Mordhau
 marox — Project Lead
Project Lead
  • Likes received 5556
  • Date joined 25 Oct '15
  • Last seen 5h

Private Message

819 5556
  • 7 Jun
 marox — Project Lead

@rob_owner said:
i think only their idle stances are different, if more is different beyond that then it isnt casually noticeable

This is correct

819 5556
  • 6 Jun
 marox — Project Lead

We're working on a devblog to get them up to date, patch notes would probably be too confusing for Kickstarter backers to see, especially those without access.

819 5556
  • 6 Jun
 marox — Project Lead

We are working on a lot of stuff in parallel, some older WIP things are being implemented and finalized, and some things are broken as a result, which prevents us shipping builds in between. Among things that are being actively implemented and are rendering the update unshippable until finished at the moment are:

  • voice commands and replacement of all voices: this takes time, as things are still coming in and we're hoping to ship the update with 3 voices at least, but only have 1 implemented right now. (That'd be a lot of UGH UGH UGH AGH AGH AGH repetition if we shipped just 1 voice)

  • new maps (or expansions of existing maps, several are going on in parallel here)

  • horse movement rework to be less annoying and glitchy, some additional animation work to make them more lively creatures when changing directions

  • locational wounds

  • perks

  • bows and crossbows

  • crosshair work (angles, indicators for various things)

There's other things at work such as work on Frontline, but we absolutely need to finish those listed up above before we're able to ship this update.

819 5556
  • 5 Jun
 marox — Project Lead

@ÐMontyleGueux said:

@marox said:
It's due to quivers not being able to be picked up, but you start with one whenever you select a bow/shortbow/xbow in the armory. So if you pick something up without the corresponding quiver your character tosses the remaining ammo away, leaving you with 1-2 arrows/bolts. It was a necessary limitation, and also a way for us to identify people who chose a dedicated ranged weapon in the armory.

You said not too long ago that there wasn't going to be quivers like in slashers because there were confusing. So you have changed your mind then ? Does that mean equipping a bow for example will automatically fill another equipment slot with a quiver or is it "invisible" to the player ?
Edit : I want to be clear, I'm not criticizing here, I'm expecting things to change. I'm merely trying to get up to date information.

They're not like the slasher quivers, in the sense that they don't take up slots, and don't require you to choose them in customization, they come automatically with the ranged weapon.

819 5556
  • 5 Jun
 marox — Project Lead

@Cswic said:
Credit goes to Von. Could interactable weapon racks be put on the "nice to have, maybe sometime in the future" list of things?

There are already weapon racks that can be seen on various maps that have weapons spawned on them. Could they be made to allow us to place weapons on them as well?

This would go nicely with a certain other mechanic that was "previewed" recently.

I don't know if they will be on maps, they would probably be broken for skirmish if they persisted between rounds. They might make it as something we ship for modders though since I did start on this some time ago for fun to see if it could be easily added. I do this from time to time to make sure it's possible to extend the game through blueprint modding sufficiently and to add required functionality if not, as well as to supply modders later on with some examples.

@Mittsies said:

@Vondulun said:
That won’t be possible with bows, Jax said if you pick up a bow you’ll only get one arrow.

That's a clever way of limiting it, I suppose.

It's due to quivers not being able to be picked up, but you start with one whenever you select a bow/shortbow/xbow in the armory. So if you pick something up without the corresponding quiver your character tosses the remaining ammo away, leaving you with 1-2 arrows/bolts. It was a necessary limitation, and also a way for us to identify people who chose a dedicated ranged weapon in the armory.

819 5556
  • 27 May
 marox — Project Lead

@Mittsies said:

@Humble Staff said:
Also devs have said that they want to implement localised wounds some time in the future. This may allow for what you are suggesting, for dents in armor, etc.

This would be great as long as the consistent full-body wound visuals remain in some form.

We are keeping it for readability, although it might be tweaked.

A lot of work was done on the localized wounds already prior to alpha shipping, it was just disabled because it wasn't complete. This is often the case with features that aren't critical/ready for testing, we just turn them off for the time being because they're not ready yet. Work has very recently resumed on the localized wounds and they're coming along nicely. It supports a variety of wound types, such as holes, accurately angled slices, general messy blotches, etc. We're also prototyping possibilities of using the same system for deformations, one by modifying the normal map (somewhat subtle, think armor dents), and one based on changing the vertex geometry (which leads to very noticeable deformations). The latter in particular it is highly experimental and may not work out at all, so don't get too excited.

819 5556
  • 19 May
 marox — Project Lead

@yourcrippledson said:

@marox said:
This was a long time coming, but we are changing to vertical FOV next build. Currently the horizontal FOV is locked to be 'constant' across aspect ratios, which is fairly pointless and creates a weird situation with different aspect ratios having different looking 1p models, with wide screens being zoomed in and the opposite being zoomed out. It also lets us standardize the viewmodel correction and fix clipping. For now this will shift the advantage to widescreens in the sense that they will have more peripheral vision, which we can account for later if it becomes an advantage issue.

Thanks for trying this Marox.

These screenshots display the issue he is attempting to improve upon. Not just affecting 21:9.

At 130 FoV and -15 Camera Distance
4:3 1280x1024 (this is closer to what I want to see of my arms and below my nose. Only without a stretched screen, or black bars, and in 16:9)

3rd 20180517093155_1.jpg
20180517093146_1.jpg
20180517093147_1.jpg
20180517093147_2.jpg
20180517093149_1.jpg
20180517093151_1.jpg

16:10 1600x1024

3rd 0180517093349_1.jpg
20180517093343_1.jpg
20180517093343_2.jpg
20180517093344_1.jpg
20180517093346_1.jpg

16:9 1920x1080 (native)

3rd 20180517093430_1.jpg
20180517093424_1.jpg
20180517093425_1.jpg
20180517093426_1.jpg
20180517093427_1.jpg

I used the same settings for all screens. Only altering Resolution.

you don't seem to lose any x-Axis visibility at all, and gain a substantial amount of Y-axis. Causing 4:3 to basically be the meta view model.

Yeah, I know. The biggest issue for us isn't so much that widescreens were being disadvantaged (which they were) but rather that the viewmodel stretching was impossible to account for, so clipping was unavoidable and we couldn't control it. This way we can decide on a maximum amount that still doesn't clip all over the place and that we can adjust for, and allow all aspect ratios to reach that FOV.

819 5556
  • 19 May
 marox — Project Lead

This was a long time coming, but we are changing to vertical FOV next build. Currently the horizontal FOV is locked to be 'constant' across aspect ratios, which is fairly pointless and creates a weird situation with different aspect ratios having different looking 1p models, with wide screens being zoomed in and the opposite being zoomed out. It also lets us standardize the viewmodel correction and fix clipping. For now this will shift the advantage to widescreens in the sense that they will have more peripheral vision, which we can account for later if it becomes an advantage issue.

819 5556
  • 7 May
 marox — Project Lead

The main issue with holding the button down and having the 240 react after the fact (or after releasing it) is input lag, which makes the input scheme quite sub-optimal. By input lag I mean the delay between pressing the button and releasing it.

819 5556
  • 4 May
 marox — Project Lead

@Frise said:
Does it not affect the timing of the grunt sound? Because this happened to me, Meph winded up a stab, the grunt sound played so I went for the chamber, but he morphed it into an overhead after the grunt, which confused the hell out of me.

At that time I had this setting at 1, with a ping of 56ms; I assumed my ping jumped at that moment leading to the grunt being played prematurely.

You are right, it does. It's because the grunt is tied to a point before release, so it does affect it. It really shouldn't, though, so I'll change the grunt to compensate for the extrapolation accordingly.

819 5556
  • 4 May
 marox — Project Lead

In this patch we changed the look interpolation equations entirely, so this is probably a result of that. The interpolation time might be too low at the moment to smooth out the stuff the server sends, but it would be good to have a high framerate video instead of a gif, and the fps this was played at.

819 5556
  • 3 May
 marox — Project Lead

Also windup. This ONLY affects the attack animation, nothing else.

819 5556
  • 2 May
 marox — Project Lead

@ÐMontyleGueux said:

@Maci said:
It makes attacks seem further into release than they actually are. If you set that to exactly your ping, you should be able to parry and chamber attacks as if you were on local play. However, it does so by skipping other parts of the animation which looks a bit janky.

From what I understand, it will try to use your current ping to extrapolate animations if your ping is below m.MaxPingExtrapolation. If your ping is higher than that, it will use the value m.MaxPingExtrapolation. For instance if m.MaxPingExtrapolation == 0.025 and ping == 20ms, you will see attacks 20 ms in the future, and setting m.MaxPingExtrapolation to 0.090 won't do anything. But, if m.MaxPingExtrapolation == 0.025 and ping == 50ms, you will see attacks 25 ms in the future, and you might want to set m.MaxPingExtrapolation to a higher value.

Not sure if that's entirely true, but that's what I understand from the name m.MAXPingExtrapolation...

Yup, this is how it works

819 5556
  • 1 May
 marox — Project Lead

@Monsteri said:
99% of "free" devs/contributors promise a lot, are very excited, then fuck around for a week and are never heard from again. Not worth the organizational effort and likely a big detriment.

This is the biggest problem. We managed to get around this by using people from the Chivalry community, because they're generally genuinely interested in the game. In addition, we've also had the luxury of allowing a lot of people learn from a very rudimentary skillset over the years, or get acquainted with the technology. We could do that again, but it takes time for people to pick things up and it takes time to teach them and deal with them in the process. Time we don't have right now. The other option is to hire people who are already fairly well acquainted with the tech, but those don't work for free and/or don't have enough time, so we can't afford them or the overhead of dealing with them right now.

819 5556
  • 29 Apr
 marox — Project Lead

@Bodkin said:
The parry stamina thing needs to also apply for shields.

Eating ~15 stamina on each block is cancer with multiple enemies

It does.

819 5556
  • 29 Apr
 marox — Project Lead

@Jax said:
Binds don't come in at the 'steepest' angle so that could be your issue.

They do have the steepest angle actually

819 5556
  • 28 Apr
 marox — Project Lead

I liked the AOC catapults, but they're a different beast. The catapult we have right now is mobile, and we won't turn it into an AOC type stationary catapult. But I'd love to have a stationary catapult that could be used in this manner, some day.

819 5556
  • 23 Apr
 marox — Project Lead

@Zexis said:

as the hitboxes there in 1P are different (3p is OK though)

How are the 1P hitboxes different on local play?

They use the 1p animations in local play, hitboxes otherwise are based on 3p animations (on dedicated servers).

819 5556
  • 23 Apr
 marox — Project Lead

@Pred said:

@marox said:
The only real difference with the messer is the model/tracers and obviously attack timings, the tracers and model seem fine and nothing particularly weird is done there, only other thing would be the timings.

Hmmm, what about the guard, is there a tracer there? The knife-like one doesn't have guard so maybe it's misleading somehow if it's using the same tracers as the one with guard?

No tracer on the guard, it's pretty much identical to the other tracer setups for swords

819 5556
  • 23 Apr
 marox — Project Lead

@Pred said:

  1. I don't know if it's direclty related to FOV, but it feels like your hitboxes extend beyond your body, especially in front of your knuckles and it's not like a random desync once every 200 plays but rather a staple of the game. Visual cue you are given is that you have plenty of room and time to react (the "Not Hit" part) you actually don't have any. Don't know if pushing back hitbox on your hands and/or making an actual chunk of the weapon hit you before doing damage (rather than first pixel of a stab) are possible options or not, but it could at least help mitigate majority of the cases, when the attack is incoming from in front of you.

Adjusting the hitboxes in this case could lead to the other guy missing but it looking to him like the hit should've landed, although I agree the hits do feel disconnected, especially the mace hit. The only sensible solution other than hitbox adjustments (which might not help enough) would be to move the 1P model (and with it the camera) further in, but this would reduce some of the visual distance which might make it harder to see things to some degree.

  1. There is something wrong with Messer because what is described in point 1 happens to involve Messer much more often and more pronounced than other weapons, especially when they miss an attack and combo.

The only real difference with the messer is the model/tracers and obviously attack timings, the tracers and model seem fine and nothing particularly weird is done there, only other thing would be the timings.