[Lowell Goudge]

first of all, let's get rid of th ebiggest myth out there. RAW data is NOT data right off the sensor.

If it were, it would be a series of varrying intensity red, green and blue dots.

see the attached link for the description of the bayer filter used on most digital cameras

Bayer filter - Wikipedia, the free encyclopedia

Aside from that, what RAW really consists of is the following.

An interpolation off the Bayer filter array to provide the user with red, green, blue and black channels at a conversion of either 12 or 14 bits per channel.

When you save in JPEG, this is reduced (by interpolation) to 8 bits per channel. One of the biggest impacts, therefore when converting to JPEG is that the steps between individual different colors or even shades within one color are larger. In many cases, this is not a problem, BUT when trying to dig detail specifically out of the shadows, this can be a problem, because the JPEG image also compresses the shadows more than a RAW image, so these "steps" between shades are larger.

The other significant difference is that JPEG has additional processing based upon camera settings, specifically contrast, color balance (white balance), noise reduction and sharpening.

Extreme compression in Jpegs, as snostorm suggests also starts to compress data by merging common elements of the image within larger blocks. this is where the real loss of image quality hits, but if you pick maximum resolution this is not significant.

If you are comfortable with your skills and knowledge of the camera settings, and / or you never take extreme shots which you know will require significant processing, JPEG can easily suit your needs.

The biggest arguments presented for RAW aside from greater color depth, is the ability to correct errors in post processing. To me the best credible argument, presented on the forum to date is that as time progresses, there may be development of better noise reduction and sharpening algorythims and as a result there is "potential" for better processing in the future. I say potential, because I am not sure it has happened yet.

For me, I shoot JPEGS almost exclusively, and have the same interests as snostorm, and have no complaints. I pay attention to white balance when shooting, amd modify WB to suit the lighting, I do the same with contrast, and I also pay close attention to exposure. I put an effort to getting it right at the time of capture, as opposed to fixing errors in simple things later.

It is everyone's personal preference, nothing more.

Don't let people tell you that shooting JPEGs makes later adjustments impossible, that is pure BS. Processing JPEGs is no different than processing RAW, you just don't have the same range of adjustments. If your exposure is close, i.e. within 1 stop it probably does not matter, but if you are out by 2-3 stops, then RAW can help some, but regardless it will not be a great image, it will still look like it was recovered from the scrap heap, just a little less so.

[Chex]

HI C1,

There are numerous Jpeg vs RAW threads on this and all fora on the web. The forum software even supplies a "Similar Threads" listing at the bottom of this page.

I'm a Jpeg shooter almost exclusively, but will be the first to admit that RAW gives you considerably more potential to edit your images. You have the ability to do significant edits with Jpegs, but you can do more with RAW. In the end, you'll probably be converting your images to jpegs for showing or printing and I'd guess that there are few, if any people who can consistently distinguish in what format any given image had been shot originally.

Jpeg compression does discard data from the original file that the sensor creates, but usually this is not as bad as it might sound. Consider that there are usually large areas of any image where each individual sensor cell is capturing the same information as the adjacent ones (I realize that each sensor is color sensitive for red, green or blue, and there are 2 green ones for every red or blue, but I'm considering one group as one cell in this example) -- this information is redundant, and all that the Jpeg compression algorithm does is accumulate a list of pixels with the same value and then re-assign them the same value for the image when the compression of the file is expanded for viewing. My K-5 gives me approximately 2 to 3:1 compression with 6-11 MB files opposed to the 16-26 MB files that DNG RAW produces. Considering the amount of redundant information in most images, this is not extreme compression, and not that much information is actually tossed out.

Jpegs use a lossy compression scheme, and each edit and save of the same file results in the possible addition of compression artifacts, so it's a good idea to always save an edited file to a different name, and if you want to re-edit the image, always start with the original again. With some RAW editing, you can save the editing workflow, start with where you left off, and undo some steps , then redo them, saving the newly edited file to a new name. You can do this with Jpeg editing, but it has to be during the original editing session, before you close the file you're working on. It usually takes at least a few edit and save cycles with the same file before the compression artifacts start to become obvious though, especially when you're starting with the highest quality jpegs out of the camera.

Still, a 14 bit K-5 RAW file will give a whole lot more leeway to pull detail out of shadows and will give extra highlight headroom than the 8 bit Jpeg will. I believe with your K10, they advertised 14 bit RAW, but it turned out to be 12 bit with 2 unused bits so there was no practical difference in DR. Properly exposed jpegs will give all the information that most need for great images. RAW gives you more exposure latitude, and a larger range of potential for editing. RAW also allows you to disregard in-camera settings for image parameters like Contrast, Sharpness, White Balance, etc. You can set up a RAW processor to make these adjustments automatically and process your files in batches.

Setting up Jpeg output to your liking and making sure that you expose at least close to correctly can cut out this intermediate step and save time and storage space, especially if you're a high volume shooter.

The argument that the camera has only a "little" computer that is less powerful than your desktop is true, but not really valid from a practical standpoint, IMO. That "little" image processor may be small in size and not as powerful as the CPU in your computer, but it's dedicated to do only what it was specifically designed to do --process images -- and it can convert RAW data into a jpeg in a few seconds at most (at least some of the lag between taking a picture and being able to review it on the LCD is due to the usual I/O bottleneck of writing to the card). I doubt that many who do batch RAW to Jpeg conversions on their home computers can claim that their setups are more efficient and can do the conversions in less than 2 sec/image with 16MP images.

The argument that memory is cheap is a good one on the surface, but there are hidden implications. Consider that I currently shoot 10-20 K shots/year. At 25MB/RAW file, that's 250 - 500 GB, and if your images are important, then you'll have at least one backup, so that means .5 - 1 TB per year at my current rate. I just bought 2 x 1.5 TB external HDs at $60 each. . . not terribly expensive, but then I tied up a whole day at least to organize the drives, do the backups, and verify them. . . and I only use HDD to backup -- if I were to use single layer DVDs at 8GB per disk, that's at least 30 disks, and they take a while to burn, and who knows how long they'll really last. BDs have 25 GB capacity, and would still take about 10 disks per year, but I haven't invested in this technology yet. I save everything but totally unredeemable shots. I've found that new technology can actually save some shots that I would have discarded 5 years ago, and I regularly spend my winter idle time going over old shots to pick out those that I might find interesting now, but didn't when I took them or in previous reviews.

From the Panny FZs through the DS, K10, K20, K-7 and now with the K-5, I have over 100K images. Because of lower resolution, these files take up a little over 500GB in Jpeg. I can view them all with ACDSee 3.1 (a very old version that I like for its speed) because they are jpegs. If they were RAW, I'd need a program that included every version of PEF for the different models or DNG for the newer ones.
ACDSee Pro will open all the RAW files. If you choose to use PEF, which is Pentax's proprietary RAW format, each new model uses a new version, so PEF users have to wait for software mfgs to release an update that includes their model. For DNG, it's an open source format, so RAW processors that already are set up for DNG, it's no problem, unless Pentax does something strange in their implementation, which I believe has happened in the past . . . I don't really know, I shoot jpegs and don't have that problem.

The argument that you can convert a RAW file to Jpeg, but not the other way around is true, but somewhat mitigated by the ability to shoot RAW + Jpeg if you want to have some insurance. The K10 was the first to have the RAW button which allows you to switch formats on the fly, and the K-5 now allows you to shoot in Jpeg and upon review on the LCD (chimping), you can choose to save the RAW information of the last file while it is still in the buffer by pressing the AE-L button. Although I appreciate all of these options, I rarely have used them, but the new ability to save the RAW file will possibly help save some shots where I've made some exposure errors. . . if I can remember to use it. Personally, I've programmed my RAW button as an instant switch to give me exposure bracketing.

I don't feel that you lose appreciable resolution with Jpegs, as long as you use *** or **** quality settings. I'm a bird shooter, and feather detail is an important element of what I like to do. Here is an example of what Jpeg can capture if everything goes right. . .



Here's a 100% crop with no editing from the lower middle of the frame,



This was shot with a K20. Higher resolution is always nice, but this is certainly good enough for my purposes. . .

It's also argued that RAW files exhibit less noise at higher ISO, but I've gotten good very high ISO performance from my K-5 with Jpegs. I've found that by setting "Sharpness" to -4, with High ISO Noise Reduction (NR) turned OFF, I can shoot Jpegs that probably rival RAW shots at the same sensitivity. I just do the NR and sharpening in PP with Topaz Denoise and InFocus, which I think does at least as well with Jpegs as any RAW NR with possibly better sharpening.

This one's with the K-5 and a Sigma 180/3.5 APO DG Macro wide open at the local Nature Center through glass. ISO 10,000



This one's with the K-5 in ordinary room lighting, using a DA 18-250. ISO 12,800



I was trying to see if miraculously any of the nose leather texture would still be present after NR, it wasn't, but I can't really be disappointed in the result -- it's ISO 12800 after all. . .

I'm confident enough in the K-5's noise performance with Jpegs at my settings that I'm at least starting off the birding season using Auto ISO 200-10000. That's about 3 stops faster than I'd shoot the K20 or K-7, and it's gotta help with the long glass and light-robbing TCs that I shoot most of the time.

I shoot pretty high volume (historically from 10K-20K per year), and to me, Jpeg *** doesn't show any artifacts that might be accentuated by PP, so I don't even shoot the highest quality compression available in the camera. Shooting RAW, I'd not only spend a lot more time in front of the computer just reviewing files, but I'd have to convert them to edit them the way I want anyway as my most used plugins don't work with RAW images AFAIK.

I've found that there are a lot of what I consider myths concerning Jpeg vs RAW. Jpegs are more challenging to set up and expose correctly, but done right, there's little to be gained shooting RAW for me, except for very unusual lighting conditions. Consider all the arguments, choose whichever you prefer for whatever reason you might choose, and don't let anyone tell you that you've made the wrong decision -- there really isn't a "right" to this controversy.

Scott

Now had you shot those pics in RAW, the level of detail you could bring through using Camera RAW, Photoshop, Lightroom.. any program would be far superior to the standard jpeg. To me it becomes the question of "Do you care enough about your pictures to do PP work?" if yes, shoot RAW... if no then Jpeg.. but thats what P&S and super zoom camera's are for.

[cats_five]

Now had you shot those pics in RAW, the level of detail you could bring through using Camera RAW, Photoshop, Lightroom.. any program would be far superior to the standard jpeg. To me it becomes the question of "Do you care enough about your pictures to do PP work?" if yes, shoot RAW... if no then Jpeg.. but thats what P&S and super zoom camera's are for.

I'd say it MIGHT rather than WOULD. And 'care enough' - a rather emotive way to put it. I have many really good pictures I took with a Canon A70. Now it's true that at the time I brought it digital photography was somewhat in it's infancy, but it (and many other cameras that have come along since) can produce great results in good conditions.

Here's another way of looking at it. If I go for a days walk, I can take the K7 plus the tripod plus the spare lens plus the spare battery plus the ND filters plus all the other kit, or I can take the A70 plus a spare set of batteries. If I'm planning a long talk I can tell you which as an unfit hill walker I'd take, unless there is something I know that is truely spectacular at the far point of the walk. Is it better to get to those wild places and take some shots, than not get there and take none?




(no this isn't a wild place except maybe on Saturday night, but it was taken with the A70)

[Chex]

If you would take a tripod for your K-7 why not for your P&S.. not a real comparison.. taking your K-7 and one standard zoom may be heavier then your A70, but the results would be far better as well. For those long walks and only taking one lens is the reason I bought the 18-135 WR for the K-5.. weather sealed with good range. Maybe bring a spare memory card and battery at most. And I do say care enough, because if you are shooting with your DSLR or any camera than can shoot RAW there is no real benefit to shooting jpeg other than file size. most camera's you can do minor on camera editing that saves the edit as a jpeg.. that IS a form of PP.

[Chex]

he argument that memory is cheap is a good one on the surface, but there are hidden implications. Consider that I currently shoot 10-20 K shots/year. At 25MB/RAW file, that's 250 - 500 GB, and if your images are important, then you'll have at least one backup, so that means .5 - 1 TB per year at my current rate. I just bought 2 x 1.5 TB external HDs at $60 each. . . not terribly expensive, but then I tied up a whole day at least to organize the drives, do the backups, and verify them. . . and I only use HDD to backup -- if I were to use single layer DVDs at 8GB per disk, that's at least 30 disks, and they take a while to burn, and who knows how long they'll really last. BDs have 25 GB capacity, and would still take about 10 disks per year, but I haven't invested in this technology yet. I save everything but totally unredeemable shots. I've found that new technology can actually save some shots that I would have discarded 5 years ago, and I regularly spend my winter idle time going over old shots to pick out those that I might find interesting now, but didn't when I took them or in previous reviews.

Having a stockpile of old jpegs from P&S's thrown on a 1.5TB drive doesn't take up much room. Now with the K-5 I intend to get 2 2TB drives and mirror them(raid1) so I always have a backup if one drive craps out. 2TB will last a long time as each year you go back and try to streamline your collection of keepers. And for those who want backups for the business side of photography, simply offering your client 1 year after the shoot before the data is lost (other than shots you want to keep for your portfolio) is more than enough time. Even then the number of shots you would normally keep for a portfolio should fit on a DL DVD or a USB memory stick (they are getting very big now days!).

[SpecialK]

There are many more adjustments and levels of adjustment in PP software than in the camera. Much of it works better with RAW than jpg files. RAW files are your proof of ownership because you never post or share them.

[Frogfish]

If you would take a tripod for your K-7 why not for your P&S.. not a real comparison.. taking your K-7 and one standard zoom may be heavier then your A70, but the results would be far better as well. For those long walks and only taking one lens is the reason I bought the 18-135 WR for the K-5.. weather sealed with good range. Maybe bring a spare memory card and battery at most. And I do say care enough, because if you are shooting with your DSLR or any camera than can shoot RAW there is no real benefit to shooting jpeg other than file size. most camera's you can do minor on camera editing that saves the edit as a jpeg.. that IS a form of PP.

Well to start with I think it is rather disrespectful to say 'if you care enough' as a way to differentiate between shooting in jpg or RAW. I recently did a shoot with a Pro and there were literally hundreds of shots - it would have taken him and his wife a month to process them as many were for the 800 guests (something that as a Pro he just doesn't have the time to do) and the very high quality jpgs from his 7D and my K5 were plenty good enough for top quality A3 prints. That is to say nothing of the other quality shooters on this forum who prefer jpg for one reason or another.

As for no difference between jpgs shot from a P&S and a DSLR - well I guess you imagine that much greater sensor size, larger file sizes, high ISO range and lower noise, the processing ability of the DSLR and all the other benefits of shooting with a DSLR count for nothing ?

For the record I shoot jpg most of the time (saving tough shots to RAW in camera after the shot if need be) and usually only shoot RAW in high contrast or low light conditions.

[Chex]

Good call SpecialK, never thought about the proof of ownership, and that is a HUGE issue now days with people ripping off good quality jpegs online and using them wherever they want to.

[Frogfish]

Good call SpecialK, never thought about the proof of ownership, and that is a HUGE issue now days with people ripping off good quality jpegs online and using them wherever they want to.

Most people don't post their 14MB jpgs on the web so I don't see any less proof of ownership there.

[Chex]

Frog, I just don't understand why someone would choose jpeg, even for events with lots of shots.. change memory cards and keep shooting. Most programs do have Batch converters so you could convert them all with little effort keep the great shots that you might actually do PP work on and dump the rest of the RAW's when your done. Yes it takes more time, but yes it gives you better QC in your final output images.. and to the client, thats what your judged on.
I don't mean "care enough" in a disrespectful manor, probably not the right way to word it.. sorry. I just can't fathom why anyone who wants the most out of their images would ever choose jpeg over RAW.

[Chex]

Most people don't post their 14MB jpgs on the web so I don't see any less proof of ownership there.

I've seen some fairly large shots, and even larger digital art pics on sites like Deviant Art and Digital blaspheme. Regardless, to be ripped off you only have to be as big as a standard desktop resolution, even down to 1600x1200 is large enough probably even smaller. It's not like they try to claim ownership.. they just don't intend to get caught.

[Frogfish]

Frog, I just don't understand why someone would choose jpeg, even for events with lots of shots.. change memory cards and keep shooting. Most programs do have Batch converters so you could convert them all with little effort keep the great shots that you might actually do PP work on and dump the rest of the RAW's when your done. Yes it takes more time, but yes it gives you better QC in your final output images.. and to the client, thats what your judged on.
I don't mean "care enough" in a disrespectful manor, probably not the right way to word it.. sorry. I just can't fathom why anyone who wants the most out of their images would ever choose jpeg over RAW.

Could it be that under 'most' conditions no-one can tell the difference between RAW processed images and high quality jpgs ? I think that may be the answer.

[Frogfish]

I've seen some fairly large shots, and even larger digital art pics on sites like Deviant Art and Digital blaspheme. Regardless, to be ripped off you only have to be as big as a standard desktop resolution, even down to 1600x1200 is large enough probably even smaller. It's not like they try to claim ownership.. they just don't intend to get caught.

True. But in those cases owning a RAW file is of no more use than owning the 14MB original jpg - both are proof of ownership but both are irrelevant in cases such as mentioned above.

[Chex]

You are saying if you had your jpeg, and the RAW file, did your PP work on the RAW file then converted it, you don't think you could tell the difference? When you do work with saturation/desaturation, Hue/color balance RAW's are far more workable. If your cataloging the event using only composition and color/sepia/B&W for your artistic arsenal then jpegs probably do the trick. On camera editing allows some extra work, but lacks the depth and detail of using a proper PP program with RAW files.
Granted you have to have the time to do the PP work, no difference from photographers doing that extra cared darkroom work with film. Simply taking the roll of film to the 1 hr photo doesn't get the same results as working it in your own darkroom for getting the colors/saturation etc you were after.

[Chex]

True. But in those cases owning a RAW file is of no more use than owning the 14MB original jpg - both are proof of ownership but both are irrelevant in cases such as mentioned above.

I'm not sure this is true.. I may be wrong, but jpeg files can be wrote over (edited and re-saved with same name) which changes the file permanently.. hence someone who knows WTF they are doing could edit out the embedded copyright info and re-save the file. A RAW file on the other hand is a read only file.. when you edit a RAW file, you only edit a small embedded file that controls the workable parameters of the file. Everything that was saved is never lost, just cataloged in a small separate embedded file.
I think the copyright usually falls in the shots of scenery, as its easy to prove with a superior copy of people or live action weather you have a jpeg or RAW its hard to debate having the superior copy of something that happens for that one moment only.. but scenery and inanimate object shots can be taken by almost anyone anywhere with close to the same results.. so those are the messy ones.