where there’s smoke there’s forum fire
TaoTeJared said:
. . . As for the compression on Raw I have always set it to 14-bit (more color info) with AdobeRGB and Lossless compressed. Many I have talked to, say they use the same settings. There must be a difference between all three but I have no clue. It seems to be the middle ground and I have never been disappointed with using that setting.
Thanks for the info, Tao!
@DaveyJ: +1 Studio.
No need for apologies for your arguments - heck, they even have me thinking again. I can't agrue jpg wouldn't be more convenient.
@studio460 - no problem. On a side note, you may want to look up or test what shoots faster if you are really trying to capture high speed stuff. Jpeg I believe is the fastest. :\ It sounds like you are shooting with flashes so I'm guessing it won't matter with the recycle times.
TaoTeJared said:
@studio460 - no problem. On a side note, you may want to look up or test what shoots faster if you are really trying to capture high speed stuff. Jpeg I believe is the fastest. :\ It sounds like you are shooting with flashes so I'm guessing it won't matter with the recycle times.
Thanks again, Tao. This weekend, yes, I'll be shooting with flash, but my primary key (SB-800) is powered with a high-voltage pack, so recycle times are wicked-fast. I may experiment with RAW at a later date, but I'll be shooting with a familiar workflow, .JPGs, on this first session. I've read this entire thread carefully, weighed both arguments, and have decided to shoot .JPG, at least this first time out. One of the most compelling arguments came from some guy named "Ken," who tends to write a lot about Nikon stuff on his site, and if his facts are accurate, he makes a pretty convincing case for .JPG.
Arguably, the greatest benefit from shooting RAW would be its claimed increased dynamic range, but I could find no empirical data to support this claim. As others have argued, I always aim to expose for the highlights anyway, and I typically don't have any white balance issues. Under controlled lighting conditions, I will be able to dictate my scene's brightness ratio to within what the sensor can handle natively. This is what we've had to do when lighting for television for decades. However, I'm also aware of the lossy nature of .JPG compression and all that it entails. Ideally, as others have stated, RAW is ideal as an acquisition format, and .JPG more appropriate as a distribution format only, but I'm only experimenting at this point, and I intend to shoot like a zillion frames this first time out. Thanks for everyone's input on this topic--there is no "right" or wrong" method to shoot, just different options for different needs.
I personally don't get much of an advantage with Raw unless it is magic hour AM sunrises. The graduations of color are much, much better. I do believe there is more info in RAW and you can bring back some of the files more if your exposure is not close.
That seems to be the underlying reasoning in the arguments for it, when you miss exposure or white balance is off. (Low light as well but they overlap a bit.)
I was taught to get the exposure and everything else is a hot knife through butter. I also always shoot with an ExpoDisk (WB neutral) and always get files that I can easily move color around on. Jpeg works for me with the way I shoot.
Personally, Jpeg or Raw makes very little difference so I shoot Jpeg. Bellow are a couple of extremes where I see the difference:
Orig was a Raw file
Orig was a Jpeg file Graduations not as smooth
Orig was a Raw file - same morning as the above file after shooting it in raw. Note the smoother graduations.
If you are looking for numbers, the true RAW numbers reside in the 14 Bit depth. As the Bit depth increases the number of recorded colors increase exponentially. Mainly it deals with the graduations and smoothness as well as sharpness of an image. For example Hasselblad's raw is 16 bit and the images that come from that are amazingly rich.
Thanks for posting those, Tao! I do see a difference in the gradients, but the second image also appeared less saturated (I know, many variables). Also, I don't argue that people seem to be "pulling" more detail out of the highlights in RAW files, I just haven't been able to locate any empirical data showing this quantitatively.
Another significant difference between the two, as I understand it, is that a preset gamma is applied to the image's characteristic curve in a .JPG capture, while in a RAW capture, it can be re-determined in post. This also, is likely a key argument for shooting RAW. With the exception of RAW-based, motion-video cameras like RED, in most modern broadcast video cameras, many of these parameters are field-adjustable (black level, gamma, knee, etc.). I guess there isn't really a parallel level of pre-capture control (over the characteristic curve) in even top-line DSLRs yet.
Also, the key argument regarding bit-depth which got my attention was this one:
"JPG has 8 bits per color per pixel and raw may have 12 bits, but here's the big catch: raw is 12 bit linear, and JPG is 8 bit log, gamma corrected or some other non-linear transform derived from the 12 bit linear data." --that "Ken" guy.
While I understand the basic concept of bit-depth, this statement seems to imply that there is "useful" data compression achieved in the in-camera gamma application to the curve, in and of itself--possibly a "good" thing, depending on the definition of "good" here. Now I don't mean to start another heated debate about this, I just want to vet some of these statements so I fully understand what's actually going on, and what, precisely, I'm giving up.
Lastly, sRGB vs. Adobe RGB: I am also choosing to shoot in Adobe RGB, since I will always be aware that I'm in Adobe RGB in post, and I will only be sending files to Adobe RGB-aware printing devices.
studio460 said:
Also, the key argument regarding bit-depth which got my attention was this one:"JPG has 8 bits per color per pixel and raw may have 12 bits, but here's the big catch: raw is 12 bit linear, and JPG is 8 bit log, gamma corrected or some other non-linear transform derived from the 12 bit linear data." --that "Ken" guy.
You should not forget that JPEG is a format that is beeing used for a long, long time. And with a good reason: it serves its purpose very well.
I think the comparison JPEG <> RAW is like slides <> negatives is very good. Slides are easier to archive (you don't need to store the negative and the print) and give brilliant colors right out of the box (if the exposure is good).
Negatives give you more freedom in post (especially b&w). Dodging, burning and different grade papers (or even multigrade)...
But these properties do not make slides better than negatives or vice versa. It is just different.
To me the most important thing is to select the file format (or film type) that fits your workflow. Why choose RAW if you don't like to do PP? On the other hand: why take a chance and shoot JPEG if PP is something you really love to do?
Take in mind that JPEG comes in different flavors, and we users do not always have control over what the camera outputs. The common flavor uses a lossy compression algorithm which reduces the image size and quality.(*)
From a lossy compressed file you can not get an exact reproduction of the original data after decompressing the file.
Also you can not get good results if you do multiple-edits on the same file (load, fix, save.. then load again, fix and re-save), due to the fact that each time the file is compressed to save the information, some of the information on the image is discarded by the JPEG lossy compression algorithm.
That is the reason, the results from @TaoTeJared on the JPEG file out of the camera have non smooth gradients compared to the JPEG files that were generated from the RAW files.. (that may be saved using a different compression factor).
The second item is the number of bits. The more bits you have per color may help you to get better results when post-processing the file.(**)
@Correlli: The reasons why JPEG is widely used are:
1) It allows to reduce the image size dramatically, making it more versatile for Web applications, paying with image quality.
2) A royalty-free format, that was the output from a community called "Joint Photographic Experts Group" (That is where the JPEG name comes).
That is, JPEG is not used because it is the best image format(***) but because its wide use.
--- Ricky
(*) The JPEG format contains options also to do a lose-less compression. but cameras do not use this format.
(**) There is also JPEG format that supports 12 bit, used in medical applications(Usually grayscale images)
(***) In the early 2000 a there was proposed a better compression algorithm to be used in JPEG files (Also called JPEG2000) to the Joint Photographic Experts Group, but because some limitations imposed by Hewlett Packard (That was a member of the comity), this enhancement was not widely adopted.
Hummm gamma. I may have to read up on that one. Not questioning it, you may have helped me find the little devil that is causing me problems with a printer lab that I send files to. I'm a PC guy, and Gamma is one colossal screw up that Apple causes. Someone down stream may not be using the embedded settings.
studio460 said:
"JPG has 8 bits per color per pixel and raw may have 12 bits, but here's the big catch: raw is 12 bit linear, and JPG is 8 bit log, gamma corrected or some other non-linear transform derived from the 12 bit linear data." --that "Ken" guy.
I understand the mathematical theory behind it. This would be one of the times I would say someone (Big K) got a little too big for his britches and took a term and miss applied it's practical use. He got stuck on the word "linear" that was published somewhere/ somehow, and then added his "JPG is 8 bit "log"" (he added log) and came to a conclusion I don't participially share. JPG is Log but I am not convinced that RAW is just Linear. There is more too it that is way above what I really care about to be honest. There is some merit in it - I would just have to hear Nikon's side on it.
To be honest, I could really care less, but my math background gets hung up on some of theses things. Mom told me that Proof is in the pudding, and the Raw works better sometimes. I still will shoot about 300 JPEGs today though.
Yes, I understand most of the arguments for/against, RAW vs.JPG, but I just haven't been able to find any quantitative data showing measurable differences in captured dynamic range. I'd also like to vet Ken's statement, since it makes for a convincing case regarding the bit-depth argument. I did see some resolution charts on another site which showed that the RAW files were able to resolve greater detail than the .JPG files, but it was very subtle.
studio460 said:
Yes, I understand most of the arguments for/against, RAW vs.JPG, but I just haven't been able to find any quantitative data showing measurable differences in captured dynamic range. I'd also like to vet Ken's statement, since it makes for a convincing case regarding the bit-depth argument. I did see some resolution charts on another site which showed that the RAW files were able to resolve greater detail than the .JPG files, but it was very subtle.
The point is photography is a visual art and if you can see the difference (even if there is no tangible way to record it) than it must be better... At least that is what us RAW junkies like to say.
kyoshinikon said:
The point is photography is a visual art and if you can see the difference (even if there is no tangible way to record it) than it must be better... At least that is what us RAW junkies like to say.
If there's increased dynamic range, this should be measurable. I guess I'll go out and buy a Kodak 21-step chip chart or something and try and see for myself.
kyoshinikon said:
The point is photography is a visual art and if you can see the difference (even if there is no tangible way to record it) than it must be better... At least that is what us RAW junkies like to say.
A wise old man once said to me "Grasshopper. You should strive to capture as much information as possible on your in-camera images."
He went onto say, "You can always throw away what you don't need or want to be in your finished files (probably JPG's for some web site), but if the info isn't there to begin with -- you can curse and scream all you like... it won't make the slightest bit of difference."
(sorry about the pun "bit"...)
The message being, capture as much as you darn well can, THEN when you are worrying about it on the computer screen, you can always reduce size/bit depth but adding it in takes something akin to magic.
For me that means RAW or at a pinch PSD or TIFF. Now *YOU* may personally find different, in which case, whatever works, but like I say... if it isn't there at the start... give up and go away (or use HDR) - now THAT'S another topic entirely. :)
TaoTeJared said:
I understand the mathematical theory behind it. This would be one of the times I would say someone (Big K) got a little too big for his britches and took a term and miss applied it's practical use. He got stuck on the word "linear" that was published somewhere/ somehow, and then added his "JPG is 8 bit "log"" (he added log) and came to a conclusion I don't participially share. JPG is Log but I am not convinced that RAW is just Linear. There is more too it that is way above what I really care about to be honest. There is some merit in it - I would just have to hear Nikon's side on it.
Sorry, Tao, I wasn't replying to your post (I didn't see it until now). Nevertheless, I think the claim has some information which may be valid. Any further discussion on this is simply beyond my technical expertise, so I just threw it out there for Testing, Jonny, Drab, or someone like that to consider.
From what I understand, applying an in-camera gamma value, is in fact a form of "compression," and the fact that the DlogE curve (i.e., the "characteristic curve") is represented logrithmically, may in fact result in the description of the .JPG's bit-depth as, "8-bit log," as an accurate statement.
That said, I have found many other qualitative examples, elsewhere online, showing RAW files capable of delivering superior end results; I just haven't seen any quantitative data.
No issues here - It would be interesting to hear what people can find.
rbid said:
@Correlli: The reasons why JPEG is widely used are:
1) It allows to reduce the image size dramatically, making it more versatile for Web applications, paying with image quality.
2) A royalty-free format, that was the output from a community called "Joint Photographic Experts Group" (That is where the JPEG name comes).That is, JPEG is not used because it is the best image format(***) but because its wide use.
Absolutely agree on that. This is what I meant by "it serves its purpose very well".
studio460 said:
Yes, I understand most of the arguments for/against, RAW vs.JPG, but I just haven't been able to find any quantitative data showing measurable differences in captured dynamic range. I'd also like to vet Ken's statement, since it makes for a convincing case regarding the bit-depth argument. I did see some resolution charts on another site which showed that the RAW files were able to resolve greater detail than the .JPG files, but it was very subtle.
Sounds to me like its time to make your own experiments to discover if there is value in RAW for your style of shooting. Also, keep in mind many of Ken's statements on this are from 2004. Also, the guy really likes to come off as dismissive in one paragraph and then in the other he's completely contradicting nearly everything he just dismissed. His article on RAW is no exception and, like most everything else he writes, it reads like he sat down 7 different times to write basically the same thing, but using slightly different language. Basically, his advice on this department is hardly reliable and hardly definitive and very, very redundant...but who am I to say. JPEG or RAW its all still photography I guess. ;]
... mmm... gamma.. this remembers me that there is an additional factor in comparing photos: color profiles and monitor/printer calibration.
What do you have to say with this issue that is also related to RAW vs. JPEG.
Let's just stick to the RAW vs JPEG discussion guys. This thread is getting too long and some of these "off-topic" points are going to get buried. Go over to the "What is "acceptable" as good image quality?" thread and bring up gamma/color space/monitors/printers over there. Otherwise we'll be repeating the same stuff all over again.
Well, after reviewing quite of few online examples of comparison photos, I've decided to shoot .JPG mostly, but for specifically "set-up" shots, I will use the RAW + .JPG shooting option in my camera. I simply didn't buy large enough CF cards to accommodate a full-day of shooting at RAW capture file sizes. I will be shooting compressed 14-bit NEF, and try converting in Adobe CS5.
You bought a D3s only to shoot it in JPEG?!
*Facepalm*
rbid said:
... mmm... gamma.. this remembers me that there is an additional factor in comparing photos: color profiles and monitor/printer calibration.What do you have to say with this issue that is also related to RAW vs. JPEG.
I have never saw anything different - it is the same for both.
Okay, I admit it . . . I have absolutely no idea what I'm doing.
You guys were right--I just shot some tests on my D3s, and I do see some aliasing (posterization) in the .JPG files (see gradient on right part of frame, below). I shot RAW + JPG, but I already deleted the card, and I don't know where Aperture put the NEFs, if anywhere. It took forever to download the CF card, so I assume it put the NEFs somewhere. I have the older version of Aperture, v2.1.4, which doesn't support something or other. I have the camera set to Adobe RGB, but when I tell CS5 to "assign to profile; working profile: Adobe RGB" it gets overly saturated (the image below is before this step). Should I go back to sRGB?
Nikon D3s: color space = Adobe RGB; picture control = standard; JPG, large, optimum; downsampled to 800 x 532 pixels.
Do I have to re-shoot and pull the NEFs directly from the CF card? How do I get them into ACR in CS5? Help!
I know we are a nikon forum but you really cant blame the JPG format for the not so good images you get from the nikon cameras. As you guys know I have a S5pro and the JPGs that come from that have amazing smoothness and DR that are way superior to the JPG that comes from the Nikon cameras that I have seen and have. So on Nikon I shoot Jpg+NEF-14 lossy Compressed and on mt S5Pro I shoot Jpg unless the scene requires very High DR then I shoot Jpg+RAF.
PS Studio Nice image BTW. Might go with a squarish crop and remove right side a bit ? Lovely hand.. is she the one you said has had hand modeling jobs?
You must log in to post.
