What they both say. Just, the third shot (church door) seems to be leaning to the right just a little bit, or am I seeing things? Lovely sunset-by-the-sea, especially when here it is a cloudy Sunday morning with no hope of relief for a couple of days...
Lovely shots Wayne. One observation though - I think you are over sharpening the images a bit - or more accurately sharpening with a too large radius in the end. If you look at the otherwise beautiful sunset shot you can see that the rocks have been over sharpened but without bringing out the fine detail.
Today is the third day of bad weather here in Stockholm, so I have not been able to go out and shoot. I'm running out of 35/1.4 images to post and I'm really scraping the bottom of the barrel now:
Thanks Charles, Paul, Philippe, and Luka!
Here is California in the Summer we have never ending sunny hot hazy days and so I look forward to the few cloudy days. Of course partly cloudy is the best.Hope you get good shooting weather soon.
Phil, I will look at the levelness of that one again. I corrected it some already and thought it was level using the lines but I will check again. Maybe a little more rotation counter-clockwise.
Luka, my sharpeneing for these shots consisted of input sharpneing in LR 2.6 at default 25/1.0/0.
Then in CS3 I did some midtone contrast sharpneing with unsharp mask at 15/35/0.
Then downsized and sharpened with unsharp mask at 50/0.7/0.
Let me know if you see anything wrong with that flow. My flow for web size shots has never been as good as the one I have for large prints.
Any other suggestions for improving web flow are appreciated. Maybe I need to read Samuli' s workflow in the Z* thread again.
Wayne, I think that last unsharp mask is too coarse. Try instead a smart sharpen, r=0.2px a=100%. Then repeat it until you get the detail you want. Step back a bit if you see that it's becoming over sharpened. The first USM is a strange choice as well. A radius of 35 pixels will smear the image rather than sharpen it.. Did you perhaps mean 3.5?
So far that is automated but after the script I optionally add one to three Smart sharpen r=0.2, a=100%.
And yes, Samuli's description of his workflow is really great - I learned a lot from his description.
Here is a comparsion the method you described (A) and the one that I use (B). I added two smart sharpens at the end of B. Open in separate tabs and flip between them:
In my experience if you are using Canon or Nikon glass, the way you resize the images is not critical. With Zeiss glass however you get typically a lot of fine detail that gets lost if one isn't careful how the resizing is done. The standard bi-cubic interpolation resize really kills detail by averaging pixels..
Excellent shot Luka, as usual I like the composition of #2, but really like artistic feel to #1 and #4.
More importantly, I know it has been mentioned by Samuli in his workflow, but it is great to see subtle differences in your workflow. It is appreciated the detailed workflow, with examples to compare.
Andreas, Look forward to seeing some shots from Spain soon.
Luka,
Thanks for your sharpening flow example and description.
One thing, is "Sharpen x2" some kind of separate step? I don't understand what your mean there.
I will try the two stage reduction with sharpening approach tonight.
I think the fine detail squashing is probably due to the single big reduction approach.
The low-amount, high radius step I was doing is a midtone-contrast enhancement and should not affect the fine detail. It should be similar to the clarity slider in LR and ACR.
I really, I mean really, tried to ignore this thread.....but you guys are not helping out here....
What should I choose .... between C/Y 35/1.4 and ZE 35/2.0 ....?
And, please don't tell me to get both.....
Here is a repost of previous image with your sharpening for web flow.
What do you think? I did two sharpens after the 1024 resize.
Thanks for sharing your flow.
Hmm.. definitely better, but still a bit over sharpened. Rocks are notoriously difficult to get just right.. The two sharpens in the end - were they smart sharpens (radius=0.2)?
Paul Yi wrote:
I really, I mean really, tried to ignore this thread.....but you guys are not helping out here....
What should I choose .... between C/Y 35/1.4 and ZE 35/2.0 ....?
And, please don't tell me to get both.....
Yep, they are different. The 35/1.4 has a more subtle drawing style while the 35/2 is sharper. The color rendering is different as well. If the 35/1.4 comes out in ZE mount, then that would be the one to pick if you had to choose one. Adapted 35/1.4 vs native 35/2 is a tossup. If you can, get both.
Paul,
I have both lenses and like them both.
For wide open f1.4 or f2 shooting I prefer the CZ 35/1.4 and also for low-light shooting.
I also prefer it for portraits and times when you want more separation between subject and background and/or more 3-d feel. For stopped down ~f8 landscape shooting I need to do some comparisons first but both do well with the edge maybe going to the Z* 35/2.
Luka,
Thanks for the feedback. Yes, last two sharpens were smart sharpen with radius of 0.2.
One question, what advanced settings do you use on these smart sharpen stages?
I just used the basic version when I did this.
Also, before I have only used USM and smart sharpening functions and never used the "sharpen" function. What algorithm does that function use? Is it similar to USM? Those first two sharpens really provide a lot of sharpening. I guess my main problem besides not using a two stage reduction process with incremental sharpening was not sharpening more before reduction. I thought people told me that causes artifacts and to only sharpen after resizing.
Also, you use bicubic sharper when reducing each time, right?
Wayne, no advanced settings just the basic one. And no - I use regular bicubic resize, not sharper. I have found that the bicubic sharper is pretty crude when it comes to sharpening.
I have no idea how the "Sharpen" function algorithm works and how it differs from USM. The reason why I'm using it is simply because I used Samuli's resize as an starting point for my own. It worked well, so I never bothered changing it.
denoir wrote:
Wayne, no advanced settings just the basic one. And no - I use regular bicubic resize, not sharper. I have found that the bicubic sharper is pretty crude when it comes to sharpening.
I have no idea how the "Sharpen" function algorithm works and how it differs from USM. The reason why I'm using it is simply because I used Samuli's resize as an starting point for my own. It worked well, so I never bothered changing it.