[noser]

First, a quick bonus comparison: How does the K-5's viewfinder compare to the K-1's?

The K-1 viewfinder is a little brighter, clearer, and bigger.

With the O-ME53 magnifying eyecup on the K-5, it's closer, but different. I wear glasses and I can't quite see the entire viewfinder of the K-5 when using it, but manual focussing is better.

Now, the main event... this is one specific test that I've always wanted to do.

How does a 50mm on a FF body look compared to a 30mm on a crop sensor? They should be pretty close, right?

Check out these two shots, which are fairly similar. If you don't peek at the EXIF text on the left side of each image, I don't think you could tell them apart. So don't peek, and scroll down...

1)


2)






(no peeking, answer below, and a twist...)






1) 30mm Sigma @ f/1.4
2) 50mm Pentax @ f/2.8

That's pretty darn close, visually.

So what happens when the 50mm is used at f/1.4? How does that look?



I'm not a huge bokeh-monster fanboy, and the world does NOT always look better at f/1.4. But if you need the extra stop, it's there, and it's a big jump in appearance if your subject distance is much closer than your background distance.

[jimr-pdx]

No doubt others can point out several differences and/or dealbreakers.
To me the PGE rule applies to both, Plenty Good Enough for anything I'd want from this view. Thanks!

[noser]

Plenty Good Enough for anything I'd want from this view. Thanks!

PGE... Very, very, very true!

For those that have the upgrade bug but can't afford it, they shouldn't feel left out.

The digital 30mm Sigma is sweet - and that's NOT the Art version, either. I think I paid $375 for it new.

For those that might need the extra bokeh for their style or subjects, high ISO performance, or big-print capability, they got ya covered.

[quh86]

if i am not mistaken, in terms of DOF APSC and FF are just 1 stop apart ?!
between f2.8 and f1.4 there are two stops, therefore it would have been more accurate to shoot the FF 50mm at f2 instead 2.8. but this is just nitpicking. very interesting to see the results of a comparison between K-5 and K-1, thank you

[noser]

if i am not mistaken, in terms of DOF APSC and FF are just 1 stop apart ?!
between f2.8 and f1.4 there are two stops, therefore it would have been more accurate to shoot the FF 50mm at f2 instead 2.8. but this is just nitpicking. very interesting to see the results of a comparison between K-5 and K-1, thank you

Yes - you are correct, but I chose the two shots that most closely resembled each other. I actually thought the 50mm f/4 shot was closer at first, but it had just a bit more contrast in the details, so the 2.8 was a much better match on second look.

Here is the f/2.0 shot for kicks. It's somewhere closer to the f/1.4 shot, as the size of the OOF objects doesn't match the 30/1.4 or the 50/2.8 shot.

[Ptitboul]

It seems that the first two shots were not taken from exactly the same point of view. The background in the center and at the right of the pictures are similar, but the background at the left changes a lot.
Therefore I am pretty sure that the size of the OOF objects don't depend only on the DOF, and that explains why you made the comparison with f/2.8 instead of f/2.0.

[noser]

It seems that the first two shots were not taken from exactly the same point of view. The background in the center and at the right of the pictures are similar, but the background at the left changes a lot.
Therefore I am pretty sure that the size of the OOF objects don't depend only on the DOF, and that explains why you made the comparison with f/2.8 instead of f/2.0.

It was still within inches / cm of each other, even though I was trying to keep apparent subject sizes the same with nearly equivalent lenses. Part of my interest in the comparison is that 30mm isn't quite right for APS-C - it's a smidge wider than ideal - so doing real world test of real lenses in real shooting conditions is far more interesting than mathematical or idealized conditions. The items in the background were much further away (several feet, about a meter for some branches) so the relative distances were still similar. I also had working distance constraints for near focus.

I could re-do it on a tripod and use a measuring tape, and I'd probably get something that splits the difference. But that's not interesting, we know all this already.

What these tests inform me (wide, normal, and tele) is how my current style must change. Since my go-to lenses are wider now on FF, I have to be much more careful of my background selection and optical distortions. I suspect that for familiar things (the human face) at similar coverage to APS-C there will be a difference where FF produces it's most flattering image by selecting longer focal lengths.

In simpler terms it means I'll have to take care not to make portraits of people where I get closer than with APS-C to get a similar image, but end up exaggerating the size of their nose.

[Ptitboul]

It was still within inches / cm of each other, even though I was trying to keep apparent subject sizes the same with nearly equivalent lenses. (...)

I agree that since you only used "nearly equivalent lenses" the comparison is difficult. The equivalent of the 50mm on the K-1 would be a 33mm on the K-5, which does not exist. However, you could have taken the two pictures from exactly the same point of view, and then cropped the picture made with K-5 to reduce the FOV to the one of the picture taken with the K-1. And then, if the picture with the 30mm is taken at f/1.4, the picture with the 50mm should be taken at f/2.3 to get exactly the same DOF. f/2.3 is not available, but f/2.2 is available when having 1/3 EV steps.

[noser]

I'm just happy the test showed these things, math aside:

1. Nearly identical shots are possible between formats when the comparison is realistic, the settings simple, and the results useful. I have a 50 f/1.2 but I chose the f/1.4 lens at it was AF and more typical for most users. Comparing the 30mm at 1.4 to the 50mm at 1.2 would be like bringing a knife to lightsaber fight (or is that the other way round?). But the point is that the K-1 isn't a magical unicorn, and even at close quarters and wide apertures both formats produce great results.

2. Outside of comparable settings, pretty much any FF + 50mm f/1.2, f/1.4, f/1.7, or maybe f/1.8 (I haven't tried my other ones yet) can out-DoF the 30mm at f/1.4 on APS-C. So if you have a Sears 55 f/1.8 or an FA 50mm f/1.7 know that it already will blow a 30mm on APS-C away. But that's a bit contrived, since if you have both systems you can just mount the 50mm on the APS-C, step back a bit, and worry less about the width of good background (as compared to using the 30mm).

I think the math is comforting from a 'this should work' perspective, but as I found out some comparison shots don't work, since the minimum focus distances of the lenses aren't the same. I didn't present those as cases where one system was better than another because that's not how shooting in the real world works... no one ever said "Gosh darn it, if only I could focus an inch closer I could get the same image as the other system with this different lens!"

Well, no one should ever say that, anyway.

Life's too short.