Nikon Coolpix 5700 severe flash underexposure problem

Hi, All!

I bought my 5700 in late July to replace my older Fuji 4900. I
bought the Nikon 5700 to replace the Fuji expecting sharper
images with less noise, which it does do well. I didn't have
the budget for a DSLR and lenses.

The 5700 produces excellent results outdoors with bright or
cloudy light, but flash performance is quite variable and
often poor. Examples of the poor pictures I got are posted to
alt.binaries.photos

I collect automobile pictures for a hobby, both digital camera
and scans. Much of the time, I am outdoors at cars shows and
city streets taking pictures some people call "street
shooting", but I also frequent large museums such as the Henry
Ford Museum and the Walter P. Chrysler Museum, both in
Michigan.

Because I want to shoot car pictures in museum settings, I
need more range than is provided by the 5700's built-in
Speedlight, so I bought a Vivitar 728 external flash with a
guide number of 92 at ISO 100.

The Vivitar 728 is an ordinary "dumb" electronic flash in that
it has *no* set-uo or zoom or any options. It's only purpose
is to pour a lot of light onto the subject. I verified at the
camera store where I bought it that it could synch up with my
5700's hot shoe and fire either by itself or with Nikon's
Speedlight, depending on how I have it set-up.

About 1/3 of the time, my flash pictures with either the
Vivitar or Nikon Speedlight are quite good. Another 1/3 are
underexposed by maybe 2 f/stops but are easily fixable in my
favorite editor, Jasc's Paint Shop Pro 9. The remainder are 4-
6 f/stops under and look like the proverbial black cat in a
coal bin.

Virtually all camera set-up functions are factory-standard, so
there is little chance I have messed up the operation of the
camera's auto exposure electronics or any other pertinent
parameter. You can see some of this in the EXIF data of my
attached example pictures. If you would like me to list key
components of my set-up, please advise which ones you want to
verify.

I requested help from Nikon's Tech Support. After asking the
obligatory newbie questions, they basically have said "that's
the way it works". Of course, they won't even talk about my
Vivitar because I wouldn't pop the $280 for a Nikon external
flash.

I've also sent about 20 example images straight from the
camera to . Again, I got stonewalled on
the Vivitar and lots of dumb questions. It is not nearly
resolved yet, and I haven't heard from Nikon since my last E-
mail to them last Friday night.

As to the Vivitar, their only recommendation was to shoot in
full manual mode using the 92 guide number.

The other suggestion was the vary the Speedlight control
called Variable Power. You will see from the pictues I've
posted to a binary group that I've tried the max VP value with
contiuing poor results.

My previous digital, a FujiFilm 4900 from early 2001, manages
pretty well in museum environments even though I only used its
built-in pop-up flash unit. Ditto for my wife's little $150
Kodak 6330. Neither produce "good" results, but exposures are
OK and very consistant.

The "bad" images I take with either the Nikon Speedlight or
the Vivitar are underexposed by 4-6 or more f/stops. They
*can* be fixed in PSP 9, but with much higher edit times than
well-exposed images, plus, digital noise is problematic from
the increased amplification of the 5700's electronics trying
to retain detail in the image. The noise is often impossible
to correct without destroying the sharpness and detail of my
car pictures.

Here is what I've done so far to relieve my problem:

Read the 5700 manual thoroughly, several times. Read the
Vivitar external flash manual thoroughly, several times.
Called Nikon Tech Support on 1-800-645-6689 twice. Contacted
Vivitar Tech Support - they said "our flash has no options -
talk to Nikon, it is their problem". Experimented extensively
with various settings on my camera (putting them back to
"default" when I was done so as not to create a problem
through user error). And, consulted with digital photographer
experts who are Cyber friends of mine on PSP's user forums.
Nothing has helped so far.

As best I've been able to tell from reading the manual and
talking to Nikon Tech Support, metering when using either the
built-in Speedlight or an external flash of any brand besides
Nikon is controlled exclusively by the light sensor on top of
the flip-up speed light. To confirm this, I tested changing my
auto exposure mode from Matrix to Spot without improvement.

The Nikon Tech Support reps I talked to, although well-
trained, patient, and thorough, could not explain to me
exactly how the 5700 determines exposure when using flash of
any kind, and suggested I include that in my E-mail to you.

At the suggestion of Nikon Tech Support, I have also
experimented at the WPC Museum by changing both exposure
compensation (the +/- button next to the shutter release) and
varying the Variable Power setting of the Speed light in Set-
up. When pictures are properly exposed, they are definitely
brighter with Variable Power set to 1 or 2, but that did *not*
improve the severe underexposure I've seen in so many of my
pictures. Ditto for exposure compensation.

I thought I might have figured out what was causing my flash
exposure underexposure inconsistency by noting the it often
occurres with dark subjects and/or when there was something in
the field-of-view that was very light colored and would
reflect a large amount of flash light to the Speedlight
sensor. But I have confirmed that this is *not* always the
case.

I've posted a 9 of example JPEGs straight from the camera that
exhibit the problems I'm describing to alt.binaries.photos
with the sugject line Nikon problem #x, and a description of
the particular image.

Hopefully with my explanation and these example pictures, you
may be able to diagnose my problem and recommend a change in
my procedures to get more reliable results.

I did not post the "good" and "almost good" pictures but the
subject distances and ambient lighting is equivalent to the
"problem" pictures. If it would help anyone reading this post,
I can easily post additional pics to alt.binaries.photos.

In conclusion, I can neither prove nor disprove that I have a
technical problem with my 5700. Nothing that Nikon Tech
Support has said so far suggests that it is broken. So far,
Nikon is still hung up on some sort of user error, which is
entirely possible. IMHO, if the 5700 flash system were broke,
I'd get bad results all the time, not the variability I'm
seeing.

I would appreciate anyone's advice as to 1) what I might be
doing wrong, and 2) what I might do to correct it.

Thank you in advance for any help.

 

1. Digital camera hotshoes are susceptible to being blown and thus damaging
the camera from a large firing voltage on the external flash unit. The
general consensus seems to be 12VDC or less. I have two Vivitars, one has a
6VDC trigger and the other has a 45VDC trigger.

2. I have experienced similar things to yours with my 5700. Some pictures
are ok, some very dark. As you have noted the camera measures the reflected
light through a sensor built into the pop-up flash. It seems to me, that the
moment it figures out it has had enough light, then that is the end of the
picture taking. So, the sensor in the pop-up controls the amount of light
given for the exposure, the simulated shutter speed must be fixed at some
setting, maybe 1/60th or even 1/100th of a second. With the addedd flash,
you get more flash than the camera is set for and the sensor shuts the
camera down right away.

3. An experiment would be to obscure the pop-up sensor with something, a
hand maybe, and see what the results look like.

 

Walking down the road minding my own busines, Richard
Tomkins
stopped me and commented courteously ...

I hadn't thought of this being an issue and certainly didn't
know the voltages can and do vary. Are you suggesting the
Vivitar may have damaged my 5700? Is there a way for me to
find out what the firing voltage on my Vivitar 728 is? Can't
find it in their puny manual.

Makes me wonder, though, since I had about a month's
experience with just the Speedlight before my Paint Shop Pro
friends convinced me to pop for an external flash.

Nikon sells two of them, one is about $280 and the other is
something like $340. Both are huge, and clearly meant for
the more serious amatuer and/or pro, using a Nikon DSLR. I
went around and around with a couple of Nikon Tech Support
folk who insisted that Nikon couldn't support a Vivitar. I
said I don't have the budget for a Nikon external - their
response was, basically, "then why'd you buy a Nikon camera
if you don't want to do things right?". Well, call me a
dummy, but I wasn't about to spend $280 on a huge flash for
a camera that cost only $550 after rebate.

Yes, both my experience and what I've been able to wring out
of Nikon Tech Support, what you describe is the case. *But*,
Nikon Tech Support doesn't see this as a defect!

Two points to consider here:

1) "Frank" from Nikon Tech Support says that something
s wrong with my 5700's flash exposure system, after
examaning the 20+ pictures under all sorts of shooting
conditions with the Nikon Speedlight and my Vivitar. Well, I
guess I'll find out in about a month - I sent my 5700 to
Nikon in NY, counting UPS each way and Nikon's turn-around
time of 7-10 business days, I might see my camera before
Thanksgiving.

2) When I first started seeing the phenomenon I described, I
found examples where the flash was bouncing off something
very white, like a car's WSW tires, flash glare on the
windshield, or maybe something in the foreground that'd
catch the full flash blast, all three of which would "fool"
the Speedlight's sensor. Then, I started noticing that the
dang thing also underexposes fairly randomly even when there
isn't anything anywhere in the scene which might cause the
Speedlight sensor to think "I've got enough light now, so
I'll shut down the exposure.

I never thought of trying that, Richard, because the manual
says that the Speedlight will throw an error message if it
can't read light coming in. But, I'll certainly try it once
I get my camera back.

BTW, with my own experiences and yours, as amplified in my
reply to you, I think that the 5700 has a design flaw for
flash photography, but I can't get Nikon to admit it (Duh!).
I have *no* trouble with my "crappy" 4.5MP Fuji 4900 or my
wife's $150 Kodak 6330 in exactly the same shooting
conditions - I tested them both when I first started seeing
the problem.

Being the pessimist and cynic that I am (in spades!), I'm
girding my self for the laconic note I expect to get when my
camera is returned from Nikon Service - "no problem found".

Richard, given that I've got $1,000 in this thing counting
CF cards, batteries, Vivitar and the like, I'm not too
enthusiastic about buying another camera anytime soon.

Nikon told me to put the 5700 into full manual mode and
shoot with my Vivitar using its 92 guide number at ISO 100.
That'd be OK if the 5700 would tell me the distance it is
focusing to! Nikon said "well, just guess!". I did test both
the Speedlight and my Vivitar in various settings in my home
using just a simple tape measure, but haven't tried that in
a museum yet.

I've been looking at these laser rangefinder thingies at
home improvement stores that people use to measure their
rooms electronically. I can get one with a range of 50 feet
for $30. There're two rubs, though: 1) the thing is about
1.5"W x 7"L x 1.25"T so it'll be bulky in my pocket, and 2)
museum curators may freak when they see me shooting a laser
beam at one of their cars!

In conclusion, maybe I'm not going crazy after all if you
also experience occasional or frequent unexplained flash
exposure differences as I do. Given that, isn't it sad that
a manufactuer as renowned as Nikon would be that damn dumb?

 

Power the flash up and read the voltage with a meter --
http://www.botzilla.com/photo/G1strobe.html

If it was too high, your camera would have probably failed by now.


I use Sunpak flashes with my CP 5000. They work really well. There is a
good bit of variance between the different ones I have, so I stick with
the lowest. I think it was $30 used.

I don't think there's anything wrong with your camera or your flash, but
I think there is a problem in how you are using it.

Does your 728 have a Nikon dedicated shoe (4 pins)?

What mode is your flash set to? If you use TTL mode, then the 5700 uses
the external sensor to squelch the flash, and if you have something near
the sensor, like your finger, it will reflect the light back and shut it
off.

If you are using the flash in an "automatic" mode, then the flash uses
it's own sensor, and you need to set the aperature according to the
automatic range in use.

If you have the flash set to automatic mode, and the camera set to
automatic mode, then the camera will try to control the flash, but will
not be able to and you will get unpredictable results.

If you use auto mode on the flash, then your guess doesn't need to be
very good at all, only in the right range. The Sunpak 36dx I use has a GN
of about 120. The middle auto range (for f/4) runs from 3.3 feet to 30
feet.

Besides -- all you need to do is leave the monitor on and review the
shots with the histogram and you will know strait off if the exposure is
OK or not! It's not like cars in museums are action shots ;-)

Bob

 

bob commented courteously ...

Beats me! The manual is pretty poor, but there are *no*
controls whatsover on the flash itself except that I can
pull out the flash head to control the width of the flash
pulse somewhat to compensate for in wide-angle, normal, or
telephoto.

As I said, I don't see an "auto" or "TTL" control or any
control on the Vivitar 728. So, I'd expect that the 1/16"
diameter sensor on the 5700's built-in Speedlight is being
used to measure the reflected light and make all the
exposure decisions.

Yeah, but if you are in full manual, you do need to be
fairly accurate with the distance given the fact that flash
light falls off as the square of the distance and you want
to get something at least mid-range in the exposure so
highlights aren't blown out nor are mid-range and shadows
turned into mud, resulting in digital noise.

I wouldn't be playing with full manual in the first place if
I could get the 5700's "auto" to work. For now, I'll have to
cool my heels until I get the blankety-blank thing back
around Thanksgiving to see if they fixed it, said "no
problem found", or as Richard Thomkins suggested (based on
exposure variability he's also seeing), live with the fact
that it's a design flaw.

The 5700 has a nifty feature where it shows the picture it
just took for 3 seconds and gives me a chance to delete it.
So, I *know* right then if it is OK, sorta OK, or crap. Of
course, there's an ancillary problem where the viewfinder
and/or LCD display looks considerably brighter than what I
see on three different PCs when looking at the actual JPEGs.

But, that's not the issue. The issue is what to do about
lousy exposure once I see that it's 6 stops under, while the
picture I took 5 feet away 2 minutes ago looks perfectly
fine.

I can't speak competantly of the flash(es) you've used nor
what your experience has been with digitals in a museum
environment, so I'm hardly arguing with you, just commenting
on what I see and what Nikon has told me.

 

I use a Vivitar 730AFN with my 5700. I have used it on the camera
hotshoe and off-camera with the SC-28 cord. I have had excellent
exposures under many different conditions. Very happy with the
combination (and price) given the 5700's lack of flash control
features.


+++++++++++++++++++++++++++++++++++
+ The News Guy(Mike) - Seinfeld Lists
+ (two mirrored sites)
+ http://membres.lycos.fr/tnguym
+ http://wave.prohosting.com/tnguym NOWTHISWORKS
+ All things Seinfeld; scripts, trivia, lists,
+++++++++++++++++++++++++++++++++++

 

Are you saying there are NO controls on the back of the vivitar flash?
Other than the power switch and the test button, that is... If there are no
controls, then you could not have tested the flash in manual mode (the
flash, not the camera) as you previously indicated!

Does the front of the flash say 728AFNi?

With a flash in auto mode, it knows how you set the camera, and it takes
care of the exposure. The flash has a sensor on the front that measures the
light. You don't need to be accurate, because the flash is. If your flash
only has TTL mode and doesn't have auto mode, then you wouldn't be able to
try that out.

Its when you have a flash in manual mode that you need to be accurate,
because then nothing is doing any metering, but if your flash doesn't have
a manual mode, then you can not try that either.

If your flash doesn't have a manual mode, then using manual settings on the
camera aren't going to help.

I would guess that the flash system on my 5000 and your 5700 are the same.
My 5000 handles the external Sunpak flash I use really well. If all else
fails, you can sell your vivitar on ebay and buy a Sunpak. Since the Sunpak
has TTL, auto, and manual, you can get really good flash pictures with any
camera.

But if you put it into play mode and roll the dial two clicks clockwise, it
will show you the histogram, and then you will know exactly what the
exposure looks like, regardless of the condition of the LCD.

We still haven't figured out how your flash operates. If it only has TTL
mode and if it isn't working right, then you probably just need a new
flash.

If the flash has an automatic mode, it's only going to be 6 stops under if
you are too far away. If the flash has a manual mode, it will only be 6
stops under if you are way too far away.

Most people in tech support don't know how to use the products they support
-- they just read scripts. It's pretty clear that the guy you talked to
didn't know how to use your flash. That seems to be the real issue --
figuring out how to use the flash, and then determing which settings on the
camera to use to support the flash.

Bob

 

bob commented courteously ...

There are not controls, such as "auto" or "manual". There is, of
course, an on-off switch, a test button, an auto check light, a
ready light, and two sliding scales to do a rough calculation of
speed and aperture at a given ISO and with a given focal lenght.

Also, the flash hood pulls out to change from 28mm to 85MM

Yes, it does.

My understanding from reading the manual is that the *only*
"auto" thing this model flash can do is "auto focus", *not*
"auto exposure". That's my interpretation of what "AF" in
727AFNi means. And, the manual cover says "728 Auto Focus
Flashgun". That's what the 1/2" square red sensor on the front
of the flash is for, *not* for the 728 to control its own
exposure.

But, it *does* work in manual mode, but there're two problems:
1) getting a somewhat accurate distance measurement and doing
something about the 5700's smallest aperture of f/8.0.

I haven't tried full manual on the 5700 in a museum environment
yet, but when I tried it in long (20-30 foot) shots around my
house, I found I could up the shutter speed past 1/125, where it
is supposed to synch, and "simulate" the correct exposure.

I understand manual flash and I understand guide numbers. It's
pretty simple: set the shutter to 1/60 or 1/125, divide the
guide number (92 at ISO) by the distance in feet, and voila!,
you get an f/stop.


Bob, I appreciate your trying to help me, but this discussion is
going nowhere. I see *nothing* on the 728 itself or in the
manual that would indicate that it does *any* exposure
calculation at all, nor is there anything that says it
communicated with the 5700 in any way.

Nikon Tech Support, in the same breath as saying they won't
support a 3rd party flash, insist that the *entire* flash
exposure determination is through the Speedlight's sensor.

I have a last question for you: Did you go to
alt.binaries.photos and actually look at the pictures I posted?
If you did, you would have seen that I get *exactly* the same
crappy very underexposed images with the 5700's built-in
Speedlight as I do with the Vivitar 728. Are you now going to
tell me I don't know how to use that also?

I want help with this, but I'm not an idiot. The Speedlight-only
should work reliably in all shooting conditions for its range
(up to 13 feet) and the Vivitar should work reliably up to its
range of 23 feet. I just don't see why you seem to think I don't
know how to use the Vivitar, but if there's something obvious
I'm missing, please point it out.

 

(much snipped)

Jerry,

This may be of NO use whatsoever to you, since I don't have the
equipment you do. But I do have 3 Nikons - one 950 and two 995s and
I've encountered similar problems with exposure variability. My
solution is rather crude, and certainly not fully automatic.

I have a Slave Flash unit (from many years ago in film photography)
and a solid state detector to trigger it. I mount it on a tripod
close enough to the subject, aim the trigger sensor at the camera, and
use Guide numbers determined experimentally. It may not be state of
the art, but it WORKS!

Just my input.

Olin McDaniel

 

Olin K. McDaniel commented courteously ...

Hi, Olin.

I can't do much right now until I get my 5700 back from
Nikon Service - fixed I hope!

As I mentioned in my OP, the problem of variability also
exists with the 5700's built-in Speedlight, not just the
Vivitar 728. Nikon basically ducked the entire non-Nikon
external flash and suggested I go fully manual.

I may have to go full manual with my Vivitar, and I'm
thinking of buying one of those home improvement store laser
rangefinders to do the distance for the guide number. As you
correctly say, it ain't state-of-the-art but it'll work.

Right now, I'd say it is 50/50 that my Nikon is broken; the
other view is that Nikon just did a shitty job of designing
the flash exposure electronics and software. I hope it's
broke and that they fix it, I'm just not holding my breath.

Thanks for your suggestions.

 

When you get the camera back in three weeks, you will find out for
sure if there is a problem with it.

I don't think you are an idiot, or I would not be conversing with
you.

I tried to find your pictures, but my server doesn't have
alt.binaries.photos. If you want to post them someplace else I'll
take a look. I don't know if it will be possible to determine
anything other than to verify that they are underexposed.

The spec sheet on vivitar's website leads me to believe that your
flash has an auto mode, and that it does not have a manual mode.
But if that's not the case, it wouldn't be the first time a
manufacturer had misleading info on it's website.

When you get your camera back, if they say there's nothing wrong
with it, and if you want to figure out how to get things to work
right, post a new message (or email me) and maybe we can figure it
out. I think there is something you're missing, and I think it
probably isn't obvious.

Bob

 

bob commented courteously ...

Bob, I'm sorry if I got pevish with you. I apologize. It's
just that so many people - mainly at Nikon - have basically
insulted my intelligence for so long I get short-termpered.

Is the E-mail in your header correct, Bob? I'll try E-
mailing you my pics.

I've looked at Vivitar's website also. And, I've re-perused
my meager manual. If I understand it right, the 728 is
*only* TTL-auto with a Nikon or Canon autofocus SLR. Not
even sure if they mean "DSLR", since it doesn't say so.

In any event, as best I've been able to determine through
testing, as well as from both phone and E-mail with Nikon,
the 5700's normal daylight/available light auto exposure
system is completely disabled when it is in "speedlight"
mode.

It determines that by the setting in Set-Up, which I have
set to "Auto". Then, I set the built-in Speedlight to be in
in Fill Flash mode, so it will always fire. If I turn my
Vivitar "on", the 5700 fires that, else, it fires the
Speedlight. But, in neither case is the auto exposure system
operating. When I want to go back to daylight/available
light, I set the Speedlight to only fire when it needs to.

Bob, I never dismiss operator error, here. But, well prior
to posting here and my multi-go-around with Nikon, I got
other opinions that strongly suggest to me that the thing is
either broken or very poorly designed. I just don't have the
variability with either my old Fuji 4900 or my wife's little
Kodak 6330 under exactly the same conditions (I tested them
and reported same to Nikon but those people are incredibly
arrogant!).

Richard Tomkins in his reply to me on Oct. 28 in this thread
confirmed that he, too, had great variability with his 5700
in Speedlight mode.

As I indicated in my previous post, if I have to, I'll go
full-manual and hope the museum guys don't go crazy when
they see me aim a laser at their cars!

 

No appology necessary!

You need to modify the email according to the instructions in the signature
line -- e.g. take out the letters "MAPS"

I do know that Nikon and Canon have different shoes, but since yours has
the letters Ni in the model name, we know yours is for Nikon. The spec
sheet mentions some older Nikon cameras, like the 4004. The 4004 was pretty
basic, so if it worked with that, it ought to work with the newer stuff too
-- it just won't support all the new features. But the 5xxx series cameras
don't support all the new features either, so that's not an issue.

When you say "fill flash" that means "always on" with the little lighting
bolt showing, right? The term "fill flash" can be used with a different
meaning too, but I see in my CP manual they use "fill flash" and "always
on" to mean the same thing. I don't do a lot of flash photography, but when
I have, it's been really reliable. I get overexposure with really close
subjects, and I get underexposure sometimes when I try to bounce off a dark
wall and the flash runs out of power, but in general it works pretty well.

(Yes, they are {arrogant}). The 4900 has a one pin ISO shoe, so when you
use the flash with it, the only thing it can do is fire the flash. The
flash aparently knows it's not on a Nikon, so it uses the other sensor
behind the red spot on the front to know when to shut down. (It has to have
a sensor, because it has a test button/indicator -- without a sensor, it
couldn't know when to light up the indicator. The main purpose of the red
light though, is to light up a scene with infrared light that the slrs can
focus on). Since the 4900 only has one pin, it cannot control the exposure,
other than by selecting an aperature before you release the shutter.

With a GN of 90, you will need to be at least 12 feet away (at f/8) to
avoid overexposure, unless you have ND filters. I'll take a look at the
pictures when you email them and see if I can see anything that gives me a
clue. I'd bet on a problem with the flash before a problem with the camera.

Here's another idea: if you deactivate the extra pins (either by opening
the flash and cutting the wires, or using plastic shims) the flash will
operate in plain old automatic mode, like it used to on your 4900.

Bob