Procedural sub-biomes to motivate more surface exploration

[Kegereneku]

Procedural means that you go random within certain parameters following a certain method. That is exactly what the continent generator did. ***Sniped for brievety***

Nice contradiction with your edit...

Procedural only mean that you generate following a method, randomness is something that must be added on top of it, they are separate things.

Ex : The planets of KSP are generated procedurally, will always appear the same way and are not randomized between players, something they don't need to.

Still, I'm giving you the doubt that it is all a semantic error...

Did you as I think (rereading 3 times your posts now) : generated ONE map where you procedurally generated the main feature (volcano, wind, rain...etc) and use a random tool to fill-up the crater/volcano...which will always appear ? OR... did you program something that actually generate a new continent/map/weather each time one play following only basic layout ?

Because, yes, if you don't attempt to randomly generate planet surface/geomes "each game", you indeed keep your geomes only procedural, not random. No matter if a noise generator is used to shape the original relief.

Still, to me it seem you still haven't understand how you are only changing the problems while solving the rover's one.

- increasing mindless click-fest.

- making rover dominant over static probes (though I'll agree it can eventually be accepted as a feature).

- worsening the balance of reward per biome/planets.

Anyway, I will stop participating to this discussion. Even if you were supporting me or I supporting you, your way of "discussing/editing" is giving me headache and I've better things to do.

ps : I should have been clearer that I meant 10 Area-of-Interest per planet, but it really should have felt obvious considering the number of them I gave as example for Duna.

I'll try to keep the quote apocalypse to a minimum.

***Sniped for brievety***

Thank you for that, but you are also giving me headache (really).

So I'll try my best to answer a few points without being unnecessarily blunt as well (note : I failed).

First, I insist you are like children who speak of stacking features that are mutually exclusive/self-defeating.

Yes you can call that a strawman, know that it is occasionally a valid rhetorical device to make someone understand his error, albeit in an irritating way.

...just like calling someone's argument a strawman is also a form of strawman, and not necessarily the constructive kind.

NEXT !

I'm one fervent defender of time-based game-mechanism...I made a few thread about it, especially go along a periodic-budget.

Still, the flaw in your 10h-scan logic is that you are still considering the Act of scanning the planet as a GOAL rather than as a MEAN to have fun.

It is a understatement to say that not everyone here play to have the planets 100%-explored, really most play for the Act of exploring which include a lot of firework. During timewarp you aren't playing, Timewarp is meant as a tool to lessen a boring parts of a game that have little to no relevance in the gameplay : waiting.

In our case, you still have to make it relevant in the first place. The act of waiting for the scan is pointless (and punishing) in term of gameplay. Even if Career-mode had periodic-budget it would still be pointless.

So once again I think you are confusing "it's realist" for a game-mechanic and simply going for the argument because you really like the idea.

NEXT : Trial & Error

Again, just because you can don't mean you should. Apollo-13 random part failure would NOT be fun. It only "could" ...assuming some miraculous Game-AI prevented it from happening every time it would make players tears their hair off and cursing.

Basically you are trying to 'win the argument' with an hypothetical situation based on pure luck. That's certainly a logical fallacy.

It's completely constructive. By that rationale, only people playing Minecraft using the same seed can discuss minecraft, or share experiences.

No, no, no... that's also not the rationales quoted at all, and that's also wrong !!!1! (1 for emphasis)

The less two player have in common, the less experience they share. The very reason Seed were created in Minecraft was to share them ! Solving what was a FLAW in randomly generated map.

Anyway the whole comparison with Minecraft make no sense ! Different game = different design. I could make thousand analogy of what you've done here but it would only detract us more.

NEXT : "God doesn't play dice"

Well, sorry but I can't imagine your game being well made seeing how ou are confusing the game having randomization features with the game rules being made up randomly. You think you are doing the former when you are doing the later (KSP's uniques planets parameters act voluntary as rules, same rules, same players experiences).

I've managed to argue without calling you stupid even once. I'll leave our relative intelligence for the reader to determine.

THAT. IS a strawman, just so you know.

I won't deny my provocative rhetoric, but keep in mind that if you DO say stupid things, reader might agree you are.

Also, you are confusing a specific use of FoW (for exploration, wargame do not randomize) with random generation and grasping at straw to justify a GOAL for players (exploration) which is a subjective opinion. Which is what I was trying to make you understand.

You would rightfully say that I cannot make that claim, just as you cannot define random as always bad.

I gave many arguments explaining in details why random generation conflict with many of the aspect of the current OR expected/wished game balance/gameplay. So your dismissing of all this as "me saying stuff is always bad" is insulting.

Not that I frankly expect you to understand anymore...

Even though you like some of my idea, I don't think you get the logic behind.

So, since it's clear this whole discussion is leading nowhere, I'm out.

I'll still read answers for courtesy.

[Mr. Scruffy]

Nice contradiction with your edit...

Procedural only mean that you generate following a method, randomness is something that must be added on top of it, they are separate things.

Ex : The planets of KSP are generated procedurally, will always appear the same way and are not randomized between players, something they don't need to.

Still, I'm giving you the doubt that it is all a semantic error...

Did you as I think (rereading 3 times your posts now) : generated ONE map where you procedurally generated the main feature (volcano, wind, rain...etc) and use a random tool to fill-up the crater/volcano...which will always appear ? OR... did you program something that actually generate a new continent/map/weather each time one play following only basic layout ?

Because, yes, if you don't attempt to randomly generate planet surface/geomes "each game", you indeed keep your geomes only procedural, not random. No matter if a noise generator is used to shape the original relief.

Still, to me it seem you still haven't understand how you are only changing the problems while solving the rover's one.

- increasing mindless click-fest.

- making rover dominant over static probes (though I'll agree it can eventually be accepted as a feature).

- worsening the balance of reward per biome/planets.

Anyway, I will stop participating to this discussion. Even if you were supporting me or I supporting you, your way of "discussing/editing" is giving me headache and I've better things to do.

ps : I should have been clearer that I meant 10 Area-of-Interest per planet, but it really should have felt obvious considering the number of them I gave as example for Duna.

[...]

I think i explained pretty well, what the generator did: First it formed the relief of the terrain with a process with a high random factor. Everything else was determined by the relief, basically. Cause, you know, typically you dont have deserts in regions with a lot of rainfall, for example. They´d usually form either at the western end of the continent or "behind" (again: west of) mountain ranges. This was not so, because it checked for coordinates to place them, but because humidity was carried by the predominant direction of the wind (westward from the eastern sea) and rained off gradually over flat terrain and massively when it hit a mountain range. This would typically be, where rivers would have their fountains, and typically, they´d flow eastwards and always downhill, until they´d reach the sea, occasionally forming lakes, to overcome dead-ends in relief, on their way to the sea.

It was intended to generate the playing grounds for an exploration game. Very much like ´random´ maps in civilization (to name just game which does this) do. Those are formed not purely based on randomness either (that´s why you can choose a ´script´ for generation - i usually use ´tectonics´, but there are many others). For some, you can, for example, set how many continents there should be. Yet, you wont know where those are, or their shapes.

If you´d replace the random parts of any of these scripts with fixed numbers, yeah, those scripts will turn out the same maps every time. Would the tectonics part of my little continental generator not work with random numbers, but determine a ´vulcano´ erupt in spot x,y at iteration z, raising the terrian there by n meters, and that for all ´vulcanoes´ and ´meteors´, it would generate the same continent every time.

The ´canvas´ forming the basis for the continental generation was simply a flat surface. Other pre-definitions: Wind goes east to west, equator is hottest, size of the map, rivers go downhill, biome-definitions... (in other words: some simplistic earth-science basics) and desired sea/land ratio (defining sea-level at leisure rather than logic). But the method could also be applied to a more defined starting state, just as well - the number of iterations on the ´tectonics´-level would then set how much the results would tend to deviate from the beginning state. I could have painted a map of africa for it and have all maps be alterations of that - problem would have been the congo-basin: I dont think the generator would have it (it´s west of mountains - sort of an anomaly). But as long as it wouldnt have to re-create really existing places, the results seemed plausible enough and stroke a nice balance between ´different each time´ and ´realistic and predictable within its boundaries´.

Bottom line (and i´d really like to drop this out of the topic, too - it´s sort of off-topic): We agree in the fact, that if there is no random part in procedural generation, the results of the procedure will always be the same, provided the input (starting state, methods, parameters, boundaries) has not changed either. If we start in the same system and we have the same definitions of where what interesting sites should spawn, then we will find them in the same places.

------

Now back to rovers - replying to your points in sequence:

- I am not sure where your line of ´mindless clickfest´ is. I dont think that considering where interesting sites might be and figuring a way to get there safely - for some will be hard to find, others hard to reach, by defintion - falls under my definition of ´mindless clickfest´. In fact, clicking should be the least part of it.

- Yepp, sort of the point of this whole thread, isnt it?

- Balancing science reward is an issue, i´ll give you this, but: Again, you assume the solution must be worse, than what we have now.

---

About my edits: Sorry, but i sometimes ponder these things when AFK and append stuff later - if i do edit stuff after a reply, it happens simultaniously (x-post).

Edited August 12, 2015 by Mr. Scruffy