Black Magic pocket cinema camera test 1 from Joseph Jagatic on Vimeo.
Test Olympus M. Zuiko Digital 17 mm f/1,8 (Focus Numerique).
E-P5 ILC Review by Steve Digicams.
Apple Aperture upgrade (Dpreview).
The GX7 vs the OM-D E-M5, battle for my affection, Round 2 – Build Quality and Ergonomics by Tyson Robichaud.
Top 10 Best Mirrorless Compact System Cameras (ePhotozine).
Micro 4/3rds goes airbourne on BBC news!
E-M5 review by Benjamin Jehne.
45mm review by Benjamin Jehne.
24 Responses
Sorry, but this test looks outright terrible. Pocket cam? Really? Would have thought it came from a cell phone of an ophthalmologist.
I know this is off topic, but I just now came inside after portrait photographing my kids with my Canon Rebel using an 85mm @ 2.8 (it can go to 1.8 but I find it gets really sharp at 2.8, not sharp enough before that), and got some nice bokeh.
I’m planning a switch to m43, and will be getting the 45mm 1.8 which is sharp wide open. Am I right in saying that because 85*1.6/2.8 = 48.57 and 45*2/1.8 = 50, which is very similar to 48.57, I can expect the same depth of field and general bokeh between these two setups? So if I frame the kids the same I shouldn’t really see much difference between these two setups, is that right? Is that how it works?
Thanks.
The 85 @ 2.8 gives you a depth of field of 1.38 meters at a 10 meter focus distance on the Rebel (an APS-C sensor).
A 45 @ 1.8 will give you 2.59 meters of DoF at the same distance with a 4/3 sensor.
Panasonic will be releasing a 42.5/f1.2 next year if you’re willing to wait. It will likely be very expensive though.
I wanted to start a news blog about m43 cameras last year, but I never got the time to keep it up to date. What I *do* have on that site, though, is a nice tool to calculate DoF. http://43news.tumblr.com/measurements Feel free to test it out.
Thanks for the link I had a look at it. But I don’t want the two setups to be at the same distance, I want the framing to be the same, so the 45mm lens will need to be closer than the 85mm. That extra closeness, plus the wider aperture of 1.8 are I think enough to make the DOF the same as it would be if using the 85mm on a crop.
That’s my theory, and I used the formula in my original one to arrive a number which I have no idea if it has a formal name or not, but I call it the DOF figure (Total EFL divided by aperture used). And then bokeh is kind of directly relevant to DOF. Of course, bokeh may be more pleasing looking for one lens over the other, but basically the amount of bokeh will directly correlate to that DOF figure. Is this right?
‘Out of focus’ and ‘bokeh’ are two completely different things.
‘Out of focus’ is, obviously, whether you can distinguish two points on a picture and this will affect depth of field. Bokeh is how the out of focus bits are rendered and depends on how the lens is built (for example the number of shutter blades) and will not affect depth of field – you can use two different 100mm lenses on the same camera one of them may give a background that is a creamy undistinguishable backdrop but with another you may be able to make out the shapes in the background.
Using your tool, the DOF figure is the same (0.22m) for these two setups described above, if I put the distance as 4 metres for the Canon system and distance as 3 metres for the m43 system. My formula seems too simple though, because it doesn’t factor in distance from the subject anywhere, or sensor size either. I’m not sure if it should. I’m only guessing that a Canon user would stand at 4m compared to a m43 user at 3m away.
AW – 75mm rather than 45mm on m43 is more comparable to 85mm on APSc (Canon Rebel). In terms of angle of view in 35mm format, 75:m43 is like 150mm and 85:APSc is like 136mm, where 45:m43 is like 90mm.
In terms of DOF, 85@F/2.8:APSc is similar to 75@F/2.2.
Thanks, yes I know the angle of view will not be the same between the two setups, but I don’t really care whether I have to stand a little further back or a little closer. The thing I’m interested in is the amount of DOF and therefore bokeh if I frame a portrait the same way as I usually frame it with my current system.
If I could afford the great 75mm m43 lens I would want to do some calculations based on that too (so your DOF comment is useful), but since I can’t see myself affording it yet, I’m trying to work out if the 45mm will do the same job as my current 85mm (which I usually always use at f2.8).
AW – OK I understand your point, although perspective (where you stand w.r.t. the subject) is always the principal component of composition.
Assuming you are using the 45mm instead of 75mm and standing closer in order to keep the size of the subject identical (say, head-and-shoulder), and your goal is to have similar background (say, 30 meters away) blob sizes to that of 75@2.2 (or 85@2.8 on APSc). You would need to shoot at about F/1.6 on a 45mm. Because of the change in perspective, the calculation is not a simple scaling any more. This is very useful: http://howmuchblur.com/#compare-2x-45mm-f1.8-and-1.6x-85mm-f2.8-on-a-0.9m-wide-subject but again, both your shooting style and the look of the pictures are going to be quite different.
oh wow that site looks great, I will spend some time on it.
Without goin into the numbers, i can assure you the 45mm f/1.8 is a very gentle rendering lens for portraits, either head or half/full figure shots. You could decide to go for the 75mm, but i think thats more of a specialy lens, with the perspective compression of tele lens more visible here. Awesome if used in the right way, but prob a bit less flexible like in tight spaces (where the limits of the 45mm fl also is already showing)
If you’d use the 1.8/45mm you have to get a bit closer than with the 85mm. The pics will turn out a bit different but I think you will like the results 🙂
I sometimes wish there could be 12-35/2 and 35-100/2 lenses comparable in size and weight to EF24-70/4 IS and EF70-200/4 IS from Canon. Of course, lighter and smaller would be even better, but the 4/3 lenses have indicated otherwise (35-100/2 is as heavy as 70-200/2.8).
+1000. 2.8 on m43 is like…just barely not enough. 2 would be perfect.
I agree – for my GH2 I traded my Legacy 28mm f2.8 for a Canon FD 28mm f2.0, and that one extra stop really has made me much happier with the OOF rendering I can achieve.
I don’t know whether Tyson Robichaud’s first round of E-M5 vs GX7 comparison was posted here, but I found it much more interesting, especially the fact that GX7’s IBIS was better at steadying long exposures: http://tysonrobichaudphotography.wordpress.com/2013/10/17/the-gx7-vs-the-om-d-e-m5-battle-for-my-affection-round-1-ibis-evfs-lcds/ !
Strange results on Tyson Robichaud IBIS test in round 1, with GX7 getting sharper results than E-M5 at long exposures. Opposite of everyone else yet. Don’t know what to think of that.
I have the impression too my em-5 sucks at slow shutter speeds, but i prefer to shut it off anyway for stills (i hate it when the ibis correction movements of the sensor cut off my models feet during critical framing)
Luminous landscape says the two IBIS systems are comparable, gives slight edge to E-M5: http://www.luminous-landscape.com/reviews/cameras/panasonic_gx7_review.shtml .
Comparable interested me.
Then I read the review and noticed that IBIS doesn’t engage during video. That’s decidedly not comparable.
Sorry I have to say it.. Worst video quality ever.
Quality is actually GOOD, BUT – this is looking like crap, because it needs post processing. The captured dynamic range is too much to just through on an 8 bit per channel encoded video and screen. You need to apply curves and compress highlight and shadow areas.
Here’s how it can look with minor curves applied.
http://imageshack.us/g/1/10376625/
Note the last image shows what the curves are for the default image of the old blurry guy.
If you’re compressing wide dynamic range, using the full brightness available to the medium is a good idea. There’s a lot of unused area in the peak highlight and shadow that isn’t touched. They compressed the contrast even more than they had to.
Thanks for showing that, looks like there is alot in the file which you can pull out.