Topics

locked Venice varying decode sensitivity

Geoff Boyle
 

I had an email earlier from Alister with a couple of stills attached, they looked very different and he asked me what Resolve was doing wrong as it gave different levels for the same image from the OCN and the EXR exported from Sony RAW.

Of course being Alsiter it was something either wrong with Resolve or that he was doing something wrong in Resolve.

He hadn't considered that Sony RAW is outputting the files in EXR from the Venice set to 500 but in a system that wants be set to 800, I'd have confirmed this but you can't alter the ISO when you're outputting EXR.

Anyway, I've checked again and again and the Sony RAW software outputs 2/3rds of a stop down.

You get the correct levels if you use the OCN in Resolve...

--
Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076

Argyris_Theos_cml
 

Geoff would you please elaborate?
Venice is supposed to be 500, right?
You shot at nominal ISO, ie 500?
Then Sony RAW Vewer treated it as 800?
Would this not mean the it should look 2/3rds Overexposed?
What am I missing?

Argyris Theos, gsc 
DoP, Athens Greece,
+306944725315
Skype Argyris.Theos
via iPhone

27 Μαρ 2018, 8:00 μ.μ., ο/η "Geoff Boyle" <geoff.cml@...> έγραψε:

I had an email earlier from Alister with a couple of stills attached, they looked very different and he asked me what Resolve was doing wrong as it gave different levels for the same image from the OCN and the EXR exported from Sony RAW.

Of course being Alsiter it was something either wrong with Resolve or that he was doing something wrong in Resolve.

He hadn't considered that Sony RAW is outputting the files in EXR from the Venice set to 500 but in a system that wants be set to 800, I'd have confirmed this but you can't alter the ISO when you're outputting EXR.

Anyway, I've checked again and again and the Sony RAW software outputs 2/3rds of a stop down.

You get the correct levels if you use the OCN in Resolve...

--
Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076

Geoff Boyle
 

Probably my bad explanation.

I wrote whilst I was angry, I'm fed up with people who admit they don't understand Resolve, ACES or EXR then making profound statements about things that are wrong.

Venice is 500 ISO, my metered tests show this.

If you take the OCN directly into Resolve it works fine.

If you output to EXR from Sony RAW it shows as 500 ISO but outputs files that are exactly 2/3rds of a stop darker than the OCN in Resolve.

If you set the ISO to 800 in Sony RAW the displayed levels increase by 2/3rds of a stop but you cannot output that as quite rightly the Sony software ignores your settings when outputting the EXRs in ACES AP0

It's not a big deal, a very small problem in new software/equipment but of course lets blow it up out of all proportion because this is the internet and we absolutely want to share our ignorance.

Cheers

Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076

On 27/03/2018 20:43, Argyris_Theos_cml wrote:
Geoff would you please elaborate?
Venice is supposed to be 500, right?
You shot at nominal ISO, ie 500?
Then Sony RAW Vewer treated it as 800?
Would this not mean the it should look 2/3rds Overexposed?
What am I missing?

Argyris Theos, gsc 

Riza Pacalioglu
 

…we absolutely want to share our ignorance.

Geoff Boyle NSC FBKS

The absolutely best phrase I heard, ever!

 

Riza Pacalioglu B.Sc. M.Sc. M.A.

Producer

South of England

 

alister@...
 

On 28 Mar 2018, at 06:32, Geoff Boyle <geoff.cml@...> wrote:

Venice is 500 ISO, my metered tests show this.

If you take the OCN directly into Resolve it works fine.

If you output to EXR from Sony RAW it shows as 500 ISO but outputs files that are exactly 2/3rds of a stop darker than the OCN in Resolve.

Does this mean your latitude test EXR’s are darker than they are supposed to be?

From what I understand Geoff has now looked at an EXR from Sony’s Raw Viewer and the X-OCN original in Resolve ACES and he see’s what appears to be a brightness/exposure difference between the two. Myself, I am seeing a chroma and hue difference. But, both of us are seeing a difference and that is not something I would expect. I’d really like to understand why, which is why I emailed Geoff to plead with him to do a test. If, following that test as Geoff states the EXR’s are being printed down (or metadata screwed up?) by Raw Viewer then I have to ask the question - does that mean that Geoff’s EXR’s are darker than they should be? I don’t actually think they are, but even so, why is there a difference.


You can download an X-OCN file from Venice and my EXR from here and try for yourself if you wish. https://my.pcloud.com/publink/show?code=XZYFs87ZIIwBsGa4kMzJVStT3iJ79z7kJaQX

My workflow is this:

I have a single X-OCN frame from Venice (500ISO). Using Sony’s Raw Viewer I export an ACES EXR of that frame. In Raw Viewer I deselect all input and ASC-CDL options and export using ACES Linear Bake. I believe this to be the correct process. Most of the options have no effect  on an ACES Bake anyway.

Then I start a fresh ACEScc project in Resolve with my input transform set to Sony Raw, output to Rec-709. I import both files. I’m using the “Sony” de-bayer option. What I see is very different hues and saturation. The Sony EXR file, being highly scientific - looks nice and as I remember the scene looking when I shot it, while the X-OCN file looks much less nice, having what to me visually appears to be a distinct pink/red color cast and is over saturated. This is not how I remember it looking when I shot it.

Looking at the scopes, the exposure of both is the same, middle grey and white are at very similar levels. But the vector scope tells a different story with some funky stuff going on with the X-OCN. I cannot easily correct this through a white balance shift and it doesn’t look like incorrect white balance. 

I use Resolve all the time, I love what it does, it’s a great tool for the money. I don’t normally use EXR’s, I normally go straight in with the original file. But having looked at the EXR’s I can see that I would much, much rather have the EXR image from Raw Viewer than the direct X-OCN image. But I don’t want to have to introduce another step into my workflow and expand the files size. My belief is I should not need to do this, my belief is that both files should look the same.

So I’d really like to understand what is going wrong without being shouted at and put down. I believe what I am doing is 100% correct. But my results say otherwise and Geoff’s own tests suggest something is not right.

I would really appreciate it if someone could take a look at these files in ACES in both Resolve and any other ACES workflow and tell me/us what they see? Is it just me or is there a problem. Geoff seems to think Raw Viewer is doing something not as expected or is not compatible with ACES.

I’m getting a little fed up with being called ignorant, arrogant or whatever, because I have identified an issue that it appears may have effected Geoffs test frames. If as Geoff has suggested Raw Viewer is printing down the EXR’s or screwing up the metadata in the EXR's then does this not suggest that Geoffs EXR’s are darker than they should be? Maybe this is a small problem, I’m NOT trying to blow it out of all proportion, but if this is the case everyone should be aware of it when viewing the files so they can make allowances. It appears that for some reason the playing field appears to be a little wonky, but no one wanted to believe me. 

Having looked at the output from Geoffs EXR’s, they look correct to me, which is why I am even more confused and would really like some clarity on where the exposure offset that Geoff see’s and the hue offset that I am seeing are coming from.

Teething problems occur in new releases, I am fully aware of that, so can we all get together and between us work out where the problem is so we can ask the right people to address it, even if it’s "Hey Alister your doing it all wrong", because right now I don’t believe I am doing it wrong, I believe the workflow doesn’t work as expected.

Alister Chapman

DoP - Stereographer
UK Mobile +44 7711 152226
US Mobile +1(216)298-1977


www.xdcam-user.com    1.5 million hits, 100,000 visits from over 45,000 unique visitors every month!  Film and Video production techniques, reviews and news.

















On 28 Mar 2018, at 06:32, Geoff Boyle <geoff.cml@...> wrote:

Probably my bad explanation.

I wrote whilst I was angry, I'm fed up with people who admit they don't understand Resolve, ACES or EXR then making profound statements about things that are wrong.

Venice is 500 ISO, my metered tests show this.

If you take the OCN directly into Resolve it works fine.

If you output to EXR from Sony RAW it shows as 500 ISO but outputs files that are exactly 2/3rds of a stop darker than the OCN in Resolve.

If you set the ISO to 800 in Sony RAW the displayed levels increase by 2/3rds of a stop but you cannot output that as quite rightly the Sony software ignores your settings when outputting the EXRs in ACES AP0

It's not a big deal, a very small problem in new software/equipment but of course lets blow it up out of all proportion because this is the internet and we absolutely want to share our ignorance.

Cheers

Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076

On 27/03/2018 20:43, Argyris_Theos_cml wrote:
Geoff would you please elaborate?
Venice is supposed to be 500, right?
You shot at nominal ISO, ie 500?
Then Sony RAW Vewer treated it as 800?
Would this not mean the it should look 2/3rds Overexposed?
What am I missing?

Argyris Theos, gsc 

Geoff Boyle
 

Alister, in you latest message you say you are very familiar with resolve but in previous messages you implied the opposite.

You like to tweak metadata, I like to just get on and use the manufacturers settings and grade later.

I do the tests, I publish the results, I explain how I've generated the material.

From then on it's up to the user.

I don't have time for your games. Stick to Facebook, they're ready for it there.

Cheers
Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076

alister@...
 

Oh come on Geoff, stop being an ass. I do know how to use ACES,  I do know my workflow isn’t broken and I’m not playing games. Open your eyes - you are seeing a problem to, are you just going to pretend it isn’t there? Because if it is a Raw Viewer print down, as you think, your test EXR’s are null and void. All that time wasted!

Until I pleaded with you to test X-OCN v EXR you hadn’t even realised there was a difference - we both know there shouldn’t be. But I guess desperately trying to point out there is a problem makes me ignorant. I’m just really disappointed that everyone was prepared to tell me to fuck off before anyone had even bothered to take a few minutes to actually look into what I was saying. Does everyone think I’m saying “the pictures look different - there’s a problem” for the sake of my ego or because - heck - there is actually something wrong.

We’ve already established that in previous versions of Resolve there were problems with the Sony IDT’s, so it is possible that there may be another problem now. But I’m open to other reasons for the cause of what I and others are seeing.

Who says you can’t use metadata if it helps get a better end result. You know as well as I do that there is no single “right” way to grade an image. Who says you can’t use LGG if that’s what gets you the image you desire. Every control does one thing, change the maths and maths is maths. Does it matter if when I am after an output of 4 whether I use 2+2 or 2x2? My choice, not yours.

You’ve finally looked at the files. There IS a problem and I don’t need to fiddle, tweak or adjust to make it happen. You know the files should be the same. Do you honestly think that Sony would be screwing up their own flagship camera footage in their own software? And this doesn’t just happen with Venice, it happens with any of the Sony cameras in raw or X-OCN at any ISO.

I’m collating the results from tests being done by other people. In Baselight the EXR’s and X-OCN look identical and the same as the EXR’s in Resolve. But I want to see more examples first to rule out any freak results. I’m trying to do this properly. And yes, once I have enough data to be sure of the result I will post it on facebook. If the result is due to an error on my part then I will say so. I’m not afraid to admit when I am wrong. Right now though this is looking highly unlikely.

Why not let others test, it only takes a couple of minutes. Lets see what they see and what they think without any name calling or other BS. Isn’t CML supposed to help people by discussing problems, and I really would like to get to the real root of this. But clearly no one will ever believe anything I say because I’m just too ignorant. So why not see what others think. Isn’t that a fair approach? Maybe I’m wrong… but maybe I’m not.

If as you are suggesting Raw Viewer is printing down 2/3rds of a stop then you really do need to add a note to your EXR’s to let everyone know they might not be right. Frankly though I don’t think there is anything wrong with the EXR’s and so far that’s the feedback I’m getting from those prepared to take a few minutes and look at this for themselves. 

If Resolve has an issue, shouldn’t we be asking BM to look into it? Or is that forbidden because they sponsor the list? If I get a confirmed Raw Viewer issue then I will be asking Sony to put it right.

I don’t want to fall out with you over this, but I am astounded by the number of people with their hands clamped firmly over their ears when someone points out a problem. You are seeing a problem, I am seeing a problem, and 4 other people that have responded to me are seeing the same problems, so lets just pretend there isn’t one so we can continue to call Alister ignorant shall we? Or shall we let others have a peek at the files for themselves and draw their own conclusions. You came to the conclusion that something was up, what do you think others are going to find? Is that a problem, it shouldn’t be.

Alister Chapman

DoP - Stereographer
UK Mobile +44 7711 152226
US Mobile +1(216)298-1977


www.xdcam-user.com    1.5 million hits, 100,000 visits from over 45,000 unique visitors every month!  Film and Video production techniques, reviews and news.

















On 28 Mar 2018, at 11:40, Geoff Boyle <geoff.cml@...> wrote:

Alister, in you latest message you say you are very familiar with resolve but in previous messages you implied the opposite.

You like to tweak metadata, I like to just get on and use the manufacturers settings and grade later.

I do the tests, I publish the results, I explain how I've generated the material.

From then on it's up to the user.

I don't have time for your games. Stick to Facebook, they're ready for it there.

Cheers
Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076

Nick Shaw
 

On 28 Mar 2018, at 12:41, alister@... wrote:

I do know how to use ACES,  I do know my workflow isn’t broken and I’m not playing games.

I get a 100% match between your ACES EXR and your X-OCN MXF in both Resolve and Baselight.

And I see no additional detail that I can dig out of the shadows of the X-OCN that I can't also access from the ACES EXR.

But your attitude means I have no inclination whatsoever to help you find out what you are doing wrong.

And now I've said all I plan to on this matter.

Nick Shaw
Workflow Consultant
Antler Post
Suite 87
30 Red Lion Street
Richmond
Surrey TW9 1RB
UK

+44 (0)7778 217 555

Geoff Boyle
 

Seconded.

Also before you criticize my negative tone try remebering that I get a lot of email offlist and messages on Facebook.

There's a limit...

Cheers
Geoff Boyle NSC
Cinematographer
EU based
+31 (0)637155076

On 28 Mar 2018, at 14:00, Nick Shaw <nick@...> wrote:
On 28 Mar 2018, at 12:41, alister@... wrote:

I do know how to use ACES,  I do know my workflow isn’t broken and I’m not playing games.

I get a 100% match between your ACES EXR and your X-OCN MXF in both Resolve and Baselight.

And I see no additional detail that I can dig out of the shadows of the X-OCN that I can't also access from the ACES EXR.

But your attitude means I have no inclination whatsoever to help you find out what you are doing wrong.

And now I've said all I plan to on this matter.

Nick Shaw
Workflow Consultant
Antler Post
Suite 87
30 Red Lion Street
Richmond
Surrey TW9 1RB
UK

+44 (0)7778 217 555

Peter Chamberlain
 

If there is a Resolve issue, we’d be happy to know.

PM me if preferred.


Peter Chamberlain 
DaVinci Resolve Product Manager




On 28 Mar 2018, at 7:41 pm, alister@... wrote:

Oh come on Geoff, stop being an ass. I do know how to use ACES,  I do know my workflow isn’t broken and I’m not playing games. Open your eyes - you are seeing a problem to, are you just going to pretend it isn’t there? Because if it is a Raw Viewer print down, as you think, your test EXR’s are null and void. All that time wasted!

Until I pleaded with you to test X-OCN v EXR you hadn’t even realised there was a difference - we both know there shouldn’t be. But I guess desperately trying to point out there is a problem makes me ignorant. I’m just really disappointed that everyone was prepared to tell me to fuck off before anyone had even bothered to take a few minutes to actually look into what I was saying. Does everyone think I’m saying “the pictures look different - there’s a problem” for the sake of my ego or because - heck - there is actually something wrong.

We’ve already established that in previous versions of Resolve there were problems with the Sony IDT’s, so it is possible that there may be another problem now. But I’m open to other reasons for the cause of what I and others are seeing.

Who says you can’t use metadata if it helps get a better end result. You know as well as I do that there is no single “right” way to grade an image. Who says you can’t use LGG if that’s what gets you the image you desire. Every control does one thing, change the maths and maths is maths. Does it matter if when I am after an output of 4 whether I use 2+2 or 2x2? My choice, not yours.

You’ve finally looked at the files. There IS a problem and I don’t need to fiddle, tweak or adjust to make it happen. You know the files should be the same. Do you honestly think that Sony would be screwing up their own flagship camera footage in their own software? And this doesn’t just happen with Venice, it happens with any of the Sony cameras in raw or X-OCN at any ISO.

I’m collating the results from tests being done by other people. In Baselight the EXR’s and X-OCN look identical and the same as the EXR’s in Resolve. But I want to see more examples first to rule out any freak results. I’m trying to do this properly. And yes, once I have enough data to be sure of the result I will post it on facebook. If the result is due to an error on my part then I will say so. I’m not afraid to admit when I am wrong. Right now though this is looking highly unlikely.

Why not let others test, it only takes a couple of minutes. Lets see what they see and what they think without any name calling or other BS. Isn’t CML supposed to help people by discussing problems, and I really would like to get to the real root of this. But clearly no one will ever believe anything I say because I’m just too ignorant. So why not see what others think. Isn’t that a fair approach? Maybe I’m wrong… but maybe I’m not.

If as you are suggesting Raw Viewer is printing down 2/3rds of a stop then you really do need to add a note to your EXR’s to let everyone know they might not be right. Frankly though I don’t think there is anything wrong with the EXR’s and so far that’s the feedback I’m getting from those prepared to take a few minutes and look at this for themselves. 

If Resolve has an issue, shouldn’t we be asking BM to look into it? Or is that forbidden because they sponsor the list? If I get a confirmed Raw Viewer issue then I will be asking Sony to put it right.

I don’t want to fall out with you over this, but I am astounded by the number of people with their hands clamped firmly over their ears when someone points out a problem. You are seeing a problem, I am seeing a problem, and 4 other people that have responded to me are seeing the same problems, so lets just pretend there isn’t one so we can continue to call Alister ignorant shall we? Or shall we let others have a peek at the files for themselves and draw their own conclusions. You came to the conclusion that something was up, what do you think others are going to find? Is that a problem, it shouldn’t be.

Alister Chapman

DoP - Stereographer
UK Mobile +44 7711 152226
US Mobile +1(216)298-1977


www.xdcam-user.com    1.5 million hits, 100,000 visits from over 45,000 unique visitors every month!  Film and Video production techniques, reviews and news.

















On 28 Mar 2018, at 11:40, Geoff Boyle <geoff.cml@...> wrote:

Alister, in you latest message you say you are very familiar with resolve but in previous messages you implied the opposite.

You like to tweak metadata, I like to just get on and use the manufacturers settings and grade later.

I do the tests, I publish the results, I explain how I've generated the material.

From then on it's up to the user.

I don't have time for your games. Stick to Facebook, they're ready for it there.

Cheers
Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076


alister@...
 

WTF. Come on guys are you even hearing yourselves. My attitude? All I asked for is to be listened to and to try and get some feedback from others, what the hell is so wrong with that? But, no, I get constantly shot down for trying to do this properly which is to have a large number of people test and then look at the results. So far from CML we have 3 tests and 3 sets of different results, so no consensus yet.

Geoff has stated - he is seeing an exposure shift, 2/3rds of a stop down. Why? He thinks Raw Viewer, in which case the EXR exposure files are incorrect as they will be 2/3rds of a stop too dark - but actually they don’t look too dark to me.

I’m seeing a color shift?  Why? What might cause a color shift in the workflow I have outlined in previous posts. I’m all ears.

Nick. Thank you for testing. I appreciate the input, I honestly do, although it leaves me more confused as from what’s been posted here, we now have Nick with no problem, Geoff with an exposure offset and me with a color shift. Off list I have 3 people with differences in Resolve and one with both files the same in Baselight.

I suspect that Geoff's EXR’s are OK, they look to be at the expected brightness and Nick's test seems to confirm this. So why is Geoff seeing an exposure offset between the EXR’s and X-OCN that Nick isn’t seeing, where is his offset coming from?

Anyone else want to let me know what you are seeing?

There is something odd going on somewhere, Geoff has a problem, so do I, but I am being berated for saying so.

It’s not a witch hunt or anti -anything campaign, I just want to figure out why 2 of us are getting differing results in systems that are supposed to produce identical results. I’m inclined to think Nick is doing something right that maybe Geoff and I are something doing wrong, or there is a bug or different software versions that are leading to variable conclusions. What is so wrong with that attitude?




Alister Chapman

DoP - Stereographer
UK Mobile +44 7711 152226
US Mobile +1(216)298-1977


www.xdcam-user.com    1.5 million hits, 100,000 visits from over 45,000 unique visitors every month!  Film and Video production techniques, reviews and news.

















On 28 Mar 2018, at 12:59, Peter Chamberlain <peter.chamberlain@...> wrote:

If there is a Resolve issue, we’d be happy to know.

PM me if preferred.


Peter Chamberlain 
DaVinci Resolve Product Manager




On 28 Mar 2018, at 7:41 pm, alister@... wrote:

Oh come on Geoff, stop being an ass. I do know how to use ACES,  I do know my workflow isn’t broken and I’m not playing games. Open your eyes - you are seeing a problem to, are you just going to pretend it isn’t there? Because if it is a Raw Viewer print down, as you think, your test EXR’s are null and void. All that time wasted!

Until I pleaded with you to test X-OCN v EXR you hadn’t even realised there was a difference - we both know there shouldn’t be. But I guess desperately trying to point out there is a problem makes me ignorant. I’m just really disappointed that everyone was prepared to tell me to fuck off before anyone had even bothered to take a few minutes to actually look into what I was saying. Does everyone think I’m saying “the pictures look different - there’s a problem” for the sake of my ego or because - heck - there is actually something wrong.

We’ve already established that in previous versions of Resolve there were problems with the Sony IDT’s, so it is possible that there may be another problem now. But I’m open to other reasons for the cause of what I and others are seeing.

Who says you can’t use metadata if it helps get a better end result. You know as well as I do that there is no single “right” way to grade an image. Who says you can’t use LGG if that’s what gets you the image you desire. Every control does one thing, change the maths and maths is maths. Does it matter if when I am after an output of 4 whether I use 2+2 or 2x2? My choice, not yours.

You’ve finally looked at the files. There IS a problem and I don’t need to fiddle, tweak or adjust to make it happen. You know the files should be the same. Do you honestly think that Sony would be screwing up their own flagship camera footage in their own software? And this doesn’t just happen with Venice, it happens with any of the Sony cameras in raw or X-OCN at any ISO.

I’m collating the results from tests being done by other people. In Baselight the EXR’s and X-OCN look identical and the same as the EXR’s in Resolve. But I want to see more examples first to rule out any freak results. I’m trying to do this properly. And yes, once I have enough data to be sure of the result I will post it on facebook. If the result is due to an error on my part then I will say so. I’m not afraid to admit when I am wrong. Right now though this is looking highly unlikely.

Why not let others test, it only takes a couple of minutes. Lets see what they see and what they think without any name calling or other BS. Isn’t CML supposed to help people by discussing problems, and I really would like to get to the real root of this. But clearly no one will ever believe anything I say because I’m just too ignorant. So why not see what others think. Isn’t that a fair approach? Maybe I’m wrong… but maybe I’m not.

If as you are suggesting Raw Viewer is printing down 2/3rds of a stop then you really do need to add a note to your EXR’s to let everyone know they might not be right. Frankly though I don’t think there is anything wrong with the EXR’s and so far that’s the feedback I’m getting from those prepared to take a few minutes and look at this for themselves. 

If Resolve has an issue, shouldn’t we be asking BM to look into it? Or is that forbidden because they sponsor the list? If I get a confirmed Raw Viewer issue then I will be asking Sony to put it right.

I don’t want to fall out with you over this, but I am astounded by the number of people with their hands clamped firmly over their ears when someone points out a problem. You are seeing a problem, I am seeing a problem, and 4 other people that have responded to me are seeing the same problems, so lets just pretend there isn’t one so we can continue to call Alister ignorant shall we? Or shall we let others have a peek at the files for themselves and draw their own conclusions. You came to the conclusion that something was up, what do you think others are going to find? Is that a problem, it shouldn’t be.

Alister Chapman

DoP - Stereographer
UK Mobile +44 7711 152226
US Mobile +1(216)298-1977


www.xdcam-user.com    1.5 million hits, 100,000 visits from over 45,000 unique visitors every month!  Film and Video production techniques, reviews and news.

















On 28 Mar 2018, at 11:40, Geoff Boyle <geoff.cml@...> wrote:

Alister, in you latest message you say you are very familiar with resolve but in previous messages you implied the opposite.

You like to tweak metadata, I like to just get on and use the manufacturers settings and grade later.

I do the tests, I publish the results, I explain how I've generated the material.

From then on it's up to the user.

I don't have time for your games. Stick to Facebook, they're ready for it there.

Cheers 
Geoff Boyle NSC FBKS 
Cinematographer 
Zoetermeer 
www.gboyle.co.uk 
+31 (0) 637 155 076


Nick Shaw
 

On 28 Mar 2018, at 13:29, alister@... wrote:

I suspect that Geoff's EXR’s are OK, they look to be at the expected brightness and Nick's test seems to confirm this. So why is Geoff seeing an exposure offset between the EXR’s and X-OCN that Nick isn’t seeing, where is his offset coming from?

I don't have Geoff's X-OCN to compare with the EXR, so can't comment there. The only matched pair of EXR and X-OCN I have is the one you posted.

And so as not to deny the help to others here, try using "No Input Transform" not "Sony RAW". The latter is a legacy IDT which is not needed for X-OCN (or any other Sony RAW format, AFAIK in Resolve 14.3). As has been said in this thread before, most raw decodes direct to ACES, and therefore needs no IDT. Applying one messes it up.

Nick Shaw
Workflow Consultant
Antler Post
Suite 87
30 Red Lion Street
Richmond
Surrey TW9 1RB
UK

+44 (0)7778 217 555

Art Adams
 

I'm not sure why people think Alistair has an attitude. To me, he's just asking questions. There are probably very few people on this list who know what is really going on, and others who think they do or have no idea. I don't think the former need to have disdain for the latter. There's no one place to learn all this stuff, and many of us don't have that kind of time anyway. 

I think we need to decide whether CML is a place to share knowledge and learn, or a place where we insult others for not knowing things that we do or for getting some very complicated issues wrong. If it's the latter, well, we already have the  YouTube comments section for that.

This kind of interaction is what keeps people from contributing to CML. If that's what CML is about -- a few people posting and thousands reading but not contributing for great of being publicly humiliated for asking questions about the complexities inherent in our industry  -- then we are definitely on track.

Maybe it's time to enable emojis so we stop projecting our own biases onto otherwise simple communications.

--
Art Adams
DP
San Francisco Bay Area




alister@...
 

On 28 Mar 2018, at 13:50, Nick Shaw <nick@...> wrote:


And so as not to deny the help to others here, try using "No Input Transform" not "Sony RAW". The latter is a legacy IDT which is not needed for X-OCN (or any other Sony RAW format, AFAIK in Resolve 14.3). As has been said in this thread before, most raw decodes direct to ACES, and therefore needs no IDT. Applying one messes it up.

Humble pie eating time.

My workflow was incorrect.

Nick was correct. Using no input transform instead of "Sony Raw” gives me a matching image, correct colours, no shift. Everything looks right. I’m feeling particularly foolish because this is something I should have tried for myself, but I didn’t think it would be wrong.

Hand in the air, I was doing things wrong. In the past I had to use Sony Raw as the input transform, but obviously that’s changed at some point and I wasn’t aware of the change. I’ve passed this information on to the others that tested for me, who were following my workflow verbatim and as a result copying my error. As soon as you get rid of the Sony Raw transform they too get a correct result. Not excusing myself, but I would not be surprised if others may be making the same mistake.

Now I’m curious as to why Geoff is seeing an offset because now I am not. But I don’t want to upset his day any more than I already have.

Thank you Nick.

I apologise to one and all. 

Apologies, apologies.


Alister Chapman

DoP - Stereographer
UK Mobile +44 7711 152226
US Mobile +1(216)298-1977


www.xdcam-user.com    1.5 million hits, 100,000 visits from over 45,000 unique visitors every month!  Film and Video production techniques, reviews and news.

















On 28 Mar 2018, at 13:50, Nick Shaw <nick@...> wrote:

On 28 Mar 2018, at 13:29, alister@... wrote:

I suspect that Geoff's EXR’s are OK, they look to be at the expected brightness and Nick's test seems to confirm this. So why is Geoff seeing an exposure offset between the EXR’s and X-OCN that Nick isn’t seeing, where is his offset coming from?

I don't have Geoff's X-OCN to compare with the EXR, so can't comment there. The only matched pair of EXR and X-OCN I have is the one you posted.

And so as not to deny the help to others here, try using "No Input Transform" not "Sony RAW". The latter is a legacy IDT which is not needed for X-OCN (or any other Sony RAW format, AFAIK in Resolve 14.3). As has been said in this thread before, most raw decodes direct to ACES, and therefore needs no IDT. Applying one messes it up.

Nick Shaw
Workflow Consultant
Antler Post
Suite 87
30 Red Lion Street
Richmond
Surrey TW9 1RB
UK

+44 (0)7778 217 555

Gavin Greenwalt
 

I dare say I am well versed in everything post, and have even written my own tools. And even I get bit frequently by user error. If 2+2=5 Allister is doing the right thing by trying to figure out why. Maybe it is user error but even user error can expose a tricky workflow bug that will bight users. I would say when dealing with camera processing it’s about 50/50 between bugs and workflow error. But even 50/50 means I’m helping identify and correct software errors. I’ve been on the other side of the fence as the lone support tech for a widely used 3D plugin, the whole job was just sorting between user errors and bugs. But the user error was also helpful because they would know what was poorly labeled or interpreted wrong or just badly designed. If there is a 2/3rds stop shift, that’s useful info. So now we have 3 different results Allister found a hue shift. Geoff found an exposure shift and Nick found zero shift. I wouldn’t say that’s not worth it examining.

Gavin Greenwalt
VFX Supervisor
Seattle, Wa

alister@...
 

Actually this whole discussions has answered many questions.

For some time I had been unhappy with the Raw and X-OCN in Resolve, and for some time I had been seeing differences between Raw Viewer and Resolve. I’ve just had a brief conversation with Richard from Blackmagic and he said that there had indeed a small problem but it had been fixed. I’m guessing that there were perhaps problems with the old “Sony Raw” transform and now that the transform is being done differently the problem has gone away. So I’m wondering just how long I’ve been looking at incorrect images, not being aware that I didn’t need to. I’m also talking to the people that told me the Resolve transform was not to Sony Specs. Perhaps they too are not aware that the problem has been fixed, because perhaps they have been making the same mistake as me.

So once again. Thank you very much Nick for taking time out of your day to look at my footage for free. I really appreciate that, as now I have a much nicer starting point for my grades. And I might need to go back and re-visit some of my recent projects too  (smiley face needed here). I owe you a beer… or two.

Alister Chapman

DoP - Stereographer
UK Mobile +44 7711 152226
US Mobile +1(216)298-1977


www.xdcam-user.com    1.5 million hits, 100,000 visits from over 45,000 unique visitors every month!  Film and Video production techniques, reviews and news.


Alex Metcalfe DoP
 

On 28/03/2018 15:00, Art Adams wrote:
I think we need to decide whether CML is a place to share knowledge
and learn, or a place where we insult others for not knowing things
that we do or for getting some very complicated issues wrong
Well said Art, I agree 100%. It's starting to feel like I'm back in the
playground.

--
Alex Metcalfe DoP
00 44 (0) 7785 557611
www.alexmetcalfedop.co.uk

Riza Pacalioglu
 

If there is a Resolve issue, we’d be happy to know.

 

Peter Chamberlain 

DaVinci Resolve Product Manager

 

I have a non-related issue with DV Resolve (surround audio monitoring). How can I get in touch with you?

 

Riza Pacalioglu B.Sc. M.Sc. M.A.

Producer

South of England

 

 

Nick Shaw
 

On 28 Mar 2018, at 15:55, alister@... wrote:

So once again. Thank you very much Nick for taking time out of your day to look at my footage for free. I really appreciate that, as now I have a much nicer starting point for my grades. And I might need to go back and re-visit some of my recent projects too  (smiley face needed here). I owe you a beer… or two.

All's well that ends well!

I probably overreacted, so I apologise too.

I certainly wouldn't claim to have never fallen foul of user error. And I know I've occasionally been too quick to say something was an issue, when in fact the error was mine.

Cheers

Nick Shaw
Workflow Consultant
Antler Post
Suite 87
30 Red Lion Street
Richmond
Surrey TW9 1RB
UK

+44 (0)7778 217 555

Noel Sterrett
 


On 03/28/2018 11:45 AM, Nick Shaw wrote:
All's well that ends well!

Those people who think they know everything are a great annoyance to those of us who do.  - Issac Asimof

Cheers.

--

Noel Sterrett Admit One Pictures info@...