VLM diary - from a clown to a respected comedian.

[Wylte]

This thread is great!  I'm sure that you documenting your journey and laying it out like this will help others in the future, as well as further development of the toolkit and tutorials for using it.

[takise]

Due to professional obligations, I have to wait until the weekend to organize the video recording setup in OBS.

 

I am very suprised and happy with the support I have received from the community here and behind the scenes, it really motivates me to continue working.
Thank you all for your support and positive reactions.

 

...to be continued soon

Edited January 4 by takise

[takise]

20 hours ago, Wylte said:

This thread is great!  I'm sure that you documenting your journey and laying it out like this will help others in the future, as well as further development of the toolkit and tutorials for using it.

 

Thank you for supporting Wylte, it really means a lot to me.

The point is to have more and more active creators in the community.
Every new mind brings new ideas and a fresh look at old problems.

We just need to make it easier for them to enter and systematize the work system.

Edited January 4 by takise

[dcbenji]

I think splitting off each individual light is ok for an EM or something without flashers and/or many blinking lights. Otherwise, each Gi light split off to generate proper raytracing generates too many lightmaps which effects performance. What i did on badcats and policeforce, for example, was just select a bulb in key areas and split those off, like 5 or 6 bulbs total. You can try splitting them all off, but if there are performance problems (lots of stutters or major drop in FPS) you may want to optimise the table this way.

[takise]

I got Covid, luckily the light version, so apart from the headache and nausea it's quite bearable.

The advantages are that I have time to prepare for the start of recording and editing materials for the diary.
I'm not going to work next week, so materials will start to appear and the publication system will be systematized.

 

I lie under the covers and read and become more and more involved in the topic.

 

One technical aspect of VLM puzzles me.
I will write in short sentences because I can get lost in English and after almost 10 years of using it on a daily basis, I still cannot think in this language and every time I translate it in my head on the fly.

 

When exporting from Blender/VLM to VPX when all scenarios are generated.

I understand that this is done on the principle: one light source in relation to the objects on the table.

 

My question is whether scenarios are then rendered in relation to the light source for each object on the table separately?

 

And the second question, is whether objects on the table that are not interact with a given light source are omitted, I mean are objects beyond the range of a given light source omitted in this process?

Does VLM have any limitations in terms of interoperability with renderers other than Cycle?

 

Generally, it's winter outside and there's snow in an amount that hasn't happened here in a decade.
Everything is surrounded by a white fairy-tale aura.

It's good that with age you start paying attention to such simple details.
For most of our lives, we just pass by in pursuit of something.

 

I wish everyone a nice weekend.

 

...to be continued

 

 

Edited January 6 by takise

[bord]

On 1/4/2024 at 8:01 PM, dcbenji said:

I think splitting off each individual light is ok for an EM or something without flashers and/or many blinking lights. Otherwise, each Gi light split off to generate proper raytracing generates too many lightmaps which effects performance. What i did on badcats and policeforce, for example, was just select a bulb in key areas and split those off, like 5 or 6 bulbs total. You can try splitting them all off, but if there are performance problems (lots of stutters or major drop in FPS) you may want to optimise the table this way.

 

@dcbenji I agree, although do you think turning off raytracing on individual VPX lights gives the same advantage? In other words, could I render each light individually and then turn raytracing off as needed to optimize performance?

[dcbenji]

On 1/6/2024 at 5:32 PM, bord said:

 

@dcbenji I agree, although do you think turning off raytracing on individual VPX lights gives the same advantage? In other words, could I render each light individually and then turn raytracing off as needed to optimize performance?

Hmm yeah i see the benefit of splitting all the GI lights, but I believe split lights generate more lightmaps and more draw calls which affect performance apart from the raytraced shadow effect. I don't think the extra geometry is so much a possible issue, but the processing of the fading (or binary on/off) of the individual lightmap meshes (as opposed to large grouped meshes). But on older tables where the GI is basically static maybe this doesn't matter.  Would have to test more, might just be something to keep in mind if you run into weird performance problems on some toolkitted tables.

[mistermixer]

Following this topic with interest.  

There are a lot of steps finishing a table so think it's hard to do all by one person, not ?

( starting a build in vpx - export to blender - swap objects and or create meches - testing renders,baking ... - 

  export to vpx ... )    And the hardest one for me will be to script all these new features.

 

Also difficult i think is when you have to add an object,light ... you forgot in the first place. Do you have to

start the entire blender proces again ?  The bakingproces lasts 8 hours i've been told.

 

Thanx for digging into this new process building a table.

 

( started with blender 4.0 giving errors importing a table, version 3.6 will do )

 

[rxhunter]

Hi @takise! I'm also a n00b with VPX who's hoping to give something back. VPX is an incredibly deep library of really cool stuff, and I'd also like to contribute. I'll be watching this thread for when I dive into table creation. Thanks for sharing!