| Author | Thread |
|
|
10/24/2006 12:39:19 PM · #1 |
| I don't know what my problem is but I can't get this picture to 150 kb. I use levels, unsharp mask, rezise and sharpen again, but it's still about double of what it needs to be. What am I doing wrong? |
|
|
|
10/24/2006 12:41:30 PM · #2 |
well you can also use the jpg quality settings to change how large the image saves at.
if you are using photoshop, choose file>save for web
then you will see a quality slider, and slide that around until it is exactly under 150kb
|
|
|
|
10/24/2006 12:45:02 PM · #3 |
|
|
|
10/24/2006 01:15:29 PM · #4 |
Saving for web is a huge benefit to reducing file size and stripping out non-image data from the image. Adding extra passes of USM actually increases file size too btw, as there are more sharp-edged contrast changes in the image. Noise reduction can drop the numbers a bit when needed.
A couple other tricks involve going into selective color adjustments, in the black channel and increasing the black level while watching the changes on-screen, but watch going over 10% at a time - the more pure black or pure white, the smaller the file.
Try experimenting with that, going to save for web each time and optimize for a 150kb file and looking at the compression factor (%) - the higher the number, the less compressed it is, and thus reducing the introduction of compression artifacts.
When all else fails, put a huge white or black border on it. (just kidding) |
|
|
|
10/24/2006 01:30:08 PM · #5 |
I just prepared an extremely detailed, ultra-wide-angle shot for possible October Free Study, and in save for web the quality level is a stunningly low 48% at 150Kb target size. However, when you open the 2-up feature and compare the 100% with the 48%, there's essentially no discernible difference, not enough to worry about.
So don't worry :-)
R. |
|
|
|
10/24/2006 03:24:43 PM · #6 |
Originally posted by Bear_Music:
I just prepared an extremely detailed, ultra-wide-angle shot for possible October Free Study, and in save for web the quality level is a stunningly low 48% at 150Kb target size. However, when you open the 2-up feature and compare the 100% with the 48%, there's essentially no discernible difference, not enough to worry about.
So don't worry :-)
R. |
I'll believe it when I see it. From the time I started counting I have not entered 23 challenges cause I couldn't get the file size down to 150k without discernable loss of quality. I can see the difference between 80 & 100 percent. |
|
|
|
10/24/2006 03:44:37 PM · #7 |
I agree.
Some of this perception has a lot to do with monitor resolution too.
When I view an image at home, on a 1600x1200 screen resolution, the detail differences are few, as the image is smaller in relation to the screen size. When viewing on a 1024x768 monitor, a lot changes and compression artifacts and jaggies really start to show up.
I personally cringe if my save for web gets around 76% and go back and figure out a way to ease the details a bit. |
|
|
|
10/24/2006 04:42:43 PM · #8 |
While my normal tendency might be to examine at high magnification for signs of compression artifacts, the truth about challenge entries is that we've only got a few short seconds to wow the voter. The vast majority will not pick over an image looking for subtle artifacts.
I agree with Robert that even at 50% quality, there's really very little difference that will be perceived by the voters. Here is an old comparison set I put together looking at file size vs. image dimensions. The "qxx" at the end of the file name is the quality when saved for web with the stated target file size. This is a "torture test" image; notice that even at 640x640 and 200k file ize, the quality is only 53!. Some of these have *really* high compression, as low as quality=19, and do show visible artifacts, but not so much as to be greatly detrimental to the impact, assuming short-term viewing. I'm not sure I'd want to submit this one at 800x800 and 150k, but it still has more visual information by far than the 640k versions. Visual impact is significantly influenced by information content, so for most images (those with more normal detail levels) the increased size will affect voter perception far more than the quality differences.
My one regret with this set is that I did not include a reference image at each resolution with high (>80%) quality. |
|
|
|
10/24/2006 04:44:12 PM · #9 |
|
|
|
10/24/2006 05:08:39 PM · #10 |
Originally posted by kirbic:
...the truth about challenge entries is that we've only got a few short seconds to wow the voter.
Visual impact is significantly influenced by information content, so for most images (those with more normal detail levels) the increased size will affect voter perception far more than the quality differences.
|
That is based on the assumption that one is only posting for the submission-voting period. It is my understanding that the once the voting is over the image is permanently posted. As far as I know you can't replace it with a less compressed version. Personally I could care less how an image of mine is scored, but I do care what it looks like. |
|
|
|
10/24/2006 05:17:09 PM · #11 |
Originally posted by strangeghost:
One of the images is high compression, one is medium, and one is low. After you've carefully inspected each image and formed your own opinions, check your answers here. |
Excellent illustration, John. |
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 09/10/2025 01:48:29 PM EDT.
