Wanderer2 Loc: Colorado Rocky Mountains

<<If the files are identical bit for bit, then, well, uh, the problem is in your imagination.>>

No way is it my imagination! The differences in sharpness are dramatic, visible in the image on my small laptop screen and really visible on my larger monitor I use for processing. It's evident even without any image magnification and with a small amount of magnification it's severe in degree. I've looked at many files that have been uploaded and everyone shows this change in sharpness. There's no way an acceptable print of any size could be made from one of these uploaded files. Thanks for taking the time to respond, I appreciate it.

Wanderer2 Loc: Colorado Rocky Mountains

Thanks, I haven't tried that. Will do.

Longshadow Loc: Audubon, PA, United States

It has to do with how they display what they have stored.

TriX Loc: Raleigh, NC

Download the file (rather than viewing online) and examine the file properties. Are they the same (file type and size) as the file you uploaded? Easy to test...

Hamltnblue Loc: Springfield PA

I'm sure the files are compressed during transmission and storage. Maybe some data is being lost when compressing already compressed files

rgrenaderphoto Loc: Hollywood, CA

Your internet connection is the issue, the low speed is introducing transmission errors. Load the Speedtest app onto your Smartphone and see if the Upload/Download speeds are what you are paying for. In this instance, the upload speed is the governing factor.

I would suggest a local RAID solution for file storage until faster bandwidth is available. Any otehr Satelite provider will have the same issues.

bleirer

-

JohnSwanda Loc: San Francisco

It seems very unlikely to me that transmission errors would just consistently reduce sharpness in files and have no other effect. More likely, it would be corrupted and would not open.

Longshadow Loc: Audubon, PA, United States

Checksum wouldn't match.

TriX Loc: Raleigh, NC

it’s EXTREMELY unlikely that it’s the internet connection. The speed has nothing to do with it. The connection is TCP over IPv4 or IPv6. Packets of data are either received intact and acknowledged or resent. There is both a checksum at the IP layer and the TCP layer. The IP link layer encapsulates IP packets in frames with a CRC footer that detects most errors, and typically the end-to-end TCP layer checksum detects most other errors. The net-net is that a poor internet connection may be slow because of resent packets or latency, but the data, when received, is correct.

JimH123 Loc: Morgan Hill, CA

I still am confused a bit about what you are saying. Are you directly viewing the image while it is still on the cloud? Or are you downloading the file and comparing the original to the downloaded file? This is what you should be doing. If you are looking at it while still in the cloud, they may have some way of downsizing the preview that you might be seeing on line. I suspect this since they need to conserve space in the cloud and may use some message effectively Zip groups of files together so that there is no wasted space within the clusters of the clouds storage space. And perhaps they can only show a crude preview from this Zipped like group of files. But the download should return it as it was.

Wanderer2 Loc: Colorado Rocky Mountains

The possibility, as described above, that the sharpness of the image in storage in the cloud may be decreased but when downloaded will be the same as the original file before being uploaded, has been mentioned by two other posters also. I had been looking at the files stored in the the cloud, Amazon Photos, and seeing the difference in sharpness there. Before posting this query I had not thought of looking at the files after downloading from the cloud, as suggested.

I've now done that, downloading the test file that I uploaded earlier today as described in my original post, and the sharpness of that file is the same as the original file before being uploaded to Amazon Photos, as one poster stated was his direct experience with Amazon Photos. The dimensions and MBs appear to be the same in the original file, the file stored in the cloud, and the file downloaded from the cloud. So apparently the degredation in sharpness seen in the images on Amazon Photos is due in some way to how they are treating the uploaded files, even though they don't seem to be compressed. Thus, I have been concerned needlessly since what matters is the sharpness of a file after I have downloaded it from the cloud, not how sharp it is in cloud storage.

Thanks to all for your help. If anyone has any further thoughts as to what Amazon Photos is doing to the uploaded files it would be interesting to know. At least now I don't have to be concerned about changing internet providers!

TriX Loc: Raleigh, NC

Glad you resolved the issue. It’s likely that the file has not been changed in the cloud, but rather, how the file is displayed when you view it online, which is probably a lower resolution to conserve internet BW and decrease the lime to load the image.

Cheers

JimH123 Loc: Morgan Hill, CA

This is good to know. They don't actually need to compress a file to save some space. A hard drive stores data into a block of space called a cluster. The size of a cluster is dependent upon the operating system requirements, and the size of the drive. For instance, a 32G SD card has a cluster size of 16K bytes and a large spinning hard drive is going to have something else. As a file is saved, it fills many clusters until all the bytes of a file are saved. Now lets suppose that the last cluster saved has just one byte. Then that entire cluster is marked as used and the remaining space of that cluster cannot be used by anything else. Now everyone is used to ZIPPING files to compress them and to make them fit in a smaller space. One of the ways this works is to two or more files, join them together so that they look like a bigger file and where that first file ended and didn't fill a cluster, the next file starts and completes that cluster. And with lots of files in a group, this happens many times.

But the files can also be compressed too. And this is a lossless compression. I have no information on how effective this compression is. I know ZIPPED RAW files may not save much space, but that's not to say they don't have some other compression ability that does even better.

Anyway, these saved, compressed files are not immediately accessible. It is likely that they have some means to scan the bigger compressed file and to extract enough of the details of an image that you can identify what you are looking for. And then when you ask for it to be returned to you, that they go to the extra effort to pull it out in its entirety. Thus the problem you were having.

Now don't confuse this method of storing files with what Facebook does. Facebook does not archive files. They make files available for viewing. But they also know that online viewing does not need gigantic files to let others see what you are posting. Thus, they feel no shame in downsizing that enormous file that you upload and to save it into something more suitable for online viewing.

TriX Loc: Raleigh, NC

Off the OP’s subject. But HD cluster size is always a compromise. Ideally, you align your writes with the cluster size to minimize wasted space, but if not feasible, smaller clusters potentially have less wasted space, but result in lower performance for large files such as images.