PDA

View Full Version : Show me that Lightwave 2018's new render engine is a worthy successor to 2015...



Paul_Boland
03-26-2018, 09:35 AM
Hi Folks.

There was a recent heated debate on the Lightwave Wiki Facebook page about the render engine in Lightwave 2018 vs Lightwave 2015. A moderator of that page has closed that thread. There is a lot of split opinions about the Lightwave 2018 renderer and if it's a worthy successor to 2015? After playing around with it for over two months, reading the online documentation from cover to cover, and watching many videos on YouTube about it, as of this point I'm not convinced. Every test I do in Lightwave 2018 is a hell of a lot slower than the same test in Lightwave 2015 and the render results are not as good. And I'm not a lone voice in the wilderness saying this.

Whenever this debate comes up, all I keep seeing from the pro-Lightwave 2018 group is "It's a new engine, you have to learn it's new ways." Well, folks, I've tried... I've tried damn hard to learn it. And all I see is an inferior engine in 2018. I decided to be fair and to try and pit Lightwave 2018 and Lightwave 2015 against each other on even terms, creating the same scene file in both packages, sorting out the render setting so they match, and texturing them the same too, and the results were jaw-dropping bad for Lightwave 2018. You can view the video on this here:


https://www.youtube.com/watch?v=4Yjgg0k7_9c

The end result...

140869
Lightwave 2015 rendered this in 46.8 seconds.

140870
Lightwave 2018 rendered this in 9 minutes 48 seconds!!

So the reason for this topic I'm posting here is, I want someone to show me, help me, to see that the Lightwave 2018 render engine is a worthy successor to Lightwave 2015! Tell me what I need to do shave 9 minutes off the 2018 render time, bringing it in par with 2015, producing a crisp, clean, beautiful render on par with Lightwave 2015 in around 46 odd seconds. I'm attaching both scene files below in the zip file. Show me where I'm going wrong, prove to me that Lightwave 2018 is a worthy successor, and I'll create a follow-up video highlighting the facts. Because right now, the render times and end results in Lightwave 2018 compared to Lightwave 2015 leave a lot to be desired...

mummyman
03-26-2018, 09:44 AM
Oh.. boy.. here we go. My day is planned!

Chuck
03-26-2018, 09:55 AM
Whenever this debate comes up, all I keep seeing from the pro-Lightwave 2018 group is "It's a new engine, you have to learn it's new ways."

That's not all that's been said and I've seen a number of posts from RebelHill, Elusive Elephant, Matt Gorner, Antti Järvelä and others who have been working with 2018 for some time that they can get equal or better quality with lower rendering times for most cases, and they've discussed the settings that need to be experimented with for those results. One of the things I have on my task list is to gather up the tips, pointers and commentary on use of the new renderer into some blog entries, which I hope will help folks along the learning curve.

That said, I certainly look forward to those folks and others taking a look at your scene and working with it. It may also turn out to be of interest for development, because as a new renderer it is likely that there is plenty of room for reviewing use cases and tuning for better performance. It is a good idea, when you have a scene that seems to be intractable in getting to a relatively equal or better time to 2018, to submit that as a bug report.

jeric_synergy
03-26-2018, 10:43 AM
That's not all that's been said and I've seen a number of posts from RenderHill, Elusive Elephant, Matt Gorner, Antti Järvelä and others who have been working with 2018 for some time that they can get equal or better quality with lower rendering times for most cases, and they've discussed the settings that need to be experimented with for those results. One of the things I have on my task list is to gather up the tips, pointers and commentary on use of the new renderer into some blog entries, which I hope will help folks along the learning curve.
While such is aggregation is GOOD, what would be even better would be if such articles were linked IN THE DOCUMENTATION.

I've been encouraging this for years, and that it would be totally under LWG's control (versus worthy but independent online articles) makes it even more of A Good Idea.

There's lots of worthy material out there that would enhance the LW documentation: I don't see any point in not using it, even if some fraction of it is ephemeral. To not have it in the dox just makes life harder for LW users.

darkChief
03-26-2018, 11:19 AM
Hi Folks.

There was a recent heated debate on the Lightwave Wiki Facebook page about the render engine in Lightwave 2018 vs Lightwave 2015. A moderator of that page has closed that thread. There is a lot of split opinions about the Lightwave 2018 renderer and if it's a worthy successor to 2015? After playing around with it for over two months, reading the online documentation from cover to cover, and watching many videos on YouTube about it, as of this point I'm not convinced. Every test I do in Lightwave 2018 is a hell of a lot slower than the same test in Lightwave 2015 and the render results are not as good. And I'm not a lone voice in the wilderness saying this.

Whenever this debate comes up, all I keep seeing from the pro-Lightwave 2018 group is "It's a new engine, you have to learn it's new ways." Well, folks, I've tried... I've tried damn hard to learn it. And all I see is an inferior engine in 2018. I decided to be fair and to try and pit Lightwave 2018 and Lightwave 2015 against each other on even terms, creating the same scene file in both packages, sorting out the render setting so they match, and texturing them the same too, and the results were jaw-dropping bad for Lightwave 2018. You can view the video on this here:


https://www.youtube.com/watch?v=4Yjgg0k7_9c

The end result...

140869
Lightwave 2015 rendered this in 46.8 seconds.

140870
Lightwave 2018 rendered this in 9 minutes 48 seconds!!

So the reason for this topic I'm posting here is, I want someone to show me, help me, to see that the Lightwave 2018 render engine is a worthy successor to Lightwave 2015! Tell me what I need to do shave 9 minutes off the 2018 render time, bringing it in par with 2015, producing a crisp, clean, beautiful render on par with Lightwave 2015 in around 46 odd seconds. I'm attaching both scene files below in the zip file. Show me where I'm going wrong, prove to me that Lightwave 2018 is a worthy successor, and I'll create a follow-up video highlighting the facts. Because right now, the render times and end results in Lightwave 2018 compared to Lightwave 2015 leave a lot to be desired...

Don't know about matching speeds, but I improved the times on the scene you shared. I switched on interpolated Gi (primary rays 32, secondary 16. Double each value if you notice splotches). Lowered the camera as samples (minimum 1, maximum 4). And reflection samples between 4-8 gives good results with noise filtering.

Also Principled BSDF takes longer than Standard materials.

Also Gi doesn't seem to make a big difference for the scene, switching it off also improves times.

Hope that Helps in some way.

prometheus
03-26-2018, 11:36 AM
I think some things look better, was testing some daz figures and using the new skin material vs old looks very nice in 2018, and alos eye reflection with dielectric etc looks very nice..and you get what you see in vpr ..propagating through to final render.
Some landscape terrain setting ..in general principled look better, though I still have to discover the Additive sharpness..and see what in lightwave 2018 that replaces that setting from 2015, addivite sharpness for rocks was great, but I do not see a way for the new shaders to work with that..or I have just missed it, it may not be necessary.

What bothers me is VPR speed refinement together with GI, where I feel it is quite a bit faster in 2015 (VPR refinement) different beast though...and despite me lowering as many values I can for rendering, it still iterates slower than the old GI in 2015, when it comes to final renders..that is a different matter that I havenīt looked in to that much.

The new volumetrics, at lower scale they look great when considering render speed, at larger real life scale of clouds, the balance between render speed and step quality isnīt acceptable, in fact the old legacy hypervoxels(within 2018) is blistering fast compared to the new volumetrics.
the old legacy hypervoxels have removed a barrier ( probably the new render changes, not hv changes) so you can overlap hypervoxels in a way with several nulls which you couldnīt do in 2015 without extreme lag.

Kryslin
03-26-2018, 11:38 AM
Paul:

While I can't match the 48 second render time, even I, rank amateur that I am, can get as good as the 2015 render in less than 2 minutes.

The key is samples (Light, Radiosity Rays, Reflection, Refraction, Subsurface Scattering) over AA passes, in nearly all cases. Also, don't use Adaptive AA. Set your number of passes and let it go (Adaptive AA increases render times by 10-15% in my recent experience). You start at 1 pass, and samples set at defaults, and then bring up the sample values where the noise is. Then, start bringing the AA passes up. The noise reduction wortkflow in the documentation does work well. There are exceptions to the samples before passes rule; High Frequency textures and Fiber FX are the two that come to mind.

kolby
03-26-2018, 12:56 PM
Here is my test.
LW 2015.3 - 26,5 sec.
140876

LW 2018.0.2 - 25,6 sec.
140875

Only disaster are fireflies. And I can't match that crazy transparent material on right sphere.

wesleycorgi
03-26-2018, 01:03 PM
Hi Kolby: can you share what you tweaked to get down to 25.6?

Morgan Nilsson
03-26-2018, 01:12 PM
Oh boy, there is so much wrong with the settings in the scene its no wonder you think the LW2018 renderer is crap...

First of all, camera samples, 10/20 adaptive AA with a threshold of 0.01? That is pure insanity. I never go above 8 as a maximum (and that is in scenes with an insane amount of edges that can cause stair-stepping artifacts, those settings has nothing to do with noise for me) and rarely needs to change minimum from 1.

Threshold I typically go around 0.05-0.08, but like all settings, it varies from a scene to scene basis.

Also, you get a ton of indirect diffuse noise with gi rays set to 1 only, around 6 is a decent value, and, if you have specific objects that suffer a lot from indirect diffuse noise, you can just go into that objects properties and force more gi rays on it to resolve the noise.

Camera samples should never, ever be the go to setting to remove noise in your scenes. Those should in 99/100 cases only be tweaked to fix issues with sharp edges that causes the stair-stepping artifacts.

rustythe1
03-26-2018, 01:36 PM
Yep, I changed camera, surfaces, (turn off ray back, glossy etc) turn off GI as its a point light in a closed box, GI has no effect, changed a few other settings and got it down to 3 seconds

kolby
03-26-2018, 01:40 PM
Hi Kolby: can you share what you tweaked to get down to 25.6?

updated contents
140877

wesleycorgi
03-26-2018, 01:43 PM
Thanks!

SBowie
03-26-2018, 01:50 PM
Paul, my friend, I want to say I feel your pain - but instead I have to say that I only look forward to feeling it. I haven't had time yet to start mounting the learning 2018 curve. I'll do so with a personal project sometime after NAB. Thankfully, I have an advantage, in that I can walk over and ask Jarrod and Matt stupid questions from time to time (since their desks are within 20 feet of mine.

This said - my impression, supported by a conversation we had at lunchtime, is that 2018 really does represent a quantum leap, albeit one that comes with a corresponding (one-time) learning curve. And I think helping people surmount that curve represents a real challenge for both NewTek and many of its longstanding LW customers (including you and me). What I've been confidently told is that some renders may indeed take a little longer, though not always - but that the payback in terms of render quality, time saved in surfacing setup, etc., more than make up for this. To paraphrase Jarrod, 'Once you get the hang of the PBR renderer, you will never consider going back'.

I'm happy you went to the trouble of posting this thread, and am sure the feedback will go a long way to helping me, and others.

Bitboy
03-26-2018, 01:58 PM
Paul, I've been doing a little interior scene and I'm pretty happy with LW 2018 so far. Still learning the new stuff, but here's my image which took ~20 minutes to render on an older i5 4430S @ 2.70 Ghz with 12 GB RAM. And I'm not very well versed in doing interiors at all, so for a first try in LW 2018 this is ok in my book. I'll probably adjust some of my settings with the advice people have been writing in this thread, so thanks for that guys :)

And even with that rendertime, I'll probably be able to bring it down a bit, since it's not fully optimized yet. I know there's a little noise in the image a few places and there's still stuff missing that I have to put in. All surfaces are Principled, no Standard materials at all. I've attached my settings and they're not that different than the standard settings, but I've adjusted the GI and Render settings some to get the quality where I wanted it.

So I don't get the complaints since I never had my interiors looking like this in 2015 with a decent render time, and that was on a faster machine as well. Just my 2 cents...

Rendered image:
140881

Render screenshot:
140885

LW screenshot:
140882

Camera settings:
140883

GI settings:
140884

Render settings:
140886

Matt
03-26-2018, 02:05 PM
There are so many things wrong with this comparison that simply are not helping. The scenes are not even close to a fair comparison.

First, you have glossy reflections on many surfaces in 2018 that are NOT even glossy, thus slowing it down unnecessarily, by comparison your 2015 has NO reflection blurring (glossy reflections equivalent)
AA settings are not even close to similar, they are set MUCH higher in your 2018 scene, because ...
You have Interpolated GI on in 2015, whereas you have Brute Force on in 2018, so you over-cranked the camera AA samples as a result
You have multiple bounces of GI in 2018 but only one in the 2015 scene

What I've found so far.

lardbros
03-26-2018, 02:57 PM
I've watched a bit of your video, and you seem like a good bloke, but after undoing a lot of issues with the file I did get it to render in around 30 seconds.

So, the major major issues I saw were things like colour space being off, GI being on for such a simple scene. The camera samples were really high... noise filter was also on, when it isn't that great for this scene really.


I turned GI off because it made naff all difference on this scene.
I personally think it looks nicer in 2018, but i prefer the physically correct lights and the much more realistic shading of 2018. I also made sure the ice ball had an index of refraction to it, which slows things down, but looks MUCH nicer.

I've attached the scene, so hopefully it'll help if you want to take it apart etc?

I understand your woes, but this is a completely and utterly different render engine, more akin to Arnold (think of films like Gravity) so it's not like it's a poor renderer, it's really not... it's v1 of Newtek's new renderer. They WILL NOT leave it here, that'd be silly. They'll continue to improve it, and make it even better. I do hope they manage to do some more to the non-photoreal side of things to assist you though.

Cheers,
Tim

PS, the attached image rendered in 30 seconds here.

MSherak
03-26-2018, 03:35 PM
Hi Folks.

There was a recent heated debate on the Lightwave Wiki Facebook page about the render engine in Lightwave 2018 vs Lightwave 2015. A moderator of that page has closed that thread. There is a lot of split opinions about the Lightwave 2018 renderer and if it's a worthy successor to 2015? After playing around with it for over two months, reading the online documentation from cover to cover, and watching many videos on YouTube about it, as of this point I'm not convinced. Every test I do in Lightwave 2018 is a hell of a lot slower than the same test in Lightwave 2015 and the render results are not as good. And I'm not a lone voice in the wilderness saying this.

Whenever this debate comes up, all I keep seeing from the pro-Lightwave 2018 group is "It's a new engine, you have to learn it's new ways." Well, folks, I've tried... I've tried damn hard to learn it. And all I see is an inferior engine in 2018. I decided to be fair and to try and pit Lightwave 2018 and Lightwave 2015 against each other on even terms, creating the same scene file in both packages, sorting out the render setting so they match, and texturing them the same too, and the results were jaw-dropping bad for Lightwave 2018. You can view the video on this here:

So the reason for this topic I'm posting here is, I want someone to show me, help me, to see that the Lightwave 2018 render engine is a worthy successor to Lightwave 2015! Tell me what I need to do shave 9 minutes off the 2018 render time, bringing it in par with 2015, producing a crisp, clean, beautiful render on par with Lightwave 2015 in around 46 odd seconds. I'm attaching both scene files below in the zip file. Show me where I'm going wrong, prove to me that Lightwave 2018 is a worthy successor, and I'll create a follow-up video highlighting the facts. Because right now, the render times and end results in Lightwave 2018 compared to Lightwave 2015 leave a lot to be desired...

For a fair test I adjusted the scenes and surfaces. So I took the settings from the 2018 scene and applied them to the 2015 for the GI and Camera. Then I adjusted the Colorspace from Linear to sRGB for both. (always use this mode and in 2018 change the picked and light to sRGB). Turned off 2018 Noise reduction since not needed for this test. I setup the materials in 2018 to match the 2015 ones. Now one thing most do not know about 2018 GI is that EVERYTHING is on even in Interp mode. This means if you are going to compare 2015 GI you have to turn on Use Transparency and Directional Rays. Can't turn these off in 2018. Once setup this is what I got from image 1 and 2. (also noticed the box is not modeled in the same spot. Anyway) Look at the times between the first two images attached. Also notice how they look close but very different. This is due to the materials not working the same way between the two even though the renderers are setup with equal GI, CS and Camera settings.

So I used my BSSRDF shader I am working on in 2015 to match the principled BSDF material in 2018. The third image is 2015 with this material. Now you can see that 2015 time is around 2018 with the correct type of materials. There are many things that 2018 will not be able to do that 2015 can with the right shader.

1) 2018 can't adjust reflections when the glossy reflections button is on. No controls since it's and on/off flag.
2) Bump maps do nothing but cause errors in 2018 since it's a surface (material) renderer. You have to use Normal maps.
3) 2018 Fireflies show up due to no control over double specular (reflections) bounce. Look at the chrome sphere in 2018.
4) 2018 transparent settings in the principled BSDF will not do caustics or shadow tinting correctly. Again since it's a surface renderer. Not a raytracer anymore.
5) 2018 shaders are not flexable. Have to stick to the rules of real-time and fixed materials. Yes means become familiar with Quixel and Algorithm systems.

There are many things different between 2015 and 2018 since the underlying renderer is not the same at all. 2015 can do more than 2018 with the same shaders (Materials). There is a little render time increase, less than 10%, but it's 99.99% accurate compared to 50% accurate. Yes principled BSDF works great, but it's made for real-time viewport since it has fixed rules. It's really is not made to replace a true raytracing engine. Main rule in principled BSDF shading system, one has to use normal maps. The bump map channel is now microsurfacing. 2018 bump maps are turned into a normal map at rendering but the CS is off internally, hence why it looks funny.

Personally it makes me sad to see the 2015 renderer die when it did not need to. Both the 2015 and the 2018 renderer could have been placed in 2018 and the materials could have worked with a unified panel from 2015. They did kill the best raytracing engine on the planet besides radiance. Only thing 2018 has over 2015 is the volumetic system and OpenGL. Course it's implementation for generating source context is lacking heavily for the volumetrics and the lack of substance and quixel inputs for opengl is disappointing. 2018 is going the route of everyone else in the industry and is great to learn if ones workflow is for real-time source generation. 2015 is better for more control and the vast amount of plugins available. But one has to really learn how lighting and surfacing work to get 2015 to render as easy as 2018. Here is a hint. Try plugging in a fresnel node into everything in 2015 and see 2018 show up. :) Happy Rendering.


PS. I have both but sticking with 2015 as my renderer. Oh by the way GI generation is the same in both 2018 and 2015. Only difference is Softness setting in 2018 hides the splotches.

By the way Lighting should really have the Inverse Distance^2 set. But did not want to adjust any of the lighting from the scene. This would be another issue in 2018 moving to LUX.

Tim Parsons
03-26-2018, 05:50 PM
Paul this rendered in 10sec. on my 8yr old i7! I watched your whole video and you repeated words like "exact same this and exact same that" and that is ultimately why you haven't grasped the new engine. It's different and you will need to learn new things. :) You make some fair points here and there but for the most part you just don't get it yet. You will, I'm sure. :) One thing I thought you might be on to was the inclusion of a noise filter, but I doubt that is an after thought of providing a poor renderer. I think noise reduction filters are an up and coming thing in rendering as this is just a first round. I personally don't use it as I think it adds more issues than it resolves. I do think what NT needs to address is either a HOT pixel filter or specialized sampling in the render engine to eliminate them. IMO, for a first pass, this new 2018 render and shading system is fantastic! Stick with it as I think your video has been thoroughly debunked by the responses so far. :thumbsup:

140900

Paul_Boland
03-26-2018, 08:11 PM
Hi Folks.

First off, thank you very much for all the very detailed analysis and feedback of the issues I was having with my scenes in Lightwave 2018 vs Lightwave 2015. I got feedback here, via PM, and on the YouTube comments too. I have to say it was a lot more than I expected to be honest and it is very much appreciated.

After following some of the suggestions, and doing more test renders, I was able to get my render in 2018 down from 9 minutes 48 seconds, to 4 minutes 23 seconds. While still a long way off from Lightwave 2015's 46 second render, it did indeed show me that my understanding of how the new render engine in 2018 works is obviously still not up to speed. That being the case, out of respect for any unintensional harshness that came across in my video on the issue (I said many times I was not taking a pot shot at Newtek or Lightwave), I have removed the video.

From all the feedback here, it seems that the new Lightwave 2018 render engine is indeed a totally different beast, with some saying it's aimed at more real-world rendering. I'm going to keep at it, keep trying to master it, to learn it, to get good render times on my eight year old system, and to try and wrap my head around my favour nightmare - the node editor! LOL!! I just hope that after being with Lightwave since version 8 - over a decade of use, that the new big change in the render engine in 2018 doesn't mark the point where my Lightwave journey ends and I'm left behind. Yes, I still have 2015, but trying to keep old software running forever is a tough battle, hence it's great to keep up to date.

So again, THANK YOU for all the detailed feedback and help, it is hugely appreciated.

samurai_x
03-26-2018, 08:24 PM
Lw 2018 renderer is good. But not for interiors or enclosed spaces at least with regards to render time.
There's no secret to lw 2018 renderer. Start with lowest samples possible and increase samples slowly until its enough to get rid of errors. Camera samples will be the last setting you increase. Easy to get good render in lw 2018.
But that doesn't mean its faster than lw 2015 :D The render time is generally pretty close, again except for interiors or enclosed spaces. Maybe we will see a cpu,gpu hybrid version of lw 2018 renderer, like Cycles, in the distant future. Then that will be faster.

papou
03-26-2018, 09:24 PM
I can understand Paul's feeling.
Lightwave could be an "out of the box" working toy for hobiest.
Tweak a bit... test here and there... render, et voilā: you have something. It's not very beautiful but it's fast.
If you learn a bit more, you can produce beautiful render.
I started like this.


Now, it's a much slower learning curve tool.
Hobiest are stopped there.
Artist who are not very technic are slowed by nodes.
For them nodes are somethings like "finish your render engine code yourself".
And there is no more Easy Texture Layer Option.

NT have dropped all these guys. And maybe it was the majority of users.

I work with lightwave since 5.6.
Lightwave is in my DNA.
Im able to go with 2018. But do i want that? not sure.
In fact i don't care about PBR. I like to mimic PBR with my tweaking skills. Do something fast that work. Maybe not totally realistic but the artistic way.
So maybe NT dropped people like me too.

That start to be a lot...

Rayek
03-26-2018, 10:58 PM
Paul this rendered in 10sec. on my 8yr old i7! I watched your whole video and you repeated words like "exact same this and exact same that" and that is ultimately why you haven't grasped the new engine. It's different and you will need to learn new things. :) You make some fair points here and there but for the most part you just don't get it yet. You will, I'm sure. :) One thing I thought you might be on to was the inclusion of a noise filter, but I doubt that is an after thought of providing a poor renderer. I think noise reduction filters are an up and coming thing in rendering as this is just a first round. I personally don't use it as I think it adds more issues than it resolves. I do think what NT needs to address is either a HOT pixel filter or specialized sampling in the render engine to eliminate them. IMO, for a first pass, this new 2018 render and shading system is fantastic! Stick with it as I think your video has been thoroughly debunked by the responses so far. :thumbsup:

140900

Certainly, 10 seconds and the quality leaves a lot to be desired. No proper reflections, noisy as heck, no anti-aliasing, hot pixels, visible banding, etcetera.

Come on, at this point I'd just switch to a real-time render engine (Unreal, Unity, Godot, Eevee, etc.) and get a high-resolution anti-aliased noiseless result in real-time and at a far higher quality as well. What's the point here? 10 seconds for an unusable render? If a game-like result is the aim, use a game render engine.

lardbros
03-27-2018, 04:29 AM
Hi Folks.

First off, thank you very much for all the very detailed analysis and feedback of the issues I was having with my scenes in Lightwave 2018 vs Lightwave 2015. I got feedback here, via PM, and on the YouTube comments too. I have to say it was a lot more than I expected to be honest and it is very much appreciated.

After following some of the suggestions, and doing more test renders, I was able to get my render in 2018 down from 9 minutes 48 seconds, to 4 minutes 23 seconds. While still a long way off from Lightwave 2015's 46 second render, it did indeed show me that my understanding of how the new render engine in 2018 works is obviously still not up to speed. That being the case, out of respect for any unintensional harshness that came across in my video on the issue (I said many times I was not taking a pot shot at Newtek or Lightwave), I have removed the video.

From all the feedback here, it seems that the new Lightwave 2018 render engine is indeed a totally different beast, with some saying it's aimed at more real-world rendering. I'm going to keep at it, keep trying to master it, to learn it, to get good render times on my eight year old system, and to try and wrap my head around my favour nightmare - the node editor! LOL!! I just hope that after being with Lightwave since version 8 - over a decade of use, that the new big change in the render engine in 2018 doesn't mark the point where my Lightwave journey ends and I'm left behind. Yes, I still have 2015, but trying to keep old software running forever is a tough battle, hence it's great to keep up to date.

So again, THANK YOU for all the detailed feedback and help, it is hugely appreciated.



With the new renderer it almost is going back to re-learning everything you knew or thought you knew, and starting again.

- Reflection doesn't really exist any longer (well, it does, but not in its old 2015 form). Where you see reflection settings in Material nodes in 2018, it is only pure mirror reflections. These cannot be blurred.
- Wherever you see Specular, this is now reflections as they were in 2015. You dial this up to control the amount of spec/reflections. On the Principled shader though, you will only see reflections increase realistically. So, you'll get brighter reflections on the edges. This shader won't work for everything, as indeed it probably won't work great for non-photoreal stuff, and some stylised things too.
- You control the blurriness of the specular reflections using either Roughness in the Principled BSDF shader, or Glossiness in the Standard shader.

With these new ideas it can be weird and confusing... but now I can't think any other way, and it makes so much more sense when doing product shots or realistic lighting.

Keep at it though! Don't give up!

samurai_x
03-27-2018, 05:43 AM
A very good explanation of the Principled BSDF. The parameters are similar so it should be applicable to lightwave.

https://www.youtube.com/watch?v=4H5W6C_Mbck

Axis3d
03-27-2018, 07:01 AM
I do kind of miss the Unified Sampling implementation that 2015 had. It was one place to go to globally take care of grain and noise adjustments. In 2018, there seems like there are many places to track down the grain issues (individual lights, surfaces, etc.).

Perhaps, something like this can be implemented in 2018?

Tim Parsons
03-27-2018, 07:13 AM
Certainly, 10 seconds and the quality leaves a lot to be desired. No proper reflections, noisy as heck, no anti-aliasing, hot pixels, visible banding, etcetera.

Take it in the context of Paul's video. He wasn't getting anything close to that in under 7min. Paul's scene is from 1991, nothing about it will ever look good. :)

kolby
03-27-2018, 07:29 AM
... You start at 1 pass, and samples set at defaults, and then bring up the sample values where the noise is. Then, start bringing the AA passes up.

I think this is wrong procedure because the AA samples (or passes) works as a multiplier for other sampling. So if you set all shading samples to have clean result while your AA is 1 and then raise AA to 6, you will multiply your shading samples by 6 without any benefit because it was clean even at AA 1 and you get about 6 times longer rendertime.
So I think the right procedure is firstly set the correct AA values to have clean edges, textures, DOF or MB and then raise carefully all sampling where needed to clean out remaining noise.

rdolishny
03-27-2018, 08:53 AM
Please consider reactivating your YouTube video for context. I never saw it.

Kryslin
03-27-2018, 11:33 AM
I think this is wrong procedure because the AA samples (or passes) works as a multiplier for other sampling. So if you set all shading samples to have clean result while your AA is 1 and then raise AA to 6, you will multiply your shading samples by 6 without any benefit because it was clean even at AA 1 and you get about 6 times longer rendertime.
So I think the right procedure is firstly set the correct AA values to have clean edges, textures, DOF or MB and then raise carefully all sampling where needed to clean out remaining noise.

What I forgot to mention is that AA passes go up, you can start reducing your samples, trying to keep the same overall amount - if 100 samples at 1 camera pass gives you a clean result, then 25 will work with 4 passes.
To make things go quicker, you'll need to enable the buffers they use in the noise reduction workflow described in the docs.

What I usually go with :
1) Is there a lot of reflection / refraction / SSS? Add more of the appropriate sample.
1a) Noisy surfaces without any reflection / refraction / SSS are taken care of by light samples,
2) Do you have an environment light? Double / Triple / Quadruple the number of samples needed.
3) Do you have any high frequency textures (like a woven mat, or Fiber FX)? More AA passes.

I've also gotten a 30s render, with the same quality of the 2015 render.

Chris S. (Fez)
03-27-2018, 12:50 PM
Please consider reactivating your YouTube video for context. I never saw it.

I disagree. The video was plain wrong and bad for business. Some bad Newtek press is warranted. Not this time. This is mostly a matter of learning curve.

prometheus
03-27-2018, 02:14 PM
How do you guys feel about the new skin material?
I think it looks great to some degree, especially the reflections, fresnel ..and may render quite a bit faster than the old skin material, however...
I miss the multiple scattering, which allowed for quite easy setup and get red ears, nostrils and fingers ..I know too little about the new skin material to reach the same effect with that kind of ease, not sure what in the new material replaced multiple scattering
or perhaps I just missed to sync the samples in the rendering settings.

roboman
03-27-2018, 09:19 PM
I took a model from a video game and have been working on texturing it in lightwave. The skin is the first thing I managed to do to a degree that I liked it. It's much better then what I was able to do in older ver. A uv texture map in one or more of the sub layers is how I got the coloring to the parts that needed it. Lots of texture maps out there for daz and poser models, smudge out some of the detail and they make a nice sub surface coloring . The patent leather shoes also look great. Most of the rest still really sucks, but working through the suck to what is good is part of learning, right :) Nodes are very slowly starting to sink in. I never touched nodes before and kind of avoided pbr systems before. I think I might come to like the pbr. Not sure I'll come to really like nodes....

vipvip242
03-28-2018, 02:14 AM
This kind of topic is just a non-sense : serioulsy, we have to use this kind of software not to render 'spheres in a box' (and/or benchmark seconds less or more) but real elaborated pictures : and the new LW 2018 render engine can archive results, very nicely, which we can't have before...

samurai_x
03-29-2018, 09:07 PM
So I used my BSSRDF shader I am working on in 2015 to match the principled BSDF material in 2018. The third image is 2015 with this material. Now you can see that 2015 time is around 2018 with the correct type of materials. There are many things that 2018 will not be able to do that 2015 can with the right shader.

1) 2018 can't adjust reflections when the glossy reflections button is on. No controls since it's and on/off flag.
2) Bump maps do nothing but cause errors in 2018 since it's a surface (material) renderer. You have to use Normal maps.
3) 2018 Fireflies show up due to no control over double specular (reflections) bounce. Look at the chrome sphere in 2018.
4) 2018 transparent settings in the principled BSDF will not do caustics or shadow tinting correctly. Again since it's a surface renderer. Not a raytracer anymore.
5) 2018 shaders are not flexable. Have to stick to the rules of real-time and fixed materials. Yes means become familiar with Quixel and Algorithm systems.

There are many things different between 2015 and 2018 since the underlying renderer is not the same at all. 2015 can do more than 2018 with the same shaders (Materials). There is a little render time increase, less than 10%, but it's 99.99% accurate compared to 50% accurate. Yes principled BSDF works great, but it's made for real-time viewport since it has fixed rules. It's really is not made to replace a true raytracing engine. Main rule in principled BSDF shading system, one has to use normal maps. The bump map channel is now microsurfacing. 2018 bump maps are turned into a normal map at rendering but the CS is off internally, hence why it looks funny.

Personally it makes me sad to see the 2015 renderer die when it did not need to. Both the 2015 and the 2018 renderer could have been placed in 2018 and the materials could have worked with a unified panel from 2015. They did kill the best raytracing engine on the planet besides radiance. Only thing 2018 has over 2015 is the volumetic system and OpenGL. Course it's implementation for generating source context is lacking heavily for the volumetrics and the lack of substance and quixel inputs for opengl is disappointing. 2018 is going the route of everyone else in the industry and is great to learn if ones workflow is for real-time source generation. 2015 is better for more control and the vast amount of plugins available. But one has to really learn how lighting and surfacing work to get 2015 to render as easy as 2018. Here is a hint. Try plugging in a fresnel node into everything in 2015 and see 2018 show up. :) Happy Rendering.


PS. I have both but sticking with 2015 as my renderer. Oh by the way GI generation is the same in both 2018 and 2015. Only difference is Softness setting in 2018 hides the splotches.

By the way Lighting should really have the Inverse Distance^2 set. But did not want to adjust any of the lighting from the scene. This would be another issue in 2018 moving to LUX.

Lots of info here.
So first, is your shader for lw 2015 publicly available?

Its too bad they did kill the best bang for the buck renderer in lw 2015.
Maybe its hidden as a cheat code in lw 2018. There was no reason to remove it. Other appz have multiple render engines in them so people that know how to use them can still use them.

Tim Parsons
03-29-2018, 09:55 PM
I disagree. The video was plain wrong and bad for business. Some bad Newtek press is warranted. Not this time. This is mostly a matter of learning curve.

And it showed great courtesy on Paul's part to take it down. Kudos to Paul.

JohnMarchant
03-30-2018, 04:32 AM
I have to be honest and say that comparisons between rendering on 2015 and 2018 are pretty pointless. One maybe slower than the other in a given circumstance, but its not like we can compare like for like because its not. Even trying to set up a scene to do a comparison is hard because you cant use the same scene and have to use work arounds to get the same render.

Yes its a new renderer and a new learning curve, and indeed yes you can make complaints about this and that in 2018 renderer, however you cant really make a fair comparison between the two and indeed why do you have to. You can run any flavour of LW you like.

I have still not used 2018 Modeler because to be honest there is not that much difference to me from 2015, well not enough to warrent the risk that allot of my thrid party plugins/scripts not working with 2018.

prometheus
03-30-2018, 02:00 PM
I have still not used 2018 Modeler because to be honest there is not that much difference to me from 2015, well not enough to warrent the risk that allot of my thrid party plugins/scripts not working with 2018.

for spline bridge I prefer the artsphere plugin instead of the new one natively.
and in camera modeling I havenīt really used, though I figure it is quite nice to have when that is needed.

But I like the new Live array tool, that one is nice....though I can do arrays in blender parametricly and change at any time, including making them operate as boolean tools also, parametric and booleans which I canīt performe with lightwave...then again the array tool modifier in blender doesnīt have that many axis to array on as Lightwave has, you would have to add another modifier on dedicated axis for full 3d array...thereīs always something isnīt it.

Thereīs still too little done with modeling tools in modeler, and layout ofcourse..so it wil not be any upgrade this time until I see where that goes.

jaxtone
03-30-2018, 10:16 PM
This was a really good question/input! How come Newtek didn't allow multiple render engines/methods?


Lots of info here.
So first, is your shader for lw 2015 publicly available?

Its too bad they did kill the best bang for the buck renderer in lw 2015.
Maybe its hidden as a cheat code in lw 2018. There was no reason to remove it. Other appz have multiple render engines in them so people that know how to use them can still use them.

cresshead
03-31-2018, 12:36 AM
drat, the video has been pulled

so did you resolve your issues?
are you using 2018 or sticking with 2015 render for now?

JohnMarchant
03-31-2018, 03:25 AM
for spline bridge I prefer the artsphere plugin instead of the new one natively.
and in camera modeling I havenīt really used, though I figure it is quite nice to have when that is needed.

But I like the new Live array tool, that one is nice....though I can do arrays in blender parametricly and change at any time, including making them operate as boolean tools also, parametric and booleans which I canīt performe with lightwave...then again the array tool modifier in blender doesnīt have that many axis to array on as Lightwave has, you would have to add another modifier on dedicated axis for full 3d array...thereīs always something isnīt it.

Thereīs still too little done with modeling tools in modeler, and layout ofcourse..so it wil not be any upgrade this time until I see where that goes.

Agreed for Modeling i have 2015 Modeler set up how i like it and with all the tools i need working. Not enough reason yet to model in 2018, but hope springs eternal. In camera modeling and texturing ive not used because there is no point for me. If i have to set up a scene and such then i can do that in layout and texture in layout.

Greenlaw
03-31-2018, 12:18 PM
While there's not a lot of differences between 2015 and 2018's modelers, the biggest reason to use 2018 for modeling is that the surfaces will be compatible with 2018's rendering system. If you go between Layout and Modeler a lot, this feature is criticial.

(Well, unless you have OD Tools. It has a plugin that lets you keep both 2015 and 2018 compatible surfaces embedded in the object.)

Being able to use Layout View in 2018 Modeler is pretty nice. I haven't used this feaure in production yet but it should be useful when I need to do camera projection mapping for VFX work.

jwiede
03-31-2018, 07:51 PM
While there's not a lot of differences between 2015 and 2018's modelers, the biggest reason to use 2018 for modeling is that the surfaces will be compatible with 2018's rendering system. If you go between Layout and Modeler a lot, this feature is criticial.

Except that previewing surfaces in Modeler 2018 doesn't work. For any surface attributes that aren't visible via OpenGL, you're now "surfacing blind" in Modeler (2018). That's a pretty serious limitation, IMO.

While surfacing in Modeler previously had limitations, the surface preview was still very useful for catching/correcting a lot of surface issues. It also allowed users to obtain a reasonable surface preview without requiring maximal OpenGL settings -- thereby allowing use of more performance-optimal OpenGL settings for Modeler viewports.

Chris S. (Fez)
03-31-2018, 07:57 PM
After all this time. Use your archive of surfaces. Or use the Hub. Not a big deal IMHO.

Chris S. (Fez)
03-31-2018, 08:12 PM
140964

jwiede
03-31-2018, 08:32 PM
After all this time. Use your archive of surfaces. Or use the Hub. Not a big deal IMHO.

:rolleyes:

Because "Do without" has worked so very well as a selling point for Lightwave to date?

ROFL!

Chris S. (Fez)
03-31-2018, 09:39 PM
:rolleyes:

Because "Do without" has worked so very well as a selling point for Lightwave to date?

ROFL!

Pick your ******* battles and get it done. We are addressing an inconvience...a small one...not a limitation.

Paul_Boland
04-01-2018, 09:53 PM
Hi Folks.

Thanks for all the additional feedback, info, and thoughts and questions added here, appreciated. Let me just give an update here and try and address what was said...

"There's no secret to lw 2018 renderer. Start with lowest samples possible and increase samples slowly until its enough to get rid of errors. Camera samples will be the last setting you increase. Easy to get good render in lw 2018."

Thanks for the info. I'm taking this approach on board and working my way through it.

"Now, it's a much slower learning curve tool.
Hobiest are stopped there.
Artist who are not very technic are slowed by nodes.
For them nodes are somethings like "finish your render engine code yourself".
And there is no more Easy Texture Layer Option.

NT have dropped all these guys. And maybe it was the majority of users."

I found these comments very powerful and speaking to me. This describes me. I worked my way from Lightwave 8 to 2015 with no real issues. But the new renderer in Lightwave 2018, and in particular the node editor, has be lost. I'm try to learn this, I'm reading the forums, watching videos online, doing lots of testing, but right now I feel like a line was drawn in the sand when Lightwave 2018 came out and I have been left behind it. I still can't understand why it's so complicated now to get great renders when Lightwave 2015 and prior was simple and easy to use for awesome results...?!?

"Keep at it though! Don't give up"

Thanks! I'm keeping it at, not ready to give up by a long shot.

"Paul's scene is from 1991, nothing about it will ever look good."

Kind of a curious comment... If Lightwave 2018 can't produce beautiful renders quickly on the simple ("1991") scenes, what does that mean for more complex stuff???

"Please consider reactivating your YouTube video for context. I never saw it."

Sorry, I deleted it and also deleted it from my hard drive. Given that the replies here showed me that I still didn't fully understand the workings of the new engine, it made my video unfair so I took it down.

"This kind of topic is just a non-sense : serioulsy, we have to use this kind of software not to render 'spheres in a box' (and/or benchmark seconds less or more) but real elaborated pictures : and the new LW 2018 render engine can archive results, very nicely, which we can't have before..."

I have to totally disagree with this post. Why were my questions, my issues, viewed as nonsense? If I want to render an enclosed room with boxes and spheres in it, why should my needs not be met just because the engine should be used for more elaborate scenes? Lightwave 2015 can do it and it can produce beautiful and awesome renders so I don't understand where you're statement is coming from???

"so did you resolve your issues?
are you using 2018 or sticking with 2015 render for now?"

Right now, my learning of Lightwave 2018 is still work in progress. Over the past few days I did some artwork for the cover of my next novel but I did it in Lightwave 2015 just because I know the engine and how to get the results I want. I am all for moving fully into 2018 but until I understand its in's and out's it currently remains a piece of software I'm trying to learn.

* * * * * * * * * *

Right now Lightwave 2015 is still my software to go to for 3D work. I'm trying, as stated above, to learn the new 2018 engine, but comments such as those above saying the new engine can't cope with basic scenes don't fill me with confidence. I'm not a professional 3D artist, I'm a hobbyist, and if I want to do some simple scene, and Lightwave 2018 can't handle it because it's designed for more complex stuff, then I fear I could indeed be left behind here. I don't want to be left behind, I don't want to be left using Lightwave 2015 as Lightwave 2018 and its future versions put a bigger and bigger chasm in from of me for rendering and bringing my imagination to life.

And then what happens a few years down the road, should Windows evolve to a point where Lightwave 2015 no longer works... Then I'm lost! I've been there before and it's not a nice place to be in.

As said, I appreciate all the additional feedback, thanks. I'm working on Lightwave 2018 in my spare time, trying to get the grip of it, but I do have to agree with the comments above about the ease-of-use of Lightwave being lost in this new version. 2015 produces crisp clean renders with ease. 2018 requires a lot of tracking down settings, making adjustments, doing test render after test render after test render to see if you have a good render now... That's a sad state of affairs... :(.

Right now, I feel left out in the cold by Lightwave 2018. The over-complexity of the render engine is not easy to use. And the node editor interface is a totally nightmare for me. I'm still trying to understand that beast and why it's infiltrated so many areas of Lightwave in this single upgrade (node editor for making lighting adjustments just seems crazy!!). As I said I'm trying, I'm trying hard to get back into the warmth of Lightwave in 2018, but it's a tough task.

I'll keep you posted...

Kryslin
04-01-2018, 10:25 PM
If you're having problems with nodes, you can simply add a layers node to the input, and texture away - just be sure to use the right layers node for the input (color, scalar, or bump) I'm just the opposite, I could never understand the old texturing system, but was creating stuff with nodes almost instantly.

CaptainMarlowe
04-02-2018, 12:37 AM
Paul, I'd like to react to some of your comments on other people comments.
I have been a hobbyist for years now, and I has always preferred nodes to layered textures. I had been accustomed to nodes even before joining LW as Vue was using them a lot, for instance. So no, all hobbyist won't be driven away by nodes.
Then you say, if simple scenes take forever to render, what will it be with complex ones ? I'd say if you have optimised your assets and scenes from the beginning for LW 2018, the more complex the scene, the easier it handles it. But that's only my own experience, of course, as I render a lot of exteriors with thousands of instances almost of the time.

pinkmouse
04-02-2018, 01:14 AM
...And the node editor interface is a totally nightmare for me. I'm still trying to understand that beast and why it's infiltrated so many areas of Lightwave in this single upgrade (node editor for making lighting adjustments just seems crazy!!)...

Paul, it may not be any consolation, but one of the reasons I haven't upgraded is that there aren't enough node interfaces yet! :D

Tobian
04-02-2018, 06:21 AM
I'm not interested in a public fight, with people who are hilariously, laughably wrong, or kicking a guy when he's down, and to be clear Paul, I'm putting you in the latter camp, so suffice it to say if you want help, ask in the right way, and ask in the right place, and this is not the right way or the right place, as it's just full of trolls and idiots, which is why I never spend any time here any more.

Paul if you need assistance, just ask, I can be reached on.. not this forum ;)

No offence but part of the reason you are really struggling, render times aside, is you have not moved on from old render technology paradigms, and have not caught up with the newer techniques and workflows, so the mind-shift you have to undergo is huge. Meanwhile, every rendering platform in the industry has already moved this way for either years or just now also. I moved my workflows over quite a few years ago, the difference though, is now, because it's designed from an ease of use perspective, I now have *LESS* work to do, not more, and that's the whole point of this paradigm shift. It's not to make your life harder, it's to make it easier. You're just fighting it because you're trying to do it the old way. The reason why it wasn't done the right way before is simply computers were too slow and didn't have the spare resources to deal with it. With careful optimisations and faster computers, this is no longer the case, and hence everyone is moving over.

I'd love to do more videos and help guides to explain it all, but I have a full time job now, and they are a lot of work. I'd love Newtek to do more, but if the engineers take the time to do this, they take it away from actually making the software, and well, the marketing department, or whatever you want to call it is currently not there. I'm not happy about it, but it is what it is. Endless complaining doesn't fix it, just find people who are going to help you, not just ***** about stuff they are also ignorant about.

SBowie
04-02-2018, 07:13 AM
... and this is not the right way or the right place, as it's just full of trolls and idiots, which is why I never spend any time here any more.Based on the comments I get and my own observations, most agree that (after the drama of last year) there has been a marked improvement in tone and discourse here in 2018. This remark is not particularly helpful (the rest arguably was).

Buckshot characterizations of this type reduce you to the level you are decrying, being at once unnecessarily provocative, overstated, and offensive to many who are here to make positive contributions (just as many have done in this specific case).

samurai_x
04-02-2018, 07:47 AM
I have a full time job now, and they are a lot of work.

Congratulations! :D
Just trolling.

Paul_Boland
04-02-2018, 02:29 PM
Thanks, folks, for the continued input. Just to address what was said above...

"If you're having problems with nodes, you can simply add a layers node to the input, and texture away - just be sure to use the right layers node for the input (color, scalar, or bump) I'm just the opposite, I could never understand the old texturing system, but was creating stuff with nodes almost instantly."

LOL!! Different strokes for different folks. I agree, with the node editor popping up in loads of software, it obviously makes sense to lots of people. For me, it's a chaotic jigsaw puzzle to be wired up just to apply a texture to an object or adjust a light property. I have, and am, making progress with it. But for me, it doesn't come easy, it doesn't make a lot of sense. But thanks.

"Paul, I'd like to react to some of your comments on other people comments.
I have been a hobbyist for years now, and I has always preferred nodes to layered textures. I had been accustomed to nodes even before joining LW as Vue was using them a lot, for instance. So no, all hobbyist won't be driven away by nodes."

I never said all hobbyists would be driven away by the node editor. I said the ease of use has gone from Lightwave with this over complex interface and that could drive some away. But as the post over shows, some people love the node editor, it makes perfect sense to them. Unfortunately for me, it doesn't so I'm currently battling against the new interface to achieve what was easy to do in the old.

"Then you say, if simple scenes take forever to render, what will it be with complex ones ? I'd say if you have optimised your assets and scenes from the beginning for LW 2018, the more complex the scene, the easier it handles it. But that's only my own experience, of course, as I render a lot of exteriors with thousands of instances almost of the time."

Again you misread what I said. What I said was, if Lightwave 2018 isn't designed for handling simple scenes, then what does that mean for complex ones? I was responding to the post that called my issues nonsense. I think if someone is having an issue with something, even if it's trivial to someone else who views it as a non-issue, it still doesn't mean it's a non-issue to the person having it. I'm having issues with the new 2018 render engine and I'm getting a lot of feedback here to help with it. And just to address what you said there about optimising my assets and scenes in 2018... I never had this problem in 2015. I never had a problem with my assets and scenes in Lightwave until I bought Lightwave 2018 and the totally new render engine and interface threw me off the tracks. LOL!! To use your words, the fact that I have to "optimise" my assets and scenes in order to get a good render, is an added level of complexity that wasn't there before.

"Paul, it may not be any consolation, but one of the reasons I haven't upgraded is that there aren't enough node interfaces yet!"

LOL!! Cheers :)!!

"Paul if you need assistance, just ask, I can be reached on.. not this forum

No offence but part of the reason you are really struggling, render times aside, is you have not moved on from old render technology paradigms, and have not caught up with the newer techniques and workflows, so the mind-shift you have to undergo is huge. Meanwhile, every rendering platform in the industry has already moved this way for either years or just now also. I moved my workflows over quite a few years ago, the difference though, is now, because it's designed from an ease of use perspective, I now have *LESS* work to do, not more, and that's the whole point of this paradigm shift. It's not to make your life harder, it's to make it easier. You're just fighting it because you're trying to do it the old way. The reason why it wasn't done the right way before is simply computers were too slow and didn't have the spare resources to deal with it. With careful optimisations and faster computers, this is no longer the case, and hence everyone is moving over."

Thanks for the feedback and thoughts on the matter at hand. As I said, I'm trying to grasp the new way to work in 2018, I really truly am. And just to say, I thought the feedback I got overall here was very good, I do view the community here as helpful and friendly. Thanks again.

Tobian
04-02-2018, 04:13 PM
Based on the comments I get and my own observations, most agree that (after the drama of last year) there has been a marked improvement in tone and discourse here in 2018. This remark is not particularly helpful (the rest arguably was).

Buckshot characterizations of this type reduce you to the level you are decrying, being at once unnecessarily provocative, overstated, and offensive to many who are here to make positive contributions (just as many have done in this specific case).

If you require me to name specific individuals who can go **** themselves, I can, but my understanding is that personal insults and attacks are not the done thing on forums. Not sure who these 'most' you refer too are, but certainly no one I know. I'll go back to my corner of the internet and leave you be then.

SBowie
04-02-2018, 05:17 PM
If you require me to name specific individuals who can go **** themselves, I can, but my understanding is that personal insults and attacks are not the done thing on forums..True, this would violate forum policy. The absence of a rule explicitly prohibiting more generalized insults, however, should not be taken as meaning it is not equally unwelcome.


I'll go back to my corner of the internet and leave you be then. Probably best for all concerned, thanks.

IgnusFast
04-02-2018, 06:38 PM
I'm actually pretty impressed with the new renderer so far. I followed some of the recommendations here for ray settings, etc. Even so, the render times are about the same. But I really like the look of the reflections...

Left is 2015, Right is 2018. Direct scene import/conversion, no work done at all. Some of the skin surfaces are too harsh, but the glass is gorgeous.

140995
140994
140996
140997
140998

Paul_Boland
04-02-2018, 07:22 PM
141000

So following info on the thread here, PM's, and other feedback from folks, I'm here. I can tell you now that the ice ball is totally the new Principle surface in 2018 and as you can see I figured out reflectiveness too. Please ignore the render time, I was following the online documentation settings and this is what I got!! LOL!!

Two questions...

1. It says to remove the fireflies to increate the light samples but I can't for the life of me figure out how to do that. I got the impression from the online documentation that object surfaces themselves can have additional light samples applied to them but how do you do it? Is it the volumetric light sample because I have volumetrics turned off?

2. I still can't figure out how to make a surface glossy like the ice ball in the 2015 render. I have tried messing around with Sheen and Clear Coating Gloss and still no joy. How do I make a surface glossy so it gets that light glossy element on it?

Nicolas Jordan
04-02-2018, 09:02 PM
141000

So following info on the thread here, PM's, and other feedback from folks, I'm here. I can tell you now that the ice ball is totally the new Principle surface in 2018 and as you can see I figured out reflectiveness too. Please ignore the render time, I was following the online documentation settings and this is what I got!! LOL!!

Two questions...

1. It says to remove the fireflies to increate the light samples but I can't for the life of me figure out how to do that. I got the impression from the online documentation that object surfaces themselves can have additional light samples applied to them but how do you do it? Is it the volumetric light sample because I have volumetrics turned off?

2. I still can't figure out how to make a surface glossy like the ice ball in the 2015 render. I have tried messing around with Sheen and Clear Coating Gloss and still no joy. How do I make a surface glossy so it gets that light glossy element on it?

To try and get rid of fireflies you have to increase your reflection samples if you have glossy reflections(blurry reflections) turned on. Increasing light samples will reduce the shading/shadow noise in a scene. I have found I need to use about the same amount of samples I used in LW 2015 to get good results. Also increase the roughness to get more spread out specularity.

samurai_x
04-03-2018, 08:12 AM
A very good explanation of the Principled BSDF. The parameters are similar so it should be applicable to lightwave.

https://www.youtube.com/watch?v=4H5W6C_Mbck

Paul, watch the video I posted earlier with the blender principle bsdf. The video goes through each parameter in the pbsdf shader.
Its the same as in lightwave 2018 just its GPU powered renderer.
Sheen has nothing to do with reflection or making things look glossy.

Paul_Boland
04-03-2018, 02:55 PM
Latest render:
141013
Got the render time down to just under five minutes. Can't get out the fireflies. I've up-ed the reflection rays to as high as 50 and can't clear them from the ice ball or reflection of the ceiling. I'm working on it...

Watching the video now, thanks!

rustythe1
04-03-2018, 03:12 PM
141000

So following info on the thread here, PM's, and other feedback from folks, I'm here. I can tell you now that the ice ball is totally the new Principle surface in 2018 and as you can see I figured out reflectiveness too. Please ignore the render time, I was following the online documentation settings and this is what I got!! LOL!!

Two questions...

1. It says to remove the fireflies to increate the light samples but I can't for the life of me figure out how to do that. I got the impression from the online documentation that object surfaces themselves can have additional light samples applied to them but how do you do it? Is it the volumetric light sample because I have volumetrics turned off?

2. I still can't figure out how to make a surface glossy like the ice ball in the 2015 render. I have tried messing around with Sheen and Clear Coating Gloss and still no joy. How do I make a surface glossy so it gets that light glossy element on it?

you need to drop your roughness as that's whats making the spec spread, make sure glossy reflections is ticked on the second tab, , image 1 here is standard materials, image 2 is bsdf with 100%spec 30%rough, image 3, you get a better spec hit with the delta material though, it can also do transparency, this is with the spec at 50% (each of these only took 5 seconds to render by the way) the reason you don't get as much spec with the bsdf is because the spec hit is not physically correct to the material, so it will never look exactly the same, as also bump maps do not function correctly with PBR, you need to try and create normal maps.
141014141015141016

rustythe1
04-03-2018, 03:18 PM
Latest render:
141013
Got the render time down to just under five minutes. Can't get out the fireflies. I've up-ed the reflection rays to as high as 50 and can't clear them from the ice ball or reflection of the ceiling. I'm working on it...

Watching the video now, thanks!

the fire flies in the ice ball are caused by refraction, either increase that, or drop the refraction to 1.1 or less (try dropping the refraction and you will see them magically vanish), the two fireflies in the ceiling are not fireflies, they are the repeat reflections of the point light and will never vanish unless you turn off glossy reflections

wesleycorgi
04-03-2018, 04:07 PM
Paul, watch the video I posted earlier with the blender principle bsdf. The video goes through each parameter in the pbsdf shader.
Its the same as in lightwave 2018 just its GPU powered renderer.
Sheen has nothing to do with reflection or making things look glossy.

Thanks for sharing this (again); very helpful.

Paul_Boland
04-03-2018, 07:16 PM
THANK YOU ALL very much for all the help here. As you can see from the latest render here, success!! Damn, I have to say this made me smile when I saw it :)!! I've definitely learned a lot over the course of this thread and I very much appreciate all the help. I won't say I'm ready yet to jump ship on 2015 and move fully into 2018, but this is definitely a step in the right direction. Still have a lot to learn, will probably be asking more questions in the future, but I'm pleased with what I got here :)!!

Thanks, Folks, much appreciated!!

141023

Kryslin
04-03-2018, 11:08 PM
Paul : To get your ice ball to look closer to the 2015 version, hook up the color output of your image node to the luminous color input. Then, raise the luminosity value to 20% - 30%.

Raise your reflection samples to get rid of most of the fireflies, and kill the reflectivity of any surface that goes not absolutely require it. Render times will drop, and fireflies will go away.

rustythe1
04-04-2018, 05:54 AM
those fire flies are not reflection sample problems, they are just reflections on the low poly geometry and refractions through the sphere, so although annoying they are supposed to be there so you will only get rid of them by creating better geometry and getting a balance of light brightness (that's why the scene is such a bad example because its inside an enclosed space) if this was out side it would be far easier to get fast clean results, take this for example, its 4k, has depth of field, and yet it rendered in less than 15 mins on my 5 year old i7!
141030141031

Kryslin
04-04-2018, 10:33 AM
rustythe1: In the time it took me to write that response(I'm typing 1 handed, because of a neck injury that affects my left arm), there were three more replies... Or, there were three or more replies after the message I was answering. More, smoother geometry (my version of the scene has 2 subd quad spheres @ 4) does not get rid of those 2 last reflection highlights. removing reflection from those surfaces that do not need it (or reducing it to extremely low amounts) does.,if they're not wanted.

Also, good looking render. Right side looks more like a photo than the left. I have yet to venture into 4k renders, mainly because the bulk of my work is in FFX, and my7 year old i7 would take hours rendering FFX at that resolution...

rustythe1
04-04-2018, 01:41 PM
Thanks, even more so if you add some post filters, and use real lens camera, (bit silly though, spend ages removing noise, then add it back to make it look photographic!!!!)
141032

prometheus
04-04-2018, 04:12 PM
Hereīs some VFX insight with Ben Snow interviewed..a bit of an old article though.
The section around Energy Conservation(almost in the end of the page) may help understanding why Newtek/lightwave opted to go for it.
https://www.fxguide.com/featured/ben-snow-the-evolution-of-ilm-lighting-tools/

Marander
04-05-2018, 03:27 AM
Thanks, even more so if you add some post filters, and use real lens camera, (bit silly though, spend ages removing noise, then add it back to make it look photographic!!!!)
141032

Rustythe, it's a good render but what I notice in every LW2018 render are the unrealistic plastic looking metals / conductor materials. In your image the car and the drainage, that ruins the picture for me. Also the reflections on the flower pots don't look right to me. Not your fault (you did amazing models and renders in the past) but an issue in the 2018 shading to me.

Qexit
04-05-2018, 05:40 AM
Rustythe, it's a good render but what I notice in every LW2018 render are the unrealistic plastic looking metals / conductor materials. In your image the car and the drainage, that ruins the picture for me. Also the reflections on the flower pots don't look right to me. Not your fault (you did amazing models and renders in the past) but an issue in the 2018 shading to me.Hm, I suspect the drainpipes and guttering are supposed to look like plastic as that is the material that would be used for them in real life. Though I have to agree that the flowerpot reflections seem a little off. Maybe a little bright for the foggy look to the rest of the scene ?

Schwyhart
04-05-2018, 08:25 AM
Hereīs some VFX insight with Ben Snow interviewed..a bit of an old article though.
The section around Energy Conservation(almost in the end of the page) may help understanding why Newtek/lightwave opted to go for it.
https://www.fxguide.com/featured/ben-snow-the-evolution-of-ilm-lighting-tools/

Very interesting article, but the grammar is really bad. It gets worse the further you get into the article, so the editor obviously got lazy.
Sorry, I know this comment is off topic, but it just annoys me when grammar in articles is this bad.

raymondtrace
04-05-2018, 08:59 AM
Very interesting article, but the grammar is really bad. It gets worse the further you get into the article, so the editor obviously got lazy.
This is an "interpretation" of a podcast interview, so you might be reading casual speech...or poor transcription. The link to the original podcast audio is at the end.

Schwyhart
04-05-2018, 10:19 AM
That makes sense. I've read a lot of FX Guide articles and this one in particular was kind of hard to read in some parts.
Carry on :)

wyattharris
04-05-2018, 12:59 PM
Hm, I suspect the drainpipes and guttering are supposed to look like plastic as that is the material that would be used for them in real life. Though I have to agree that the flowerpot reflections seem a little off. Maybe a little bright for the foggy look to the rest of the scene ?

The pots also look like they are reflecting something like studio lighting or whatever is on the camera side of the scene. Light panels as opposed to the car. Am I seeing that right?

jaxtone
05-01-2018, 12:52 PM
...

erikals
05-01-2018, 01:41 PM
The pots also look like...

the pots seem to exclude the sky+fog reflection.
one can tell as the sky reflection is too bright.

...just a guess.

thomascheng
03-15-2019, 12:14 PM
Are you guys considering denoising solution for AMD GPUs? Such as AMD Wavelet Denoiser? Not everyone is on board with Nvidia GPUs.

gar26lw
03-15-2019, 07:23 PM
Based on the comments I get and my own observations, most agree that (after the drama of last year) there has been a marked improvement in tone and discourse here in 2018. This remark is not particularly helpful (the rest arguably was).

Buckshot characterizations of this type reduce you to the level you are decrying, being at once unnecessarily provocative, overstated, and offensive to many who are here to make positive contributions (just as many have done in this specific case).

you summed it up nicely, thankyou.
clearly someone thinks they are superior, unfortunately that’s not the impression i am getting.

i’m glad the original poster started this conversation, got some good info for everyone and reasonable discussion. i guess he didn’t have as much luck on facebook with the higher caliber of patron that frequents there.

SBowie
03-16-2019, 06:34 AM
clearly someone thinks they are superior, unfortunately that’s not the impression i am getting.Well, someone else is a little cranky. ;)

Resurrecting this to take a swipe at someone is sort of the definition of the unnecessary and provocative "Buckshot characterizations of this type" you just commended me for criticizing. Water under the bridge, nothing to see here, let's move along.

erikals
03-16-2019, 06:57 AM
http://forums.newtek.com/images/misc/quote_icon.pngWater under the bridge, nothing to see here, let's move along.http://forums.newtek.com/images/misc/quote_icon.png

+5

gar26lw
03-16-2019, 07:05 AM
Well, someone else is a little cranky. ;)

Resurrecting this to take a swipe at someone is sort of the definition of the unnecessary and provocative "Buckshot characterizations of this type" you just commended me for criticizing. Water under the bridge, nothing to see here, let's move along.

yes the irony is not lost ;)

gdkeast
03-16-2019, 06:24 PM
From my perspective, everything you need to know has been amazingly explained here:
www.rebelhill.net/html/rhlsr.html

Without these tutorials, I'd be probably saying the same thing; but with them, all is answered. And this is why I'm so grateful to Rebel Hill.

gar26lw
03-16-2019, 08:42 PM
should they be bundled with lw docs?
i don’t think the render settings are that complicated but i do think the unified system of the past was a good setup.
it’s seems self evident that some thought about render setting presentation/doc’s would not go amiss in making things more accessible for the layman.
maybe some render wizard or pop-up hints?
maybe preset scenes or render settings for common scenarios?

rustythe1
03-17-2019, 09:03 AM
its not really about the "render settings", but more about the light and materials and how the render engine works, that's where most people are going wrong in regard to things like fireflies, so preset scenes probably will not help, following the method is what works, and there are some nice little tidbits in the tutorials that help with features you thought you had lost from the old engine!

Chris S. (Fez)
03-17-2019, 09:38 AM
I love the new renderer but Fprime progressive final rendering was genius and is sorely missed. It is essentially idiot-proof in terms of render settings. Sometimes, for print projects in particular, you render overnight. Whatever quality the renderer arrives at by the time work begins the next day, you call it done.

RebelHill
03-17-2019, 09:41 AM
Yep... there really is no such thing as a "common scenario", because that would mean a common scene. Following the workflow is pretty straightforward once you have the basic parts understood, and makes it all very quick and easy to work with.


Without these tutorials, I'd be probably saying the same thing; but with them, all is answered. And this is why I'm so grateful to Rebel Hill.

Cheers, glad you're pleased.

gar26lw
03-17-2019, 11:07 AM
I love the new renderer but Fprime progressive final rendering was genius and is sorely missed. It is essentially idiot-proof in terms of render settings. Sometimes, for print projects in particular, you render overnight. Whatever quality the renderer arrives at by the time work begins the next day, you call it done.

the ability to carry on from a point was also awesome, if nothing had changed.
i think modo borrowed fprimes progressive ability.

Chuck
03-18-2019, 09:07 AM
Are you guys considering denoising solution for AMD GPUs? Such as AMD Wavelet Denoiser? Not everyone is on board with Nvidia GPUs.

We don't comment on future plans, but we would recommend that you please submit a feature request for the development team's consideration.

sadkkf
03-18-2019, 09:28 AM
Yep... there really is no such thing as a "common scenario", because that would mean a common scene. Following the workflow is pretty straightforward once you have the basic parts understood, and makes it all very quick and easy to work with.



Cheers, glad you're pleased.

I second this praise. Your videos have improved my quality of work immensely.

You are the hero Lightwave needs.

gar26lw
03-18-2019, 03:50 PM
We don't comment on future plans, but we would recommend that you please submit a feature request for the development team's consideration.

i think there is an intel denoiser too.

http://www.cgchannel.com/2019/01/intel-releases-open-source-denoiser-open-image-denoise/

different to the one in lw?

erikals
03-18-2019, 04:17 PM
there are also DPont denoiser tricks for LW2019, demonstrated by Gerardo Estrada
https://forums.newtek.com/showthread.php?158546-Denis-has-updated-most-of-his-plug-ins-for-2018-2019!!!!/page5