How to get better bokeh using the Extreme

disjecta

disjecta

Lucky Duck
So undoubtedly, most of you have seen teh ongoing debate about good and bad bokeh and there are some who have objected to the Extreme's bokeh, saying it is too diffused or whatever.

This is a very bright adapter and it needs to be "calmed down" a little. When you are shooting with your Nikon or Canon lens wide open, you are going to get a softer bokeh and, in the Extreme's case, it's going to look diffused and ghost-like. If you shut down the lens a little to, say, f4, the bokeh is dramatically improved.

From a practical standpoint, this works because you shouldn't be shooting with the lens wide open all the time anyway. I am posting an example of what I am talking about:

Lens used in this case was Nikon 50mm f1.8

Lens open to f1.8 sample first with crappy bokeh:

bokeh_f1.8.jpg



Lens closed down to f4, bokeh much improved:

bokeh_f4.jpg
 
Why you stealing my advice, Disjecta? (that was a joke, by the way.)
 
Last edited:
This is a good point on sharpening up bokeh. One thing that we discovered in testing with the Brevis/50mm f1.4/CF1L on the scope is that you can actually stop down the 35mm lens from f1.4 to f4.0 and see almost no light loss. It's not something we advertise, but any Brevis user can attest to it. Take a look at time 0:53 in the autumn footage. It's an 85mm at f4.

So stop down your lenses a few stops if you can! The benefits to lenses that are soft wide open, as well as bokeh improvements are nearly universal.
 
Does that have less to do with the adapter than it does the actual Lens, though? It's pretty much the same thing with the Letus and the M2. Same Lens and all. Even in Disjecta's two stills, the apparent light increase from 4 to 1.8, or decrease from 4 to 1.8, is very minuscule to begin with.

I mean I'm probably wrong, but that seems to make more sense than the adapter since you're stopping the lens down... not the adapter.
 
It has more to do with how the lens is interacting with the GG at a certain iris opening. I made adjustments to the A1's exposure to get approximately the same reading so, as far as light loss or gain, the above stills are not a good example of that. I was only demonstrating the bokeh difference.
 
hey Disjecta,

Do you have any f/2.8 images both from a faster lens and from one who's minimum is f/2.8?

thanks for the stills - great help you are :)
 
Khol, a 35mm lens aperture closes 50% from it's previous value for each f stop, meaning you should require double the light each time you do it. This is the physics of the lens..period. So stopping down a 35mm lens from f1.4 -> f2.0 -> f2.8 -> f4.0 should require in increase in the amount of light required by 8x. In other words, your 1000 watt light, should now be eight 1000 watt lights (we're ignoring luminous efficy for you sticklers out there). What is very, very different for the Brevis is that stopping down the 35mm lens from f1.4 to f4 as in my example would only require another 500 watt light...not 8 x 1000 watt lights. This was illustrated very well in Phil Blooms review (and not using our new CF1Le) which shows the Brevis in the range of 4 stops more efficient than the other two adapters tested with the 50mm lens at f8. Phil actually should have stopped down the HVX nearly 2 more stops on the Brevis grabs for proper exposure.

What does this mean? Brevis owners can (and should) dial in 1 or 2 stops on their 35mm lenses to get into the lens sweet spot. They can do this with near zero light loss, meaning the camera's aperture would remain untouched. If you do have a Brevis unit and find this hard to believe...tether to a scope and explore this for yourself :-) My general rule of thumb is to shoot with both the taking camera and the taking lens in their sweet spots with respect to sharpness. The Brevis On Ice (HV20) and Autum compilation were shot with both cams in the f4 to f5.6 range, and lenses in the f4 range too. The adapter design took care of the rest of the sharpness issues :-)

In any case, Steve's point on stopping down lenses to sharpen them up and clean up bokeh is a hot tip for image quality. The fact that his footage looks as good as it does is not an accident! Just remember that every adapter will differ in terms of what the effect is.
 
Last edited:
Heh there you go defying the laws of physics again, Dennis! :p

Anyway from what I've tested with my letus economy, there isn't *that* much difference between f1.8 to f4, although it's definitely noticable. I doubt it's anywhere as efficient as the brevis claims to be, though.
 
Just take frame grabs and compare them on a scope in your NLE...then you'll know exactly. Take a few with your camera pointed at a well lit white wall at f2.8 and then f4 (bare) to compare. This is one of the reasons I'm an advocate of testing adapter light loss with a lens mounted up, and then tested wide open, down to f4, then f8.
 
I've tested the f1.4 to f4 thing and it's true. You get a stop or less of light (not measured just eyed and wowed) when closed down.
 
MalcolmOng said:
Heh there you go defying the laws of physics again, Dennis! :p

Anyway from what I've tested with my letus economy, there isn't *that* much difference between f1.8 to f4, although it's definitely noticable. I doubt it's anywhere as efficient as the brevis claims to be, though.
Click to expand...

I'm doing the same thing and not seeing a huge difference, although the laws of physics say that there's definitely a difference from 1.4 - 4 (on my 50mm anyway).

Will have footages today when I get back from riding around with the Letus EX.
 
OK, so will someone educate me on why the first grab is "bad" bokeh?

When I've been looking at bokeh, I am looking for smooth edges, the closer the bokeh is to a circle, the better the bokeh is IMO. So the more the bokeh looks like polygons, the more "bad" it is to me. When things go out of focus in my own eyes, they make circles of light, not polygons of light. So my evaluation of bokeh has always been how "real" does it look.
 
Ryan, the circular aspect of specular highlights in bokeh has to do with the lens itself and the blade structure. The less blades, the more of a diamond shape you are going to get. All definition in the picture has fallen apart in the first still and it has taken on a diffused or ghost look to it. There is nothing natural about it at all. You may prefer it but it's not regarded as good bokeh. The second still is way more in the direction of classically desireable bokeh.

Everything is subjective, of course, and bokeh tends to be one of those areas where there are as many opinions as there are variations in bokeh itself. Subjectively, there is no right or wrong, technically there is.
 
I'm not saying that I like the first example better overall- but I do like the circular nature of the bokeh in the first better, as compared to the second one that has a more polygon look / feel to it.

So what I hear you saying is that the first is "bad" because of the overall ghosting in the image, whereas with the second the ghosting is gone, making it "good" overall. Is that correct?
 
Yes, and if you were using high quality cinema lenses, you would see a circular pattern around the highlights....and the integrity of the image is maintained more in the second one.
 
Taking a look at the other examples you have provided, I really like the bokeh in this image- it is circular, and the image retains definition, free of ghosting:
soccerbokeh1.jpg


However I really do not like th bokeh in this image due to the polygon look, as well as you can actually see the edges of the blades in the red light:
soccerbokeh5.jpg
 
disjecta said:
Yes, and if you were using high quality cinema lenses, you would see a circular pattern around the highlights....and the integrity of the image is maintained more in the second one.
Click to expand...

Thanks- I'm with you now, and I think I finally get what people are talking about. I have been looking at just the circular areas of the bokeh, and not the whole image overall. That makes a lot more sense. Thanks :)
 
In the top still, the highlights are not actually circular, they are exactly the same shape as the bottom still. The difference is that the highlights are so small in the top frame, they are perceived as circular. The same lens was used on both of these shots and the same f-stop.
 
disjecta said:
In the top still, the highlights are not actually circular, they are exactly the same shape as the bottom still. The difference is that the highlights are so small in the top frame, they are perceived as circular. The same lens was used on both of these shots and the same f-stop.
Click to expand...

Very interesting- good to know- so size does matter ;) . There are a ton of variables that go into the perception of the rendition of the bokeh. Thanks for providing the examples, as well as educating me on all of that.
 
You must log in or register to reply here.
Back
Top