Letus Flipped GG´s and Light Loss at DVinfo.net

Holger Leonhard · May 22nd, 2006, 06:40 PM

I built a rig with the letus flipped and a HC-3. First pictures are beautiful, but I found a really big light loss of approx. 4 - 6 stops - compared the cam without letus and letus with a canon FD 50mm / 1.4 with full open Iris.
Other Comparison: when I film in a room with a ~60W light bulb without adapter, the Image is o.k., not too bright, but also not too dark with max Gain. With adapter it gets very dark.
What´s your observation ?
Is there better - brighter - GG´s available ?

Quyen Le · May 23rd, 2006, 10:45 AM

I wonder what method are you using to measure light loss? What I do is set shutter speed to 60, put the adapter on where light is sufficient and write down the f-stop. Then do the same (same frame) without adapter and compare the f-stop. The current GG doesn't lose that much light. Thanks.

Quyen

Francois Poitras · May 23rd, 2006, 12:10 PM

Yes, Quyen suggests a good method here.

Here is what I personally do to measure light loss with an adapter.

1. Place a chart on a wall in a room with no external light. Light the chart sufficiently (camcorder around F/4 - F/5.6).
2. Put the camcorder on a tripod and frame the chart using the exact same zoom setting as you would with the adapter on.
3. Shoot the frame with the camcorder only at different aperture settings at every aperture increment between, say open and F/8. Record the different settings vocally as you change the aperture settings.
4. Install the adapter and move the cam-adapter closer to have the same frame of the chart.
5. Turn on the adapter and shoot the chart (framed identically) at different aperture settings in the same range as with the camcorder only.
6. Capture both in your NLE and compare brightness at the different aperture settings. If you get about the same brightness when cam alone is at F4 and adapter is F2, the adapter plus lens loses 2 F/stops.

This is the method I use. It can be debated whether the camcorder alone should be set at the same zoom setting it is at with the adapter on, or placed at the same distance from the chart and zoomed back. The latter would also add the light loss from the extra zoom-in necessary with the adapter.

Holger Leonhard · May 25th, 2006, 11:26 AM

o.k. now I did a more precise test (shutter = 50):

The Motiv without Letus, Camera 0db F4:

http://www.deepcode1.com/downloads/l...4_0_gain_0.png
_________

The Motiv with Letus, Camera +6db F1.8
Glass: Canon 50mm F 1.4 full open

http://www.deepcode1.com/downloads/l..._50mm_F1_4.png
_________

The Motiv with Letus, Camera +9db F1.8
Glass: Vivitar 28mm F 2.0 full open

http://www.deepcode1.com/downloads/l..._28mm_F2_0.png

everything original but deinterlacing and some color correction, because with Letus the colors appear different.

so the difference is 4.0/0db to 1.8/+9db with a 2.0 Glass and
4.0/0db to 1.8/+6db with a 1.4 Glass

Francois Poitras · May 25th, 2006, 12:42 PM

Quite interesting. I had suggested to use F4 as your reference (camcorder only) image because I did not think you would need to revert to gain. In order to get a precise "F/stop only" measurement, it looks like you would need even more light and use a higher F/stop range with the camcorder-only as a reference (F2 to F8, for example).

Still, I believe your test as it is gives a good indication.

Holger Leonhard · May 26th, 2006, 12:50 AM

So I do the maths - result:

F 4 to F 1.4 : 4 - 2.8 - 1.8 - 1.4 is 3 stops
0 db to 6 db Gain is 2 stops
Together it is 5 stops light loss. That´s a lot.

@Quyen - are you shure my Adapter has the latest GG included ?

Holger Leonhard · June 10th, 2006, 11:32 AM

o.k was very busy the last weeks - now jobs ar done and some time for another test ..
I found new results which I can not explain, really.
I tested now with more light (500 W tungsten, no filters, approx. 1 m distance)

Here the scene shot without letus, gain 0db, shutter 50, HC-3 iris = 4,

http://www.deepcode1.com/downloads/l...4_gain_0db.jpg

And here the scene shot with letus, gain 0db, shutter 50, HC-3 iris = 2,
Glass: Canon 50mm 1.4 full open

http://www.deepcode1.com/downloads/l..._50mm_F1_4.jpg

In this test I have a light loss from F4 to F2, thats only 2 (!) stops light loss !!!
My apologies to Quyen, there must be something going wron on the first test, I really cannot explain whaht.
I repeat the first test later ..

Forrest Schultz · June 12th, 2006, 12:15 PM

Holger, is it possible that you can post some footage sometime, i really like the look of the HC3 and want to see some more samples from it. thank you very much

Holger Leonhard · June 13th, 2006, 03:45 PM

Here some family & friends / bbq stuff grabs I did for testing. I hope the current job gives me time to do some additional footage from munich in the next weeks, here is a great summer festival and of course the soccer world cup with lots of VERY hot brazilian girls .. ;-)
The Letus performs great under good light conditions, but I have the feel when the light dims the light loss increases rapidely. At daylight just ~2 stops, at evening it rises to 4-5 stops. It seems like the GG soaks a fix part of the incoming light.
On some of the sunlight images I am not shure if It was the Vivitar 28-80 @2.8 or the Canon 50mm @ 1.4. At Daylight, the Vivitar Zoom is wonderful - it gives shallow DOF and lots of light, on evening you urgently need primes like the Canon 50mm 1.4, but it is VERY hard to get the focus on moving objects.

http://www.deepcode1.com/downloads/l...q/bbq_mike.jpg
http://www.deepcode1.com/downloads/l.../bbq_sarah.jpg
http://www.deepcode1.com/downloads/l...q_things_1.jpg
http://www.deepcode1.com/downloads/l...q_things_2.jpg
http://www.deepcode1.com/downloads/l...q/bbq_vino.jpg
http://www.deepcode1.com/downloads/l.../bbq_cigar.jpg
http://www.deepcode1.com/downloads/l...q/bbq_dish.jpg
http://www.deepcode1.com/downloads/l...bq/bbq_dog.jpg
http://www.deepcode1.com/downloads/l...bbq_fire_1.jpg
http://www.deepcode1.com/downloads/l...bq_flowers.jpg
http://www.deepcode1.com/downloads/l.../bbq_melon.jpg

Holger Leonhard · June 13th, 2006, 03:51 PM

addy: viewed on the TV, all the images look much brighter. I did the framegrabs straight from VLC and saved them as jpg with 85% quality, no visible loss against PNG but much smaller.

Forrest Schultz · June 13th, 2006, 07:12 PM

awesome! thank you very much

Dennis Wood · June 13th, 2006, 10:39 PM

Holger, you'll find that if you test a variety of scenes and light levels, your loss will vary.

Holger Leonhard · June 14th, 2006, 12:47 AM

yes. Maybe the values the camera shows is not correct - unfortunately the HC-3 (the HC-1, too) don´t show the iris, shutter and db values while recording, but the values are recorded into the datastream and can be seen on Playback.
Also I don´t know if the stop - values shown of the HC-3 are correct - it´s a consumer cam.
On some of the pictures in sunlight it is possible that I used the vivitar @2.8 instead the canon 50mm. But all the pictures marked as "Vivitar 2.8" are done this Glass. So you cann assume to need a Stop of "4" on the Camera when using a 2.8 stop on the Glass and shooting in bright sunlight.

Holger Leonhard · June 14th, 2006, 12:50 AM

are any DVX / Letus flip users here to post some low light images with exact stop and db values ? would be interresting!

to the HC-3: The cam is sensitive very close to the FX-1 we found, maybe the FX-1 is 1/2 stop or so better, but not much. Also you can boost the HC-3 to +18db without too much noise, but on full gain there is some loss of sharpness and loss of color.

Bob Hart · June 14th, 2006, 10:52 AM

An Australian DP did some tests on the Mini35 and there is a web page with his article on the subject.

As I recall it, cumulative light loss through groundglass based relay systems at the tighter apertures is much less than at wide apertures. This might go some way to explaining inconsistent results from tests and measures metioned above.

May 22nd, 2006, 06:40 PM	#1
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	Letus Flipped GG´s and Light Loss I built a rig with the letus flipped and a HC-3. First pictures are beautiful, but I found a really big light loss of approx. 4 - 6 stops - compared the cam without letus and letus with a canon FD 50mm / 1.4 with full open Iris. Other Comparison: when I film in a room with a ~60W light bulb without adapter, the Image is o.k., not too bright, but also not too dark with max Gain. With adapter it gets very dark. What´s your observation ? Is there better - brighter - GG´s available ?

May 23rd, 2006, 10:45 AM	#2
Quyen Le Major Player Join Date: Jun 2005 Posts: 356	I wonder what method are you using to measure light loss? What I do is set shutter speed to 60, put the adapter on where light is sufficient and write down the f-stop. Then do the same (same frame) without adapter and compare the f-stop. The current GG doesn't lose that much light. Thanks. Quyen __________________ Affordable Adapter http://www.adapterplace.com

May 23rd, 2006, 12:10 PM	#3
Francois Poitras Regular Crew Join Date: Nov 2004 Location: Québec (Canada) Posts: 133	Yes, Quyen suggests a good method here. Here is what I personally do to measure light loss with an adapter. 1. Place a chart on a wall in a room with no external light. Light the chart sufficiently (camcorder around F/4 - F/5.6). 2. Put the camcorder on a tripod and frame the chart using the exact same zoom setting as you would with the adapter on. 3. Shoot the frame with the camcorder only at different aperture settings at every aperture increment between, say open and F/8. Record the different settings vocally as you change the aperture settings. 4. Install the adapter and move the cam-adapter closer to have the same frame of the chart. 5. Turn on the adapter and shoot the chart (framed identically) at different aperture settings in the same range as with the camcorder only. 6. Capture both in your NLE and compare brightness at the different aperture settings. If you get about the same brightness when cam alone is at F4 and adapter is F2, the adapter plus lens loses 2 F/stops. This is the method I use. It can be debated whether the camcorder alone should be set at the same zoom setting it is at with the adapter on, or placed at the same distance from the chart and zoomed back. The latter would also add the light loss from the extra zoom-in necessary with the adapter. __________________ François

May 25th, 2006, 12:42 PM	#5
Francois Poitras Regular Crew Join Date: Nov 2004 Location: Québec (Canada) Posts: 133	Quite interesting. I had suggested to use F4 as your reference (camcorder only) image because I did not think you would need to revert to gain. In order to get a precise "F/stop only" measurement, it looks like you would need even more light and use a higher F/stop range with the camcorder-only as a reference (F2 to F8, for example). Still, I believe your test as it is gives a good indication. __________________ François

June 14th, 2006, 12:47 AM	#13
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	yes. Maybe the values the camera shows is not correct - unfortunately the HC-3 (the HC-1, too) don´t show the iris, shutter and db values while recording, but the values are recorded into the datastream and can be seen on Playback. Also I don´t know if the stop - values shown of the HC-3 are correct - it´s a consumer cam. On some of the pictures in sunlight it is possible that I used the vivitar @2.8 instead the canon 50mm. But all the pictures marked as "Vivitar 2.8" are done this Glass. So you cann assume to need a Stop of "4" on the Camera when using a 2.8 stop on the Glass and shooting in bright sunlight. Last edited by Holger Leonhard; June 14th, 2006 at 11:13 AM.

May 25th, 2006, 11:26 AM	#4
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	o.k. now I did a more precise test (shutter = 50): The Motiv without Letus, Camera 0db F4: http://www.deepcode1.com/downloads/l...4_0_gain_0.png _________ The Motiv with Letus, Camera +6db F1.8 Glass: Canon 50mm F 1.4 full open http://www.deepcode1.com/downloads/l..._50mm_F1_4.png _________ The Motiv with Letus, Camera +9db F1.8 Glass: Vivitar 28mm F 2.0 full open http://www.deepcode1.com/downloads/l..._28mm_F2_0.png everything original but deinterlacing and some color correction, because with Letus the colors appear different. so the difference is 4.0/0db to 1.8/+9db with a 2.0 Glass and 4.0/0db to 1.8/+6db with a 1.4 Glass

May 26th, 2006, 12:50 AM	#6
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	So I do the maths - result: F 4 to F 1.4 : 4 - 2.8 - 1.8 - 1.4 is 3 stops 0 db to 6 db Gain is 2 stops Together it is 5 stops light loss. That´s a lot. @Quyen - are you shure my Adapter has the latest GG included ?

June 10th, 2006, 11:32 AM	#7
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	o.k was very busy the last weeks - now jobs ar done and some time for another test .. I found new results which I can not explain, really. I tested now with more light (500 W tungsten, no filters, approx. 1 m distance) Here the scene shot without letus, gain 0db, shutter 50, HC-3 iris = 4, http://www.deepcode1.com/downloads/l...4_gain_0db.jpg And here the scene shot with letus, gain 0db, shutter 50, HC-3 iris = 2, Glass: Canon 50mm 1.4 full open http://www.deepcode1.com/downloads/l..._50mm_F1_4.jpg In this test I have a light loss from F4 to F2, thats only 2 (!) stops light loss !!! My apologies to Quyen, there must be something going wron on the first test, I really cannot explain whaht. I repeat the first test later ..

June 12th, 2006, 12:15 PM	#8
Forrest Schultz Major Player Join Date: Mar 2005 Location: Atwater, CA Posts: 246	Holger, is it possible that you can post some footage sometime, i really like the look of the HC3 and want to see some more samples from it. thank you very much

June 13th, 2006, 03:45 PM	#9
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	Here some family & friends / bbq stuff grabs I did for testing. I hope the current job gives me time to do some additional footage from munich in the next weeks, here is a great summer festival and of course the soccer world cup with lots of VERY hot brazilian girls .. ;-) The Letus performs great under good light conditions, but I have the feel when the light dims the light loss increases rapidely. At daylight just ~2 stops, at evening it rises to 4-5 stops. It seems like the GG soaks a fix part of the incoming light. On some of the sunlight images I am not shure if It was the Vivitar 28-80 @2.8 or the Canon 50mm @ 1.4. At Daylight, the Vivitar Zoom is wonderful - it gives shallow DOF and lots of light, on evening you urgently need primes like the Canon 50mm 1.4, but it is VERY hard to get the focus on moving objects. http://www.deepcode1.com/downloads/l...q/bbq_mike.jpg http://www.deepcode1.com/downloads/l.../bbq_sarah.jpg http://www.deepcode1.com/downloads/l...q_things_1.jpg http://www.deepcode1.com/downloads/l...q_things_2.jpg http://www.deepcode1.com/downloads/l...q/bbq_vino.jpg http://www.deepcode1.com/downloads/l.../bbq_cigar.jpg http://www.deepcode1.com/downloads/l...q/bbq_dish.jpg http://www.deepcode1.com/downloads/l...bq/bbq_dog.jpg http://www.deepcode1.com/downloads/l...bbq_fire_1.jpg http://www.deepcode1.com/downloads/l...bq_flowers.jpg http://www.deepcode1.com/downloads/l.../bbq_melon.jpg

June 13th, 2006, 03:51 PM	#10
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	addy: viewed on the TV, all the images look much brighter. I did the framegrabs straight from VLC and saved them as jpg with 85% quality, no visible loss against PNG but much smaller.

June 13th, 2006, 07:12 PM	#11
Forrest Schultz Major Player Join Date: Mar 2005 Location: Atwater, CA Posts: 246	awesome! thank you very much

June 13th, 2006, 10:39 PM	#12
Dennis Wood Major Player Join Date: Mar 2005 Location: Ontario, Canada Posts: 916	Holger, you'll find that if you test a variety of scenes and light levels, your loss will vary.

June 14th, 2006, 12:50 AM	#14
Holger Leonhard Regular Crew Join Date: Oct 2005 Posts: 71	are any DVX / Letus flip users here to post some low light images with exact stop and db values ? would be interresting! to the HC-3: The cam is sensitive very close to the FX-1 we found, maybe the FX-1 is 1/2 stop or so better, but not much. Also you can boost the HC-3 to +18db without too much noise, but on full gain there is some loss of sharpness and loss of color.

June 14th, 2006, 10:52 AM	#15
Bob Hart Inner Circle Join Date: Dec 2003 Location: PERTH. W.A. AUSTRALIA. Posts: 4,476	An Australian DP did some tests on the Mini35 and there is a web page with his article on the subject. As I recall it, cumulative light loss through groundglass based relay systems at the tighter apertures is much less than at wide apertures. This might go some way to explaining inconsistent results from tests and measures metioned above.