It would be nice if a costume slot included body data so that we could be different shapes, sizes, colors, and genders. For instance Bruce is more than a little smaller and pinker than the hulk.

Does anyone recall if this particular feature has been discussed?

"THE TITANS ARE COMING! THE TITANS ARE COMING!"

Hero_Zero wrote:

It would be nice if a costume slot included body data so that we could be different shapes, sizes, colors, and genders. For instance Bruce is more than a little smaller and pinker than the hulk.

Does anyone recall if this particular feature has been discussed?

Yes the CoT Devs have already told us they are breaking up the "costume data" into two types of save files. One will be the "costume files" that will only contain the costume items used for a costume slot. The other will be a "body slider" type file that will contain all the "body data" for a costume slot.

The advantage of having these broken up into two types of files is that you'll be able to save/load costumes separately from body shapes so it'll actually be much easier to create/organize things like having "Bruce Banner" costume slots and "Hulk" costume slots.

Again like I've said these things have actually been covered like 6 or 8 times over the last few years already. ;)

stonejaguar wrote:

Lothic wrote:

Yes the CoT Devs have already told us they are breaking up the "costume data" into two types of save files. One will be the "costume files" that will only contain the costume items used for a costume slot. The other will be a "body slider" type file that will contain all the "body data" for a costume slot.

This answers my question. Assuming you can make a macro with costume change tied to a power in CoT, I will be able to make my stone/stone tanker type the way I had imagined it. This is why I like the CoT community. Thanks!

You're welcome. :)

Just be aware that CoH imposed a 30 second timeout between costume changes and it's possible CoT will do the same. What this means is that you could press one button to switch from costume slot A to costume slot B and it would switch instantly. But once you did that you'd have to wait 30 seconds before you could switch again to costume slot C.

The reason I'm mentioning this is that I did a large amount of experimenting in CoH with tying "powers to costume changes" like you're talking about doing. While it sometimes resulted in fun/useful results you have to understand that the power activations could easily get "out of sync" with costume changes and the keybinds could end up being more cumbersome than they were worth.

It'll be interesting to see if CoT will make any of that easier or harder to work with.

That’s weaksauce, gimme shadow clones. :D

@stonejaguar, I'm not sure you really got the answer you were looking for.

From your OP, it looks like you would also be happy with a costume change that occurred upon "entering combat." As with most MMO, entering combat can occur upon a number of different stimuli: [list][*]When you hit the [i]change stance[/i] hot-key (assuming there will be one)[*]When attacking a target[*]when being attacked by a target[color=red]*[/color][*]When activating a non-targetable (such as ground targeted AoE) power[*]whenever the game tells you that you are in combat, such as at the start of a scripted event[/list]

I think it would be a nice feature if a costume change emote were to proc whenever a character changed statues into and out of combat.

We've already discussed some in-combat effects, such as auras and blinkies, but I'm not sure if the subject of an actual combat costume change has been mentioned before your post.

[color=red]*[/color]There have been discussions about being attacked while in travel movement mode and whether or not it will take a character out of travel movement or not, and by extension, whether that would alse enter a character into combat.

Huckleberry wrote:

@stonejaguar, I'm not sure you really got the answer you were looking for.

From your OP, it looks like you would also be happy with a costume change that occurred upon "entering combat." As with most MMO, entering combat can occur upon a number of different stimuli: [list][*]When you hit the [i]change stance[/i] hot-key (assuming there will be one)[*]When attacking a target[*]when being attacked by a target[color=red]*[/color][*]When activating a non-targetable (such as ground targeted AoE) power[*]whenever the game tells you that you are in combat, such as at the start of a scripted event[/list]

I think it would be a nice feature if a costume change emote were to proc whenever a character changed statues into and out of combat.

We've already discussed some in-combat effects, such as auras and blinkies, but I'm not sure if the subject of an actual combat costume change has been mentioned before your post.

[color=red]*[/color]There have been discussions about being attacked while in travel movement mode and whether or not it will take a character out of travel movement or not, and by extension, whether that would alse enter a character into combat.

This is an overall valid suggestion: Having costume changes be triggered by various different external events.

But I would still repeat my word of caution about the possible 30 second costume change timer. Again we don't know if CoT will follow CoH in imposing its own costume change timeout but if it does then pretty much anything that might secondarily trigger a costume change might easily get unintentionally stuck or "out of sync".

Obviously once we know how CoT intends to handle this timeout issue then it'll become much more clear how being able to trigger "secondary costume changes" might work.

blacke4dawn wrote:

Lothic wrote:

I would still repeat my word of caution about the possible 30 second costume change timer. Again we don't know if CoT will follow CoH in imposing its own costume change timeout but if it does then pretty much anything that might secondarily trigger a costume change might easily get unintentionally stuck or "out of sync".

Obviously once we know how CoT intends to handle this timeout issue then it'll become much more clear how being able to trigger "secondary costume changes" might work.

Iirc MWM has said that there will be a costume change CD due to that the costume will have to be "compiled/assembled" client side, and that interrupting that would be sub-optimal. I do hope that they have a good caching system though so that when someone changes back to a costume you've already seen you won't have to "re-compile/re-assemble" it, at least for some time. That caching could also help if two or more characters have the exact same costume.

Yes I understand the needs of a "costume change" in terms of the impact it has on the game's resources. The biggest impact isn't so much your local client side switch as it is on the message processing required to send your costume information to all the surrounding players to update how their clients render your character's costume on their screens. If players were allowed to change their costumes multiple times in a row without a timeout they could easily "overload" the system and cause other players around them a ton of lag and grief (in the negative MMO sense of that word).

Regardless of all the impacts associated with a costume change it's clear there WILL need to be some kind of timeout to prevent griefing in CoT. My reasonable question now in 2018 is whether or not the game would really require a full 30 seconds for such a timeout.

I always suspected (even back when playing CoH) that pegging that timeout at a full 30 seconds was just a hyper-conservative arbitrary value the Devs settled on "just to be sure" everybody would be able to play without any impacts. I'm wondering now that most peoples' computers and Internet connections are typically much more capable than they were back in 2004 whether setting the timeout for this to say 10 or 15 seconds would be completely adequate for the purpose it's serving. Again I understand the timeout needs to be "long enough" to be useful but at the same time it really also ought to be as SHORT as reasonably possible.

I'm hoping the Devs of CoT will consider this point during their beta testing. If they discover that setting this timer to say 10 seconds is simply not long enough via experimental testing they could always increase the value before launch.

TheInternetJanitor wrote:

Is there a serious reason to need to switch costumes more than once every few seconds outside of trying to link costume changes to power use? Because linking it to power use only works if you can change costumes as fast as you can use the powers you want associated with it. Assuming the change can't be that fast for the sake of argument, what else is left that makes it necessary to minimize the cooldown?

Well again I'll be the first to agree that a costume change cooldown of "some" duration is likely always going to be needed just to prevent griefers. Even if the game could easily process the message traffic of a player trying to switch their costumes as quickly as they could press the buttons to do it it would still be annoying as heck to have a person standing in front of you trying to induce an epileptic seizure via multiple hyper-quick changes. But I would argue that having that timeout be a full 30 seconds is [b]also annoyingly too long[/b] for those who would have much more reasonable/legitimate uses to switch multiple times in a short period of time.

Apart from the desire to effectively link "powers to costume changes" without the ever-present side-effects of those things too easily getting "out of sync" there's the hassles of "arbitrarily waiting" for the timeout to expire for things like costume contests, various RP situations and the simple annoyance of switching to the wrong outfit and having to wait 30 seconds just to fix that.

So the compromise solution being proposed here is that having the costume change timeout timer be set to a value of say 10 seconds would still serve the purpose of effectively preventing costume griefing while also greatly reducing the overall "annoyance factor" of those people who might have a "legitimate" reason for making several quick changes.

Another compromise alternative would be to allow people to switch their costumes as many times as they'd want within say a very tiny span of time (say 5 seconds) BEFORE the 30 second timeout kicks in. This would allow legitimate costume switchers the chance to quickly fix a switch mistake BEFORE having to endure the timeout while still preventing another player whose only intent is griefing from being able to annoy anyone else for longer than 5 seconds.

Either way the goal of reducing CoH's 30 second timer in CoT is to prevent the griefing while ALSO minimizing the hassle imposed on players not intent on griefing.

Project_Hero wrote:

TheInternetJanitor wrote:

Is there a serious reason to need to switch costumes more than once every few seconds outside of trying to link costume changes to power use?

Long multi-costume anime style transformation sequences, of course!

For when you want to show every bit of your transformation!

The key difference is that the person who'd want to change through several outfits as much as say once every 5 or 10 seconds (i.e. anime style transformation sequences) is NOT trying to grief anyone with that. It's the idiot who's frantically pressing their costume change buttons 6 or 8 times a second to try to piss people off who needs to be stopped.

Unfortunately the arbitrary 30 second timer in CoH managed to stop the griefing by overwhelmingly screwing over the legitimate costume switcher. I'm simply suggesting that the "griefer" in this scenario can still be stopped without being so draconian against the effectively "innocent" player.

TheInternetJanitor wrote:

So I'm not sure how it'll be handled in CoT but one way to do what you describe and still avoid many problems with quickly swapping costume is to have client side throttling on this.

I'd be all for the game being smart enough to do the kind of "costume data throttling" you're talking about. And I agree that even if a given player's client has not been "updated" with another player's very latest costume data that would not really be that critical because as you say it's all basically "cosmetic" anyway.

But the bottleneck that causes the lagging is not anyone's local client doing the "switching" of the costume items that appear on the screen. That part really doesn't take that much "extra processing" at all because all the art already exists pre-generated on the client. What causes the lags is the extra network traffic not being able to keep up with everyone moving around and popping off powers on your screen. Once the network in your local area starts getting bogged down with [b]ANY[/b] kind of extra costume messaging (even "throttled" as you'd suggest) then things like "rubberbanding" start to happen. This is how costume change griefing works and why it's such a bad thing.

Any solution that prevents costume change griefing must eliminate the extra network traffic in order to be effective.

TheInternetJanitor wrote:

The rest of the time it would work as normal, just with a possibility for a faster costume cooldown for everyone since that client side safety net exists.

Another idea to "mitigate" the extra costume data traffic would be to control this problem the same way they controlled people spamming too much on the chat channels in CoH. Basically the way that worked was if you started to spam a huge number of chats as fast as possible the game would eventually (after a few seconds) detect the spamming and automatically cut you off from being able to send any text chats for like a full minute. The net effect was that people who never spammed were never punished.

So to apply that idea to costume changes maybe the game could be made "smart enough" to detect that if you tried to spam a bunch of costume changes too quickly (let's say like more than 5 quick changes during 5 continuous seconds) the game could just simply shut off your ability to send costume change commands for a full minute as "punishment". This would effectively eliminate anyone's ability to grief people via "spamming" costume changes while still allowing non-griefers to successfully fire off randomly quick changes when desired. The game could even have a second layer of detection that could keep "moderately" quick changes under control. For example maybe if you tried to do more than say 8 changes per minute (averaging once every 7.5 seconds) the game could again put you in a timeout for a full minute. The net effect here is that the average player would only be able to accomplish costume changes at best once roughly every 9-10 seconds without punishment which really should be fast enough for everyone playing this game. Keep in mind this scheme would actually allow for occasional bursts of several quick changes within the span of a few seconds if desired.

Again the whole point of this is to prevent costume change abuse/griefing while still allowing a reasonable amount of changing that could happen more than the "once per 30 second" eternity that CoH imposed.

TheInternetJanitor wrote:

Having lots of powers used in an area and having costumes changed with big flashy animations very quickly is going to hinder local system resources waaaay before a network traffic bottleneck will matter. Not only that, but connection speeds have increased drastically in recent years. While CPUs and GPUs have also been increasing, games are expected to use that power to look good. The amount of network traffic to tell a local client to trigger an animation, though? That doesn't need to change a great deal. The muscle work is done locally. The network traffic just tells the local client to flex that muscle.

Basically this means that a local throttle would be an effective strategy to get a good result with little downside for this situation.

Just to be clear I'm only "reporting" what the original CoH Devs always claimed was the answer to the question why there was a costume change timeout timer in the first place.

The answer to that, according to them, was that without a clear safeguard to prevent griefing you could potentially (as a worst case scenario) have several players stand together in a crowded area (i.e. an open zone event) and start spamming costume changes as many times as they could push the buttons to do it. The net effect of the network having to try to broadcast all those multiple "costume change messages" to everyone else around them could collectively generate so much extra lag for everyone that it could grind everyone's actions to a halt. Again I'm simply relating to you pretty much word for word what was told to us many years ago in CoH to explain why the costume timeout timer was arbitrarily set to the relatively huge value of 30 seconds. There was never any mention about how much processing power individual clients needed to have/use to support the actual costume changes themselves - the focus was always on network loading (or in this case [i]overloading[/i]). Take this information with whatever sized grain of salt you'd like... *shrugs*

So what does that mean for CoT? As you say (and as I implied several posts ago) a lot has changed since 2004 in terms of hardware capabilities of both client and server-side systems. It could very well be possible that some combination of "throttling" and/or "intelligent dynamic thresholds" could be the answer for this instead of a simplistic arbitrary 30 second timeout. Unfortunately the Devs of CoT have yet to fully comment on this particular design issue so we can only speculate at this point what their plans for CoT are concerning this.

All I can tell you is that having the 30 second timeout in CoH was definitely a HUUGGGEE annoyance for those of us who wanted to be able to change costumes relatively quickly without any desire or intention to cause any griefing to anyone else. Anything the Devs of CoT can do to reduce/eliminate that timer while also preventing any associated griefing would likely be much appreciated regardless of how that's ultimately implemented.

Tannim222 wrote:

You’re mostly correct. Animations work a little differently. They aren’t only loaded client-side (you can’t for example remove or change an animation file so you never see it on another player), are keyed by power activations, which also hooks into the combat engine. When one player activates a power, everyone sees the animation and that causes more network traffic.

The problem is when systems start getting lagged because they struggle to keep up, everything compounds. The network traffic tracks everything going on, every step a character makes. collider detections, combat data, etc, which can ultimately result in getting lagged out. But in between “runs smoothly” and “lagged out” there is “too laggy to play”. Where frame rate drops and you’re in slide-show land.

Frequent costume swaps can cause this to happen when you consider the larger implications - as I noted earlier, large groups of people in one area changing costumes every several seconds will not be good for low end system players.

I wonder if you could build location into this logic for the '30 second duration' that CoH had. Like, if in an instance, X seconds since it will affect a max number of players equal to how many are on the team. Or, if X number of other PC's within X ft/yds, X seconds (this would be for areas in the open-world), etc. If this is even implementable, it could reduce the strain in those key locations. But I guess that would depend on if that distance metric would detect other teams/players in the same instance but on a different team.

So a player who is off playing on the edge of the map with no other PC's within 1,000 ft/yds could potentially have no CD on this ability since there is no one else around to affect. Or those who are in instances solo'ing.

One thing we looked into which is possible, but we can’t promise at this time is creating a morph system tied to certain powers or even a set based on Momentum that allows for certain transformations. Like a hand turning into a giant clawed hand, turning into a werewolf at the full stage etc.

The possibility exists, and it may be possible to have a set of morphs work like a prop. But we have certain hurdles to deal with and while the base testing was possible, it requires more resources than we can apply for launch so we have put it off until later.

TheInternetJanitor wrote:

Thanks Tannim! The point I was making though is that much of what you just described (worrying about particle effects, how many layers of costume to draw at once, etc) are concerns for local system resources. You said yourself you were testing with a low end laptop that had trouble loading shaders - these are local system concerns. CPU, RAM, GPU. An old laptop and a supercomputer can be using the same network interface card and internet connection.

Am I mistaken in assuming the game is only going to send a relatively simple message from the server to the client no matter how fancy the costume or animation is? All you need to send is "use animation number 75 and load these textures and parts".

The resulting animations and effects might bring a low end computer to tears due to impact on GPU, CPU, etc, but the network traffic impact is identical and even a low end connection and network interface can handle it.

If the issue is truly network traffic then this would be a problem when *any* significant amount of network traffic was needed, and have zero to do with costume changes specifically.

For example, if network traffic alone was truly the bottleneck, you would not be able to have a significant amount of players in the same area doing *anything* because that would generate similar amounts of network traffic. Talking, fighting enemies, even just walking around.

Again it's not an issue of the "costume change data" messages being small or big size-wise or not. The problem is having griefer(s) blast out a bunch of those messages to overload everyone else around them.

If you'll recall back in CoH there was always a lag problem whenever a bunch of players got together [b]regardless[/b] of costume changes. With a bunch of people near each other you get a huge increase in the amount of messages being sent related to player positions, power activations, combat result messages and so on. The game always had big problems with handling crowds - this is why the Devs had so much trouble getting base raiding to work and why they actually [b]reduced[/b] the number of people that could participate in Hami raids as the game evolved.

So it stands to reason with all that messaging chaos going on if you were to add a dozens/hundreds of extra costume data messages to that mix (via griefers pressing their costume change buttons as quickly as possible with no timeouts) that it could easily lag everyone out.

Bottomline network lag was a very significant problem/limitation in CoH and costume change griefing was a big enough threat to that that the Devs imposed a near-ridiculously huge costume change timeout timer (30 seconds) to deal with the issue.

Lothic wrote:

Yes I think there are a number of ways the Devs could make this thing "smart enough" to react more intelligently than just applying a relatively simplistic "X second timeout timer".

Again the main goal of course is to prevent costume change griefing, but that should not have to come by punishing non-griefers. Between the "throttling" idea to the "text chat" style implementation to your "player proximity" idea there has to be some way to make this thing much more "friendly" than the way CoH handled it.

I recall it came up that the Unreal Engine has built-in proximity features that determine the rendering detail based upon the performance of our clients and I also expect that we will be able to adjust it in some game controls settings window. So it could very well be that the processing burden associated costume changes will just be in the noise and/or included in the calculations of those. But I'm now just surmising rather than speaking from actual knowledge.

Tannim222 wrote:

One thing we looked into which is possible, but we can’t promise at this time is creating a morph system tied to certain powers or even a set based on Momentum that allows for certain transformations. Like a hand turning into a giant clawed hand, turning into a werewolf at the full stage etc.

The possibility exists, and it may be possible to have a set of morphs work like a prop. But we have certain hurdles to deal with and while the base testing was possible, it requires more resources than we can apply for launch so we have put it off until later.

Oooh, me likey!