Beautiful shots Brian. Exquisite detail, very nice composition and wonderful colours.
A flurry of questions for you (if you don't mind my asking via your thread) re. downsizing etc.
When I get my shots initially, I often have the more intricate detail of the compound eyes of the subjects however when I'm resizing for the web, in most cases the detail is lost. I understand naturally due to compression etc. I'll lose some detail but it seems so many others manage to retain really good sharp eye detail.
E.g in your Xanthogramma, Volucella and Helophilus portrait shots, you've retained finer detail.
What I'm doing essentially at the moment is taking my 5184x3456 Raw file, doing whatever needs doing to it in Lightroom. Then I'll export as a full res 16bit TIF where I'll make a few more minor amendments to it. Noise reduction, sharpening. I'll then save that as it is. At this point, I'll reduce the image down to 1296px wide (25%). I often find finer detail can be lost even at this point. I'll save as a level 12 quality Jpeg in CS6 and upload to my gallery. Of course then I'll link the 1024px wide image on forums for the benefit of those whom may not have super high resolution screens or are working in a desktop space smaller than say 1680x1050 for example.
Is there a specific formula to retaining detail when downsizing images? I would be very grateful if you could let me know your downsizing process from Tif (if you work via Tif) to Jpeg and if you work on any kind of per pixel / percentage downsizing ratios. Also, do you use Bicubic, Bilinear etc.?
Typically, I've just experimented with Bicubic and Bilinear and found that bilinear retained a lot more detail but I'm wondering whether that is the ideal method.
Lord VetinariRegistered UsersPosts: 15,901Major grins
edited June 12, 2013
Thanks for the comments Ian,
re downsizing, I don't do anything special just downsize a quality 12 jpg in PS using bicubic from full size to 180 DPI 9cm short side. As I've commented before I do resharpen the downsized pic.
I have seen and tried doing the recommended downsizing of a series of steps not more than about 20% of the picture size- this does preserve more detail but is just too time consuming for my processing. If I think I'm losing too much detail then I will show a 100% crop of the photo.
I'll give your methods a go and see how some of my images fare.
As a matter of interest. Here's two copies of the hoverfly I recently posted. Linked as opposed to embedded.
This first one is a TIF downsized to 1296px wide using Bicubic, saved as JPEG, uploaded to my gallery and then the 1024px wide image from the gallery linked (as it's the 1024px ones I always show for those with lower resolution screens
And this one is the exact same shot resized and saved in the same way with the exception of using the Bilinear option when downsizing. As you can see, there's a noticeable difference in the eyes with almost no difference to the background softness.
Comments
A flurry of questions for you (if you don't mind my asking via your thread) re. downsizing etc.
When I get my shots initially, I often have the more intricate detail of the compound eyes of the subjects however when I'm resizing for the web, in most cases the detail is lost. I understand naturally due to compression etc. I'll lose some detail but it seems so many others manage to retain really good sharp eye detail.
E.g in your Xanthogramma, Volucella and Helophilus portrait shots, you've retained finer detail.
What I'm doing essentially at the moment is taking my 5184x3456 Raw file, doing whatever needs doing to it in Lightroom. Then I'll export as a full res 16bit TIF where I'll make a few more minor amendments to it. Noise reduction, sharpening. I'll then save that as it is. At this point, I'll reduce the image down to 1296px wide (25%). I often find finer detail can be lost even at this point. I'll save as a level 12 quality Jpeg in CS6 and upload to my gallery. Of course then I'll link the 1024px wide image on forums for the benefit of those whom may not have super high resolution screens or are working in a desktop space smaller than say 1680x1050 for example.
Is there a specific formula to retaining detail when downsizing images? I would be very grateful if you could let me know your downsizing process from Tif (if you work via Tif) to Jpeg and if you work on any kind of per pixel / percentage downsizing ratios. Also, do you use Bicubic, Bilinear etc.?
Typically, I've just experimented with Bicubic and Bilinear and found that bilinear retained a lot more detail but I'm wondering whether that is the ideal method.
Thanks for any advice you could give to me.
My Smugmug gallery
re downsizing, I don't do anything special just downsize a quality 12 jpg in PS using bicubic from full size to 180 DPI 9cm short side. As I've commented before I do resharpen the downsized pic.
I have seen and tried doing the recommended downsizing of a series of steps not more than about 20% of the picture size- this does preserve more detail but is just too time consuming for my processing. If I think I'm losing too much detail then I will show a 100% crop of the photo.
Brian V.
http://www.flickr.com/photos/lordv/
http://www.lordv.smugmug.com/
I'll give your methods a go and see how some of my images fare.
As a matter of interest. Here's two copies of the hoverfly I recently posted. Linked as opposed to embedded.
This first one is a TIF downsized to 1296px wide using Bicubic, saved as JPEG, uploaded to my gallery and then the 1024px wide image from the gallery linked (as it's the 1024px ones I always show for those with lower resolution screens
http://ipc.smugmug.com/Close-Ups-Macro/Macro-7D-Canon-100L/i-zZfC4Hf/0/XL/IMG_1899-XL.jpg
And this one is the exact same shot resized and saved in the same way with the exception of using the Bilinear option when downsizing. As you can see, there's a noticeable difference in the eyes with almost no difference to the background softness.
http://ipc.smugmug.com/photos/i-fRSHL5g/0/XL/i-fRSHL5g-XL.jpg
My Smugmug gallery