Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

DR ..a good discussion!!

Really telling in the forehead:


attachment.php


attachment.php
 

Attachments

  • Screen Shot 2018-04-15 at 7.45.07 PM.jpg
    Screen Shot 2018-04-15 at 7.45.07 PM.jpg
    8.6 KB · Views: 0
  • Screen Shot 2018-04-15 at 7.44.55 PM.jpg
    Screen Shot 2018-04-15 at 7.44.55 PM.jpg
    7.7 KB · Views: 0
Is Helium going to be added to the list soon? I've played with some r3d's from it with Xyla charts in the past but would love to see a new test too!
 
Bjorn,

You'll have to ask luigi. I believe he wanted to test recovering highlights between the cameras. Have you seen the original clips he made available for testing?
 
Also,

That was a Gemini vs Alexa and not a Monstro vs Alexa as Luigi pointed out to me. The pulled down highlights were part of my first attempt to try to bring back some of the highlights on the cars. I probably just left them there and forgot with the further attempt.


Tiff of original Alexa .ari out of Resolve.


arri_1.1.32.jpg



Tiff of original Gemini .r3d out of Redcine-X

A005_C006_04147_B.0000000_F.0000000.jpg
 
Last edited:
Björn,

Example is Gemini VS ALEXA NOT MONSTRO.
Rand, please placed the 3 raw frame , ALexa, Gemini and Monstro;) thank you.

HEre RD3 Gemini + ALexa :
https://www.dropbox.com/s/nirg531k1u8tkuy/highlight.zip?dl=0

Here R3D Monstro:
https://www.dropbox.com/s/2p6ehjtbfar74lk/A003_C006_0414ON.0000000F.R3D?dl=0

SAME APERTURE and SAME LENS.


Björn Monstro has about a difference of 1 stop in the highlights compared to ALEXA but better than 1 stop in the shadows as well as a better noise, but maybe you can answer me a question.
Is it possible on larger sensor the light hit on the whole sensor size is higher and therefore Monstro explode first?
Since if I have to have the same depth of field of s35 I have to close almost 1 stop.
Or am I saying a bullshit?

I hope you understand from my bad English.
 
Last edited:
luigi,

I'm getting a "404" error message for both links.
 
Luigi,

these are by no means the definitive looks of the cameras but just my grade of them.


Alexa


Alexa_brought_down_graded.jpg




Gemini


gemini_grade_less_blue.jpg



Monstro


monstro_grade_3.jpg
 
"Same Depth of Field'

Monstro

same_depth_2_monstro2.jpg



Now with


Alexa


Alexa_brought_down_graded.jpg




Gemini


gemini_grade_less_blue.jpg
 
Last edited:
Patrick Tresch said:
I don't think it's the issue. Exposure shouldn't be determined by sensor or film size.

BUT, IMHO, it's just that the base iso to have the same DR in the highlights from mid gray and + or - the "same" noise floor is 800 on Alexa and 1600+ on Monstro.

Pat
Click to expand...

PAt, but 'if we want to get the same depth of field of a s35 we have to close the diaphragm. So if I have to make a comparison with ALexa or Dragon 5k, I should have the same type of image or as similar as possible.
So at T5.6 on Monstro on Alexa or dragon 5k the diaphragm is more open.
This would lead to having similar high lights or better to say the total DR would be equal from medium gray up.
Or am I wrong?
 
luigivaltulini said:
PAt, but 'if we want to get the same depth of field of a s35 we have to close the diaphragm. So if I have to make a comparison with ALexa or Dragon 5k, I should have the same type of image or as similar as possible.
So at T5.6 on Monstro on Alexa or dragon 5k the diaphragm is more open.
This would lead to having similar high lights or better to say the total DR would be equal from medium gray up.
Or am I wrong?
Click to expand...

Ok so if you close the iris to get the same depth of field, do you compensate the exposure by changing the shutter speed or ND filter?
 
Patrick Tresch said:
Ok so if you close the iris to get the same depth of field, do you compensate the exposure by changing the shutter speed or ND filter?
Click to expand...

ND filter...obviously. Unless you are shooting charts that don't move from a tripod that doesn't move. Then shutter speed is irrelevant.
 
I feel like when you’re comparing cameras, giving them different exposures makes the images much more different and introduces more variables than having different DoF.
 
Patrick Tresch said:
Ok so if you close the iris to get the same depth of field, do you compensate the exposure by changing the shutter speed or ND filter?
Click to expand...

I put ND to ALexa.
By doing this I noticed the same depth of field in the image with full frame Monstro.
This deduction led me to say that if we have and want the same depth of field on ALexa s35 and Monstro full frame VV the Highlights are practically at the same breaking point.
As you say, I could at this point raise the ISO to 1600 and close the diaphragm of 1 stop on Monstro or put an ND on ALEXA and open 1 stop . But still Monstro has less noise and more resolution.
I hope I made myself understand eheheh
 
luigivaltulini said:
I put ND to ALexa.
By doing this I noticed the same depth of field in the image with full frame Monstro.
This deduction led me to say that if we have and want the same depth of field on ALexa s35 and Monstro full frame VV the Highlights are practically at the same breaking point.
As you say, I could at this point raise the ISO to 1600 and close the diaphragm of 1 stop on Monstro or put an ND on ALEXA and open 1 stop . But still Monstro has less noise and more resolution.
I hope I made myself understand eheheh
Click to expand...

When making an objective measurement (of any kind), it is helpful to keep that measurement in mind. A general principle of the Scientific Method is to eliminate as many extraneous variables as possible so that the effect of only the most important (hopefully one) is most visible. If you want to measure sensitivity, or noise, or resolution, or contrast, or depth of field, or perspective, or color accuracy, or highlight clipping, or size, or weight, or power consumption, or battery life, or resale value, or ... it is best to set up a test that really shows that makes everything else equal so that the single variable can be evaluated. The more variables you have to look at AT THE SAME TIME, the more you venture from quantitative fact to qualitative narrative.

If it's too distracting to see very different DOF between two imaging systems, using an ND filter to eliminate the DOF variable is a good idea. If the ND itself doesn't disrupt the test in other ways. So first you have to prove that the ND changes exposure but not any other "important" criteria, and once that's done, you can use it to eliminate the DOF difference. If the ND introduces a color shift (either due to IR pollution or some other reason), then it's not necessarily helping your test.

Many years ago some newbies posted a test comparing RED with an EOS 5d2 and found no real differences in image quality. What they really demonstrated was that if you ignore enough variables and then overlook any results that contradict your initial narrative, you can claim anything. A good test is clear in what it's trying to measure (not "I just threw on some lenses to see what would happen"), clear in its methodology, transparent with its data (not "the R3D files are too big to post, but trust me when I say that these thumbnails prove you cannot see a difference"), and faithful in its summary findings (not "the chromatic aberration you see in these tests never bears out in real-world scenarios").
 
Michael Tiemann said:
When making an objective measurement (of any kind), it is helpful to keep that measurement in mind. A general principle of the Scientific Method is to eliminate as many extraneous variables as possible so that the effect of only the most important (hopefully one) is most visible. If you want to measure sensitivity, or noise, or resolution, or contrast, or depth of field, or perspective, or color accuracy, or highlight clipping, or size, or weight, or power consumption, or battery life, or resale value, or ... it is best to set up a test that really shows that makes everything else equal so that the single variable can be evaluated. The more variables you have to look at AT THE SAME TIME, the more you venture from quantitative fact to qualitative narrative.

If it's too distracting to see very different DOF between two imaging systems, using an ND filter to eliminate the DOF variable is a good idea. If the ND itself doesn't disrupt the test in other ways. So first you have to prove that the ND changes exposure but not any other "important" criteria, and once that's done, you can use it to eliminate the DOF difference. If the ND introduces a color shift (either due to IR pollution or some other reason), then it's not necessarily helping your test.

Many years ago some newbies posted a test comparing RED with an EOS 5d2 and found no real differences in image quality. What they really demonstrated was that if you ignore enough variables and then overlook any results that contradict your initial narrative, you can claim anything. A good test is clear in what it's trying to measure (not "I just threw on some lenses to see what would happen"), clear in its methodology, transparent with its data (not "the R3D files are too big to post, but trust me when I say that these thumbnails prove you cannot see a difference"), and faithful in its summary findings (not "the chromatic aberration you see in these tests never bears out in real-world scenarios").
Click to expand...

thanks Michael, you're always precious with your comments;) love this.
 
You must log in or register to reply here.
Back
Top