 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
| PhotoKB HomeDiscussion Groups Digital Photography Digital PhotoDSLR CamerasZLR CamerasPoint & Shoot CamerasFilm Photography 35 mmLarge FormatMedium formatDarkroomFilm and LabsOther EquipmentPhoto Technique Nature PhotographyPeople PhotographyTechnique GeneralGeneral Photo Topics General TopicsAustralian PhotographyUK PhotographyDirectoryPhoto Clubs |
Photo Forum / Digital Photography / DSLR Cameras / April 2006Why many settings arguments don't matter with RAW
I often see some religious discussions regarding certain camera settings, and wonder why they go on at all in a world of RAW pictures. (None of this post is relevant if you shoot in JPG.) sRGB vs Adobe RGB: Since this conversion happens after interpretation of the RAW image data, this is a function of your RAW converter, not the camera. Does this setting have any implication at all outside of a recommendation to your converstion software? Photo buffs far and wide argue the merits of Adobe RGB as a camera settings, but from my interpretation of the RAW image data, the color space of these images is unique to each manufacturer, and any conversion to another color space happens after/during the de-mosaicing step. This means that the camera setting is useless when shooting in RAW. Sharpness, Contrast, etc: This setting shouldn't mean anything when shooting RAW, since this is something that is performed after de-mosaicing. (I once posted images for comparison on a photo site, and had people beat me up for not posting what my sharpness setting in the camera was. They didn't seem to understand that this step wasn't performed in the camera.) Whitebalance: Same, no effect on RAW. In fact, the only real "settings" that matter to a raw image are the actual exposure (in as much as you can set exposure compenstion for an image), and ISO/aperature/shutter speed. I sort of wish that SLRs would simply remove these options when shooting in RAW, so people stop having some notion that they do anything. Am I mistaking any of the features of my Camera, and how they are applied to images? -drew >I often see some religious discussions regarding certain camera > settings, and wonder why they go on at all in a world of RAW pictures. [quoted text clipped - 31 lines] > > -drew dtype, as an addendum to your comments: I've shot shot slide film for years -- literally, as you know, what you shot is what you got --- and RAW gives me the same type of results and feedback that slide film did. It allows me to tell the story I want to communicate with my photographs which is the bottom line of "writing with light". To me, RAW avoids the problem of interference by some set of algorithms that make me guess whether it was me or the camera that messed things up -- shooting RAW tells me that it was the operator and not the equipment! And as a learning tool for new photographers or for experienced ones learning a new camera, RAW is invaluable. Lastly, and to me this is very important to me, I find RAW cuts down on the amount of time spent in post processing -- I don't have either the time nor the inclination to do a bunch of "adjustments" to an image that perhaps were a result of the camera settings in the first place. I'm a professional photographer, not a Photoshop expert. P.S. I like my vegetables raw too.....:) Mike > dtype, as an addendum to your comments: I've shot shot slide film for > years -- literally, as you know, what you shot is what you got --- and RAW > gives me the same type of results and feedback that slide film did. It > allows me to tell the story I want to communicate with my photographs > which is the bottom line of "writing with light". What utter rubbish, surely the reverse is true, surely RAW gives you maximum opportunity to change everything afterwards? > To me, RAW avoids the problem of interference by some set of algorithms > that make me guess whether it was me or the camera that messed things > up -- shooting RAW tells me that it was the operator and not the > equipment! And as a learning tool for new photographers or for > experienced ones learning a new camera, RAW is invaluable. > Lastly, and to me this is very important to me, I find RAW cuts down on > the amount of time spent in post processing -- I don't have either the > time nor the inclination to do a bunch of "adjustments" to an image that > perhaps were a result of the camera settings in the first place. I'm a > professional photographer, not a Photoshop expert. If you have the settings right in jpeg mode surely that will take less effort to get to the final product? > P.S. I like my vegetables raw too.....:) Me too. > Mike >> dtype, as an addendum to your comments: I've shot shot slide film for >> years -- literally, as you know, what you shot is what you got --- and RAW [quoted text clipped - 4 lines] >What utter rubbish, surely the reverse is true, surely RAW gives you maximum >opportunity to change everything afterwards? I agree. RAW is more like a negative, where you can do everything you want in the printing process, and jpeg more like slide film (though, of course you can do editing with jpegs as well). >If you have the settings right in jpeg mode surely that will take less >effort to get to the final product? Especially for "native" raw-converters, like Nikon Capture that reads all the metadata and displays is that way (color space, sharpening, cruve etc). Other third party converters don't read them (but white balance). -espen  Signaturehttp://www.seland.org/ >I sort of wish that SLRs would simply remove these options when >shooting in RAW, so people stop having some notion that they do >anything. The settings don't effect the raw data but they're used to generate the thumbnail/preview and/or the accompanying JPEG. You wouldn't want to lose the ability to view images on the camera would you? In addition they are sometimes used to give a hint to the raw converter. They're harmless, so why worry about them? SignatureJohn Bean > The settings don't effect the raw data but they're used to > generate the thumbnail/preview and/or the accompanying JPEG. > You wouldn't want to lose the ability to view images on the > camera would you? True there, although on that LCD, color space and sharpening/etc probably don't have much meaning. I've never shot with an accompanying JPEG, so I hadn't thought of that. > In addition they are sometimes used to give a hint to the > raw converter. They're harmless, so why worry about them? Yes, I can see their value, especially for accompanying JPEGs. As for hints to the RAW converters, I'd venture that since those hints are generated from the RAW data itself, and not from other camera settings (most of the time), raw converters are technically better prepared to make these judgements on their own. The exception might be if the photographer wanted to provide these hints at the time of taking a picture, and have those remembered through the workflow process. I think that it just isn't always known that these settings don't have meaning to your RAW workflow. It certainly wasn't to me at first. I spent a good bit of time internally considering color spaces to shoot in before coming to the realization that it didn't matter in the least. In reading 20 different flamewars on usenet, flickr, and dpreview on the subject, I never saw anyone raise his hand and say "Just shoot in RAW, and it doesn't really matter". (I'm sure folks have, but not enough for me to have seen it in several discussions.) Part of my post was also a question. Is my understanding incorrect? (Preview and accompanying JPEG are already two things I hadn't thought of.) -drew > > The settings don't effect the raw data but they're used to > > generate the thumbnail/preview and/or the accompanying JPEG. [quoted text clipped - 28 lines] > (Preview and accompanying JPEG are already two things I hadn't thought > of.) When I open a Nikon NEF (RAW) file in Photoshop, one of my options is to convert as shot. Those are my settings right there. Frankly, though, I like PhaseOne's Raw capture software, and I don't think it even uses those settings for the previews. SignatureRegards, Matt Clara www.mattclara.com >> In addition they are sometimes used to give a hint to the >> raw converter. They're harmless, so why worry about them? [quoted text clipped - 4 lines] >(most of the time), raw converters are technically better prepared to >make these judgements on their own. Not so. Some converters use more or less of the camera settings metadata, but they all have the ability to use some of it. Every converter I'm familiar with has at least a "use camera WB" option. Where does it get that from then? ;-) SignatureJohn Bean >>>In addition they are sometimes used to give a hint to the >>>raw converter. They're harmless, so why worry about them? [quoted text clipped - 9 lines] > of it. Every converter I'm familiar with has at least a "use > camera WB" option. Where does it get that from then? ;-) WB is the only setting that carries over with ACR. > Whitebalance: Same, no effect on RAW. I believe WB does effect the RAW capture. It's probably not real critical in most cases but it's like EC for different colors and will reduce or increase the data captured to have WB set correctly. This is why Nikon was trying to encrypt their WB although it's not that important, the RAW converter can guess the intended WB or an operator can chose it manually. The exposure can also be adjusted in RAW conversion but it's not as good as shooting it correctly in the first place. The jpeg contrast settings will effect the histogram used for judging exposure in the field, although it's arguably most useful to turn down contrast & saturation so that resembles RAW more closely. Also, yes I do shoot RAW+jpeg to speed up download & the culling process, then I only download the best RAW files & keep many of the 'seconds' in basic jpeg just in case I come back & want a little different view or something (it happens). Often I won't even bother converting the raw because the jpeg is fine, later if I decide to print large, then it's worth fiddling with RAW. > I believe WB does effect the RAW capture. It's probably not real > critical in most cases but it's like EC for different colors and will > reduce or increase the data captured to have WB set correctly. Is this the case? Can a Bayer-pattern sensor adjust sensitivity on a per-color basis? I guess it is technically feasible, but wasn't a feature of sensors that I was aware of. Are you sure that this isn't just a hint to the RAW processor that gets handled after demosaicing? -drew >> I believe WB does effect the RAW capture. It's probably not real >> critical in most cases but it's like EC for different colors and will [quoted text clipped - 4 lines] > feature of sensors that I was aware of. Are you sure that this isn't > just a hint to the RAW processor that gets handled after demosaicing? The Nikon D2x is the only one (that I'm aware of) that does white balance at the sensor. For the rest, it's simply a metadata setting that the raw processor uses (or not). The D2x varies gain on individual channels to do white balance better, and then adds metadata so you can proceed from there during raw conversion just as before. (It's adding an additional step, not replacing one with the other... you can still adjust it in raw conversion, but you get additional benefit from getting it closer in the shooting process.) SignatureJeremy | jeremy@exit109.com > The Nikon D2x is the only one (that I'm aware of) that does white balance > at the sensor. For the rest, it's simply a metadata setting that the raw > processor uses (or not). The D2x varies gain on individual channels to > do white balance better, That is quite interesting, and new information to me. Thanks! In certain settings (ones which have high saturation of a particular color), that could in theory get you quite a bit in terms of additional range, as you can effectively expose for each channel separately. Pretty cool stuff. I wonder how well it works in practice. -drew >>The Nikon D2x is the only one (that I'm aware of) that does white balance >>at the sensor. For the rest, it's simply a metadata setting that the raw >>processor uses (or not). The D2x varies gain on individual channels to >>do white balance better, Apparently the nikon cameras do some processing before the raw is written. Adjusting gain on individual channels will only introduce round off error and ultimately a loss in signal to noise ratio. Apparently the nikon D70 raw is not lossless, see: Is the Nikon D70 NEF (RAW) format truly lossless? http://www.majid.info/mylos/weblog/2004/05/02-1.html The above article indicates the D70 raw records only 683 levels from the 12-bit A-D. The canon raw data is truly raw and can be used to derive sensor characteristics, like read noise and full well capacity in electrons, see: Procedures for Evaluating Digital Camera Noise and Full Well Capacities; Canon 1D Mark II Analysis http://www.clarkvision.com/imagedetail/evaluation-1d2 Roger > That is quite interesting, and new information to me. Thanks! > [quoted text clipped - 4 lines] > > -drew >>> The Nikon D2x is the only one (that I'm aware of) that does white >>> balance [quoted text clipped - 6 lines] > Adjusting gain on individual channels will only introduce > round off error and ultimately a loss in signal to noise ratio. But the data gathered should be more useful because it's already closer to where you want to be (if the WB was set right). I do appreciate your comment in another thread about how doing high-bit raw conversion is actually less work because less is screwed up but doesn't this achieve the same objective? > Apparently the nikon D70 raw is not lossless, see: > [quoted text clipped - 3 lines] > The above article indicates the D70 raw records only > 683 levels from the 12-bit A-D. Hmmm "The new Nikon D70's predecessor, the D100, had the option of using uncompressed or compressed NEFs. Uncompressed NEFs were about 10MB in size, compressed NEF between 4.5MB and 6MB. In comparison, the Canon 10D lossless CRW format images are around 6MB to 6.5MB in size. In practice, compressed NEFs were not an option as they were simply too slow (the camera would lock up for 20 seconds or so while compressing). The D70 only offers compressed NEFs as an option, but mercifully they have improved the performance. Ken Rockwell asserts D70 compressed NEFs are lossless, while Thom Hogan claims: Leaving off Uncompressed NEF is potentially significant--we've been limited in our ability to post process highlight detail, since some of it is destroyed in compression." > The canon raw data is truly raw and can be used to derive > sensor characteristics, like read noise and full well capacity [quoted text clipped - 14 lines] >> >> -drew SignaturePaul Furman http://www.edgehill.net/1 Bay Natives http://www.baynatives.com >> Apparently the nikon cameras do some processing before the raw is >> written. >> Adjusting gain on individual channels will only introduce >> round off error and ultimately a loss in signal to noise ratio. AFAIK, only the D200 (and maybe the D2X) do this before writing the RAW data. But They do it on the analogue signal before doing the A/D conversion and thus increasing the dynamic of the (12 Bit) conversion. Of course it can't increase the absolute noise level of the sensor pixels and thus will either increase or decrease the signal to noise ratio of the color channel affected, depending on the direction of the modification (that is similar to the ISO setting) done to same. -Michael >>> Apparently the nikon cameras do some processing before the raw is >>> written. [quoted text clipped - 10 lines] > > -Michael If done on the analog signal with an amplifier with less noise compared to the signal, it could boost signal-to-noise ratio on low ISO settings. Low noise cameras are currently limited by the 12-bit A-D converter at low ISO. At higher ISO, the amplification will not help because the noise from the signal (Poisson statistics of the electrons) dominates. If done on the 12-bit data, there would be round off error and at low ISO the signal/noise is A-D limited anyway (more so with lower signals, like shadows). If done on 16-bit data, then round-off error should be small because you have 4 extra bits, but the signal/noise is still A-D limited at low ISO. At high ISO neither digital nor analog gain will help because the signal is, again, Poisson statistics limited (photon noise limited). Roger >>>> Apparently the nikon cameras do some processing before the raw is >>>> written. [quoted text clipped - 31 lines] > >Roger You know how people into high end audio often have separate D/A converters? It would be interesting if the data could be output from the camera and then sent to an "outboard" 16 bit A/D. You could do this with an evaulation board from a sensor manufacturer. -Rich In message <442D3141.6040507@qwest.net>, "Roger N. Clark (change username to rnclark)" <username@qwest.net> wrote: >At high ISO neither digital nor analog gain will help >because the signal is, again, Poisson statistics limited >(photon noise limited). I don't agree with that; from all I've seen, shadow readout noise is weakest, relative to absolute signal strength, at the higher amplifications. Based on the trends in-camera in my 20D, I don't know why they'd stop at ISO 1600 for amplification; 3200-level amplification would have to plummet in quality to be not worth using. Even if there were no improvements in absolute_signal-to-noise, a decrease in posterization and line-offset-banding would reduce noise a bit (posterization is mainly ignored, and line-offset-banding, although highly visible to humans, is insignificant in the composite standard deviation). Amplification tends to be a very good thing, in terms of absolute_signal-to-noise, and improves right up to ISO 1600 with the recent Canons. Why should it stop right there? As far as your comments on low ISOs go, I think it is pretty clear that even though 12-bit digitization is too coarse for ISO 100, that's not its only problem. An ISO 100 RAW exposure is *far* noisier than an ISO 1600 exposure with the same amount of light hitting the sensor, even after the latter has been posterized to the level of the former. The low digitization merely emphasizes the poor readout of absolute signal levels at low ISOs. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > In message <442D3141.6040507@qwest.net>, > "Roger N. Clark (change username to rnclark)" <username@qwest.net> [quoted text clipped - 10 lines] > absolute_signal-to-noise, and improves right up to ISO 1600 with the > recent Canons. Why should it stop right there? At ISO 1600 the gain is 0.8 electrons per 12-bit A-D number (DN). At lower ISO gain is several electrons/DN. Once you reach a gain of 1, there is no advantage to higher gain. With read noise of 3 to 4 electrons, you are then digitizing the noise quite well. Note on a P&S camera, like an canon S60, the gain is 0.7 electron/DN at ISO 400. On smaller pixel cameras, the gain reaches 1 at lower ISOs. > As far as your comments on low ISOs go, I think it is pretty clear that > even though 12-bit digitization is too coarse for ISO 100, that's not [quoted text clipped - 3 lines] > low digitization merely emphasizes the poor readout of absolute signal > levels at low ISOs. On my 1D Mark II, the noise analysis shows that the camera has a read noise of 3.9 electron at ISO 1600 at a gain of 0.8 electrons/DN. At ISO 100, the gain is 13 electrons/DN, so with noise in the A-D converter of around 1 DN, one would expect a little more then 13 electrons of read noise. The measured read noise is 16.6, close to prediction, and mostly due to quantization. See: http://www.clarkvision.com/imagedetail/evaluation-1d2 To further illustrate the gain effects, at ISO 800 the read noise is 4.04 e/DN, so close to the iSP 1600 value of 3.90. At ISO 3200, the noise was measured at 3.93 e/DN, so no improvement over ISO 1600. I'm now doing maximum ISO at 1600, and usually 800 as you get higher dynamic range at ISO 800 than 1600 or 3200, with essentially the same noise characteristics. Roger > noise is 4.04 e/DN, so close to the iSP 1600 value of 3.90. In case anyone was confused by my typo, iSP should have been ISO >AFAIK, only the D200 (and maybe the D2X) do this before writing the RAW >data. Nope. The D200's RAW data for a daylight shot has a cyan-green cast, just like most other DLSRs. It is in the native WB of the camera, and has not been altered. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>AFAIK, only the D200 (and maybe the D2X) do this before writing the RAW >>data. > > Nope. The D200's RAW data for a daylight shot has a cyan-green cast, > just like most other DLSRs. It is in the native WB of the camera, and > has not been altered. Of course it has more green but it seems to be altered in the A/D conversion if I'm reading this correctly: http://www.nikonimaging.com/global/technology/scene/10/index_02.htm "With conventional digital cameras, reproduction of gradation depends solely on the results of digital measurement, eventually causing gaps in gradation. To minimize such gaps in gradation, the D2X introduces our unconventional approach of analog adjustment for white balance in the first stage of signal processing." You can argue that is trying to compensate for a lesser sensor but they wouldn't do it if it didn't help improve performance. >> Nope. The D200's RAW data for a daylight shot has a cyan-green cast, >> just like most other DLSRs. It is in the native WB of the camera, and >> has not been altered. >Of course it has more green but it seems to be altered in the A/D >conversion if I'm reading this correctly: >http://www.nikonimaging.com/global/technology/scene/10/index_02.htm >"With conventional digital cameras, reproduction of gradation depends >solely on the results of digital measurement, eventually causing gaps in >gradation. To minimize such gaps in gradation, the D2X introduces our >unconventional approach of analog adjustment for white balance in the >first stage of signal processing." >You can argue that is trying to compensate for a lesser sensor but they >wouldn't do it if it didn't help improve performance. The D200 and the D2X are different cameras; you are assuming that they do WB the same. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > The D200 and the D2X are different cameras; you are assuming that they > do WB the same. How is it different? I'm interesteed in the D200. >> The D200 and the D2X are different cameras; you are assuming that they >> do WB the same. > >How is it different? >I'm interesteed in the D200. I said the D200 doesn't WB before writing the RAW. You quoted something from Nikon about the D2X. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > Nope. The D200's RAW data for a daylight shot has a cyan-green cast, > just like most other DLSRs. It is in the native WB of the camera, and > has not been altered. Per definition, the RAW data can't have a "cast" at all, as it always needs to be viewed using a propriety algorithm (using additional parameters), specified for that Camera type and that picture. -Michael >> Nope. The D200's RAW data for a daylight shot has a cyan-green cast, >> just like most other DLSRs. It is in the native WB of the camera, and >> has not been altered. >Per definition, the RAW data can't have a "cast" at all, as it always >needs to be viewed using a propriety algorithm (using additional >parameters), specified for that Camera type and that picture. That is not true. RAW data is not invisible, cryptical data in a black box, except in many people's minds. It is a linear bitmap, with red, green and blue filters understood to be over the sensor pixels. At any given ISO, a different amount of photons within the passed band are needed to reach RAW saturation (or sensor saturation), based on the color filter. Digitals average somewhere around 60% as sensitive to red as green, and 75% as sensitive to blue than green. This means that white light records in a ratio of about 60:100:75 (R:G:B), AKA cyan-green. It means that a purplish-magenta records at about 100:100:100 (white RAW data). Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >But the data gathered should be more useful because it's already closer >to where you want to be (if the WB was set right). I do appreciate your >comment in another thread about how doing high-bit raw conversion is >actually less work because less is screwed up but doesn't this achieve >the same objective? I think he was talking about manipulating the already-digitized data, as integers. That serves no real purpose that can't be done in a better way. The rounding errors add imprecision and posterization noise (which may not be especially visible in normal use). I've seen this in the RAW data of at least one Nikon; every 7th RAW value was absent in one channel, and every 6th was missprobably be more precise if it were simply multiplied by 6 and divided by 7 for the former, and multiplied by 5 and divided by 6 for the latter, to return it to its original state. Integer scaling is only reversible when the scalar is greater than 1. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >I've seen this in the RAW >data of at least one Nikon; every 7th RAW value was absent in one >channel, and every 6th was missprobably be more precise if it were >simply multiplied by 6 and divided by 7 for the former, and multiplied >by 5 and divided by 6 for the latter, to return it to its original >state. It should have read: I've seen this in the RAW data of at least one Nikon; every 7th RAW value was absent in one channel, and every 6th was missing in the others. The RAW data would probably be more precise if it were simply multiplied by 6 and divided by 7 for the former, and multiplied by 5 and divided by 6 for the latter, to return them to their original state. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > I believe WB does effect the RAW capture. It's probably not real > critical in most cases but it's like EC for different colors and will > reduce or increase the data captured to have WB set correctly. This is > why Nikon was trying to encrypt their WB although it's not that > important, the RAW converter can guess the intended WB or an operator > can chose it manually. I don't believe this to be the case. The RAW data is what it is. Changing temperature is simply an algorithm applied during post processing. SignatureThomas T. Veldhouse Key Fingerprint: 2DB9 813F F510 82C2 E1AE 34D0 D69D 1EDC D5EC AED1 >>I believe WB does effect the RAW capture. It's probably not real >>critical in most cases but it's like EC for different colors and will [quoted text clipped - 5 lines] > I don't believe this to be the case. The RAW data is what it is. Changing > temperature is simply an algorithm applied during post processing. I guess I was thinking of the D2x with the WB protected. http://www.nikonimaging.com/global/technology/scene/10/index_02.htm "With conventional digital cameras, reproduction of gradation depends solely on the results of digital measurement, eventually causing gaps in gradation. To minimize such gaps in gradation, the D2X introduces our unconventional approach of analog adjustment for white balance in the first stage of signal processing." > Am I mistaking any of the features of my Camera, and how they are > applied to images? I think certain noise reduction choices also affect the RAW image. In particular, I am referring to "long exposure noise reduction". SignatureThomas T. Veldhouse Key Fingerprint: 2DB9 813F F510 82C2 E1AE 34D0 D69D 1EDC D5EC AED1 >> Am I mistaking any of the features of my Camera, and how they are >> applied to images? > >I think certain noise reduction choices also affect the RAW image. In >particular, I am referring to "long exposure noise reduction". AKA "dark fram subtraction". Absolutely correct, it does effect the raw data. SignatureJohn Bean >>>>> "John" == John Bean <waterfoot@gmail.com> writes: John> AKA "dark fram subtraction". Absolutely correct, it does John> effect the raw data. Well, it has an effect on the raw data, because it *a*ffects the raw data. Affect = verb Effect = noun SignatureRandal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095 <merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/> Perl/Unix/security consulting, Technical writing, Comedy, etc. etc. See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training! >>>>>> "John" == John Bean <waterfoot@gmail.com> writes: > [quoted text clipped - 5 lines] >Affect = verb >Effect = noun My English references all indicate that "effect" can be a noun or a verb. Perhaps your version of English differs. "Have a nice day" SignatureJohn Bean >>>>>>> "John" == John Bean <waterfoot@gmail.com> writes: >> [quoted text clipped - 10 lines] > My English references all indicate that "effect" can be a > noun or a verb. Perhaps your version of English differs. Time for someone to effect his escape... > My English references all indicate that "effect" can be a > noun or a verb. Perhaps your version of English differs. As a verb it has a different meaning from the verb "affect". SignatureJeremy | jeremy@exit109.com >>>>>>>"John" == John Bean <waterfoot@gmail.com> writes: >> [quoted text clipped - 8 lines] > My English references all indicate that "effect" can be a > noun or a verb. Perhaps your version of English differs. True, but your usage of effect was still wrong. Greg Signature"All my time I spent in heaven Revelries of dance and wine Waking to the sound of laughter Up I'd rise and kiss the sky" - The Mekons >>>>>>>>"John" == John Bean <waterfoot@gmail.com> writes: >>> [quoted text clipped - 10 lines] > >True, but your usage of effect was still wrong. Since this is a not an English language group it's not really something I want to debate, but my deliberate usage of "effects" affects the intended meaning. I'll admit the wording was sloppy but I was trying to convey that using dark frame subtraction effects [a change] to the raw data. Given the omission of the implied words it was probably a bad choice. Now can we get back to the subject in hand? ;-) SignatureJohn Bean >Since this is a not an English language group it's not >really something I want to debate, but my deliberate usage [quoted text clipped - 3 lines] >Given the omission of the implied words it was probably a >bad choice. As you originally wrote the sentence, it means that the very existence of the raw data results directly from the dark frame subtraction. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< The settings for the camera are sometimes used as a starting point for the Raw software, most software at least uses the White Balance, but Nikon Capture uses allmost all (or all) settings. This means a faster workflow for all pictures which do not need any postprocessing. Losing this feature, the 'camera' settings. Would make postprocessing a need for a larger number of the pictures. In General the settings should improve the pictures on average. So Statistically less pictures need postprocessing with the 'correct' settings. For all the pictures that need other settings than at the time of taking the picture Raw can be used. Most 'Slide' people should be familliar that you can't postprocess and so should use the right parameters when taking the pictures. Now you have the best off both worlds. Use the best parameters and still be able to postprocess where you have 'goofed' / 'didn't have the time', 'the situation was to difficult to judge in the field'. But different people use different workflows. So I can imagine that some people ignore all the settings and always start to work from scratch with the raw. (I don't). ben >I often see some religious discussions regarding certain camera > settings, and wonder why they go on at all in a world of RAW pictures. [quoted text clipped - 31 lines] > > -drew >This means a faster workflow for all pictures which >do not need any postprocessing. Losing this feature, >the 'camera' settings. Would make postprocessing >a need for a larger number of the pictures. ? All raw require post processing by definition. -- Ed Ruf Lifetime AMA# 344007 (Usenet2@EdwardG.Ruf.com) http://EdwardGRuf.com > ? All raw require post processing by definition. No they don't. Most camera's are capable of taking raw together with a jpg. So if the picture is not worth bothering about, or the jpg is sufficient the raw does not need any post processing. What I actually meant was that even if you want to use the raw, it doesn't need any postprocessing where you have to do work as an operator. You can just batch process all raws with the parameters as in the camera set. As I do mostly because the jpg in my camera is only a basic jpg. So the only thing I do with the raw is to make a fine jpg. The bottom line is, if the jpg is sufficient, you do not need the raw at all. > -- > Ed Ruf Lifetime AMA# 344007 (Usenet2@EdwardG.Ruf.com) > http://EdwardGRuf.com >> ? All raw require post processing by definition. > >No they don't. They do... >Most camera's are capable of taking raw together with a jpg. ...the in-camera jpeg is a product of the RAW. (snipp) >The bottom line is, if the jpg is sufficient, you do not need >the raw at all. That's true. Or I would say: If the jpeg is sufficient for your kind of publishing (web, print etc) you don't need the raw data. -espen Signaturehttp://www.seland.org/ >>Most camera's are capable of taking raw together with a jpg. > > ...the in-camera jpeg is a product of the RAW. this is a matter of definition. Most people rever to raw only when it is available to the user. We (that is me and a lot of other people) would not accept a camera which produces only jpg's to be advertised as a raw capable camera. So the word RAW is for me and most people reserved for the external use of the internal format of the camera. If you want to call every internal format of a camera RAW, there is the problem that whe do not have an adequate name for the external only RAW. Although language is not exact, I think on the word RAW there is at least some consences how people use the word. (The OP uses the word raw in this way). So I like to distinct between the word "internal format" (which every camera has at some point) and RAW which is only provided with the higher end camera's. > That's true. Or I would say: If the jpeg is sufficient for your kind > of publishing (web, print etc) you don't need the raw data. To get back to the discussion of the OP. With this last 'statement', paremeters of the camera are still inportant, for this jpg. And for automatic batchprocessing of the RAW. So many settings DO matter even when using RAW. Ben >>>Most camera's are capable of taking raw together with a jpg. >> [quoted text clipped - 18 lines] >(which every camera has at some point) and RAW which >is only provided with the higher end camera's. "internal format" just another word for jpeg? What "high end cameras?" A Canon 350D or Nikon D50 aren't very high end. Signature"I have been a witness, and these pictures are my testimony. The events I have recorded should not be forgotten and must not be repeated." -James Nachtwey- http://www.jamesnachtwey.com/ >>So I like to distinct between the word "internal format" >>(which every camera has at some point) and RAW which >>is only provided with the higher end camera's. > What "high end cameras?" > > A Canon 350D or Nikon D50 aren't very high end. My Canon S60 has raw. >"internal format" just another word for jpeg? No. >What "high end cameras?" > >A Canon 350D or Nikon D50 aren't very high end. They are, even if they're cheaper than other. -espen Signaturehttp://www.seland.org/ >>So I like to distinct between the word "internal format" >>(which every camera has at some point) and RAW which >>is only provided with the higher end camera's. > > "internal format" just another word for jpeg? No jpeg is derived from the internal format, jpeg is NOT an internal format. > What "high end cameras?" There is a difference between 'higher' and 'high'. A child which is 'higher' than it's brother does not have to be 'high'. So when I refer to 'higher' end camera's I do not neccesary mean high end camera's. There is a difference in 'high' and 'higher', I do not know how this is called in English or American. > A Canon 350D or Nikon D50 aren't very high end. No but they are higher than most point and shoots. Ben. >> A Canon 350D or Nikon D50 aren't very high end. >No but they are higher than most point and shoots. They may be. But as this discussion group is called rec.photo.digital.slr-systems, a 350D or D50 fall into the category of low-end examples. >>> A Canon 350D or Nikon D50 aren't very high end. >>No but they are higher than most point and shoots. > > They may be. But as this discussion group is called > rec.photo.digital.slr-systems, a 350D or D50 fall into > the category of low-end examples. In my original comment I wanted to make the distinction between camera's which have a RAW format and camera's which do not have a RAW format. Sorry I did not realise that some people in this group would only consider the DSLR's, while I was refering to all types off camera's. (Including point and shoot). --- So I like to distinct between the word "internal format" --- (which every camera has at some point) and RAW which --- is only provided with the higher end camera's. Sorry if I wasn't clear on that. ben > In my original comment I wanted to make the distinction > between camera's which have a RAW format and camera's > which do not have a RAW format. My 3.3MP Oly P&S does TIFF format but it's so slow as to be impractical, I think that's true for ZLR's as well. >> In my original comment I wanted to make the distinction >> between camera's which have a RAW format and camera's >> which do not have a RAW format. > >My 3.3MP Oly P&S does TIFF format but it's so slow as to be impractical, >I think that's true for ZLR's as well. TIFF is not RAW; TIFF is basically a JPEG without JPEG compression artifacts; a very inefficient file format compared to RAW. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>> In my original comment I wanted to make the distinction >>> between camera's which have a RAW format and camera's [quoted text clipped - 5 lines] >TIFF is not RAW; TIFF is basically a JPEG without JPEG compression >artifacts; a very inefficient file format compared to RAW. How about JPEG2000? >>TIFF is not RAW; TIFF is basically a JPEG without JPEG compression >>artifacts; a very inefficient file format compared to RAW. >How about JPEG2000? No. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>> In my original comment I wanted to make the distinction >>> between camera's which have a RAW format and camera's [quoted text clipped - 5 lines] >TIFF is not RAW; TIFF is basically a JPEG without JPEG compression >artifacts; a very inefficient file format compared to RAW. Tiff is far more than JPEG with out compression. http://www.digitalpreservation.gov/formats/fdd/fdd000022.shtml Why JPEG is bad... http://en.wikipedia.org/wiki/Jpeg note TIFF 24bit RGB JPEG 8bit. If you think TIFF is just JPEG with out compression artifacts you need to do more study. Signature"I have been a witness, and these pictures are my testimony. The events I have recorded should not be forgotten and must not be repeated." -James Nachtwey- http://www.jamesnachtwey.com/ >>TIFF is not RAW; TIFF is basically a JPEG without JPEG compression >>artifacts; a very inefficient file format compared to RAW. >Tiff is far more than JPEG with out compression. >http://www.digitalpreservation.gov/formats/fdd/fdd000022.shtml >Why JPEG is bad... >http://en.wikipedia.org/wiki/Jpeg This is all "JPEG compression". What you have before the compression is the same thing as a 24-bit TIFF. >note TIFF 24bit RGB JPEG 8bit. Cameras that offer both TIFF and JPEG use 24-bit color (8 per channel) for both. >If you think TIFF is just JPEG with out compression artifacts you need >to do more study. If you did a little more study yourself, you'd know that the cameras that offer both TIFF and JPEG differ only in JPEG compression. Same Color space used, basically the same transfer curve; the JPEG is merely lossily compressed. Both are in a very different class of data than the RAW data (which is generally not an output option with cameras that offer TIFF). Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >If you did a little more study yourself, you'd know that the cameras >that offer both TIFF and JPEG differ only in JPEG compression. Same [quoted text clipped - 3 lines] >Both are in a very different class of data than the RAW data (which is >generally not an output option with cameras that offer TIFF). You're doing fine, right up until that last parenthetical clause. Some cameras offer TIFF, some cameras offer RAW, and some offer both. Saying that RAW isn't generally available on cameras that offer TIFF is rather too broad a claim. (I don't actually know of any cameras that offer 16-bit-per-component TIFF, but I wouldn't be totally surprised to find this is an option on some of the studio scanning backs. But apart from that I think you're still safe in claiming that TIFF throws away some low-order data bits) >>If you did a little more study yourself, you'd know that the cameras >>that offer both TIFF and JPEG differ only in JPEG compression. Same [quoted text clipped - 9 lines] >Saying that RAW isn't generally available on cameras that offer TIFF >is rather too broad a claim. Hardly. Only a small minority that offer either offer the other as well. >(I don't actually know of any cameras that offer 16-bit-per-component >TIFF, but I wouldn't be totally surprised to find this is an option on >some of the studio scanning backs. But apart from that I think you're >still safe in claiming that TIFF throws away some low-order data bits) Most camera TIFFs are 8-bit per channel, and clip the RAW data about a stop below maximum in the green channel, and up to 2 stops in the red channel, for daylight WB. The resolution of the highlights is also lessened in conversion to 8-bit TIFF. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>Some cameras offer TIFF, some cameras offer RAW, and some offer both. >>Saying that RAW isn't generally available on cameras that offer TIFF >>is rather too broad a claim. > >Hardly. Only a small minority that offer either offer the other as >well. The majority of cameras don't offer TIFF at all - it's either JPEG only, or JPEG and RAW. I'd agree with you if you were claiming that TIFF isn't generally available on cameras that offer RAW. But that isn't what you said - you said that most cameras that offer TIFF don't offer RAW as an option. That isn't true for Nikon (who only offer TIFF, as far as I can see, on their D2 bodies - those have RAW), it doesn't seem to apply to Canon (who don't appear to offer TIFF), and it isn't true for Pentax (who offer TIFF and RAW on all their DSLRs, but apparently don't offer TIFF on their point-and-shoots). I'm sure there are some cameras that offer TIFF, but not RAW. Against those, though, are all the Pentax DSLRs, and the top-of-the-line Nikons. That's enough to make me question the claim that most cameras with TIFF support don't offer RAW as an output format. Not, mind you, that I think it's *sensible* to use TIFF if you have RAW (the file size is larger, and has less useful information content). >>>Some cameras offer TIFF, some cameras offer RAW, and some offer both. >>>Saying that RAW isn't generally available on cameras that offer TIFF [quoted text clipped - 12 lines] >it isn't true for Pentax (who offer TIFF and RAW on all their DSLRs, >but apparently don't offer TIFF on their point-and-shoots). The Majority of cameras don't offer RAW either. Signature"And he gave it for his opinion, that whoever could make two ears of corn, or two blades of grass, to grow upon a spot of ground where only one grew before, would deserve better of mankind, and do more essential service to his country,than the whole race of politicians put together." _Gullivers Travels_ Part ii. Chap. vii. "Voyage to Brobdingnag" Jonathan Swift 1667-1745 >>>>Some cameras offer TIFF, some cameras offer RAW, and some offer both. >>>>Saying that RAW isn't generally available on cameras that offer TIFF [quoted text clipped - 14 lines] > >The Majority of cameras don't offer RAW either. That would be the "JPEG only" cameras I mentioned above. In any case, while it's true, it's irrelevant to the discussion. >>> A Canon 350D or Nikon D50 aren't very high end. >>No but they are higher than most point and shoots. > >They may be. But as this discussion group is called >rec.photo.digital.slr-systems, a 350D or D50 fall into >the category of low-end examples. In terms of ruggedness, weather-proofing, buffer speed, number of ciustom function, AF capability, MPs etc, yes, but in terms of per-pixel imaging, there really isn't much of a direct correlation between price and quality of RAW data. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>>> A Canon 350D or Nikon D50 aren't very high end. >>>No but they are higher than most point and shoots. [quoted text clipped - 7 lines] >imaging, there really isn't much of a direct correlation between price >and quality of RAW data. BS! Look at the Canon 1DSMkII RAW and also Leap Aptus Backs and get back to us. There's a direct correlation between price and the quality of the RAW and it's all related to Sensor size. Signature"I have been a witness, and these pictures are my testimony. The events I have recorded should not be forgotten and must not be repeated." -James Nachtwey- http://www.jamesnachtwey.com/ >So I like to distinct between the word "internal format" >(which every camera has at some point) and RAW which >is only provided with the higher end camera's. A camera's internal format (the digital data through the pipeline after first pre-processing cycles) and raw is about the same, just that the raw-file have a file header, including meta-data. >So many settings DO matter even when using RAW. Still, very few. -espen Signaturehttp://www.seland.org/ >> ? All raw require post processing by definition. > >No they don't. Yes they do. You have to process the file. You might use all the in camera defaults in say Nikon Capture, but you still have to process the NEF file, even if it is to extract the jpg contained within. >What I actually meant was that even if you want to use the >raw, it doesn't need any postprocessing where you have to >do work as an operator. You can just batch process all raws >with the parameters as in the camera set. Well, that's NOT what you said. You're still having to process the raw files. -- Ed Ruf Lifetime AMA# 344007 (Usenet2@EdwardG.Ruf.com) http://EdwardGRuf.com Ed Ruf (REPLY to E-MAIL IN SIG!) wrote: > Yes they do. You have to process the file. You might use all the in camera > defaults in say Nikon Capture, but you still have to process the NEF file, > even if it is to extract the jpg contained within. On some Canon cameras, separate raw and jpeg files are stored. I preview the jpegs, and choose the best images for raw processing, so most images I never touch the raw file except to delete it or archive it. Roger >Ed Ruf (REPLY to E-MAIL IN SIG!) wrote: > [quoted text clipped - 6 lines] >so most images I never touch the raw file except to delete it or >archive it. Same on the Nikons if you want to store both. But one can just extract the embedded jpeg if you don't want to do that. Especially on the D70 this makes sense as the raw+jpeg option is limited to the lowest quality jpeg, sam as the embedded one, iirc. On the D200 you can choose to have higher quality jpegs stored with raw if you want to. In doing this you're making a conscious choice to give up storage space in the camera, instead of post processing later. The old "you can pay me now, or you can pay me later" choice. Pick the one which suits your needs. -- Ed Ruf Lifetime AMA# 344007 (Usenet2@EdwardG.Ruf.com) http://EdwardGRuf.com Ed Ruf (REPLY to E-MAIL IN SIG!) wrote: >>Ed Ruf (REPLY to E-MAIL IN SIG!) wrote: >> [quoted text clipped - 15 lines] > processing later. The old "you can pay me now, or you can pay me later" > choice. Pick the one which suits your needs. The advantage of separate raw and jpeg files is 1) no need for extraction, 2) fast and simple image viewing with no need for processing, and 3) disk storage systems like the Epson P2000, P400, nixview vista, and others will show jpeg images if separate but not raw or jpegs embedded in raw. On some canon cameras, you can select the jpeg quality as well as separate jpeg files. All cameras should have these option. Roger >>> ? All raw require post processing by definition. >> [quoted text clipped - 3 lines] > defaults in say Nikon Capture, but you still have to process the NEF file, > even if it is to extract the jpg contained within. No they don't, this is not a requirement, not by law, not by agreement. As explained under the "No they don't", there are lot's of situations where the NEF file doesn't get processed at all, other than being copied or deleted. >>What I actually meant was that even if you want to use the >>raw, it doesn't need any postprocessing where you have to [quoted text clipped - 3 lines] > Well, that's NOT what you said. You're still having to process the raw > files. Yes that's why I corrected my self with what I meant. Sorry I was not clear about admitting my error. (My error being that I didn't consider batch processing als postprocessing, which it is offcourse.). > -- > Ed Ruf Lifetime AMA# 344007 (Usenet2@EdwardG.Ruf.com) > http://EdwardGRuf.com All raw require post processing by definition. >>>No they don't. >> [quoted text clipped - 26 lines] >> Ed Ruf Lifetime AMA# 344007 (Usenet2@EdwardG.Ruf.com) >> http://EdwardGRuf.com Back to the original discussion of RAW versus JPEG raised by "dtype": To possibly avoid the unavoidable of more hair splitting by some of the newsgroup members: yes, you must "post process" RAW images because they need to be converted to produce a "publishable" print. And if you shoot hundreds of images per day, which format would you prefer to post process? My definition of "publishable" -- they're not snapshots for a family album nor low res images for the web -- they're images for editorial or fine art print. To attain those publishable images, would you prefer to use RAW of JPEG files? Do you (and the publisher/printer) want RAW or JPEG files? All the data or interpolated data? RAW seems to be the consensus answer. Thus a reasonable response to the original question of RAW versus JPEG would be to ask what is the intended use of the image. A bad analogy: Do you like dehydrated meat or the real thing? Personally, I like my meat RAW so I can cook/burn it to my own taste.....:) By the way, Pete, "rubbish" was a little harsh don't you think......methinks you must be a Nikon user??? Hee, hee...... Mike > Back to the original discussion of RAW versus JPEG raised by "dtype": To > possibly avoid the unavoidable of more hair splitting by some of the > newsgroup members: yes, you must "post process" RAW images because they > need to be converted to produce a "publishable" print. And if you shoot > hundreds of images per day, which format would you prefer to post process? Actually, to be clear, I never intended to compare RAW versus JPEG. In fact, my original message stated that JPG shooters should ignore the thread. :) My question was whether or not certain camera settings were at all relevant when shooting in RAW, such as color space, sharpness, whitebalance, etc. I've heard some good arguments why some folks might still find those useful, but also some good reasons why in certain workflows they might be ignored completely. All in all, it has been an informative discussion. As for my own workflow, when using something like Lightroom (Adobe beta) or Aperture (won't Apple ever come out with a Universal binary?), I may continue to ignore these settings completely, as long as I don't mind some processing of each picture, or automatic processing by the software. To each his own, but I have enjoyed the thread. -drew Mike Rooney wrote: >> Back to the original discussion of RAW versus JPEG raised by "dtype": To >> possibly avoid the unavoidable of more hair splitting by some of the [quoted text clipped - 23 lines] > > -drew Drew, you mean an original question/comment to this newsgroup was misinterpreted and the discussion may have gone off on a tangent??? Unbelievable!!! :) Sorry. Mike > All raw require post processing by definition. >>>> [quoted text clipped - 51 lines] > >Mike Concerning Nikon and Canon. At default settings, which cameras product JPEG images closer to their RAW images? In other words, which camera applies the least amount of modification, compression to JPEG images. Reason I asked is that I noticed of two separate cameras (not Canon or Nikon) the default compression for their "high resolution" JPEG files was hugely different, 2.7:1 versus nearly 8:1 which pretty much dictated shooting RAW exclusively with the camera using higher compression. |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2009 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.