[mcgregni]

Also sometimes a quick 'jiggle' of flash compensation, say try-0.5, then -0.7, can help. The distance indication is only a guide, and with other factors (such as lens zoom setting seeing a wide scene) can be pushing your subjects beyond the minimum power/distance the flash is capable of.

Does it work better when you stop down to F5.6? Also what flash head zoom setting are you using? You can try a wider one to reduce flash exposure if all else is at the limits ....or move the subjects further away ....

---------- Post added 30-06-16 at 13:40 ----------

You'll only get the hang of this if you really seize firm control over all aspects of both flash and ambient exposures.

[sinus007]

Hi,
Thanks. I'll try all the suggestions later tonight.
I don't usually use direct flash - most of the time bounce and/or diffuser.

[beholder3]

Hi,
To answer some of the questions:
shooting in Av
ISO100
F2.8
1/60s

If you want to identify the cause of the error, use what I described above. Not open up the aperture. Not Av. Not long shutter times.

The difference between P-TTL and A is about 2 stops.

That there is a difference between two completely different metering systems is to be expected.
In order to be able to analyze it really for you, you need to share the two raw files (not JPG).
We need to see the raw histograms. Only they can tell us, if the photo is correctly exposed.

Maybe you just like it a little darker and the light metering in the flash prefers darker outputs. Then this whole thread has a simple solution: To get your subjective preference in PTTL you always dial in -2 flash exposure compensation in the flash or on the camera.

I mean, I can use A mode or I can find the right settings in M mode and get a more or less good lighting. But why to have P-TTL flash and not use the feature.

Misconception. The camera in M mode still means P-TTL at work.

You do need to understand something about maths: Normally without flash you have only three variables: Time & ISO & F-stop. To get really consistent results you should fix two (like ISO and Fstop in AV Mode) and let the little computer calculate just one: shutter times. One degree of freedom is good for predictability. If you turn on AutoISO in normal shooting mode you already have many, many correct exposure settings: all are combinations of ISO and shutter time.

Using a flash you get a fourth variable into play! Strobe light output. Many people are not aware of that.
That is why I suggest using M-Mode on the camera, where you set ISO, shutter and F-stop. Because now PTTL can/must only adjust one thing left: flash output.

The worst thing you can do is use Tv or Av and AutoISO and Flash. Because the absolutely correct exposure now is hundreds of triplets of parameters, a three dimensional space of correct solutions.

[Flasher]

Just a thought sinus, try jiggling the flash in the hot shoe and dont lock the lever. Sometimes my flash doesnt connect the P-TTL pins on my K3 and I have to twist it slightly to get it to work. It still fires but at full power. Also make sure it is sliding in all the way, could be something stopping it.

To fault find start with the basics and then get more complicated.

These flashes are upgradable and there were previous versions that didn't work properly on the latest cameras but I suspect (not confirmed) that if it works on the K-3 it should work on the K-1.

---------- Post added 07-01-16 at 08:45 AM ----------

The things I would check and do one at a time so you know which one was the fault are: (I'm suggesting all things even though you sound like you know what you are doing but just to be sure)

Check the seating in the hot shoe.

Check the exposure compensation of the camera (bit of a doh statement but just in case!)

Check the flash settings in the camera. Press the flash four way flash button and select the first selection (flash on) and compensation = 0

Check the flash is connected in P-TTL. The distance scale on the flash should show when the shutter is half and it should move up and down when the Aperture is changed on the camera. Zoom the lens (listen for the noisy motor in the flash).

Not sure if there is anything in the new K-1 menus that effect the flash but it's worth a look (I only have a K-3)

[sinus007]

@Flasher Thanks. I did all wiggling and jiggling. Didn't help.
@Beholder3 Thanks, I'll do the test shots as you suggested.

[Flasher]

One strange thing I have just had a play with my K3 and when in Av the minimum shutter is 1/125 at f2.8 even in a dark room. You mentioned 1/60.
Turn the flash off and see what the shutter is (should be slow in a darkened room), turn the flash on and see if the shutter changes up to the minimum. Point the camera out the window to daylight and the shutter should go to the max shutter (1/160 or 1/180 on K3 and perhaps 1/200 on K-1?)

Of course the K-1 may be lower min shutter and I am using 1/3 EV stops not 1/2 so that may be the difference.

[sinus007]

Beholder3,
Here's the link to the folder on Google drive with the test shots.
https://drive.google.com/folderview?id=0B6WzqzArbiT7dUhFZ0FzNVF6OTg&usp=sharing

All settings as you suggested. I used 24-70 lens. Distance's about 5 m.
@24mm
1113 P-TTL mode
1115 A mode

@70mm
1118 P-TTL mode
1120 A mode

Thanks again.

[Flasher]

Keep in mind that if all the settings/distance etc are all the same (emphasis on the Same settings!) and it works on the K-3 then it should work on the K-1, simple as that. Changing modes, distance f-stop compensation should not be necessary.

I have just tried my K-3 and it underexposes by about 1-2 stops on Av in a darkened room (used histogram to determine that). A exposes correctly (as expected.) It does the same in M (as beholder said M uses P-TTL)

Both of my AF540 give the same results. If yours is over exposing then it sounds like the flash needs a firmware upgrade or the camera is not metering P-TTL correctly OR you have not set the K1 camera to the correct settings.

Have you updated the K-1 firmware Sinus?


I've just seen the photos, will be interested to see the diagnoses.

[sinus007]

Yes, I did put the latest f/w in K-1.
As for K-3, I didn't compare K-1 and K-3 with that flash with the same settings. I just remember that I didn't have issue of overexposing with K-3 unless I was too close to the subject.

[Flasher]

Yes, I did put the latest f/w in K-1.
As for K-3, I didn't compare K-1 and K-3 with that flash with the same settings. I just remember that I didn't have issue of overexposing with K-3 unless I was too close to the subject.

Oh ok I misunderstood your original statement. As mentioned I tried my K-3 with your settings and the flash worked fine (slightly underexposed.). I'm using a Sigma 28-70 f2.8 but that shouldn't make any difference.

I looked at the exif and the only thing I could see is that you are using AF-C (release priority). Just in case there is a bug in the K-1 try using AF-S.

Just a question with the photos, the second photos "A flash" did you set the flash to f8? It looks like it is still set to f2.8

---------- Post added 07-01-16 at 11:21 AM ----------


And can you put up a raw photo showing the problem with f2.8 in Av. Thanks

[sinus007]

Flasher,
I believe all shots are @F8.0

[Flasher]

Flasher,
I believe all shots are @F8.0

That's a bit strange then, the second shots should be correctly exposed if you set the camera to f8 and the flash A mode to f8.

A mode is independent from the camera and should expose correctly if the flash iso and f value match the camera. Try removing the flash from the camera in A mode at f8, it should expose correctly (if you have an RF trigger).

Also just checking but you can't select A mode when the camera is set to Av (you have to turn the camera off to select A)

Looks like your flash maybe faulty. Have you tried a different lens?

[sinus007]

Flasher,
I didn't set anything on the flash when it was in A mode. Sorry.

[Flasher]

When your camera is in Manual mode you press the "mode" button on the flash and it cycles through to "A". On the flash screen you should see the iso and the F number. Pressing the S button selects the F number (start blinking) and you can select the number to match your camera F setting using the S thumbwheel. Press S again. Do the same with iso.

That then sets the flash to expose for F8.0 at iso100. It uses its own sensor on the red lens part of the flash to measure the exposure. This is independent of the camera settings so if you set the camera to F2.8 the image will be over exposed just like if you set the camera to F8.0 and the Flash A mode to F2.8 then it will be underexposed.

What do you see on the back of your flash wen you select A mode?

[beholder3]

Beholder3,
Here's the link to the folder on Google drive with the test shots.
https://drive.google.com/folderview?id=0B6WzqzArbiT7dUhFZ0FzNVF6OTg&usp=sharing

All settings as you suggested. I used 24-70 lens. Distance's about 5 m.
@24mm
1113 P-TTL mode
1115 A mode

@70mm
1118 P-TTL mode
1120 A mode

Thanks again.

Thanks, I think we are getting closer here. Obviously the "brightest" P-TTL image is 1118.

Here is the Raw Histogram:


Overexposed starts at EV 3 in that scale. So you see not one of the 4 RGGB channels is seriously overexposed, even the brightest green channel still has 2/3 of a stop free headroom. The JPG preview confirms this is a perfectly well exposed image.

The 24mm P-TTL image is good as well.

If these P-TTL images appear "too bright" to you it really is just a matter of taste and you can and should finetune this by using flash exposure compensation on flash or in camera.

This leads me to the assumption that maybe your use of setting up things for A mode is not perfect. The A mode images obviously are underexposed.

If you just forgot to adjust F-stop to F8 and I do push the raw file by the missing 3 stops in software they are much closer. I would say the remaining difference then is 2/3 of a stop. And that easily is the difference between the two different metering methods:
The camera was set by you to matrix metering which always is somewhat difficult to predict in the details. The flash metering is done by one photo cell which does not differentiate anything in the image just average, so there are bound to be different results.

If the latter is what you prefer set Flash ExpComp to -2/3 in PTTL and you should be good to go.