How much greater latitude? at DVinfo.net

Steve Mullen · January 28th, 2007, 10:14 PM

After I published pix of a person inside a coffee shop who was perfectly exposed, while the outside objects behind him through windows were only moderately over-exposed -- folks asked if I could quantify how much greater latitude I thought the V1 had.

Today I took a near $25,000 HD camcorder to the coffee shop and recreated the situation.

Unlike the V1 in which a single exposure could capture the entire range, with this camera the inside exposure was correct at f/2.8 The outside was not only overexposed, but a white "mist" covered came from the window. This was from a $10,000 lens, so it wasn't flare. To get the outside exposed to look like the V1 did -- I had to move to f/5.6. Now the inside was way too dark.

Therefore, the V1 had almost 2-stops greater latitude.

However, the HD camcorder -- unlike the the V1, was not set for maximum latitude. I forgot. :) So the real increase is more like 1-stop. The extra stop makes life EZier.

Obviously, this was not a 2/3-inch CCD HD camera, so the extra latitude is relative to THIS camera and not a $50K to $100K camcorder. I'm not going to post pix for obvious reasons. But, this camera does use CCDs.

I have to think a CMOS pixel accepts a very wide light range and can output a signal that's proportionate to this input.

-------------------

I have found one can push the V1's exposure a bit too far, however. For awhile I was working with zebra Off and pushing the signal right to the top of the Histogram. This creates a "Jughead" look. After turning on Zerba at 100IRE, I have kept the top of the signal to the orange line -- or worse-case, mid-way between the orange line and the top of the histogram. This looks better and matches the fact that many NLE's clip at 100IRE. So I expose for a bit of Zebra on the highlights.

Tony Tremble · January 28th, 2007, 10:48 PM

How do you know if the the latitude is due to CMOS or from the technowizardy of the EIP?

I haven't noticed any difference between the V1 and XH-A1 which I thought I would given all the press about increase of latitude of CMOS chips. If there is any difference it must be very small.

TT

Mikko Lopponen · January 29th, 2007, 03:22 AM

It's probably the EIP and 14-bit processing. The better latitude only appears on good lighting as the EIP can map pixels, in low light situations there would be no difference.

Stephen van Vuuren · January 29th, 2007, 09:18 AM

Quote:

Originally Posted by Tony Tremble

How do you know if the the latitude is due to CMOS or from the technowizardy of the EIP?

This is the million dollar question with this camera and though I've asked for info and footage before to test latitude (which I realize would be a pain to shoot), I
m surprised few people are obsessed (as I am) with latitude.

I sold my XL1 and bought the first DVX100 I could get my hands on to get progressive images with much more latitude (thanks to cinegamma primarily). I still have a DVX100a now instead of HD because none of the low-cost HD cams I've seen beats the DVX in the combination of latitude, sensitivity and noise.

If the V1 pulls more stops of latitude, that's big news as that's what 35mm motion picture film still holds over DV & HD. However, recent DSLR tech is now more sensitive, lower noise, higher latitude and higher rez than 35mm film (some very interesting tests on this at Luminous Landscape's site).

So the day is coming with motion images - the question is when, how and how much?

I've seen V1 images that appear to offer much great latitude than most other digital video cameras but then I've seem images that look much like the mediocre video we all shoot around. So it is actual latitude, processing, combo? Is it related to the issues we've seen?

However, there is no "Barry Green" or "Adan Wilt" of V1 shooters out there shooting some reasonablely well thought out controlled shots to shed more light on exactly how this camera & settings works (with at least one well know camera for comparison). If I had some spare cash or credit, I would buy one and do it.

Anyone? Anyone?

Steve Mullen · January 29th, 2007, 01:29 PM

Quote:

Originally Posted by Stephen van Vuuren

So it is actual latitude, processing, combo?

Does it matter HOW?

Remember, you don't get 1.5 to 2 stops by simply shooting. You have to set-up the camera correctly -- then make a WIDE profile. You only want to use WIDE in high contrast situations.

In any case, the fact some footage looks great and some bad has far more to do with the shooter than the camera.

Bill Ravens · January 29th, 2007, 01:43 PM

Ok, so I don't have a lot of scientific proof, but, I do have a fair bit of field experience.

scene 1, act 1: I have been shooting a great deal of outdoor scenes with a Canon XL2, DV, 24fps, high desert, full sun at mid-day, with some of the talent in and out of the shadows of low trees. A tuff lighting situation anywhere....and one where fill lights are necessary. Without fill, exposing for the shadows, the skies weren't that deep hi-desert blue, but washed out pastel blue or white. Exposing for the skies, and the shadows were featureless. No amount of fiddling could fix that.

scene 2 act 1: repeat previous but with a JVC HD110 shooting HDV, 24p. Expose for the shadows, and whoa! wait a minute, is that a deep blue sky I see? Indeed it is. How utterly surprising and happy that makes me. Quick check to frame the shot in DV, same camera.....skies a little less blue, but still better than the XL2.

Stephen van Vuuren · January 29th, 2007, 01:48 PM

Quote:

Originally Posted by Steve Mullen

Does it matter HOW?

Remember, you don't get 1.5 to 2 stops by simply shoting. You have to set-up the camera correctly -- then make a WIDE profile. You only want to use WIDE in high contrast situations.

In any case, the fact some footage looks great and some bad has far more to do with the shooter than the camera.

If the extra latitude is achieved by DSP alternation of portions of the frame (e.g. highlights/shadows), it has great potential for artifacts or unexpected results in various lighting conditions. Any V1 shooter would want to know exactly how this works.

If the latitude is mostly native SNR on the sensor, then a nice ASA/ISO info/noise levels, stops etc. would be useful data.

What I'm looking for is actual images shot in controlled fashion with at least one known camera as a base - right now I have not seen this type of footage posted.

Kevin Shaw · January 29th, 2007, 01:52 PM

Quote:

Originally Posted by Stephen van Vuuren

So the day is coming with motion images - the question is when, how and how much?

Based on initial reports it sounds like we may get what you're after in a few more weeks with "Red," at a starting price of around $25-30K for a functional rig. That doesn't help those of us with more modest budgets, but it points the way to better things in the future at affordable prices. I doubt there's any technical reason why a digital movie camera can't surpass the quality of film at some point; it's just a question of improving digital technology enough to translate light effectively.

Ten years ago few professionals believed that digital photography would ever amount to much; today most camera companies are abandoning development of film-based still cameras. Welcome to the 21st century...

Stephen van Vuuren · January 29th, 2007, 01:57 PM

Quote:

Originally Posted by Kevin Shaw

Based on initial reports it sounds like we may get what you're after in a few more weeks with "Red," at a starting price of around $25-30K for a functional rig. That doesn't help those of us with more modest budgets, but it points the way to better things in the future at affordable price

I agree RED is clearly the future. And if I had $30K, I would have a pre-order in :)

Chris Medico · January 29th, 2007, 02:20 PM

Quote:

Originally Posted by Stephen van Vuuren

This is the million dollar question with this camera and though I've asked for info and footage before to test latitude (which I realize would be a pain to shoot), I
m surprised few people are obsessed (as I am) with latitude.

I sold my XL1 and bought the first DVX100 I could get my hands on to get progressive images with much more latitude (thanks to cinegamma primarily). I still have a DVX100a now instead of HD because none of the low-cost HD cams I've seen beats the DVX in the combination of latitude, sensitivity and noise.

If the V1 pulls more stops of latitude, that's big news as that's what 35mm motion picture film still holds over DV & HD. However, recent DSLR tech is now more sensitive, lower noise, higher latitude and higher rez than 35mm film (some very interesting tests on this at Luminous Landscape's site).

So the day is coming with motion images - the question is when, how and how much?

I've seen V1 images that appear to offer much great latitude than most other digital video cameras but then I've seem images that look much like the mediocre video we all shoot around. So it is actual latitude, processing, combo? Is it related to the issues we've seen?

However, there is no "Barry Green" or "Adan Wilt" of V1 shooters out there shooting some reasonablely well thought out controlled shots to shed more light on exactly how this camera & settings works (with at least one well know camera for comparison). If I had some spare cash or credit, I would buy one and do it.

Anyone? Anyone?

Stephen, I have a V1 and live just to the east of you in Burlington/Efland. I'm happy to come over sometime and let you shoot with the camera and put it against what you normally use.

I've made my choice and I'm going to make the most of it. If there is anything that I can do to help others I'm happy to do it.

Also maybe you could give me a lesson in good shooting. ;)

Chris

Stephen van Vuuren · January 29th, 2007, 05:45 PM

Quote:

Originally Posted by Chris Medico

Stephen, I have a V1 and live just to the east of you in Burlington/Efland. I'm happy to come over sometime and let you shoot with the camera and put it against what you normally use.

I've made my choice and I'm going to make the most of it. If there is anything that I can do to help others I'm happy to do it.

Also maybe you could give me a lesson in good shooting. ;)

Chris

I don't know about a lesson but we can certainly meet up and shoot some A/B footage with DVX100 on latitude, 24p rendering etc. Contact me via www.sv2studios.com

January 28th, 2007, 10:14 PM	#1
Steve Mullen HDV Cinema Join Date: Feb 2003 Location: Las Vegas Posts: 4,007	How much greater latitude? After I published pix of a person inside a coffee shop who was perfectly exposed, while the outside objects behind him through windows were only moderately over-exposed -- folks asked if I could quantify how much greater latitude I thought the V1 had. Today I took a near $25,000 HD camcorder to the coffee shop and recreated the situation. Unlike the V1 in which a single exposure could capture the entire range, with this camera the inside exposure was correct at f/2.8 The outside was not only overexposed, but a white "mist" covered came from the window. This was from a $10,000 lens, so it wasn't flare. To get the outside exposed to look like the V1 did -- I had to move to f/5.6. Now the inside was way too dark. Therefore, the V1 had almost 2-stops greater latitude. However, the HD camcorder -- unlike the the V1, was not set for maximum latitude. I forgot. :) So the real increase is more like 1-stop. The extra stop makes life EZier. Obviously, this was not a 2/3-inch CCD HD camera, so the extra latitude is relative to THIS camera and not a $50K to $100K camcorder. I'm not going to post pix for obvious reasons. But, this camera does use CCDs. I have to think a CMOS pixel accepts a very wide light range and can output a signal that's proportionate to this input. ------------------- I have found one can push the V1's exposure a bit too far, however. For awhile I was working with zebra Off and pushing the signal right to the top of the Histogram. This creates a "Jughead" look. After turning on Zerba at 100IRE, I have kept the top of the signal to the orange line -- or worse-case, mid-way between the orange line and the top of the histogram. This looks better and matches the fact that many NLE's clip at 100IRE. So I expose for a bit of Zebra on the highlights. __________________ Switcher's Quick Guide to the Avid Media Composer >>> http://home.mindspring.com/~d-v-c Last edited by Steve Mullen; January 28th, 2007 at 11:11 PM.

January 28th, 2007, 10:48 PM	#2
Tony Tremble Regular Crew Join Date: Oct 2006 Location: United Kingdom Posts: 451	How do you know if the the latitude is due to CMOS or from the technowizardy of the EIP? I haven't noticed any difference between the V1 and XH-A1 which I thought I would given all the press about increase of latitude of CMOS chips. If there is any difference it must be very small. TT

January 29th, 2007, 03:22 AM	#3
Mikko Lopponen Major Player Join Date: Mar 2006 Location: Helsinki, Finland Posts: 393	It's probably the EIP and 14-bit processing. The better latitude only appears on good lighting as the EIP can map pixels, in low light situations there would be no difference.

January 29th, 2007, 01:43 PM	#6
Bill Ravens Join Date: Nov 2001 Location: santa fe, nm Posts: 3,264 Images: 10	Ok, so I don't have a lot of scientific proof, but, I do have a fair bit of field experience. scene 1, act 1: I have been shooting a great deal of outdoor scenes with a Canon XL2, DV, 24fps, high desert, full sun at mid-day, with some of the talent in and out of the shadows of low trees. A tuff lighting situation anywhere....and one where fill lights are necessary. Without fill, exposing for the shadows, the skies weren't that deep hi-desert blue, but washed out pastel blue or white. Exposing for the skies, and the shadows were featureless. No amount of fiddling could fix that. scene 2 act 1: repeat previous but with a JVC HD110 shooting HDV, 24p. Expose for the shadows, and whoa! wait a minute, is that a deep blue sky I see? Indeed it is. How utterly surprising and happy that makes me. Quick check to frame the shot in DV, same camera.....skies a little less blue, but still better than the XL2.