Jeepers Creepers, Where'd Ya Get Those Peepers? Part 3.

EUREKA!

Quick show of hands: how many of you use protective UV filters on your lenses?

I'm going on 45 years of using interchangeable lens cameras, and have bought the old saw that it's better to break a filter than the front element of your lens, and there is no degradation caused by a filter.

Right.

Except with a Sigma 120-400mm f/4.5-5.6 APO DG HSM telephoto at full zoom.

To recap (more here and here): I had some unexpected results when comparing a micro four thirds camera (Olympus OM-D E-M10) and an APS-C camera (Sony A77II). All I wanted to know was, how much difference is there between MFT and APS-C. I was surprised that the smaller sensor E-M10 images appeared to be sharper than the larger APS-C sensor images (more on sensors here). But, I was trying to recreate a realistic shooting environment by handholding and using a fairly fast shutter speed. I decided that I needed to be a little more disciplined if just good image quality was my sole objective.

So, I repeated all my tests using a tripod. With image stabilization enabled. And, with image stabilization disabled (as recommended by the manufacturer). And, using the self-timer to further reduce the possibility of causing camera shake when pressing the shutter release. And, oh, my, did I ever get some weird results with the A77II. There were some kind of strange artifacts present when stopped down to f/11. Like a double exposure or something. And, it got worse at f/16! The mystery deepened.

I was ready to blame the lens and throw it under the bus, figuratively speaking. Then, in the middle of the night, it hit me - there was one other variable: the freaking "protective" UV filter that cost all of about $5 sitting in front of my almost $1,000 lens. I know, I know, it is so close to the lens it should cause no discernible degradation of the image.Well, surprise, mateys. It does. And, here's the proof:

Without Filter and With Filter

I would not have believed it if I hadn't seen it with my own two eyes. Multiple times. After doing some research, I've found that you can pay close to $100 for a "good" filter, so maybe that's my problem - a $5 Tiffen just isn't a match for a Hoya or better. In the mean time, however, I think I'll just go filterless.

And, I can get back to comparing MFT to APS-C!

Aye, lassies an laddies, we've uncovered t'culprit, an 'is name be Tiffen...

Jeepers Creepers, Where'd Ya Get Those Peepers? Part 2.

The saga continues...

Previously on Jeeper Creepers: A young lad in North Dakota discovers the allure of photography when he finds his father's Kodak "Pocket" camera and learns about aperture and shutter speeds by opening the back of the camera and watching what happens when the shutter release is pressed (more here). Later he graduates to a Brownie Hawkeye, then a 35mm rangefinder, and then a SLR with interchangeable lenses. He is hooked...

In my last post, I described some unexpected results when comparing a micro four thirds camera (Olympus OM-D E-M10) and an APS-C camera (Sony A77II). It might be fairer to say I am comparing lenses, namely, the Olympus M.Zuiko 75-300mm f/4.8-6.7 II MSC ED zoom and the Sigma 120-400mm f/4.5-5.6 APO DG HSM. All I wanted to know was, how much difference is there between MFT and APS-C. I got more than I bargained for...

In my earlier comparisons I was surprised that the smaller sensor E-M10 images appeared to be sharper than the larger APS-C sensor images (for more on sensors, click here). But, I was trying to recreate a realistic shooting environment by handholding and using a fairly fast shutter speed. I decided that I needed to be a little more disciplined if just good image quality was my sole objective.

I tried a number of things, with use of a tripod probably being the most important. Although, quite frankly, with the image stabilization built in to both the E-M10 body and the A77II body, I'm not sure it really made that much difference, particularly when using shutter speeds of 1/500 sec and faster. Still, using a tripod took the camera shake variable out of the picture (pardon the pun). Even with a tripod, however, there are a couple of other considerations, which we will explore as we go along.

First let's agree on some fixed constant. Most photographers would agree that the best image quality should be at some aperture other than wide open (subject to spherical aberration, coma and astigmatism) or closed down to the minimum (diffraction reduces sharpness). It's simple physics. A lens can not be optimized across its entire surface, so in almost all cases the optimum image quality is somewhere in the middle aperture range, in fact usually one or two stops smaller than the maximum in order to allow the most light possible while providing the optimum lens corrections. For example, here is the difference between f/6.7 and f/11 on the E-M10:

Olympus 75-300mm f6.7 vs f/11 @300mm (100% crop)

F/16 looks marginally better, but interestingly f/22 is less sharp. As the lens approaches its smallest opening diffraction rears its ugly head and IQ suffers. For our purposes let's stay at f/11.

So, how much difference does a tripod make? When I compare the Olympus handheld shot at 1/2000 sec against a tripod shot at the same f stop but a shutter speed of 1/500 sec, I see virtually no difference at max zoom (300mm). I would hope I never shoot at any less than 1/500 sec except for extreme cases. However, for the sake of argument, let's compare the shots on a tripod. Oops. We already did. The comparison shots above were taken on a tripod.

Next comes the argument that shooting on a tripod can be worse UNLESS you disable the IS (image stabilization) feature. I have seen some pretty extreme examples - supposedly any movement is picked up by the IS and an anti-movement adjustment mechanism goes into a loop, picking up the movement caused by the mechanism. I don't see this with these two cameras.

So, we are down to lenses, or more properly, the Sigma lens. Here is just the Sigma compared at max aperture (f/5.6) and close to optimum (f/11):

Sigma 120-400mm f/5.6 vs f/11 @400mm (100% crop)

There is some kind of strange aliasing going on in the f/11 shot. Like a double exposure or something. And, it gets worse at f/16! The mystery only deepens...

Any suggestions? The only thing I can think to do is get my hands on another 400mm and see what kind of results I get.

Jeepers Creepers, Where'd Ya Get Those Peepers?

No, I am not a peeping Tom. A peeping Chuck, perhaps.

Yes, I am in danger of becoming a "pixel peeper". A pixel peeper is someone who magnifies photographs on a computer screen to critically evaluate image resolution at the pixel level. However, it is commonly used as a derogatory remark as explained in the Urban Dictionary.

Late in April, shortly after receiving my new Olympus OM-D E-M10, I was doing some comparison of image quality and came up with the following shocking sample. Note that the sharpest image seems to be from an old Canon SX50 with a 1/2.3" sensor, not the APS-C Sony A55 or the new Micro Four Thirds Oly. (SX50 on the left, then E-M10 and A55 - you may have to click on the image to see the difference) A fluke, you might say. I might agree. Unfortunately I didn't have time at the time to pursue the issue (closing up in Florida and heading back to KC), and now the SX50 is gone (took an unfortunate dip in the lake) and the A55 is also gone (upgraded to an A77 M2). So, back to the drawing boards

It is just too weird to think that a super small sensor could yield images comparable to APS-C or MFT, so I decided to concentrate on a comparison between the E-M10 and the brand spanking new A77 M2. Results have been different from my first attempt, but are still puzzling. Even though having a smaller sensor than A77 (MFT is 61% the size of APS-C), the E-M10 appears to have significantly better sharpness. What the heck?

 I started by setting both cameras up to what I considered to be a typical configuration for sailing or birding shots: shutter preferred and fairly fast (1/1000 or faster), medium ISO (greater than 200 but less than 1,000), which would probably mean shooting most of the time wide open, maximum aperture. I ended up using a shutter speed of 1/1600, ISO 400 and max aperture on a Sigma 70-400mm (f/5.6) and the Olympus 150-300mm f/6.7. And, this is what I got:

Again, you may have to click on the image to see the difference, but just look at the window frames and I think you'll agree that the E-M10 images are sharper. So, thinks I, lenses typically aren't at their best wide open, let's stop down a couple stops. And this is what I got (f/11):

Not much difference, is there? Note that I cranked the ISO up to 1600 in order to stop down and from a noise standpoint there isn't much difference compared to the original ISO 400 shots.

Oh, dear. What can the matter be? Well, we all know that zoom lenses are not at their best at maximum zoom, so I backed both lenses down to the 35mm equivalent of 300mm (instead of 600mm). Results? Not that much different:

What is left to explain this phenomenon? A 300mm vs a 400mm lens? Brand name lens vs independent? 

My plan is to clamp these babies on a tripod even though I probably won't do that very often in reality. Any insight into this dilemma will be greatly appreciated... 

Update: I failed to mention that the images above were 100% crops of these windows.