IMPORTANT, please note ... whenever you visit a Pbase gallery for the first time, or if you have clicked on a size smaller than "original size" while visiting this or any other Pbase gallery, please be sure to click on "original" size, otherwise you will be presented with a low quality JPEG as versus the high quality JPEG (the size doesn't change, only the quality level of the JPEG compression); you only have to do this once, on the very first picture you look at.
Alternatively, on the very first picture you can add "/original" to the URL in the address bar and hit "refresh" or "go" as the case may be; again, you only have to do this once (on the very first picture). This "/original to the URL" approach is the sure-fire/no-mistake/best way to ensure you will see the highest quality JPEG presentation.
Both the 50D (firmware version 1.0.2) and 40D (firmware version 1.0.8) files have been interpolated to 4230x2820 pixels to simulate prints of equal paper size, and sharpened (after the interpolation) using Photoshop's (version 8.0) USM set to 200,0.2,0. The test target used was created to give an idea of both shadow and highlight detail in addition to high contrast resolution including for both positive and negative background transitions.
Note 1: Both cameras were tested using RAW format and developed via DPP. Using DPP's Preferences, both luminance and chrominance sliders have been set to zero; this was done prior to developing any files using the latest DPP (version 3.5.0.0). Also, the Contrast, Color Tone, Color Saturation, and Sharpness are set to 0, 0, 0, and 3, respectively. For both cameras, the Picture Style was set to Standard.
Note 2: Not that it matters but FWIW the 50D's in-camera "Auto Lighting Optimizer" has been disabled.
Note 3: The lens used in this test was a Canon EF 50mm f/1.4 USM. For both cameras, the lens was focused manually in Live View Mode.
Note 4: For both cameras, in-camera sharpness was set to "3" and all noise reduction was disabled.
Note 5: The 50D is very slightly overexposed relative to the 40D; to compensate for this the 50D files (and "only" the 50D files) have been adjusted by -0.2 EC.
Note 6: FWIW the object to sensor plane distance in this resolution test was 13.3 feet.
Joe thanks for the great service to the community. Your first class reviews are very useful to all of us. They answer with both real data we can see and well reasoned arguments. Thanks.
Guest
13-Oct-2008 19:42
Great job, have you done any AF comparisons? Especially AI servo tracking.
John
11-Oct-2008 15:19
Thanks a lot Joseph this is really cool. I have a 40D and was wondering just how good the higher ISOs were on the 50D. great job, thanks for posting!