larryepage Loc: North Texas area

There are ordinarily two parameters associated with capturing and saving JPEG image files.

The first specifies how many of the pixels from the image are retained. The choices are usually something like Basic, Standard, and Fine. Of those three, Fine saves the largest number of pixels, usually the entire number that are in the original image, or the maximum number that the sensor can provide. Standard captures fewer pixels...generally about half of them, or maybe half in each direction. Basic saves the smallest number of picels...sometimes as few as 10% of those captured by the sensor...essentially a thumbnail.

The other parameter controls how severely the file is compressed. The choices are usually Small, Medium, and Large. Large files are compressed the least...around 40%. Medium files are compressed more, and Small files are compressed the most...90% or even a little more.

The more the file is compressed, the less able software is able to accurately restore it. The more lossy it is. And if the file is stored with fewer pixels to start with, then the resolution is reduced, whether there is compression loss or not.

So if JPEGs are captured and stored as Small/Basic, there is resolution loss to begin with, followed by loss from the extreme compression. JPEGs captured and saved as Large/Fine will retain all of their initial resolution, and will lose little or nothing due to compression.

Choices made with camera setup or Save parameters will make a huge difference in the subsequent quality of inages.

Wallen Loc: Middle Earth

It depend on the compression ratio. If too low, then JPEG artifacts will happen even on the first get go. As for printers, it may be printing with the nozzles not calibrated to their sharpest position. low quality materials both ink and paper can mess things too. The person seeking answers says the raw files were satisfactory but loosing sharpness when printed. I do not believe this is a sensor nor a lens issue.

But you got me thinking with PP sharpening. Over sharpening images can actually do the opposite.

Wallen Loc: Middle Earth

Actually my comment was a general overview. Specifics will happen only if i know exactly what you are doing and what software you are using. Perhaps you can give a little more detail?

kymarto Loc: Portland OR and Milan Italy

Indeed. Since you can't print a RAW, or even view it for that matter, what are you using to convert and view the RAW vs the jpg?

jerryc41 Loc: Catskill Mts of NY

I find that JPEG need less work than raw, which always need processing, especially for sharpness.

Pistnbroke Loc: UK

The whole content of this thread does make one wonder why anyone should use raw. Certainly as far as Nikon is concerned the raw processor NX-D is built into the camera and you have sharpness/contrast/etc controls also built into the camera. All I ever do is crop and on occasions adjust brightness for consistency but you can get Perfectly Clear to do that for you

Linary Loc: UK

The image posted of the spider measures a mere 333 x 500 pixels. At 300 ppi that is 1.11" x 1.69"
If printed on a 4 x 6 card the resolution is down to 83 ppi.

If that is what the OP is doing, this surely explains the softness.

editorsteve

Jpeg compresses by taking a bunch of pixels (2×2 all the way up to 128x128 sometimes) and creating a pattern that approximates what the pixels contain. The pattern is mathematically described so it reduces file size but creates fuzziness. Upping the contrast increases the apparent sharpness but that just fools the eye.

Pistnbroke Loc: UK

Quote ….that just fools the eye.

Personally and I don't know if it applies to everyone but I use my eyes to view my pictures.

larryepage Loc: North Texas area

I think investigation will show that this isn't really what happens with JPEG compression. If you read the standard, the compression algorithm does work with blocks, but only up to 8x8 pixels. What it then does is generate a "word" (or words) of some length less than 64 words that represents not a combination of pixels, but rather the bit pattern that represents that block of pixels...the same way that Zip compresses text files. The big problem here is that folks think that JPEG files are images. They are not...no more than raw files are images. They are simply files which can be reassembled into images by a JPEG viewer. It just happens that JPEG viewers are very commonly available. The problem is that when you try to compress a file and represent a whole big bunch of bits with a smaller number of bits, uncertainty gets introduced into the decoding process. The greater the ratio between the original number of bits and the saved number of bits, the greater the uncertainty and therefore the greater the opportunity for inaccurate decoding. This happens when choosing the smallest file size.

There is a second variable associated with JPEGs. That is the number of pixels to save. One choice is to select 'highest quality,' which saves all of the pixels from the original image. This has absolutely nothing to do with compression. If a lower 'quality' is selected, then pixels have to be discarded. There are two basic ways to reduce the number of pixels. The first is to select, for instance, to keep every other pixel in each row and every other row of the image (or every third one, or every fourth one). That will obviously remove detail from the image. The second would be, as stated, to combine a group of pixels into a single pixel, perhaps representing the average of the group or perhaps done some other way. This will clearly result in a loss of sharpness and detail. I will be honest and tell you that I don't know how most software accomplishes this reduction. But what I do know is that it is the editor's choice, and it happens only if chosen to happen. It can also be completely avoided by choosing to save at the highest quality level.

Terminology varies around these choices. You have to read your own camera manual and understand what your software calls them, then make the best choice.

I am not advocating JPEG over raw here. But sometimes JPEG may be a requirement or may fulfill a real need. There are real differences, and there can be critical limitations, but with a little knowledge they can be managed.

TheShoe Loc: Lacey, WA

JPEGs are quite democratic in their dropping of data. That is determined by the conversion and compression algorithms without applying any need test to any datapoint.

htbrown Loc: San Francisco Bay Area

The jpg that comes from your camera is more than a 'translation' of the raw file. It has been post-processed in a number of ways. For example, when you set white balance on your camera, it doesn't affect the raw file, but it does inform how the camera adjusts the jpg. One of the adjustments your camera makes is sharpening, and this may explain the difference in what you're seeing.

Delderby Loc: Derby UK

Are you suggesting that if I adjust the sharpening in - camera, that this could then be cancelled out by the data being discarded?

larryepage Loc: North Texas area

I realize that you are asking TheShoe...but if you choose less compression and higher quality, you can cancel almost all of it out...

jamesl Loc: Pennsylvania

When you create the Jpg, check and make sure the Quality is set to 100 (or highest available quality). That is as close to uncompressed as you can get with a Jpg.