K3 aggressively underexposes with flash?

NotaxPen · 04-07-2014, 04:57 PM

Originally posted by awaldram

1 Camera subject distance + processing = min distance (bounce irrelevant)
easiest processing direct min = 2.3ft anything else greater required distance. (bounce = more processing )
ergo
"When I back up a little so the camera is more than about 3.5 feet from the subject, the bounce flash exposures are correct every time"

2 Manual lens means no matrix metering , no p-ttl no pre trigger no processing so I don't understand your reference to it.

3 But you have proved that P-ttl and distance are the variables that effect exposure at min distance !!

---------- Post added 07-04-14 at 01:15 PM ----------

af58af2 = min 1/256 = 1/33,000 pulse
fgz540 = min 1/64 = 1/14,000 pulse

min GN @ 50mm ioso100, fgz540 = GN5, af58af2 = 2.5

My Manual Focus lens. is an "A" lens, doesn't that work with matrix metering and PTTL? The K-3 manual says it does, and it sure seems to work fine.>

I have the 540FGZII, it has min 1/256.
It doesn't matter whether the min. distance is 2.3 or 2.5 feet, when the inconsistency happens at 3.5 feet it's not because I'm closer than the spec because I'm not.

We're still talking apples and oranges here with distance. Flash to subject distance is not the same thing as camera to subject distance unless the flash is on top of the camera pointing forward.

I think what I may be missing is understanding why the camera processes the amount of light from the pre-flash differently when in bounce mode than when direct. Shouldn't it just measure how much light comes back regardless of the path it travels?

In any event, the problem can easily be avoided, I'm just curious because I notice similar behavior as the OP.

---------- Post added 04-07-14 at 04:12 PM ----------

Originally posted by chickentender

Not trying to start anything here, but browsing through this thread simply makes me glad I never use TTL. The time spent here, IMO would be better served just nailing manual flash settings. If you're shooting primarily fixed subjects and product type shots, this is a no-brainer for me. Another case of the camera tech trying to be smarter than the photographer (and failing). The only time I really consider using TLL is in rapidly changing lighting conditions, different subjects, different distances and no time to dial it in manually.
(here ends the voice of dissent)

I agree about the tech trying to be smarter than the photographer and failing. I typically would only use PTTL if I'm moving around, such as a wedding or event. In most cases I generally prefer to use manual because I know what to expect.
However, I am impressed at how accurate the PTTL exposures are with the K-3 and AF540FGZII when I don't get close to it's limits.

derekkite · 04-07-2014, 06:54 PM

I was shooting with an F100 2.8 and Sigma ring flash on the weekend. If I was less than about a foot away from the subject it overexposed by just under two stops, if I backed away the exposure was fine. Pttl. I haven't done much with this combination, this was really my first try. I have been generally quite impressed by the flash exposure on the K-3 in different circumstances, but it seems there are corner cases where it goes wrong.

awaldram · 04-10-2014, 04:01 AM

My point is
The flash duration is typically described with two numbers: t.5 is the length of time for which the flash impulse is above 0.5 (50%) of the peak intensity, while t.1 is the length of time for which the impulse is above 0.1 (10%) of the peak
once the camera -> subject distance is such that you inside T .1 then you can no longer have an accurate exposure, If the camera shuts off the flash at a normal T.1 time then the result will be a seriously underexposed shot.
If it tries to extend the exposure you will get a seriously over exposed shot.

It is not related to 'bounce' as it's the light reflected from the subject to the film plane not the light from the flash -> subject -> film plane.

Where bounce comes in is in intensity (bounce requires more flash power) as you increase intensity so does t.1 increase and hence min distance increases.

so you can never have the same min distance for direct and bounce and bounce min distance will always be greater than direct.

Note what happens to your distance scale when you bounce, There is a reason it switches off.!

awaldram · 04-10-2014, 04:07 AM

I suppose it time to put my money where my mouth is

My contention is that min exposure distance is relative to the flash in use not the camera body.

Providing your operating withing the P-ttl window you can take flash images at any distances down to touching the glass

so lets try it
all shot with SIgma 50mm Macro and Metz 15ms-1 @4inches from front element
1st image f4 exceeds P-ttl windows
2nd image f10 nicely inside window
3rd image shows ambient light level without flash
4th down to 2inches and P-ttl still operating fine.

Attached Images

View Picture EXIF

View Picture EXIF

awaldram · 04-10-2014, 04:12 AM

And to show bounce is irrelevant
1st image bounced (controlling flash blocked with 100% visible light filter)
2nd how setup for bounce

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Pentax flash with a Pentax K3	bscott	Pentax K-3 & K-3 II	3	03-05-2014 01:49 PM
K3 with built-in flash issue - any idea?	AlexZav	Flashes, Lighting, and Studio	6	03-01-2014 02:30 PM
K3 Flash	Bigdomino	Pentax K-3 & K-3 II	2	02-07-2014 03:31 PM
K3 "different" behaviour with pop-up flash	jackassp	Pentax K-3 & K-3 II	4	12-20-2013 03:32 PM
K3 with Metz flash	Jimlad	Pentax DSLR Discussion	2	12-07-2013 02:59 PM

04-07-2014, 04:57 PM	#61
NotaxPen Senior Member Join Date: Feb 2011 Posts: 123	Originally posted by awaldram 1 Camera subject distance + processing = min distance (bounce irrelevant) easiest processing direct min = 2.3ft anything else greater required distance. (bounce = more processing ) ergo "When I back up a little so the camera is more than about 3.5 feet from the subject, the bounce flash exposures are correct every time" 2 Manual lens means no matrix metering , no p-ttl no pre trigger no processing so I don't understand your reference to it. 3 But you have proved that P-ttl and distance are the variables that effect exposure at min distance !! ---------- Post added 07-04-14 at 01:15 PM ---------- af58af2 = min 1/256 = 1/33,000 pulse fgz540 = min 1/64 = 1/14,000 pulse min GN @ 50mm ioso100, fgz540 = GN5, af58af2 = 2.5 My Manual Focus lens. is an "A" lens, doesn't that work with matrix metering and PTTL? The K-3 manual says it does, and it sure seems to work fine.> I have the 540FGZII, it has min 1/256. It doesn't matter whether the min. distance is 2.3 or 2.5 feet, when the inconsistency happens at 3.5 feet it's not because I'm closer than the spec because I'm not. We're still talking apples and oranges here with distance. Flash to subject distance is not the same thing as camera to subject distance unless the flash is on top of the camera pointing forward. I think what I may be missing is understanding why the camera processes the amount of light from the pre-flash differently when in bounce mode than when direct. Shouldn't it just measure how much light comes back regardless of the path it travels? In any event, the problem can easily be avoided, I'm just curious because I notice similar behavior as the OP. ---------- Post added 04-07-14 at 04:12 PM ---------- Originally posted by chickentender Not trying to start anything here, but browsing through this thread simply makes me glad I never use TTL. The time spent here, IMO would be better served just nailing manual flash settings. If you're shooting primarily fixed subjects and product type shots, this is a no-brainer for me. Another case of the camera tech trying to be smarter than the photographer (and failing). The only time I really consider using TLL is in rapidly changing lighting conditions, different subjects, different distances and no time to dial it in manually. (here ends the voice of dissent) I agree about the tech trying to be smarter than the photographer and failing. I typically would only use PTTL if I'm moving around, such as a wedding or event. In most cases I generally prefer to use manual because I know what to expect. However, I am impressed at how accurate the PTTL exposures are with the K-3 and AF540FGZII when I don't get close to it's limits.

04-07-2014, 06:54 PM	#62
derekkite Pentaxian Join Date: Sep 2011 Location: Nelson B.C. Posts: 3,782	I was shooting with an F100 2.8 and Sigma ring flash on the weekend. If I was less than about a foot away from the subject it overexposed by just under two stops, if I backed away the exposure was fine. Pttl. I haven't done much with this combination, this was really my first try. I have been generally quite impressed by the flash exposure on the K-3 in different circumstances, but it seems there are corner cases where it goes wrong.

04-10-2014, 04:01 AM	#63
awaldram Veteran Member Join Date: May 2008 Location: Hampshire Photos: Gallery Posts: 732	My point is The flash duration is typically described with two numbers: t.5 is the length of time for which the flash impulse is above 0.5 (50%) of the peak intensity, while t.1 is the length of time for which the impulse is above 0.1 (10%) of the peak once the camera -> subject distance is such that you inside T .1 then you can no longer have an accurate exposure, If the camera shuts off the flash at a normal T.1 time then the result will be a seriously underexposed shot. If it tries to extend the exposure you will get a seriously over exposed shot. It is not related to 'bounce' as it's the light reflected from the subject to the film plane not the light from the flash -> subject -> film plane. Where bounce comes in is in intensity (bounce requires more flash power) as you increase intensity so does t.1 increase and hence min distance increases. so you can never have the same min distance for direct and bounce and bounce min distance will always be greater than direct. Note what happens to your distance scale when you bounce, There is a reason it switches off.!