On 11-08-24 6:59 PM, Tom Cakalic wrote:
It's occurred to me that many digital camera tests including the most
popular sites do not give an accurate representation of noise
generated by the camera sensor(s).
While they may do a fairly decent job of reporting on noise at a given
ISO, there seems to be no accounting for length of exposure, length of
continous camera operating time, ambient temperature, or in the case
of night shots, shadow noise, which is far more obvious, than noise
say on a light blue British postage stamp... all of which should
affect noise levels.
I attempted a series of 10 10-sec exposures @ ISO 800 last night with
the K-7. Ambient temperature around 60 degrees F. I had noise
reduction turned off. The camera had been operating for< 3 minutes
before starting.
First the camera gave up taking the shots after 5 exposures because I
only allowed a 1-sec interval between shots. However I was mortified
at the amount of noise present, which was totally out of line with
examples seen on dpreview. I'll try again tonight with NR on.
Thoughts?
I hope you are shooting RAW, but talk of NR on and off suggests JPEG. NR
settings have no effect on RAW, where the image is delivered to you,
well, raw. If you inspect JPEGs you are muddling noise performance with
JPEG and NR algorithm behaviour.
And what do you mean by "gave up taking shots"? Are you referring to
the (perfectly normal) pause because of a full image buffer needing to
be processed?
Have you studied the studies available at http://www.dxomark.com/ ?
-bmw
--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow
the directions.
