M2 Footage looks awful
- Thread starter jdorsey
- Start date
That may well be, but the camera's ND filter simply darkens an overexposed image that has already lost some (or much) of its definition when it was projected onto the GG of the DOF adapter.Matty_g said:
If you were to put a ND filter in front of the lens attached to the DOF adapter, you'd get a better image.
Lenilenapi
Carbonite Member
I've heard this before from Dennis at Brevis about not using the in -camera filters but whatever the problem is it must be subtle because I haven't seen it yet.
If you did a comparison test what would the internal filter version look like that would be so bad? I haven't done any careful comparisons but it's always looked fine when i was doing set-up tests.
Re: focus - I always ( underline ALWAYS) carry a blowup poster of a resolution test chart (kinkos will do it cheap) on my jobs with the adapter. it makes for a very easy way to double check your focus. I ALWAYS double check that frequently during the day . The nice thing is that it you can do it with the lens on . I only do a focus on the ground glass once to get me close. its much easier to find the sweet spot between corners and center with the chart. It's more prescise and confirms that you haven't drifted.
jdorsey, - How much time did you spend with the adapter before going to shoot with it? It takes time to get used to. Also exterior panoramas will only reveal the softness and aberrations inherent in the adapters.
- Lenny Levy
If you did a comparison test what would the internal filter version look like that would be so bad? I haven't done any careful comparisons but it's always looked fine when i was doing set-up tests.
Re: focus - I always ( underline ALWAYS) carry a blowup poster of a resolution test chart (kinkos will do it cheap) on my jobs with the adapter. it makes for a very easy way to double check your focus. I ALWAYS double check that frequently during the day . The nice thing is that it you can do it with the lens on . I only do a focus on the ground glass once to get me close. its much easier to find the sweet spot between corners and center with the chart. It's more prescise and confirms that you haven't drifted.
jdorsey, - How much time did you spend with the adapter before going to shoot with it? It takes time to get used to. Also exterior panoramas will only reveal the softness and aberrations inherent in the adapters.
- Lenny Levy
jdorsey
Active member
ozduc said:
When you asked the first time, I forgot that I at some point--though I'm not sure if it was with this shot--I switched out the lens to my Nikon 28-35mm which had a polarizer on it.
I took two shots of this particular golf hole. One with the polarizer and one definitely without--both look equally *bad*.
jdorsey
Active member
Lenilenapi said:
I spent very little time with the adapter...probably 15 min. of footage total...and most of that I was testing other things. I had two good jobs lined up before I ever tested my camera or the M2. It's no excuse for not doing more testing...but I was unaware of how delicate and involved the whole process is--and, obviously, overconfident.
jdorsey said:
What were your camera settings. Video grain like that is caused by underexposure. I have used Nikon lenses for years and the 35-70 zoom is about the most unsharp lens that Nikon made. I had one but sold it years ago.
Try using faster, sharper lenses. I have seen some good results with the M2 but it does take time and experience to set-up properly for best results.
siniarch
Veteran
Hey Guys,
Though I agree with you guys that you should NOT use the ND filter in the camera to controll exposure, I was corrected on this point to some degree.
The problem is subtle because the Ground Glass does not behave as we think it does. We (or rather I) think (thought) that the Ground Glass was an element much like the CCD chips on the camera. That it has limits. Like the CCD. so if it gets over exposed, it cuts off the anything brighter than 100 ire, or darker than 0 ire. but the fact is that the Ground Glass is more like a lens element. When underexposed or over exposed, it doesn't clip the highlights or shadows. So running an ND filter after the fact will still show detail on the Highlights.
But No matter what people say, I do believe that the Ground Glass Losses some resolution. it is after all ground glass. So, some of the detail might be lost and some of the light might be blended once going through it. So I cannot imagine that an ND filter after the Ground Glass will look "EXACTLY" the same as a ND filter before the Ground Glass. The Key word is "EXACTLY". I do believe thought that the difference is so minute that most people won't notice it, and that for the most part it shouldn't make a difference.
Barry Green corrected my thinking about the GG not clipping any light. He in fact did a test where he exposed the image correctly through the lens with an ND filter in front of the lens, and then he over expose the image on the GG by 256x(or something like that) of light and corrected it with the ND in the camera and found that the highlights showed the detail after the ND was turned on.
Again I can't imagine it was exactly the same. But he couldn't see the difference.
my 2cents.
Though I agree with you guys that you should NOT use the ND filter in the camera to controll exposure, I was corrected on this point to some degree.
The problem is subtle because the Ground Glass does not behave as we think it does. We (or rather I) think (thought) that the Ground Glass was an element much like the CCD chips on the camera. That it has limits. Like the CCD. so if it gets over exposed, it cuts off the anything brighter than 100 ire, or darker than 0 ire. but the fact is that the Ground Glass is more like a lens element. When underexposed or over exposed, it doesn't clip the highlights or shadows. So running an ND filter after the fact will still show detail on the Highlights.
But No matter what people say, I do believe that the Ground Glass Losses some resolution. it is after all ground glass. So, some of the detail might be lost and some of the light might be blended once going through it. So I cannot imagine that an ND filter after the Ground Glass will look "EXACTLY" the same as a ND filter before the Ground Glass. The Key word is "EXACTLY". I do believe thought that the difference is so minute that most people won't notice it, and that for the most part it shouldn't make a difference.
Barry Green corrected my thinking about the GG not clipping any light. He in fact did a test where he exposed the image correctly through the lens with an ND filter in front of the lens, and then he over expose the image on the GG by 256x(or something like that) of light and corrected it with the ND in the camera and found that the highlights showed the detail after the ND was turned on.
Again I can't imagine it was exactly the same. But he couldn't see the difference.
my 2cents.
jdorsey
Active member
HVXguy said:
My Settings:
SCENE NAME :SATURA
00005000: 8 ; DETAIL LEVEL: 0
00005001: 8 ; V DETAIL LEVEL: 0
00005002: 8 ; DETAIL CORING : 0
00005003: 12 ; CHROMA LEVEL: 4
00005004: 8 ; CHROMA PHASE: 0
00005005: 8 ; COLOR TEMP: 0
00005006: 9 ; MASTER PED: -10
00005007: 5 ; A.IRIS LEVEL: -3
00005008: 1 ; NEWS GAMMA: OFF
00005009: 5 ; GAMMA : CINELIKE D
0000500A: 0 ; KNEE: AUTO
0000500B: 3 ; MATRIX: CINE-LIKE
0000500C: 1 ; SKIN TONE DTL : OFF
0000500D: 0 ; V DETAIL FREQ : THIN
0000500E: 1 ; OPERATION TYPE: FILM CAM
0000500F: 0 ; FRAME RATE: DEFAULT
00005010: BE026801 ; SYNCRO SCAN :
I'm glad you said that about the 35-70 zoom. I have used it on photoshoots and found it to be inferior to my 28-35. I didn't know why. So now I know, it's just a bad lens.
CinemaElectronika
Well-known member
jdorsey said:
Sorry, I get confussed regarding compression codecs but H.264 Decoder 640x360, AAC Stereo is a very good one... give a try.
Good Luck.
H
Lenilenapi
Carbonite Member
jdorsey,
One thing to consider in creating settings for the adapter is to raise your detail settings. I have often set them from +5 to +7 to compensate for the softness of the adapter and haven't seen any negative artifacts though its probably a bit noisier. these were talking heads though and different images might require different settings. Detail is ultimately a taste issue but i believe Barry said he found +2 or +3 to be a good general setting for this camera.
Also my tests and also Adam Wilt's find that cinegamma is softer so I bump up the detail a point or 2 with it. Not everyone has found that though so check for yourself.
I assume you realize that you are crushing your blacks out there with a -10 Ped level, but again this comes down to taste and it seems like you are going for a saturated intense look.
Better luck next time with the adapter.
- Lenny Levy
One thing to consider in creating settings for the adapter is to raise your detail settings. I have often set them from +5 to +7 to compensate for the softness of the adapter and haven't seen any negative artifacts though its probably a bit noisier. these were talking heads though and different images might require different settings. Detail is ultimately a taste issue but i believe Barry said he found +2 or +3 to be a good general setting for this camera.
Also my tests and also Adam Wilt's find that cinegamma is softer so I bump up the detail a point or 2 with it. Not everyone has found that though so check for yourself.
I assume you realize that you are crushing your blacks out there with a -10 Ped level, but again this comes down to taste and it seems like you are going for a saturated intense look.
Better luck next time with the adapter.
- Lenny Levy
Barry_Green
Moderator
Doesn't happen. The glass doesn't "expose" anything, only the CCD (or CMOS) does. You can't "overexpose" glass.THoff said:
Don't agree, I'd have to see some evidence of this. I've used the M2 (briefly), Mini35, Letus35FE, never had an issue with using in-camera ND filters...
Barry_Green
Moderator
This makes no sense. The image isn't "exposed" at all until after the ND filters. A ground glass doesn't expose an image, the very concept is alien. Only a CCD, CMOS, or actual film can expose an image. Ground glass is ground glass, not an image capture/exposure system.THoff said:
Should make absolutely no difference whatsoever. Other than the in-camera filters being more convenient and dustproof, that is. I really can't understand this whole ground-glass-exposure line of thinking, it's just foreign to me. I really don't believe that's how it works...
I ran a test to verify this claim. I exposed the Letus35FE system properly at a low light level, and took a shot. Then I juiced up the light so it was something like 256x as strong, and put on the camera's ND filters and set the iris for proper exposure. The shots look *identical*. There is no truth to the concept that the ground glass is being overexposed.
Unless I'm not understanding what someone's talking about...
Last edited:
HVXguy said:
Before saying the 35-70mm is the most unsharp lens, you forgot to say which one you talking about. There is about 8 different 35-70 mm lens on the Nikon serie. Look here for info : http://www.photosynthesis.co.nz/nikon/lenses.html#35-xx
and here for review : http://www.fredmiranda.com/reviews/showproduct.php?product=112&sort=7&cat=28&page=1
http://www.nikonians.org/nikon/nikkor-70mm/70mm_shootout2.html
The 35-70mm AF 1:2.8 D is one of the sharpess at any aperture. ; )
Last edited:
OK, let's reverse the scenario. Instead of overexposing the GG and trying to recover from 100 IRE by adding ND filters, let's badly underexpose and go with 0 IRE.
Suppose you have a futuristic sensor that lets you amplify without any noise whatsoever, so you dial in a bunch of gain. If the entire image on the GG is at 0 IRE, how much gain do you think is necessary to get a well-defined image?
Once two image elements of different brightness hit either 0 IRE or 100 IRE due to incorrect exposure, you can't really go back and separate them, because at this stage the difference in brightness between the two elements disappears. You can either brighten or darken both, but not independently.
Suppose you have a futuristic sensor that lets you amplify without any noise whatsoever, so you dial in a bunch of gain. If the entire image on the GG is at 0 IRE, how much gain do you think is necessary to get a well-defined image?
Once two image elements of different brightness hit either 0 IRE or 100 IRE due to incorrect exposure, you can't really go back and separate them, because at this stage the difference in brightness between the two elements disappears. You can either brighten or darken both, but not independently.
Barry_Green
Moderator
The image doesn't have IRE on the ground glass. The image doesn't get exposed by the ground glass. It's just another piece of glass. Like a window, or a lens element. It doesn't expose.
Let me say it again: I exposed a scene at x footcandles (say, 100 or so, I don't remember). Then I ramped up the light to 25,600 footcandles and put in two stops of iris and six stops of ND filter. The resulting image was identical to the 100-IRE image. The ground glass loses nothing, there is no "exposing".
Let me say it again: I exposed a scene at x footcandles (say, 100 or so, I don't remember). Then I ramped up the light to 25,600 footcandles and put in two stops of iris and six stops of ND filter. The resulting image was identical to the 100-IRE image. The ground glass loses nothing, there is no "exposing".