Loading...
  OR  Zero-K Name:    Password:   

Can we make windgens not depend on altitude?

26 posts, 487 views
Post comment
Filter:    Player:  
Page of 2 (26 records)
sort

CZrankpsaniac
11 days ago
I am just posting this here because many people are complaning about windgens. Some people probably don't know that on some maps windgens are better then Fusions, Singus and even Adv. Geos. Because of this ( and the fact that they cost 35M, so they have no initial investment) many people will build 100 windgens instead of a Singu. The advantages to building thrm are: Harder to destroy (can be built on hills in the back only accesible by Fleas), no initial investment (Singu makes cost after is is built, 100 windgens contribute even if they have not been all built) and lack of a glorious shiny explosion.

100 Windgens have one disadvantage: LAG. This alone should be enough reason to make them unsuitable for late game.

Making Windgens produce 0.4-2.5E regardless of altitude would make them reasonable. It would also give people reasons to build Fusions. And it would make maps with currently ridiculous wind better. And it would mean more shiny Singu explosions.


+0 / -0


DErankkatastrophe
11 days ago
Lag is definitely a thing. Maybe rework the model so it doesn´t move?
+4 / -0



EErankAdminAnarchid
11 days ago
(edited 11 days ago)

quote:
Lag is definitely a thing. Maybe rework the model so it doesn´t move?

How do you know it's the model moving, and not for example grid calculations that causes the lag?

Capping windgen top output below singu seems fair as long as wind is still better on average than solar.
+0 / -0


LVrankSenaven
11 days ago
Fusion is already reasonable to rush from start in team games. Or even two. Especially in maps where is no wind or is pretty bad.
I wouldn't be against option that you can switch off windgen rotation in game options. Because there is some maps where players build them 300 and its massive lags.
+2 / -0


LVrankSenaven
11 days ago
(edited 11 days ago)

quote:
How do you know it's the model moving, and not for example grid calculations that causes the lag?

We assume it because 300 fleas moving around also cause seriously lags.
+1 / -0



TRrankFirepluk
11 days ago
(edited 11 days ago)

Develobs just need to move wind animation from cpu to gpu and make rotation speed constant
who gives a fuck after all if it lags like a dying horse, considering stupid zk is already pretty much single threaded :\
[Spoiler]

Same goes for flea/puppy collision detection and pathing ... those little shitters should be allowed to go -through- units just to improve late game performance... 3 FPS ain't fun
+5 / -0


PLrankAdminSprung
11 days ago
quote:
quote:
How do you know it's the model moving, and not for example grid calculations that causes the lag?

We assume it because 300 fleas moving around also cause seriously lags.

I don't think windgen spinning requires pathfinding.
+1 / -0

ZArankAstran
11 days ago
I like the variable element introduced by windgens. It rewards the people who take the time to study maps and altitudes. If lag is the problem then I agree that a constant rotation or even a still windgen is a better idea.
+0 / -0



AUrankAdminGoogleFrog
11 days ago
(edited 11 days ago)

The tree shader demonstrates that we may well have the technology for GPU spinning. The first thing to do would be to check whether spinning is actually an issue.

I like the windgen height boost, it gives wind some positional dependence.
+3 / -0


DErankkatastrophe
11 days ago
I like the height-boost as well.
If the movement contributes to the lag, it might be useful to get rid of it. Every little bit helps.
+1 / -0


LVrankSenaven
11 days ago
quote:
I don't think windgen spinning requires pathfinding.


If you send all fleas in straight line in flat map it takes so much pathfinding calculations (especially if distance isn't very big) to destroy performance? Its not even about 300 fleas, its enough with 40 - 50. Especially on so bad CPU as AMD.
+0 / -0


LVrankSenaven
11 days ago
(edited 11 days ago)

quote:
3 FPS ain't fun

Are u kidding? I playing most on 1 fps or even in back time when somebody build windgen farm. So much resigned games in larger maps because of 1 min back in time.
However game on 30 fps already feels like playing TBS if you trying to play seriously. Singu rush or nuke rush of course is not included.

quote:
yea. yea... tell me about MT support... how much does it bring? +15% fps which ends on 2nd core? :P


EGrankzerver was only one who bake in ZK forum promising stories about MT. And when devs asked him for code he didn't give it as i remember. Plus never show his face here again. So I don't have even fantasies about this 15% plus. Plus also zillions new problems with such new engine would appear.
Only thing which makes me sad is that i choose AMD over Intel and start playing ZK only after than. If i would know i would buy Intel and would not save this ~200 euros worth money. No, AMD FX series isn't so bad for medium gaming. And plus you have nice furnace in pc case who warms you in cold climate.
+2 / -0

USrankHun_Tzu
11 days ago
(edited 10 days ago)

Please no. That's the whole point of them. That's why they're fun.

Mexes and geos are a fight for horizontal territory. Windgens are a fight for vertical territory. And solars are just boring and meant as backup. Fusions are large boring solars with more fragility. Singus are beautiful liability nukes that require planning for isolation+defense, which is cool, unless maybe terra+razors defends them too easily in a boring way? (idk, I haven't really played FFA-speed-lobster much)

However, can we please remove the random changing wind? and just make it alternate between 25th percentile and 75th percentile every 30 seconds? As in, if the range in the old silly random-wind days (what were we thinking back then?) of ZK was 0.1-2.5, now it switches between 0.7 and 1.9 (average 1.3), if 1.3-2.5 -> then alternate between 1.6 and 2.2 (average 1.9). That way you have the exact same averages, but no randomness, and all you variation fanboys get to keep your variation.

Also, while we're at it, let's nerf solars. Seriously, who wants to see solars be the best option on almost every map even for your main base energy? Solars should be for vulnerable or isolated territory (such as overdrive for single mexes), like they are in real life as power sources for isolated electronics. Solars ingame can also be good connecting power to the frontline, which is fine, although I think that can be made a lot better, too.

Speaking of map-dependency, the way the wind-map algorithm combines with the maps we tend to have, makes it so that there's a ton of maps with a wind range of 0.0-0.8 or whatever, a few with 1.0-2.4, and not a lot in between. I'd be interested in an overhaul of the wind-map algorithm so that there's an approximately even distribution of wind-power among our map pool (either biased towards less wind or honestly I wonder if a mostly even distribution of map-averages wouldn't be so bad). [Edit: Tell me if I'm mistaken here, I don't have all the map stats in a table in front of me.]

I have a lot more thoughts on this and not the time to explain everything, but here's what I'd like to test out:


- Windgen changed to non-random as described above
- Wind-map algorithm changed to either
A) (very roughly, these numbers are just an example of the algorithm's style) 40% of common maps with an average 0.3 wind, 30% with .8 average, 20% with 1.2 average, and 10% with 1.7 average
or B) 25% 0.3 wind-average, 25% .8, 25% 1.3, and 25% 1.8

- Solars cost increased 20% (test a range between 15%-40%) from 70 -> 85
- Solars grid range increased +25%

- Windgen health increased +10% (test a range between 10%-30%)

- Pylon cost reduced 50%, 200 -> 100
- Pylon nuke-on-death damage reduced 50%, bullshit -> bullshit/2

- Fusion cost reduced 25% (25%-50%) [intended to make fusion not such an unrealistic jump in 1v1 or competitive settings]
- Fusion's stats all reduced 25% (25%-50%) (energy production, health, explosion damage)

- Geothermal's cost decreased 25% (10%-50%)[They don't seem remotely useful to me in a competitive setting, given their underwhelming cost benefits over solar/wind, low health, pylon connection requirements for overdrive, their coarse-granularity cost-jump (expensive things that come in large chunks means that you have to build them way too early or spend time with a power deficit, and spend much longer without payback during construction), their inability to choose their location in an convenient and defendable location, and their obvious locations marked with massive beacons for the enemy)
- [this one's just something I'd like to see] Eliminate geo's explosion-on-death (reduced to the minimum appropriate range+damage) [it's absolutely infuriating to me that something labeled "geothermal" would be dangerous. Geothermal plants IRL are the safest, tamest, most boring, most innocuous things you could possibly imagine. Flagpoles are more dangerous than geothermal plants.] Either that, or change the name to fusion reactor, and name fusions geothermals and make them safe.

- Current heavy weapon power requirements reduced from 50 E to 10-40 E, but have them actually consume energy while firing (after each shot, they consume their respective energy while reloading)
- Give power requirements to Bertha (50?), Nuke (for stockpile; not for firing), Anti-Nuke (for firing, although anti's mechanics are so janky), DRP (100?), Zenith (500?), and Starlight (1400?), and have them consume said energy while firing. Discount their prices by their equivalent cost in fusion construction.


Actually yeah, I love that. Make it so that Starlight is practically free (1000 metal-ish) and just an energy-focusing laser that is entirely based on energy infrastructure. I actually love that so much, I would love to see this.

So like, you have to build a massive energy farm and defend it to keep starlight operational. You can have other projects using the energy in the meantime, and then when you turn on Starlight, you choose to cannibalize all the energy from those projects and divert it to the starlight. So to get starlight up as soon as possible, you power-down all your overdrive, berthas, antis, cloaked units etc. and have to continuously maintain your energy grid requirements. Have Zenith be similar but not as extreme, so enemies can power it down and make any stockpiled comets fall on you.

Typed on phone, pardon the typos.
+1 / -0

AUrankisaach
10 days ago
I think energy in general in ZK is very confusing and I still dont understand what is happening alot of the time. Im not even sure how to view if a map is good for wind or not. My suggestion would be make windgens -only- depend on altitude and not map specific.

Map dependent wind turns it into a noob trap on some maps. Memorizing the wind levels on 30 different maps is boring metagaming that dosent add anything meaningful as far as competitive play and takes game balance out of the hands of the developers and into whoever last updated the map 8 years ago.
+0 / -0

PTrankraaar
10 days ago
(edited 10 days ago)

- if the problem is gameplay incentives and economy snowballing: use the "zkminwindmult" mapoption on more of those maps with cliffs to reduce the minimum wind to sane values (0.3-0.5)

- if the problem is performance, I have a magical solution: triple the cost, attributes and build footprint of wind gens. This way you get a third of the performance cost for a given metal investment.
+2 / -0

USrankHun_Tzu
10 days ago
(edited 10 days ago)

I believe I found the algorithm on some wiki/guide, and I believe it's standardized.

In fact, I believe the exact opposite is a serious issue that mapmakers complain about often: not being able to set the wind value for the map peak - resulting in janky or non-existant workarounds.

And exactly, I didnt know performance was an issue, but I don't see any reason windgens need to have such fine granularity anyway. So the solution would definitely be to just uniformly multiply them. A lot of windfarms are super-annoying to construct and path around, too, because they're so small. And I really don't care if that eliminates their chain-destruct feature either. Again, like geos, wind turbines IRL are stupidly innocuous.
+0 / -0

CZrankpsaniac
10 days ago
Is there any reason why you people like the height boost? I like it in theory, but in practice it doesn't seem to result in better gameplay.
+0 / -0

USrankSteel_Blue
10 days ago
(edited 10 days ago)

People are complaining about not knowing about wind mins, so let's make it easier to see. We have economy overlay, which shows reclaim and metal, but for zero-k the metal graph is irrelevant like 98% of the time, and sometimes doesn't even reflect Mex spots on maps where zk spots are hacked into place(I think the map named alkion wastelands is a good example, name could be wrong). If we could get some height lines labeled for their min wind in the economy map, that would be cool.
+1 / -0


USrankFealthas
10 days ago
Most of the time the decision of whether or not to build wind gens is based entirely on the map, not the terrain of the map. Some maps just have like .7 min everywhere and there is no choice BUT to build wind. Others have like a .1 min and you can't build wind because it will drop to .2 and ruin your eco for a bit.
+1 / -0


DErankkatastrophe
10 days ago
on a sidenote, why don´t work windgens on obsidian, but solars do?
+1 / -0
Page of 2 (26 records)