SGpro rev. 3 vs Letus35 Extreme - Part I

marco0782

marco0782

Well-known member
My friends and I did a pretty intensive test of these two adapters over the weekend. Ultimately we were unable to perform as ambitious of a test as I had promised earlier, but I hope that the results are as useful to you as they were to us.

I will be posting our results in a few different parts. This first part will demonstrate the difference in bokeh quality, sharpness and light loss using these two lenses:

Nikon 50mm AIS f/1.2
Nikon 85mm AIS f/1.4


Both cameras were recording at 1080i/24pa and with neutral scene file settings except Master Ped set to -2, Cinelike V gamma and Cine Matrix.

Note:
Unfortunately when we tested these lenses wide open there was a slight difference in the way exposure was set on the different HVX's, resulting in the SGpro images to be slightly brighter than the Letus (it would have been easier if we were testing both cameras simultaneously, but we only had one copy of each lens). In order to be as consistent as possible, I have only posted the images at f/2.8 and f/5.6.


The point of focus in these images was the ARRI oval logo on the side of the light.

50mm Test

SG50mmf28-f20.jpg

SGpro - Lens at f/2.8, HVX at f/2.0

LEX50mmf28-f34.jpg

Letus35 Extreme - Lens at f/2.8, HVX at f/3.4

SG50mmf56-f20.jpg

SGpro - Lens at f/5.6, HVX at f/2.0

LEX50mmf56-f20.jpg

Letus 35 Extreme - Lens at f/5.6, HVX at f/2.0


85mm Test

SG85mmf28-f20.jpg

SGpro - Lens at f/2.8, HVX at f/2.0

LEX85mmf28-f28.jpg

Letus35 Extreme - Lens at f/2.8, HVX at f/2.8

SG85mmf56-f20.jpg

SGpro - Lens at f/5.6, HVX at f/2.0

LEX85mmf56-f20.jpg

Letus35 Extreme - Lens at f/5.6, HVX at f/2.0


Conclusions


Bokeh:

The SGpro has more defined bokeh circles. Once you stop down on the Letus the circles become more defined, but they still lag behind the SGpro. This ultimately comes down to a matter of preference, but the SGpro's bokeh is more filmlike throughout the aperture range.

Sharpness:

They are both equally sharp to my eyes. If you look closely the Letus appears to have slightly more depth of field which may give the illusion of increased sharpness.

Light loss:

The SGpro loses roughly one stop more light than the Letus, and that is without a flip module.


Coming soon:

Part II will focus on how both adapters perform with high shutter speeds and small apertures.
 
Last edited:
Although it wasn't extensive, it is still a good example of the two's bokeh rendition. Although the Sgpro's bokeh is much more filmic, the letus extreme does gain an extra stop of light, as well as a flip, which for those who's cirumstances would benefit from these bonuses are a better option.

Thanks for this!
 
Boy I'm, kind of mystified by the differences between tests that people have run.
As I recall Tim Naylor found his SGPr0 Rev 3 to be less than 1/2 stop slower than the Letus (wasn't it even less than 1/3rd stop or is that my imagination?) , and I found the Letus to be slightly slower than the SGPro Rev2 (Wayne says rev 2 & Rev 3 should be the same). Kholi said SGPro 1/2 stop slower than Letus.

Now you guys have a full stop difference.

Marco, what did you mean by one HVX's exposure being set differently?

When I ran tests we found one HVX to be 1/4 -1/3 stop faster than the other even though they had the same settings.

Also marco - where are you posting your images? Do you use the member upload center or post them on another site entirely?

Lenny Levy
 
The entire stop difference here is a little confusing. But, someone mentioned havin' to zoom in further versus havin' to zoom in very little. These should create difference in light loss, yes?

I'm not sure how you got that the SG was faster than the Letus. I can't see that being at all accurate. That would mean the SGpro Rev2 loses less than half a stop?

To me, as well, both images look pretty good in these stills. Lookin' for the next part!
 
Marco, what did you mean by one HVX's exposure being set differently?
Click to expand...
When I tested the lenses wide open on the SGpro I should have stopped down on the HVX more than I did. Compared to the wide open tests on the Letus, the SGpro test results were noticeably brighter due to my mistake and I didn't want to cause confusion.

When I ran tests we found one HVX to be 1/4 -1/3 stop faster than the other even though they had the same settings.
Click to expand...
We used two different HVX's for this test. I have no idea if one is faster than the other, I did not think that would be an issue or it would have been tested.

Also marco - where are you posting your images? Do you use the member upload center or post them on another site entirely?
Click to expand...
They're posted on my webserver.
 
Last edited:
If you're confused about stop loss, look at the pics with the taking lenses set to f/5.6. The Letus is clearly about a stop faster. We had identical zoom-in on both cameras and lens-to-subject distances.
 
Marco,
I don't doubt your results, and I can see the comparison you're describing. but I am surprised that different people have gotten such different results doing the same tests. I wonder if it is correct that the SGPro2 is the same as the 3. Wayne ought to know though.

I was quite surprised to find the exposure difference between 2 HVX's to be as big as ours were in tests. Don't know if it could be even more extreme or not, but that can screw up testing fast.

One issue in my experience is that the response of these adapters to f-stop changes on the 35mm lens is not linear - and my guess is it has something to do with how much the diffusing material is directing light from the wider parts of the iris (hope that makes sense.)

The result is that one adapter could be faster at 5.6 and another faster at f1.4. I know absolutely that this is true between f1.4 and f2.8 on the 35mm lens because I did alot of testng. The Brevis CF1L for example doesn't even see the difference between f1.4 and 2.8. Other adapters show a full 2 stops and some show only 1 stop. Its weird.


I didn't test all the way to 5.6 though. I seem to recall one of the adapters really responding very fast to changes below f4 so it was getting much darker as the stops got smaller. Might have been the SGPro but I don't recall for sure.

Did you use a waveform to match the exposures?

Lenny Levy
 
The Sg actually responds like it should. 1 stop down, is 1 stop down. Though at the very top, from F1.2 to 1.4... it doesnt respond linearly.
 
Thanks Tim, I do seem to remember that the SG and the Letus were relatively similar in linearity on my tests so that probably wouldn't be very important on these comparisons.

I'm pretty excited about your upcoming tests with SG Flip vs the Letus BTW. Any chance you can get a Brevis Flip to try on the SG as well? I thought there was someone out there in NYC who had one.

Lenny
 
Tim and i arent concerned with the brevis right now. we want to know which of the two adapters we own are the best.
 
I didn't get as big a difference as these tests. We tested on the same camera for the reasons Lenny stated. I suspect a discrepancy in the cameras' sensitivity.

1.2 to 1.4 is 1/2 a stop.
 
Timur.
Actually was only referring to the Brevis flip on an SG. I also love the look of the SG and am curious about the Letus flip vs the Brevis on the SG. Especially since the Brevis is supposed to correct for offset.

Hien implied to me he has something coming out soon that will allow a correction though so that should be good.


I'd like to test it myself but I can't afford to keep buying all of these things.

Anybody in the San Francisco area buying either Flip who wants to run some tests?

Lenny Levy
 
Actually after experiencing a bad left side edge sharpness on my LE, I've become interested in Brevis again now that I've heard they have an X/Y adjustment. This should be standard since image centering is not. It may be the one thing that'll have me sell my LE or my HVX.

Does anyone know if the XDcam EX is centered?
 
Tim Naylor said:
Actually after experiencing a bad left side edge sharpness on my LE, I've become interested in Brevis again now that I've heard they have an X/Y adjustment. This should be standard since image centering is not. It may be the one thing that'll have me sell my LE or my HVX.

Does anyone know if the XDcam EX is centered?
Click to expand...

I'm sure the Le brothers are already on the adjustment thing. Til then you could actually just zoom in a smidge to correct it. Admittedly, after I set up for 17mm (widest I have and impractically so) I have never had soft edges.
 
I only looked briefly at the EX-1 + Letus Extr but tomorrow I will have it up. I think mine was not centered however though maybe not as bad as on my HVX.
I agree Tim- its a big issue to me, but I also concur with Kholi that eventually Letus will fix it. When is the question and what will they do for those of us who bought the first version. Theya re pretty responsibe guys though so i expect the best.

Will there be any advantage to Brevis or Letus flip aside from that? i suspect they will be functionally similar, Brevis will prob be lighter, but optics will probably trump all for me.

I know there's a guy in NYC who said he had that Brevis flip for testing on one of these threads. Maybe you can find him.
 
Lenny, this isn't flaming but I have to be curious: if you can't tolerate the tiny amount of edge softness when using 28mm and 35mm (50mm+ doesn't exhibit this), how can you tolerate the edge softness inherent to the naked Brevis unit because the achromat's too close?
 
Kholi,
You may be confusing me with someone else. I haven't complained about the 28mm or 35mm or any lens in particular. Actually i haven't had time to test the individual lenses on the different adapters at all.

The issue I did grind into the ground (admittedly) is that all of the adapters I tested had a different optimum plane of focus for the center and the corners of the frame. I saw this for sure on the Brevis and on the Letus Extreme. My memory is that the SGpro was better but I wouldn't swear to it. I think my Brevis CF1L and my Letus are similar but I haven't checked carefully and I would have to check it with each lens. If anything memory tells me the Brevis was better but I could be completely wrong.

I also ground into the ground that I only trust focus when it is done on a chart after the 35mm lens is up. So i'm dubious about reports that one adapter is sharper than another unless a chart is used for careful set-up. Also is it center focus that's sharper or overall focus after finding a compromise back focus point?

I think a chart is the only way to really see what's going on. Its easy to download one as a file and get it blown up at Kinko's then carry it around.

For the record though, I don't like any soft corners. I don't know what you are saying about the Brevis having too strong an achromat though. You might be right but mine had pretty decent center to corner.
Alot of my bread and butter are interviews and it doesn't matter for that stuff.

- Lenny
 
You must log in or register to reply here.
Back
Top