Please review my article for accuracy!
I just wrote an article about using the various exposure modes in the D100 and other cameras. It is not yet complete, since I have not added images to affirm my assertions, but will be soon. Before I start using the article, or donate it to Nikonians.org, would you please read it and see if you note any inaccuracies. Thanks!
See the raw text at:
#1. "RE: Please review my article for accuracy!" | In response to Reply # 0vfnewman Basic MemberWed 28-Aug-02 11:45 AM
Your article is pretty good, but there are a few points that stood out to me:
Well, this varies according to what type of photographer you are. If you are only interested in taking snapshots, and will never try to do anything with the images later, except maybe reprint them on your color printer, or take them to a processor for prints, then the JPEG mode is fine.
I don't know of a single sports shooter or PJ that shoots in anything other than JPG. JPG has limitations, but for plenty of professionals, the benefits (file size, speed of file writing, and lack of need for post-processing) outweigh the liabilities. I wouldn't make JPG sound like a snapshooter-only format.
JPEG converts the image from 12 to 8-bits and then compresses the image, permanently throwing away large quantities of image data in the process.
TIFF goes from 12 to 8 bits too. I also wouldn't say it throws away "large quantities of image data". There is the issue of "significant bits". This is discsussed better than I can do it here: http://www.scantips.com/basics14.html
I know this discusses scanners, but the underlying principal is the same. The camera has a pretty good idea which bits is should keep and toss. Not perfect, mind you, but not bad.
You can later uncompress the JPEG image by letting the computer add data back, but it does this by "interpolating" the data, which means it makes an educated guess about where and what data to add back.
Please explain your basis for saying the computer adds data or interpolates a JPG upon opening. I don't think it does.
When a JPEG image is made generally only about 25% of the original image data is maintained.
Again, please explain the basis of this statement.
I don't think your discussion makes an adequate distinction between bit depth and data compression. IMO, you should treat the two more clearly. The above statement is a good example. Besides the fact that I don't agree with the 25% number, you don't say whether the loss is in shadow/highlight detail (bit depth) or in resolution/detail (JPG-induced data loss) or some combination of both.
JPEG mode can be used and modified to a degree, but your image will degrade VERY quickly upon manipulation.
I disagree with this. Done properly, there is little degradation.
Surely, within the next year or two, a standard will emerge, and this problem will go away.
I'd be shocked if that happens, but that's strictly my opinion and I am only adding this as an editorial comment. You accuracy is not in question.
You should also mention that Nikon Raw image formats require the use of special software to open them. Until recently, if you wanted to open a NEF, you had to pay a ridiculous amount of money to get Nikon Capture. Now Nikon View, available free, can open and manipulate NEF files, but it's a little limited in capabilites and won't do batch processing, which Capture does for (I think) $150. Bibble and Quimage are less expensive options from third-party developers. But as of right now, QImage (my favorite) does not decode NEF files from the D100, only the D1/X/H.
Last, it is important to remember that as of right now, I'm not aware of any monitor or printer that works in more than 8 bits. You can open a NEF, but you'll only see it in 8 bits. You can convert it to a high-bit TIFF, but again you'll only SEE 8 bits of it. You DO have the luxury of choosing the 8 bits you want upon converting the NEF to anohter format. The same applies for printers. You CAN choose the most important bits (which mainly means shadow detail), but we are still in an 8-bit world. That, of course may well change.
#3. "RE: Please review my article for accuracy!" | In response to Reply # 1Wed 28-Aug-02 02:52 PM
Thanks for the excellent points you brought up about my unfinished article. I will indeed print this out, research the particular points you mention, and either modify or reinforce my article.
This is great! Where else in the world could I go to find expertise, and people willing to help with things like this, except here at Nikonians?
As a new writer, I am trying to learn the writing process, and write accurate articles in the process. I see the need for very careful research before a particular statement is made. I am attempting to do this, but at the same time, I find the varied opinions of people in our group are extremely valuable for the writing process. I simply can't get away with many opinion statements without getting called out on it by someone, such as yourself, with an equally strong opinion in the opposite direction.
I personally do not like JPEG, since I am wary about throwing away any data. I have found that my own JPEG images with my D100 are NOT as sharp, or as sharpenable as the RAW conversions to TIFF are. It seems that the JPEG mode of the D100 is throwing away enough data that I cannot sharpen the image as well in Photoshop 6.0. The guys at DPReview.com (see quotes below) agree with me on this in their very detailed D100 review. In fact, I have cut out a couple of direct quotes from this review (backed up by actual images) where they are discussing this very issue:
"There is a noticeable difference in sharpness between JPEG Fine and RAW, the PC based RAW conversion software has much more processing power available and does appear to apply slightly more sharpening to the image before output (default settings were used to convert the RAW image). Getting back to in-camera formats the JPEG Normal setting doesn't seem to introduce too much in the way of visible artifacts and would seem to be more than acceptable for everyday shooting, especially if you have a limited capacity storage card. Smaller output size images are of course very sharp and detailed, reducing 6 million pixels down to 3.3 (2240 x 1488) does produce a very sharp and smooth image."
"It's fairly clear to see from the side-by-side crops from our standard resolution chart that shooting RAW and converting to JPEG later will deliver a sharper image with more visible detail. Despite having both the camera JPEG and RAW convert on the same sharpening settings the RAW converted file is obviously sharper."
They do admit that JPEG is "more than acceptable for everyday shooting," but I hope that I am doing more than making snapshots. I have been discussing this matter with an associate today, and he has asked that, instead of conjecture, that I should make some actual images that tests my type of shooting (Nature Photography, Landscapes, etc.), and how JPEG might apply. I will do so, and get back to this thread with the results.
Anyone else out there have any strong opinions about this subject? Thanks in advance!
#2. "RE: Please review my article for accuracy!" | In response to Reply # 0
As noted above this seems to downplay the value of JPEG quite a bit. I don't know if I'd make some of the statements you did. There are many famous pro wildlife shooters that only use the JPEG mode.
You might want to post this over at:
and see what kind of response you get. This is a BB run by Moose Peterson and David Cardinal. It is fairly specific to shooting issues related to Nikon Digital SLRs. You might get lot's of feedback from people who are using these camera every day.
#4. "RE: Please review my article for accuracy!" | In response to Reply # 2Wed 28-Aug-02 02:58 PM
Tell me please, what particular statements do you object to? Is it primarily the statements about the JPEG issue, or others?
I do know that Moose Peterson and others always use JPEG mode, and I understand why. I, too, like the high speed writes and higher capacity on my microdrives that JPEG gives me. Even Moose Peterson admits in his D1x review that he does not need "large files." That may well be the point of disagreement here. I DO need large files, the bigger the better, which is why I look down on JPEG. Maybe I should not be so adamant about it though.
Anyway, I just shot 177 images at a face painting session and at home found it much more difficult to get high quality results out of the JPEG images at 8" x 12" 300dpi printing resolutions. In other words, if I took a RAW image and compressed it to a JPEG with Photoshop, it looked great and was easy to sharpen. But if I took a camera prepared JPEG and tried to use it for sharpening purposes it did not perform nearly as well. I could not get the razor edge look in their eyes and hair that I so love.
I enlarged the images to an equivalency of and 8 x 12 size print at 300 DPI, and found a MUCH "grainier" look to the image.
Has anyone else had this experience?
#7. "RE: Please review my article for accuracy!" | In response to Reply # 4sjn Basic MemberThu 29-Aug-02 01:35 AM
>Tell me please, what particular statements do you object to?
>Is it primarily the statements about the JPEG issue, or
I was primarily objecting to the JPEG statements. It was starting to sound like some of the Canon ads a few months back! It seemed to definitely say JPEG is only good for snapshots and worthless for anything else. While I agree with you about many of the issues about JPEG throwing away data, there are many non-snapshot applications where it works perfectly - as is evidenced by how many pros use it for other than snapshots.
I think I would feel better about the article was giving JPEG a fair hearing. Talk about the cases where it works really well (not just snapshots) - this approach is more of a "here's all the facts, draw your own conclusions based on what you need the photo for" might gain more respect from readers.
The reason I suggested posting a copy to NikonDigital is that there are many advanced amatuers and pros using JPEG on a dailing basis. They may be helpful to you in exploring some of these cases where JPEG really is acceptable. David Cardninal over there is also the author of DigitalPro (image mgmt/editor tailored for Nikon digital SLRs) and may be willing to give you some more technical feedback on the data loss due to compression.
Hope that helps!
#10. "RE: Please review my article for accuracy!" | In response to Reply # 4
#5. "RE: Please review my article for accuracy!" | In response to Reply # 0
I enjoyed reading your article. I too have a D100 and shoot mostly JPEG, mostly because I only have one 256 meg. card and would only get 25 raw pictures. I have been thinking about getting a IBM microdrive which would give me room for over 100 raw pictures. People have commented about heat build up in the camera, durability issues and a 8 time write speed making buffer capacity an issue on action shots in raw mode. My point is one other suggestion to highlight in your article might be a sentance or too on memory card size, camera buffer speed in different modes ie jpeg, raw and tiff and lastly pros and cons of different storage media.
#6. "RE: Please review my article for accuracy!" | In response to Reply # 0
Check out these webpages for some VERY interesting JPEG facts. My contention is that JPEG is permanently throwing away large portions of your image data, and that manipulating (re-editing and saving) an existing JPEG degrades the image even furthur. Everything I am reading on the net, so far, seems to back up this belief. My own experience has shown this to me with my JPEG images.
#8. "RE: Please review my article for accuracy!" | In response to Reply # 6vfnewman Basic MemberThu 29-Aug-02 09:55 AM
Let's take a couple steps back here. This is getting confusing.
1. After looking at your article again, I think it looks more like a treatise that turns into an editoral. You start talking about different file options and end up with quite an anti-jpg slant. You should focus on it being one or the other.
2. This thread is straying from the origial request of a critique of your article. Do we still want to talk about the article, which has a great deal of good potential, by the way, or why you think jpg is bad?
As far as the links in response #6 I don't see anything shocking. JPG is a lossy compression scheme, no doubt about it. And manipulating and re-saving a jpg will cause more loss. That's why the best way to handle jpgs is to open the original and save it as a tiff or psd and THEN work on it, resaving as you go in a lossless format.
By the way, what exactly is your workflow and what software are you using for adjustments and printing? For the jpgs you were not happy with, how did you perform the sharpening. Photoshop's USM is not the ideal route. What are you using for a printer? I have shot plenty of jpgs on my D100, which with a little sharping in PS, and printed on an EPSON 1270 through QImage are absolutely gorgeous and sharp. You ARE shooting your jpgs in the Fine mode, right?
#12. "RE: Please review my article for accuracy!" | In response to Reply # 8Thu 29-Aug-02 03:50 PM
Indeed I have been using Photoshop 6.0's Unsharp Mask for sharpening my JPEGS.
I welcome any pointers to a better method. Are you using something other than Photoshop, or something within Photoshop that I am not aware of. Thanks!
#15. "RE: Please review my article for accuracy!" | In response to Reply # 12
#9. "RE: Please review my article for accuracy!" | In response to Reply # 6
I'd be very interested to see if you follow Scott's suggestion to post your article to nikondigital.org. I think those guys might give you a real run for your money! If you could hold your own over there, you'll get by here. I get the feeling that the D100 is your first digital? I'd be disapointed if you decided the JPEG format was "snapshot" grade based on the D100 experience alone. The Dpreview comments you quote are in fact re the D100; not the JPEG format. Come back to us w/ the same quotes for D1H, D1X, EOS 1d or D60??? Please be open minded about this - for your own sake and reader's as well... (I'll now go look at the sites you posted - to remain open minded myself).
I see that you have revisited the 4stops statement over in the other thread. Hope you saw Alan''s comments on standardising RAW format - very interesting (and completely logical).
Lastly, I am looking forward to hearing back from you re FOV crop (per offline discussions).
We all look forward to your final article - wishing you all the best.
#13. "RE: Please review my article for accuracy!" | In response to Reply # 9Thu 29-Aug-02 04:00 PM
Yes, I did speak too soon (as usual) about the EV range of the D100. I have now determined in my own testing that is it almost as capable as negative film in holding a range of light values. It held over five stops of value in my test photos. So, it is just a bit better than transparency film, with the caveat that any very bright sources of light will lose detail more quickly than with any film I've used so far. I still preach the opinion, "Expose for the highlights, adjust for the shadows!"
I, too, am intrigued with the 1.5 FOV crop vs. Image magnification issue, and will be addressing that very soon in my D100 review. I am shooting some side-by-side images with both my F5 and my D100 at the same focal length, with the same lens, and we will examine the images soon.
Thanks for the cool suggestion!
#11. "RE: Please review my article for accuracy!" | In response to Reply # 0
I could have survived this better if I had been insulting everyone's grandmother or something.
I have seen the light! This is indeed my first digital camera, other than a cheap Kodak, and my wife's CoolPix 990. I have been primarily a 35mm shooter with my beloved Nikon F5, until now!
In deference to those of you more expert than myself, I have stepped back and reexamined my article from the standpoint of the 10,000 e-mails I received immediately upon starting this thread. I rewrote it with much more balance than before, and giving a much better light to JPEG and TIFF modes.
I still am not a died in the wool native mode JPEG shooter, because I have seen how much nicer my RAW image conversions from Photoshop 6.0 look in comparison to camera generated JPEGS. Plus, I own a big pile of IBM one-gigabyte microdrives. And, I just love post-processing images.
I think that there's a big difference in a professionals and an advanced amateur's look at this subject. A PRO has a real workflow, and has to get the job done RIGHT NOW, with maximum quality in the allotted time. I agree that JPEG is most capable in that respect, and makes a very nice image.
Myself, being merely a lover of photography, and having the luxury of a more leisurely workflow, really would prefer to lovingly handle my own images in my own way. I do not want ANY of my original images tampered with, any more than I want to find little bugs eating my Provia transparencies. RAW mode is the closest thing to a reusable piece of film that I can get in digital. If I were a Moose Peterson, or any one of you professionals who are actually making a living with your cameras, I suppose I would seriously reconsider JPEG as a primary shooting mode.
As it is, I am merely a highly opinionated amateur photographer/writer, trying to accumulate enough digital images to sell some stock photos. I love this D100. It has opened up new aspects of photography to me. I am taking hundreds of images per week with this new tool. Maybe someday soon, I'll be a PRO also.
I am so happy with the camera that I am trying to write articles about it and digital photography in general. Am I yet an expert... well... probably not, but I do have some really good research material in the Internet, and some really intelligent Nikonians to keep me in line.
What have we accomplished in this thread other than filling up my e-mail box with letter bombs and stern lectures? I think we have accomplished a lot! I now have a more palatable and balanced article, with still a good dollop of opinion, but more usuable information. And, we have all enjoyed each other's company in discussing a controversial issue.
If you still see issues with my article, I would welcome your comment. When this article is finally done, with a bunch of pictures, hopefully it will be of some benefit to mankind.
May we ALL keep learning together!
#14. "RE: Please review my article for accuracy!" | In response to Reply # 11slow_driver Basic MemberFri 30-Aug-02 12:22 PM
Great to see you approaching this with a good attitude. In fact, there are not many who would so openly mark the article as "open for review" before it is actually published. So, good luck in your work; I had better go read the updated article...