[Tea Monster]

icecoldduke, on 01 June 2016 - 04:11 PM, said:

.... When I get physics working...

WUT?!!!

[icecoldduke]

Tea Monster, on 01 June 2016 - 04:19 PM, said:

WUT?!!!

Yes PolymerNG will have physics. I said this awhile back . Physics is something I would like to get in after I get the renderer more flushed out.

[Mark]

After hearing that, I'm willing to bet my paycheck that TM is approaching his wife right now with an ear to ear grin and the biggest boner ever.

[Kyanos]

icecoldduke, on 01 June 2016 - 04:21 PM, said:

Yes PolymerNG will have physics. I said this awhile back

icecoldduke, on 03 December 2015 - 06:27 PM, said:

4) Model should get updated to skeletal models, no physics will be in at first.

Tell us more about how physics based animation will work.

[HiPolyBash]

Trooper Dan, on 01 June 2016 - 09:57 AM, said:

I'm sure that the new renderer will be put to good use. But I don't buy into this idea that there are lots of people out there hankering for more Duke Nukem, and if only there were a flashy renderer which made it look like a modern game they would flock to the community and make new content. Also, the first really good projects that have lasting value are almost certainly going to be created by the usual suspects around here -- people with a lot of Build/EDuke32 experience and true creative inspiration. Setting out with the goal of "pushing everything to the absolute limits" doesn't sound very inspiring from a creative standpoint and is likely to result in a flashy tech demo which won't hold people's attention for long.

Duke Nukem Reloaded did exactly that. When the original images of the rooftop circulated everyone collectively ejaculated all over themselves to see a classic they've grown up with looking gorgeous and they wanted to contribute to it. I'm sure a graphical showcase getting press coverage would attract new blood.

[icecoldduke]

Drek, on 01 June 2016 - 05:24 PM, said:

Tell us more about how physics based animation will work.

Duke picks up alien, Duke throws alien through boxes that are stacked on top of each other, boxes fall down. Tada. Physics based animation was not the proper term, but you will able to use physics to make things feel more immersive. Also I would like to have ragdoll support eventually. I might try to get a friend of mine at work to start working on physics stuff, but we'll see.

Do you have a scenario, besides fans, that would require vertex animation, that couldn't be done with skeletal animation or through scripted physics events?

[Mblackwell]

er... Do you know how clipping and collision work in the engine and what kinds of data gets returned and how it ties into actor members and gets hooked into scripting?

[Danukem]

HiPolyBash, on 01 June 2016 - 05:29 PM, said:

Duke Nukem Reloaded did exactly that. When the original images of the rooftop circulated everyone collectively ejaculated all over themselves to see a classic they've grown up with looking gorgeous and they wanted to contribute to it. I'm sure a graphical showcase getting press coverage would attract new blood.

The Duke Nukem Reloaded screens were shown in 2010. In May of 2011, Duke Nukem Forever was released. Unfortunately that changed the landscape because it was a current-gen game (for that year) and was so disappointing. I do not believe that the same level of excitement would exist today, but I would be happy to be proven wrong.

[HiPolyBash]

I didn't say that it'd be to the same level, just that it'd be likely to attract new blood. The point is it'd still be likely to attract people to the community, maybe not to the same degree that Reloaded did at the time but it'd surely result in some people coming out of the woodwork given a modern base to create content on without any legal wrangles. Everytime any classic game is given a modern overhaul news of it spreads like wildfire, there's no reason to think that this would be any different.

The point is showcase style content that is designed to show off the capabilities of PolymerNG has a reason to be created if not simply for publicity it'd at the very least show content creators who once had an interest in Duke3D but are now more just casual observers that there's something cool in the works that'd serve as an avenue they could explore without the issues that they would have experienced in the past ranging from legal to lack of support for modern technology and optimization which were serious hurdles to overcome.

[Micky C]

Trooper Dan, on 01 June 2016 - 10:04 PM, said:

The Duke Nukem Reloaded screens were shown in 2010. In May of 2011, Duke Nukem Forever was released. Unfortunately that changed the landscape because it was a current-gen game (for that year) and was so disappointing. I do not believe that the same level of excitement would exist today, but I would be happy to be proven wrong.

Don't forget that all the people working on DNR were most likely already fully familiar with the engine. A lot of people would have to learn several things from scratch for Build.

[Tea Monster]

HiPolyBash, on 01 June 2016 - 05:29 PM, said:

Duke Nukem Reloaded did exactly that. When the original images of the rooftop circulated everyone collectively ejaculated all over themselves to see a classic they've grown up with looking gorgeous and they wanted to contribute to it.

I didn't. Those first images weren't that good looking either. Dozens of people had tried to do this in a variety of different engines and got shut down with C&Ds. My first thought was "Why has this guy been allowed when everyone else was not?" What have you done behind the scenes to be allowed to do this when everyone else was shut down?

Secondly, when he was presented with this unique opportunity, which far more competent and talented people have been refused, he screwed the pooch. He had the one golden ticket and he blew his nose in it and tossed it away.

Those are my thoughts on the matter.

OK. Rant over.

Edit (due to not wanting to clog up the thread with OT stuff): If Fresch hadn't buggered up his chance at making a free Duke mod and had released Reloaded to to community, it could have been very much like the current Unreal Tournament, with hundreds of people making new maps and content for an open engine and game. That would have blasted Duke modding into orbit and may even have put the King back onto his throne. THAT is what Fred threw away and why I think it's such a crime.

Hopefully with what is happening with NG, we can try and make some of that finally happen.

[HiPolyBash]

The point was that the images circulated everywhere, the general public grew excited, and it attracted dozen of people who were willing to contribute to it which is exactly the point I'm trying to make when it comes to the case of content being created to showcase the renderer to attract people to the community.

[Kyanos]

icecoldduke, on 01 June 2016 - 05:33 PM, said:

Do you have a scenario, besides fans, that would require vertex animation, that couldn't be done with skeletal animation or through scripted physics events?

Not until I get to try out "scripted physics events" in build.

About models, can you make it so we see a model if the mesh extends outside of the current sector the sprite is in and the camera isn't looking at that sector?

[icecoldduke]

Drek, on 02 June 2016 - 04:23 AM, said:

Not until I get to try out "scripted physics events" in build.

So can we agree to reserve judgement until the physics system is implemented? .

Drek, on 02 June 2016 - 04:23 AM, said:

About models, can you make it so we see a model if the mesh extends outside of the current sector the sprite is in and the camera isn't looking at that sector?

Yea I can do that, sounds like a good idea .

[Kyanos]

icecoldduke, on 02 June 2016 - 05:52 AM, said:

So can we agree to reserve judgement until the physics system is implemented? .

Physics, specifically ragdoll skeletons will be a huge plus. Currently models tend to clip through everything, dead troopers are the worst for this IMO.

[Mblackwell]

I would suggest learning how clipmove works.

[icecoldduke]

Mblackwell, on 02 June 2016 - 07:21 AM, said:

I would suggest learning how clipmove works.

I know how clipmove works and your right it would be hard to preserve existing functionality. My first instinct would be, as I've said before, have a legacy path that allows existing mods/legacy HRP stuff to run, but this legacy path would disable all NG effects. Then we would expose new functionality to the script code that deals with proper physics.

[Mblackwell]

It's not about HRP stuff, it's about the entire game engine relying in clipmove.

[icecoldduke]

Mblackwell, on 02 June 2016 - 07:40 AM, said:

It's not about HRP stuff, it's about the entire game relying in clipmove.

If your taking about what it would take to make new content, that hooks up the physics engine, that would allow a lot of the existing code to work properly, I'm not sure yet what the path of least resistance would be. If you would like to look into that for me, that would be fantastic . If not I'll deal with this after the renderer is more flush out.

[Mblackwell]

No, I'm saying that every piece of code including projectiles, player movement, etc is all bound to clipmove (both in actual calls, and expectations of how it works and what it returns).

[icecoldduke]

Mblackwell, on 02 June 2016 - 08:08 AM, said:

No, I'm saying that every piece of code including projectiles, player movement, etc is all bound to clipmove (both in actual calls, and expectations of how it works and what it returns).

I get what your saying , I don't have a good answer for you right now that would follow the path of least resistance. I have to look into the issue more, which won't happen until the renderer is flushed out more. When I do start to look at it, or I get someone else to, then we can discuss the best way forward. For right now I am expecting I will be able to get a physics system working, so I'm moving forward with that assumption. If that changes then we can re-evaluate at that time. My sole question if we can get physics working with scripted physics events and skeletal animation working, is there any reason to have vertex animated objects?

[Mblackwell]

Is there any reason not to other than Xbox One?

[Spiker]

Quick question about the new renderer: albedo or diffuse?

[icecoldduke]

Mblackwell, on 02 June 2016 - 08:55 AM, said:

Is there any reason not to other than Xbox One?

Performance and memory. I'd rather weed out the performance crushing system's early on, especially ones that don't yield good results, no matter the performance impact scale between min spec and high end hardware. I want to use the resources wisely .

[Mblackwell]

But it's not a AAA game, it's a modding tool. The people creating the mod decide what things they need from the full available list and what spec they target. Limiting options arbitrarily is a bit silly.

[icecoldduke]

Mblackwell, on 02 June 2016 - 09:13 AM, said:

But it's not a AAA game, it's a modding tool. The people creating the mod decide what things they need from the full available list and what spec they target. Limiting options arbitrarily is a bit silly.

I'm not limiting options arbitrarily, I have a solid reason behind this, I want to maximize performance and vram usage.

[Mblackwell]

You're at the wrong end of it.

You want to maximize performance and vram usage for the theoretical game you are not building. Your renderer should be as fast as possible no matter what's thrown at it and the modders are the ones who are actually MAKING a game and decide what content goes into it and if the performance cost means something to them based on what systems/hardware they target.

[icecoldduke]

Mblackwell, on 02 June 2016 - 09:19 AM, said:

Your renderer should be as fast as possible no matter what's thrown at it a

You throw bad content at the renderer, there is only so much you can do to maximize performance. No one has thrown a good scenario at me yet that would make me thing supporting vertex animation with NG graphics is a good idea. Remember vertex animated objects will work with NG effects turned off. This conversation is only about making new content for PolymerNG.

[Hendricks266]

You need to drop the "legacy mode" mindset, or this is never getting merged (without modifications).

I'm not trying to make an ultimatum, simply stating that the more stuff that conflicts with mainline development goals, the more work merging creates (for me, in all likelihood) and the further it delays unification.

[Mblackwell]

Let's see, so that people don't have to rebuild decades of content so they can use some post process shaders?
Or so they can build things like animated static props (rotating objects in particular) without waiting for things which are incompatible with Duke/Build to be implemented?
So that people can build content that works on all renderers without modification (and can have extra features in the newest renderer)?