flash exposure area?"

Ok, I just thought of a simple experiment. Two lights in two groups hitting the same target with the same power. If it overexposes the target then I'm wrong. If the exposure is correct then I'm right. Right?

Kraig

"The wisest follow their own directions" -Euripides
"I thought there would be more elephants" -C. Columbus

>if the flash exposure area is on the subject's
>cheek, and both group A and B reflect off that cheek, wouldn't
>the "base" power be determined by A+B? And the
>actual power of each flash determined by the relative FEC?

No, I don't think so, because right after each group sends its monitor preflash, the commander sends the power of that group right back. It doesn't wait to see if other groups will hit that spot. This indicates that the bases the power of each group solely on the amount of reflected energy from the monitor preflash for that group.


>I think I made that clear as mud. Basically, I think I'm
>asking "Won't CLS compensate for overlap if the overlap
>is in the flash exposure area?"
>
>Ok, I just thought of a simple experiment. Two lights in two
>groups hitting the same target with the same power. If it
>overexposes the target then I'm wrong. If the exposure is
>correct then I'm right. Right?

I have tried this, and I see no indication of a change in power based on whether the flashes overlap. As far as I can tell, they simply add, which I'm guessing is why they set each group to underexpose by a stop or so.

Russ
Russell MacDonald Photography
Nikon CLS Practical Guide

Kraig:

CLS will not recognize multiple speedlights in a given group illuminating the subject because they all fire at the same time during preflash sequence. CLS will return the same exposure compensation value (within the + or - three stop limits) to all units in that given group.

CLS will not recognize if speedlights in a different group will overlap with units in other groups because their preflash sequences occur at different times. Should units in multiple groups overlap, they will be additive, overexposure should result and the photographer will have to enter negative exposure compensation values for appropriate groups in the on-camera master/commander.

Remember, CLS has no idea how many speedlights are configured in each group and where they are all pointed.

Russ's test confirms this. Thanks Russ!

Edited to correct missing word in first sentence: "not".

Regards,
HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal,

I don't think the number of flashes per Group would be relevant.

What I envisioned is something you might call "matrix CLS".
The contribution for the Master and each CLS Group would be measured at several points on the frame as opposed to taking a single measurement.
Then, using simplified linear algebra, powers (weights) would be computed for the Master and Groups to obtain "even" or "correct" lighting.
Perhaps someday. . .

Regards,
Bill

Visit me at My site

Bill:

Nice catch! The number of units in a group is not relevant.

I missed the work "not" in the first sentence when I was interrupted with a phone call.

Omission corrected above.

Agreed, if every lens used with "matrix CLS" provided distance data and enough sensors are used for CLS, a somewhat more intelligent version may appear. This still leaves the reflectance issue to deal with.

Perhaps if the shutter were up while the preflash sequence occurs, and we had sufficient processing horsepower to fire all three groups, look for distance data, identify overlapping hot spots, etc., across the sensor itself, and figure out which group(s) are involved, we could greatly improve things. I still don't see an easy way around the reflectivity issue.

From another perspective, if the on-camera CLS system received data from the remote units (e.g., model, zoom setting, head position, distance from subject, azimuth setting for camera and speedlights, etc.) things could get really interesting. As it is now, I suspect that the on-camera logic simply sees a burst of light from each configured group as you suggested.

Your thoughts?

Regards,


Thanks,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Amazing work! I will have to come back and study this.

Lakeesh:

Let me know if you have any questions.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hello Hal;

I'm new here, in fact it was this very post which came up on a search which made me register at least for 25 days

Kudos to you for your experiment - nicely executed.

My interest is in the timings for the communication flashes, so if you would have had a 5th strobe in multiflash mode set to 100Hz it might have given a crude timebase?

Anyway my interest stems from an idea to increase CLS range/envelope by using an IR gelled strobe (connected to an optical slave) to act as a sort of signal repeater (or indeed a sort of CLS hub) for the main SBs. If the CLS data comm rate is around 100Hz or less it just may be possible?

best,

John

John,

Much faster than 100Hz.
See SB-800 measurements


Bill

Visit me at My site

Thanks Bill

John:

Welcome to Nikonians! Thank you for joining us.

Bill's link to Alson van der Meulen's SB-800 measurements is the best we have to date concerning electronic details of the CLS technology. I worked with Alson several years ago when my SB800 herd was in its infancy. What is really needed is a four channel recording oscilloscope. NAS and other syndromes keep getting in my way.

There are a couple ways of extending the range of Nikon speedlights that may be of use to you.

1) Place all on-camera and remote units in the SU-4 mode. In this manner, one can daisy chain them somewhat farther than the Nikon specified range. I haven't tested this to see just how far I can go, but I have easily gone sixty feet and more. SU-4 of course places you in a strictly manual mode, with no TTL capability.

I have also fired remote SB800s in CLS TTL mode at sixty feet or so at night, with careful positioning of on-camera master/commander and remote units.

2) Current generation Pocket Wizards (Plus II and MultiMax units) are radio frequency (RF) and will extend the range to an advertised 1,600 feet. These products are also manual mode only, with no TTL capability. I recently acquired a half dozen of each of these units and am working out some complex, timed interval night shots with moving law enforcement vehicles.

Recently announced Pocket Wizard models claim to provide RF CLS TTL capability at similar ranges. The Canon models are supposed to ship in a few months and the Nikon versions later this year.

Thanks again for joining us. I hope you find your Nikonian experience rewarding.

Regards,


HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal,
I heard you recently mesmerized a Nikonians Academy i-TTL, CLS class in Phoenix with a full presentation on this subject.
Thank you.

Have a great time
JRP (Founder & Administrator. Nikonian at the north-eastern Mexican desert) Gallery, Brief Love Story, The Team
Join the Silver, Gold and Platinum members that help this happen; upgrade. Join your personal web site to the Nikonians WebRing
Make sure you check our workshops at The Nikonians Academy

JRP:

Thank you for your comments.

Winston and I always have a great time together. The class size (14) is just right and provides a lot of individual, hands-on time with the students. I think everyone went home with some new skills and insights. It is always rewarding to see "the lights come on" with CLS students. I am meeting one of Winston's students today at a local camera store for additional discussion.

I continue in spare moments to assemble my two day CLS, etc. workshop. I'm giving a two hour CLS intro at a local camera store in April to test a couple presentation ideas. I have added pocket wizards to the initial agenda, as they are now rising to greater levels of visibility with the recent PW announcements of their new Canon and Nikon TTL models. It will obviously be a while before I can get my hands on a pair of the Nikon models, but there is much to cover with the current offerings, particularly the MultiMax units.

Thanks again for your comments.

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal;

Thanks for your warm welcome.

Yes I am aware of the mandraulic way of working with the SU4 system both M and A modes. In fact I have several of the SU4 widgets which help to keep old units like the SB25 in the fold.

The flexTT5/1 system from PW looks interesting, from what I understand they hijack the comms and broadcast it wireless using their proprietary system. Look forward to the reviews when they come in thick & fast.

I was hoping for a cheap way of extending CLS but the frequency Nikon uses is very high. It surprises me that Nikon haven't produced a CLS hub themselves.

best,

John

John:

The idea of a CLS hub, while intriguing, has some practical considerations, circuit delays and timing not the least of them.

Example: If we stick to the current three remote groups A, B and C, and we have remote speedlights in all three groups on the upwind side (my newly coined CLS technical term) of the relay hub and additional units in all three groups on the downwind side, and the relay is accomplished optically, how does the on-camera unit sort them all out? The speedlights in Group A on the upwind side of the relay hub are not likely illuminating the same subjects as those on the downwind side. Additionally, the relay hub will begin reporting the downwind units slightly after the upwind units begin and they will overlap. Current optical signaling is limited to one basic channel. Yes, we could do optical frequency division multiplexing, but this would likely drive sale prices beyond anything most photographers would pay. Adding an artificial timing delay in the relay node simply slows down the preflash sequence, which is already fairly long when three remote groups are reporting. See the problems?

As we have already seen with pocket wizards and the Sekonic meters, using radio frequency (RF) signaling requires three sets of different frequencies to cover communications licensing in North America, Europe and Japan. This adds to design and manufacturing costs and higher prices.

The current non-CLS pocket wizards offer vastly greater range and are easily configured. Yes, one must work strictly in manual mode, with no TTL capability, but I shoot that way almost exclusively so it is not a problem for me.

If, by extending CLS, you mean more groups, we already have a solution for that providing you are using SB900 and SB800 units. By placing them all in the SU-4 mode, each individual speedlight can be configured independently of the others and the three group limit disappears. Once again, you are strictly in manual mode with no TTL capability.

At the rate we are going, we will soon need a small, portable USB keyboard simply to configure our speedlights. (grin!)

Much to think about.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hi Hal;

Sorry - when I spoke about extending CLS I was referring mainly to range or blackspots.

If we talk about extending groups then the holy grail for me would be similar to a wireless DMX type strobe setup with the mini-mixer sitting on camera hotshoe. I love manual remote control.

best,

John

John:

Take a look at the Pocket Wizard MultiMax units if you have not already done so. These units give you four remote groups (A, B, C, D) plus one on-camera called L, for Local.

With these units, you can produce timed interval shots a couple different ways, with intervals in the hours if required. I've had a bit of fun experimenting with a half dozen of these units firing at 0.0, 1.5, 4.0 and 8.0 seconds after shutter opening. Following initial setup of the remote units, all control of which groups will fire is done at the on-camera unit. Yes, PWS will fire a pair of speedlights from each unit, so long as they are identical and do not have excessively high trigger voltages (>250 volts according to the manual)

Current PW range is given as 1,600 feet, absent any thick concrete walls, metal buildings, etc. I've not tested this yet, but it is certainly much greater than the optical CLS, which I have fired at night in excess of sixty feet or more.

Thanks for your comments.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Great continued discussion on this topic.

I'm going to make a guess/prediction that the CLS system will remain at three groups for a very practical reason - the time that it takes to complete the CLS signaling and preflash sequence. The more groups available, the longer the sequence will take, and there is a practical limit to how much time can be spent in this sequence prior to exposure in real life photography.

In fact, some of the links previously referenced look at this issue in relation to eye blinking at the time of exposure. At the time, the D70 and D2X were current models, and it was found that the preflash sequence using the built-in flash on the D70 operated slowly enought for the exposure to capture the blink caused by the light from the preflashes, while the D2X (SB800 as commander) was rapid enough that the exposure occurred before the eye could blink.

The analysis was one of the factors that has led me to believe that the preflash signalling protocol is not highly depending on a timing pulse because of the apparent differing speeds with which these two camera models could accomplish the preflash sequence, making me believe that there is something in the on-off pattern of the preflash that controls the response from the remotes.

nl

Neal:

Nice to see you again. Hope all is well.

Agreed: Three is probably a practical limit for current CLS technology. At some point, with more than three remote groups plus an on-camera master engaged, the assorted preflash circuit delays will become intolerable as the milliseconds add up. Faster on-camera and remote unit circuitry and related technology are possible, but at what cost? At some point, consumers will refuse to pay it.

Without the four channel recording oscilloscope that I keep talking about, I am limited to the approximate scaling mentioned earlier in this thread for timing analysis, based on the 1/30 sec shutter time. The unknown here is how constant is my rotational velocity on the D2X capturing the preflash and image capture pulses.

I believe the preflash sequence functions asynchronously, given that a pulse from the on-camera master/commander is used to initiate the exchange with each remote group. The only truly synchronous interval appears to be that between the trailing edge of the image capture pulse and the simultaneous firing of the master and all remotes. This interval is short enough, that very slight differences between remote units (microseconds?) are not a problem.

Think how far we have come from the days of flash powder!

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hal - what a great thread. Thank-you for the CLS analysis. I had wondered if signaling was by light or if radio frequencies were also involved. Now I know. Very interesting. I just bought an SB-600 for my D90 and am having a lot of fun with it. Peter

Peter:

Welcome to Nikonians and thank you for your comments.

CLS technology is a truly marvelous tool. I hope you are using your SB600 in wireless remote mode with your D90.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

How did I miss this gem? Thanks Hal.

I think I'm missing something. I get the feeling that iTT does not consider the individual outputs of the groups? IOW, if you had two groups set to 0 "compensation", the shot would be one stop overexposed?

I always assumed it was calculating based on each group illuminating the same subject whether it was or not. It doesn't make sense that the exposure is set for each group independently. With 3 groups, you could be as much as 1.58 stops over.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

>I think I'm missing something. I get the feeling that iTT does
>not consider the individual outputs of the groups? IOW, if you
>had two groups set to 0 "compensation", the shot
>would be one stop overexposed?
>
>I always assumed it was calculating based on each group
>illuminating the same subject whether it was or not. It
>doesn't make sense that the exposure is set for each group
>independently. With 3 groups, you could be as much as 1.58
>stops over.

No, I know from experience that images with overlapping flashes on the subject will not normally be overexposed. In fact, I usually have to increase compensation to get proper exposure.

I think I know why.

The question has been asked many times here on Nikonians and other forums as to why, when using a single flash wirelessly in TTL, it always underexposes. I think the system assumes you may have multiple flashes and that overlap and overexposure could be always be a possibility, so it underexposes each group intentionally by a stop or so.

Russ
Russell MacDonald Photography
Nikon CLS Practical Guide

Ah...OK. I'm just starting to use it (CLS), albeit with QTTL wireless and have been adjusting for the additional unit being added. It seemed to cause slight overexposure.

I guess I'll have to get out my calibration target and mess with it initially.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

>Ah...OK. I'm just starting to use it (CLS), albeit with QTTL
>wireless and have been adjusting for the additional unit being
>added. It seemed to cause slight overexposure.

OK. I was only talking about Nikon SB's. I haven't used Quantums wirelessly.

Russ
Russell MacDonald Photography
Nikon CLS Practical Guide

Hi Russ,

Yes, I knew that, but would think they work the same way. I have also noticed over exposure with my pop up speedlight used with an SB800.

In the past, I have used my SB900 with an SB800 as two key lights to gain a stop with large groups, or to illuminate wide FOVs. In that case, I guess they were one group, so I would not have noticed.

Scott Chapin
Powder Springs, GA, USA
Nikonians Team Member

A note to Hal
That was a wonderful example of "kitchen table" research, well done.

Have you considered using sound-cards to use as a 2 channel oscilloscope for each of pair of monitored channels? Four channel cards are not too expensive to really simplify things. Some scope software have pretty versatile data capture options, and best of all, there are several software packages that are free. Their crystal controlled clock, low jitter and 44.1Khz to 192Khz sampling rate at 16-24 bits makes for some pretty high res measurements. Some sound cards even handle DC. An AC-only input can be used if the slower measured signal is gating a known HF signal(, in duplex mode the sound card provides any AC signal you want, makes pulse width measurements a snap by just counting the carrier transitions under a measured pulse.

I have the needed test equipment but not enough camera gear to duplicate your experiments using a 4 channel DSO or logic analyzer. The logic analyzer would need photo-transistor sensors but that would be a simple to create optic>logic level signals.

My TEAS(Test Equipment Acquisition Syndrome) overpowers my NAS. Good used test equipment is cheaper than good camera gear. There are exceptions, my RF service monitor cost me 3 times what a D3X costs. That could have gotten me a few good lenses, which right now I would probably use more often.

I built my own wireless triggers (and studio 600w/s strobes) but have hardly used them since a week or so after I built them in the fall. The SB900 and Sigma 530Super do everything I need(or can afford) right now.

By the way, for low power signaling, there are overlapping RF spectrum in the 3 main markets, which means MMCs are dirt cheap for those frequencies as the entire RF section of a wireless system.

Maybe you need a few less SB800s and I need one less HP DSO:>)

Stan
St Petersburg Russia

Visit my Nikonians gallery.

Stan:

Thank you for your kind words. These images have proven very useful in conveying the CLS preflash and image capture pulse logic to users.

The good news: Between us we may be able to decode some of the Nikon CLS pulses.

The bad news: We are half a world apart.

Thanks for joining this thread and for the great advice should I overpower my NAS long enough to purchase the scope. I hadn't considered sound cards ... great idea!

At the moment, I am building a herd of Pocket Wizards. With two more MultiMax units (on order) I will have six Plus IIs and seven MultiMax units. Following more initial testing here at home, I will be setting up some night shoots with moving law enforcement vehicles, etc. The goal is multiple images of a single moving vehicle along a path, with the final one right in front of the camera. The MultiMax units are perfect for this with their extensive interval timing features. Now, I am pretty much limited to capturing their light trails

I was in St Petersburg for a couple days a few years ago. Clearly not enough to thoroughly see the city, but enough to get a taste of its very rich history and architecture. Simply beautiful!

Thanks again for joining us.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Hi Hal
That is quite a project and sounds really interesting.

Since you have been here and for a couple days, was it possibly on a cruise ship? Most of the 190 major cruise ship port calls in the summer come for 2 or 3 days. If it was a cruise ship we may have met without knowing it. I am in the business of providing tours for cruise ship passengers who arrive each summer by the hundreds of thousands...quite a change from my long career in the music recording industry.

I moved here 6 years ago and seldom go back to my house in Marin County, just north from San Francisco, 3 times in 6 years. I have a ton, literally, we weighed it thinking I could ship it over here, of test gear, mostly HP, Tek, Motorola, Sound Technology, and much more sitting in a storage unit in California. I've brought over what I can fit in my suitcases each time I've returned, leaving out any clothes.

One thing you could really use in exploring the CLS light pulse train is an Arbatrary Waveform Generator. When you think you are close, program the generator to create the same pattern and test it in a reproducible convenient way. To bad I won't be back for months because I could get one out of storage and loan it to you.

Do you think there is a modulation scheme that involved amplitude? One post mentioned that, but I would bet against it in such a noisy path with so many reflections and ghost paths. It would be hard to tell until finer looks are seen into a single burst to see what it is made of. My hunch is that it is a simple pulse position coding without getting too fancy. There is not that much data to send and it is not that fast of a data rate.
Next time I have access to some more of my gear I will try to dissect a burst to see what is inside the envelope you displayed so well in the photos. I am amazed that hand panning could have such a constant interval between events, that must have taken a lot of practice or you just have a steady hand.
A local community college or university might have an electronics club that has donated gear from manufacturers. That might be a source of enthusiastic helpers who just happen to have DSO and Arb Generator needed to inspect the pulses and then test the theory by controlling the slaves with a programmed coding of what you find using the ArbGen. If that code was data stream secret was broken, there could be lots of DIY projects that could take the lighting system to the next level. Maybe doing the adjustments as discussed for multiple light sources on the target, or repeaters or computer control of the whole works for specialized requirements just like what you are working with now. It would also allow same channel individual control of any number of flashes using RF spread spectrum techniques that would open up all sorts of possibilities.

All very interesting.....
Good luck with your sequenced shot project...

Stan
St Petersburg Russia

Visit my Nikonians gallery.

Stan:

Yes, we were there on an NCL cruise of the Baltic Sea. Great time! We took several of the tours: Cathedrals, Hermitage, Peter Hauf (sp?), etc.

Alson in Holland took a look at some of the pulses a couple years ago. Go here to see his work. This is as much as we have to go on to date. We all realize that this is an academic question.

The CLS coded pulses must include channel, group and power level data at a minimum, and possibly more. Example: When the test button is pushed, the remote groups will fire sequentially, at one second intervals so the user can check to see if they are receiving the data. Firing all at once might be difficult with multiple groups spread over some distance.

You probably know that the PW people have announced two new products that convert the Canon and Nikon wireless speedlight technology to RF. They claim full TTL compatibility for the units. The Canon units are possibly shipping now, with the Nikon units scheduled for second quarter. I have been in touch with the PW people. As soon as I can get my hands on a set of the Nikon products, I will test them. As I understand it, they are picking the signals up with induction coils of some kind, not optical sensors. Interesting. If so, they have likely already decoded the pulses and know what is going on.

Stan, I am supposed to be retired. I'm busier now than I was when I was working! I threaten to get a real job so I can get some time off.

The school project is interesting. We have a DeVry Technology school here and I have given some thought to approaching them with such a project. My fear is that it would take more of my time than I have to spare.

Thanks for your interesting exchange. Let me know what you think of Alson's traces. Let's stay in touch.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

I think the interpretation of the images may be in error. From reading the stuff I can find, the very first bunch of pulses (just a block under the first column) are for remotely setting up the speed lights. The first case is to setup A/B/C for TTL mode. The 2nd case would be setting up A/C for TTL and B to manual.

For each TTL group, the master first send a group select command, then send the pre-flash (low) command. If the camera sees a weak reflected light, it will ask for a 2nd pre-flash with higher intensity.If the return from first pre-flash is good, it will skip the 2nd pre-flash. The reason for this is to save the energy on the remote flash.

The camera calculates (no calculation needed for manual group) the amount of flash needed for each group. It transmit the amount data for all groups in a batch (the block from master under Column All). After a pre-determined time, all groups fire the flash using the amount data received.

Max

I have made some tests and hopefully understood the CLS wireless flash sequences. Here are the posts I made on my blog.

http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-i/

http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-ii/

http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-iii/

Your comments and suggestions are welcome.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

Hi Max
You've done some good detective work. The use of a sound card as a digitizing scope made sense, I was going to do that myself as described in last spring. But with a business to run and lots of socializing experiments just sort of got pushed lower on the priority list.
The coding as you displayed is actually a little simpler than I assumed, thinking there would be timing reference pulses.
What is the X axis time scale calibration?
When I built my wireless RF link I used a coil taped to the outside of the flash to read the pulses and just retransmitted them without decoding them. That will still work ok but knowing the signaling it would allow designing a more elegant receiver that mounted inside the remote flash to get access to the optical detector in a mixer that would allow it to respond to either the RF receiver or flash signalling without selecting one or the other. So a mix of units could be used in the same setup.
If someone wanted to package a kit or even have a run made up such a system could be a cheap and better performing alternative to the very expensive PW. The only reason for being so expensive is that it is such a small niche market that no one has bothered to jump into it. The upper limit of what a receiver should retail for would be about $80 and a little less for a transmitter. Due to the extended range, RF channel selection might be needed. Currently the flash link is pretty short ranged so more than one Nikon CLS user at the same event probably do not interfere with each other. Instead of channelized operation, Spread Spectrum all on the same 2.4Ghz band would allow use of low cost MMM's and a logical virtual channel, like WiFi could be used. The band used for data links is one of the only bands that are universal in all the ITU regions and license free.That would mean the RF modules would be so cheap due to their being used in a billion products ranging from wireless headsets to garage door openers.
Good work...
Stan
St Petersburg Russia

Visit my Nikonians gallery.

Stan,

This is not all detective work though. There are actually open literatures that have most of the details. All I did was filling in some of the details that are not clearly documented. Just think about how Nikon can protect their IP.

I like your idea about the transmitter. I was thinking about something very simple: an IR LED that connects to my computers headphone jack so I can simply play back the signals from a computer. A cheap SU-800!

All units are seconds in the charts.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

Max:

I knew somebody would eventually show up with a proper oscilloscope for decoding the CLS pulses. My ancient, single channel, vacuum-tube Tektronix scope is inadequate for several reasons. I have been tempted to purchase one of the new four channel, recording, Tektronix units, but other NAS needs always prevailed. You confirmed what I suspected, but could not prove: it is a fairly simple binary coding system.

Have you had time to look at the FP sync pulse to see if it is a single, prolonged pulse, or a series of shorter pulses? The jury is still out on that question. I tried capturing it using the same technique I used for the CLS coding (two cameras, one rotating during the pulse, etc.) and suspect that it is continuous, but again, cannot prove or disprove it.

Thanks for a nice piece of work.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

Yes, I did look at FP. The pulse duration is ~1/200s. The output during the whole pulse is very uniform. I think it has to be pulses because dumping full power takes less time than that. The frequency is too high for the 96kHz sampling rate I used.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

Max:

Once the Xenon tube voltage reaches ignition potential, and discharge begins, is it possible to drop the voltage and/or current flow, yet maintain the discharge at a lower level in an effort to stretch a single pulse to ~ 1/200 second (0.005 sec)?

My curiosity continues. If the FP mode is a series of pulses, that implies a rise time for each individual pulse, and the possibility of non-uniform illumination as the moving shutter slit traverses the image plane. Yes, a frequency => 96kHz may be adequate to mask that.

Your thoughts?

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

>I have made some tests and hopefully understood the CLS
>wireless flash sequences. Here are the posts I made on my
>blog.
>
>http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-i/
>
>http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-ii/
>
>http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-iii/
>
>Your comments and suggestions are welcome.
>
>Max

Max,

This is great work! I had hoped someone would come up with something like this. I never thought of using the audio channel of the PC. Very ingenious!

I haven't fully gone through all the details yet, but I will, and I expect to finally know whether my previous speculation has been correct about how multiple Remotes work together.

Now, we can really understand what is going on!!

Thanks!!

Russ
Nikonian Team Member
Russell MacDonald Photography
Nikon CLS Practical Guide

>
>Max,
>
>This is great work! I had hoped someone would come up with
>something like this. I never thought of using the audio
>channel of the PC. Very ingenious!
>
>I haven't fully gone through all the details yet, but I will,
>and I expect to finally know whether my previous speculation
>has been correct about how multiple Remotes work together.
>
>Now, we can really understand what is going on!!
>
>Thanks!!
>
>Russ
>Nikonian Team Member
>http://russmacdonaldphotos.com">Russell
>MacDonald Photography
>>"http://NikonCLSPracticalGuide.blogspot.com"]Nikon
>CLS Practical Guide

After some tests and thinking, I think the Nikon CLS isn't too dumb when it comes to managing multiple groups. I think it is safe to say the flash output of a remote group is affected by the other groups. Or the camera is smart enough to manage the outputs of all remote groups as a whole so the final exposure is more likely desirable.

I covered that in the 4th post of the subject.

http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-iv/

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

>After some tests and thinking, I think the Nikon CLS isn't too
>dumb when it comes to managing multiple groups. I think it is
>safe to say the flash output of a remote group is affected by
>the other groups. Or the camera is smart enough to manage the
>outputs of all remote groups as a whole so the final exposure
>is more likely desirable.
>
>I covered that in the 4th post of the subject.

Max,

I read your discussion about how multiple remote flashes operate.

My results were quite different. Whenever two remote TTL flashes overlapped, the image did increase in brightness by about 0.7 stops(assuming both were set to the same power).

Here are a couple of things I have noticed:

1) Books with shiny covers are not good subjects when you are up close. The reflected specular highlights can cause the flash power to decrease unpredictably. This doesn't make a big difference, but it might partially explain why your test didn't show much brightness increase when you used two overlapping flashes.

2) It's not usually a good idea to use use TTL-BL when on the SC-28 cord. The TTL-BL equations require that the distance from the subject to the camera equals the distance of the subject to the flash. Also, TTL-BL works best when the camera and flash are colocated. The background ambient must be the same for the flash as for the camera. Bright areas in the background (a window allowing outside light for instance) from the camera position will cause the flash power to increase to try to balance with it, but if the flash is not co-located with the camera and pointed in the same direction, its power may be increased drastically trying to balance the area that the camera sees.

3) TTL on the SC-28 cord works normally, as though the flash were mounted on the camera.

Russ
Nikonian Team Member
Russell MacDonald Photography
Nikon CLS Practical Guide

>Max,
>
>I read your discussion about how multiple remote flashes
>operate.
>
>My results were quite different. Whenever two remote TTL
>flashes overlapped, the image did increase in brightness by
>about 0.7 stops(assuming both were set to the same power).
>
>Here are a couple of things I have noticed:
>
>1) Books with shiny covers are not good subjects when you are
>up close. The reflected specular highlights can cause the
>flash power to decrease unpredictably. This doesn't make a big
>difference, but it might partially explain why your test
>didn't show much brightness increase when you used two
>overlapping flashes.
>
>2) It's not usually a good idea to use use TTL-BL when on the
>SC-28 cord. The TTL-BL equations require that the distance
>from the subject to the camera equals the distance of the
>subject to the flash. Also, TTL-BL works best when the camera
>and flash are colocated. The background ambient must be the
>same for the flash as for the camera. Bright areas in the
>background (a window allowing outside light for instance) from
>the camera position will cause the flash power to increase to
>try to balance with it, but if the flash is not co-located
>with the camera and pointed in the same direction, its power
>may be increased drastically trying to balance the area that
>the camera sees.
>
>3) TTL on the SC-28 cord works normally, as though the flash
>were mounted on the camera.
>

Russ,

Thanks for your comment.

I noticed the specular reflection in the initial tests so in the second batch of the tests I tried to avoid them by arranging the camera-subject-flash angles. I originally shot just the wall and the results were similar.

In the tests, I changed the mode on the flash to TTL (no BL) when shooting with the SC-28 connection.

I will do some more tests to investigate this further.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

I did some new tests with more appropriate setup and I came to the same conclusion.

http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-iv/

It is difficult to know why we had different results. My guesses are: 1) differences in camera, 2) differences in the test setup.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

>I did some new tests with more appropriate setup and I came
>to the same conclusion.
>
>http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-iv/
>
>It is difficult to know why we had different results. My
>guesses are: 1) differences in camera, 2) differences in the
>test setup.
>
>Max

I have to agree that you are not seeing the same addition of the two groups that I did.

I was using a D200 when I ran my tests. I now have a D3. I should try it again.

My previous understanding of the system timing was that each group was analyzed, and a power calculation was immediately sent out to that group, in which case it could not possibly know about any of the other groups, and therefore could not take their contributions to the image into account.

I was looking at your timing sequences in your part 2, but I don't see any analysis of the 'remote flash output amount' when there are two groups in use. Did you study this sequence to understand the format? I would assume it is an address followed by a power setting for each group.

Russ
Nikonian Team Member
Russell MacDonald Photography
Nikon CLS Practical Guide

>
>I was using a D200 when I ran my tests. I now have a D3. I
>should try it again.

Were you running the latest firmware on D200 when you did the tests?

>
>My previous understanding of the system timing was that each
>group was analyzed, and a power calculation was immediately
>sent out to that group, in which case it could not possibly
>know about any of the other groups, and therefore could not
>take their contributions to the image into account.
>
>I was looking at your timing sequences in your part 2, but I
>don't see any analysis of the 'remote flash output amount'
>when there are two groups in use. Did you study this sequence
>to understand the format? I would assume it is an address
>followed by a power setting for each group.

Nope. The flash output amount for all groups are sent out last, before the trigger pulse. I described the detailed format in Part III of my posts. The format is like this:

{Channel}{CMD Code}{Amount for Group A <Amount for Group B> <Amount for Group C>}.

Details are here:

http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-iii/

Please share the results if you indeed get a change to run the tests again.

Thanks.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

>I
>>don't see any analysis of the 'remote flash output
>amount'
>>when there are two groups in use. Did you study this
>sequence
>>to understand the format? I would assume it is an address
>>followed by a power setting for each group.
>
>Nope. The flash output amount for all groups are sent out
>last, before the trigger pulse. I described the detailed
>format in Part III of my posts. The format is like this:
>
>{Channel}{CMD Code}{Amount for Group A <Amount for Group
>B> <Amount for Group C>}.

I'm not sure I agree with this.

When there are multiple remotes, how would each remote know which power setting was for it? For instance, if there were only a single group assigned to group C, where would the the power setting pulses occur? Immediately or delayed? If they came immediately after the Channel ID, how would the group C remote know these were for it? How would it know that there was no group A or B?

If the Group C pulses always occur a certain amount of time after the channel ID, then I can see how it would work if the remote had a timer to count the delay.

Russ
Nikonian Team Member
Russell MacDonald Photography
Nikon CLS Practical Guide

Russ,

There is the 'channel id' code, then the 'cmd code' and for the power setting there are 3 bytes that follow the 'cmd code' byte 1 is the 'A' channel, byte 2 is the 'B' channel and byte 3 is the 'C' channel. If a channel is not used then the value of byte for that channel is '00'. Each byte for the power level is 8 bits long with each bit having a value of '0' or '1'.

No power 0/0 = 00x or 00000000
Full 1/1 = 01x or 0000001
One half 1/2 = 0Cx or 00001101
one quarter 1/4 = 1Bx or 00011001
One eighth 1/8 = 25x or 001000101
One xixteenth 1/16 = 31x or 00110001
One thirty seccond 1/32 = 3Cx or 00111101
One sixty forth 1/64 = 49x or 01001001
One one hundred twenty eight 1/128 = 55x or 01010101

The bits are timed intervals after the command code.

George
My Nikonian Galleries

I understand code assignments to the power. That's not the issue.

My question arises from this comment in the author's work:

"The amount data for multiple groups are concatenated together. If there is just one group, the number may be shorter than 8 bit since zero bits are not transmitted. However if there are more then one group configured, the amount data for the first group(s) is always 8 bit (padded with dummy bits)."

What I am asking is, if there is no Group A and no Group B, then where will the power byte for group C be located in time? Is it padded out to its normal position with a long string of zeros? Not according to the author.

And if it is not padded with zeros, how does the Group C Remote know that the data right after the '0101' word is its data?

This is why I thought that the first 2 bits in the power byte must be dedicated to addressing which group the power setting was for.

Russ
Nikonian Team Member
Russell MacDonald Photography
Nikon CLS Practical Guide

It appears that a 00x parameter value is sent.

George
My Nikonian Galleries

>What I am asking is, if there is no Group A and no Group B,
>then where will the power byte for group C be located in time?
>Is it padded out to its normal position with a long string of
>zeros? Not according to the author.


It is padded with zero so the remote flashes don't get confused.

>This is why I thought that the first 2 bits in the power byte
>must be dedicated to addressing which group the power setting
>was for.

No, it is not the case. I have updated part iii of my post to include an example in which group A is off but group B is on.

http://dptnt.com/2009/11/nikon-cls-advanced-wireless-lighting-part-iii/

The system is quite robust. Even if a remote group is configured in master unit and it is not actually powered on, no flash in other groups will get confused to assume its position.

All remote units see the complete information from the master unit and it can pick up the correct information intended for it.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

>If the Group C pulses always occur a certain amount of time
>after the channel ID, then I can see how it would work if the
>remote had a timer to count the delay.

The time it takes to transmit 16 bits of information.

Btw, there is a flash amount cmd code (4 bit) after the channel ID, but before the amount data.

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.

Max:

I suspect we all agree: the CLS remote exposure compensation algorithms running in various Nikon camera bodies and speedlights will remain unknown, until somebody with a logic analyzer and reverse assembler reverse engineers them.

I may be wrong, but I'm guessing that the on-camera master system does not know how many speedlights exist in eeach remote group. If they did, this implies that each remote unit in a given group identifies itself during the preflash sequence. It also implies that each unit in a remote group would be aware of the others, so that their responses would be timed to avoid conflicts.

Perhaps each remote unit in a given group fires at full power during the preflash sequence. The on-camera system could recognize the total of N remote units and derive the number in each group. This falls apart when one considers that a remote group could consist of one SB900, one SB800, one SB600, one SB-R200, etc. In order to sort this out, the on-camera system would have to know what each unit in a remote group is, as well as how many in the group. Getting sticky.

Assuming the on-camera system does not know how many units are in each group, how does it determine the power level to send to each unit within the group? Assume the on-camera system decides that a given group should fire at X power level. For one remote unit in the group, this would mean "Fire at power level X." For two remote units, this would mean "Each of you fire at X/2 power level." Etc.

Have you given any thought to looking at the pulses returned by the speedlights in the remote groups during the preflash sequence? A second channel on your scope plugged into the PC port on a remote SB900/800, so you could see the timing relationships between the remotes and the on-camera master system?

Might be interesting.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

I think the power level for the group is sent and each unit in the group uses the the same value.

The test level for a group would be at a level less then the full power to allow for a fast recycle. And the camera's sensor would pick up the contributing light for each group and compute the needed value for the exposure combining all the groups. And then how is the overall exposure and each group's power level computed? There has been a response from Nikon UK that there is a different algorithm for the pop-up vs. mounted Speedlight. But Nikon UK may not have the real answer.

But what happens when you have SB-600 and SB-800/900 mix since the power levels available for the SB-600 are different from the SB-800/900. My guess is that the SB-600 ignores the value for the 1/128 power level.

George
My Nikonian Galleries

Russ, Max et al:

Agreed: The on-camera system probably sends back one power level for each group. This still leaves unanswered the question: what if each remote group has multiple units?

On several of my law enforcement night shoots, I would have two or three distinct, separate groups: a distant patrol car with two or more lights on it, a backup officer with one or two lights on him, an arresting officer at the window of a stopped car with three or more lights. In these setups, the light from any one remote group would have little, if any, effect on the other two groups due to distance and setup. In all my night shots, the remote SB800s are off-camera, so the system cannot identify the location of any lights. All the camera can see is the light reflecting off the subject(s). How does the system know that there are three distinct subjects, instead of just one? How does it know if the light from all three groups is converging or diverging? Good questions.

In my top image at the start of this thread (all three groups in TTL), there appear to be three pulses emitted by the on-camera master for the three remote groups, and two pulses returned from the remote units. Look at the shape and intensity of the on-camera and remote pulses. Is it possible the remote pulses may include encoded data based on their perception of light reflected from "their" subject? Nothing to date suggests the on-board Camera/SB900/800/SU800 configuration cannot receive such data. The decoding logic is there when on-board SB900/800 units are used as master, and perhaps with the SU800 as well, even though it is never used as a remote unit.

The popup speedlights may be another case, which might account for the different algorithm you mentioned. Unless Nikon has isolated the bus from the PC port when in remote mmode, it may be a simple chore for Max to plug into a remote SB800/900 and take a look at the preflash pulses returned to the on-board master.

I may have to set up the above test again, this time with two or more remote SB800s in each of the three groups. This would tell us if all remotes within a given group respond simultaneously or not to the preflash sequence pulses from the on-board master.

If the SB600 ignores the 1/128 power level direction, and fires at the closest thing to it (1/64), it will be on stop too bright, which may be significant.

The deeper we get into this, the more questions arise.

Regards,

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

>Russ, Max et al:
>
>Agreed: The on-camera system probably sends back one power
>level for each group. This still leaves unanswered the
>question: what if each remote group has multiple units?

Wouldn't this work itself out automatically?

Let's say there was only one remote in a group, and it fired its preflash and the camera measured for simplicity '6'. Then, let's say that whenever an energy level of 6 is measured it assigns a power level of 1/2th for that flash.

Now, let's put three remotes in the same group all aimed at the same spot. They all fire their preflashes simultaneously, so now the energy measured by the camera is '18' (3 times as much). This will cause the power calculation to come up with a power of 1/3 as much or '1/6th'. So now each of the 3 remotes will fire at 1/6th power and when all 3 remotes are added together it will total a power level of 1/2, just like with a single remote.

Russ
Nikonian Team Member
Russell MacDonald Photography
Nikon CLS Practical Guide

It could be that simple ... Then again ... ?

HBB in Phoenix, Arizona
Nikonian Team Member

Go here for a list of membership upgrade benefits.

Photography is a journey with no conceivable destination.

The three pulses from the mater are 1) pre-flash command, which tells which group should do the pre-flash, 2) pre-flash trigger low, and 3) pre-flash trigger high. The last two are just simple double-pulses.

The remote unit emits two pre-flashes immediately after seeing the pre-flash trigger signal. The second one should be stronger than the first one. The 2nd one is requested if the camera sees a weak return from the first pre-flash. It will skip the 2nd pre-flash if the first one is good. At least this is the case with a SB-800 as a master. With the built-in flash of D200, it always asks for two pre-flashes no matter what.

Remote unit does not communicate anything back. I emits only the pre-flashes and the final flash.

All remote flashes in the same group will fire with the same flash output amount, at the same time.

Max

Visit My Blog. Visit My Nikonians Gallery.

Visit my Nikonians gallery.