Chaos Corona Forum
Chaos Corona for 3ds Max => [Max] Bug Reporting => Topic started by: JG_monomiru on 2024-11-22, 15:53:25
-
Recently i installed Corona 12. update 1 and it doesnt render specific light objects (various Polygonobjects with corona light material) in each of their own Light-Select Renderelement - just the main ForrestPack_Object containing all light objects when added to a light-select element.
Before corona 12.update 1 it was behaving that way
FP-object containing multiple Lightobject A and Lightobject B
in Corona 12 hotfix 1 and lower
Lightobject_A --> Lightselect_element A = light only A
Lightobject_B --> Lightselect_element B = light only B
FP-object --> Lightselect_element_FP = dark no light
in Corona 12 update 1
Now those individual light-objects get ignored and it only renderes the FP_object
Lightobject_A --> Lightselect_element A = dark no light
Lightobject_B --> Lightselect_element B = dark no light
FP-object --> Lightselect_element_FP = shows light of A and B combined
This is actually something that makes postproduction very complicated if i cant separate light-sources in Light-Selct / Lightmix anymore... and need for each light-object an individual FP-Object to be created.
Not sure if thats something the new corona verison messed up - or if its on behalf of updated Forrest.Pack
max 2025 / corona 12 update 1 / ForrestPro 9.1.5
ps. It does work with max 2024 / corona 12 hotfix 1 / ForrestPro 9.1.5
So i doubt its caused by forrestpack.
-
same problem here, forest with Lightobject stop to render after corona 12 update 1
max 2021 / corona 12 update 1 / ForrestPro 9.1.5
-
I use this workflow at times, but my current projects don't require it at the moment.
I'll stay tuned to see what the result is. I would expect that everything that worked in 12H1 should work on 12U1.
-
As did I... it was a surprise when I had an older Project rerendered and didn't paid attention to the corona-version i used. And then realized all my Light-Select Renderelements were "empty" - not showing specific objects light.
But regarding the Topics about Forrest Pack and C-12U1 there seem to be more issues between those plugins.
-
Hi,
I was able to reproduce this. I have logged it in our system to be further reviewed by our devs. We will let you know once we have any update on this.
(Report ID=CMAX-1419)
-
This behavior may or may not be correct or wanted?
If LightSelect element has a forest object selected, I would expect that all the forest objects are in this element - this is the case for Corona 12 Update 1.
While if a only a single light is added to LightSelect element, I would expect only that light to be in the lightselect, not also all its instances in Forest object.
If this was somewhat useful, this - light source and all its instances in Forest object in the same lightSelect element - can be brought back.
But selecting forest object as lightselect element should definitely have the complete Forest object in it.
Let me know what you guys think.
-
Well the desired Workflow which has been usable since dawn of corona was that you can have obj/Ids for each object and the Forrest-Object is not represented as an individual Objeckt_Ids in render mask elements.
That way it was resonable and good, that you have the same "logic" with lights and light-select-render-elements. So you don't have a combined Forrest-Pack light select, just the individual objects on their own.
It would be great if there could be a Forrest-Pack_- Lightselect itself... but for my workflow i would prefer to have the used light-items as seperate light-select by itself.
Otherwise i would need to have seperate Forrest-Pack objects for different types of Light-distribution.
Usually I use them for exterior rendering, where you need lots of lights in various parts of the building and various appearance... like linear light for floor spaces etc., suspended lights for office spaces, maybe round suspended lights for lobby or "meeting" zones... and maybe spot lights or floor lights for conference spaces or else.
So if it could work like it was working all the years since light-mix was a thing, that would be great - also it would make it much more practical to work on older projects. Otherwise we need to start saving older corona-render versions for each older project ... or even worth start over with setting up render elements and also later reconfiguring post-set-ups.
-
I know Object-Ids and Material_Ids behaved differently in Corona and Vray with Forrest-Pack...
when i worked with Vray before corona-render, i was a bit surprised i didnt had all the forrest-pack object as just one ID... but then i really discovered the potential to have each-distributed object with their own ID... Specially when working with RGB-Masking.
Now after almost 10 Years of working with Corona I definitly don't want this to change...
And also working with Light-Mix for a fair amount of time, i definitly don't want the old "behaviour" to be changed.
So definitly preferring individual sub-lights over a combinded forrest-pack in light-select element. If there is a technical reason why you can't have both.
-
Will this Issue be resolved in a future Update?
For now 12u1 is unusable without it.
-
Just in case someone missed it - same story with FP and masking. You can't add objects into the mask include/exclude as before, only the entire FP object. I think those problems have the same roots.
-
Hi,
Letting you know that this has been fixed internally (reverted to previous behavior).
For the masking, can you please show, in a simple scene, what you mean, so that can be checked too?
-
Sure, here it is:
first image - chaos scatter. Scattered objects - box, sphere and teapot included in r,g,b
second image - forest pack. Same setup, but FP objects aren't visible in the mask(previously it worked the same way as chaos scatter)
third image - FP object included in R channel of another mask. Not 100% sure, but it look like previously it wasn't possible to include the entire FP object in the mask
I just noticed - I messed up with the render setting on the screenshots (didn't show the mask settings), but I think you get the point.
-
I my guess what you are referring to, but please be more specific:
E.g.
X mask element's Y mode worked this/that way in Corona 12 HF1 (for ChaosScatter object and/or FP object) and changed in Corona 12 Update 1.
Or something similar.
It will help to both understand and report the issue the best.
-
previously FP worked the same way as Chaos Scatter.
I have 3 objects - box, sphere, teapot.
Then I scatter these objects with forest pack.
To see items scattered by FP I need to include box, sphere and teapot into cshading_mask. That's how it worked before. That's how chaos scatter works.
But now it has no effect on FP. If I include those objects - I can't see FP items in the mask.
Simply - the second screenshot should look the same as the first one, but it doesn't.
-
I believe what you are saying is the original issue (for lightmix layers), as it is a matter of how/what to include in a (masking) layer.
I have tested it with the fix, and it is indeed working as you say - an object and all its instances in forest object are included in the mask.
-
it seems the validation or beta of hotfixes are not tested thoroughly enough.
as those things seem to mess up production workflows ... previously i could mostly switch to a new version of corona without so much of fear everything breaks apart. As it might with a new version of 3dsmax and not version-compatible plugins ... when they are released later even sometimes a half year later.
Not sure what happend with corona 12 update 1 ... but releasing an version which alters behaviors as harsh as this is not good. You should consider revoking 12 update 1 or releasing a fix quite soon.
sorry for being that harsh - but I'm supportive of corona for a long time and with increased subscription prices i might be also more demanding.
-
Word by word my point of view.
Good Luck
-
it seems the validation or beta of hotfixes are not tested thoroughly enough.
as those things seem to mess up production workflows ... previously i could mostly switch to a new version of corona without so much of fear everything breaks apart. As it might with a new version of 3dsmax and not version-compatible plugins ... when they are released later even sometimes a half year later.
Not sure what happend with corona 12 update 1 ... but releasing an version which alters behaviors as harsh as this is not good. You should consider revoking 12 update 1 or releasing a fix quite soon.
sorry for being that harsh - but I'm supportive of corona for a long time and with increased subscription prices i might be also more demanding.
There are infinitely many test scenarios, and that is why we highly encourage to try or test a few scenes/scenarios, as close as possible to production (with a backup of course), so such things are found and tackled. If this was brought up during RC/DB stages, it would have definitely been handled. As a note, this was not planned to change at all, but with FP IVY introduction, it made us do non-minor last minute changes, which brought to masking behavior change.
In my honest opinion, this is not a harsh change, and even the desired behavior (selecting a single object and having all its instances inside other objects) is weird for me. What if you want only that single element, without instances? That is why, before making a decision, I have asked if this is the wanted behavior, if you guys think there is something to improve.
Demand as highly as needed, that is welcome surely, but also make sure that you contribute to finding and fixing thing as best as possible, in as constructive way as possible.
-
Well the behavior was there for many years... At least for object id and also since light mix was a thing.
So we adapted workflows upon it. So it doesn't seem like a nice idea to suddenly change it... And then ask later if that new behavior is something we want.
-
and also its not as odd as it sounds to have some objects in your scene as "master" and then distribute them with forrest pack - so i can easily exchange furniture or lights or whatever when the demand is there for it.
With having unique masks (rgb by object-id) you always can select this specific object and then do some "post-pro" touchups....
Also with lights its quite an efficient workflow when having multiple differen light objects as master and want to distribute them in certain ways and later maybe exchange just the object and not the position.
So you dont need to mess around with 100s of instances in your scene-explorer...
and the fun part is then for photoshop or whatever... you have each light as a light-select element where you can use that in photoshop to enhance certain areas...
Much faster then always figuring out all the perfect light-setup inside lightmix or even in 3dsmax before...
Sometimes its speedier just do it in post (even though its not as popular as it was 20 years before)
The examples shows ... scene explorer only has 5 objects for this light setup ... 4 different ligths and one forrest-object ...in this "new behaviour" i would have to creat for each light a forrest-object by itself ... or distribute each light as instance.. which is quite a pain in the end.
and sorry for sounding "pissed" before - i appreciate if you introduce new options... but it was weird asking after that sudden and maybe unwanted change appeared, if we would prefer that instead of the old one... i mean we got workflows now on the older behaviour and might even need to work on older projects. i dont really want to install every now and then an older version just to keep working on older projects. that might be ok for really old projects where you kept track of plugins and renderengines and might even just use an old version of max... but not something with "minor" versioning.... like from 12hfx1 to 12update1.
-
There are infinitely many test scenarios, and that is why we highly encourage to try or test a few scenes/scenarios, as close as possible to production (with a backup of course), so such things are found and tackled.
No one would expect such fundamental changes when reading "Added support for ForestPack Ivy Generator". In fact, many would not even have the newest FPP installed, so they won't care. It's entirely ok to change things/try out something but it happened in major version 12 as an "update"? (for whatever reason this was chosen - first time in Corona history). In a major versions "update" I do not expect such changes at all.
Basically it's very cool that the feature got support, otherwise the yelling would have started later anyway. But it is definitely a very basic change. It does not only affect lightMix and masks, it also affects VFB raycast picking. Now it is not possible any more to quickly pick that nasty, wrong looking FPP custom object to fix it. Instead, you get the FPP object itself.
By the way, that discussion is there for a long time (https://forum.corona-renderer.com/index.php?topic=15201.0) and as mentioned previously, users adopted the way Corona handled FPP and took opportunity from it because there has been no movement or commitment. So scenes use the "old" behavior, also v12 ones. Now that update in the same major versioning. I'm still not sure if it was an accident and some unknown side effect because of the last minute changes or a known implication. Basically there should be more time to test for users and developers anyway. Usually you are swamped with work and occasions to test are rather rare. But I'm repeating myself again.
In my honest opinion, this is not a harsh change, and even the desired behavior (selecting a single object and having all its instances inside other objects) is weird for me.
You can say the same about how hierarchies are handled. Isn't it weird that you get multiple objects in mask when including a single one? Yes it seems. But here the same: users adopted that behavior and link objects to another to have them semi automatically included in a mask. You would break again many scenes when changing this. As said: all ok in another major Corona version where users can expect fundamental changes - and if it gets announced.
I hope you get my point - I got yours. Meanwhile, as of FPP 9, edge mode in IR seems to be broken which is worse :)
Good Luck
-
There are infinitely many test scenarios, and that is why we highly encourage to try or test a few scenes/scenarios, as close as possible to production (with a backup of course), so such things are found and tackled.
No one would expect such fundamental changes when reading "Added support for ForestPack Ivy Generator". In fact, many would not even have the newest FPP installed, so they won't care. It's entirely ok to change things/try out something but it happened in major version 12 as an "update"? (for whatever reason this was chosen - first time in Corona history). In a major versions "update" I do not expect such changes at all.
Basically it's very cool that the feature got support, otherwise the yelling would have started later anyway. But it is definitely a very basic change. It does not only affect lightMix and masks, it also affects VFB raycast picking. Now it is not possible any more to quickly pick that nasty, wrong looking FPP custom object to fix it. Instead, you get the FPP object itself.
By the way, that discussion is there for a long time (https://forum.corona-renderer.com/index.php?topic=15201.0) and as mentioned previously, users adopted the way Corona handled FPP and took opportunity from it because there has been no movement or commitment. So scenes use the "old" behavior, also v12 ones. Now that update in the same major versioning. I'm still not sure if it was an accident and some unknown side effect because of the last minute changes or a known implication. Basically there should be more time to test for users and developers anyway. Usually you are swamped with work and occasions to test are rather rare. But I'm repeating myself again.
In my honest opinion, this is not a harsh change, and even the desired behavior (selecting a single object and having all its instances inside other objects) is weird for me.
You can say the same about how hierarchies are handled. Isn't it weird that you get multiple objects in mask when including a single one? Yes it seems. But here the same: users adopted that behavior and link objects to another to have them semi automatically included in a mask. You would break again many scenes when changing this. As said: all ok in another major Corona version where users can expect fundamental changes - and if it gets announced.
I hope you get my point - I got yours. Meanwhile, as of FPP 9, edge mode in IR seems to be broken which is worse :)
Good Luck
I get your points about changes/updates for version numbering/naming. The changes were last minute (in dev timescale) and were done quickly to support the new IVY (IVY update came a week or so before planned release), which brought the changes, so nothing intentional. But when hierarchy selection/inclusion will be fixed, it will be intentional, as it is reported as a bug/feature request (as an option). I get that the behavior is adopted, but that doesn't mean it is correct or expected, and doesn't it need to be changed or improved? If it is clear that the behavior should not be changed, so be it, at least this discussion brough it up.
Edge mode with IR is known and afaik already fixed internally.
(Report ID=CMAX-1424)
-
There are infinitely many test scenarios, and that is why we highly encourage to try or test a few scenes/scenarios, as close as possible to production (with a backup of course), so such things are found and tackled.
I get that the behavior is adopted, but that doesn't mean it is correct or expected, and doesn't it need to be changed or improved?
Still weird you state that a behaviour which was there since many years is "not correct" ... if it weren't correct you could have mention it before. Now that ForrestIvy is there suddenly you change things...reagarding Forrestpack behaviours. It wasn't a problem for many years, but seems to be now.
For me ForrestIvy behaves a lot like RailClone and i'm ok with railclone not having unique IDs for each subobjects, rather one object-id for the main RC-object - same with Lightmix.
But its a different scenario - i can work with this and have adapted to it.
With Forrestpack it was a cool feature you can have individual IDs/Lightmixselects per suboject and not the whole - even though in some scenarious this might be wanted. But then you could bring it as an option or figure things out with itoo... to add a option-button if masks/lightselects will be individual per suboject or for a FP-object. Then artists have more freedom with controling it.
Other than that I haven't tested it with cryptomatte yet - might be nice to have both masking options in one rendering - like whole FP object in cryptomatte and individual sub-objects in Cmasking-mask
-
which brought the changes, so nothing intentional. But when hierarchy selection/inclusion will be fixed, it will be intentional, as it is reported as a bug/feature request (as an option).
So all fine, kind of accident while trying to push in as much features as possible. Good to read about having sort of hierarchy handling option in mind. I know quite a few people struggling with the current behavior as well as others using it happily as it works currently.
Edge mode with IR is known and afaik already fixed internally.
That's good news, thanks!
Good Luck
-
Despite I was a bit ranting on certain things.
Thanks its being addressed and taken care of.
Merry Christmas to the Developers and Team!