Scale Megathread

No, no of course not, my wording was just trying to simplify things. Of course it's not going to be perfect first try, but this is missing the point which is whether or not to attempt the first balance pass before or after all components are included. "Balance" here can refer to unit balance or size balance (relating to this thread).

I consider two kinds of balance - Macro, which the large scale balance, involving large adjustments and fundamental changes to get all units roughly "in line" with each other. Then comes the micro balance, involving much smaller changes to individual units where needed. I am referring here to the macro balance phase, as introducing large amounts of content after this has occurred can necessitate a second round of macro balance, so it only makes sense to wait to perform this until all the content is in.

If you guys watch the youtube video released today, I think it will be very clear that the terrain scale was altered a bit. I think it looks much much better! Go check it out if you haven't already:

It seems the landscape elements (mountains/rift) are sized up, which is way better as they become real natural frontiers instead of useless props.
But the scale between the planet and the units seems to be the same, there's still this cartoonish superdeformed feeling from very tiny planet.

Scale between the units and the planet can be changed by increasing planet radius.

"bugfixes balances tweaks and gunships" if they'd spent all the time they said it would take to redo the props you'd think he'd have mentioned it.
I don't see the difference.

me neither

It's been mentioned before, but I'd like to throw in my two cents. Unit scale and unit speed are related, but not extremely important. If units were scaled down a little, I don't think their speed would look too odd. bots are supposed to be fast, and tanks and the commander already appear quite slow. Unit scale is not heavily related to planet scale (at least directly); planets have a limited max size right now, but could be larger, so I don't think that's too important either.
Unit scale is important relative to the scale of terrain features (and those features are important relative to planet scale). If unit size were shrunk by 15% or so, it might help, but I think the suggested 30% size decrease is overkill. The real problem here, and I'm sure a lot of people feel the same way, is that the current terrain 'props' are too simple and poorly scaled. Right now the desert biome has a few good props that allow a slightly higher ground on which to build (though I don't know if these actually provide tactical advantage, they have minimal elevation differences and they're usually too sparse to be of use), but most other props are simply holes in the ground or small natural walls.
If and when these terrain features are changed (and there are a number of other aspects of it I think should be changed for gameplay and looks, but writing about those would be going quite off-topic, and I don't know the terrain system's limitations), I have a feeling unit scale won't be criticized nearly as much.

I've been thinking this for a while and while building scale can be decreased to better fit planet sizes, i do think units need be scaled up in relation to their factories... The best example is the Leveler. Huge double barreled tank, looks puny coming out of such a huge factory...

Ships do it way better in tis approach i think...

So my two cents:

Overall size down, size relation units - production facility: up

they are being changed, they've been changed. haven't you noticed the new small inclined plateau that starts off a bit above land and ends up a bit under? it's meant to create high ground low ground but it's too small for the units to even interact with! a texture pattern on the ground is what it's comparable to.

the units are still way to big in comparison to their props and the planets way too small as well in comparison to it's props.

Make the props bigger, not the units smaller.

I don't care, either provide a solution due to the planet size slider.

You mean the desert biome plateaus? I mentioned those. And yes, I also mentioned that the height difference seems too small to matter much even for those. A whole lot more needs to be changed than just that, and yes, I know things are being changed, but we don't yet know in which ways. Mountains are currently small spikes with abrupt sides; there are no mountain ridges or cliff faces that affect terrain height, mountains do not have as wide a footprint as they should and air units have no problem passing over them. IMO cliff faces should be fairly common in many biomes, especially along shorelines, but I don't know if it's feasible to implement that or not.
As for saying that planets are too small relative to their props, I can only disagree with you. We have limited planet size right now, and the props seem fine relative to larger scale planets.

Would it be possible to bring out a patch where we the player base could test around with like a set of unit sizes which is defined in the match setup?
Also gamespeed and or energy/metal rates as well!

Also build in a way to easily poll which game modes are used the most or ask the nice PMMA guy to add it at least for PMMA users if you haven't done the first one already.

That doesn't sound like to bad of an idea. It would just be difficult to manage, let say a random match in the game screen has Metal rate set at (random number) game speed set at (random number) Unit sizes set to, i don't know, (big). That one test could be a horribly not fun version of the game because the planet selected was to small to fight on.

A good test needs strong, simple independent variables and one or two dependent variables to see clear results honestly.

Where would the poll exist, and what is PMMA?

I can simply agree with all of this. While biome concepts are just concept, i can still remember my excitment when i saw the awesome desert biome. We're far from even 10% of what the desert biome concept promises. I easily understand this is something probably not easy to implement in a procedural stuff, that Uber is working on all fronts, but i also admit that sometimes it's a pain to wait for things to change. Hopefully all this is going to accelerate when Uber is done with core features (well actually, biome complexity is some kind of core feature).

by polling I mean polling data, this doesn't necessarily include people being polled
obviously the settings would need to be obvious
all options I mentioned were part of TA as well as SC and SC2 (ok not the scaling, but there were mods to do that easily)
so there goes the rest of your argument
PMMA is a modmanager, tho i just realized I meant PAstats not PMMA

WHO DARETH AWAKEN ME FROM THY SLUMBER??

The settings need to be obvious? Again, this is a problem to the user.. just describe how you would have the poll settings oriented and I won't be confused

I know the game speed settings in Starcraft 1 in the options menu. Are you saying, have that in the options menu and have people adjust it to a setting they find fitting when they battle against an AI?
Do you want the metal and energy to function the same?

Also, I was never arguing against you, just stating facts.

well how would a speed-, ressource-scale or unit scale setting not be obvious?
most people know what it does, most people had math
<common sense the words ratio and scale, in most languages it's more or less the same word

I'm sorry I didn't explain the poll thing at all (google polling computer science)

If they set up a mechanism to poll certain data that would happen automagically,
and they do this for a load of other things, they need to do that to properly optimize their game
(polled data for example is how many matches were played, who logged on when
and played games when, so adding a function that logs the use of sliders as well isn't much of a problem.

Gamespeed is really obvious, I doubt there is anyone that could misinterprete
Ressource collecting rate is as well, it's even in the game,
just that you can't adjust it for everyone, just yourself and the AI,
this is for testing purposes (testing new maps) I guess.


Yeah, I want the metal/energy function to be the same, it gives people the possibility to play a game at a pace they like, and believe me it's easy for people to come to an understand which gamespeed or ressource rate they like.

People would hopefully be protected by common sense from chosing tiny planets and gigantic units, but all the thread is about downsizing the units, see poll results.

Even if people were too daft to understand it, trying it once had never hurt anyone,
life is about trying, if you don't try, you don't succeed.


I was put off by that, a test can be anything, obviously there's worse and better tests, but that's like opinion?
Lots of people died testing things for the better of human survival (natural poisons, Marie Curie and radiation)
In science we sadly don't have many strong, simple independent variables since life is pretty holistic, tho we try everything to make perfect testing environments, because at a point you need what I quoted from you, the same way I made my ~10000 words long summary/explanation of the System Manager, tho I should urgently add the current 12 planet limit (dude I could swear I used 13 planets yesterday, only one that didn't load was the one with 93)

anyway without that and the word horribly I would have probably put more time into explaining polling and would have been friendlier I guess, as well as all the fallacies (strawman, leading the question)
especially the random number one made me go into mental defense (random number? inbetween where? any number? how'd that be feasible? if that'd be released devs would have tried them out, adjustable gamespeed (adjustable EVERYTHING) has been around for a long time. <common sense



So I wondered if you lost/ignored your common sense or just use all those eristic arguments for other reasons.


With the 'test' I had in mind, they would have polled this data over a long time, they then could statistically analyze it,
then they could see how many people chose 0.7 unit scale 1.2 metal ratio and 1.5 energy ratio
that way they could see
how many people chose what and what was chosen most, from that data they could come to a conclusion.

They want to know how much to resize the units, let's say 50% of people chose 0.8, 30% chose 0.6, 10% chose 0.5, 5% chose 1.2 and another 5% chose 0.7 then the average of that would be 0.725 and it'd be stochastically the unit size most people would be pleased with, tho with the game setting and universe it's not likely that such a scale goes further than 1.5 in testing and 1.2 in release (it would increase the performance hit twice fold, larger units would have to be rendered and since tiny planets would be too small for a lot of players, they'd just bump up the planet size (in a thread I made i explained how the planets relate to conventional 2D RTS maps and what performance hit which sizes have.

This is the same way the evil corporations test and find out everything about us, they poll data on us, or gather, dunno I'm just used to using the term poll in this context, not every word means the same for everyone and most words have one to two dozen meanins, tho there's a few that have less, and I'm obviously ignoring grammatical particles (particle being a perfect example, for someone who writes a graphic engine particles are fancy stuff, to a physicist they're an elusive breed that like to do odd stuff, like be in two places at once or be a wave while being matter. To me a particle is a lot of things, but in this context) tho even they often have different meanings.

Why do you insult me, I was genuinely curious on what poll methods you would choose.

so you're saying there would be a sandbox option when YOU vs AI that you can then adjust :

• Game speed
• Metal income
• energy income

okay, that sounds feasible. I wouldn't be against this idea, maybe have it so that the settings are locked in the beginning. However, wouldn't it be easier to auto run the AI's and see how well they expand or demolish each other and then test that "setting" that gives a good pace and resource collection on the masses?

To better write this statement. Let me give you an example

People who perform better on exams eat an apple before the exam.
The Independent variable would be taking away apples from 50% of the test takers and have the other half eat an apple before hand and see what the exam scores would be.

Maybe I didn't properly read your statement, but you didn't specify what would be adjustable variables and in what context. And it seems that you took offense to that. I'm sorry I guess.