underexposing with the sgpro.

jenningsp

jenningsp

Well-known member
underexposing with the sgpro.

got my sgpro a few days ago and borrowed a camera off my friend.

i've never trusted other peoples light loss test. now you can not trust me :D
the sgpro loses 1 stop with a f/1.4 lens.
it is also important to note that you lose about 2/3rds of a stop from zooming in as well. but no body counts that.

here are some stills
all shot with a nikkor 35mm f/1.4 AI and the sony A1p in HDV 25p. 6dB of gain. no post.


1204731991.jpg

1204732099.jpg

1204732154.jpg

1204732233.jpg
 
Last edited:
cool grabs.

But the zoom in factor is quite significant in discussing the total light loss per adapter. However since each camera needs a different amount of zoom, it is difficult to put a finger on exactly the total light loss.

But the fact that you have 6db gain is adding light, so the total light loss still seems to be around 2 stops.
 
PaPa said:
But the fact that you have 6db gain is adding light, so the total light loss still seems to be around 2 stops.
Click to expand...

don't understand your logic here. total light loss with an f/1.4 lens at f/1.4 is 1.66 stops (if you choose to include the zoom loss)

most people state the adapters light loss without the zoom loss. so it's total light loss is 1 stop with an f/1.4 lens at f/1.4.

then you've got the people who minus the lens' light loss. an f/1.4 lens loses 1 stop. so if you minus that, you could say the sgpro loses 0 stops.

YAY!!!! :D
 
Knowing that the Letus extreme does in fact lose less light, I will explain how it works with my XH A1

The Adapter on it's own loses approximately a half stop. My camera's widest aperture with at full wide is F1.6. Generally speaking, i shoot at -3db gain. To zoom in to get full frame, the camera must zoom to Z78, which puts my cameras aperture from 1.6, to 2.6 ( which is equivalent to a stop and a third ), I generally shoot at F2 on my 35mm lens, I am not sure how much light that loses, that means in the end, we are losing ( even htough the adapter is extremely efficient ) a total of approximately 2.5 stops. Now i could compensate by adding gain from my -3db but then i have to decide if i want to add grain.

That to say, i don't find it the same by adding gain to the camera and then saying that my adapter now loses less light because i am compensating by camera gain, because you are effectively doing the same thing as boosting in post. And since you are shooting at +6db gain with the adapter, you are therefor also compensating.

Unless i have misunderstood something?
 
ignore the 6db of gain. that's only refering to the stills, not my light loss tests.

it sounds as if my sgpro loses less light than your lex...
 
so far the tests have proven the opposite, which is what leads me to believe there is something wonky in one of our calculations.
 
Jenningsp,
its great to have more people post there findings about light loss, but unless you explain in detail how you arrived at that conclusion it just goes on the junk heap of everybody else's .

Lenny Levy
 
Jenningsp, that is sure a lot of noise, ouch.... unless thats the look you going for... congrats on the Sgpro


Jon
 
matty g - the sony a1p. the pal version of the sony a1u. like the z1/fx1 the pal version does a decent fake 25p. the ntsc version is useless.

its native iso is like 80 so it's not so crash hot in lowlight. hence all the noise jon.

and lenny/papa, trust me. i know what i'm doing :D
 
Even using the HV20 and the Sgpro in low light situations didnt give me that kind of noise....


Hmmm, shows Ill never buy a sony A1


Jon
 
a lot of the noise is because i under exposed and used b-press. the hv20's iso is about 120. it's also got a bigger chip and more resolution. so it is better in lowlight than the a1p.
 
jenning, its not that i am saying that you do not know what you're doing, but what i would like ot know from you are the following:

1) What is your camera's aperture at full wide?
2) What is your cameras aperture once you zoom in to academy frame?
Record how many stops of light you lose

Now, Take the naked camera and expose for medium grey. Aim to expose for F 8.
Now without adding any sort of gain, add the sgpro, a lens at F2, zoom in, and again, expose for medium grey and see now what you're camera's f stop is at. Because the fact is that the letus extreme does not lose more light than the SG pro, this is seen visible by the type of bokeh the sgpro produces.

If you manage to get around to doing this, please post your findings. Because as Leni said, unless you can explain how you came to your conclusion about the light loss of the sgpro, there is really no credibility.
 
Last edited:
alright. i got out my waveform monitor and did some super duper accurate tests for you papa. i double checked everything and it was all done under very controlled circumstances. :D

i evenly lit a wall and set up may camera on a tripod. here are the results.

a difference of 25 on the waveform is 1 stop for this camera.


stock camera wide open zoomed out = 75
stock camera wide open zoomed into academy = 63
sgpro f/1.4 = 40
sgpro f/2.0 = 38
sgpro f/2.8 = 30


so as you can see the total light loss including the zooming and sgpro and an f/2.0 lens is 37 or 1.5 stops.

it's interesting that there is basically no change from f/1.4 to f/2.0. i've seen this with the brevis but i didn't think the sgpro would react like this.

what do you think?
 
Jenningsp,

"a difference of 25 on the waveform is 1 stop for this camera."

Where did you get this information from? I doubt its completely accurate across more than very few stops and only points to the difficulty of doing these tests.

No camera will be consistent across the IRE range regarding how much difference a stop makes.

In general higher exposures show more change as you open the iris and lower values show much less. That is unless you get into the knee.
Thus it would be very unusual if the difference between 60 and 80 is the same as 20 to 40. It all depends on your camera's gamma curves though. You can see it on a chip chart.

Barry suggested a long time ago the only accurate way to judge exposure and light loss would be to always bring the camera to the same point on the waveform by using a dimmer or some other way to alter the light (move it or add a second light etc). Then read the light levels with a meter.
If it takes double the footcandles of light to get to the same point on the waveform then you've lost 1 stop of light. its accurate and that's what I tried to do in my tests.

What you noticed with the SG though I do believe. i didn't bother to check between 1.4 and 2.0 but I've always said that this happened with all adapters not just the Brevis. It is only more extreme on the Brevis but they alll do it to some extent. Its in the nature of the optics of these adapters. One more reason i was irritated when people called the Brevis impossible and trickery.

Keep testing man.

lenny levy
 
lenny. i got "25" from looking at my waveform monitor and opening up the aperture on the stock camera 1 stop. i then double checked it by adding 6dB of gain to make sure it was the same. 6dB of gain being 1 stop.

yeah the linearity of the curve differs but i'd say there was less than a + or - 5% difference.. i did these test several different ways. by opening aperture, by adding gain and then finally using a spot meter through the adapter. it's all pretty spot on. :D

i took your advice and tested the center and edge focus on the gg. MY GOD does it make a difference. if you push just past the center point you get an average focus and it's awesome. can't thank you enough for that one.

i also spent a few hours with a monitor testing what was the minimum aperture and shutter you could use before the gg showed up in a bright sky.

i won't bore you with all the details but with a shutter of 1/50 and 1/100 don't go past f/4. f/4 is good. but don't got past it. in 99% of your footage you'll be fine up to f/32 but in those bright sky shots. f/4 is the the minimum. trust me.

more testing!!!
 
Sounds like you double checked that stuff carefully.
1.5 stops is more than I saw on the HVX, wonder if its the difference between cameras.

Are you in academy on the SG?

Lenny
 
mostly using academy for now because the little sony A1p vignettes like a bastard with the sgpro.
 
You must log in or register to reply here.
Back
Top