Re: Histogram question
AFAIK, they average the channels. The only one showing separate channels is some Fuji. Regards Albano --- Glen [EMAIL PROTECTED] wrote: I've been told that a lot of DSLR cameras use only the green channel when calculating the histogram, instead of the full RGB data. Does the Pentax *istDS display a true RGB histogram, or not? It seems that there have been times when the histogram hasn't been entirely accurate for me. I got clipped highlights sometimes when the histogram didn't suggest any such thing. thanks, Glen Albano Garcia Photography Graphic Design http://www.albanogarcia.com.ar http://www.flaneur.com.ar __ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com
Re: Histogram question
At 12:11 AM 12/1/2005, David Mann wrote: On Dec 1, 2005, at 12:40 PM, Rob Studdert wrote: I just did a QD test using my *ist D and the camera histogram responds similarly to green and red. However what I didn't check is how the histogram is calculated, possibly if one channel is clipped and the others are low the aggregate RGB histogram will display as unclipped? The driver for my scanner tends to round low values to zero when displaying so I'm finding it very easy to blow out the highlights if I'm not careful. I think there is a good chance that the Pentax either rounds some of the low values to zero or perhaps doesn't combine three channels before providing the histogram results. I've noticed that sometimes the histogram will appear to indicate that the brightest value captured is a mid-tone, maybe 128 out of 256. However, if I adjust the white point lower in my editing software to correct the overly dark scale, small highlight areas will sometimes begin to clip, well before I have gotten anywhere near what the histogram says is the brightest value recorded. Sometimes, the clipping only occurs in one or two channels, which makes the highlight area look discolored. There might still be detail in the highlight from the one or two channels that didn't clip, but it still doesn't look good because of the color shift. In general, the histogram is a big help in setting exposure and contrast settings on the camera, but there are still some cases where I think it gives misleading information. Some sort of clipping indicator would be a nice addition to the camera's preview features, or possibly provide a numerical readout of the very brightest and darkest pixels captured in the image, as measured on a 0-255 scale. take care, Glen
Re: Histogram question
Albano Garcia [EMAIL PROTECTED] wrote: AFAIK, they average the channels. The only one showing separate channels is some Fuji. The Canon 1D-II and 1Ds-II will display separate R, G and B histograms. So will the Sigma SD10, if you're interested! ;-) -- Mark Roberts Photography and writing www.robertstech.com
Re: Histogram question
I got it from here: One issue that has come up is that with the possible exception of the Fuji S2 Pro all other current DSLRs only display overall luminosity histograms and not individual channel brightness. This means that it's possible to blow out one of the R G or B channels without realizing it. Here's what Thomas Knoll has to say on the matter... Full article here: http://www.luminous-landscape.com/tutorials/expose-right.shtml Regards Albano --- Mark Roberts [EMAIL PROTECTED] wrote: Albano Garcia [EMAIL PROTECTED] wrote: AFAIK, they average the channels. The only one showing separate channels is some Fuji. The Canon 1D-II and 1Ds-II will display separate R, G and B histograms. So will the Sigma SD10, if you're interested! ;-) -- Mark Roberts Photography and writing www.robertstech.com Albano Garcia Photography Graphic Design http://www.albanogarcia.com.ar http://www.flaneur.com.ar __ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com
Re: Histogram question
Albano Garcia [EMAIL PROTECTED] wrote: I got it from here: One issue that has come up is that with the possible exception of the Fuji S2 Pro all other current DSLRs only display overall luminosity histograms and not individual channel brightness. This means that it's possible to blow out one of the R G or B channels without realizing it. Here's what Thomas Knoll has to say on the matter... Full article here: http://www.luminous-landscape.com/tutorials/expose-right.shtml OK. But if you check the detailed reviews on a site like DP Review, you'll be able to actually see the three-channel histograms on the Canon 1D-II and 1Ds-II and Sigma SD10. --- Mark Roberts [EMAIL PROTECTED] wrote: Albano Garcia [EMAIL PROTECTED] wrote: AFAIK, they average the channels. The only one showing separate channels is some Fuji. The Canon 1D-II and 1Ds-II will display separate R, G and B histograms. So will the Sigma SD10, if you're interested! ;-) -- Mark Roberts Photography and writing www.robertstech.com
Re: Histogram question
Mark Roberts wrote: Albano Garcia [EMAIL PROTECTED] wrote: I got it from here: One issue that has come up is that with the possible exception of the Fuji S2 Pro all other current DSLRs only display overall luminosity histograms and not individual channel brightness. This means that it's possible to blow out one of the R G or B channels without realizing it. Here's what Thomas Knoll has to say on the matter... Full article here: http://www.luminous-landscape.com/tutorials/expose-right.shtml OK. But if you check the detailed reviews on a site like DP Review, you'll be able to actually see the three-channel histograms on the Canon 1D-II and 1Ds-II and Sigma SD10. Remember that the Fuji S2 is an old model, introduced 3-4 years ago and superceded by the S3 Pro last year. The review likely hasn't been updated to reflect the newer camera's capabilities. -Adam
Re: Histogram question
On 1 Dec 2005 at 10:22, Glen wrote: In general, the histogram is a big help in setting exposure and contrast settings on the camera, but there are still some cases where I think it gives misleading information. Some sort of clipping indicator would be a nice addition to the camera's preview features, or possibly provide a numerical readout of the very brightest and darkest pixels captured in the image, as measured on a 0-255 scale. I believe that the histogram display data is extracted from the embedded JPG file which is affected by camera settings. Until the histogram is a function of the RAW data the histogram or a clipping indicator is only going to be a guide not an absolute indicator. Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: Histogram question
At 06:15 PM 12/1/2005, Rob Studdert wrote: On 1 Dec 2005 at 10:22, Glen wrote: In general, the histogram is a big help in setting exposure and contrast settings on the camera, but there are still some cases where I think it gives misleading information. Some sort of clipping indicator would be a nice addition to the camera's preview features, or possibly provide a numerical readout of the very brightest and darkest pixels captured in the image, as measured on a 0-255 scale. I believe that the histogram display data is extracted from the embedded JPG file which is affected by camera settings. Until the histogram is a function of the RAW data the histogram or a clipping indicator is only going to be a guide not an absolute indicator. Okay, but I've had the histogram fail to show clipping when shooting in JPEG mode. If the histogram is based on JPEG mode, and the shooting mode is JPEG mode, then an ideal histogram should always indicate the presence of any clipping, shouldn't it? take care, Glen
Re: Histogram question
On 1 Dec 2005 at 17:46, Glen wrote: Okay, but I've had the histogram fail to show clipping when shooting in JPEG mode. If the histogram is based on JPEG mode, and the shooting mode is JPEG mode, then an ideal histogram should always indicate the presence of any clipping, shouldn't it? No I don't believe so, what you are viewing an aggregate RGB histogram, if there is clipping on any one channel it may not be displayed in the RGB histogram. Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Histogram question
I've been told that a lot of DSLR cameras use only the green channel when calculating the histogram, instead of the full RGB data. Does the Pentax *istDS display a true RGB histogram, or not? It seems that there have been times when the histogram hasn't been entirely accurate for me. I got clipped highlights sometimes when the histogram didn't suggest any such thing. thanks, Glen
Re: Histogram question
On Nov 30, 2005, at 13:11, Glen wrote: I've been told that a lot of DSLR cameras use only the green channel when calculating the histogram, instead of the full RGB data. Does the Pentax *istDS display a true RGB histogram, or not? It seems that there have been times when the histogram hasn't been entirely accurate for me. I got clipped highlights sometimes when the histogram didn't suggest any such thing. Sounds like something that could be checked by taking a full-frame photo of something that is entirely blue, entirely red, and entirely green. Assuming the metering is about 18% grey for all 3 targets, I would expect the histogram (if monochromatic) to look wildly different between the 3 photos. (with the understanding that the idea of pure green/red/blue targets is more theoretical than achievable, but it should be close enough to get the idea, doncha think?) -Charles -- Charles Robinson [EMAIL PROTECTED] Minneapolis, MN http://charles.robinsontwins.org
Re: Histogram question
On 30 Nov 2005 at 14:11, Glen wrote: I've been told that a lot of DSLR cameras use only the green channel when calculating the histogram, instead of the full RGB data. Does the Pentax *istDS display a true RGB histogram, or not? It seems that there have been times when the histogram hasn't been entirely accurate for me. I got clipped highlights sometimes when the histogram didn't suggest any such thing. I just did a QD test using my *ist D and the camera histogram responds similarly to green and red. However what I didn't check is how the histogram is calculated, possibly if one channel is clipped and the others are low the aggregate RGB histogram will display as unclipped? Rob Studdert HURSTVILLE AUSTRALIA Tel +61-2-9554-4110 UTC(GMT) +10 Hours [EMAIL PROTECTED] http://members.ozemail.com.au/~distudio/publications/ Pentax user since 1986, PDMLer since 1998
Re: Histogram question
On Dec 1, 2005, at 12:40 PM, Rob Studdert wrote: I just did a QD test using my *ist D and the camera histogram responds similarly to green and red. However what I didn't check is how the histogram is calculated, possibly if one channel is clipped and the others are low the aggregate RGB histogram will display as unclipped? The driver for my scanner tends to round low values to zero when displaying so I'm finding it very easy to blow out the highlights if I'm not careful. The preview window is nice and big so it's not too much of a problem... just a little annoying. - Dave
Re: Histogram question
I would expect that to be true of all digital cameras as exposure (luminance) is taken only from the 2 green pixels (each interpolated pixel is made up of two green and one each red and blue) which most closely match the response of the human eye. (Disclaimer: the above is vastly simplified, if you want a detailed 5 paragraph explanation someone else will be glad to supply it) graywolf http://www.graywolfphoto.com Idiot Proof == Expert Proof --- Charles Robinson wrote: On Nov 30, 2005, at 13:11, Glen wrote: I've been told that a lot of DSLR cameras use only the green channel when calculating the histogram, instead of the full RGB data. Does the Pentax *istDS display a true RGB histogram, or not? It seems that there have been times when the histogram hasn't been entirely accurate for me. I got clipped highlights sometimes when the histogram didn't suggest any such thing. Sounds like something that could be checked by taking a full-frame photo of something that is entirely blue, entirely red, and entirely green. Assuming the metering is about 18% grey for all 3 targets, I would expect the histogram (if monochromatic) to look wildly different between the 3 photos. (with the understanding that the idea of pure green/red/blue targets is more theoretical than achievable, but it should be close enough to get the idea, doncha think?) -Charles -- Charles Robinson [EMAIL PROTECTED] Minneapolis, MN http://charles.robinsontwins.org
