Digital ND filter question xa10

Mikko Topponen · November 10th, 2011, 02:02 PM

Quote:

Originally Posted by Rainer Halbich

What I don't understand is how a DSLR can adjust its ISO but a Digital video camera can't.

gain = ISO. Same basic thing.

Don Palomaki · November 11th, 2011, 08:27 AM

I would say that gain is more like processing time,0 dB gain representing the nominal rcommended developer and processing time; increased gain being a bit like push processing.

ISO would correspond to the sensitivity of the sensor picture elements; e.g., aperture, shutter speed, and incident light on the subject,required to produce an optimal imge at 0 dB gain. Of course traditionally video does not speak to ISO, more often the ratig has been the aperture required to produce something like 50 IRE from 1000 LUX on a 50% gray card at 1/60 shutter (or somethign like that, the figures I gave are not actuals to my knowledge).

Don Palomaki · November 22nd, 2011, 07:03 AM

A bit more on the ND thing.

If we set the camcorder to Manual and manual exposure mode we get the Aperture, shutter, and gain displayed and can adjust them. They can be adjusted independently. The ND setting is coupled with the aperture setting, not separate.

As we adjust the aperture we find that after f/4 we get steps in ND (and marks for 1/2, 1/4, etc.) until we run out of ND filter range, then the aperture stops continue to f/5.6 etc. We can see changes in image brightness as we move between ND setting marks inplying that the filter effect is continuous, e.g. a ND wedge that moves through the light path.

It all make sense now. It is one less thing the user has to manage, fitting for a camcorder at this price point, and I am content to NOT have to deal with turning the ND ON/OFF it for the shooting I do..

Rainer Halbich · December 16th, 2011, 10:05 AM

Quote:

Originally Posted by Tom Hardwick

Bear in mind that the v'finder aperture readout is an extrapolation of ND + iris blade position. You may well be told you're shooting at f/5.6 but that's because the camera's actually shooting at f/2.8 with 2 stops of light being soaked by internal ND.

It's pretty easy to see the ND filter at work if you look down into the lens, and my Panasonic MX300, SD900, Sony PDX10 and so on have all had undocumented ND and unreliable aperture readouts.

I can only see the iris working, I don't see a ND filter in the lens.

The xa10 gives the F-stop / ND value. it is not combined into the f-stop reading as you say it is.

I agree, a rotating ND filter directly behind the iris is a possibility, but if this is true is must be a EXTREMELY close to the iris and I doubt that this is the case. This will also need to be a very large graduated filter because the effect will be different if the iris is wide open vs a small iris. A larger filter can make it more gradual. It is a possibility but I feel 20% positive about the theory.

Rainer Halbich · December 16th, 2011, 10:10 AM

Quote:

Originally Posted by Mikko Topponen

gain = ISO. Same basic thing.

Are you sure? It makes sense to me bacause more gain = more niose and more litht, just like hight iso + more niose and more light.

But why not call it the same thing?

Jeff Harper · December 16th, 2011, 10:15 AM

Yes, ISO and gain mean the same thing for our purposes. Why different names? Who knows. Wikipedia will have the answer, for sure.

Rainer Halbich · December 16th, 2011, 10:15 AM

Quote:

Originally Posted by Don Palomaki

ISO would correspond to the sensitivity of the sensor picture elements; e.g., aperture, shutter speed, and incident light on the subject,required to produce an optimal imge at 0 dB gain. Of course traditionally video does not speak to ISO, more often the ratig has been the aperture required to produce something like 50 IRE from 1000 LUX on a 50% gray card at 1/60 shutter (or somethign like that, the figures I gave are not actuals to my knowledge).

I am not sure about your ISO theory, According to me the ISO is ONLY the light sensitivity of the film/ digital image sensor. It has nothing to do with SS, Aperture, ND value, WB or frame speed. Or maybe I don't understand what you are saying?

Rainer Halbich · December 16th, 2011, 10:25 AM

Quote:

Originally Posted by Don Palomaki

A bit more on the ND thing.

If we set the camcorder to Manual and manual exposure mode we get the Aperture, shutter, and gain displayed and can adjust them. They can be adjusted independently. The ND setting is coupled with the aperture setting, not separate.

As we adjust the aperture we find that after f/4 we get steps in ND (and marks for 1/2, 1/4, etc.) until we run out of ND filter range, then the aperture stops continue to f/5.6 etc. We can see changes in image brightness as we move between ND setting marks inplying that the filter effect is continuous, e.g. a ND wedge that moves through the light path.

It all make sense now. It is one less thing the user has to manage, fitting for a camcorder at this price point, and I am content to NOT have to deal with turning the ND ON/OFF it for the shooting I do..

That is all good and I also think it is awesome, but it does not answer the question.

If this is a digital ND filter which lowers the iso/gain, many people will fry their sensors because they filmed the sun rise or sunset thinking that it is a optical filter...poor camera sensor...

Tom Hardwick · December 16th, 2011, 10:26 AM

Quote:

Originally Posted by Rainer Halbich

I can only see the iris working, I don't see a ND filter in the lens.

The xa10 gives the F-stop / ND value. it is not combined into the f-stop reading as you say it is.

All the camcorders I've looked at that have no external ND switching use internal ND Rainer. I'd be very surprised if the XA10 bucked the trend. But I haven't looked at this camera so I didn't list it as one of these.

Any internal ND placed at the iris blade position can be tiny. Even when wide open it still only needs a tiny filter. And it doesn't have to be graduated - the ND is just inserted into the light path a little bit at a time. It's completely out of focus, as are the iris blades of course.

tom.

Tom Hardwick · December 16th, 2011, 10:50 AM

Rainer - you started this thread asking about a 'digital ND filter' and I thought we'd all talked you out of it. Obviously we haven't.

Rainer Halbich · December 16th, 2011, 01:01 PM

Quote:

Originally Posted by Tom Hardwick

Rainer - you started this thread asking about a 'digital ND filter' and I thought we'd all talked you out of it. Obviously we haven't.

One day when AVCHD becomes outdated I am going to strip my camera and I am going to find out.
I will report back in 5 years.

For now I will assume that it has something to do with the sensor sensitivity, because I don't want to burn my sensor when I record the sunset.

Jeff Harper · December 16th, 2011, 03:10 PM

Ranier, I believe that in science they say that the simplest explanation is the most likely the correct one, or something to that effect.

In this case, the most likely and commonsense approach would be that Don is correct. First, no one here, so far, has ever heard of a ND filter that is purely electronic, but I understand that just because we've not hear about it does not mean it is not so. But the very idea sounds, well, to not seem logical, at least to me, but I'm no expert, so what do I know.

Secondly, if this were a new type of ND filter that operates the way you fear that it does, Canon would likely mention it, if not brag about it. It would likely be well known.

Thirdly the characteristics of the ND filter on the XA10 seen to be that of a genuine ND filter. When it kicks in the images look like a real ND filter is being used. If it acts like a rabbit, and smells like a rabbit, it is probably a rabbit.

Fourth, if the sensor was particularly more sensitive and prone to danger from sunsets, etc, then Canon would probably mention it, because they would not want to deal with the many returns of damaged cameras.

Long story short, the safest, most logical bet is that the camera uses a traditional ND filter. Just because you can't see it doesn't mean it is not there.

Just turn it off, buy a variable ND filter or a regular one, and you won't have to worry at all.

Don Palomaki · December 16th, 2011, 08:00 PM

Quote:

Yes, ISO and gain mean the same thing for our purposes.

Not quite, except perhaps in the narrow taxonomy of the DSLR that has morphed the definition of ISO. In the DSLR they use ISO to refer to how the image exposure was calculated (see the past paragraph below) and factor in DSP results as well.

ISO ~ sensitivity of the CCD/CMOS in absolute terms (e.g., microvolts per photon)
Processing time/temperature ~ Gain, 0 dB gain being the recommended developer time/temp point.for the rated ISO, or amplification of microvolts in the pixel sensor.

The primary reason for ISO is to allow metering of light and then setting exposure for the optimal result on film. You need the meter and ISO because with film there is no instant feedback as you get with a camcorder in the viewfinder and zebra (if you use it). You have to soup the film to see the result.

Photons reaching the film/CCD/CMOS pixel is a combination of the aperture, exposure time, light on the subject, subject reflectivity, and any losses due to filters. It is a bit easier than speaking in photons per sqmm for, say, 20 IRE.

Because you can translate aperture and shutter speed for, say 20 IRE, you can compute an equivalent ISO.

Not to be confused with people exposing film at a different ISO and then push or pull processing to obtain the negative density they want, or art effect they are trying to achieve.

Jeff Harper · December 16th, 2011, 09:29 PM

Don, on my DSLR type camera, I use the ISO setting exactly as I use the gain on my videocamera, the less the better. In that way it is the same thing for us, and for the many of us that don't know the differences between the two, the long and short of it is that they are the same thing, for practical day to day use, right?

I raise the gain on my Xa10 just as I will raise the ISO on my GH2, under the exact same conditions: when I need a brighter image and I can't increase exposure any further and I'm at the slowest advisable shutter speed for the conditions I'm shooting in.

Kawika Ohumukini · December 18th, 2011, 01:07 AM

I'm convinced this is a physical filter that moves completely in. I zoomed all the way in and put a rectangle HDV-Z96 LED light facing slightly off to the side facing into the lens. When I went one click past f/4 I could see a reflection of my entire rectangle light appear. When I went back a click the light reflection disappeared. It appeared to be a filter rotating from the bottom up to cover the opening completely. As I continued to increase the f-stop the iris didn't appear to change size so I'm wondering how they're handling the continuous change.

November 11th, 2011, 08:27 AM	#32
Don Palomaki Wrangler Join Date: Sep 2001 Location: Northern VA Posts: 4,488	Re: Digital ND filter question xa10 I would say that gain is more like processing time,0 dB gain representing the nominal rcommended developer and processing time; increased gain being a bit like push processing. ISO would correspond to the sensitivity of the sensor picture elements; e.g., aperture, shutter speed, and incident light on the subject,required to produce an optimal imge at 0 dB gain. Of course traditionally video does not speak to ISO, more often the ratig has been the aperture required to produce something like 50 IRE from 1000 LUX on a 50% gray card at 1/60 shutter (or somethign like that, the figures I gave are not actuals to my knowledge). __________________ dpalomaki@dspalomaki.com

November 22nd, 2011, 07:03 AM	#33
Don Palomaki Wrangler Join Date: Sep 2001 Location: Northern VA Posts: 4,488	Re: Digital ND filter question xa10 A bit more on the ND thing. If we set the camcorder to Manual and manual exposure mode we get the Aperture, shutter, and gain displayed and can adjust them. They can be adjusted independently. The ND setting is coupled with the aperture setting, not separate. As we adjust the aperture we find that after f/4 we get steps in ND (and marks for 1/2, 1/4, etc.) until we run out of ND filter range, then the aperture stops continue to f/5.6 etc. We can see changes in image brightness as we move between ND setting marks inplying that the filter effect is continuous, e.g. a ND wedge that moves through the light path. It all make sense now. It is one less thing the user has to manage, fitting for a camcorder at this price point, and I am content to NOT have to deal with turning the ND ON/OFF it for the shooting I do.. __________________ dpalomaki@dspalomaki.com

December 16th, 2011, 10:15 AM	#36
Jeff Harper Inner Circle Join Date: Jun 2005 Location: Cincinnati, OH Posts: 8,425	Re: Digital ND filter question xa10 Yes, ISO and gain mean the same thing for our purposes. Why different names? Who knows. Wikipedia will have the answer, for sure.

December 16th, 2011, 10:50 AM	#40
Tom Hardwick Inner Circle Join Date: Mar 2003 Location: Billericay, England UK Posts: 4,711	Re: Digital ND filter question xa10 Rainer - you started this thread asking about a 'digital ND filter' and I thought we'd all talked you out of it. Obviously we haven't.

December 16th, 2011, 03:10 PM	#42
Jeff Harper Inner Circle Join Date: Jun 2005 Location: Cincinnati, OH Posts: 8,425	Re: Digital ND filter question xa10 Ranier, I believe that in science they say that the simplest explanation is the most likely the correct one, or something to that effect. In this case, the most likely and commonsense approach would be that Don is correct. First, no one here, so far, has ever heard of a ND filter that is purely electronic, but I understand that just because we've not hear about it does not mean it is not so. But the very idea sounds, well, to not seem logical, at least to me, but I'm no expert, so what do I know. Secondly, if this were a new type of ND filter that operates the way you fear that it does, Canon would likely mention it, if not brag about it. It would likely be well known. Thirdly the characteristics of the ND filter on the XA10 seen to be that of a genuine ND filter. When it kicks in the images look like a real ND filter is being used. If it acts like a rabbit, and smells like a rabbit, it is probably a rabbit. Fourth, if the sensor was particularly more sensitive and prone to danger from sunsets, etc, then Canon would probably mention it, because they would not want to deal with the many returns of damaged cameras. Long story short, the safest, most logical bet is that the camera uses a traditional ND filter. Just because you can't see it doesn't mean it is not there. Just turn it off, buy a variable ND filter or a regular one, and you won't have to worry at all.

December 16th, 2011, 09:29 PM	#44
Jeff Harper Inner Circle Join Date: Jun 2005 Location: Cincinnati, OH Posts: 8,425	Re: Digital ND filter question xa10 Don, on my DSLR type camera, I use the ISO setting exactly as I use the gain on my videocamera, the less the better. In that way it is the same thing for us, and for the many of us that don't know the differences between the two, the long and short of it is that they are the same thing, for practical day to day use, right? I raise the gain on my Xa10 just as I will raise the ISO on my GH2, under the exact same conditions: when I need a brighter image and I can't increase exposure any further and I'm at the slowest advisable shutter speed for the conditions I'm shooting in.

December 18th, 2011, 01:07 AM	#45
Kawika Ohumukini Regular Crew Join Date: Oct 2011 Location: San Diego, CA Posts: 185	Re: Digital ND filter question xa10 I'm convinced this is a physical filter that moves completely in. I zoomed all the way in and put a rectangle HDV-Z96 LED light facing slightly off to the side facing into the lens. When I went one click past f/4 I could see a reflection of my entire rectangle light appear. When I went back a click the light reflection disappeared. It appeared to be a filter rotating from the bottom up to cover the opening completely. As I continued to increase the f-stop the iris didn't appear to change size so I'm wondering how they're handling the continuous change.