Vehicles Feel too "Floaty" Missing Details

Yeah I know it from there as well. I just tested if my idea of how it works is still correct and it seems to be unchanged. It is basically one way: You can add units and units you added give you some very limited data. Units created by gameplay mechanics are not listed.

Logging them by using JSON.stringify yields this:

First the sim object, then every unit object. The 2 units are the commanders. These are the only units ever added to the array, because they are spawned via the spawn unit cheat.
When this was logged I already had build like 10 walls. I still only have 2 units in the units array.
The field "dead" on the units is updated by the sim, that is how the script knows a player lost the game. It polls that information.

So basically with this the script can spawn units and for the units it spawned itself it can tell if they are still alive or if they are dead. That is all it can do afaik.

So that thing is not an iterator or any other representation, but merely a variable which is monitored (and partially updated) by the sim....

Btw.: What exactly is that "sim" object anyway? Or the "server" object? Both come from imports which are not in the file system.

If these things are anything like the modules in node.js, then this doc could eventually help:
http://nodejs.org/api/modules.html#modules_the_module_object

I have no idea where they come from but they seem to be very limited way to communicate with the native layer. However there is close to no API to do any sort of gameplay scripting. Everything is limitied to the bare minimum the current gameplay needs. Which isn't much.

Well, then let's see if the apparently node.js inspired module system can give more insights.

Thanks.
So GL_MAP_PERSISTENT_BIT with immutable storage it is.

EDIT: Looking at the particle system documentation, my feeling is that an on-GPU update for particles would be a rather complex shader.

Yep, that's why it's not a GPU particle system. We can't do millions of particles, but we can do hundreds of thousands with a more control.

I actually agree with this (in principle and in part*); though i might have challenged this idea before on the basis of a general increase in match time leading to fans of 15-20min max games having a rage. Its well past time we took a look at this again (though perhaps in another thread?) with the increased eco of commanders allowing fast and first factories it feels as if the rapid production of units (made even more rapid in beta i think?) isn't needed as much/at all any more.

*the zerg part is where we differ, PA's scale was always intended to be magnitudes greater and losing that for some treads will be lame IMO, but production should go down and hp should go up somewhat also IMO.

It's not just treads, it's a whole plethora of missed graphical opportunities because of the sheer number of units in this game. But it's more then that, it's not just about graphics. One has to ask, is the current level of units needed to make interesting and meaningful gameplay? Can that be achieved with slower build times and more health? Would more HP make strategy and unit placement more important? I honestly think the game would be more interesting and play better with slower build times and larger hp pools. I am not a big fan of the current one shot mechanics, it strips strategy out of the game and turns it into a numbers game instead of a strategy game. TA players slower but better, with more personality in units and a bit more build strategy. It still manages to keep a huge sense of scale that PA is going for.

But you are right, perhaps it's best for another thread. I still think it has some merit in this thread because it's one of the largest limiting factors as why we can't have some graphical polish on basically everything (supposedly, I bet there are some workarounds).

I also think blaming the lack of graphical polish on unit numbers is a cop out on Ubers part. Look what modders did with naval trails! Why isn't that in the vanilla game? You'd think Uber would have enough sense to add something as simple as that. It doesn't take a genius to see that water and environments in generally are severely lacking. The first thing I thought when I built naval units was "This doesn't look right, something is off here." Same goes for land based units.

Understand the lack of some of these features might have something to do with performance but these were influenced as much by the zoom level at which the game is typically played, and possibly budget constraints if not actual financial budget but also time; And yes the art team is separate from the others but we don't know all the details, they might be retained on part time basis for instance or any number of solid explanations as to why they're not working 24/7 to satisfy the 'need' for frills that won't get noticed more than a couple of times a game if that. This isn't me saying i don't want them in some way shape or form, I do, but I am saying you're coming across quite a bit too demanding IMO, and not just you. I highly recommend you make that other thread, there are game-play considerations for slower built, tougher units as well as mere performance and gloss.

Hmm, I didn't feel like I was being particularly demanding. I am ready for some change and improvements though. I remember reading similar posts about the graphical improvements and lacking features nearly a year ago. Yet, little has been done nearly a year later for a lot of those complaints. I think it's important to stay adamant about wanting these features so one day hopefully they are worked on. I follow many other indie developers and they definitely iterate on these types of things faster. Based on my experiences with other companies I would definitely describe Uber on the slow side of things. That is my experience, maybe not yours. I also don't agree with a lot of Ubers design decisions and I think there are a lot of things that could have be done better. In fact, a lot of the graphical effects you'd come to expect in ANY game are completely missing from PA. A lot of these (like ship trails) you'd expect at release and I find the complete absence astonishing.

I am semi-tired of hearing it will get fixed or it will get there eventually. If I come back to these forums another year from now will they be fixed? How about 2 years? How many years have to pass before it's acceptable to complain about such things not getting addressed? I hope you get where I am coming from. You can have patience to a degree, but at some point there has to be a cutoff. For me that cutoff is now, I paid them to support and help their development nearly a year ago knowing the game wasn't finished. I've waited, and nothing has happened. So now I am making forum posts to get my opinion across on things I'd like to see improved / fixed. That's sort of what forums are for...

Uber achieved a miracle to create a game like PA within 2 years. They basically started with a blank file of nothingness.
To compare Supreme Commander took 4 or 5 years.
Uber may be independent, but the game they are working on has the technical complexity of AAA titles.
Also yes the graphics stuff certainly is "lacking" in details. However I don't think it is generally required to have all those nice little graphics features you describe. They are all only nice to have. Makes sense PA has not many, because the time constraints on PA have been pretty harsh so far.
I would not be unhappy if the graphical style of PA doesn't change much in the future. In the end I am mostly watching icons fight icons anyway.

this won't solve anything but merely delay it .. and yes i am here for metalavalanches as was the idea of @neutrino

do we realy need trails?
do we realy need scorchmarks ?
do we realy need missilesmoketrails?

if your game runs slow you could turn those off and still have the same gameplay ... what you suggest however is rightout gameplaychanging and that i disagree with ..

Actually, we need neither, at least not server side, as there are better alternatives for the information domain. Simple smoke particle emitters client side are sufficient for the aesthetics.

All 3 serve the same purpose, giving you a persistent hint of what happened even if you were unable to attend due to multi tasking.

And that's the issue, you NEED some type of working tracer which goes beyond a notification in the Chronocam. If something fired at you, you wanna know exactly where the shot came from and what unit it came from, without examining the projectile manually. If something passed your vision, you wanna know exactly what you last saw and what it looked like when you lost track of it. Otherwise, reading that information is a very tiresome process which forces you to jump around in Chronocam if you want to access all available data.

Scorchmarks and trails give you instant access to SOME of that information, at least you get a persistent quantor for the amount of units or destruction events for the recent past. By no means the best solution, but in lack of alternatives.

Oh, and yes. They give a "used look" to the battle field. That's somewhat typical for most RTS, the gamefield changes in appearance as it's cluttered with leftovers during the battles and/or as natural resource patches are being used up. That simply gives the player a feeling of progression, even if they happen to be locked in a stale mate with no gains for the past 10 minutes.


Ok, it's not exactly what the OP asked for, he wanted a more aesthetically pleasing client side unit movement physics, but the thread derailed in that direction.

its pretty simple the more stuff you put up the more resources are needed to calculate that stuff ... and i rather reduce purely grafical/cosmetic/aestetic shikki mikki than change anything that has to do with actual gameplay ...

It might be not what the original poster asked for, but I like your reasoning much more than just demanding aesthetics for itself.

Why not have both? Hire another art member on the team if the current one is busy pumping out 100'n something commanders.

I am sure Uber will gladly do so if you pay for that.

I"m sure Uber is rolling in dough from all those commanders in the store Uber spent resources designing instead of other core features riiiight?!?!

Well they don't really have much choice though do they? They promised them as part of the KS campaign- maybe a bad decision (although we'd need the numbers to be sure of that) however I did note when they attempted the new Kickstarter for HR they *didn't* offer any custom units even at the 1k tiers. The main thing is they've well on their way to finishing the custom commanders, so hopefully it won't be too long and the art people will be free for other things (have you seen the new walls in the PTE btw? Mini shield segments rather than blocks, a really nice improvement ).

Who knows. I sure hope they are. But considering they had to let go people shortly after launch to be able to savely plan for longer development I doubt that. Uber has shown in the past that they are willing to hire more people if they have the money. For example during the initial 90$ sale PA was #1 in the steam sale charts for a while. Shortly after a few new devs appeared in the forums "I have been hired because sales were good".