Not logged inOpenClonk Forum
Up Topic Development / Developer's Corner / Revival Release Design Document
- - Date 2019-09-15 13:51
Parent - - By Clonkonaut Date 2019-09-10 18:38
Alright. I am not nearly finished with this, as you can see. Many more topics buzz around in my head but I haven't yet found the time to put these into words. But I hope this will already give a first impression on the scope of the project I have in mind.

https://wiki.openclonk.org/w/Revival_Release_Design_Document
Reply
Parent - By Marky Date 2019-09-10 18:42
Good effort :) Even less "lets change everything" than I thought.
Parent - - By Marky Date 2019-09-10 18:49
Regarding the movement stuff: Delegating hardcoded movement stuff to script logic might help? So instead of walk procedure => stick bottom vertex to the landscape / scale procedure => stick left or right vertex to the landscape we could just define "stick this vertex to the landscape now" (maybe even several, for a transition between walk and scale)?
Parent - By Clonkonaut Date 2019-09-10 22:18
As there are no solutions given, it's an open discussion. If you think you have a good idea, feel free to share it or test around. I'd rather have a new thread for that instead of a discussion branch in an unrelated topic.
Reply
Parent - - By Zapper [de] Date 2019-09-14 20:24

>Delegating hardcoded movement stuff to script logic might help?


I would say it wouldn't help at all. Even if it might "feel" different for scripters, for the actual outcome it doesn't matter whether something is "hardcoded" in the engine or "hardcoded" in script. And improving it in the engine might even be a lot easier than doing that in script (e.g. because of the speed advantage).
Parent - By Marky Date 2019-09-20 16:04
I just thought that defining "vertex n attaches to landscape now" is different to "procedure walk always attaches the bottom vertex". Also, in script you can adjust the behavior per definition, in the engine you can only code the behavior globally.
Parent - - By Zapper [de] Date 2019-09-14 20:22
One thing that I noticed about the shape maps commentary. In the beginning you say "the game looks a bit different each release in a feeble attempt to somehow work around the problems through a path of low resistance[...]" and I think that modifying the shape maps fits well into that category :)
We had the problems with uneven landscapes way before the shape maps. In most games that are not adventure-style rounds (but rather settlement), you'll mostly get stuck in your own holes and often barely notice the shape maps. So I guess a "sustainable" fix that doesn't fall into the category of minor tinkering along the path of lowest resistance would change the movement per-se. E.g. different vertex behaviour or even allowing "teleports" over certain height differences to make the overall experience better.

But other than that. All valid points. I guess the vertex/movement system is one of the parts of the game where radical changes could have the highest impact on playability :)
Parent - By Clonkonaut Date 2019-09-15 11:26
I'd say you mostly notice the shape maps in arena-style maps where landscape shaping is of low priority but you frequently want to aim something at others and then suddenly get stuck somewhere (but I agree that changes in vertex behaviour can have a heavy impact on how walking works).
Reply
Parent - - By Clonkonaut Date 2019-09-15 11:43
(Splitting this up because it's a different point)

I am also aware that a few points in the list (especially the UX stuff I have in mind) will lead to changes in how the game looks. I just hope that we might come up with one final change to core systems and can then confidently go on. If we now identify lurking problems (or deem those non-problems through discussion) and come up with a satisfying solution, we just might get a game that finally loses its overall 'still in development' touch.
Reply
Parent - By Zapper [de] Date 2019-09-15 13:51
Maybe we should spend some time watching youtube Let's Plays again :)
I always found it super helpful to see how new players get to know the game and what they complain about
Parent - - By Newton [de] Date 2019-09-20 09:20
A good effort! I'd like to first comment on a few

> Shape Maps: Traversing the landscape


An easy task that is not too involving with a clear definition of done! Sounds good, I'm in! Will you remind me next week, perhaps we can meet to do it together?

> Corner Climbing: Bad animations, bad movement


PeterW once suggested to experiment making the Clonk (and thus basically everything) much smaller (factor 2 or 3). The vertices of the Clonk would be closer together, not allowing stray pixels to get between. The landscape in turn would be more blocky in relation to all the other objects in the game, but because of PeterW's shader and Sven's shape maps, still look good. In Terraria, there are no problems maneuvering in the landscape. In Clonk, single pixels have always been a nuisance and bring no benefit to the game.
(I'd not be available to do that though.)

> Landscape Manipulation


I do not see such a problem with "destroying the landscape". It is part of the gameplay challenge to make the mine accessible rather than digging the shortest path to the gold. I do not expect that after a round of Clonk, I end up with a beautiful or easily traversible landscape.

I forgot how the loam and the bucket worked. But anyway, here is an idea for a new loam behavior:
- clicking and pressing down shows a circular progress bar at the point where the loam will be placed. After 1s or so of pressing down, the loam is placed
- the loam is placed as a big chunk of earth. The chunk is so big that one won't get the problems with scaling
- loam could be an ammunition-type of object that is collected automatically while digging

---

More generally, from my point of view, what's missing is a clear vision of what must be in the core game in order to be considered as done. And then focus on that. Few people want to play unfinished games.
Parent - - By Clonkonaut Date 2019-09-22 13:02
I do see a problem in landscape manipulation when all you get is something you cannot really go though without a hassle. Terraria is a good example because you can easily build stuff in Terraria that makes it easy to walk through everything. Partly because in Terraria you remove the landscape in chunks rather than in pixels.
Reply
Parent - - By Marky Date 2019-09-22 19:47
What if the shovel instead showed some kind of preview what kind of "corridor" you'd remove (similar to the metal bar pack) and when confirmed you do a slow animation similar to the pickaxe?
Parent - By Pyrit Date 2019-09-23 22:27
That sounds like a radical gameplay change. But I lke it.
It's closer to the previous titles: Give tasks and let your clonks do stuff, while you can go and do something else.
But it contradicts the design choice: The player should not sit there and passively watch the clonks do stuff.
Parent - By Clonkonaut [de] Date 2019-09-24 08:23
I had something similar but not necessarily slower in mind. I will soon write it down (but of course anyone may suggest different approaches).
Reply
Parent - By Zapper [de] Date 2019-09-24 09:40
To quote a wise man "Even worse, the game looks a bit different each release in a feeble attempt to somehow work around the problems through a path of low resistance"
While your suggestion would make tunnels through earth straighter, it doubt it would mitigate the problem. From my experience, you do not usually get stuck on earth chunks but rather on rock etc. that has been blasted to weird shapes. It might make it a bit easier to dig straight lines, but comes with different drawbacks (less natural, boring for player, might not allow for other things that are possible with the current shovel (e.g. round shapes).

I guess if we really want to fix the problem of getting stuck, we would have to revisit the vertex system - at least for the Clonk. I don't think that's an impossible task and I think that is the only thorough solution without trying here and there to fix some minor issues
Up Topic Development / Developer's Corner / Revival Release Design Document

Powered by mwForum 2.29.7 © 1999-2015 Markus Wichitill