I have the “Size” column shown in dophin but it only seems to show file sizes in fractions of KiB’s, MiB and so on. For example a file might have size 921.4 KiB. I’d prefer to see the exact size in bytes, ex: 943549. I now workaround that to get the exact size by right clicking the file, then using “Properties” to see the exact size like it might show for the 921.4 KiB file as
Size: 921.4 KiB (943,549)
I suspect they might have done the limitation on the “Size” column to not be exact bytes because a big file would need a very large number string but for an integer bytes value the “KiB” label wouldn’t be needed. So you’d save about 4 spaces (the space and three letters of the unit label) by going exact.
For example my zoom rpm file shows in Properties:
197.5 MiB (207,089,528)
so my hypothetical “Size” column string would be 207089528 (opting to drop the commas). So the “197.5 MiB” is still using nine spaces for its string. So both are 9 spaces for the string. So I suspect you only start using more space in the Size column for files above 1GB, which are rather rare. So it seems to be there’s little reason not to have exact file sizes in the Dolphin “Size” column.
So is the exact size just not possible and this would have to be a change request?
Hi - the ability to set some other lower level of rounding, more precise than the default, looks to be the topic of this feature request in the KDE Bugtracking System: 498820 – Increase file size precision in view
This looks like something that will just require someone with the time, ability and interest to jump in and figure out how to implement 
In the meantime, you can also add yourself to the CC list for that issue in Bugzilla to get email notifications whenever new comments are added, or the status is changed. Making sure the “Add me to CC list” option is checked, then choosing “Save Changes”, will add you.
My purely personal preference for the current method of displaying sizes isn’t based on the visual length of the string, but based on how I think about file sizes. I typically want to get a sense of scale first - basically, what order of magnitude is the file’s size? (bytes, kilobytes, megabytes, gigabytes) Then, if I need more precision, it’s typically at the level of “120 megabytes vs. 80 megabytes”. I understand there are folks with other use cases, but I personally don’t usually need to compare the sizes of kilobyte+ files down to the individual bytes.
Just sharing that for what it’s worth - not much, since I’m not a developer or pro-level user or anything, but maybe helpful context for someone!