Topics

locked Raw feelings (was: Venice varying decode sensitivity)

Adam Wilt
 

I'm not sure why people think Alistair has an attitude. To me, he's just asking questions. 

I agree. I think Alistair had the misfortune to raise his hand in the middle of an “oh, Geoff, why didn’t you do the tests the way I wanted them done?” whinge-fest and so he got taken down by the same anti-aircraft guns trained on the complainers. “I’m sorry, Mrs. Chapman, your son was caught in the crossfire. It’s wartime; it happens. We present you, with sympathy, his sole earthly remains: a slightly singed SxS card.”

I find it interesting that Geoff and Alistair, two experienced people who spend all their time futzing with this sort of thing, are having difficulties getting the same result. I’m not confident enough of my ACES workflow skills to suss this one out, and I’m a neurotic tech geek—what hope is there for the regular cinematographer to be able to get her images safely and predictably through this pipeline? 

Yes, it’s just growing pains, as ACES is still relatively new and its integration into existing toolkits is still evolving, never mind IDTs and SDKs for brand-new cameras. Even so, it seems there’s a lack of clarity on how all the three- and four-letter lego bricks are supposed to be stacked to get the desired output. Discussions like this one may get fractious at times but they’re essential to reveal the ambiguous or ill-understood links in the workflow. Geoff and Nick may snort with impatience (understandably so) at the pathetic fumblings of the muggles, but we muggles are struggling with these same processes and often these discussions are the only way we can figure out what’s really going on.

Please, let’s turn the heat down and focus on the problem. Even if we in the peanut gallery never learn why Geoff and Alistair can’t get the same result, it’s still been a valuable exercise, for now we know that there be dragons out there, somewhere between the Venice and Resolve.

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

Emojis: combining the ambiguity of cryptic icons—“Is that an eggplant in your pocket or are you just happy to see me?”—with pre-teen aesthetic sensibilities. To misquote Dave Barry, they’re a quick and easy way to make any communication less attractive (and less communicative). I’ll put up with a fair amount of mudgobbing and deadcatting on CML, but if emojis start appearing, I’m outta here. Not that I have an opinion or anything.

Adam Wilt
technical services: consulting / coding / camerawork
Vancouver WA USA (no, not that Vancouver, the other one)

alister@...
 

I think I made some mistakes in the way I spoke to some people in the discussions and I apologise for that. But this is a great resource and Nick Shaw and others helped me get to the bottom of my problem.

So my hands are up - a problem with my workflow. But as some people have noted, sometimes these things are not always obvious and when you’ve been doing something one way because that was the way it was supposed to be done, if you don’t know it’s changed and the change doesn’t totally break the software, it’s difficult to even know that you are doing something wrong. Let alone know what it is you are doing wrong. If I hadn’t tested my own workflow by periodically comparing the EXR’s and X-OCN files I might not have even known that things were not right and I wonder how many people in the real world actually do those kind of checks. It makes me think of “If it aint broke, don’t try to fix it”. How many people might be using a broken workflow and not know it, because the images I was getting didn’t look wrong until you saw them side by side with a correct image, then you see the problems. I’ve suggested to Peter Chamberlain that the seemingly now redundant Sony Raw transfer option be removed altogether. 

There are so many variables in these workflows now that identifying problems like this can be tough. If I had the camera pointed at a chart and was looking at what I had been seeing (a slight red tint) on a monitor I might of just put it down to “that’s the way that camera looks”. Bringing up a test chart on the monitor wouldn’t have helped because the chart would most likely be some other file format not affected by this problem.

So can I propose a CML project to create color charts that are encoded into all the main raw and log formats that we use. It would make trouble shooting issues like this much easier and provide a way to perform periodic tests for workflow errors. I’m not sure how practical this is for the raw formats, perhaps we can get the manufacturers to help, but I think it’s something that could prove to be a valuable resource.


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.


Colin Elves
 

I’m not sure if this is the right list to say this (actually I’m pretty sure it’s not but since it’s come up here...) I have to say I feel like there’s been a noticeable shift in tone on CML since the introduction of the new system.

It’s become a lot more immediate and consequently a bit more ‘down the pub’ - 90% of the time I think this is a positive change. I’m really enjoying the new CML, I’m finding it a lot easier to engage with conversations, to follow threads and to learn more! All of which is great. And I am also enjoying what seems to be a much lighter tone in many posts.

But the other 10% of the time things have got noticeably ‘nasty’: The queen of cinematography thread, the PUA, this thread one or two others incidents (and yes, I know I’ve also been bad). 

Not sure what to do about this. Or if anything can really be done. I’m just saying it’s something I’ve noticed.

Colin Elves
Director of Photography
London/Berlin. 


On 28 Mar 2018, at 21:42, alister@... wrote:

I think I made some mistakes in the way I spoke to some people in the discussions and I apologise for that. But this is a great resource and Nick Shaw and others helped me get to the bottom of my problem.

Paul Curtis
 

On 28 Mar 2018, at 21:42, alister@... wrote:
So can I propose a CML project to create color charts that are encoded into all the main raw and log formats that we use. It would make trouble shooting issues like this much easier and provide a way to perform periodic tests for workflow errors. I’m not sure how practical this is for the raw formats, perhaps we can get the manufacturers to help, but I think it’s something that could prove to be a valuable resource.

That's a great idea. There are various macbeth charts generated electronically and if they could be turned into camera original formats that can be matched by real footage that would be really interesting (along with full range grads and resolution targets). But the challenge would be to generate RAW versions of them as if they came from the camera. That way people can sanity check their workflows with an artificial, yet perfect sample full range image.

This would also probably need the help of the manufacturers as well.

You also make a great point in that workflows are inherently complicated simply because they are joining together bits of software and tech from many companies. With a developer hat on i can read between the lines of what's going on usually but even so i've been caught out before, so what chance does someone who just needs to use the software have...

cheers
Paul

Paul Curtis, VFX & Post | Canterbury, UK

Geoff Boyle
 

OK, I've been away for 24 hours and had a great time, however...

Was I hostile? damn right I was!

This was something that spilled over into email and Facebook and people here are not aware of comments made elsewhere.

I've worked hard for more than 20 years to make sure that these evaluations are independent and objective. I don't appreciate people saying that sponsors have an influence because I've made it damn clear that they don't. I've done this to the extent of losing sponsors because they don't like what has been said about their kit.

Instead of making statements, not asking questions, that something is wrong, why not try quietly talking to people and trying to establish what exactly is happening.

We've discovered some major things in testing in the past that haven't become general knowledge because we talked quietly to the manufacturers and got it sorted. Probably the biggest of those was a new camera that worked perfectly at 24, 23.97 & 29.76 but had a sensitivity drop of over 3 stops at 25fps! No, you don't know which camera it was and I'm not telling you. It was sorted quietly.

Now I have managed to celebrate my wedding anniversary I will get on quietly trying to find out what is happening with my Venice material and talking to a lot of people, as I normally do, and publishing the results when I have them.

This will of course be delayed because of various pagan festivals occurring at the moment.

The new system is causing the issues I thought that it would, the only solution is much tighter moderation and that's a pain for everyone.

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

Leigh Hatwell
 

Hi all

Bloody well said!

+1

Cheers
Leigh
Workflows and more tech stuff
Paris

.

Le 28 mars 2018 à 9:15 PM, Adam Wilt <adam@...> a écrit :

I'm not sure why people think Alistair has an attitude. To me, he's just asking questions. 

I agree. I think Alistair had the misfortune to raise his hand in the middle of an “oh, Geoff, why didn’t you do the tests the way I wanted them done?” whinge-fest and so he got taken down by the same anti-aircraft guns trained on the complainers. “I’m sorry, Mrs. Chapman, your son was caught in the crossfire. It’s wartime; it happens. We present you, with sympathy, his sole earthly remains: a slightly singed SxS card.”

I find it interesting that Geoff and Alistair, two experienced people who spend all their time futzing with this sort of thing, are having difficulties getting the same result. I’m not confident enough of my ACES workflow skills to suss this one out, and I’m a neurotic tech geek—what hope is there for the regular cinematographer to be able to get her images safely and predictably through this pipeline? 

Yes, it’s just growing pains, as ACES is still relatively new and its integration into existing toolkits is still evolving, never mind IDTs and SDKs for brand-new cameras. Even so, it seems there’s a lack of clarity on how all the three- and four-letter lego bricks are supposed to be stacked to get the desired output. Discussions like this one may get fractious at times but they’re essential to reveal the ambiguous or ill-understood links in the workflow. Geoff and Nick may snort with impatience (understandably so) at the pathetic fumblings of the muggles, but we muggles are struggling with these same processes and often these discussions are the only way we can figure out what’s really going on.

Please, let’s turn the heat down and focus on the problem. Even if we in the peanut gallery never learn why Geoff and Alistair can’t get the same result, it’s still been a valuable exercise, for now we know that there be dragons out there, somewhere between the Venice and Resolve.

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

Emojis: combining the ambiguity of cryptic icons—“Is that an eggplant in your pocket or are you just happy to see me?”—with pre-teen aesthetic sensibilities. To misquote Dave Barry, they’re a quick and easy way to make any communication less attractive (and less communicative). I’ll put up with a fair amount of mudgobbing and deadcatting on CML, but if emojis start appearing, I’m outta here. Not that I have an opinion or anything.

Adam Wilt
technical services: consulting / coding / camerawork
Vancouver WA USA (no, not that Vancouver, the other one)

Colin Elves
 

Well. I don’t know about others but I consider the CML tests to be the gold standard - everything else seems to be marketing of one form or another. So please do keep it up!

In terms of the new system: do we really need tighter moderation? We’re all adults here, right? I’m not sure you fine people need to be doing more than you already do.

I think the most effective moderation was Jessica (I think) pointing out that these messages go to 4,000+ professionals in our industry - so don’t say anything you’d regret your peers hearing.

Colin Elves
Director of Photography
Berlin/London.


On 29 Mar 2018, at 11:55, Geoff Boyle <geoff.cml@...> wrote:
The new system is causing the issues I thought that it would, the only solution is much tighter moderation and that's a pain for everyone.
Cheers
Geoff Boyle NSC FBKS
Cinematographer
Zoetermeer
www.gboyle.co.uk
+31 (0) 637 155 076

alister@...
 


Just maybe, if instead of immediately taking an aggressive, defensive stance, someone had read what I was saying, half of the arguments would not have happened. I stand by my assertion that the Sony Raw transform that is in the list of ACES transforms is not to Sony specs and if anyone takes the time to try it, they will see that it tints your footage. Yesterday Blackmagic’s Peter Chamberlain confirmed with me that there had been problems with it and he should know. I certainly wasn’t saying it wasn’t right as a put down on Resolve, I couldn’t do my job without Resolve and I think its an excellent product. 

But now…. as I have finally found out, you are not supposed to use it in version 14.3 of Resolve, because if you do it messes up the footage. You don’t use the listed Sony Raw input transform, you must use No Input Transform - which in my mind goes against my understand of how ACES works - Camera Specific IDT, lots of colour science stuff in the middle, RRT, ODT (or combined output transform) - nice pictures come out at the end. Of course I now understand that Resolve is automatically setting up the transform, but I think most people can see why someone might make the very same mistake.

But you know - no one was listening. I described my workflow step by step very early on in the discussion, stating that I was using the Sony Transform and getting a defective result, but all I got was “impossible, you're ignorant, you don’t know what you are doing” None of that is helpful, none of that gets to the bottom of the problem. Once someone actually looked at what I was doing (thanks again Nick) and read through my workflow the solution came instantly, but I practically had to beg people to take a look. It shouldn’t be like that.

I too had a very interesting email inbox over the last few days too, a lot of people with exactly the same problem as me, that Resolve and Raw Viewer were producing differing results for them too. Now I know why and I can pass that on to them. But my god, it was so, so much harder to get to the bottom of why than it should have been. It could have been done and dusted within minutes if it had been discussed rather than all the dismissive, aggressive put-downs.

I’m no angel here, I didn’t help myself at times, frustrated perhaps because no one bothered to ask - why are you saying what you are saying? These workflows are complex and as it is plain to see there are things that can trip people up. I bet everyone has experienced something like this at some point, which is why places like CML are important. But…...

There isn’t much point in a discussion list if you can’t discuss or dare not put your head over the parapet in fear of being shot at. 

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 29 Mar 2018, at 11:55, Geoff Boyle <geoff.cml@...> wrote:

OK, I've been away for 24 hours and had a great time, however...

Was I hostile? damn right I was!

This was something that spilled over into email and Facebook and people here are not aware of comments made elsewhere.

I've worked hard for more than 20 years to make sure that these evaluations are independent and objective. I don't appreciate people saying that sponsors have an influence because I've made it damn clear that they don't. I've done this to the extent of losing sponsors because they don't like what has been said about their kit.

Instead of making statements, not asking questions, that something is wrong, why not try quietly talking to people and trying to establish what exactly is happening.

We've discovered some major things in testing in the past that haven't become general knowledge because we talked quietly to the manufacturers and got it sorted. Probably the biggest of those was a new camera that worked perfectly at 24, 23.97 & 29.76 but had a sensitivity drop of over 3 stops at 25fps! No, you don't know which camera it was and I'm not telling you. It was sorted quietly.

Now I have managed to celebrate my wedding anniversary I will get on quietly trying to find out what is happening with my Venice material and talking to a lot of people, as I normally do, and publishing the results when I have them.

This will of course be delayed because of various pagan festivals occurring at the moment.

The new system is causing the issues I thought that it would, the only solution is much tighter moderation and that's a pain for everyone.

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

Geoff Boyle
 

Very early on in this event I told you that you didn't need an IDT in Resolve for Sony material, you chose to ignore that.

I explained that Resolve used the Sony SDK, you queried that. Denied it actually, or maybe it was an out of date one.

But Alister, I'm sure you'll reply to this with further distortions.

Go ahead, you can have the last word.

I don't have time to waste on this and I really don't want CML to descend into a slagging fest, which it hasn't done very often.

I set this up to exchange information, I want it to work in a civilized way when possible but when people keep insisting on ignoring information and distorting events, spreading it across several different system, I have to respond.

I don't know why you have to do this, you've done it on Facebook and now you're doing it here.

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

alister@...
 

Perhaps one significant take-away from the discussions is that it does actually show that Geoffs approach to supplying EXR’s so that everything is in the same format is absolutely valid.

If you have a broken viewing workflow, at least all the files will be broken by the same amount when you view them. 

I’d still like to have access to the original files as transforms and other factors change over time (winking smiley face). 

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.