Rolling Shutter??

Tom Hardwick · February 23rd, 2009, 05:23 PM

Just to make things even clearer, hopefully, f stops are mathematically derived from the ratio of the iris opening to the focal length.

So you could have two lenses side by side, both with the same focal length and maximum aperture. One could be made of the finest glass and be beautifully multi-coated. The other could be made from cloudy plastic and have dead insects stuck to one of the glass elements.

Guess what - they're both the same specification, but one lens passes a lot more light than the other. OK, now we have two lenses with the same spec but one is made up of 8 elements, the other has four. The 8 element lens is probably better corrected, but it certainly won't pass as much light even if they both claim (accurately) that they're f/2.8.

So real and proper cameras have lenses marked in T stops. T for transmission, and this sorts the men from the boys. Generally a 20x f1.6 zoom won't pass as much light as a 10x f/1.6 zoom simply because there are far more elements in the long zoom's lineup. But if both lenses are marked T 1.6 then they will indeed both pass the same amount of light. The 20 x zoom will be manufactured with a wider maximum f stop to counteract its greater light losses. Bench testing determines the lens' T stop - it has nothing to do with the f stop.

We're sold f stops because it looks more impressive. Remember your lens is still an f/1.6 even with the lens cap on, and it's transmitting no light at all.

tom.

Adam Gold · February 23rd, 2009, 05:56 PM

Quote:

Originally Posted by Tom Hardwick

f stops are mathematically derived from the ratio of the iris opening to the focal length..

This is a critically important point, and I'm glad you made it. It belongs in the lens ramping discussion as well. Most of those who are most hysterical about this issue seem to fail to understand that the f-number is changing as you zoom, merely because the ratio of iris size to focal length is changing, and you aren't losing my nearly as much light as the numbers would indicate. When you go from, say, 1.4 to 2.8 as you zoom, it only means the focal length has doubled -- you are not really losing 75% of the light -- the iris size is relatively unchanged and your only light loss is as the elements shift (so T-value would change).

As you and others have pointed out, to have no ramping, you'd need a lens with the diameter of a dinner plate -- which the pro studio and sports cameras do, even though they still have relatively small 2/3" chips. And the iris would have to open up dramatically as you zoomed to maintain the same "f-stop."

Most of this hysteria over this non-phenomenon is just uninformed nonsense.

Jeff Harper · February 23rd, 2009, 06:33 PM

We're indeed fortunate to that you see fit to straighten us out Andrew.

Maybe someone will tell us soon that these things are actually benefits, but that since we don't understand them we can't appreciate them. :)

Stelios Christofides · February 24th, 2009, 12:00 AM

Wow guys!!! what a wonderful site this is with all the info you can get. You know, you can google everything, but it's not the same, because here, you get people who interact with each other and you "hear" other opinions too. I have learned so much in these forums that I would have never learned anywhere else.

Stelios

Jeff Harper · February 24th, 2009, 12:14 AM

I agree Stelios. There is LOTS to learn here!

Tom Hardwick · February 24th, 2009, 03:50 AM

Quote:

Originally Posted by Adam Gold

you aren't losing nearly as much light as the numbers would indicate. When you go from, say, 1.4 to 2.8 as you zoom, you are not really losing 75% of the light

Not quite on your side here Adam. If your f/1.4 lens has smoothly changed into an f/2.8 lens you'll need to quadruple the light in the room to get the same exposure on your chips.

Or to put it another way. If you can film at f/1.4 with two 100 watt lamps lit (say) then at f/2.8 you'll need to turn on eight 100 watt lamps to get the same exposure. This shows how dramatic a two stop loss can be, and the Z5 loses two stops, wide to tele.

tom.

Todd Clark · February 24th, 2009, 09:43 AM

Quote:

Originally Posted by Adam Gold

This is a critically important point, and I'm glad you made it. It belongs in the lens ramping discussion as well. Most of those who are most hysterical about this issue seem to fail to understand that the f-number is changing as you zoom, merely because the ratio of iris size to focal length is changing, and you aren't losing my nearly as much light as the numbers would indicate. When you go from, say, 1.4 to 2.8 as you zoom, it only means the focal length has doubled -- you are not really losing 75% of the light -- the iris size is relatively unchanged and your only light loss is as the elements shift (so T-value would change).

As you and others have pointed out, to have no ramping, you'd need a lens with the diameter of a dinner plate -- which the pro studio and sports cameras do, even though they still have relatively small 2/3" chips. And the iris would have to open up dramatically as you zoomed to maintain the same "f-stop."

Most of this hysteria over this non-phenomenon is just uninformed nonsense.

"This non-phenomenon is just uninformed nonsense" is your opinion!! At the end of the day when you zoom all the way in you loose to much light too be of any benefit plain and simple!!!

Tom Hardwick · February 24th, 2009, 09:51 AM

Quote:

Originally Posted by Todd Clark

when you zoom all the way in you loose to much light too be of any benefit plain and simple!!!

Calm down Todd. This situation *only* occurs when light levels are so low that even using +18dB of gain you're forced into using apertures wider than f/3.4. For most of a camcorder's life this just doesn't happen - in fact I'd think that for most filmmakers using ND filters is much more common.

tom.

Jeff Harper · February 24th, 2009, 10:16 AM

As you know Tom, I feel similarly as Todd. But what you say it absolutely correct, 90% of the time it is a non- issue, maybe 95%!

I think for typical shooting, for most any typical shooter, it really is not a big deal. I agree on that wholeheartedly. We know that is true because so few complain about it.

On the other hand, for the few of us for whom the indoor extreme closeups at about 10 feet are important, the lens ramping goes from being unimportant to "Oh my God!".

I remember how I freaked out when I discovered it. I didn't even know it existed. There I am about 8 feet or less, and I could not fill my LCD with my bride's face...because the exposure changed and it was unusably dark.

I do feel hopeful with your post Tom re: the V7. That, again was really great info, and for the likes of me it seems like the solution.

I've said how god-awful the rolling shutter can look, and I meant it. But I can and will live with it.

Adam Gold · February 24th, 2009, 02:24 PM

Quote:

Originally Posted by Tom Hardwick

Not quite on your side here Adam. If your f/1.4 lens has smoothly changed into an f/2.8 lens you'll need to quadruple the light in the room to get the same exposure on your chips.

You could be right, but I don't think it scales that way. Going back to your earlier post, if we were talking t-stops that would undoubtedly be true as those are an actual measure of light. But with f-stops, as you pointed out, these are merely numerical expressions of physical measurements and have nothing to do with light transmission. There really isn't an exact link between f number and, say, lumens.

Here's an experiment: Take a wide shot full open (say 1.6) and then zoom in fully. Go to your NLE and zoom up the wide shot so it matches the tele you took in the cam. Even though there are at least two, maybe three, stops difference between either end of the zoom, is the luma of the CU really only 25% of the wide shot? (Obviously you'd have to shoot on full manual with fixed gain and shutter.) I'm betting it isn't, but will happily be proven wrong if it means we all learn something.

Greg Laves · February 24th, 2009, 04:22 PM

After reading the latest postings, and Adams post from yesterday, in particular, I did a test with my Z7. I have a broadcast (style?) Fujinon lens for my Z7. It will maintain f1.4 from full wide to full telephoto. Since I have doubts that anyone's eye is so finely calibrated to tell very minimal exposure differences, I used the zebras to evaluate the exposure. And what I found was that from full wide to full telephoto a grey object would show virtually negligible change throughout the entire range of the lens. And realistically, I would never need to adjust the exposure to compensate for the zoom. No matter what the zoom was, the 70% zebra indicated that the subject was correctly exposed right at f1.6 to f1.7.

Jeff Harper · February 24th, 2009, 04:24 PM

That is awesome Greg. Got to get me one.

Terence Murphy · February 24th, 2009, 05:02 PM

Quote:

Originally Posted by Jeff Harper

Maybe someone will tell us soon that these things are actually benefits, but that since we don't understand them we can't appreciate them. :)

This is obviously an intentional benefit -- if the lens held at f1.6 through the whole zoom, then the depth of field at 20x, focused on a subject 10 feet away, would be so shallow that you couldn't get the entire face in focus! The tip of the nose to the ears would be too far off the focal plane to be sharp! So that F3.4 is a good thing :-)

Yes, I jest.

-Terence

Greg Laves · February 24th, 2009, 05:08 PM

Quote:

Originally Posted by Terence Murphy

This is obviously an intentional benefit -- if the lens held at f1.6 through the whole zoom, then the depth of field at 20x, focused on a subject 10 feet away, would be so shallow that you couldn't get the entire face in focus! The tip of the nose to the ears would be too far off the focal plane to be sharp! So that F3.4 is a good thing :-)

Yes, I jest.

-Terence

I thought that was why Sony gave us only 1/3" chips, so the DOF wouldn't be too shallow.

Tom Hardwick · February 25th, 2009, 02:13 AM

[QUOTE=Greg Laves;1017775]I have a broadcast (style?) Fujinon lens for my Z7. It will maintain f1.4 from full wide to full telephoto./QUOTE]

What is this Fujinon lens you have on the Z7 Greg, and how much zoom does it have? The stock 12x Fujinon that comes with the Z7 most certainly doesn't have an f/1.4 maximum aperture - it ramps from a nominal f/1.6 to f/2.0.

tom.

February 24th, 2009, 12:00 AM	#64
Stelios Christofides Trustee Join Date: Jul 2003 Location: Nicosia, CYPRUS Posts: 1,080	Wow guys!!! what a wonderful site this is with all the info you can get. You know, you can google everything, but it's not the same, because here, you get people who interact with each other and you "hear" other opinions too. I have learned so much in these forums that I would have never learned anywhere else. Stelios __________________ My Blog: http://steliosc.blogspot.com "I hope for nothing, I fear nothing, I am free" Nikos Kazantzakis

February 23rd, 2009, 05:23 PM	#61
Tom Hardwick Inner Circle Join Date: Mar 2003 Location: Billericay, England UK Posts: 4,711	Just to make things even clearer, hopefully, f stops are mathematically derived from the ratio of the iris opening to the focal length. So you could have two lenses side by side, both with the same focal length and maximum aperture. One could be made of the finest glass and be beautifully multi-coated. The other could be made from cloudy plastic and have dead insects stuck to one of the glass elements. Guess what - they're both the same specification, but one lens passes a lot more light than the other. OK, now we have two lenses with the same spec but one is made up of 8 elements, the other has four. The 8 element lens is probably better corrected, but it certainly won't pass as much light even if they both claim (accurately) that they're f/2.8. So real and proper cameras have lenses marked in T stops. T for transmission, and this sorts the men from the boys. Generally a 20x f1.6 zoom won't pass as much light as a 10x f/1.6 zoom simply because there are far more elements in the long zoom's lineup. But if both lenses are marked T 1.6 then they will indeed both pass the same amount of light. The 20 x zoom will be manufactured with a wider maximum f stop to counteract its greater light losses. Bench testing determines the lens' T stop - it has nothing to do with the f stop. We're sold f stops because it looks more impressive. Remember your lens is still an f/1.6 even with the lens cap on, and it's transmitting no light at all. tom.

February 23rd, 2009, 06:33 PM	#63
Jeff Harper Inner Circle Join Date: Jun 2005 Location: Cincinnati, OH Posts: 8,425	We're indeed fortunate to that you see fit to straighten us out Andrew. Maybe someone will tell us soon that these things are actually benefits, but that since we don't understand them we can't appreciate them. :)

February 24th, 2009, 12:14 AM	#65
Jeff Harper Inner Circle Join Date: Jun 2005 Location: Cincinnati, OH Posts: 8,425	I agree Stelios. There is LOTS to learn here!

February 24th, 2009, 10:16 AM	#69
Jeff Harper Inner Circle Join Date: Jun 2005 Location: Cincinnati, OH Posts: 8,425	As you know Tom, I feel similarly as Todd. But what you say it absolutely correct, 90% of the time it is a non- issue, maybe 95%! I think for typical shooting, for most any typical shooter, it really is not a big deal. I agree on that wholeheartedly. We know that is true because so few complain about it. On the other hand, for the few of us for whom the indoor extreme closeups at about 10 feet are important, the lens ramping goes from being unimportant to "Oh my God!". I remember how I freaked out when I discovered it. I didn't even know it existed. There I am about 8 feet or less, and I could not fill my LCD with my bride's face...because the exposure changed and it was unusably dark. I do feel hopeful with your post Tom re: the V7. That, again was really great info, and for the likes of me it seems like the solution. I've said how god-awful the rolling shutter can look, and I meant it. But I can and will live with it.

February 24th, 2009, 04:22 PM	#71
Greg Laves Major Player Join Date: Nov 2007 Location: Corpus Christi, TX Posts: 640	After reading the latest postings, and Adams post from yesterday, in particular, I did a test with my Z7. I have a broadcast (style?) Fujinon lens for my Z7. It will maintain f1.4 from full wide to full telephoto. Since I have doubts that anyone's eye is so finely calibrated to tell very minimal exposure differences, I used the zebras to evaluate the exposure. And what I found was that from full wide to full telephoto a grey object would show virtually negligible change throughout the entire range of the lens. And realistically, I would never need to adjust the exposure to compensate for the zoom. No matter what the zoom was, the 70% zebra indicated that the subject was correctly exposed right at f1.6 to f1.7.

February 24th, 2009, 04:24 PM	#72
Jeff Harper Inner Circle Join Date: Jun 2005 Location: Cincinnati, OH Posts: 8,425	That is awesome Greg. Got to get me one.

February 25th, 2009, 02:13 AM	#75
Tom Hardwick Inner Circle Join Date: Mar 2003 Location: Billericay, England UK Posts: 4,711	[QUOTE=Greg Laves;1017775]I have a broadcast (style?) Fujinon lens for my Z7. It will maintain f1.4 from full wide to full telephoto./QUOTE] What is this Fujinon lens you have on the Z7 Greg, and how much zoom does it have? The stock 12x Fujinon that comes with the Z7 most certainly doesn't have an f/1.4 maximum aperture - it ramps from a nominal f/1.6 to f/2.0. tom.