Color Correcting the Dragon Box - 3 Part Spectacular

Discussion regarding the entirety of the franchise in a general (meta) sense, including such aspects as: production, trends, merchandise, fan culture, and more.

Moderators: Kanzenshuu Staff, General Help

HakkaiBills93
Beyond-the-Beyond Newbie
Posts: 472
Joined: Fri Jul 07, 2017 3:04 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by HakkaiBills93 » Sun Jun 30, 2019 6:59 am


Please educate yourself on what average delta E is, I'm getting of tire of your ignorance. And by the way, there is no color cast in my sailor moon match, every color was matched except the white patch being a little off.
Educate yourself on what is a color cast as if your white have a tint (like there is and unlike you i gave evidence of that) THEN you have a color cast (as the tint don't affect only the white) so no it's not a "superior" job
it's not cause you match your references that it make the result worth or better if the references is tinted all i point is that dbox aren't screenshot, and if you were truly doing that superior things we already have seen something worth but....NOT

How is it disrespectful when saying one is better than the other? Old software will get outclass by newer software, it's the same thing for the color correction world. 3 way color corrector was superseded by gamma-gain-offset color corrector; the white balancing method using curve or levels was superseded by temperature and tint adjustment; selective color was superseded by AB grid; what is the problem? Stop being ignorant and stupid.
because it's YOUR OPINION (as long as you don't give the evidence on the supposed superiority, it's an opinion, words you said worth absolutely zero, when i said something i gave example unlike you) you are litterraly shit on others way but you don't even show where they are inferior.
This sailor moon screenshot have tint so have a color cast (a color cast don't affect only the white color, i invite you to learn more about it)
(here a remind of how look your white colors) so affect more or less all colors
Image
you have a delta value that mean you match your references and? it suck and won't work for every frame, no lut can adjust all colors value, saturation etc for all frame, they used one frame reference and apply them to the whole, so like i said you'll get something like that as a result if you used season set or bright cels
Image

instead of keeping the details from the source
Image

the last time i talked about this serious issue that will affect bright and dark scene, you used retarded explaination as energy should be bright bla bla bla, yes but not to the point destroying lines etc, try it and adjusting contrast etc after the lut will never revert the process
if you want this kind of result OK it's your taste but calling it superior when you show a tinted result and thinking that this kind of overbrightness is logical is just stupid
after you said "using the level set ones" okay but then it will be too darker as they are crushed in black , there is no reliable full sources (not as much cels as needed) that can used this method

you talked about xerox code but there's no clue that the xerox code for the anime should be allways okay, animators made some errors etc, you can't say 100% a thing is superior when you didn't even have made your things supposed superior working perfectly lossless.

if you are able to show a result that won't take forever to settle , don't have a color cast and neither destroy this kind of bright or dark area and automatically did everything scene by scene then i'll reconsider it being superior way to do but as long as you don't do anything in that way, it's just another way, neither superior, neither inferior, it just look like "how to color match a reference when you litterally know nothing into color correction", as you can't say something will work perfectly before being done, you open your mouth loudly, take people from really high but litterally only use no act, only words, big mouth, small dick like i said , this kind of provocation is to see if it's really words only like i now know it is.

that's why i wanted the challenge, show me where it's superior or just shut up . Someone that can't even gave solid evidence about what they said should close their mouth


and the result you show is totally not showing anything superior than what can be reach with another software, yours also have his own limit that i let you figured quickly cause a whole episode is not a simple screenshot and if you really did it with a brain (not the kind of stupid argument like "what is blown is supposed to be blown" -_-"

about the white balance there's others way to do like by rgb adjustment too , all method have his own advantage and their weakness

You talk a lot but you litterraly show nothing that's what bother me, a simple screenshot don't show nothing really superior , if your goal is not to have a white balance properly done then forget to color match, your result will need a white balance that will affect others colors anyway
why i never did a color correction is first cause of time lacking but i know that even a color matching couldn't be lossless either ,the way it work can perfectly fit some sources with the right references (but at least you can get the same result with others software as long as you know how to use them) but it's the dragon box topic, and dragon box is seriously irregular in all way about colors, brightness, etc

so once again show evidence or clue, stop talking uselessly SHOW IT!!!!!! OR FUCK YOU!!!!!!!!!!!!!! show that you have guts and take the challenge, all people here show how pathetical and stupid you are
Stop making things up when you are clueless about color matching program, none of that even make sense.
SERIOUSLY!!!!!!!!!!!!? who show litterally no clue and no evidence? you show a pathetic screenshot with issue and you dare to called your shit "superior" they don't make sense cause you are too retarded to understand that's all, small dick with no guts

color grading by color matching consist into making your sources all colors turning into the same colors as the reference shot, all defined by a .lut created by matching programs that's why you need a reference shot AND a source shot, BUT in the dragon ball color correction thread (why i don't totally care about sailor moon it's out of topic as we talked about "color correcting the dragon box" and anyway yours don't even have a good white meaning that or your references or your way sucks (added to the judgement) )

The only "release" that you called "reliable" have overbrightness and crushed black, meaning that if you use it as reference , it will be also crushed black and overbright in the result if your references isn't properly white balanced like the season set, it won't be properly white balance in the result, it's pretty logical, and even by matching one image YOU SIMPLY CAN'T (at least until you show serious clues about i being wrong) said that the scene by scene matching won't blow out details in bright or dark area and have accurate white balanced in the result
You also have no clues that the xerox code applied to film footage and that all frames was having exactly the same colors, (animators made mistake etc)

User avatar
lansing
Beyond Newbie
Posts: 177
Joined: Mon Jan 06, 2014 1:49 am

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by lansing » Sun Jun 30, 2019 12:43 pm

crabshank1 wrote:
Sun Jun 30, 2019 4:49 am
lansing wrote:
Sat Jun 29, 2019 8:00 pm
The accuracy is crazy considering the heavy amount of artificial color cast.
This made me lol.
I have no idea what you're talking about

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Sun Jun 30, 2019 1:25 pm

lansing wrote:
Sun Jun 30, 2019 12:43 pm
crabshank1 wrote:
Sun Jun 30, 2019 4:49 am
lansing wrote:
Sat Jun 29, 2019 8:00 pm
The accuracy is crazy considering the heavy amount of artificial color cast.
This made me lol.
I have no idea what you're talking about
You complaining about colour casts after what you've said about white balance.

User avatar
lansing
Beyond Newbie
Posts: 177
Joined: Mon Jan 06, 2014 1:49 am

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by lansing » Sun Jun 30, 2019 1:53 pm

crabshank1 wrote:
Sun Jun 30, 2019 1:25 pm
lansing wrote:
Sun Jun 30, 2019 12:43 pm
crabshank1 wrote:
Sun Jun 30, 2019 4:49 am


This made me lol.
I have no idea what you're talking about
You complaining about colour casts after what you've said about white balance.
I don't get your point, what are you talking about?

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Sun Jun 30, 2019 1:54 pm

The power of crushing part 2:
Image

0.12 applied to 1280x960 image the downscaled to 640x480. Crushed on the right.

Sidenote: there was a mistake in the code for my grey dithering, new version's on Github https://github.com/crabshank/HLSL-Resha ... ither.hlsl and in my OneDrive version of AvsPmod https://1drv.ms/u/s!AjHPdv1_75reiQwQwkHob1qDTTi9.

User avatar
lansing
Beyond Newbie
Posts: 177
Joined: Mon Jan 06, 2014 1:49 am

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by lansing » Sun Jun 30, 2019 2:08 pm

HakkaiBills93 wrote:
Sun Jun 30, 2019 6:59 am

Please educate yourself on what average delta E is, I'm getting of tire of your ignorance. And by the way, there is no color cast in my sailor moon match, every color was matched except the white patch being a little off.
Educate yourself on what is a color cast as if your white have a tint (like there is and unlike you i gave evidence of that) THEN you have a color cast (as the tint don't affect only the white) so no it's not a "superior" job

...

This sailor moon screenshot have tint so have a color cast (a color cast don't affect only the white color, i invite you to learn more about it)
This is my reference image for the sailor moon shot
Image
The reference white patch doesn't have a color cast, and the tint on the resulting white patch is not affecting all color in my match, it was proven by the average delta E, so it is not a color cast. It simply means that the white patch is a little off in the match. This is like saying a match that hits all patches (including white) with red being a little off is having a color cast, this is pure ignorant.
HakkaiBills93 wrote:
Sun Jun 30, 2019 6:59 am
because it's YOUR OPINION (as long as you don't give the evidence on the supposed superiority, it's an opinion, words you said worth absolutely zero, when i said something i gave example unlike you) you are litterraly shit on others way but you don't even show where they are inferior.
What? Are you drunk? I had literally detailed the reasons here, it's the ease of usage (workflow) and the accuracy. You're not going to get those reference color with white balance, curve or rgb parade, selective color maybe but it's going to be time consuming with no means of measure of accuracy.
HakkaiBills93 wrote:
Sun Jun 30, 2019 6:59 am
the last time i talked about this serious issue that will affect bright and dark scene, you used retarded explaination as energy should be bright bla bla bla, yes but not to the point destroying lines etc, try it and adjusting contrast etc after the lut will never revert the process
if you want this kind of result OK it's your taste but calling it superior when you show a tinted result and thinking that this kind of overbrightness is logical is just stupid
you said "using the level set ones" it will be dark so , there is no reliable sources (and anyway there won't be as much needed) that can used this method

if you are able to show a result that won't take forever to settle , don't have a color cast and neither destroy this kind of bright or dark area and automatically did everything scene by scene then i'll reconsider but as long as you don't anything in that way, it's just another way, neither superior, neither inferior, it just look like "how to color match a reference when you litterally know nothing into color correction"

that's why i wanted the challenge, show me where it's superior or just shut up, it remind all yound jerk that open their big mouth but never do anything unless being 10 against one

show it or shut up


and the result you show is totally not showing anything superior than what can be reach with another software, that's a new way to do cc but in no case it's better, it have his own limit that i let you figured quickly cause a whole episode is not a simple screenshot
about white balance can be done by rgb adjustment too , all method have his own advantage and their own disadvantage
You talk a lot but you litterraly show nothing that's what bother me, a simple screenshot don't show nothing really superior , if your goal is not to have a white balance properly done then forget to color match, your result will need a white balance that will affect others colors anyway
Stop making things up when you are clueless about color matching program, none of that even make sense.
stop being arrogant and more idiot , it's YOU that said things but show 0 evidences about your things being "superior" if it's truly superior why didn't you take my challenge of cc? me using my way, you using your way?

you take people from high but seriously where are your clues, evidence, thoses that an inferior way can't reach?
show them , opening loud a big mouth but don't have the guts to show anything behind is why i called you small dick

or you show truly that your way of cc is superior or you stop talking in that way

color grading by color matching consist into making your sources all colors turning into the same colors as the reference shot, all defined by a .lut created by matching programs that's why you need a reference shot AND a source shot, BUT in the dragon ball color correction thread (why i don't totally care about sailor moon it's out of topic as we talked about "color correcting the dragon box" ) the only "release" that you called "reliable" have overbrightness and crushed black, meaning that if you use as reference a source overbrigth in the white and crushed black in the black, it will be also crushed black and overbright in the result if your references isn't properly white balanced like the season set, it won't be properly white balance in the result
What in the blue hell are you talking about? Let me guess, this whole childish arguments about overbrightening is you trying to be a smart ass putting up another straw man about me matching to the season set, when I had already stated a year ago that I'm matching to cel.

HakkaiBills93
Beyond-the-Beyond Newbie
Posts: 472
Joined: Fri Jul 07, 2017 3:04 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by HakkaiBills93 » Sun Jun 30, 2019 4:47 pm

This is my reference image for the sailor moon shot
Image
The reference white patch doesn't have a color cast, and the tint on the resulting white patch is not affecting all color in my match, it was proven by the average delta E, so it is not a color cast. It simply means that the white patch is a little off in the match. This is like saying a match that hits all patches (including white) with red being a little off is having a color cast, this is pure ignorant.
it can also be like that by other way, once you properly correct one scene in after effect, it will be the same for the whole scene having the same colors issues it would be more impressive if your lut was working for a whole episode without any trim or anything like that, like i said i didn't see anything that make this way really superior than any other ones

What? Are you drunk? I had literally detailed the reasons here, it's the ease of usage (workflow) and the accuracy. You're not going to get those reference color with white balance, curve or rgb parade, selective color maybe but it's going to be time consuming with no means of measure of accuracy.
of course i read it, i just explain that the ONLY FULL references you can have among the two you called reliable are the season set, you don't have and you'll never get as much cels needed) you didn't know what can be done by other way that's what i stated before , it's not as much time consumming than your way as an evidence if i was really 100% sure of one reliable references i would have done cc for the whole series as you didn't have even started doing it . at that time you can't even show anything about the supposed color palette cause you lack references for that.

With after effect, you can have into one preset, some filters, my preset allow me to do everything and the only thing i have to do for each part trimmed is the white balance , everything else is already automaticaly done. it's a big advantage against your method that can't be done in that situation where you lack what to create the palette.


What in the blue hell are you talking about? Let me guess, this whole childish arguments about overbrightening is you trying to be a smart ass putting up another straw man about me matching to the season set, when I had already stated a year ago that I'm matching to cel.
No it's a perfect example that's all, even if you got a nice cel with beautiful and accurate color , like this one

https://www.flickr.com/photos/65846913@ ... 014408632/

and match it to the dbox , it will blown out things, as you can see the white on this cel is really terribly white, but dbox white never should be white but shade of grey else it will destroy bright scenes what i point is not the matching capacity, it's the amount of requirement that can't make it superior to any other things, and matching this cel with color match tools. you'll probably have the right exact things in the dbox for this screenshot but as long as you don't proove me the opposite with a video (a screenshot is not worthy to check that) but in the same part of the episode i know one or two part where it will cause issue.

what pissed me off is only that you called it "superior" but nothing shown couldn't be done by other way or look exceptionnal
have you a references cel for each scenes, in each part for all episodes? NO , i know you are trying to create a matching palette using xerox code but what i said is that you didn't take the source footage in consideration, and nothing can say if in one frame gohan skin use one xerox code and into another one another one for lot of differrent reason, it's impossible to predict, or said it just because you stated it is, i have watch countless release having countless differrent colors, even the broadcast recording from kei don't have the same colors as the season sets (colors are lighter, not so dark etc) to do the right color correction, you need to know for each shot, the exact colors, exact value, exact saturation etc and that you can't said it

SO YES your approach is really interesting but as long as you didn't make it real finished and checked everything that is lossless then you simply CAN'T compare it to another way and said it's inferior simply cause as our old way use manual stuffs that only need knowledges and one references, yours require far more things that aren't easy to get. In one year, where is the palette? where is the automated stuffs? of course it can't be done yet as it's not so easy to do, need lot of things and you couldn't tried or do it, so you can't be sure of the outcome without having even do it in lot of episode
sailor moon maybe have reliable sources like i have for saint seiya so this can be really helpfull, it's nice, no one will say that tools sucks BUT you talk more than what you have done and here is a dragon ball color correction topic
nothing shown allow you to called it "superior" as long as you didn't encode all episodes with all presets, checked them without any issues etc...and proove that it was really lossless.

with our way unlike yours , with a references source, we can as much doing a white balance, alter hue or colors to match any references as long as we have a nice reference , selective colors or other plugin can do that even if you do'nt think so, i can show you when you want, and from every scene a color correction, i can create presets for selective colors and keep 98% of the time the same colors , it's not really a matter of time consumming, it's the footage issue and being 100% sure about true reliable references that make most of the people doing wrong things, if i used the season set colors as references no matter how i did it by matching tools or after effect or others it's possible, but i'll crush white and black and it sucks) your way need as much sources screenshot and as much references shot for each screenshot to do scene by scene color correction, a lut work only for one source, if your footage have differrent color cast, differrent colors alterations, your lut simply won't work on them as the source will be differrent and lut can't adapt to them

I don't know (or care) about sailor moon footage, try the one i know and that is in topic (after beeing so rude after someone that was doing something else than cc but on the dbox, you should be the last one doing off topic using sailor moon stuffs in a dragon ball topic, it's dragon ball ONLY here.

try to cc the web dll (not bluray one, only web dll) version of the dragon ball movie 1 whole opening (only) using 1 and only lut with reference that you want and then post the clip, i am pretty sure you'll get several issue for that as there is hard issue on this opening you simply can't for sure correct with your way

you can use this cel as reference
Image

if you want to impress me do it and show it (and i am not talking about a screenshot only that is so easy to trick, show the whole opening as a video, cause it's not on that screenshot that there will be the issue)
show your guts to give solid evidence that your things is superior, don't show to everyone that you are a big mouth coward that only can use word but litterally can't give evidence (as you are neither an expert or a professionnal or even someone with a solid reputation about it)

DO IT then we'll talk about it, color correct screenshot by color match won't impress anyone here

User avatar
lansing
Beyond Newbie
Posts: 177
Joined: Mon Jan 06, 2014 1:49 am

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by lansing » Sun Jun 30, 2019 9:28 pm

HakkaiBills93 wrote:
Sun Jun 30, 2019 4:47 pm
This is my reference image for the sailor moon shot
The reference white patch doesn't have a color cast, and the tint on the resulting white patch is not affecting all color in my match, it was proven by the average delta E, so it is not a color cast. It simply means that the white patch is a little off in the match. This is like saying a match that hits all patches (including white) with red being a little off is having a color cast, this is pure ignorant.
it can also be like that by other way, once you properly correct one scene in after effect, it will be the same for the whole scene having the same colors issues it would be more impressive if your lut was working for a whole episode without any trim or anything like that, like i said i didn't see anything that make this way really superior than any other ones
What are you talking about? That explanation was to prove that there is no color cast in my match where you meaninglessly trying so hard to attack on with your limited knowledge.
HakkaiBills93 wrote:
Sun Jun 30, 2019 4:47 pm
What? Are you drunk? I had literally detailed the reasons here, it's the ease of usage (workflow) and the accuracy. You're not going to get those reference color with white balance, curve or rgb parade, selective color maybe but it's going to be time consuming with no means of measure of accuracy.
of course i read it, i just explain that the ONLY FULL references you can have among the two you called reliable are the season set, you don't have and you'll never get as much cels needed) you didn't know what can be done by other way that's what i stated before , it's not as much time consumming than your way as an evidence if i was really 100% sure of one reliable references i would have done cc for the whole series as you didn't have even started doing it . at that time you can't even show anything about the supposed color palette cause you lack references for that.
Comparing differences between color correction methods has nothing to do with color references. Your logical thinking is a complete mess.
HakkaiBills93 wrote:
Sun Jun 30, 2019 4:47 pm
What in the blue hell are you talking about? Let me guess, this whole childish arguments about overbrightening is you trying to be a smart ass putting up another straw man about me matching to the season set, when I had already stated a year ago that I'm matching to cel.
No it's a perfect example that's all, even if you got a nice cel with beautiful and accurate color , like this one

https://www.flickr.com/photos/65846913@ ... 014408632/

and match it to the dbox , it will blown out things, as you can see the white on this cel is really terribly white, but dbox white never should be white but shade of grey else it will destroy bright scenes what i point is not the matching capacity, it's the amount of requirement that can't make it superior to any other things, and matching this cel with color match tools. you'll probably have the right exact things in the dbox for this screenshot but as long as you don't proove me the opposite with a video (a screenshot is not worthy to check that) but in the same part of the episode i know one or two part where it will cause issue.
Oh man you are really clueless on almost everything related to color correction. Have you ever scan anything with a scanner? There is no complete white or black in prints in the real world. The darkest representation of a print scan is about rgb(15,15,15) and the brightest around rgb(250, 250, 250), therefore it is impossible to blown up or crushed your match when you're matching to cel scan.

Your cel image right there is not a good reference as well because it was altered as the white hits 255, it was so obvious.
HakkaiBills93 wrote:
Sun Jun 30, 2019 4:47 pm
what pissed me off is only that you called it "superior" but nothing shown couldn't be done by other way or look exceptionnal
have you a references cel for each scenes, in each part for all episodes? NO , i know you are trying to create a matching palette using xerox code but what i said is that you didn't take the source footage in consideration, and nothing can say if in one frame gohan skin use one xerox code and into another one another one for lot of differrent reason
What is a xerox code, what are you even talking about?
HakkaiBills93 wrote:
Sun Jun 30, 2019 4:47 pm
it's impossible to predict, or said it just because you stated it is, i have watch countless release having countless differrent colors, even the broadcast recording from kei don't have the same colors as the season sets (colors are lighter, not so dark etc) to do the right color correction, you need to know for each shot, the exact colors, exact value, exact saturation etc and that you can't said it
This is the exact reason why I posted my finding about how color palette fits into their character model design a couple of months back, hello? And you are the first one to reply on that post, you are just pathetic.
HakkaiBills93 wrote:
Sun Jun 30, 2019 4:47 pm
with our way unlike yours , with a references source, we can as much doing a white balance, alter hue or colors to match any references as long as we have a nice reference , selective colors or other plugin can do that even if you do'nt think so, i can show you when you want,
You literally just tried it on the sailor moon sample and it was a disaster.
HakkaiBills93 wrote:
Sun Jun 30, 2019 4:47 pm
try to cc the web dll (not bluray one, only web dll) version of the dragon ball movie 1 whole opening (only) using 1 and only lut with reference that you want and then post the clip, i am pretty sure you'll get several issue for that as there is hard issue on this opening you simply can't for sure correct with your way

you can use this cel as reference
if you want to impress me do it and show it (and i am not talking about a screenshot only that is so easy to trick, show the whole opening as a video, cause it's not on that screenshot that there will be the issue)
show your guts to give solid evidence that your things is superior, don't show to everyone that you are a big mouth coward that only can use word but litterally can't give evidence (as you are neither an expert or a professionnal or even someone with a solid reputation about it)

DO IT then we'll talk about it, color correct screenshot by color match won't impress anyone here
What? Why would I do that? One lut is not going to work for the entire video and there is not enough reference. What is the logic?

User avatar
Ajay
Moderator
Posts: 6076
Joined: Sun May 26, 2013 6:15 pm
Location: Surrey, UK
Contact:

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by Ajay » Sun Jun 30, 2019 9:43 pm

Absolutely ridiculous behaviour in here.

You have three options:

1. Take a step back, cool off, and return to your debate in a civilised manner, entirely free of snide comments, insults, and general disregard for forum etiquette.

2. Stop this argument where it is, and move on.

3. Continue the way you are and receive account strikes and/or extended bans, as a result.

Up to you.
Follow me on Twitter! // Super Animation Catalogue
[url=https://twitter.com/Yuyucow][b]kViN[/b][/url][i] on Tadayoshi Yamamuro[/i] wrote:
A fun drawing killed his family and now his life's goal is to oppress them all.

HakkaiBills93
Beyond-the-Beyond Newbie
Posts: 472
Joined: Fri Jul 07, 2017 3:04 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by HakkaiBills93 » Mon Jul 01, 2019 6:18 am

Ajay wrote:
Sun Jun 30, 2019 9:43 pm
Absolutely ridiculous behaviour in here.

You have three options:

1. Take a step back, cool off, and return to your debate in a civilised manner, entirely free of snide comments, insults, and general disregard for forum etiquette.

2. Stop this argument where it is, and move on.

3. Continue the way you are and receive account strikes and/or extended bans, as a result.

Up to you.

Very hard to keep calm when someone is taking people from very high use only theories but can't even show any evidence as is "superior" way isn't even ready and dare to split on other way like an arrogant fool. I hate this kind of arrogant guys who think knowing everything but aren't even able to proove what they said. he talk arrogantly about a way that he didn't have settled yet, it's like if i was saying color correction with after effect is the best and all other sucks but i can't even produce the series with after effect and give enough evidence of all other limitation
What are you talking about? That explanation was to prove that there is no color cast in my match where you meaninglessly trying so hard to attack on with your limited knowledge.
working with screenshot and motion video is differrent things, i don't use technical words of course but what i learnt from practice is that theories are not everything in color correction, but i know far better than you issues from the dragon box footage since the time i worked on it and why this irregularity in dragon box footage will also be an issue for you (or at least show me i am wrong by other things than "theories" , lumetri color is supposed to give a true white balance in theories but in practice it don't work all the time for your example as it means having one exact referece for each alteration and that you can't tell as you can't even produce the db opening with what you have)
Comparing differences between color correction methods has nothing to do with color references. Your logical thinking is a complete mess.
But the thing i am saying is that you don't have color references, have you cels for the whole series with all characters with all outfits from all episodes? NO and like i said season sets can't be accurate no matter your carddass stories, kei which are like a dragon ball bible with all knowledge he have about db and if he said it's not accurate, then it's not accurate, you litterally only have carddass as explanation to say it's accurate, don't have any broadcast material, rare booklet, books from the first run(which was having accurate colors etc)
Oh man you are really clueless on almost everything related to color correction. Have you ever scan anything with a scanner? There is no complete white or black in prints in the real world. The darkest representation of a print scan is about rgb(15,15,15) and the brightest around rgb(250, 250, 250), therefore it is impossible to blown up or crushed your match when you're matching to cel scan.

Your cel image right there is not a good reference as well because it was altered as the white hits 255, it was so obvious.
Oh man, we talk about footage not screenshot, what i said is not the result on one frame but in motion on the others in a whole video
you'll match a screenshot perfectly and? if it was color matching photos then i would have totally agree with you as being the best quickier and lazier way to reach it but what i am allways saying and that you are totally unable to show as you deliberately ignore everything i ask you in the way to check if i am really wrong or not (as i said you can say everything you want as long as you don't show it in motion with evidence it's only speculation and nothing more)

the cels i gave you is for the opening from Ajay sooner in this topic lot of cels have this kind of brightness, and for the flickr account not from me either and it's also like that, meaning that you can't rely on all cels and this limit more your work

you can use any reference you want, just do the db movie 1 opening clip with your lut and show the result in video clip here, there is things that even color matching simply can't correct and that's what i allways pointed about it as you beeing too much arrogant about your supposed superior way but litterally produce zero evidence , zero episode, you litterally in one year did ZERO result then for something supposed to be better and quickier i just laugh a lot. what make me aggressive is the way you talk superior about something without having see if your things will work. it's too early to any kind of judgement as long as you didn't have made all required test, see all issues and solved them
What is a xerox code, what are you even talking about?
i said it wrongly i agree as the xerox code line don't have anything to do here, but you were showing character sheet from 2008 jump festa like if it can apply to an anime almost 30 years earlier, as you can and should have been understood, color in the series change regularly, there were animator colors goof like coloring error, you can't rely on official character sheet colors for a whole serie
more if you use it from several actual source like 2008 jump festa that use others colors than what was use decade ago, even in some movies color palette where differrent like in dbz movie 7-9 etc.. you simply don't have the reference and can't predict with that actual lack how it should look, all we can do actually is guessing crossing all release and see which things are the same as some cels etc, in no case it will be accurate 100% in that way BUT yours either, the references limitation is something that apply for all methods, if you don't have reliable reference , no matter if you use color match, after effect, it's the same.
This is the exact reason why I posted my finding about how color palette fits into their character model design a couple of months back, hello? And you are the first one to reply on that post, you are just pathetic.
i tried to help you cause i am curious to see it, and i am kind unlike you, all your participation are limited to criticized, no help, just said "that's wrong" "that's innacurate" "that's off topic" but don't help, and litterraly funny from someone that didn't even settle his own stuffs lacking essential stuffs to work... i don't said shit like you when i don't understand anything like you with crabshank, judging his stuffs, attacking him cause it's off topic (but you use sailor moon screenshot for a topic related to dragon ball XD ridiculous as you are more off topic than him as he can explain that he work on dbox footage but you it's sailor moon)

i am not the kind of guy that will said "it's shit" before seeing it done, i trust that color matching can help mostly when you lack knowledge or be lazy to do it otherway but because unlike you i am not the kind of guy that litteraly split on others work calling them "inferior" without at least show example, visual arguments to show where it's really wrong, i don't say anything, but when you produce zero stuffs yourself except few screenshot that you can't judge something that isn't even ready and called it superior as it's only speculation yet, in no case, you can know what issue you'll reach and where it is supposed to be superior. If you think you are the only one that have tried all the tools , you are wrong and i am not the only one either that have face the issue i talked about. dr dre tools have been used by several people far before you even know what it was.

i asked you to do db movie 1 opening only from web dll is cause there is some serious issue that i know color match tools won't help you more than any other way, i could have made the video clip myself to show it but you'll said i did it wrongly etc, so i want you to show your work on it, (as you use superior way, master everything better than everyone else etc, know litterally everything and don't see the need to give by video evidence , that you are true, people just have to trust the theories you said as it is with litteraly nothing worthy as an output
one screenshot is not one movie
You literally just tried it on the sailor moon sample and it was a disaster.
have you read what i said? i am not a sailor moon fan, i don't care about this stupid anime for girls idiotic to the death , i judge the white balance only as it's the only thing i can judge in something i can't know , i used my dragon ball preset with dragon ball colors alteration, it's just lazy to adapt on something i don't have any references (and don't care about) , in no way i said it was accurate, just properly white balance unlike yours and i show it. i don't have any cels, references or anything about sailor moon and i don't care it's a DRAGON BALL TOPIC stop acting like a retarded , it's easy to alter color, hue , actual software are very nice for that, i, yours have a slight shade in the white, the color picker show it to you (unless you are able and smart enough to read a color picker) you can't deny it
What? Why would I do that? One lut is not going to work for the entire video and there is not enough reference. What is the logic?
the entire opening exactly, YOU FINALLY HIT THE POINT on what i am saying about why you are arrogant and that it is really bad from yourself to called other things inferior when you are so limited to what you have and can't do anything if your reference have slight issue like this white one, you just can't do it so where is it truly superior? if you don't have anything to correct or can't use what you find once it have an issue where can it be superior?

When you have one solid reference, this tools can work but when you don't have...you can't produce anything, for example this cel i gave you with a white is really fine for after effect as you just have to white balance and match other colors, don't taking the brightness into consideration or just have to lower a little all color brightness , with your way it's not possible, if you use it, you blow output , color matching will match it no mater what with the issue and it can't be revert after as what is blown away is blown away taking db 109 toei overbright scan as a solid evidence that was what overbright can't be reverse fully

i allways show you, explain you that there is a serious limitation with your way as you need hard things to get as references and can't do anything without them, you can only work with unaltered references, but other software can also do by other way , as long as you trimmed it scene by scene it's the same, so no it's a differrent approach , not really better or only for people like you or limited with other softwre knowledge as long as you don't have as much references needed you can't do anything or not even a quickier way as anyway you'll have to solve stuffs scene by scene, there is no way you can automatically create lut that know what is your source and how should look the output, you can't know if it's even possible if you didn't did it.

it was allways my goal, not to show that this tools is bad , just that in no way it can be called "superior" more when you litterally are unable to do the series or even a small clip from the opening movie cause you lack reference and didn't even did your project in totality with the ability to say what was possible and what was not possible, you make a statement far before having test enough all limit of your way and that you can only see them by testing them on countless part of the dbox footage, in that way you have acted arrogantly and really like an idiot.

dragon box footage sucks, you have several colors issues sometimes from one frames to the next one (watch the yamcha vs invisible man after bulma show her boob you'll understand what i talked about, some episode start with one frame yellowish to the death but the next is better etc it's countless issues all along the series and movies, ep 35 is overdark since the second half etc), some frames don't have the same contrast, brightness etc, so you need at least as much references than any color alteration and you simply can't gathered as much references with no accurate release of the show, it's not like if all cels was availlable publicly and all well preserved, some are badly scanned.

that's all i wanted to say and fight hard with you

color matching is a great tools of course, your way is interesting ? of course but calling it "superior" far before your way being ready is FAR TOO EARLY and ARROGANT TO THE DEATH tried the db opening 1 from movie 1 web dll with the reference you want (Toei show another one with kamesennin not long time ago), you'll see at least 2-.3 scenes where it will give a wrong result ,

User avatar
lansing
Beyond Newbie
Posts: 177
Joined: Mon Jan 06, 2014 1:49 am

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by lansing » Mon Jul 01, 2019 12:00 pm

HakkaiBills93 wrote:
Mon Jul 01, 2019 6:18 am
What are you talking about? That explanation was to prove that there is no color cast in my match where you meaninglessly trying so hard to attack on with your limited knowledge.
working with screenshot and motion video is differrent things, i don't use technical words of course but what i learnt from practice is that theories are not everything in color correction, but i know far better than you issues from the dragon box footage since the time i worked on it and why this irregularity in dragon box footage will also be an issue for you (or at least show me i am wrong by other things than "theories" , lumetri color is supposed to give a true white balance in theories but in practice it don't work all the time for your example as it means having one exact referece for each alteration and that you can't tell as you can't even produce the db opening with what you have)

...

that's all i wanted to say and fight hard with you

color matching is a great tools of course, your way is interesting ? of course but calling it "superior" far before your way being ready is FAR TOO EARLY and ARROGANT TO THE DEATH tried the db opening 1 from movie 1 web dll with the reference you want (Toei show another one with kamesennin not long time ago), you'll see at least 2-.3 scenes where it will give a wrong result ,
Okay I think I'm done with this, there's no point to argue with someone who has no logical thinking and reasoning skill and who doesn't even know when his points were technically countered, and I'm getting tire of pointing out all your straw man arguments repeatedly while you keep on stacking them.

HakkaiBills93
Beyond-the-Beyond Newbie
Posts: 472
Joined: Fri Jul 07, 2017 3:04 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by HakkaiBills93 » Mon Jul 01, 2019 12:16 pm

Okay I think I'm done with this, there's no point to argue with someone who has no logical thinking and reasoning skill and who doesn't even know when his points were technically countered, and I'm getting tire of pointing out all your straw man arguments repeatedly while you keep on stacking them.
okay we'll see who are wrong and it will destroy your arguments

here your reference

Image

your reference is perfect about white look the shirt and the eyes

here is side by side your color match and the reference
Image

it's not tinted? are you sure? is it fine? it is that thing the "superior" way of cc? anyone can see that outfit aren't white it can't be called superior

little comparison between what you split on and yours
your "superior" way of cc on the left against "inferior" one on the right
Image

also check my black line and your black line, it affect all colors, the skin also look affected

and now, my result and your reference shot
Image
mine is far closer not perfect either but it's far closer than your result

you can use any theories you want, as long as your result won't fit your words, you are just humiliate yourself. Your attitude is far too arrogant and stupid and irrespectful, you are not able to make a color palette or lut database, you didn't settle anything yet, so your way is only a theorie in your mind, even with right theories you can failed like here only cause you didn't take all data into consideration, it make sense; you have nothing to work, so didn't test, barely on few screenshot and when you seems to have tried on sailor moon you couldn't even notice the issue only cause the differrence between the result and the source was so high that you probably deliberately ignore the issue.
And worst, you are so arrogant that without nothing really concret or having test the things enough, you take people from high with nothing in your hand, just learn modesty, there is allways better than you like there is better than me (as i act arrogant with you deliberately cause it's funny to see how pathetic thoses guys are when all they said are litterraly destroyed) , i know where my cc have weakness and i am as modest to say that it's not the best and keep respecting others work and try to help them, like i did with you where you only complain and show your way as "the savior of the color correction" (BIG FAIL) you can't win against me with so poor example and when you are wrong, theories worth ZERO and aren't everything .


PS: delta value i don't care forget it, only the result matter it's like saying that as all others colors are accurate, the green tint on broly bluray is not important, white balance is essential, and you are far than an average error rate , anyone can see it clearly
Image

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Mon Jul 01, 2019 11:54 pm

MAJOR UPDATE:

I've just added a white balance option (as the default) to my remapping page that calculates all the remapping options to white balance and lets you choose in an easy way:
Image

https://github.com/crabshank/HLSL-Resha ... ogram.html

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Tue Jul 02, 2019 2:32 am

crabshank1 wrote:
Mon Jul 01, 2019 11:54 pm
MAJOR UPDATE:

I've just added a white balance option (as the default) to my remapping page that calculates all the remapping options to white balance and lets you choose in an easy way:
Image

https://github.com/crabshank/HLSL-Resha ... ogram.html
I must say, make sure you are really sure that the elements of your white balance calibration image are supposed to be greyscale. E.g. I've had Piccolo's cape skew my data twice.

My code had a mistake where it was giving the mappings the wrong way round, I have now fixed this.

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Tue Jul 02, 2019 5:19 am

A whole episode white balance for DBZ E291 managed to do this (mine on left):
Image

and here are my remapping points to prove that I didn't cheat and just blow out the highlights:
Image

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Tue Jul 02, 2019 8:36 am

After testing I have concluded that for great quality, the upscale for my dithering should be at least 4x.

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Thu Jul 04, 2019 9:49 am

Just uploaded a tutorial on how to white balance using my code https://youtu.be/sYWT4F9ktDQ

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Thu Jul 11, 2019 12:12 am

Second attempt at Z E268. Managed to get a constant gi colour throughout the episode:
https://www.youtube.com/watch?v=bQsaro2A4EI

This is Dither (0.93 grey S-curve; just added S-curve dither check out my Github, crabshank is my name on there) + WB + Hue adjustments + Crushing (0.12).

crabshank1
Beyond Newbie
Posts: 123
Joined: Tue Nov 09, 2010 6:26 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by crabshank1 » Thu Jul 11, 2019 3:46 am

crabshank1 wrote:
Thu Jul 11, 2019 12:12 am
Second attempt at Z E268. Managed to get a constant gi colour throughout the episode:
https://www.youtube.com/watch?v=bQsaro2A4EI

This is Dither (0.93 grey S-curve; just added S-curve dither check out my Github, crabshank is my name on there) + WB + Hue adjustments + Crushing (0.12).
Some samples, mostly from the recap and preview (they look really cinematic), I'm very pleased with them :) (Left is mine, right is DBox):

Image

Image

Image

Image

Image

Image

Image

Image

Image

Image

Image

Image

Image

Image

Image

N.B. I think this new S-curve grey dither is why it looks like this.

HakkaiBills93
Beyond-the-Beyond Newbie
Posts: 472
Joined: Fri Jul 07, 2017 3:04 pm

Re: Color Correcting the Dragon Box - 3 Part Spectacular

Post by HakkaiBills93 » Thu Jul 11, 2019 7:52 am

you goes into the funimation direction, overbrightness, oversaturation or too much vibrant , even the color cast is stronger in the result

Post Reply