I've been evaluating the new 20D and deciding if I want to keep it over my 10D. I love the 20D's speed without a doubt but am also very concerned with image quality. I just took both of these shots with the same lens (17-40L) on a tripod at f22. Both were on auto white balance, AV mode, and ISO100. Both are 100% crops and not edited. Which one do you think is better? Care to guess which is the 10D and which is the 20D? - todd
Both images are the same size. You must have changed the position of the zoom or camera to compensate for the extra pixels in the 20D, so this could effect the outcome.
The zoom may have slighly moved but that's it. I checked that both were wide open. So which do you think is sharper or do you think both are close? The top one, which I won't say is which camera yet, overexposed the pic just a little or the bottom one underexposed.
One thing, though. The f/22 is not the best setting for 17-40, the difraction is really starting to show up. I think for a test like this you want to use a lens at its max resolution so the overall resolution wouldn't depend on the lens but rather on the camera. In this case, I'm afraid (and I might be wrong) the both cameras might outperform the lens. I would use the 17-40 at f5.6 or f/8
"A" is the 10D and "B" the 20D. I"m just concerned that the 20D is not as sharp as the 10. The 20 seems to do better on WB and exposure, but I'm almost thinking that the 10 is sharper. I've been shooting with both all afternoon and trying to decypher. Which one do I keep? Ahhh......
The more accurate color is "B" the 20D. Just wondering which is a better image. I know you guys/gals can't tell from two images, but I appreicate the comments.
B (20D) is much better, more dynamic range and accurate color balance. I don't think there is a noticable difference in sharpness, but I agree that shooting at f/22 will cause diffraction, and the 20D will therefor look softer at 100%, as it would in any situation where the lens is not resolving to the limits of the sensor. Shoot your 50/1.8 at f/11 and you'll have a much better comparison.
One thing I've noticed is that with the same lens and ISO the 20D has about a 1 stop advantage in metering. It's noticeably louder than the 10D but so much quicker in operation.
Here's another unedited 100% crop and EXIF. I don't think that's too bad for sharpness. 20D. Of course smugmug is reducing the image a bit:
Make - Canon
Model - Canon EOS 20D
Orientation - Top left
XResolution - 72
YResolution - 72
ResolutionUnit - Inch
DateTime - 2004:10:09 15:57:29
YCbCrPositioning - Co-Sited
ExifOffset - 196
ExposureTime - 1/250 seconds
FNumber - 5.60
ExposureProgram - Shutter priority
ISOSpeedRatings - 800
ExifVersion - 0221
DateTimeOriginal - 2004:10:09 15:57:29
DateTimeDigitized - 2004:10:09 15:57:29
ComponentsConfiguration - YCbCr
ShutterSpeedValue - 1/250 seconds
ApertureValue - F 5.60
ExposureBiasValue - 0.33
MeteringMode - Multi-segment
Flash - Not fired, compulsory flash mode
FocalLength - 17 mm
UserComment -
FlashPixVersion - 0100
ColorSpace - sRGB
ExifImageWidth - 3504
ExifImageHeight - 2336
InteroperabilityOffset - 9250
FocalPlaneXResolution - 2886.36
FocalPlaneYResolution - 2885.81
FocalPlaneResolutionUnit - Inch
CustomRendered - Normal process
ExposureMode - Auto
WhiteBalance - Auto
SceneCaptureType - Standard
Maker Note (Vendor): -
Macro mode - Normal
Self timer - Off
Quality - Fine
Flash mode - Not fired
Sequence mode - Single or Timer
Focus mode - One-Shot
Image size - Large
Easy shooting mode - Manual
Digital zoom - None
Contrast - High , +1
Saturation - High , +1
Sharpness - High , +1
ISO Value - 32767 (other)
Metering mode - Evaluative
Focus type - Auto
AF point selected -
Exposure mode - Tv-priority
Focal length - 17 - 40 mm
Flash activity - Not fired
Sequence number - 0
White Balance - Auto
Image Type - Canon EOS 20D
Firmware Version - Firmware 1.0.5
Owner Name - unknown (
Camera Serial Number - 420112606
If true that means ISO 100 on the 20D is like using ISO 200 on the 10D, and so on.
If ISO 800 on the 20D meters same as 1600 on the 10D, and you add the superior low noise performance of the 20D, does that mean that the 20D can shoot over 1 stop faster than the 10D and produce a comparable or superior image? Looking at DPReview's 20D summary, the additional brightness of the 20D at the same settings is quite apparent, as is the superior low-noise characteristics of ISO 400-1600. (http://www.dpreview.com/articles/canoneos20d/page15.asp)
Especially interesting is that D20 noise levels at ISO 1600 match the 10D at ISO 800. If there already is a 1 stop advantage on the 20D, that means that shooting ISO 1600 on the 20D is like shooting ISO 3200 on the 10D yet getting the image quality of 10D ISO 800.
Indeed one explanation could be that the 20D has an extra stop of dynamic range in the highlights, so the overall image is 1 stop brighter despite being properly exposed. If true, amazing.
Tubby wrote:
One thing I've noticed is that with the same lens and ISO the 20D has about a 1 stop advantage in metering. It's noticeably louder than the 10D but so much quicker in operation.
I also don't see any significant difference in sharpness between the two. Right from the start (before I knew which was which), I knew I liked the B image better.
I was never displeased with my 10D's images. And I still wouldn't be. It's just that I prefer what my new 20D gives me even more.
B is better image IMO. Aside from better exposure and more accurate color balance, although it appears colder, it is sharper too ... just look at the + sign on the calculator.
