If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
noise levels Canon g6 -Oly 8080
anyone that knows how those camereas compare to each other regarding to
image noise? thinking of "real" iso values, seems like G6 is 320 at 200 and 640 at 400.... |
#2
|
|||
|
|||
In article ,
"-T-" wrote: anyone that knows how those camereas compare to each other regarding to image noise? thinking of "real" iso values, seems like G6 is 320 at 200 and 640 at 400.... www.dpreview.com has noise analysis and sample photos in their reviews. http://www.dpreview.com/reviews/canong6/page14.asp http://www.dpreview.com/reviews/olym...0wz/page15.asp Hop to the next page to see actual night photos. The one thing missing is temperature sensitivity. Some are very bad in warm climates. On a hot summer night, my old Oly 4040 would take a picture of pure snow at 1/4 second, ISO 400. |
#3
|
|||
|
|||
In message ,
Kevin McMurtrie wrote: www.dpreview.com has noise analysis and sample photos in their reviews. Using the camera's arbitrary exposure, and arbitrary ISO values. The real test would be with manual exposure, normalizing the exposure in the output. We need comparisons of two cameras shooting the same light at the same aperture and shutter speed, not two cameras set to the same "ISO number". The files should then be "leveled" so that the average level is the same, for comparing noise. Of course, this leaves a choice of "ISO values" to compare, but those could be done in slices in the graphic. -- John P Sheehy |
#5
|
|||
|
|||
In article ,
wrote: In message , Kevin McMurtrie wrote: www.dpreview.com has noise analysis and sample photos in their reviews. Using the camera's arbitrary exposure, and arbitrary ISO values. The real test would be with manual exposure, normalizing the exposure in the output. We need comparisons of two cameras shooting the same light at the same aperture and shutter speed, not two cameras set to the same "ISO number". The files should then be "leveled" so that the average level is the same, for comparing noise. Of course, this leaves a choice of "ISO values" to compare, but those could be done in slices in the graphic. If you're looking for the best photo quality with low sensor illumination, I don't think you'd want a point'n'shoot. None of them are very good. They use strong software noise filtering which does severe damage to the details of a photo. Simply comparing noise levels won't work. You'll have to compare error levels to a very clean reference photo. Trees, grass, roof shingles, and bricks are common victims of noise filtering. The Canon 20D is probably today's champion when it comes to image quality with low sensor illumination. How much light hits the sensor depends on what lens you put on it. |
#6
|
|||
|
|||
In article ,
wrote: In message , Kevin McMurtrie wrote: www.dpreview.com has noise analysis and sample photos in their reviews. Using the camera's arbitrary exposure, and arbitrary ISO values. The real test would be with manual exposure, normalizing the exposure in the output. We need comparisons of two cameras shooting the same light at the same aperture and shutter speed, not two cameras set to the same "ISO number". The files should then be "leveled" so that the average level is the same, for comparing noise. Of course, this leaves a choice of "ISO values" to compare, but those could be done in slices in the graphic. If you're looking for the best photo quality with low sensor illumination, I don't think you'd want a point'n'shoot. None of them are very good. They use strong software noise filtering which does severe damage to the details of a photo. Simply comparing noise levels won't work. You'll have to compare error levels to a very clean reference photo. Trees, grass, roof shingles, and bricks are common victims of noise filtering. The Canon 20D is probably today's champion when it comes to image quality with low sensor illumination. How much light hits the sensor depends on what lens you put on it. |
#7
|
|||
|
|||
In article ,
wrote: In message , Kevin McMurtrie wrote: www.dpreview.com has noise analysis and sample photos in their reviews. Using the camera's arbitrary exposure, and arbitrary ISO values. The real test would be with manual exposure, normalizing the exposure in the output. We need comparisons of two cameras shooting the same light at the same aperture and shutter speed, not two cameras set to the same "ISO number". The files should then be "leveled" so that the average level is the same, for comparing noise. Of course, this leaves a choice of "ISO values" to compare, but those could be done in slices in the graphic. If you're looking for the best photo quality with low sensor illumination, I don't think you'd want a point'n'shoot. None of them are very good. They use strong software noise filtering which does severe damage to the details of a photo. Simply comparing noise levels won't work. You'll have to compare error levels to a very clean reference photo. Trees, grass, roof shingles, and bricks are common victims of noise filtering. The Canon 20D is probably today's champion when it comes to image quality with low sensor illumination. How much light hits the sensor depends on what lens you put on it. |
#8
|
|||
|
|||
In message ,
Kevin McMurtrie wrote: If you're looking for the best photo quality with low sensor illumination, I don't think you'd want a point'n'shoot. None of them are very good. They use strong software noise filtering which does severe damage to the details of a photo. Simply comparing noise levels won't work. You'll have to compare error levels to a very clean reference photo. Trees, grass, roof shingles, and bricks are common victims of noise filtering. I forget about those cameras sometimes. I was thinking more along the lines of RAW data, actually. And of course, a test that included detail at various levels of resolution would be even better than a grey card. The 20D has very low noise at high ISOs, and manages to show detail very well through whatever noise is there, but that may be because the anti-aliasing filter in the 20D is a tad too weak; I can make it do color moire at will, although I have never seen it in any "real" pictures I've taken: http://www.pbase.com/jps_photo/image/37173876 This is an LCD laptop screen, from about 5 feet away with a very sharp 90mm lens. -- John P Sheehy |
#9
|
|||
|
|||
In article , -T- says...
anyone that knows how those camereas compare to each other regarding to image noise? thinking of "real" iso values, seems like G6 is 320 at 200 and 640 at 400.... If I'm not mistaken, the G6 uses a 7MP 1/1.8" CCD, while the 8080 uses an 8MP 2/3" CCD. Therefore the pixels of the 8080 are bigger, so noise levels should be lower. -- Alfred Molon ------------------------------ http://groups.yahoo.com/group/Olympus_405080/ Olympus 5060 resource - http://myolympus.org/5060/ Olympus 8080 resource - http://myolympus.org/8080/ |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Canon 20D lenses: Canon vs Sigma | Alex Vilner | Digital Photography | 169 | October 10th 04 10:11 PM |
Canon 20D noise levels | Jerry | Digital Photography | 36 | September 18th 04 06:04 AM |
Software that determines noise levels | Justin F. Knotzke | Digital Photography | 5 | September 10th 04 03:15 AM |
CANON - The Great Innovator (was: CANON – The Great Pretender) | Steven M. Scharf | Digital Photography | 104 | September 3rd 04 01:01 PM |
Canon 10d or Nikon D70. | Dmanfish | Digital Photography | 102 | August 18th 04 12:26 PM |