Does Letus Extreme lose more light than Brevis

[Barry_Green]

Forgive the frustration. Several claims have been made that appear to defy scientific fact -- such as the idea that using an adapter can actually "gain" light.

So I guess I'm faced with two choices -- either let seemingly outrageous claims go unchallenged, or have to buy every product that comes on the market just to find out if it can do what it says it can. Fine. I'll buy a Brevis. I have no need for it, I've bought and/or tested and/or used every adapter on the market but the MovieTube, I've tested them in all manner of circumstances and find them all to behave similarly to one another and in accordance with the laws of physics. But now we have one particular product that makes claims (either from the manufacturer or from users) that simply do not correlate with the rest of the observable body of evidence, and according to St. Mad and others, I'm not allowed to point out that something just simply doesn't make sense here? That there is no "free lunch," there's always a piper to pay, there's always a compromise or a tradeoff, and if you're seeing evidence that seemingly defies observable and reasonable/rational explanation, there's got to be some downside or side effect.

And the only way to get to the bottom of it is to test it. I've taken the rational approach of repeatedly asking around for other users to loan a unit so I could get to the bottom of it, and that hasn't happened (which is, of course, perfectly understandable). I guess the only recourse is to torch a thousand bucks?

Fine. So be it. Maybe while I'm at it I'll torch another seven thousand on an EX1 just so I can be allowed to talk about that emperor's clothes too?

I mean, what's the alternative? I guess this is the same spot lenny came to, where he had to buy three of 'em. And even then we get conflicting reports where Tim Naylor and Lenny's tests directly contradict each other... maybe there's no other way.

 

[Kholi]

Part of the claim was that it behaves in this extraordinary way while the others don't.

Also, I was trying to contribute a solution - to get a ball rolling.

Thanks for the input.

Others don't. My LetusEx doesn't defy physics. My M2 Didn't defy physics. I get what you're saying, the actual point was that the Brevis was the one that he wanted to test about the defying o'physics.

Which means it's as simple as getting Barry a unit and the Cinescreen in question to test for himself.

 

[jenningsp]

do it barry, do it! buy that ex1 and the brevis. do you own an sgpro? buy that too!!!

does anyone know why the spinners footage looks 100x more awesome than ALL of the non-spinners...? it's not just the bokeh or the extra diffusion... it's something else...

for some reason the "in-focus" stuff from the non-spinners looks like it's video. so if you've got a scene where everything is in focus, it looks like video...
but with a spinner, all the footage looks awesome!!!! all the in-focus stuff looks like silk.

i'm not sure why this is... everything i've scene from the M2 and the SGpro looks awesome! but half the stuff i see from the brevis and the letus just doesn't excite me, some of it does look awesome, but some doesn't, it's about 50/50.

i own a letus35a myself and it exhibits the same non-awesomeness quality as the other non-spinners. i've shot some awesome footage with it, but i've also shot some non-awesome stuff.

i am one of those people who obsesses about how my image looks. i can see the difference very clearly. i know some other people can too. i just want to know why?

awesome

 

[Kholi]

do it barry, do it! buy that ex1 and the brevis. do you own an sgpro? buy that too!!!

does anyone know why the spinners footage looks 100x more awesome than ALL of the non-spinners...? it's not just the bokeh or the extra diffusion... it's something else...

for some reason the "in-focus" stuff from the non-spinners looks like it's video. so if you've got a scene where everything is in focus, it looks like video...
but with a spinner, all the footage looks awesome!!!! all the in-focus stuff looks like silk.

i'm not sure why this is... everything i've scene from the M2 and the SGpro looks awesome! but half the stuff i see from the brevis and the letus just doesn't excite me, some of it does look awesome, but some doesn't, it's about 50/50.

i own a letus35a myself and it exhibits the same non-awesomeness quality as the other non-spinners. i've shot some awesome footage with it, but i've also shot some non-awesome stuff.

i am one of those people who obsesses about how my image looks. i can see the difference very clearly. i know some other people can too. i just want to know why?

awesome

You're blaming the adapter for something the shooter takes responsibility for. Gonna tell me that Phillip Blooms LetusEX footage looks like that?

AND, upon looking at the Phillip Bloom Brevis shots further... My theory is that it's a huge hot spot.

 

[jenningsp]

i've seen most of phill's footage. his letus and brevis stuff with CF1 and CF3 exhibits the same thing i'm seeing on the other footage...

phill tends to shoot static shots, and the "difference" i'm seeing only really becomes apparent when everything starts moving, like people walking, moving water (this is a big one) that sort of thing...

also nighttime stuff looks a lot better than daytime stuff with the non-spinners.

phill likes to add vignetting in post to a lot of his footage. so that might explain the "hot spotting"

again it's 50/50, half the stuff from non-spinners looks great, the other half doesn't look so great. it's all better than no adapter at all though

 

[Kholi]

i've seen most of phill's footage. his letus and brevis stuff with CF1 and CF3 exhibits the same thing i'm seeing on the other footage...

phill tends to shoot static shots, and the "difference" i'm seeing only really becomes apparent when everything starts moving, like people walking, moving water (this is a big one) that sort of thing...

also nighttime stuff looks a lot better than daytime stuff with the non-spinners.

phill likes to add vignetting in post to a lot of his footage. so that might explain the "hot spotting"

again it's 50/50, half the stuff from non-spinners looks great, the other half doesn't look so great. it's all better than no adapter at all though

Definitely not doubting your observations. I notice a lot of stuff that others tend not to as well so I'll keep an eye out. I think most of it has to do with things outside of the adapter realm, though: composition, shot choice, etc. There's lighting to consider: The M2 has been used on more narrative work than any other adapter. Given that track record, you're chance of seeing really GOOD work versus Run-of-the-mill-back-yard-trees is greater.

A great example of narrative work with the LetusEX is Jack Daniel Stanley's "A Little Mouth To Feed".

 

[Douglas Villalba]

I don't always agree with Barry, but I think that anyone that has knowledge of how light works has to agree with him on this one.

Simply put my HVX without an adaptor has a ASA (ISO) rating of 320. Now add the LEX with a lens and it drops down to 200. Brevis user and builder claim that at f 2.8 the Brevis is between 1 and 2 f stops faster than the other adaptors. If the Brevis is only one stop faster the LEX (I use the LEX because I own one) that would make the ASA raring 400 on the HVX. That would mean that the Brevis is creating light since without and adaptor I started with an ASA rating of 320.

I could understand that their GG is not too sensitive to over exposure and you see no actual difference from f 1.4 - 2.8.

If I were a manufacture of an adaptor and really believe my claims, I would jump on a plain to show Barry my product. Even if just for the exposure I would get my product.

I believe that all the adaptors have + & -. Some LOOSE less light than others. Some you are able to use faster shutter speed, but by the law of light traveling they all have to loose light.

I would also offer my studio in Miami to conduct a review just for fun. I already own the LEX and HVX. My friend owns the M2, LEX and recently ordered the SG. If you would like be included just drop me a note. Preferably if you own any of the other adaptors.

 

[jenningsp]

yes the context, professionalism, composition and lighting have a great deal to do with how the image looks. but what I'm talking about goes beyond that.

it's an inherent trait that spinners give the image. there is a quality that i can't quite put my finger on. i'll call it "awesomeness" for now...

different adapters, namely their GG's, give a certain "look" too the image. and for some reason spinning adapters look "awesome" 100% of the time and non-spinners look "awesome" only 50% of the time.

i actually watched half of "a little mouth to feed" when it came out, but my download stuffed up and never finished it. still, it still had that same quality i'm talking about.

 

[USLatin]

Barry... are you really going to buy a Brevis? Dennis, why don't you let Barry pay shipping and let him have a Brevis for weekend free of change if at all possible... this should definitely be nothing but good for your product, especially if it is finally proclaimed the fastest adapter out there. I think a review by Barry that concluded just that would translate to the acceptance of it's supremacy in terms of speed for 80% + of DVXuser members. That definitely can't be bad for sales :thumbsup:

Jenningsp, maybe it has to do with consistency on all the GG's effects on the picture all the way to the shorter shutters given the higher speed of the grain?

 

[Dennis Wood]

Ok, here it is. Forgive the audio but we fired this together at the end of the day...should be fairly self explanatory. Please right click and "save as". The file is a 35MB quicktime file that runs about eight minutes.

http://www.cinevate.com/images/lightloss.mov

The camera used was the XH-A1, with no presets loaded (default config) with everything set to manual. AGC was off of course. We've tested on 3 different cameras (HV20, XH-A1, XL2) to confirm that when baselined to approx 50 IRE on the waveform monitor, a 1 stop difference (on the camera)corresponds to ~20 IRE on the monitor. We used Adobe OnLocation on the laptop, with output directed via HDMI to the 42" LCD so you could see clearly what was going on.

Seriously, we don't ask folks to buy the adapter for reviewing purposes...never have, and never will As Barry has already mentioned, we were all lined up to send one for the shootout...and we were apparently the only adapter manufacturer that was willing to do that. At least one third-party review of the flip vs non-flip Brevis has been done, and should be up any day now.

Please, if you've invested your hard earned dollars in one solution or the other, then the best thing you can do is shoot with it, learn it's quirks and limitations, and ignore this thread. If you'd like to see what I think is some very cool stuff, check the clip...and try it with what you're using. At the worst, you'll figure out how much light to bring along, or leave at home for your next shoot Does it lose light? Sure. Does it lose what you'd predict based on convention (that doesn't apply here)...absolutely not. Moving from f1.4 to f5.6 on the 50mm lens mounted to the Brevis flip with CF1Le resulted in about the same light loss as stopping the bare camera down 1 stop.

 

[Kholi]

Okay, I suck at reading waveforms and was hoping to see some actual footage. But as mucha s I suck at reading waveforms, does it read that with the Brevis on there's a non-uniformity with light to begin with.

The sides are lower than the center, and the bar's got lumps in it from the get go.

Hot Spotting?

Edit: Considering if it were a hotspot in the Cinescreen (is that what it's called?) then the hotspot blows colors out? Is that why the colors of the Brevis Unit in Phil Bloom's tests look strange?

Does anyone have a Brevis unit in Los Angeles?

 

[USLatin]

Dennis! I see you! hehe... thanks for the little video!

I am thinking aloud here... does this behavior happen with any lens or only with certain ones? Or does it happen with all lenses and just slightly differently with each?

Okay, I suck at reading waveforms and was hoping to see some actual footage. But as mucha s I suck at reading waveforms, does it read that with the Brevis on there's a non-uniformity with light to begin with.

The sides are lower than the center, and the bar's got lumps in it from the get go.

Hot Spotting?

Yes, it seems the center and either side are slightly different, but at times it was backwards... so not always necessarily "hot" spotting... I'd love to see a white page on a waveform with a different adapter to see if this is also present and just how pronounced if it is

 

[USLatin]

I wonder if the adapter not being screwed in has something to do with some of this edge fall-off?

but man... a little hot-spotting for a big chunk of light...? I gotta see some of that flipped HD hosted somewhere! I have to look for that drop-off on the edges and gauge how pronounced they are...

but what I'd really like to see is some footage of the same subject (something that would help point out hot-spotting) and see the same thing all different stops on two or three lenses, one being a focal length far from 50mm just to change things up a bit

 

[Dennis Wood]

The behaviour changes of course with different lenses, but the relative exposure curve is very similar. We always baseline with the 50mm f1.4 as it's nearly ubiquitous. I've always indicated that slower zoom lenses work amazingly will with CF1Le...and that light efficiency is why.

The sharp edge falloff on the right/left represent the black edges of our imaging element gate in the overscan area. The thickness of the line is coming from an underexposed imaging element...note the shutter speed! What you're seeing there is the grain itself in the imaging element that was not moving. You can actually use the scope with a stopped down 35mm lens to set dead-accurate back focus by observing the grain effect on the scope.

 

[Kholi]

The behaviour changes of course with differnent lenses, but the relative exposure curve is very similar. I've always indicated that slower zoom lenses work amazingly will with CF1Le...and that light efficiency is why.

The sharp edge falloff on the right/left represent the black edges of our imaging element gate in the overscan area. The thickness of the line is coming from an underexposed imaging element...note the shutter speed! What you're seeing there is the grain itself in the imaging element that was not moving. You can actually use the scope with a stopped down 35mm lens to set dead-accurate back focus by observing the grain effect on the scope.

Thanks for explaining that.

 

[USLatin]

Ok, but the higher exposure should still be a straight line till it gets to the very edges and overscan yet I see an ever so slight belly in the middle... it looks to be about 5% difference at most which given that it is a gradual curve it should be very hard to notice and it isn't as pronounced under all settings... right?

1.4f and 2f and 2.8f
seems to be a bit darker in the center by far less than 5%, and also the curve seems to be completely consistent at all of these

4f
seems to be completely flat

5.6f
seems to have a slightly irregular hot spot in the center

Now if this is from the grain just as the line is not a thin line from grain adding "dark spots" of grain all over the screen and it is the grain that will have the hot spot in the center at 5.6f or ever so slight "cold" spot in the center at 1.4, 2, and 2.8 then there will be irregularities only in the exposure of the grain across the screen... but if the grain is not consistent then the image projected isn't consistent either because they are one in the same...

which also brings me to what I think is a realization that the GG allows for different "exposures of the same area by one "grain" being darker than the one next to it and this might help mimic the dynamic range of our eyes? well... I'd ignore that las thing if you know what's good for you cause it is a thought of which I have little grasp of if at all and even if I were able to explain it it is veryl likely WRONG anywhay

 

[USLatin]

BTW it looks like I might have to get more ND filters than just the one filter equivalent to half the built-in ND's in the HVX which I was planning on getting!

I really want to see some flipped Brevis HD shot on the HVX now!

hehe...

 

[Lenilenapi]

Look guys therte is absolutely nothing in what I have described that defies any laws of physics or optics. Now I do think it has sometimes been described inaccurately as "gaining light" which is I agree is impossible. Dennis may have used that expression at one point and I told him I thought he was describing it backwards. But if you'll bear with me a moment I will try to explain what is happening as I see it.

The Brevis CF1L with its less dense screen seems to be "faster" than the others which makes perfect sense as it is diffusing less thus allowing more direct light through. With the 35mm lens at 2.8 the difference in speed (or better "transmission") is noticable so the Brevis seemed perhaps as much as a stop faster in my tests. It is also noticably a less "diffused" image which by the way is not neccessarily the same as sharpness

However related to the greater transmission it seems to not react to the wider f-stop as you go from f 2.8 to f1.4. This is NOT a physical impossibility and presumably it must have something to do with the viewing angle we are at with respect to the image on the screen. It is probably akin to vignetting but it is probably not quite the same and Please Please Please - don't use this as an excuse to claim the Brevis will vignette more, because I haven't seen that. Its possible but I haven't seen it, nor have my clients. Lets make vignetting another thread OK.

At any rate, because it doesn't react to the wide f stops, it loses its speed advantage as we open up to f1.4, because for some reason the denser screens seem to pick up more of the light from the wider apertures.

This is not illogical or crazy in the least. You all agree that you have seen this phenomena on the other adapters but not to such a degree. Barry noted that some don't seem to notice anything from f2 to f1.4.
I seem to recall that the Redrock ( the densest of these screens) reacted the most linear between f2.8 and f1.4. I believe it was almost 2 full stops. Thus the Redrock became comparatively faster with the wide open lenses.

Though I haven't done any careful tests , I have noticed that there can be non-linear behaviour on the other end as you close down the iris. On one of these guys I remember seeing the light reduce more and more as the f stop narrowed. Can't remember which one.

Does this make any sense to you guys.
As with the DOF issue - these adapters act in odd ways that we don't expect. It doesn't neccessarily mean that the observers are quacks, but that we need to change our understanding. That said it also important to be skeptical and not trust every wacky claim you read on these pages.

By the way. I have no experience with any other claims or tests run by Dennis or anyone else on this forum, and I can't get Dennis test to play so a healthy "no comment"from me.

Lenny Levy

 

[Lenilenapi]

Incidently, I don't think Tim's and my tests were so far off as to be in "direct contradiction" His Letus was a touch faster, and mine a touch slower than the SGPro but they were less than a 1/2 stop off the SG in both cases. and I was using a Rev 2 and he had a Rev 3.

Wayne swears they are the same but Tim thought that his Rev3 was maybe slower than his previous version. Everything else about Tim's test I pretty much concurred with, though I felt it was unfair to talk about sharpness without carefully checking focus on a chart as part of the test.


lenny

 

[Michael Friedman]

The math Barry cited applies if you were shooting just the imaging element from the front (like a movie screen).

Instead, you are shooting through it (it is translucent) which means that your camera is still gathering light from the outside world. This accounts for little to no change at wide exposures since the opening of the aperture is still (likely) as wide as the imaging element. You are 'seeing' straight through - edge to edge.