This is my first message in this list, I've been reading you for a while and now I decided to write. First of all, excuse me for my english, wich is sometimes far from being "readable". What Gary Lawton says about JPG versus TIFF is completely correct. Every image saved in JPG format loses information about its detail, even it's saved at the maximum quality. Wich makes JPG a good format is indeed its capabilities to reduce de file size while mantaining good details in the image. The compression algorythms that JPG uses on saving the image are specially minded to make very difficult to our eyes to see the imperfections. So, what's wrong with JPG? Almost nothing. Almost! for example, JPG is not suitable for: . Making huge copies of the photo, cause one then can easily take a very closer look and see the imperfections of the image. The imperfections that allowed JPG to reduce the image size. . If you plan to make intensive digital work with the image, soon or later you'll find the JPG imperfections amplified (the so-called "artifacts"). TIFF, in the other hand, doesn't throughs any bit of information out of the original image. In fact, TIFF also supports a kind of compression (LZW), but it's a "non-degrading" compression, like the one we use in ZIP files. (JPEG compression can be compared, this way, with the MP3 compression, or the video compression in DVD movies, wich are, in fact, different generations of the same original digital compression techniques) About what Howard says, I understand he refers to practical facts: At the end, who cares about that tiny (almost invisible) differences? I understand him. But the fact is that, as I said above, to catch the differences between an 8 bit/channel image and a 16 bit/channel image, you'll probably have to print it huge enough so you can examine the color differences. Even so, it's almost impossible to see. Probably, in that case, you'll found little imperfections in the color gradients, in form of color bands (in the 8 bit/channel image, of course). And even all of this, you'll probably wont find any huge printing system capable of reproducing 281.474.976.710.656 different color tonalities (the theoretical number of different colors you can get with a 16bit per channel image. The fact is that the real application for 16bit/channel images (except for some rare applications), is for, as I said above, digital retouching. With 16bit/channel the digital retouching algorythms has a lot more information to work with, and specially when you are working on black and white images, 16bit/channel will be very useful on conversions, and the feared "histogram gaps" will be very rare. I can talk more about histogram gaps and 16bit/channel treatment on black and white images, if you find it interesting. I wrote an article specifically about this, wich is online here: http://www.fotopunto.com/?a=articles&aa=view&article_id=20 I'm sorry, the article is written in spanish. Greets -----Mensaje original----- De: owner-photoforum@xxxxxxxxxxxxxxxxxxxxxx [mailto:owner-photoforum@xxxxxxxxxxxxxxxxxxxxxx]En nombre de Howard Enviado el: miércoles, 15 de septiembre de 2004 17:54 Para: List for Photo/Imaging Educators - Professionals - Students Asunto: That old jpg v. RAW argument again... Gary Lawton wrote: > A JPG file is an 8bit compressed TIFF file. However, JPG compression is > "lossy" meaning that data is thrown away during the compression process. > > The conversion back to TIFF will not recover data that was lost during > the compression. Nor can it recover the 8bits of data that were > discarded from the 16bit sensor reading. > > It'll work, it just wont give you the best quality > > Gary > > Peeter Vissak wrote: > >> I don't have RAW in my pretty simple camera, but I've compared finest >> JPG to TIFF that I've shot in a row and these shots have been identical. >> Well, to my eye. >> So I shoot JPG and batch convert (IrfanView!!!) to TIFF immediately and >> work in TIFF later on. >> Saves a lot of processing/writing time (this small thing is so uglily >> slow!) and of course memory. >> >> Peeter >> >> Three points... 1 Can anyone actually see the difference in the final image between TIFF and JPG at maximum quality (and not repeatedly saved and recompressed)?? 2 Can you actually see the difference between an 16 bit TIFF and an 8 bit TIFF?? 3 And in any case, if the same image were shot simultaneously on Canon 20D, Nikon D100, Pentax *st, Sigma 10D, Kodak etc., in RAW format and converted to TIFF without any manipulation, I'll bet that these would almost definitely show distinct differences when printed.... Just as using different films gave differing results under as identical a set of conditions as possible! So what DO we all mean by the "best quality"?? Howard
