No, they seem to be nearly identical. Since Evengelos said they were different adapters it might even be the same lens.
Which is looking more appealing?
- Thread starter Evangelos Achillopoulos
- Start date
No, they seem to be nearly identical. Since Evengelos said they were different adapters it might even be the same lens.
Evangelos Achillopoulos
Well-known member
I have to behave... and don't open my mouth... until tomorrow...
I just want to post something because otherwise I could not handle it...
Agrhhh.... :lipsrseal
I just want to post something because otherwise I could not handle it...
Agrhhh.... :lipsrseal
Ted Ramasola
Veteran
or yellowish green and magenta. depending on the optics used.
Its not necessarily found only on footage with adapters, the fujinon 16x of the the JVC HDxx series cameras exhibit the magenta/greenish CA at wide open, when its coupled with an adapter with a poor achromat, the CA gets pronounced.
Its not necessarily found only on footage with adapters, the fujinon 16x of the the JVC HDxx series cameras exhibit the magenta/greenish CA at wide open, when its coupled with an adapter with a poor achromat, the CA gets pronounced.
Last edited:
Evangelos Achillopoulos
Well-known member
lol... dude you better not be playing games to embarrass anyone... too much of that going on over here lately.
There is NO embarrass don't worry... my intention is to educate and learn both me and you guys...
Evangelos Achillopoulos
Well-known member
Whatever... but the polls are for the first impression... not if you put it on a microscope... first impression is what we vote...
Yup, gotcha loud and clear on that. Keep my vote how I originally placed it, A. Sorry, can't help pixel peeping at DVXuser.
Evangelos Achillopoulos
Well-known member
I want to thank all for participating in this experiment…
So its time to put the cat out of the bag…
Facts
On both cases the subject to camera distance was about the same with one foot difference, light setup was the same on both images.
A image.
Shoot with SGblade/GG.relay configured for full frame 35, front lens Canon FD 85mm F1.2 L @ wide open, camera GG.relay at F2.8 ½ which is the maximum without loss in resolution but with a slight bloom effect (the sweet spot would be F4 in which the relay perform the best). The ground glass was RR1 which is on this setup slower than RR3 at about ¾ of a stop.
B image
Shoot with Pro35 MKII, front lens ZEISS 50mm F1.3 @ wide open, iris of Pro35 was on ND 0.6 (Pro35 doesn’t has an aperture reading). The adapter has an academy framing.
Light performance
To compare the performance we should use RR3 that is faster than RR1, which is closer to Pro35 GG performance, by at least a 2/3 of a stop. To measure the exact difference we should use the same frame size (academy like Pro35 to judge accurately also the possible vignette) and lastly to use the same front lens in both setups because I have measure differences in light transition on this old Zeiss that some times are reaching a full stop brighter when compared to Canon or Nikon SLR lenses. Pro35 performs the same as GG.relay F4 when its stopped down by 2/3 of a stop (thats ND 0.2).
My measurements have shown a difference (for exactly the same lenses, RR3 in the same level of sharpness) of ½ stop, being faster the Pro35.
I can’t say that Letus Pro/ultimate is brighter than Pro35 if I take in account the sharpness. Someone could say ¼ of a stop faster the Letus Pro from Pro35 if you try few combinations with different lenses and be more forgiving on the edge to edge performance. It would be nice if someone post res charts for Letus Pro and compact. Even though I have access to the Letus products if I will do that everyone would say its not right.
The rest of the image and the core of the question.
First thing that someone can notice on A is the less compressed image (or let’s say crowded image) that’s due to use of FF35 optics and frame size. The image has better geometry and more natural. The background also is much better in terms of geometry. The longer lenses do the trick.
Second (that contest to be the first) on A, is the deeper DoF that results from the FF35 difference and the effects of the RR1. The DoF is so deep that adds to the slight blooming of the relay from the white shirt a tad more bloom due to defocus. Also there are some traces of CA but this is because of the maximum tolerable aperture of the relay and the wide open performance of the front lens.
When I was coloring the image I notice that the Pro35 had thinner blacks and it wasn’t so easy to put out details from the hairs of the subject, actually I tried hard to achieve what you see on B on the other hand A was much easier to get the details on blacks. That alone shows a better MTF on the A in the low contrast situations. SGblade /GGrelay will produce much more latitude in the image than Pro35 in low light situations, which will result more flexible image in CC and less noisy that in some cases would compensate the light loss. With Pro35 you will need more fill light to soften shadows, I reserve the opinion that thes might be from the front lens performance
To CC I was trying to much the gray background with a priority to the skin tone using a spilt screen on a calibrated D65 sRGB monitor in Color. The A was immediately more organic than the B without trying too hard. B was technically ok but something was missing… I have a feeling that the organic look on A comes from the slight blooming or deep DoF that makes more obvious the blooming effect.
My conclusion
What I notice when I finished was that the starting point of A was worse than the B and at the end when everything was corrected the A was the winner… that was what strike me and made me to start the poll… so 3 out of 5 would select A… so A is the winner and that is amazing…
My conclusion is that artifacts and imperfections when combined with accurate geometry and details on blacks are what make an image more appealing.
Even that the B was technically perfect and A was less perfect, A was the clear winner. Image A was more three dimensional and real.
If the test was with RR3 then the light performance would be much, much better and the DoF would be just a bit thinner. So a stop down of ½ for the GG.relay would vanish the blooming and the slight CA.
Moreover my conclusion is that FF35 gives sexier images and that is essential in our times to be able to choose between FF35 and S35 depending on then application.
So my design decision to make a modular relay that can easily be converted with minimal cost from FF35 to S35 was correct. Since FF35 competition in the rental market will be a major factor in the future.
P.S. Lenny you were right...
So its time to put the cat out of the bag…
Facts
On both cases the subject to camera distance was about the same with one foot difference, light setup was the same on both images.
A image.
Shoot with SGblade/GG.relay configured for full frame 35, front lens Canon FD 85mm F1.2 L @ wide open, camera GG.relay at F2.8 ½ which is the maximum without loss in resolution but with a slight bloom effect (the sweet spot would be F4 in which the relay perform the best). The ground glass was RR1 which is on this setup slower than RR3 at about ¾ of a stop.
B image
Shoot with Pro35 MKII, front lens ZEISS 50mm F1.3 @ wide open, iris of Pro35 was on ND 0.6 (Pro35 doesn’t has an aperture reading). The adapter has an academy framing.
Light performance
To compare the performance we should use RR3 that is faster than RR1, which is closer to Pro35 GG performance, by at least a 2/3 of a stop. To measure the exact difference we should use the same frame size (academy like Pro35 to judge accurately also the possible vignette) and lastly to use the same front lens in both setups because I have measure differences in light transition on this old Zeiss that some times are reaching a full stop brighter when compared to Canon or Nikon SLR lenses. Pro35 performs the same as GG.relay F4 when its stopped down by 2/3 of a stop (thats ND 0.2).
My measurements have shown a difference (for exactly the same lenses, RR3 in the same level of sharpness) of ½ stop, being faster the Pro35.
I can’t say that Letus Pro/ultimate is brighter than Pro35 if I take in account the sharpness. Someone could say ¼ of a stop faster the Letus Pro from Pro35 if you try few combinations with different lenses and be more forgiving on the edge to edge performance. It would be nice if someone post res charts for Letus Pro and compact. Even though I have access to the Letus products if I will do that everyone would say its not right.
The rest of the image and the core of the question.
First thing that someone can notice on A is the less compressed image (or let’s say crowded image) that’s due to use of FF35 optics and frame size. The image has better geometry and more natural. The background also is much better in terms of geometry. The longer lenses do the trick.
Second (that contest to be the first) on A, is the deeper DoF that results from the FF35 difference and the effects of the RR1. The DoF is so deep that adds to the slight blooming of the relay from the white shirt a tad more bloom due to defocus. Also there are some traces of CA but this is because of the maximum tolerable aperture of the relay and the wide open performance of the front lens.
When I was coloring the image I notice that the Pro35 had thinner blacks and it wasn’t so easy to put out details from the hairs of the subject, actually I tried hard to achieve what you see on B on the other hand A was much easier to get the details on blacks. That alone shows a better MTF on the A in the low contrast situations. SGblade /GGrelay will produce much more latitude in the image than Pro35 in low light situations, which will result more flexible image in CC and less noisy that in some cases would compensate the light loss. With Pro35 you will need more fill light to soften shadows, I reserve the opinion that thes might be from the front lens performance
To CC I was trying to much the gray background with a priority to the skin tone using a spilt screen on a calibrated D65 sRGB monitor in Color. The A was immediately more organic than the B without trying too hard. B was technically ok but something was missing… I have a feeling that the organic look on A comes from the slight blooming or deep DoF that makes more obvious the blooming effect.
My conclusion
What I notice when I finished was that the starting point of A was worse than the B and at the end when everything was corrected the A was the winner… that was what strike me and made me to start the poll… so 3 out of 5 would select A… so A is the winner and that is amazing…
My conclusion is that artifacts and imperfections when combined with accurate geometry and details on blacks are what make an image more appealing.
Even that the B was technically perfect and A was less perfect, A was the clear winner. Image A was more three dimensional and real.
If the test was with RR3 then the light performance would be much, much better and the DoF would be just a bit thinner. So a stop down of ½ for the GG.relay would vanish the blooming and the slight CA.
Moreover my conclusion is that FF35 gives sexier images and that is essential in our times to be able to choose between FF35 and S35 depending on then application.
So my design decision to make a modular relay that can easily be converted with minimal cost from FF35 to S35 was correct. Since FF35 competition in the rental market will be a major factor in the future.
P.S. Lenny you were right...
Last edited:
Evangelos, there isn't a thing you can derive from such comparison. Everything was different, no one thing was isolated. IMO the only thing you can get from it is that people chose A over B, barely.
That said, I think if the pose had been the same, and the grade had been closer, people would have chosen B.
I do however agree with this:
But I think you are on to something since this touches on a big point of that whole "is FF35 the future of cinema?", which makes it very interesting.
I love to see a poll with the same adapter, same GG, same camera, using primes from a close matching set, but set up for FF35 and S35 recorded areas on the GG and fine tuned in post to match as perfectly as possible. Isolating geometrical representation.
That said, I think if the pose had been the same, and the grade had been closer, people would have chosen B.
I do however agree with this:
I didn't consider you might be shooting different projected areas on the GGs. Not sure if it is the first thing people notice though (until you point it out). I think the first thing that pops out is the subject's pose and expression, and color.
But I think you are on to something since this touches on a big point of that whole "is FF35 the future of cinema?", which makes it very interesting.
I love to see a poll with the same adapter, same GG, same camera, using primes from a close matching set, but set up for FF35 and S35 recorded areas on the GG and fine tuned in post to match as perfectly as possible. Isolating geometrical representation.
Last edited:
Lenilenapi
Carbonite Member
Evangelos,
I think I guessed correctly because when you originally posted this in another thread it was in the context of a discussion about FF vs Academy.
I must confess I got a little lost in the description of all the parameters and it reminded me how hard it is to do comparisons of this sort. Jeff (Dino) and I ran into this when trying to compare the B4 Pro35 with Zeiss lenses to a Letus B4 relay with Ultimate/Nikon, and a Letus Extreme/Nikon on an Ex-1. At one point we pointed my Letus Extreme on an EX-1 and the Pro35 at the bright garage windows. The Ex/Letus combo showed a kind of bloom of overexposure with CA blue edges - not good . The Pro35 was clean but showed far more halation and blooming that spread throughout the picture. Where did the Ex-1 CA come from - the wide open Nikon, the Letus optics , the EX ? Was the bloom on the Pro 35 due to a much more diffuse GG?
If anything what I come away with is that facile generalizations often hide complex truths, it seems your new relay looks pretty good at first glance and above all that you are working damn hard over there and putting your own products to some honest and rigorous testing. Can't remember ever seeing a manufacturer post comparisons with the competition when the results were actually ambiguous.
That was fun. What's next weeks puzzler going to be?
Lenny Levy
I think I guessed correctly because when you originally posted this in another thread it was in the context of a discussion about FF vs Academy.
I must confess I got a little lost in the description of all the parameters and it reminded me how hard it is to do comparisons of this sort. Jeff (Dino) and I ran into this when trying to compare the B4 Pro35 with Zeiss lenses to a Letus B4 relay with Ultimate/Nikon, and a Letus Extreme/Nikon on an Ex-1. At one point we pointed my Letus Extreme on an EX-1 and the Pro35 at the bright garage windows. The Ex/Letus combo showed a kind of bloom of overexposure with CA blue edges - not good . The Pro35 was clean but showed far more halation and blooming that spread throughout the picture. Where did the Ex-1 CA come from - the wide open Nikon, the Letus optics , the EX ? Was the bloom on the Pro 35 due to a much more diffuse GG?
If anything what I come away with is that facile generalizations often hide complex truths, it seems your new relay looks pretty good at first glance and above all that you are working damn hard over there and putting your own products to some honest and rigorous testing. Can't remember ever seeing a manufacturer post comparisons with the competition when the results were actually ambiguous.
That was fun. What's next weeks puzzler going to be?
Lenny Levy
Last edited:
Evangelos Achillopoulos
Well-known member
I want to add something to that...
This test wasn't supposed to be shown because of the totally wrong setup of the SGblade/GG.realy to be compared (as Roberto points) with Pro35... that was a quick test that Chris done in Israel totally out of my loop... we discussed the problems afterwards and he agreed also that is not appropriate to show...
BUT when we had the above discussion (that Lenny points) and due to the engineering honesty that drives me, when I saw the starting point and then after CC how striking it was, I thought that is a good test to make it public... to understand and educate me as well as you guys... so I took the opportunity and post the poll...
Yes these tests are very difficult to perform, I have done them for my internal use just to understand where am I... I don't know if there is a meaning to post them since I have a competing product. The thing is that I know very well were I'm standing. My aim as I had said in the past was to make a solution that is balancing all the options to the optimum by taking in account the current state of the art and how business is changing.
So I build a component (the GG.relay) that is modular, it maximizes the client investment, it is shorter than the big Pro's (in S35 setup), it is light efficient as the equivalent compact (or better) and its future proof...
What would be next? I don't know probably my new web... with a big FaQ for relays and adapters...
Roberto I'm glad that you got something from this simple test even though it wasn't too professional...
Yes, it was glimpse of the FF35 future that we have to learn to respect...
Last edited:
Of course there is the whole harder to focus debate. But, it is very nice to see side by side. However it would have to be the same quality of lenses, really just two distortion "free" lenses. So if both lenses had nearly identical distortion values and everything else was also the same, then you could really isolate the physical S35 vs. FF35 element which, as you point out, is a BIG deal!
One more thing though, I would probably do a series, a pair more at the wide end, a normal pair, and a long pair... since 85mm is considered the sweet spot and the most loved length.
One more thing though, I would probably do a series, a pair more at the wide end, a normal pair, and a long pair... since 85mm is considered the sweet spot and the most loved length.
Lenilenapi
Carbonite Member
If I get Evangelo's adapter I hope to do a comparison with his Relay/Blade on a Panny 3700 next to EX-1 ( both with Nikons) and a 7D and 5D, hopefully also get together with Dino to compare to his Letus B4 relay. Pro35??? if we have time & resources.
Easier said than done. These tests are incredibly time consuming and its very hard to isolate the variables. Ever notice how many people here have described very thorough comparisons they planned to make but you never saw the results?
Lenny
Easier said than done. These tests are incredibly time consuming and its very hard to isolate the variables. Ever notice how many people here have described very thorough comparisons they planned to make but you never saw the results?
Lenny
I know. Happens all the time. This one for example doesn't sound likely:
:violent5:
He doesn't have to. The whole FF35 vs. S35 thing sounds interesting so I am just pointing out that because 85mm is the popular kid on the block to really isolate the differences between S35 and FF35 you'd need a few pairs.
:beer:
Lenilenapi
Carbonite Member
You've laid down a gauntlet I see - I make no promises and don't hold your breath - but there's a chance it could all happen. I have done a lot of these in the past between Letus, Brevis and Blade.
Actually even harder than shooting the test is recording and posting the results.
But I'll probably stick to one lens. You're right about multiple lenses to really be exact, its just such a pain I know I won't do it.
Lenny