K-7 long exposure bug?


iceblinker

Link Posted 04/02/2010 - 21:28
With my K-7 with the latest firmware, after a 1 min exposure with Bulb mode, the camera is locked up for a further minute as if it is doing dark frame subtraction NR, but I have Slow Shutter Speed NR turned OFF!

Is there something wrong with just my camera or do all K-7s do that?

The sensor also makes a buzzing noise during the extra minute even though SR is turned off. That's another bug.
~Pete

Mike-P

Link Posted 04/02/2010 - 21:54
I may be talking total rubbish but I seem to remember reading somewhere that it can only be turned off for a maximum of 30 seconds and after that it defaults to on.

Don't ask me where I read or when .. or infact if I ever did
. My Flickr

davex

Link Posted 04/02/2010 - 22:00
Quote:
it can only be turned off for a maximum of 30 seconds and after that it defaults to on

If that`s correct it could be the only thing that would put me off buying one.
Somebody please confirm

Davex.
K5 + 8mm-500mm zooms and primes
Please feel free to play with any images I post.
My flickr: link

iceblinker

Link Posted 04/02/2010 - 22:14
You are correct, Mike, thanks. I've just had it verified on the DPR forum as well.

Bit disappointing, although I don't do many long exposures.
~Pete

Photon

Link Posted 04/02/2010 - 23:12
I think I had a similar K20D problem: I was trying to take long time night shots of a star configuration, in the hope of showing star trail effect of our worlds rotation within the universe. I had left the shutter open via bulb release cable and a power pack (transformer) for approx four hours in the back garden. To my dismay on closing the shutter there was no image on the rear monitor display panel and the camera appeared to have died, so I foolishly took the battery out and tested the battery voltage. By experimenting I have since concluded that on long exposures, the time to write info to the memory card is approximately equal to the exposure time plus 12 seconds. Thus an exposure of 30 minutes requires an additional 30 minutes 12 seconds to transfer digital data to the card, so you need to close the shutter then leave the camera on the tripod for that additional time! Alas I have not yet taken those star configuration shots - too dam cold at the moment; I don't want to risk condensation within the camera when I take it back indoors or trusting leaving it in the shed overnight!
All five minute jobs take a minimum of eight hours!

iceblinker

Link Posted 04/02/2010 - 23:50
Photon, The extra time after the exposure is not time spent writing to the memory card. It is actually an extra exposure the camera makes with the shutter closed for the sake of noise reduction. The noise of the dark frame is subtracted from the first exposure.

This can't be fully turned off with the K20D either. It just has "on" and "auto" options (Slow Shutter Speed NR). The K-7 adds "off" but it doesn't really mean always off!

What a long wait four hours would have been!
~Pete

Photon

Link Posted 05/02/2010 - 01:01
Iceblinker, Thank you for that enlightenment, excuse my ignorance; my digital experience is less than a year and I've still a lot top learn. It was a long time to wait for such a disappointment and the fear that I had damaged the camera, I had set my watch alarm and went inside to do something else on the computer.

Excuse my ignorance again but where does this noise come from, is it constant for a given serialised camera and is it constant for all ISO settings within that camera? if so could could it not be programmed, at the time of manufacture, into the chip for that individual serialised camera, thus resolving this time delay problem?

After closing of the shutter on long exposures could I take the camera off the tripod and stow in my camera bag, without switching off, whilst the chip is performing this subtraction process, without adversely affecting the image quality?
All five minute jobs take a minimum of eight hours!

iceblinker

Link Posted 05/02/2010 - 04:14
The noise (interference) comes from the analogue circuitry of the sensor and the surrounding electronics. The higher the ISO, and the longer the exposure, the greater the amount of noise.

The pattern of noise is random, I think, so I don't really understand dark frame subtraction. Someone else will have to explain how it actually works.

Yes you can stow the camera in your bag while it gets on with the process without affecting image quality, because the shutter is closed the whole while. That is if your battery has enough juice! It might be wise instead to use a mains adapter for a mighty long exposure.
~Pete

womble

Link Posted 05/02/2010 - 06:17
I thought dark frame subtraction got rid of things like "hot pixels"? I'll admit I had just assumed that! You wouldn't be able to use a four hour exposure around here as there is so much light pollution the image would just be washed out.

K.
Kris Lockyear
It is an illusion that photos are made with the camera… they are made with the eye, heart and head. Henri Cartier-Bresson
Lots of film bodies, a couple of digital ones, too many lenses (mainly older glass) and a Horseman LE 5x4.

My website

MostlyHarmless

Link Posted 05/02/2010 - 08:55
I believe it does handle the hot pixels, but also other mostly-repeatable thermal noise ('amp glow') from the camera as well. Astrophotographers will take and stack multiple 'light' frames - the actual image - to handle random noise, then stack multiple 'dark' frames (DFS frames basically) and subtract that to handle repeatable noise, and also use a stack of 'flats' evenly illuminated 'blank' frames that show up issues in the optical system (vignetting mostly).

Neal

bretti_kivi

Link Posted 05/02/2010 - 09:24
that's (a bit more detailed than I would have gone), but my opinion, too. There is a not-quite random set of noise on many high-ISO images or long exposures and dark frames would see hotpixels off very nicely.

I need to do some experiments on this myself, especially if we get to go to the summerhouse at easter: there I have 0 light pollution and electricity (yay!).

Bret
my pics: link
my kit: K3, K5, K-01, DA 18-55, D-FA50 macro, Siggy 30/1.4, 100-300/f4, 70-200/2.8, Samsung 12-24/f4, Tamron 17-50, and lots of other bits.

iceblinker

Link Posted 05/02/2010 - 09:40
Thanks for the info, Neil & Bret. That's interesting.

While active, can the temperature of the sensor be affected by the location of the camera?

If so, perhaps, ideally, the camera should be left in situ on the tripod after all, rather than be packed in a bag while it does the DFS.
~Pete
Last Edited by iceblinker on 05/02/2010 - 09:42

womble

Link Posted 05/02/2010 - 12:23
I would not pack it in the bag but at least you could move it indoors somewhere safe and go to bed!

I remember other threads on here where our astro fans were sticking with the K10D because you could turn off DFS. I cannot see any problem with that as you don't need fast burst mode or AF for astro photos of hour long durations!

Cheers, Kris.
Kris Lockyear
It is an illusion that photos are made with the camera… they are made with the eye, heart and head. Henri Cartier-Bresson
Lots of film bodies, a couple of digital ones, too many lenses (mainly older glass) and a Horseman LE 5x4.

My website

iceblinker

Link Posted 05/02/2010 - 12:38
womble wrote:
I would not pack it in the bag but at least you could move it indoors somewhere safe and go to bed!

But if it's warmer indoors, I'm wondering if that will affect the noise?

Or maybe the ambient temperature is virtually irrelevant compared to the heat generated inside the camera?
~Pete

MostlyHarmless

Link Posted 05/02/2010 - 13:58
The astrophotographers don't move anything - same temperature, everything. So yes it would affect the noise - but they are chasing minute amounts of signal in lots of noise, so I think I'd try taking it inside and seeing if I was happy with the results.
Neal
Last Edited by MostlyHarmless on 05/02/2010 - 13:59
Add a Comment
You must be registered or logged-in to comment.