You were asking what polish is...
weezl
Join Date: 2008-07-04 Member: 64557Members, Reinforced - Shadow
I think the most important part is that everything is in synch! Literally: "What you see is what you get!".
From small details like comm not seeing dropped weapons and seeing enemy particle effects way outside LOS.
To the most crucial things like the current delay built into all interaction between players which for example makes the skulk die as soon as he notices the marine shoot, but the marine has emptied 1/3 mag already at a still skulk. When both sides should have (nearly indistinguishably) identical version of the encounter.
This is completely gamebreaking for me right now!!!
Same problem as marine; when you SEE the skulk being in melee range he actually has already bit you 1-2 times! This also causes skulks to have "1m long teeth" which is completely absurd and so infuriating to die to.
To here comes also the retardedly stupid random skulk twitchy movement when it's hopping around like a spazzed kid on caffein as you're trying to aim at it with the rifle. It's like trying to stab a fly with a needle!
Another thing is responsiveness and ability to command as com. As in being able to micro arcs and whips in crucial situations!
Right now this is insanely frustrating/impossible due to lacking features and various issues. Like the need to double click for order bug.
Un/rooting has to be done individually while all selected ARCs can un/deploy.
Arcs seem to be extra screwed up this patch (212) with the blank firing problem which just adds to their UN-EASE OF USE!
The other major issue is they seem to need UNINTERRUPTED vision of a target during the firing cycle to hit it, otherwise a blank shot AND PLAYER LOS DOESN'T HELP ANYTHING!
In NS1 the cycle is [shoot then medium cooldown], that's why LOS and scan guarantees atleast the first salvo.
In NS2 the cycle is [long charge then shoot], AND if anything goes wrong during a cycle the shot will be blank - pure BS!
The only sure way is to overscan. And now with the retardedness of scans costing TEAMRES and not energy this issue becomes really big!
There's a reason why StarCraft 1+2 are the worlds by far leading RTS, and why C&C3 f-ing isn't.
The reason is SC does things right - support abilities cost ENERGY, units+buildings cost RESOURCES. This is INTUITIVE and is also the established standard in the gaming business for that reason.
C&C3 had support abilities costing resources (among a couple other differenses) and who remembers that game 1y after playing through the SP campaign? - No one.
Also, in SC, there's no unit that can place a limited amount of free offensive structures that die with it!!!! All buildings cost resources!!!!
Meds and ammo are "created" just like buildings, should absolutely cost TR, also from a pure gameplay perspective.
Nano armor/build and scan are energy based support abilities and should therefore cost freaking energy!
Make them: 1. more expensive like 50 or 100 energy (scan, maybe 30-40) and 2. make the starting energy of CC half of armor, that gives incentive to get multiple command points!
As for aliens it makes even less sense to remove energy - THEY ARE SUPPOSED TO BE AND SUPPOSED TO ACT ORGANIC!
Even if it makes sense from a pure gameplay perspective in comp context, as far as comps go you should have a f-ing-bright-lit, white-walled game on lowest settings where enemy players are red boxes and teammates are green boxes, cos this is what they would do given the chance (at that point you could just play Warsow or QuakeLive) so i would take comp opinions with a grain of salt.
Anyway, how can one building chain-cycle through alot of saved up res (say 10 cycles)? Wouldn't it get exhausted? It's not like an electrical apparatus, but a living organism.
Here's a thought example:
Can you as a human lift a really heavy stone (say 50KG) up 1m? Probly once, maybe twice - then you'd get exhausted and need rest. After maybe 3-4 lifts you need to rest for a long time, then after 5-6 lifts you can no longer do it no matter how long you rest - you need to recover. This sounds realistic and how living beings work, yes?
How about this: if you have 10 energy drinks you can lift 15 times, 20 drinks 25 lifts, 30 drinks 35 times, etc. Absurd, right?
To sum it up what polish is: what you see determines what you expect and something polished behaves as expected.
edit:
no alien spawning when eggs are blatantly available is stupid.
bilebomb firing farther than Gl is back now that BB returned to gorge...is also stupid.
sentries are pointless, take absolutely no dmg, a skulk kills one in 5 secs - BAM TEN f-ing TR GONE POOF! stupid.
MACs still need a universal auto-repair (that can be switched off for buildings/mech-units/players* separately)
*off by default. or and maybe not auto, but marine has to use idle MAC.
From small details like comm not seeing dropped weapons and seeing enemy particle effects way outside LOS.
To the most crucial things like the current delay built into all interaction between players which for example makes the skulk die as soon as he notices the marine shoot, but the marine has emptied 1/3 mag already at a still skulk. When both sides should have (nearly indistinguishably) identical version of the encounter.
This is completely gamebreaking for me right now!!!
Same problem as marine; when you SEE the skulk being in melee range he actually has already bit you 1-2 times! This also causes skulks to have "1m long teeth" which is completely absurd and so infuriating to die to.
To here comes also the retardedly stupid random skulk twitchy movement when it's hopping around like a spazzed kid on caffein as you're trying to aim at it with the rifle. It's like trying to stab a fly with a needle!
Another thing is responsiveness and ability to command as com. As in being able to micro arcs and whips in crucial situations!
Right now this is insanely frustrating/impossible due to lacking features and various issues. Like the need to double click for order bug.
Un/rooting has to be done individually while all selected ARCs can un/deploy.
Arcs seem to be extra screwed up this patch (212) with the blank firing problem which just adds to their UN-EASE OF USE!
The other major issue is they seem to need UNINTERRUPTED vision of a target during the firing cycle to hit it, otherwise a blank shot AND PLAYER LOS DOESN'T HELP ANYTHING!
In NS1 the cycle is [shoot then medium cooldown], that's why LOS and scan guarantees atleast the first salvo.
In NS2 the cycle is [long charge then shoot], AND if anything goes wrong during a cycle the shot will be blank - pure BS!
The only sure way is to overscan. And now with the retardedness of scans costing TEAMRES and not energy this issue becomes really big!
There's a reason why StarCraft 1+2 are the worlds by far leading RTS, and why C&C3 f-ing isn't.
The reason is SC does things right - support abilities cost ENERGY, units+buildings cost RESOURCES. This is INTUITIVE and is also the established standard in the gaming business for that reason.
C&C3 had support abilities costing resources (among a couple other differenses) and who remembers that game 1y after playing through the SP campaign? - No one.
Also, in SC, there's no unit that can place a limited amount of free offensive structures that die with it!!!! All buildings cost resources!!!!
Meds and ammo are "created" just like buildings, should absolutely cost TR, also from a pure gameplay perspective.
Nano armor/build and scan are energy based support abilities and should therefore cost freaking energy!
Make them: 1. more expensive like 50 or 100 energy (scan, maybe 30-40) and 2. make the starting energy of CC half of armor, that gives incentive to get multiple command points!
As for aliens it makes even less sense to remove energy - THEY ARE SUPPOSED TO BE AND SUPPOSED TO ACT ORGANIC!
Even if it makes sense from a pure gameplay perspective in comp context, as far as comps go you should have a f-ing-bright-lit, white-walled game on lowest settings where enemy players are red boxes and teammates are green boxes, cos this is what they would do given the chance (at that point you could just play Warsow or QuakeLive) so i would take comp opinions with a grain of salt.
Anyway, how can one building chain-cycle through alot of saved up res (say 10 cycles)? Wouldn't it get exhausted? It's not like an electrical apparatus, but a living organism.
Here's a thought example:
Can you as a human lift a really heavy stone (say 50KG) up 1m? Probly once, maybe twice - then you'd get exhausted and need rest. After maybe 3-4 lifts you need to rest for a long time, then after 5-6 lifts you can no longer do it no matter how long you rest - you need to recover. This sounds realistic and how living beings work, yes?
How about this: if you have 10 energy drinks you can lift 15 times, 20 drinks 25 lifts, 30 drinks 35 times, etc. Absurd, right?
To sum it up what polish is: what you see determines what you expect and something polished behaves as expected.
edit:
no alien spawning when eggs are blatantly available is stupid.
bilebomb firing farther than Gl is back now that BB returned to gorge...is also stupid.
sentries are pointless, take absolutely no dmg, a skulk kills one in 5 secs - BAM TEN f-ing TR GONE POOF! stupid.
MACs still need a universal auto-repair (that can be switched off for buildings/mech-units/players* separately)
*off by default. or and maybe not auto, but marine has to use idle MAC.
Comments
<b>It depends on the outcome you want.</b>
Anyway, the one thing that stands out to me is your issues with latency, the skulk dying just as he realises the marine is shooting etc. I've seen that, it has happened to me and I know it's frustrating. Before I respond to that I want to go over some background info to make sure we're on the same page.
<u>Latency</u>
As a general rule, there are two factors that contribute to latency. One is how fast the game runs, particularly on the server but also your PC. The other is a real world limitation imposed by the laws of physics; the distance between you and the server. In order for your compter to tell the server and then relay that to the other person's computer the data has to travel over a physical distance. Your ping plus the other guys ping. This is the minimum number of miliseconds for that kind of stuff to happen, even if the game code is instantaneous. Unfortunately phsyical distance isn't the only factor affecting ping because, just like a freeway, traffic can back up and increase your travel time.
What you see on your screen, what the server thinks is going on and what the other guy sees is all different for any one moment in time. Hit prediction can make up for that by adjudicating the claims made by each game client, but it will never be able to relay that a guy is shooting at you in advance of when the server first finds out.
The best ping I've ever had, 25ms, was to a sever in the same city as me. Probably on average, people are happy to play with a ping between 50 and 70. Above 100 you start to notice the kind of stuff you're talking about. The point that a lot of people don't seem to realise is that in many ways, the other guy's ping is just as important as your own.
Also check out <a href="http://www.humanbenchmark.com/tests/reactiontime/" target="_blank">this test</a> on reaction speeds. I found it interesting because the brain can kind of trick you in to thinking it reacts instantaneously...
<u>Back to the issue</u>
So if you already knew all of the above then I trust that you're not asking UWE to bend the laws of physics, and that you're just concerned with server performance. We all know server performance is a problem and UWE have said many times, many different ways that they know, they're working on it and it's one of the highest priorities. The success of the spark engine hinges on fixing server performance because when the game launches, the wider community will not tollerate the need for highly overclocked servers like we've had access to during the beta. The burden is already on them.
What you can do to improve the game experience for yourself:
<ul><li>Check that the server you're playing on is running well - type net_stats in the console and check the tick rate. 30 or more is good. If you start noticing problems, check again because the tick rate varies over the course of the game, generally getting worse as more structures are built.</li><li>Check that your computer is running NS well - type r_stats in the console and check your FPS. If you're getting below 30 it's just like having a higher ping.</li><li>Take a look at the ping of all the players. This will give you an idea of what to expect. When Europeans or Americans join the Australian server I play on, they have 200+ pings and it really changes the game experience. Once you're aware of that you can factor it in to the way you play. Eg: you're less inclined to poke your head around a corner to take a peek because the server will probably decide that it's been blown off before your game client starts to say "hey, you're being shot at".</li></ul>
#87 :)
<strike>Ooh, I got #48 with my 203.4 ms! :p</strike> That link has the potency to completely derail this thread..... ;)
EDIT: 45. Angelusz just seconds ago 5 198.6
EDIT2: Since I like the idea of it (and to keep this thread clean) I've made a new thread in offtopic, to play around with this test: <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=119336" target="_blank">over here!</a>
---
As for on-topic. Weezl, I know your pain. I dare even say that everyone that currently plays NS2 does. Its been said often before, and I can understand if you get tired reading it, but you'll have to wait for UWE to get to the polishing part. Better yet, a great deal of their current time is spent on this. There's quite a list of bugs that need to be fixed and the performance issues are known. UWE has the fullest intention to solve every problem you list in this thread. They just need time.
Don't worry, it won't end up being a game with messy controls, latency issues and a load of bugs. Upon release, they intend to have it running like a charm. Just give it time.
Also Kudos to Khyron. Very good post.
Also Kudos to Khyron. Very good post.<!--QuoteEnd--></div><!--QuoteEEnd-->
Agreed, i'm a little dissapointed at the progress on improving performance, if they fixed that it would make me want to play ns2 more than any additional features such as exo suits would. As it stands now I think to my self, 'hm some ns2 would be fun... But nah I can't be bothered with the cumbersome online gaming experience due to server and client side performance'. That said, i've still have had some fun times playing ns2, that is when i'm feeling patient...
I wrote this when the change came in to highlight another flaw in this system:
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->First off I will concede that the current implementation is the first
pass and hasn't really had had a chance to be balanced. Unfortunately
I believe that, at least on the marine side, the single resource model
WILL NOT SCALE with variations in player count.
Increased players mean that the number of commander abilities that can
be used on a per player basis will always decrease as the player count
increases. If Tres flow is increased based on number of players to
allow the comm to have consistent support capabilities across varied
game sizes, marines will afford upgrades far too quickly. This has
been seen in the latest patch where the marines seem to be swimming in
Tres in very early stages of the game. I'm guessing that varying the
cost of tech based on game size is something you do not wish to do
(even more complicated).
Giving the commander free abilities with soft-cooldowns could be a
solution, however cooldowns in my opinion don't fit in with an RTS
experience and I recall you stating in the past a preference to avoid
cooldowns. So the choices seem to be either reduce the amount of
support a commander can give in larger games, or allow large games to
progress very rapidly to endgame.
A POSSIBLE ALTERNATIVE
______________________________________________________________________________
A solution that is apparent to me is a simplification of the energy
model seen in earlier stages of NS2 in three steps.
1) Unify all commander energy into a single pool with a much lower
maximum value.
If a beacon costs 15 energy then a cap of perhaps 20 energy would be
appropriate. Scan and shield would cost 5 meaning the spamming of any
one ability or medpacks/ammo could be risky in the event of a base
rush by aliens. A single number on the commander display would be
obvious to the player and not overly complicated.
2) Energy regeneration could then be based on a combination of the
player count and, to increase their importance, POWER NODES.
Power nodes are a very low priority for aliens to destroy and are
often left untouched after an extractor is destroyed. Increased
importance of territory control can only deepen strategy and one of my
favourite things to see in NS2 is the power cutting out in a room,
pitching it into complete darkness.
3) Maximum energy cap could then be increased with additional command stations.
One thing that has declined from earlier versions of NS2 is the
offensive usefulness of a second CC.
______________________________________________________________________________<!--QuoteEnd--></div><!--QuoteEEnd-->
I feel that the game is being over-simplified, especially as most problems people have with learning to command stems from a lack of learning resources and opportunities (guides, tutorials, practice mode (bots)).
Reducing the amount of resource pools do not simplify the game. The opposite is the case. If you have an own resource pool for every building (as it was before), there is nothing complex in it. You don't need to think about using this one ability in this situation or spare it in favor of your economy. You don't need to think if you heal and nano this lone two pushing marines or if you research weapons2. You just do it, when the need of this ability arises. The only difficult you have to handle is the UI. Finding the right building and knowing how much energy every building has left. In my opinion, a good RTS shouldn't be about how good you can battle the UI. It should be about strategic and tactic decision making.
The less resource-pools you got, the more complex the decisions you have to make become. Because every use of one ability costs you progress, instead of having no-brainers you have to use before the energy reaches its max.
I believe the support role of the commander should not have trade-offs affecting the overall team strategy, introducing the idea that giving a player a medpack isn't worth it. This leads to frustration on the field and encourages base minding tactics as commander. This is why I suggested a much lower energy cap so energy management is important and over use dangerous when considering base rushes.
The idea is not, that giving a medpack isn't worth it. It should be a decision, not a chore (you have to do, before your energy maxes out). With energy dropping, a medpack was nearly mandatory. So as soon as one commander doesn't drop medpacks, he is damaging the team. Leading to rage against noob-coms. Don't get me wrong here. The change hasn't made it easier for the com. But with more options and less mandatory tasks, he got more variety in play and it is more forgiving not following one single tactic.
Base minding tactics are also not bad. They add variety. They are only one option. One way how the commander can play. You can play aggressive with nano and meds, but will get your upgrades later or you can play defensive, but upgrading your marines so you come out stronger in the midgame where you can change your tactic to aggressive with medpack drops and nano-pushes.
You have to consider what is fun for the player though. These players you are ignoring are real people, the development of this game should be aiming to improve the average round overall for all involved to appeal to a wider audience. This comes back to polish. The whole noob comm thing is like I said stems from the game being new, complex with little help for the new player.
If you are giving a comm more options the game may be more "complex". If the added options are bad (boring), you have an unpolished game that is less likely to succeed.
As for delays, you're talking about the regular ones.
I'm talking about a SET one that is less known (even though I've only read about it shortly on these forums), it's something like "amount of lag-compensation-time" and apparently it is 100-150ms for games in general. But it was set to 250ms for NS2 cos the netcode/server/client -performance, Max was talking about. IIRC it was lowered to 150 and it improved gameplay. But if it was lowered, maybe there are other factors contributing to the current situation, among other those you mentioned.
The human reaction time is not really relevant when it comes to perceived smoothness and perceived "micro" predictability.
A friend once tried to argue that since the human eye has a delay of ~40ms (according to him) any fps that is faster than that goes to waste. I disagreed. He didn't listen. So we had to settle it by him playing Q3A at 25fps vs me at 60fps.
And now that I have a 120Hz screen, playing NS1 @58fps is a slideshow vs. 125 fps (the odd numbers are due to firing delay minimas at certain fps amounts in GldSrc)
@Angelusz Well said, thx. I also hope for that outcome!
As for resources, UWE, please just make it make sense.
I'm an aspiring comp myself and nothing against them, but many of them if not most care solely about gameplay mechanics and nothing else. Cos at the highest level that's the only thing that matters. Imagine how all the highest regarded games in (modern-game) history would be if 90% focus was just on mechanics...
This also goes for static defenses, they are basically not viable in NS2 atm, marine ones are completely sucky, and alien ones have A BUNCH of quirks to them. They're weak so you need to put all 3 together for some sort of strength, then you need to babysit them, since you only have 3 and only in one spot, so you end up being tied to a lifeform and to a map spot. Not dynamic at all for both sides! And why specifically THREE??? And FREE??? It's a resource based game, where resources dictate how many you can build and the total amount built at a given time.
As in southpark s02e14: IT DOES NOT MAKE SENSE!
But your argument was that things were being oversimplified?
Not what was "fun".
That's something far from quantifiable and is inherently opinionated and biased, varying greatly from person to person.
Whereas the amount of choices and strategies is numerically measured.