FS7: ND Filters when wide open in bright light
- Thread starter Soldier
- Start date
cameramanben
Well-known member
Thanks Liam,
yes, I should have been more clear, some things are not necessarily a good idea, but still possible. However, at the start of the thread it was mentioned that gain is locked in CinEI.
If you apply a LUT - as an MLUT, no need to bake it in - exposure will shift with CineEI, giving the same effect as gain. A lot of the time I don't need six stops of headroom, so CineEI down a stop or two and I still have 4 - 5 stops headroom but the same visual effect as a ** STUPIDITY ALERT ** 0.9 or 1.2 ND ** END OF STUPIDITY (FOR NOW) - that should be 0.3 or 0.6 ND **. I also like that I can set 1/3 stop steps - OK you only have three options on the FS7, but that lets you fill the gaps between the physical NDs.
Waveform and zebras will also shift, so remain useful. This may well all be obvious - particularly to anyone who has shot raw - but I am aware of people who work off the log signal, sometimes through not knowing about the worth of MLUTs. Over on the Sony forum, Simon Bailey mused about false colour MLUTs, I wrote up some code to generate them, so I bunged a few in the links above.
It is also rather a lot easier to bounce between settings on the F5/F55 - got to have some advantages at the price!
Anyway, sorry to have galumphed into setting your blood boiling,
Ben
yes, I should have been more clear, some things are not necessarily a good idea, but still possible. However, at the start of the thread it was mentioned that gain is locked in CinEI.
If you apply a LUT - as an MLUT, no need to bake it in - exposure will shift with CineEI, giving the same effect as gain. A lot of the time I don't need six stops of headroom, so CineEI down a stop or two and I still have 4 - 5 stops headroom but the same visual effect as a ** STUPIDITY ALERT ** 0.9 or 1.2 ND ** END OF STUPIDITY (FOR NOW) - that should be 0.3 or 0.6 ND **. I also like that I can set 1/3 stop steps - OK you only have three options on the FS7, but that lets you fill the gaps between the physical NDs.
Waveform and zebras will also shift, so remain useful. This may well all be obvious - particularly to anyone who has shot raw - but I am aware of people who work off the log signal, sometimes through not knowing about the worth of MLUTs. Over on the Sony forum, Simon Bailey mused about false colour MLUTs, I wrote up some code to generate them, so I bunged a few in the links above.
It is also rather a lot easier to bounce between settings on the F5/F55 - got to have some advantages at the price!
Anyway, sorry to have galumphed into setting your blood boiling,
Ben
Last edited:
Liam Hall
Veteran
Ben, my blood's not boiling. Not even close, I just didn't want to have to write a long post. But, in the interests of clarity:
No it won't. Exposure will stay exactly the same. Your VF monitor will change, but the underlying recorded signal will be exactly the same. In CineEI, exposure will only change if you adjust aperture, shutter or ND.
Gain affects the whole recorded signal, it's too simplistic to say it is the same as pushing or pulling exposure using Cine EI or any other method.
0.9 is 3 stops, 1.2 is 4. The FS7 has 1/4, 1/16, and 1/64 which equates to 2, 4 and 6 stops. Cine EI only drops to EI500 on the FS7 so there is absolutely no way you can shoot wide open outside in bright sunshine with a fast prime without the use of additional ND.
Waveform doesn't work with LUTs on the FS7. Waveform doesn't work in S&Q mode. Zebras won't shift either, they will activate at the exact same IRE they are set for.
Yep. FS7 is slightly quicker with the new firmware but it's still very slow for a modern camera and doesn't have anything like the functionality you are used to on the F55. It's still a great camera, produces a wonderful image and is considerably cheaper than a f5 or f55
No it won't. Exposure will stay exactly the same. Your VF monitor will change, but the underlying recorded signal will be exactly the same. In CineEI, exposure will only change if you adjust aperture, shutter or ND.
Gain affects the whole recorded signal, it's too simplistic to say it is the same as pushing or pulling exposure using Cine EI or any other method.
0.9 is 3 stops, 1.2 is 4. The FS7 has 1/4, 1/16, and 1/64 which equates to 2, 4 and 6 stops. Cine EI only drops to EI500 on the FS7 so there is absolutely no way you can shoot wide open outside in bright sunshine with a fast prime without the use of additional ND.
Waveform doesn't work with LUTs on the FS7. Waveform doesn't work in S&Q mode. Zebras won't shift either, they will activate at the exact same IRE they are set for.
Yep. FS7 is slightly quicker with the new firmware but it's still very slow for a modern camera and doesn't have anything like the functionality you are used to on the F55. It's still a great camera, produces a wonderful image and is considerably cheaper than a f5 or f55
Last edited:
cameramanben
Well-known member
Liam,
thanks for that. Re the 0.3, 0.6, 0.9 ND thing. Oops! Sorry about that; what a stupid schoolboy error and I am deeply embarrased! I was posting during gaps in shooting and my brain managed to mangle dB and filter points. What an idiot! I'll be kicking myself over that one for a while!.
As for the exposure shifting with LUTs - the whole point about what I said is that is that things change in the viewfinder but not on the recording. You tell the camera to change EI, you see the change in the viewfinder (if you have LUTs on) the camera records exactly what it would record at ISO800, IS2000 or ISO3200 but includes a bit of metadata, then post software (such as Catalyst) applies the correct exposure adjustment. The net result is (supposed to be) visually exactly the same as baking in 'gain' in camera, but with the advantage of keeping the dynamic range. Plenty of software currently doesn't read and apply ISO data, so I've spent time learning how to produce mathematically precise exposure adjustment LUTs.
I'm generally careful about using the term gain, as it is basically supposed to be a multiplication, whereas log exposure adjustment is based on offsets, albeit offsets that vary with IRE level. It comes from the relationship log(ab) == log(a) + log(b). In the case of pulling to a low ISO with a restricted bit-depth codec (10-bit XAVC), in data terms there is literally no reduction in image quality per stop in applying an exposure adjustment in post over doing it in camera. In both cases you will reduce the headroom to white clip though. OK, the way I put it before may be a simplification, but is this blah blah last paragraph really more helpful? ;-)
Waveforms are there in S&Q on the F5/F55, but didn't know about no waveform in CineEI on the FS7. I've done two or three shoots with the FS7, two baking-in LUTs so in CineEI mode. On the F5/F55, waveform is only available when both the viewfinder and SDI2 are set to MLUT on. I've had a look in the FS7 manual and it mentions:
If you have to have the SDI1&Internal Rec on that is quite a restriction, but explains why I could see it whilst baking in. Thanks for the info on that and what a drag! The manual also mentions no waveform when SDI out is set to 4k UHD or where 4k/2k SDI output conversion is edge cropped. I've only shot HD and 2K with the FS7 so far. What happens with everything except SDI1&Internal Rec set to MLUT 'On'?
Then zebras. On the F5/F55 zebras will map to the IRE of whatever the viewfinder is showing. Have the viewfinder set to show an MLUT and, say, 65% zebras will appear at the viewfinder's 65% IRE level. This means that when the CineEI level changes, lets say down a stop, the viewfinder will immediately get darker and the zebras will have changed position. Choose an appropriate LUT with known IREs for known reflectances and zebras are a really useful tool. If it doesn't work like that on the FS7 I'm a bit surprised as that suggests a major restriction to CineEI. Really grateful to know that. The false colour approach might be worth a go, though personally one set of zebras with an aperture of 3% is about as much as I can stand on the picture - and I can quickly turn it on or off!
Again, feel so, so, so, so, so, foolish about the ND values thing - I seem to have turned into a blithering idiot over the last couple of days. I managed to leave my mattebox behind on a shoot yesterday (ironic eh? ;-) ). Tree pollen, phase of the moon, er ,er, ok I'm just being thick! Mind you, I really didn't mean to suggest that CineEI shifts replace NDs but can work with them. If you want to keep wide open being able to tweak up and down a hair's - 1/3 or 2/3 stop - comes in really handy. Alternatively, running around and not wanting to start pulling filters in and out then a stop and a half of pull at the flick of a switch (well turn of a knob on my camera) is really quick.
Thanks for the info (and the burning sense of shame) though and I'm off to look for my car keys and phone ;-)
Ben
thanks for that. Re the 0.3, 0.6, 0.9 ND thing. Oops! Sorry about that; what a stupid schoolboy error and I am deeply embarrased! I was posting during gaps in shooting and my brain managed to mangle dB and filter points. What an idiot! I'll be kicking myself over that one for a while!.
As for the exposure shifting with LUTs - the whole point about what I said is that is that things change in the viewfinder but not on the recording. You tell the camera to change EI, you see the change in the viewfinder (if you have LUTs on) the camera records exactly what it would record at ISO800, IS2000 or ISO3200 but includes a bit of metadata, then post software (such as Catalyst) applies the correct exposure adjustment. The net result is (supposed to be) visually exactly the same as baking in 'gain' in camera, but with the advantage of keeping the dynamic range. Plenty of software currently doesn't read and apply ISO data, so I've spent time learning how to produce mathematically precise exposure adjustment LUTs.
I'm generally careful about using the term gain, as it is basically supposed to be a multiplication, whereas log exposure adjustment is based on offsets, albeit offsets that vary with IRE level. It comes from the relationship log(ab) == log(a) + log(b). In the case of pulling to a low ISO with a restricted bit-depth codec (10-bit XAVC), in data terms there is literally no reduction in image quality per stop in applying an exposure adjustment in post over doing it in camera. In both cases you will reduce the headroom to white clip though. OK, the way I put it before may be a simplification, but is this blah blah last paragraph really more helpful? ;-)
Waveforms are there in S&Q on the F5/F55, but didn't know about no waveform in CineEI on the FS7. I've done two or three shoots with the FS7, two baking-in LUTs so in CineEI mode. On the F5/F55, waveform is only available when both the viewfinder and SDI2 are set to MLUT on. I've had a look in the FS7 manual and it mentions:
If you have to have the SDI1&Internal Rec on that is quite a restriction, but explains why I could see it whilst baking in. Thanks for the info on that and what a drag! The manual also mentions no waveform when SDI out is set to 4k UHD or where 4k/2k SDI output conversion is edge cropped. I've only shot HD and 2K with the FS7 so far. What happens with everything except SDI1&Internal Rec set to MLUT 'On'?
Then zebras. On the F5/F55 zebras will map to the IRE of whatever the viewfinder is showing. Have the viewfinder set to show an MLUT and, say, 65% zebras will appear at the viewfinder's 65% IRE level. This means that when the CineEI level changes, lets say down a stop, the viewfinder will immediately get darker and the zebras will have changed position. Choose an appropriate LUT with known IREs for known reflectances and zebras are a really useful tool. If it doesn't work like that on the FS7 I'm a bit surprised as that suggests a major restriction to CineEI. Really grateful to know that. The false colour approach might be worth a go, though personally one set of zebras with an aperture of 3% is about as much as I can stand on the picture - and I can quickly turn it on or off!
Again, feel so, so, so, so, so, foolish about the ND values thing - I seem to have turned into a blithering idiot over the last couple of days. I managed to leave my mattebox behind on a shoot yesterday (ironic eh? ;-) ). Tree pollen, phase of the moon, er ,er, ok I'm just being thick!
Mind you, I really didn't mean to suggest that CineEI shifts replace NDs but can work with them. If you want to keep wide open being able to tweak up and down a hair's - 1/3 or 2/3 stop - comes in really handy. Alternatively, running around and not wanting to start pulling filters in and out then a stop and a half of pull at the flick of a switch (well turn of a knob on my camera) is really quick.Thanks for the info (and the burning sense of shame) though and I'm off to look for my car keys and phone ;-)
Ben
Last edited:
Liam Hall
Veteran
Yes, this is how it works on the FS7; Zebra IRE value remains the same - which was my point for clarity. The way you wrote it before I felt someone could have misunderstood and thought the IRE value changed. Nit picky I know, but I was on a roll
cameramanben
Well-known member
Ben not good yooze words proper! Ben promise try very bestest in future
Last edited:
SpareWheel
Active member
Ben,
Waveforms are there in S&Q on the F5/F55, but didn't know about no waveform in CineEI on the FS7...
I have followed your posts with interest and have found many of your comments valuable
But following on from what Liam has said, without owning the FS7 yourself or using it extensively you are making a number of statements which are simply not valid. It might be useful if you obtained the FS7 and comment with a little more knowledge. Under these conditions it just makes you look uninformed at best and down-right silly at worst.
Claire
Waveforms are there in S&Q on the F5/F55, but didn't know about no waveform in CineEI on the FS7...
I have followed your posts with interest and have found many of your comments valuable
But following on from what Liam has said, without owning the FS7 yourself or using it extensively you are making a number of statements which are simply not valid. It might be useful if you obtained the FS7 and comment with a little more knowledge. Under these conditions it just makes you look uninformed at best and down-right silly at worst.
Claire
cameramanben
Well-known member
Cheers for the kind words Claire!
I'll grant you mixing up my dBs and filter points is a howler of the first order - I want to lock myself in a dark room until Christmas for that one.
Unless I have completely misread him, most of the problems Liam had with what I wrote were to do with my inarticulacy; my rubbish writing just made it sound as though I was being factually incorrect (see the bit about zebras above).
As for the waveform in CineEI thing, I have shot with the FS7, I have used the waveform in CineEI, I happened to be baking in.
For the first few months after the F5 and F55 got waveforms, pretty much without fail at least a couple of times a week someone would start a thread on the Sony forums asking 'why no waveform in CineEI'. Of course you could get the waveform, but what many people didn't realise - because it is not obvious and is deep in the manual - is that in the 'Video->Monitor LUT' menu the values for SDI2 and Viewfinder have to match - either both on or both off. I instinctively set both to on, including when I have shot with the FS7, and because I also had SDI1&Internal Rec set to on to bake in, the waveform was on and I had no reason to believe that the camera behaved any differently from the other F's (after all, no waveform when shooting CineEI the way it is supposed to be, ie with SDI1&Internal Rec set to off, is pretty silly).
Looking in the manual, that unfortunate situation does sound like the case.
As a well chastened ignoramus, could someone with an FS7 to hand confirm for me that when in the Video->Monitor LUTs menu with Viewfinder, SDI1 and HDMI all equally set to 'on' or 'off' and in the Video menu with SDI set to 'on' and Output Format set to 'SDI' that waveform / vector / histogram is not available, either in HD or with 4k/2k to HD conv set to 'Letterbox'. I'd be most grateful.
The zebra, false colour, and basis of log recording and exposure shifting stuff still holds true.
You can't believe how much I regret ever having posted to this thread - I didn't mean to tell anyone what to do, only make some suggestions that I have found useful (yes, including on an FS7) and that I thought in no way controversial. I'll try to keep schtum unless I have something incontrovertably true or reasonably obvious to say, but to be honest I rather enjoy joining in chats with people I see as contemporaries with a similar interest. I'd also say that I haven't been looking on this forum for long enough to know how things are around here and I'm sure everyone is fundamentally well informed, but as a general experience ownership of a camera, or even extensive use, is no barrier to ignorance or misunderstanding. I count myself in there as two years in I'm still learning and adjusting what I think I know about the F5 and F55.
And yes, Ben is the ideal name for a camera person ;-)
Now off to turn the lights out, turn on the Leonard Cohen and enjoy my shortcomings,
ben (don't deserve a capital 'B')
I'll grant you mixing up my dBs and filter points is a howler of the first order - I want to lock myself in a dark room until Christmas for that one.
Unless I have completely misread him, most of the problems Liam had with what I wrote were to do with my inarticulacy; my rubbish writing just made it sound as though I was being factually incorrect (see the bit about zebras above).
As for the waveform in CineEI thing, I have shot with the FS7, I have used the waveform in CineEI, I happened to be baking in.
For the first few months after the F5 and F55 got waveforms, pretty much without fail at least a couple of times a week someone would start a thread on the Sony forums asking 'why no waveform in CineEI'. Of course you could get the waveform, but what many people didn't realise - because it is not obvious and is deep in the manual - is that in the 'Video->Monitor LUT' menu the values for SDI2 and Viewfinder have to match - either both on or both off. I instinctively set both to on, including when I have shot with the FS7, and because I also had SDI1&Internal Rec set to on to bake in, the waveform was on and I had no reason to believe that the camera behaved any differently from the other F's (after all, no waveform when shooting CineEI the way it is supposed to be, ie with SDI1&Internal Rec set to off, is pretty silly).
Looking in the manual, that unfortunate situation does sound like the case.
As a well chastened ignoramus, could someone with an FS7 to hand confirm for me that when in the Video->Monitor LUTs menu with Viewfinder, SDI1 and HDMI all equally set to 'on' or 'off' and in the Video menu with SDI set to 'on' and Output Format set to 'SDI' that waveform / vector / histogram is not available, either in HD or with 4k/2k to HD conv set to 'Letterbox'. I'd be most grateful.
The zebra, false colour, and basis of log recording and exposure shifting stuff still holds true.
You can't believe how much I regret ever having posted to this thread - I didn't mean to tell anyone what to do, only make some suggestions that I have found useful (yes, including on an FS7) and that I thought in no way controversial. I'll try to keep schtum unless I have something incontrovertably true or reasonably obvious to say, but to be honest I rather enjoy joining in chats with people I see as contemporaries with a similar interest. I'd also say that I haven't been looking on this forum for long enough to know how things are around here and I'm sure everyone is fundamentally well informed, but as a general experience ownership of a camera, or even extensive use, is no barrier to ignorance or misunderstanding. I count myself in there as two years in I'm still learning and adjusting what I think I know about the F5 and F55.
And yes, Ben is the ideal name for a camera person ;-)
Now off to turn the lights out, turn on the Leonard Cohen and enjoy my shortcomings,
ben (don't deserve a capital 'B')
Last edited:
Liam Hall
Veteran
Ben, don't beat yourself up, we're all good
You are right the Waveform works if you are burning in the LUT. It's a mystery to me why it disappears the rest of the time...
Back on topic:
One thing that hasn't been mentioned much here or anywhere else is how well this camera handles infrared - just as well considering it's ISO2000 and we all want to shoot at F/1.2 most of the time...
You are right the Waveform works if you are burning in the LUT. It's a mystery to me why it disappears the rest of the time...
Back on topic:
One thing that hasn't been mentioned much here or anywhere else is how well this camera handles infrared - just as well considering it's ISO2000 and we all want to shoot at F/1.2 most of the time...
Lenilenapi
Carbonite Member
Ben ,
Looking at the LUT's you made for download:
Can you explain the difference between your LC709AExt.cube in the (S-Log3-MLUTs folder) and the LUTs in the "S-Log3-LC709AExtended-DtoL-Resolve" folder (that's Data to Legal right?)
Also in the latter folder - what is the difference between the
LC709AExt-up-1p33.cube and the
LC709AExt-up-1.cube
thanks, I'm working hard to get a handle on this and get used to Resolve.
Lenny
Looking at the LUT's you made for download:
Can you explain the difference between your LC709AExt.cube in the (S-Log3-MLUTs folder) and the LUTs in the "S-Log3-LC709AExtended-DtoL-Resolve" folder (that's Data to Legal right?)
Also in the latter folder - what is the difference between the
LC709AExt-up-1p33.cube and the
LC709AExt-up-1.cube
thanks, I'm working hard to get a handle on this and get used to Resolve.
Lenny
Grug
Veteran
I shoot with the same sensor in my F5, though I only rate the camera at 1000 ISO. But in bright Australian sun, I'm using the internal 1.8 ND (6 stops) with an additional Tiffen 1.2 ND (4 stops), for a total of 10(!) stops of ND. This is enough to get me back down to a T/2 or T/2.8 in even the brightest conditions.
The internal NDs on these cameras are great, but I always carry a 1-4 stop ND kit and some ND grads with me to work in combination with the internal filters. I highly recommend Tiffen's 'Water White' NDs for this purpose. I have zero issues with IR contamination.
cameramanben
Well-known member
Hi Lenny,
The only difference between the S-Log3-MLUTs folder and the S-Log3-LC709AExtended-DtoL-Resolve folder is that the MLUTs one is explicitly clipped between values 0-1.09. LUTs are generally floating point values, with the input range being from 0-1. The output range for cube files can be anything, but is usually clipped 0-1.0. what 0, 1 and 1.09 actually reference is open, but a common setup (including the way Sony MLUTs work) is for the inputs to be the data range (ie 0=10-bit 0=-7%IRE, 1=10-bit 1023=109%IRE), with the output scale of 0-1 being legal range (0=10-bit 64=0%IRE, 1=10-bit 940=100%IRE). As it happens the Sony cameras are happy to take values up to 1.09 which is fortunate as it allows for MLUTs which go into extended range, as these do.
Clipping 0-1.09 just makes sure that nothing untoward will happen in camera, but not clipping in post ensures minimal loss of information which may come in handy. This is particularly relevent as the post ones include exposure adjustment, where even with a kneed curve such as LC709(A) increasing the ISO results in a highlight range that can go beyond extended range. By not clipping it is still intact for you to pull back down in post as needed. That is the big advantage of CineEI adjustment in post, handled correctly all the dynamic range should still be available until your final render.
To be honest, Sony looks are clipped 0-1 as are LUTs generated by Resolve, but I just feel if you can avoid clipping...
The basic workflow in Resolve (as recommended by Sony), is to have clips treated as data range, then work on them monitoring and treating the waveform scopes as legal range. This is the 'DtoL' thing - it's the same as the original Sony look profiles. As pointed out to me by a DIT, the Lumetri plugin in Adobe CC seems to work correctly when the LUT is data in to data out, hence the other set.
LUTCalc is quite enthusiastic in cleaning filenames, which includes removing full stops / periods. Rather than change things back after saving, I put a 'p' in instead. up 1p33 is up 1 1/3 Stops, up 1 is up 1 stop. It's to go with the camera's 1/3 stop step options in CineEI.
Hope that all makes sense!
Ben
The only difference between the S-Log3-MLUTs folder and the S-Log3-LC709AExtended-DtoL-Resolve folder is that the MLUTs one is explicitly clipped between values 0-1.09. LUTs are generally floating point values, with the input range being from 0-1. The output range for cube files can be anything, but is usually clipped 0-1.0. what 0, 1 and 1.09 actually reference is open, but a common setup (including the way Sony MLUTs work) is for the inputs to be the data range (ie 0=10-bit 0=-7%IRE, 1=10-bit 1023=109%IRE), with the output scale of 0-1 being legal range (0=10-bit 64=0%IRE, 1=10-bit 940=100%IRE). As it happens the Sony cameras are happy to take values up to 1.09 which is fortunate as it allows for MLUTs which go into extended range, as these do.
Clipping 0-1.09 just makes sure that nothing untoward will happen in camera, but not clipping in post ensures minimal loss of information which may come in handy. This is particularly relevent as the post ones include exposure adjustment, where even with a kneed curve such as LC709(A) increasing the ISO results in a highlight range that can go beyond extended range. By not clipping it is still intact for you to pull back down in post as needed. That is the big advantage of CineEI adjustment in post, handled correctly all the dynamic range should still be available until your final render.
To be honest, Sony looks are clipped 0-1 as are LUTs generated by Resolve, but I just feel if you can avoid clipping...
The basic workflow in Resolve (as recommended by Sony), is to have clips treated as data range, then work on them monitoring and treating the waveform scopes as legal range. This is the 'DtoL' thing - it's the same as the original Sony look profiles. As pointed out to me by a DIT, the Lumetri plugin in Adobe CC seems to work correctly when the LUT is data in to data out, hence the other set.
LUTCalc is quite enthusiastic in cleaning filenames, which includes removing full stops / periods. Rather than change things back after saving, I put a 'p' in instead
. up 1p33 is up 1 1/3 Stops, up 1 is up 1 stop. It's to go with the camera's 1/3 stop step options in CineEI.Hope that all makes sense!
Ben
Last edited:
Lenilenapi
Carbonite Member
Aaaah .... Thanks for the clarification. I assumed that 33 and 66 had to do with the type of LUT . That is a big difference . Ha p=
point - cute!
1/3 stops is very nice .
Slowly the fog is lifting for me on all this.
....no doubt more questions to come. .
point
- cute!1/3 stops is very nice .
Slowly the fog is lifting for me on all this.
....no doubt more questions to come. .
I'm based out of Los Angeles and shoot on the Fs7. On bright sunny days, I've had to stop down to f11 with the built-in 1/64 ND filter in place. I am not shooting in this type of environment enough to justify investing further in IRND's. Plus, I don't mind stopping down a little in bright sunlight because the image more closely matches what my eye sees. However, with that in mind, a fixed IRND of 1-stop (.3) will bring you back down to f8. 2-stops (.6) will take you down to f5.6. I don't think you'd want to go lower than that unless you are going for a stylized kind of look.
Doug Jensen
Carbonite Member
I don't. It can reduce the dynamic range of the camera and if I am outside in bright light I probably need all the DR I can get.
If the internal NDs aren't enough, you've got to add a glass filter in front of the lens. There's really no other acceptable option.
If the internal NDs aren't enough, you've got to add a glass filter in front of the lens. There's really no other acceptable option.
Harry Lime
Well-known member
Are any shooters seeing color shifts due to IR contamination, when using very strong ND filters?
https://www.youtube.com/watch?v=_T9u-1Ib6vo
thx
https://www.youtube.com/watch?v=_T9u-1Ib6vo
thx