C1-6.0.1--Is NR better?
This is actually a continuation of a discussion that has been going on in another thread on the subject of noise reduction in C1-6. Phase One released an update today--v.6.0.1--which is said to improve some high ISO quality issues, whatever that means.
In the other thread, I had posted image samples that had been processed in ACR and C1-6 for comparison. I think that it may be of some interest to see how the same image looks after NR in v.6.0.1. Are any improvements appreciable? The sample image is a 100% crop of a photo of some blue jeans taken with a Pentax K-7 at ISO 3200. The image was underexposed in the camera and boosted by one stop in raw, making it equivalent to an ISO 6400 exposure. I will repost the baseline image along with renditions from ACR, C1-6 and C1-6.0.1. In all versions, I have tried to produce the best combination of NR and retention of detail.
Comments are welcome. I hope that others are testing the NR capability of the latest update as well.
Rob
C1-6, no NR
http://gallery.photo.net/photo/12089555-lg.jpg
ACR
http://gallery.photo.net/photo/12089553-lg.jpg
C1-6
http://gallery.photo.net/photo/12089554-lg.jpg
C1-6.0.1
http://gallery.photo.net/photo/12099393-lg.jpg
In the other thread, I had posted image samples that had been processed in ACR and C1-6 for comparison. I think that it may be of some interest to see how the same image looks after NR in v.6.0.1. Are any improvements appreciable? The sample image is a 100% crop of a photo of some blue jeans taken with a Pentax K-7 at ISO 3200. The image was underexposed in the camera and boosted by one stop in raw, making it equivalent to an ISO 6400 exposure. I will repost the baseline image along with renditions from ACR, C1-6 and C1-6.0.1. In all versions, I have tried to produce the best combination of NR and retention of detail.
Comments are welcome. I hope that others are testing the NR capability of the latest update as well.
Rob
C1-6, no NR
http://gallery.photo.net/photo/12089555-lg.jpg
ACR
http://gallery.photo.net/photo/12089553-lg.jpg
C1-6
http://gallery.photo.net/photo/12089554-lg.jpg
C1-6.0.1
http://gallery.photo.net/photo/12099393-lg.jpg
0
-
I did not notice any NR improvements in CO6.0.1 too. Maybe PhaseOne shoud more detailed describe which exactly improvements take place in new releases... now users have to explore them 😊 0 -
If I may, I'd suggest a studio image which contains high and low frequency details and smooth gradients in all the channels. Yesterday I was playing with a Pentax K-7 ISO 6400 pef downloaded from the dpreview comparison widget used in the recent reviews. My conclusion is that with such a demanding image I need to disable NR and sharpening in C1, do the colour NR in ACR and then send it to Noiseware (for some reason I can't get as good results with Topaz DeNoise which everyone seems to be praising -- it might be that Noiseware is more intuitive for me).
Also, the concept of "getting the best" out of each conversion is highly subjective so although in practical use "getting the best" is what counts the most, yet if you want to really compare the differences between v.6 and 6.0.1 I'd suggest using the same settings and comparing outside C1 (in an external viewer).0 -
[quote="sankos" wrote:
... yet if you want to really compare the differences between v.6 and 6.0.1 I'd suggest using the same settings and comparing outside C1 (in an external viewer).
I have made such comparition and compared them in FastStone Image Viever - images looks very identical, just a couple of 'single pixels' were removed/added.0 -
Don't expect NR changes in 6.0.1 for DSLR images. A few tweaks were added for Phase One backs.
-Lionel0 -
My own take from working with this image and others is that any changes in v.6.0.1 are subtle, at best. The Color noise slider may be a little more potent (i.e. it removes chroma noise at a lower level,) and the character of the grain may be slightly smoother and less blotchy. ACR stll produces a smoother result, but at the cost of losing detail. So, it's a trade-off, but I think that many people would prefer ACR's NR for the simple reason that they consider all noise to be evil. We should also bear in mind that pixel peeping at 100% or larger does not give a true sense of how an image will look in real life. So, I conclude that although C1 is not the best program for NR from images with very high levels of noise, for most other images, it is more than adequate.
Rob0 -
Hi Rob.
I have read your posts, and I agree with your comments. I am happy you found the small improvements in 6.0 compared to 5.2.1. I was just commenting on the subtle changes from 6.0 to 6.0.1 for Phase One backs.
This being said, you can expect more general improvements for high noise images in future releases.
-Lionel0 -
[quote="Lionel12" wrote:
Hi Rob.
I have read your posts, and I agree with your comments. I am happy you found the small improvements in 6.0 compared to 5.2.1. I was just commenting on the subtle changes from 6.0 to 6.0.1 for Phase One backs.
This being said, you can expect more general improvements for high noise images in future releases.
-Lionel
Lionel,
Thanks for your contribution. So the latest update only affects Phase One backs. Well, as I said, the changes in v.6.0.1 are "subtle, at best." I'm glad to hear that Phase One is actively working on improving the NR function. I actually think that it's pretty good right now, even if not best in its class.
Rob0
Post is closed for comments.
Comments
7 comments