"Edit with photoshop" downsizes the image
Why is this happening?
Original file size: 5472 x 3648 turns into 5306 x 3538 in Photoshop.
-
My guess would be that the smaller size is a result of C1's lens correction.
0 -
I don't think so, I imported the same file into Lightroom, applied lens correction, and opened it in Photoshop to find it matched the original file dimensions.
The C1 PSD file compared to the LR PSD file is identical - only ~100px smaller on each side.
0 -
I don't think so, I imported the same file into Lightroom, applied lens correction, and opened it in Photoshop to find it matched the original file dimensions.
I just tried an experiment - Raw exported to JPEG:
With lens correction: 3597 x 5406 - 21.8M
Without LC: 3830 x 5756 - 24.5M
You can hardly compare LR with C1. They are different apps from different developers.
1 -
The lens correction depends upon the focal length, of course. It is weaker (or none) for telephotos, but larger for wide angle optics.
0 -
Lens correction is always going to manipulate the interpretation of the values of pixels. In some cases this may mean a geometric adjustment that would result in a non-rectangular result. And maybe some poor quality pixels that might be better discarded than retained.
C1 makes assessments in such situations and crops according to those assessments but with the option to change the default crop applied or ignore it totally. The results may vary in subtle ways from what a camera manufacturer will do to its own jpgs when compared to the unmolested RAW files - but not by much in the cases I have looked at in the past when investigating the results.
The difference seems to be that the manufacturers, having published the figures for output dimensions for thier products, move things around so that those dimensions are achieved. Presumably LR does the same.
In both cases they will be introducing additional adjustments around the edges of interpreted images to achieve the target published numbers and accepting those dimensional adjustments is not optional.
For C1 the adjustments can be optional. The final dimensions are controlled by the settings for the output process.
For anyone worried about nominal observed "loss" of pixels - if you are expecting to Edit externally with PS or any other applications and want to retain the natural dimensions you should be able to set up an output definition that will ignore the crop when the interpreted file is sent to the external application and then do whatever you want to it to retain the crop, eliminate the crop or re-stretch the result to suit the intended purpose.
LR, I suspect, simply follows the camera manufacturer's decision and "tweaks" the results to fit the published dimensions for out-of-camera files. It's easier for them. Mostly, no one will question the effects of the crop so long as the file dimensions are as "expected".
1 -
I just tried an experiment - Raw exported to JPEG:
With lens correction: 3597 x 5406 - 21.8M
Without LC: 3830 x 5756 - 24.5M
You can hardly compare LR with C1. They are different apps from different developers.
I stand corrected! I wonder what's the reasoning for going the route that results in losing pixels versus how LR does it.
Thanks for clearing this up.
0 -
For anyone worried about nominal observed "loss" of pixels - if you are expecting to Edit externally with PS or any other applications and want to retain the natural dimensions you should be able to set up an output definition that will ignore the crop when the interpreted file is sent to the external application and then do whatever you want to it to retain the crop, eliminate the crop or re-stretch the result to suit the intended purpose.
Even when I bring it into PS with "Edit with PS" option with "ignore crop" enabled, it still downsizes. It's not a huge issue for me. If it's user error on my part, then I wanted to make sure I corrected it.
Thanks for your detailed explanation!
0 -
Kevin,
Downsize by the same amount as if the crop is not ignored?
If so it may simply be that PS is reading an internal value from the file or the file size as adjusted, includes the cropped out areas. I'm not sure how one would provide a useful image dimension figure reliably for a file with significant pin-cushion or barrel distortion.
Ultimately there is every chance that the file will eventually have some sort of re-sizing or pixel count adjustment associated with it before output so it's all a bit of a moot point really.
Loss of some sort of vital-to-the-image-result edge detail is a different matter but then that would really be a matter of framing and the required framing might not be possible for some reason. Retaining edge detail but having it distorted might also be acceptable in some instances. But they are, excuse the pun, "edge cases" in this context.
I don't have PS but I will try to see what Affinity does when I get the opportunity.
I suppose there is another possibility in that edge pixels may have some compromises and after processing the RAW data for a basic rendition to a visible image C1 (or any other application) may come up with a different usable pixel count than the manufacturer publishes. That, however, is likely to be very rare if it happens at all.
0
投稿コメントは受け付けていません。
コメント
8件のコメント