PDA

View Full Version : LW2018 - Enable Bump ?



erikals
01-15-2018, 02:06 PM
hi there,

anything like Enable Bump or similar available in LW2018 ?

Sensei
01-15-2018, 02:43 PM
(not exactly the same)

Object Properties,
Surface Displacement,
double click it, to open panel, and adjust Distance.
And plug something to Displacement.

erikals
01-15-2018, 03:06 PM
ah!..

excellent!  :king:

prometheus
01-15-2018, 03:53 PM
Blah double post

prometheus
01-15-2018, 03:56 PM
Itīs quite not the same, and the function of actually tweaking a bump map and have it displace at the same time is broken, as mentioned..surface displacement may actually be easier for displacement, but it also requires more dig in to to get the displacements, an for tweaking bump and displacement at the same time, you obviously need to feed both displacement output and bump output in to respective inputs, so I actually think the previous way was a bit shorter and easier in steps taken within the workflow, in 2018 you have to enter both the primitive properties surface displacement and check it to be on, then set a distance as you would with enable bump in deform tab previoulsy, then you have to enter node graph in the material and further on search for your procedural and add it, then go through the hoops of connecting it.

Personally I see a tendency of going too much nodal without things being hardwired in easier setup buttons a bit troublesome, not only here..but within the new volumetric system as well and some other stuff, it worries me a bit that Lighwave will loose a bit of the old fast and easy setup workflow.

JamesCurtis
01-15-2018, 04:25 PM
I agree with you on these points!

creacon
01-15-2018, 04:39 PM
A displacement map and a bump map are 2 completely different concepts.
It's good they're separated now. (and if you used the node editor in LW2015 they already were)

creacon

erikals
01-15-2018, 04:47 PM
i have to disagree a bit, since the impact is minimal.

looking at how much faster LW2018 workflow wise is in general.

though yes, there are 1 or 2 things that could be improved.
that said, overall, i find the new system much better, once i got used to it.

(Turbulence just had an update by the way, check Kat's video)

Gi and PBR is way easier / better overall in LW2018.

while LW2018 is not perfect, it's very welcome.

prometheus
01-15-2018, 04:53 PM
A displacement map and a bump map are 2 completely different concepts.
It's good they're separated now. (and if you used the node editor in LW2015 they already were)

creacon

The way they worked together is not different concepts, individually yes, depends on your own workflow which kind you like.

prometheus
01-15-2018, 04:57 PM
i have to disagree a bit, since the impact is minimal.

looking at how much faster LW2018 workflow wise is in general.

though yes, there are 1 or 2 things that could be improved.
that said, overall, i find the new system much better, once i got used to it.

(Turbulence just had an update by the way, check Kat's video)

Gi and PBR is way easier / better overall in LW2018.

while LW2018 is not perfect, it's very welcome.

Thanks for the heads up on turbulence, I need to install the latest demo and try and see how it works now with 2018.
As for dsagree..we all have our rights to that of course, and yes..maybe overall it has improved, but I honestly can not judge that in a sum up yet, still much more to test, but what I have seen when getting acess to textures ..procedurals, setting up volumetrics..I am not that fond of that Tendency to disconnect features and leave it up to users to go deeper in the nodes to re-connect the ease of tweaking a function.
But ..to early to say, I just see tendency of this occouring in some places, and I hope it doesnīt start to popup everywhere.

erikals
01-15-2018, 05:02 PM
link :)
http://www.liberty3d.com/2018/01/free-video-new-turbulencefd-build-1433/

yes, some features could be added, sort of like a default preset.
be sure to send feature requests.
i have some upcoming ones.

prometheus
01-15-2018, 05:05 PM
Enhanced workflows..
I just recently answered a question here in the forums, about copying particle emitter settings, and with 2018 it is now easy to copy a particle emitter setting to several other layers at once, if you use the scene editor, donīt think that was possible with 2015, so that is a good one for 2018.

http://forums.newtek.com/showthread.php?155709-How-do-you-copy-emitter-settings-from-one-layer-to-multiple-layers-in-Layout&p=1533444#post1533444

----
Working and setting up a volumetric effect is however with the new system much harder in almost anything, it wonīt work on point clusters, you can not just pull down a hypertexture list, that requires entering nodes and search for whatever texture you need and then plug it in, what used to be tiny half seconds, now takes several seconds in that regards, setting up a motion effect within the new system was made easy with hypertexture effect speed values and various effect, where we now have to add nulls and dig in deeper in the workings, shading model is also harder to use.

We do not have a new replacement for liquid particles either, the new system doesnīt come with a surface mode as hypervoxels has, neither sprites.

I see the new system as half implemented currently and Not a real replacement, but an addition to an old system, though I would have liked them to make it easer to acess the old hv system.
+++ allows for more realistic hero clouds, though it will not be able to produce godrays and large sky vistas alo ozone, vue, terragen or ogo taiki in a decent speed render result.

+++
On the other hand, I think they took a right path with the volumetric scattering and the new lights, except that I would have liked a sprite function within some lights, as we had in the old volumetric system, that is all gone now, and even not replacable really, I also miss a true distant light working with volumetrics, this new distant light doesnīt work with the volumetric intergrator unfortunately.

prometheus
01-15-2018, 05:32 PM
A displacement map and a bump map are 2 completely different concepts.
It's good they're separated now. (and if you used the node editor in LW2015 they already were)

creacon

Your comment got me wondering, why is it good they are separated now?
by the way, they have always been seperated in essence, the function of adding a bump displacement was asked for or a feature added that seemed good to have, it wasnīt there in the older lightwave versions, so how do you figure it is good they are seperated?

If the workflow or new additions with surface displace makes it not useful anymore? that I have yet to be convinced about, I could ofcourse be wrong.

erikals
01-15-2018, 06:19 PM
though it will not be able to produce godrays and large sky vistas
from what i've seen it should look fine and render in a decent time (?)

GodRays
http://static.lightwave3d.com/marketing/lightwave_2018/release/img/volumetrics/lightwave3d-volumetrics-trees.jpg
Shadow
http://static.lightwave3d.com/marketing/lightwave_2018/release/img/volumetrics/lightwave3d-volumetrics-ed-209.jpg
Self Shadow
http://static.lightwave3d.com/marketing/lightwave_2018/release/img/volumetrics/lightwave3d-volumetrics-cloud5.jpg

there is also an Earth w/Clouds scene in the content files (full version only, not trial)

prometheus
01-15-2018, 06:33 PM
from what i've seen it should look fine and render in a decent time (?)

GodRays


there is also an Earth w/Clouds scene in the content files (full version only, not trial)

Not fast rendertimes, itīs just volumetric lights in trees, not within a volumetric item such as clouds....terragen and vue still does this faster and better.
Erikals...none of those samples are really godrays, itīs just volumetric beam scattering in trees, not the per definition godrays, crepuscular rays ..which is rays casts from clouds obscuring light, I am almost certain of that..
Edit..may be wrong in the classification, if now crepuscular rays can be defined as such rays cast with other objects like trees, but it is till not in reference to the long rendertimes on what I ment about clouds and volumetric mixing.

And the thing is..tree samples like this and some in the trial content and the new volumetrics is a great addition, and helps sunbeams also in settings with windows, chearch and similar, but itīs not any samples of rays cause by the new volumetric voxel system, Lino posted a sample some time ago with that, and I did some samples the other day with the new system..but to be aware of, rendering times will still be slow despite the new engine, and

I am not talking about samples like the forrest, and if you try and make godrays with the new volumetrics, it means you Can not use the distant light..it doesnīt work with the new volumetric system, you are forced to use point, spherical or linear or some of the other lights, but distant light wonīt work, that poses an issue of getting the other lights positioned right, you need either sunspot or sunsky motion if you want to tweak the lightīs elevation and azimuth, but if you use that, you will be having a hard time getting the light volumetric to propagate through the atmosphere all the way to through a cloud, very very difficult..if not impossible...

with such lights as point light and spherical ( I recommend spherical) you need to position the light not too far behind the clouds (which physicly isnīt really right) for it to take an effect of casting godrays, so it is acheivable by faking the light a bit and position it behind the cloud volume slightly, but there is then a couple of problems, the clouds really need a distant light to light up and break shadows within the cloud volume, if you only have a point light within the cloud, it will look horrible, if you then however clone the light and change to distant light, you will get a better ligthing in the cloud but also having the other light cast godrays.


Thereīs also a way of using the new system in the same way I did before with hypervoxels, turn of the volumetric scattering for lights, and simply use a volumetric item with very..very low density and let it catch rays from any light type, and in this case you can use a distant light as well, with a fog item like that, I can also set fog height and depth to control where the fog is lying, wherelse the fog scattering may be much more difficult with setting various nodes up for height en length of the global scattering.

Another note, some clouds sample in the trial, are very sized down in scale, coupl of meters I think, once you start to scale cloud items up to more physical scale, it will become much slower, so I am not overthrilled with the speed of the new system, I like the looks of them in terms of shading and softness.
hypervoxels old legacy is also benefitting from the new renderer, it allows for multiply nulls vith hvs..without freezing the system (not possible before)as I mentioned and renders as fast as the new volumetrics, if not faster actually, and with the new sampled lights, it also renders with details not possible before..and that is with the old hypervoxels that is.

The clouds and lighting scattering in clouds and atmosphere is more accurate and quite speedy in terragen, but for fly through clouds and hero clouds specially designed, Lightwave may be up for the challenge...and terragen isnīt the nicest software in terms of using and UI.

erikals
01-15-2018, 06:43 PM
Erikals...none of those samples are really godrays, itīs just volumetric beam scattering in trees, not the per definition godrays, crepuscular rays ..which is rays casts from clouds obscuring light, I am almost certain of that..
not as far as i know.

- you need to add fog/haze/atmosphere near the ground, (in addition to the clouds) that's what causes the GodRay effect.
- there is also a "Sun heats Earth and water vaporizes upwords" but that is a bit different.
- there is also rain, of course.

so 3 different scenarios, all should be possible. if volumetrics works the way i think/hope.

however, only in theory, i guess we'll know in the near future after some more testing.

[agh] 3 am... ...

prometheus
01-15-2018, 06:55 PM
not as far as i know.

- you need to add fog/haze/atmosphere near the ground, (in addition to the clouds) that's what causes the GodRay effect.
- there is also a "Sun heats Earth and water vaporizes upwords" but that is a bit different.
- there is also rain, of course.

so 3 different scenarios, all should be possible. if volumetrics works the way i think/hope.

however, only in theory, i guess we'll know in the near future after some more testing.

[agh] 3 am... ...

not quite, fog haze you donīt add, that is a real physical property:9 with lightave and the volemetric scattering, you really donīt add it either as fog or haze near the ground...it is added all over the scene and doesnīt really follow any physical formula of being haze or fog on the ground.
And for real life, itīs more particles in the air, not specificly fog or haze near the ground.
Rain...not sure what you mean there, water particles in air can have various density or molecule size I think, causing light to pass through in various ways.
And letīs take a look at "certified facts" :) in wiki :)
https://en.wikipedia.org/wiki/Crepuscular_rays

But I dare you to try and get godrays from behind a cloud with the new volumetrics, and try and keep the clouds looking lightened up properly as well...and if you do, tell me the rendertimes as well :)
So little time though.

hereīs a sample from 2018 version, but with the trick of volumetric fog item, not volumetric scattering, so itīs no light set to be volumetric...Ps...Check the density setting of 0.0002

http://forums.newtek.com/attachment.php?attachmentid=139143&d=1515018487

erikals
01-15-2018, 07:20 PM
think i see what you mean now.

more tests needed...

[edit] i got "something" working with distant light.

too late though... 3.40 am... ...

Cageman
01-15-2018, 07:36 PM
Wait a minute!

This thread was about "Enable Bump" and it turned into a thread about Volumetrics....

Can you guys please stay on topic?

erikals
01-15-2018, 07:38 PM
we're just busy. :)

ok, you are right, will do.

prometheus
01-15-2018, 08:04 PM
Wait a minute!

This thread was about "Enable Bump" and it turned into a thread about Volumetrics....

Can you guys please stay on topic?

Not necessarely, we can try..but that is how a discussion go quite often, it overlaps other areas...didnīt realize the forum rules are so strict about that :)
But advice is taken, should have been directed to a volumetrics thread perhaps.

I rarely see a issue with someone going off topic in other threads transcending to other lightwave aspects, unless it becomes pure garbage, but thatīs just me.

Edited...I bumped the old thread, and started to post what we strayed to on that thread for volumetrics and godrays...

http://forums.newtek.com/showthread.php?152475-Volumetric-tricks-for-atmosphere-and-working-with-sk_sunsky

gar26lw
01-16-2018, 01:40 AM
Itīs quite not the same, and the function of actually tweaking a bump map and have it displace at the same time is broken, as mentioned..surface displacement may actually be easier for displacement, but it also requires more dig in to to get the displacements, an for tweaking bump and displacement at the same time, you obviously need to feed both displacement output and bump output in to respective inputs, so I actually think the previous way was a bit shorter and easier in steps taken within the workflow, in 2018 you have to enter both the primitive properties surface displacement and check it to be on, then set a distance as you would with enable bump in deform tab previoulsy, then you have to enter node graph in the material and further on search for your procedural and add it, then go through the hoops of connecting it.

Personally I see a tendency of going too much nodal without things being hardwired in easier setup buttons a bit troublesome, not only here..but within the new volumetric system as well and some other stuff, it worries me a bit that Lighwave will loose a bit of the old fast and easy setup workflow.

i completely agree. it’s bad, real bad. i’m keeping modo in mind because of this.

creacon
01-16-2018, 06:06 AM
You might want to control the amplitude of your bumpmap and your displacement separately.
Or use a procedural noise for displacement that uses less frequencies and a lower "Small Scale"
Or tone down your bump with distance while the bump stays the same.

And so on

creacon



Itīs quite not the same, and the function of actually tweaking a bump map and have it displace at the same time is broken, as mentioned..surface displacement may actually be easier for displacement, but it also requires more dig in to to get the displacements, an for tweaking bump and displacement at the same time, you obviously need to feed both displacement output and bump output in to respective inputs, so I actually think the previous way was a bit shorter and easier in steps taken within the workflow, in 2018 you have to enter both the primitive properties surface displacement and check it to be on, then set a distance as you would with enable bump in deform tab previoulsy, then you have to enter node graph in the material and further on search for your procedural and add it, then go through the hoops of connecting it.

Personally I see a tendency of going too much nodal without things being hardwired in easier setup buttons a bit troublesome, not only here..but within the new volumetric system as well and some other stuff, it worries me a bit that Lighwave will loose a bit of the old fast and easy setup workflow.

prometheus
01-16-2018, 07:11 AM
You might want to control the amplitude of your bumpmap and your displacement separately.
Or use a procedural noise for displacement that uses less frequencies and a lower "Small Scale"
Or tone down your bump with distance while the bump stays the same.

And so on

creacon

Well, nothing in lightwave 2015 prevented you from skipping enable bump, and just enter the displacements with nodes as we sort of can now with 2018, and for separately controlling amplitude, you could always balance the enable bump distance strength values, if you raise the actual bump opacity, you lower the distance if needed, so you always had the option of using enable bump and managage that as almost one with fast easy setup, or you could go nodal wich takes a bit longer to set up.

I just think the way 2015 worked with that option was a good addon and easy fast to set up, while the standard material in 2018 doesnīt provide any means of enabling it that way..you need to go nodal, adding yet another texture, hook it in to the right slots, itīs way more slower in workflow I think, I have nothing to say about the flexibility and the power behind nodes, I get that..but nodes inevitable brings in a moment of being more complex and taking longer time to set up, not specificly for this case, itīs valid for other types of set up as well...I can only hope that the lightwave team may hear these aspects of the workflow, and try to keep things connected to the lighwave philosophy of making it easy and fast to use, going all nodal in every little corner will stray away from that, for good or bad..who knows.

going slight off topic again, bringin down a simply texture for volumetrics for instance, I just canīt get it why a drop down list with the most common textures cant be available, and if you need to, enter nodal and change it or connect it with other texture nodes, but please keep the fast workflow and do not make it harder to use because of neglecting this kind off stuff.

gar26lw
01-16-2018, 07:32 AM
iI think if lwg keep things simple to setup, with fast settings and easy controls, then also offer a way to convert to nodal then we will have a win win situation. fast for previz and setup, fast to go nodal, complex and deep in nodal.

plenty of 2015 was on that path. with the simple texture buttons and the nodal options. I think that way was awesome and should be reintroduced, even if under the skin, its really all nodal

prometheus
01-16-2018, 07:40 AM
iI think if lwg keep things simple to setup, with fast settings and easy controls, then also offer a way to convert to nodal then we will have a win win situation.fast or previz and setup, fast to go nodal, complex and deep in nodal.

plenty of 2015 was on that path. with the simple texture buttons and the nodal options. I think that way was awesome and should be reintroduced, even if under the skin, its really all nodal

I agree completely, thatīs how I would want it to work as well, and if done right I see nothing to be bothered about from anyone of the side, those who loves nodal will have no issues entering and changing, and those who prefer direct acess to procedural textures within a certain module function can have that two, and only enter nodal editors when needed...What I can not see is if this simply wouldnīt be doable ..or require extensive amount of work to get it done.

On the other hand, they sort of had it in 2015, with hypertextures for instance, if you simply want to add a texture, you go directly to where it says texture and pull down your texture from the list, and if you want to go more advanced, enter node editor, though in that case the texture choosen wasnīt already setup in the node editor, wether or not they could pull that off as well, but otherwise not 100% needed, you can simply pull down in the node editor what you need and plug it in to the final output node, as we do today, but a complete node connection within the nodes representing the choosen texture in the main drop down menu would be great.

gar26lw
01-16-2018, 07:46 AM
I agree completely, thatīs how I would want it to work as well, and if done right I see nothing to be bothered about from anyone of the side, those who loves nodal will have no issues entering and changing, and those who prefer direct acess to procedural textures within a certain module function can have that two, and only enter nodal editors when needed...What I can not see is if this simply wouldnīt be doable ..or require extensive amount of work to get it done.

I think it would be doable. it was half way there in 2015.3. I am starting to appreciate how good 2015.3 was.

"you don't know what you've got til its gone.."

prometheus
01-16-2018, 07:51 AM
I think it would be doable. it was half way there in 2015.3. I am starting to appreciate how good 2015.3 was.

"you don't know what you've got til its gone.."

If they listen and try and get back to that easy acess workflow, I think it can be good, I donīt think itīs impossible ..the question is if they will do it or not.
As for how good 2015.3, well for me it was a crashing disaster, it was fast and workflow wise good, but stability the worst version I ever tried...as for 2018, it seems more stable..but yet to early to tell for me.

Small changes like hitting the subpatch toggle in 2015 did not work with VPR active, you had to turn that off then on again to make it work, it is now fixed in 2018, adding an object in 2015 always created double surfaces, which often made me change the wrong settings when standing on the empty null surface, that is now fixed in 2018, particle emitter settings can no be copied over to several layers at once etc, not possible in the lightwave 2015 version.

gar26lw
01-16-2018, 08:15 AM
yes it is better but the point i’m trying to make, if some dev is reading this, is that they were on the right track with 2015.
just do more of that and allow for industry standards in navigation and a few other ui and layout things. it would/could be really good.
im toying with the idea of doing a video of all the good bits from other 3d apps that i believe should be in lw. if they were, oh baby, it would be so good

prometheus
01-16-2018, 08:47 AM
yes it is better but the point i’m trying to make, if some dev is reading this, is that they were on the right track with 2015.
just do more of that and allow for industry standards in navigation and a few other ui and layout things. it would/could be really good.
im toying with the idea of doing a video of all the good bits from other 3d apps that i believe should be in lw. if they were, oh baby, it would be so good

Itīs like trying to merge all the good parts from various countries and make it happen in one country, it wonīt happen:D

but I know what you mean, my choice would be some kind of hybrid of blender, modo , lightwave...with build in fluid system, would tear out the skin modifier, sculpting in blender, weigth paint and assignment, mix it in a lightwave environment, with a touch of graphic UI scheme in modo (except for vertical tabs) and get modoīs preset and content system, and a few other things in there, but with the general looks of lightwave if needed too, a switch option of having either text ala lightwave UI or modo UI.

hypersuperduper
01-16-2018, 11:11 AM
It seems to me that some bump/displacement control consolidation would be fertile ground for plugin devs. I wouldn’t mind some cel shading/edge rendering consolidation as well. Origami digital are you listening?

gar26lw
01-19-2018, 08:50 PM
Itīs like trying to merge all the good parts from various countries and make it happen in one country, it wonīt happen:D

but I know what you mean, my choice would be some kind of hybrid of blender, modo , lightwave...with build in fluid system, would tear out the skin modifier, sculpting in blender, weigth paint and assignment, mix it in a lightwave environment, with a touch of graphic UI scheme in modo (except for vertical tabs) and get modoīs preset and content system, and a few other things in there, but with the general looks of lightwave if needed too, a switch option of having either text ala lightwave UI or modo UI.


lw could be

http://forums.newtek.com/attachment.php?attachmentid=139620&d=1516420155


instead of

http://forums.newtek.com/attachment.php?attachmentid=139621&d=1516420166




139620
139621

prometheus
01-20-2018, 09:29 AM
lw could be

http://forums.newtek.com/attachment.php?attachmentid=139620&d=1516420155


instead of

http://forums.newtek.com/attachment.php?attachmentid=139621&d=1516420166




139620
139621

We have laws against doughnutīs in Sweden, it attracts to many folks with the characteristics of a certain homer simpson.

gar26lw
01-20-2018, 05:24 PM
hehe, i wonder if the ux we are talking about was what rob was contributing and due to his departure things went away from 2015 ?