PC Alpha 18 Dev Diary!!

[khzmusik]

Fair question. And it's the main reason I suggested the template Difficulty settings approach. Where there is a Difficulty Option, but one that simply tweaks the other available setting options. Similar to how video settings work. Where you can choose the main suggested settings for video, but you can tweak the rest of the settings and it switches to custom.

I am absolutely on board with this idea - but instead of just having "difficulty," do an even more general preference template, say, "play style."

For example, one option would be "dead is mostly dead" where bedrolls and land claims are deactivated, and everything disappears permanently from your toolbelt and backpack on death.

This would be especially useful if it could be opened for modding. Say, you have a "farmer" play style with a bunch of settings, and it includes a mod that has all sorts of extra farming goodies; but the mod wouldn't be loaded (or would contain logic to not activate features) if you weren't in that play style.

 

[Jackelmyer]

I'm assuming by LOD you mean something like the "impostor POI" models you see in the distance? When a player renovates a POI its "impostor" doesn't change, causing a kind of "pop in" effect when the imposter is swapped for the actual POI.
Not speaking for TFP but I can see at least a couple problems with that.

1. They would have to generate LOD "impostors" for areas that don't have POIs in the first place, e.g. when you build a base in the woods from scratch.

2. From what I understand, "impostor" POIs are models, and generating models on the fly is, to put it mildly, difficult. Even if possible in Unity, it would increase loading times dramatically, especially for folks with mid-to-low-end hardware.

3. It would be a problem on MP servers, since those generally don't restart very often.

The "impostor POI pop-in" issue is one I've noticed myself, but it's one of those things that I don't think can be easily fixed, so I didn't even bother filing bug reports about it.

Of course, if I'm totally wrong about all of this... then TFP, feel free to fix it.

I don't know if it would be that hard or that bad.

Currently generating the imposter in the editor is pretty darn quick. The game could snapshot periodically changed structures via a queue to slowly update those imposters over the course of a game. If it's pushed off to a background low priority thread... You really shouldn't notice an impact.

At least, that's how it seems based on prefab imposter generation in the prefab editor. I get the impression that the imposter is generated by severely reducing the quality of the POI textures overall and removing some blocks all together. So it doesn't seem to require the POI to even be rendered. Maybe run through a renderer, but not actually rendered.

Dunno. *shrug*. Seems doable.

 

[Jay_ombie]

I don't know if it would be that hard or that bad.
Currently generating the imposter in the editor is pretty darn quick. The game could snapshot periodically changed structures via a queue to slowly update those imposters over the course of a game. If it's pushed off to a background low priority thread... You really shouldn't notice an impact.

At least, that's how it seems based on prefab imposter generation in the prefab editor. I get the impression that the imposter is generated by severely reducing the quality of the POI textures overall and removing some blocks all together. So it doesn't seem to require the POI to even be rendered. Maybe run through a renderer, but not actually rendered.

Dunno. *shrug*. Seems doable.

Could be an optional thing or have a 7 day LOD Re-render....

 

[Jackelmyer]

Could be an optional thing or have a 7 day LOD Re-render....

To be fair, I only know how to build .Net queueing systems and multi-thread applications. Nooooo idea how hard the actual imposter generation code is. Simply that TFP seems to already have it. And for fair sized Prefabs I've built, it seems to take less than a couple seconds to generate the imposter.

That said, I've never bothered to look of the game is pegging all cores on my machine while doing it. *shrug*

Disclaimer: I... am not a TFP developer. I am... An avid talking out his rearer. So... Grain o salt on my overvalued 2 cents.

 

[khzmusik]

I don't know if it would be that hard or that bad.
Currently generating the imposter in the editor is pretty darn quick. The game could snapshot periodically changed structures via a queue to slowly update those imposters over the course of a game. If it's pushed off to a background low priority thread... You really shouldn't notice an impact.

At least, that's how it seems based on prefab imposter generation in the prefab editor. I get the impression that the imposter is generated by severely reducing the quality of the POI textures overall and removing some blocks all together. So it doesn't seem to require the POI to even be rendered. Maybe run through a renderer, but not actually rendered.

Dunno. *shrug*. Seems doable.

Maybe. But, now you're talking about doing this at runtime, which has even more issues.

Say, you're on a MP server. Party member A is at a POI slowly rebuilding, repainting, etc. Party member B is far away, looting.

Now B comes back to that POI that A is renovating. What imposter POI is he supposed to see?

A is still updating it constantly, possibly in dramatic ways. Is the "impostor POI generator" supposed to just continually run in the background, generating imposter POIs whenever A makes a block change? Otherwise B will see a "pop in" anyway.

And, that's not even considering the resources it is taking to make these impostors, which could be better spent on things like more zombies or something.

I'm not saying it's impossible. It just seems like a lot of work (for TFP and in terms of computer resources) for not much value.

I'd much rather TFP continued to work on other things like RWG, NPCs, new content, balance issues, or increased mod support. Even if you consider this a performance optimization (which is definitely a WIP), I'd rather they spent their time on quashing lag/stutter issues or shoring up memory leaks.

 

[Slice]

Has there been any word on whether caves will ever return to RWG?

 

[JCrook1028]

We're not making the game to extend 4000 hours to 5000 hours of fun, but targeting 20-30 hour customers who the hair looks just fine to.

That is the most discouraging post from that I have ever read. Going for the quick hit and run consumer is the polar opposite of what this game deserves.

 

[AtomicUs5000]

I thought I deleted that one L shaped one that was kind of repetitive/obnoxious. Can you screenshot?

This is what I am referring to. There really isn't anything wrong with the LOD model, but maybe that you have to get too close for this to change to the higher detail model... or perhaps there is a missing stage?

 

[madmole]

Which I...er...suggested quite some time back btw...

Moving papers around on the desk is confusing activity for accomplishment.

 

[madmole]

I think you misunderstood me. Im not saying zombies need a haircut. Come on that's crazy!
What im saying is will there be randomized colors on clothes and possibly hair to make the zombies feel more diverse like in like in Mumpfy's NEW ZOMBIE TEXTURES mod.

Fight the same 20 zombies for 20-30 hours is pretty boring!

There is no way to change their clothes color, they are all skinned to one texture. You might be able to put a tint mask on the clothing or something but at that point you might as well make another zombie once you go that far, the cost is the same.

 

[madmole]

With how a player sculpts and recreates the scenery as the game goes on, LOD changes quite radically.
If it possible in the coming future to have say on first load up that the game would regenerate the LOD so to take into account and reflect the changes with map POI and the player's own base building ?

There is a modder who is working on this, if it works out we'll use his code.

 

[madmole]

Has there been any word on whether caves will ever return to RWG?

There are no plans I know of, at least not yet. Man made underground poi caves could spawn, but we're not planning any procedural stuff yet. Maybe down the road, first thing is to put out the fires.

 

[Ayrton]

a query why can not I run and jump at the same time? I am in the stable 17.2 b27

 

[AtomicUs5000]

There is no way to change their clothes color, they are all skinned to one texture. You might be able to put a tint mask on the clothing or something but at that point you might as well make another zombie once you go that far, the cost is the same.

If you have the byte array of the image pixels for the texture, couldn't a copy of it be made on the fly with changes to RGBA values at those pixels that can change... if not random color offset, then static maps with predefined offsets could be fast.

 

[Lonestarcanuck]

data/config/blocks.xml
What fun is it if you have invunerable blocks? Mod it in and see for yourself. You will be bored and go WTF was I thinking.

yeah so I made a Bedrock block fort...you are right, its no fun when you can sit there with my van helsing 100 shot crossbow with exploding bolts and rain holy hell on them and not get hurt.

 

[Jackelmyer]

Maybe. But, now you're talking about doing this at runtime, which has even more issues.
Say, you're on a MP server. Party member A is at a POI slowly rebuilding, repainting, etc. Party member B is far away, looting.

Now B comes back to that POI that A is renovating. What imposter POI is he supposed to see?

A is still updating it constantly, possibly in dramatic ways. Is the "impostor POI generator" supposed to just continually run in the background, generating imposter POIs whenever A makes a block change? Otherwise B will see a "pop in" anyway.

And, that's not even considering the resources it is taking to make these impostors, which could be better spent on things like more zombies or something.

I'm not saying it's impossible. It just seems like a lot of work (for TFP and in terms of computer resources) for not much value.

I'd much rather TFP continued to work on other things like RWG, NPCs, new content, balance issues, or increased mod support. Even if you consider this a performance optimization (which is definitely a WIP), I'd rather they spent their time on quashing lag/stutter issues or shoring up memory leaks.

Dude I absolutely agree on the priorities. Imposter generation at runtime would be waaaay low on the totem pole in my lil world of what comes next.

Was just talking about the feasibility. *shrug*. Was an interesting topic. Felt like chatting it up. :-D

 

[madmole]

That is the most discouraging post from that I have ever read. Going for the quick hit and run consumer is the polar opposite of what this game deserves.

Its not that, its being realistic about the demographic audience. Most people have better things to do then play one game for thousands of hours.

Most AAA games have about 6-10 hours of play, so aiming for 20-30 with hobby potential is still a very admirable design goal. The fact people play it 5000 hours or more is just bonus. I'm happy if I buy a game I play for more than 2 hours. If I can get 20-30 hours out of a game I'm super happy with it. We do want to create games that can become a hobby, but we can't expect people to play for thousands of hours and people who have played it for thousands of hours can't expect us to keep making it fun for them.

Then your on a slippery slope of adding late game crap 2% of players see. Its better to have less content that is super engaging for 20 hours than a ton of content that is meh or blocked.

Like I have to wonder how many players actually legit got tek gear in Ark on vanilla servers lol. What a waste of time making all those models and code for that stuff nobody ever got to use it unless they cheated.

 

[Hek Harris]

Most people have better things to do then play one game for thousands of hours.

Totally. And I blame you for that

 

[Ayrton]

@Madmole What do you think of this idea

Add 100 skin to the zombiez that we already have, and that they are random, then this will add more realism and variety, this makes the zombies not very repetitive, the question of where to remove the 100 skin?

-There comes the fans of this game, open a thread or forum where you can upload skin and select the best 100 and then add them in the next alpha, the skin have to be donated by the fans, with the number of moderators I do not think it is very difficult to reach 100 and add them in the next alpha

What do you think about this?

 

[Jackelmyer]

Its not that, its being realistic about the demographic audience. Most people have better things to do then play one game for thousands of hours.
Most AAA games have about 6-10 hours of play, so aiming for 20-30 with hobby potential is still a very admirable design goal. The fact people play it 5000 hours or more is just bonus. I'm happy if I buy a game I play for more than 2 hours. If I can get 20-30 hours out of a game I'm super happy with it. We do want to create games that can become a hobby, but we can't expect people to play for thousands of hours and people who have played it for thousands of hours can't expect us to keep making it fun for them.

Then your on a slippery slope of adding late game crap 2% of players see. Its better to have less content that is super engaging for 20 hours than a ton of content that is meh or blocked.

Like I have to wonder how many players actually legit got tek gear in Ark on vanilla servers lol. What a waste of time making all those models and code for that stuff nobody ever got to use it unless they cheated.

That is definitely the response I would expect from a company looking to flip games. Fair enough. That's the bulk of the industry.

And I don't mean that in an accusational way. I think you undervalue what you guys have built and what it yet could still be. Which I imagine a lot of people will feel by reading that post of yours.

Never the less. It's your guy's game. Totally fair.