 |
 | 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 / July 2005Canon's are not noisless
It looks like you need very specific conditions to avoid noise, even with Canon DSLRs. Take a look at this shot. Look at the green areas. This is 400ISO with a Rebel XT. The shot exposure seens right on, yet I clearly see noticeable noise in the green. Given the small size of the image (relative to the native image size out of the camera) it is very noticeable. Is the green channel the main contributor to the noise? http://www.pbase.com/myirwin/image/46013664 > It looks like you need very specific conditions to avoid noise, > even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 3 lines] > camera) it is very noticeable. > Is the green channel the main contributor to the noise? It's hard to say with only a reduced-size JPEG, but what you're seeing in the green areas looks to me at first blush like shadow clipping in one color channel. On the other hand, if you consider that shot to be "noisy", I think you're better off waiting on digital photography entirely for a few more years. :) I don't find it objectionable at all.  SignatureJeremy | jeremy@exit109.com > It looks like you need very specific conditions to avoid noise, > even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 5 lines] > > http://www.pbase.com/myirwin/image/46013664 This one here is even noisier. http://www.pbase.com/myirwin/image/46013661 Something's is strange because this shot http://www.pbase.com/rkircher/image/45323916 was shot at 800 and this shot http://www.pbase.com/rkircher/image/45360178 was at 1600 and they both look less noisy to me. This one http://www.pbase.com/rkircher/image/45323915 was at 1600 and where it's noisy as you'd expect it doesn't seem proportionally noisy compared to the sample you posted. Anyway here you can see the noise is real bad in the blue and black areas so I don't think Green has anything to do with it. All these shots were done at night and I'd done no Noise reduction. BTW the original shots were done with a 300D just like my shots. -- Rob >> It looks like you need very specific conditions to avoid noise, >> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 26 lines] > > Rob My guess was that it was underexposed at the outset, and pulled out, that seems to add to the noise levels, in my experience. SignatureSkip Middleton http://www.shadowcatcherimagery.com >>> It looks like you need very specific conditions to avoid noise, >>> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 30 lines] > My guess was that it was underexposed at the outset, and pulled out, that > seems to add to the noise levels, in my experience. The OPs sample image appears as if they was taken on a rainy day so you may very well be correct. I took a look at some of the other galleries, however, and they too have pics that are very noisy at 400ISO. I'm by no means a pro here but it seems to me that I'm getting better results across the ISO range then what's been posted. Makes me thing it may be a post processing issues. -- Rob >> It looks like you need very specific conditions to avoid noise, >> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 21 lines] > >BTW the original shots were done with a 300D just like my shots. Your third shot seems to have more moderate contrast than the first two as well. -Rich >> It looks like you need very specific conditions to avoid noise, >> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 7 lines] > >This one here is even noisier. http://www.pbase.com/myirwin/image/46013661 Has anyone noticed that the motion blur in the background is *exactly* horizontal? There isn't a trace of any deviation at all. Perhaps the panning didn't blur the background enough, so someone added motion blur in software, and then added some noise to make it look real. The first thing I noticed about these images was the perfect horizontal streaks all over the background. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > Has anyone noticed that the motion blur in the background is *exactly* > horizontal? There isn't a trace of any deviation at all. Perhaps the [quoted text clipped - 3 lines] > The first thing I noticed about these images was the perfect horizontal > streaks all over the background. That would seem to me to be consistent with perfectly horizontal panning, no? IAE, this whole thing is almost a troll, and I can't possibly refute the anti-canonical, so I will point out that the plural of Canon is: Canons.(!) SignatureJohn McWilliams >> Has anyone noticed that the motion blur in the background is *exactly* >> horizontal? There isn't a trace of any deviation at all. Perhaps the >> panning didn't blur the background enough, so someone added motion blur >> in software, and then added some noise to make it look real. >> The first thing I noticed about these images was the perfect horizontal >> streaks all over the background. >That would seem to me to be consistent with perfectly horizontal >panning, no? No. There should be some deviation. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > > [quoted text clipped - 12 lines] > > No. There should be some deviation. Why ? I am not doubting your word, just wanting to understand. john >>>>The first thing I noticed about these images was the perfect horizontal >>>>streaks all over the background. >>>That would seem to me to be consistent with perfectly horizontal >>>panning, no? >> No. There should be some deviation. >Why ? I am not doubting your word, just wanting to understand. Why do you think that there is a reasonable chance of perfectly horizontal motion for that many pixels? The horizontal streaks are *perfect* in these images. Not even 1 degree of deviation. No up-and-down jitter along the streaks. Like this: green -----------------------------------------> green white -----------------------------------------> white green -----------------------------------------> green Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< > It looks like you need very specific conditions to avoid noise, > even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 5 lines] > > http://www.pbase.com/myirwin/image/46013664 No, I've found that the blue channel is the main culprit, as did DPReview. SignatureSkip Middleton http://www.shadowcatcherimagery.com >> It looks like you need very specific conditions to avoid noise, >> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 7 lines] > >No, I've found that the blue channel is the main culprit, as did DPReview. Personally, since it's background, I don't find the noise that objectionable, just noticeable. -Rich >>> It looks like you need very specific conditions to avoid noise, >>> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 11 lines] > objectionable, just noticeable. > -Rich Neither did I, in fact, had my attention not been drawn to it, I wouldn't have noticed. SignatureSkip Middleton http://www.shadowcatcherimagery.com >> It looks like you need very specific conditions to avoid noise, >> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 7 lines] > >No, I've found that the blue channel is the main culprit, as did DPReview. DPReview evaluated that image? Or, are you referring to noise in general? Noise is usually strongest in either the red or blue channels because those usually have a weakest signal in the exposure, and the wwhite balancing amplifies there noise and they are highly posterized, to boot. Green is not usually a problem, because most common light sources have ample green, and the green-filtered bayer sensels are up to a stop more sensitive than the other channels. If you had a light source very lacking in green, the green channel would be noisy. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>> It looks like you need very specific conditions to avoid noise, >>> even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 18 lines] > other channels. If you had a light source very lacking in green, the > green channel would be noisy. Noise in general. SignatureSkip Middleton http://www.shadowcatcherimagery.com > It looks like you need very specific conditions to avoid noise, > even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 5 lines] > > http://www.pbase.com/myirwin/image/46013664 It says you're shooting with a 300D...considerably nosier then an XT especially at higher isos. SNIP > The shot exposure seens right on, yet I clearly see noticeable > noise in the green. I have no idea of the kind of postprocessing the image went through, so exposure could have been off. > Given the small size of the image (relative to the native image > size out of the camera) it is very noticeable. Besides the possibility of it being a crop, down-sampling (unless done well) will exaggerate noise. Bart >Besides the possibility of it being a crop, down-sampling (unless done >well) will exaggerate noise. The image looks like it was either downsized with nearest neighbor, or downsampled and then highly sharpened. The motion-blur looks suspiciously artificial, as well. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< >>Besides the possibility of it being a crop, down-sampling (unless done >>well) will exaggerate noise. > >The image looks like it was either downsized with nearest neighbor, or >downsampled and then highly sharpened. The motion-blur looks >suspiciously artificial, as well. 1/125 sec panning with a speeding racecar sure sounds like it would produce it's own motion blur so why would the guy need to fake it? As far as the jagged lines, just downsizing may be enough to produce that effect without sharpening, I think. I can ask the poster and see what he says. -Rich Seriously Rich I think you need to get a life, background noise in a panning shot is desirable. > It looks like you need very specific conditions to avoid noise, > even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 5 lines] > > http://www.pbase.com/myirwin/image/46013664 > Seriously Rich I think you need to get a life, background noise in a > panning shot is desirable. I was going to add to the comments above that the noise does add to the motion of the shot. -- Rob >> Seriously Rich I think you need to get a life, background noise in a >> panning shot is desirable. > >I was going to add to the comments above that the noise does add to the >motion of the shot. And yet somehow, I doubt I'd be reading that if the camera were other than a Canon. -Rich >>> Seriously Rich I think you need to get a life, background noise in a >>> panning shot is desirable. [quoted text clipped - 5 lines] > than a Canon. > -Rich You started it! > "RichA" <none@none.com> wrote in message >> And yet somehow, I doubt I'd be reading that if the camera were other >> than a Canon. > > You started it! Cogent mentation is not RichA's strong point. >>> Seriously Rich I think you need to get a life, background noise in a >>> panning shot is desirable. [quoted text clipped - 4 lines] > And yet somehow, I doubt I'd be reading that if the camera were other > than a Canon. Exactly. If it was an olympus they'd be saying "That would have been a great shot if not for all the noise"... SignatureStacey >>>> Seriously Rich I think you need to get a life, background noise in a >>>> panning shot is desirable. [quoted text clipped - 8 lines] > great > shot if not for all the noise"... Can't resist, can you? SignatureSkip Middleton http://www.shadowcatcherimagery.com >>>>> Seriously Rich I think you need to get a life, background noise in a >>>>> panning shot is desirable. [quoted text clipped - 10 lines] > > Can't resist, can you? With an Oly the startup time would have caused you to miss the shot altogether. :-) >>> Exactly. If it was an olympus they'd be saying "That would have been a >>> great [quoted text clipped - 4 lines] > With an Oly the startup time would have caused you to miss the shot > altogether. :-) LOL!!! -- Rob >>>> Exactly. If it was an olympus they'd be saying "That would have been a >>>> great [quoted text clipped - 6 lines] > > LOL!!! I feel bad but it is the truth, they suck at startup. > -- > > Rob >>>>> Exactly. If it was an olympus they'd be saying "That would have been a >>>>> great [quoted text clipped - 8 lines] > > I feel bad but it is the truth, they suck at startup. I laugh because my 300D is slow at startup too but I don't think I've missed a shot because of it. Besides it was a funny comment in a thread where a few seem to just want to stir up sh##. -- Rob >>>>>> Stacey >>>>> [quoted text clipped - 14 lines] > > Rob There is some pretty inane rubbish. ;-) >>>>>> Seriously Rich I think you need to get a life, background noise in a >>>>>> panning shot is desirable. [quoted text clipped - 13 lines] >With an Oly the startup time would have caused you to miss the shot >altogether. :-) You mean people wait for the cars to round a bend before turning on their cameras? >>With an Oly the startup time would have caused you to miss the shot >>altogether. :-) > > You mean people wait for the cars to round a bend before turning on > their cameras? Sure don't you? I stand there with the lens in my pocket and the camera off waiting.... SignatureStacey >>>>> Seriously Rich I think you need to get a life, background noise in a >>>>> panning shot is desirable. [quoted text clipped - 13 lines] > > Can't resist, can you? It's the truth. If this was posted as being from a sigma, they'd be saying the colors are off etc. The people here look at the exif data, then they make their coments.. SignatureStacey >>> Exactly. If it was an olympus they'd be saying "That would have been a >>> great [quoted text clipped - 8 lines] > the colors are off etc. The people here look at the exif data, then they > make their coments.. I didn't. I just looked at the image. After all, that's what's important, isn't it? SignatureSkip Middleton http://www.shadowcatcherimagery.com >>"Stacey" <fotocord@yahoo.com> wrote in message >>> [quoted text clipped - 7 lines] > the colors are off etc. The people here look at the exif data, then they > make their coments.. "The people here"..... speak for yourself. The day you can avoid being defensive or offensive about the equipment anyone uses will be a brighter day. SignatureJohn McWilliams Coach: "Are you just ignorant, or merely apathetic?" Player: "Coach, I don't know, and I don't care." >Seriously Rich I think you need to get a life, background noise in a panning >shot is desirable. [quoted text clipped - 8 lines] >> >> http://www.pbase.com/myirwin/image/46013664 Why? Isn't a blurred background enough? >>Seriously Rich I think you need to get a life, background noise in a >>panning [quoted text clipped - 11 lines] > > Why? Isn't a blurred background enough? If I had the original untouched photo then I might feel confident to make a comment, what was done in post production and with what tools? >It looks like you need very specific conditions to avoid noise, >even with Canon DSLRs. Take a look at this shot. Look at the green [quoted text clipped - 5 lines] > >http://www.pbase.com/myirwin/image/46013664 Oversharpening looks like the source of visible noise here. Look how aliased the lettering and trim are. Signature<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<> John P Sheehy <JPS@no.komm> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>>< |
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.