Image Quality

[Jeff D. Welker]

Yesterday I had an unplanned experience that let me compare JPEG versus RAW IQ. As I've mentioned in prior posts, I'm using my Dad's Sony a350 while I make a decision on what gear I'm going to purchase. It was a soupy day at Willie; however, it was a chance for me to work in less than ideal conditions and see what I could make of the opportunities. Since I became involved in this particular addiction, I've always shot in RAW. To make a long story short, the camera got set on JPEG instead of the RAW capture mode. I was blissfully unaware of this issue and shot a bunch of images. When I got home and began the process of "culling" shots in LR3 I immediately noticed two issues. First, there seemed to be excessive noise in every image (they were all shot at 100 or 200 ISO). Second, the detail sharpness was definitely less than I was used to seeing out of this camera/lens combo. Bottom line, the IQ was noticeably degraded compared to recent captures.

For several minutes I was really scratching my head trying to figure out what the heck was going on with these images. Initially, I was blaming the poor lighting conditions. Next, I thought maybe I had improperly set the exposure mode. It took a while; however, I finally figured out that they were all shot in JPEG instead of RAW. Now don't get me wrong, the reduced contrast in yesterday's soup will definitely cause a slight reduction in detail sharpness; however, not to the extent I was seeing. Below are two similar images. The first was shot yesterday in JEPG mode and the second was shot about a week ago in RAW mode. As you know, uploading to Flickr results in both images being JPEG's. Nevertheless, I hope you can see the difference in image quality.





Like you, I've read the pros and cons for JPEG vs. RAW. This experience has convinced me that shooting in RAW mode is the way to go - for me. Even though these soupy shots were never going to be "book worthy", I would have had much more latitude to improve them if I had been starting with RAW instead of the lesser data in JPEG. I recognize that shooting JPEG has some advantages over RAW (i.e. less of a memory card hog, camera can shoot faster and write quicker to the buffer/memory, etc). That being said, I think the RAW's positives out weigh the JPEG's.

Have you had similar experiences? What is your preference and why?

[Joe Copalman]

Poor lighting and a darker subject are the two primary culprits - you have less light available and a subject that reflects less of it.  I'm guessing the TAV-8B was shot at a much lower shutter speed than the AV-8B, which allows for more noise, blur, etc.  There's a reason many shooters don't even bother taking shots in such light.

Try it again shooting RAW+Large JPEG and compare the results.  You really need to be looking at identical images to make the call (and the edge will more than likely go to RAW again, but by a significantly smaller margin than demonstrated by the shots above). 

[Jeff D. Welker]

I appreciate the comments Joe. I believe dad's Sony a350 will shoot RAW/JPEG simultaneously. I'll give that a try and see what the results show. I may even go back out today and see how RAW works in this soup compared to JPEG. Maybe I'm way off in my assessment. It appeared to me that cropping JPEG's resulted in more noise than a similar crop in RAW; however, maybe it was primarily the poor conditions from yesterday. As the scientist's say; "further study is needed".

Here is the EXIF data for both images.

EXIF Data for JPEG Image
Camera - Sony DSLR-A350
Exposure - 0.001 sec (1/800)
Aperture - f/4.5
Focal Length - 75 mm
ISO Speed - 200
Exposure Bias   - 0 EV

EXIF Data for RAW Image
Camera - Sony DSLR-A350
Exposure - 0.001 sec (1/1000)
Aperture - f/7.1
Focal Length - 160 mm
ISO Speed - 200
Exposure Bias - 0 EV

Shutter speed and fstop are close. ISO identical. Someone will need to help me understand what the EXIF is telling me that will help explain the noise/IQ issues. If I'm reading it correctly (and I may not be) the RAW imaged was cropped more than the JPEG to achieve similar compositions. This makes me think the RAW had the potential for increased noise compared to the JPEG. Tell me if I've left the reservation on this conclusion.

I hear you on why some don't shoot in these lighting conditions. I'm still early in my learning curve and want to get in as much shooting as possible - even if they won't be keepers. It helps me work on many of the skills that appear critical for producing good images. When shooting landscapes, I loved this flat even light for small compositions of flowers, rocks, tree stumps, etc - the kind of stuff that does not include sky. If I could have gotten up close and personal with any of the aircraft yesterday, I would have likely concentrated many images on detail compositions.

[jslugman]

I think Joe's onto it about shooting the very same image and writing it two different ways. If you only have one variable (file format) then it's easy to make the call; but if you also have different focal lengths, aperture values, and crops then you now have (at least) four variables in the soup. And remember that the JPEG process is tossing information away to give you those advantages, perhaps the info it's tossing is what makes the difference.

Back to the airfield for you, sir! 

[Jeff D. Welker]

I think Joe's onto it about shooting the very same image and writing it two different ways. If you only have one variable (file format) then it's easy to make the call; but if you also have different focal lengths, aperture values, and crops then you now have (at least) four variables in the soup. And remember that the JPEG process is tossing information away to give you those advantages, perhaps the info it's tossing is what makes the difference.

Back to the airfield for you, sir! 

I will proceed with my assignment and report back. Thank you kind sirs.