Monday, April 9, 2012

Canon 5D Mark II vs Mark III–Image Quality Comparison Part II–RAW Conversion

In my previous article entitled Canon 5D Mark II vs Mark III–Image Quality Comparison, I did my test using the in-camera JPEG of both bodies using the camera default settings (which means Standard Picture Style). This is what you’d see embedded in your video files and the default processing from RAW via DPP unless you changed the picture style. This caused some naysayers to declare the results of my testing invalid, so in all fairness I decided to see if there was merit to their claims.

How I Tested

Using the ISO 100 RAW files from my original test, I have obtained new results. This section explains how I came up with those new results.

image

To make things fair, I used the version of DPP that came with the 5D Mark III as that’s the newest version at this time. DPP is the only product that you can use for RAW conversion comparisons with any accuracy because all other RAW processors are reverse engineering Canon’s CR2 file format and applying their own guesses on processing.

In addition, the faithful picture style must be used for RAW processing to see what was really in the RAW image without any picture style enhancements. Now it turns out that DPP by default will try to use a new unsharp mask sharpening method on the RAW, so I had to set both to the same values as you can see here:

The RGB and NR/ALO processing is the same between both (files can be viewed here), but Lens must also be updated to make sure both are processing the image the same way:

UPDATE: I discovered what my issue was with the color space problem wasn’t a problem after all. I had mistakenly uploaded the wrong shooting info file to the blog. There still seems to be a slight color shift that appears to be how the “As Shot” in-camera white balance works, but color space wise they are the same. All of the other images were uploaded correctly and the results remain the same. My apologies for the confusion.

Here’s the shooting information reported by Canon for both camera bodies AFTER my DPP adjustments (although for some reason the picture style value remains at the “as shot” value of Standard despite being changed to Faithful):

I exported the files as 16-bit TIFF files, opened them in Photoshop CS5.1 and saved them as JPEG’s in the sRGB color space. You can access those images by clicking on the 100% crops below.

The Results for ISO 100

I obtained my crops using the same PSD template I used last night and got the following results:


5D Mark II - ISO 100

5D Mark III - ISO 100

Conclusion

Yes, it appears a big portion of the improvement of the 5D Mark III comes from its in-camera RAW processing, and if you aren’t careful you’d see the same improvement via DPP. However, to my eyes (and going beyond the color shift) it seems there’s still a slight improvement favoring the 5D Mark III.

Now personally, I don’t know anyone who processes their photos this way so I think the real-world differences are still more accurately reflected in my previous article entitled Canon 5D Mark II vs Mark III–Image Quality Comparison. However, if you are a 5D Mark II owner looking for a reason not to upgrade, then maybe this will make you feel much better.

Why no high ISO results?

Simply put, I have a life and I have to get to work today. If the differences are this subtle here, you can easily infer similar results at the higher ISO’s. I suspect the in-camera noise handling at the RAW level of the 5D Mark III is still probably two stops better, but I’m happy to leave that as an exercise for someone else to prove or disprove as I’ve got to get to work.

Order a 5D Mark III today from Amazon, Adorama or B&H.

Related Articles

I’ve got more to come on the 5D Mark III including a big article just for parents. Check back to learn more!

Disclosure

I may make a commission if you make a purchase using links found in this article.

If you enjoyed this article, please support future articles like this by making a donation or saving money by using my discount coupon codes. Either way, your support is greatly appreciated!

Click here to learn more about how this blog is funded. Residents from Arkansas, California, Georgia, Illinois, Kansas, Maine, Michigan, Minnesota, Missouri, Nevada, New Jersey, North Carolina, Ohio, Rhode Island, Tennessee, Vermont, Washington, West Virginia, and Wyoming (and where prohibited by law) may be prohibited from using the links to make purchases, so please consider making a donation instead. This blog is intended for freelance writing and sharing of opinions and is not a representative of any of the companies whose links are provided on this site.

The opinions provided are of Ron Martinsen alone and do not reflect the view of any other entity

No comments: