Player Node Limitations?

[Jervan]

Just wondering if anyone knows if:
1) There is a limit to how many nodes a model can have?
2) Mounting a "hand" player to the players hand, then unhiding nodes to show a wielded weapon is feasible.

I'm doing this to allow for simple weapon customisation / skin colors. Thanks for any knowledge!

[Ad Bot]

[-Jetz-]

2) Mounting a "hand" player to the players hand, then unhiding nodes to show a wielded weapon is feasible.

I'm doing this to allow for simple weapon customisation / skin colors. Thanks for any knowledge!

Technically possible, just really awkward for the most part. Most scripts for items, images, and mounted players in the game and other add-ons were made without anything like this in mind. Couldn't even begin to guess at the number of issues you'd have to resolve. I can't think of anything that would actually prevent a setup like this from working, but wiring it up might become a pain unless you hard-code everything.

[lordician]

I am currently (together with whatever else i am trying to work on at the same time) working on something similar to your 2nd point.
Mounting a playertype with an empty model as a hand as it were.
I do not hide the node, no need for that.
But Jetz is right, there will be a lot of issues and things not working right with every item.
You would have to have items specifically made to be able to be mounted on a fake hand basically.
The code for the items (shooting them, equipping them all that) has to be made so that it keeps this in mind.
It is possible, but it requires a bit more effort and won't work 100% with every weapon.

It seems like we are on the same line on some RPG kinda things, heh.
I will edit this post in a minute with like a preview of what i got so far.
Maybe it will help you on your way, maybe give some answers.
I am just saying, it is not done yet, but the concept works.

Quick-Edit:
As for mount nodes, the Torquescript Quick Reference Guide says there is a limit of 32 mount nodes.
As for like the model nodes, i am not sure.
Pretty sure you meant the model nodes and i am sure there is a limit.

EDIT:
I got the question wrong so never mind about me posting what i got for the preview.
I do not want to derail the topic with my now not related code. :X

[Jervan]

As far as what I'm planning, the "weapon" (whatever it ends up being) would be part of the "handPlayer" model, it'd just be a hidden model node. The item mounts this handPlayer onEquip. Then I unHide the nodes depending on what type of upgrades the player has and what weapon it is. Then spawn the relevant projectile when the player right clicks, or do the relevant blocking method when they right click (framework already in place). The only down side is that it limits weapon speeds to their weapon class, so there is less variety there.

If it's true about the model nodes being limited then I'm thinking one handed swords all work the same, so those animations can have a swordHand. Then I would just do it by weapon animation requirements. If there ends up being too many sword upgrades then I can have a fancySwordHand and a basicSwordHand, etc.

If anyone knows of this limit it'd help a lot knowing what it is before I reach it, I'm trying to make client load times as short as possible!

[Conan]

armyunit confided once in me he had a model with 140 bones - taking this into account i dont think theres really any limitations to nodes other than the possible lag due to the extra faces and stuff in the complex hand. i know bones aren't nodes, but i presume a bunch of these bones had a corresponding mesh. i dont think you'd need more than 20.

and even if you keep the actual datablock count low, the complex model associated with said bot would probably take a while to DL.

[lordician]

AH, then i got you wrong.
What i am working on is something so you can equip an item in both hands without needing another image.
You are talking about a playertype being all the weapon models in one, i mean, yeah that could work i guess.
It would cut on the images datablocks, that is for sure.
On the other hand, i think images with their state machines are something hard to replace in that way.
Emitters have to be done via an image which you have to mount to the player for another example.

All in all it would create some more hassle which i am not sure about if it is worth it.
I guess it has to do with what you said about upgrades to the weapon, being actually visible on the weapon?
In that case it surely makes it more interesting to do it that way.

[Jervan]

The pros outweigh the cons in that case, yeah! :D The only danger was superficial limits.

Pros:
Customization (combinations of parts and coloring :D)
Much less animating (big time sink)
Save datablocks
Easily linked with combat system

Cons:
Weapon speeds dependent on frames instead of stateTime
Potentially harder on GPUs
Loss of using the image state system (RIP, flaky piece of )

Your system sounds really cool lordician. Thanks again Conan

[lordician]

Weapon speeds dependent on frames instead of stateTime
Your system sounds really cool lordician.

Oh yeah right, animation speeds, i forgot about that, but yeah, that is a con. :C
And thanks man.

I do not want to hijack this topic, but if you are interested, here is what i have so far.
It is part of my Player_Roleplayer project i started to get back into things.
But there are so many things that i wanted to do i just never finished anything of it ever since.
But i am working on it, slowly.
The code which i thought would be useful would be the support script: "Support_AdvEquip".
This allows modders to make weapons that can be dual wielded (without needing two images, but that is still possible) and a way to make equipping work for that.
And of course something so that players can use the tool in the offhand.
To test it currently i use something you might have seen a topic about earlier, my advanced melee support script for which i included some custom weapons to test both the advanced melee and the advanced equipping.
There is still some work to be done, but it is progressing quite well.
I think both things are at least at like 50 to 75% done already.

Sorry, that is all my hijacking for now, i will make a topic for the support scripts when they are done.
I did not mean to (attempt to) hijack the topic but i do also not want to look like someone who claims he makes stuff and never shows it. :/
So i figured, might as well just post the preview after all.

But yeah, excited to see how your weapon players will work out.
The idea is not new. A lot of gun creators have already thought of this but never gotten around to actually do it.
But i think this is a great opportunity to do so.
For RPG's the customization it will offer will do a lot.

[Jervan]

I reaaally like the idea you have for sleep mechanics and downed players in the combat. If you aren't working on a specific RPG / theme you could team up with me? We do have similar ideas and it'd share the load. I've also got a build of >150k bricks done so it's mostly just about coding systems and content now PM me if you're interested!

EDIT: also in the spirit of not being a BLer who just talks about things, I'll update this topic with the combat is working cause if it works it's pretty cool :D

[Jervan]

As promised
https://youtu.be/dtyWtQXMSeI

Require more knowledge:
https://forum.blockland.us/index.php?topic=299063.0

[portify]

the game will write garbage data (well, incorrectly read data) when you it sends a node color update for any shape that has 256 nodes or more which can break recoloring, randomly crash the game or disconnect players, but the effective limit is 2,147,483,647 nodes

[lordician]

I guess the first part is from personal experience?