I bought a Speedbooster, couldn't help but put it against the res chart...

Samuel H

Samuel H

Major Contributor
I need to know my gear as well as possible, how it works best, how to avoid its downsides.

I got the speedbooster so that my 24mm could be a real wide angle, but then my question was: if I have to take a portrait, should I go for the 90mm f/2.8 on the speedbooster, or for the 50mm f/1.4 on a standard adapter?

Tests here:
http://www.similaar.com/foto/lenstestsff/lenstests_speedbooster.html

My conclusion: the SpeedBooster delivers a small sharpness increase. Plus, the DoF and brightness advantages make it a winner. But it's not as big an improvement as actually going Full Frame.
My 90mm Elmarit with the SpeedBooster is slightly sharper than both the 50mm or the 90mm on APS-C without SpeedBooster. But it is not as sharp as the 90mm on an actual Full Frame. If all I have is the NEX-5N and I have to take a portrait, 90mm+SB is slightly better (sharper in the center, about the same in the corners) than 50mm with the standard adapter. But 90mm on FullFrame would be a lot better.

But the DoF and brightness advantages make the SB even better. I can take that portrait with the following settings:
* APS-C with standard adapter, 50mm lens set at f/2
* APS-C with SpeedBooster, 90mm lens set at f/2.8
With the same ISO and shutter speed, I'll get the same FoV, DoF and exposure in both cases. Now, looking at the tests, it's easy to see which option has the lead: because I can close the lens a bit, the one with the SpeedBooster is sharper both in the center and in the corners! Not as good as actually going Full Frame, but an improvement for sure, with no downsides whatsoever (because I've also checked that bokeh is basically not affected)
 
Thanks for the test. Looking at the small pictures at the end of the article, it looks to me that speedbooster is not even as sharp as the same lens on the APS-C sensor on its own. You can see it in most of the shots but most clearly 50mm. Also there's chromatic aberration, blueish glow visible, that wasn't there without the speedbooster. I'm not very convinced of speedbooster's abilities just by looking at this test. Did you use Leica R to NEX speedbooster or did you use EF to NEX speedbooster with normal Leica R to EF adapter added in the middle?
 
^ I think what you're seeing is a difference in contrast induced by the different cameras with different image profiles. It's a very flawed test, because I didn't have all the cameras at hand at the same time and had to reuse old shots for the comparison.

I got the LeicaR-NEX speedbooster (cheaper and with a tighter fit than the other option)
 
I'd really appreciate if you could shoot more scientific test by comparing the resolution of lets say your leitz 50mm and some other leitz lens with and without speedbooster... so that would be four shots altogether, or eight if you for example shot one with maximum aperture and one with like f4.0. But I understand that while taking the shots is fast, setting them up is not. If you don't have time or interest to do it, it's ok. But I'd really appreciate to see the results because I would make a buying decision according to the results. If you happen to have NEX photo camera just a photo of resolution chart or something else that allows me to compare the middle and corner sharpness would be enough, but video is ok too although i think photo would show the difference better since it has more resolution. I just spent small fortune on Leica R lenses and I'm thinking if I should get BMCC MFT with speedbooster now that BMCC got price drop. I'm interested not only to see the resolution, but also if the speedbooster affects the color tone of the picture so uncorrected shots, taken same time in same conditions would be really informative. Would be very important to know if speedbooster is viable option and can shots with and without it be mixed and still retain the "leica" look. If speedbooster fails, then getting BMCC MFT would be sort of pointless.
 
Let's see... 50mm f/1.4 and 90mm f/2.8, both with and without speedbooster, would that suit you? both on the NEX-5N, which is the camera I have at hand right now

(this is the case I'm most interested in, since I know my 24 is worse than my 35, and that my 35 is worse than my 50, so in the wide-angle side of things I always want to use the speedbooster, but I also know that my 50 is better than my 90, so I only want to use the 90 if the speedbooster is enough of an advantage)
(I don't mind redoing a simple test, since I also want to look for CA, which my usual test doesn't show)
 
That would be great, Thanks! Please shoot RAW frame with your 5N so that resolution is not lost in jpeg compression. If you can do both maximum aperture and something smaller like f4.0 or 5.6 in addition then it would be enough

If you don't have any place to share the photos, you can register to this site https://copy.com?r=1xIPl1 and install the client, you get a folder on your computer and then you can share stuff in it online for others too.
 
ok, I'll try to do it this evening; I'll post only jpgs on the site, but I'll send you a link to the raw files on a PM
 
Well, the new results are online (scroll down):
http://www.similaar.com/foto/lenstestsff/lenstests_speedbooster.html

The bad news is, I got good news for myself: back when I only had a T1i, I considered my 90mm to be my sharpest lens, but for some reason it had underperformed a bit when I reshot all my tests with the T2i. Now it's working great again, which absolutely kills my tradeoff: 90mm with speedbooster is clearly sharper than 50mm now.

But anyway, there it is, and now you can look for differences in contrast, color cast, etc, since the new shots are all done with the same camera (NEX-5N) and with the same settings (including WB).
(my guess: no color cast or contrast differences to talk about; the slight magenta cast you may see in the 50mm f/1.4 SB shot I think is the result of a very small error in focus: at such wide aperture, stuff goes green or magenta depending on the direction of the error)

And my conclusion remains: the speedbooster gives me an extra stop of light, and the FoV and DoF of a full frame camera, with no negative effect on sharpness or bokeh, so it's a great buy. I think it gives me an advantage in terms of sharpness too, but it's quite small, definitely smaller than moving to a real Full Frame sensor.

(take for example the 90mm f/8 center shot without the speedbooster: when that shot was taken -I always use 2s delay- I wasn't standing still, I was walking away from the tripod, without touching it; I thought it wouldn't be a problem, but you can clearly see it's A LOT softer than anything else; that small amount of motion transmitted to the tripod through the hardwood floor has a much bigger effect than anything else we're seeing here)

There are some raw files here, if anybody wants to have a go at them: www.similaar.com/aarchive/speedbooster_raw.zip
(I'll take them down in a week or two)
 
Last edited:
Thanks for doing the test, downloading the RAWs. Will take a look after I finish my work today. On JPGs to me with or without speedbooster, most of the time the results look very same
 
I checked the RAWs and I think speedbooster is OK to use. It doesn't seem to make pictures much sharper if at all.. but at least it doesn't ruin them. Color seems same too. What haunts me most in this test is how poorly 50mm lens did... I can't believe it is that soft. How far from that "chart" were you when you shot it? Maybe i should check if my 50mm is also that weak.
 
Yep, that's my conclusion too: no negatives, and you get the FoV and DoF of full frame.

And I must say, that 50mm is REALLY sharp, and not just "for a 37-years-old lens". I tested it extensively against a Canon 50mm f/1.4 a couple of months back and there was no contest, the Leica was clearly better in every respect. It's just that the 90mm did REALLY well in this new test.
http://www.similaar.com/foto/lenstestsff/lenstestsa.html

In any case, the chart is around 80cm wide IIRC.
 
Quick update on the speedbooster:

* I wasn't 100% sure the conclusion from the studio tests was that bokeh is not affected, so I've been shooting some real-world comparisons and I don't think bokeh is any worse with the speedbooster than without it

* with the speedbooster I've lost infinity focus with two lenses: the 24mm f/2.8 and the 90mm f/2.8; they both focus to infinity (and beyond) on a 5D2 and on my 5N with a regular adapter, but with the speedbooster I have to stop them down to f/4 if I want the horizon to be in focus; this may be a problem with my unit or a more general one; a side effect is that all focus markings will go down the toilet (I taped a white strip on my most-often-used lenses and marked the focus points with my speedbooster and my camera; it'a better now than it ever was, but it did take me an afternoon of work)
 
Does't speed booster have an infinity focus (back focus) adjustment in the unit?
I'm wondering whether you can maintain parfocal quality of canon zoom lenses with it.

Has anyone tested that?
 
* backfocus adjustment: I haven't checked that, I'll look into it

* parfocal: I don't have a single zoom lens to test that with :)
 
Cool, I'll have to play with that, thanks!

(I don't like that they try to blame this on my lenses: they focus perfectly without the speedbooster!)
 
Samuel, All lenses have to be set for the correct distance to the target chip or film plane (back focus) . its an extremely precise adjustment. Adapters oftne throw this off. Hence speed booster has an adjustment for this. Especially for zooms it can have a radical affect on performance.
 
You must log in or register to reply here.
Back
Top