Fredrick Loc: Former NYC, now San Francisco Bay Area

So, I'm attaching two images. I just edited the second image slightly from the first one (got rid of some twigs in the lower right hand corner). However, when I hover over the first jpeg image, it says dimensions are 6000 X 4000, and file size is 3.52MB. When I hover over the second jpeg image, it also says dimensions are 6000 X 4000, but now says the file size is 11.3MB!

I took a Fuji RAF (raw) file, and edited it in Luminar 4 (beta version) to create the first jpeg image. I then took that image into PSE 14 and did the slight edit to create the second jpeg image. That image grew from 3.52MB to 11.3MB! I don't even understand how a 6000 X 4000 file (from an APS-C crop sensor) could only be 3.52MB in the first place? I guess there's a lot I don't understand about file sizes, that I thought I did.

Any help would be greatly appreciated! Thanks in advance.

(Download)


(Download)

frankraney Loc: Clovis, Ca.

Both images are 6000 x 4000 pixels. This is 24 megapixels, or the size of the image the camera recorded........in pixels. The raw and jpeg from camera will both be this size.

The second sizes you mention are file size as stored on disk......this will be different for file types and amount of data in the file..typically the raw file size will be about twice the jpeg on disk, which is in megabytes, this is different......but they both are still still the same size photo.

Pixel size is photo size.......
Byte size is disk storage size.
They are different....
Also depending on the editing program and how it is edited will change the file storage size in bytes.

Hope this helps

cahale Loc: San Angelo, TX

Open both in your version of Elements. Look at Camera data in "File Info." The original is 1920 X 1280 @ 240p. The edited version is 6000 X 4000 @ 240p. There's the file size increase.

CHG_CANON Loc: the Windy City

Hopefully, all your questions should be addressed by this post: Recommended resizing parameters for digital images

Longshadow Loc: Audubon, PA, United States

The size of the base images are the same (6000x4000).
All you did was change the "content" of the image.
The editor stored the information (which is different than the original) it needs to render the image,
that's why the file sizes are different.

Every time one "edits" an image, the resultant file size will be different because of different image data for the edit.

Not sure why it would really matter.

jerryc41 Loc: Catskill Mts of NY

As far as I'm concerned, numbers and photography don't mix. Too confusing! I can do math, or I can take pictures. I take pictures.

Longshadow Loc: Audubon, PA, United States

I don't worry about them either!

GoofyNewfie Loc: Kansas City

Numbers aside, I really like the image!

frankraney Loc: Clovis, Ca.

John_F Loc: Minneapolis, MN

File size in bytes is pixel width x pixel depth x pixel bit depth / 8 . In Photoshop (Elements too) each RGB color channel takes on 256 values (this is 1 byte). So each pixel would have 24 bit depth. So a 6000 x 4000 image should yield a 72 MB file. Your 11. MB file is after a 7:1 compression.

Fredrick Loc: Former NYC, now San Francisco Bay Area

Thank you ALL for your feedback! It was very helpful and I really appreciate it. Love this forum.

As an aside, the original sky in that image was a dull gray. I was able to totally replace the sky with literally “1 click” in the beta version of Luminar 4. Look how it filled in the new sky perfectly between the branches. Think about how long and laborious that would take to accomplish in other editing programs. It had about 15 sky replacements in its new AI Sky Replacement filter. Each Sky Replacement would also slightly change the exposure/colors in the rest of the image to complement the new sky. It has other AI filters as well that are amazing and greatly speed up the editing process. I think AI filters have a great future in editing programs going forward.

Luminar 4 ships November 18, I believe. You may want to download a 30 day free trial.

RichieC Loc: Adirondacks

Well, there is no quick way of answering this without your eys glazing over... An electronic image is machine code- ultimately 0's and 1's. each pixel is kept track of... and assigned a number. It only looks like an image when your computer accesses the numbers and processes them inot the glowing phosphors on your screen ( there are different technologies - but you get the idea).

JPEG is a mathematical compression algorithm that groups similarly coded pixels and averages them by actually changing them to the very same number, then writes a shortcut to map these wherever each pixel appears in the image. This makes the code smaller. The higher the compression setting you choose, the more pixels in a wider selection of similar colors are gathered and given the same number. Less code, but the image at some point will become noticeably graphic.

JPEG does an excellent job, as you can't see all the slightly different colored pixels most of the time.. but every time you save the image, the algorithm is re-applied to an image it already compressed/averaged before. This is all a one way street, the image degrades a little, whether you can see it or not - this is not a knock or an opinion or arguable- it is a simple fact with the way the JPEG algorithm works. .

The problem is, nuances in shadows or highlights can be lost and not available to use to coax out hidden details from. I could get into 8 bit vs 12 or 14 bit and computers run on 16 bit, but it all has to do with what your camera captured in each pixel and the bigger the number the more accurate the color.

Open a 14 bit raw file, process it, and save it in a lossless format like tiff, or PSD ( Photoshop) or native RAW ... whatever and save to compare... , and then save the same image as an 8 bit jpeg ( which is a "lossfull" format- detail is lost on every successive save- not that you can necessarily, at normal viewing distance, see this). Comparing the two, you will look at what appears to be the very same image, but the file sizes will be way different and the actual numerical data/detail- depending on how well the image was captured in the first place, has much more info in it- you very well may not be able to see this with your naked eye!

Basically, while the image size in pixels are the same, the files size differences in your case most likely points to a more complicated image, more numbers to keep track of= bigger files size.

Fredrick Loc: Former NYC, now San Francisco Bay Area

Thanks! It makes sense to me now that different editing programs would have different compression ratios. I was using a beta version of one program in addition to another program. Never thought about that.

Fredrick Loc: Former NYC, now San Francisco Bay Area

Thanks for that explanation!

amfoto1 Loc: San Jose, Calif. USA

If you are willing to do an edit like that, the image would also be nicer if the beach ball nets, people and houses were cloned out! Those are all "minor background" items, so I wouldn't be concerned about it "ethically". It's not like a journalistic image, where there's little to no tolerance for even small "corrections" to an image.