ISO invariance for the K-3iii

**Robert N** · 11-20-2022, 09:12 AM

A question mainly for the astro photography experts out there who use the K-3iii. Can you advise the ISO invariance characteristic of this camera (i.e. over what ISO range can image exposure be increased without incurring more noise). Having a modern sensor, I would expect invariance to start at a fairly low level. Also from an invariance perspective, what ISO level for astro/night photography can be utilised to avoid highlights being burned out.
Also from a dynamic range graph I have seen, it looks like the native ISO range is 100 - 51,200.
Looking forward to your replies, which may start an interesting debate.

**JensE** · 11-20-2022, 09:37 AM

That really depends on your intended post-processing. Judging by https://www.photonstophotos.net/Charts/PDR_Shadow.htm#Pentax%20K-3%20III and, considering that Pentax usually handles things in full ISO stops (scaling -0.3 and +0.3), a test of ISO100, 200 and 400 with your post-processing workflow seems to be sufficient, higher settings don't seem to improve things further.

**UncleVanya** · 11-20-2022, 10:41 AM

This might help:
https://www.photonstophotos.net/Charts/PDR_Shadow.htm#Pentax%20K-3%20II,Pentax%20K-3%20III

The k-3ii isn’t the same sensor but you can see the effect of the accelerator on shadow improvement. The k-1 vs k-1ii shows a similar relationship so I think it’s valid.

In my opinion you should try iso 100, iso 400, and iso 6400. Compare those and see.

simon_199 · 11-21-2022, 12:03 PM

Originally posted by JensE

That really depends on your intended post-processing. Judging by https://www.photonstophotos.net/Charts/PDR_Shadow.htm#Pentax%20K-3%20III and, considering that Pentax usually handles things in full ISO stops (scaling -0.3 and +0.3), a test of ISO100, 200 and 400 with your post-processing workflow seems to be sufficient, higher settings don't seem to improve things further.

This is an interesting point, not completely unrelated from the issue of iso invariance. Is ISO on the K-3iii handled in full stops of analog amplification and 1/3 stops are "digital gain" ? As far as I understand the relevant plot on PhotonstoPhotos to assess that aspect of the ISO behavior should be the one linked below. They say it's not meant for comparing different cameras quantitatively but only the qualitative behavior. Cases where the 1/3 stops are possibly to be avoided should look like the 5d, with constant plateaus, the K-3 seems very linear with no plateaus, whereas the mark 3 is kinda in-between...

**JensE** · 11-21-2022, 03:51 PM

Originally posted by simon_199

As far as I understand the relevant plot on PhotonstoPhotos to assess that aspect of the ISO behavior should be the one linked below.

Both K-3 and K-1 seem to switch in full ISO increments, even though the largely ISO-invariant sensors doesn't show it as clearly as the 5DM2 with its nice plateaus in this graph. The K-1 II and KP have a very wavy 'photographic dynamic range' in the higher ISO range, in full EV periods, indicating that also the noise reduction parameters seem to be optimized at full increments. The K3-III does seem to be more fine-tuned. An additional complication for Pentax might be that while the nominal ISO increments are equally spaced by 1/3 EV, the actual implementation, as recently discussed, better matches -0.3 EV / +0.3 EV around the full ISO stops in terms of captured raw values, likely also contributing to the overall waviness - photostophotons uses the nominal ISO.

11-20-2022, 09:12 AM	#1
Robert N Site Supporter Join Date: Oct 2017 Location: Sale, Cheshire Posts: 249	ISO invariance for the K-3iii A question mainly for the astro photography experts out there who use the K-3iii. Can you advise the ISO invariance characteristic of this camera (i.e. over what ISO range can image exposure be increased without incurring more noise). Having a modern sensor, I would expect invariance to start at a fairly low level. Also from an invariance perspective, what ISO level for astro/night photography can be utilised to avoid highlights being burned out. Also from a dynamic range graph I have seen, it looks like the native ISO range is 100 - 51,200. Looking forward to your replies, which may start an interesting debate.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Pentax K-1 II with class leading maximum dynamic range and ISO invariance	beholder3	Pentax K-1 & K-1 II	13	05-14-2018 07:46 AM
ISO invariance for Pentax DSLR cameras?	edri	Pentax DSLR Discussion	18	01-24-2018 09:31 PM
Iso invariance	GUB	Photographic Technique	44	11-16-2017 12:42 AM
ISO Invariance ETTL	indy	Photographic Technique	14	03-30-2016 06:42 AM
ISO invariance and PP possibilities (link)	IchabodCrane	Digital Processing, Software, and Printing	5	08-27-2015 10:13 AM

11-20-2022, 09:37 AM	#2
JensE Site Supporter Join Date: Dec 2015 Location: Leipzig Photos: Gallery \| Albums Posts: 1,972	That really depends on your intended post-processing. Judging by https://www.photonstophotos.net/Charts/PDR_Shadow.htm#Pentax%20K-3%20III and, considering that Pentax usually handles things in full ISO stops (scaling -0.3 and +0.3), a test of ISO100, 200 and 400 with your post-processing workflow seems to be sufficient, higher settings don't seem to improve things further.

11-20-2022, 10:41 AM	#3
UncleVanya Loyal Site Supporter Join Date: Jul 2014 Photos: Gallery \| Albums Posts: 28,452	This might help: https://www.photonstophotos.net/Charts/PDR_Shadow.htm#Pentax%20K-3%20II,Pentax%20K-3%20III The k-3ii isn’t the same sensor but you can see the effect of the accelerator on shadow improvement. The k-1 vs k-1ii shows a similar relationship so I think it’s valid. In my opinion you should try iso 100, iso 400, and iso 6400. Compare those and see.

11-21-2022, 12:03 PM	#4
simon_199 Pentaxian Join Date: Oct 2014 Posts: 614	Originally posted by JensE That really depends on your intended post-processing. Judging by https://www.photonstophotos.net/Charts/PDR_Shadow.htm#Pentax%20K-3%20III and, considering that Pentax usually handles things in full ISO stops (scaling -0.3 and +0.3), a test of ISO100, 200 and 400 with your post-processing workflow seems to be sufficient, higher settings don't seem to improve things further. This is an interesting point, not completely unrelated from the issue of iso invariance. Is ISO on the K-3iii handled in full stops of analog amplification and 1/3 stops are "digital gain" ? As far as I understand the relevant plot on PhotonstoPhotos to assess that aspect of the ISO behavior should be the one linked below. They say it's not meant for comparing different cameras quantitatively but only the qualitative behavior. Cases where the 1/3 stops are possibly to be avoided should look like the 5d, with constant plateaus, the K-3 seems very linear with no plateaus, whereas the mark 3 is kinda in-between...

11-21-2022, 03:51 PM	#5
JensE Site Supporter Join Date: Dec 2015 Location: Leipzig Photos: Gallery \| Albums Posts: 1,972	Originally posted by simon_199 As far as I understand the relevant plot on PhotonstoPhotos to assess that aspect of the ISO behavior should be the one linked below. Both K-3 and K-1 seem to switch in full ISO increments, even though the largely ISO-invariant sensors doesn't show it as clearly as the 5DM2 with its nice plateaus in this graph. The K-1 II and KP have a very wavy 'photographic dynamic range' in the higher ISO range, in full EV periods, indicating that also the noise reduction parameters seem to be optimized at full increments. The K3-III does seem to be more fine-tuned. An additional complication for Pentax might be that while the nominal ISO increments are equally spaced by 1/3 EV, the actual implementation, as recently discussed, better matches -0.3 EV / +0.3 EV around the full ISO stops in terms of captured raw values, likely also contributing to the overall waviness - photostophotons uses the nominal ISO.