Movement property tweaks feedback

[TotesAmaze]

I wanted to have a discussion with you guys about Dodge Landing Speed Factor (DLSF), as i've been thinking more about it lately as I've played with movement. For those that don't know this is the slowdown you get on landing a dodge. Try dodging sideways while facing a wall and its most noticeable. At first I saw it as a necessary balance for dodging, "you've had your speed now you trade it for some slowdown!" So like repaying the debt of the dodge speed.

However, I've found I quite like the smoother feeling to the movement when this value is less effective. Of course this would make dodging even more the movement of choice at any time, seeing as you have less of a downside for it, so I've been thinking of other ways to balance it. Epic introduced auto sprint as a new and more logical option for just getting somewhere faster, but if the sprint was more of a build of momentum from the moment you land a dodge or jump, it could allow the DLSF to be smoother while still overall lowering the players speed as they keep canceling out the auto sprint build up if they dodge constantly.

Anyhow words are useless so I'll play around with settings in a blueprint, but I've noticed this hasn't had much attention yet.

[Gilanguar]

I wanted to have a discussion with you guys about Dodge Landing Speed Factor (DLSF), as i've been thinking more about it lately as I've played with movement. For those that don't know this is the slowdown you get on landing a dodge. Try dodging sideways while facing a wall and its most noticeable. At first I saw it as a necessary balance for dodging, "you've had your speed now you trade it for some slowdown!" So like repaying the debt of the dodge speed.

However, I've found I quite like the smoother feeling to the movement when this value is less effective. Of course this would make dodging even more the movement of choice at any time, seeing as you have less of a downside for it, so I've been thinking of other ways to balance it. Epic introduced auto sprint as a new and more logical option for just getting somewhere faster, but if the sprint was more of a build of momentum from the moment you land a dodge or jump, it could allow the DLSF to be smoother while still overall lowering the players speed as they keep canceling out the auto sprint build up if they dodge constantly.

Anyhow words are useless so I'll play around with settings in a blueprint, but I've noticed this hasn't had much attention yet.

I prefer it off but it does allow minimal bunnyhopping. A side effect of the DLSF is when you land all momentum is killed. If you time a jump perfectly after landing a dodge with no DLSF you will bounce forwards even if you don't give any movement input. I did a video agggees ago demonstrating it. Personally I like it although it isn't everyones cup of tea.

[TheWhiteDragon]

I wanted to have a discussion with you guys about Dodge Landing Speed Factor (DLSF), as i've been thinking more about it lately as I've played with movement. For those that don't know this is the slowdown you get on landing a dodge. Try dodging sideways while facing a wall and its most noticeable. At first I saw it as a necessary balance for dodging, "you've had your speed now you trade it for some slowdown!" So like repaying the debt of the dodge speed.

However, I've found I quite like the smoother feeling to the movement when this value is less effective. Of course this would make dodging even more the movement of choice at any time, seeing as you have less of a downside for it, so I've been thinking of other ways to balance it. Epic introduced auto sprint as a new and more logical option for just getting somewhere faster, but if the sprint was more of a build of momentum from the moment you land a dodge or jump, it could allow the DLSF to be smoother while still overall lowering the players speed as they keep canceling out the auto sprint build up if they dodge constantly.

Anyhow words are useless so I'll play around with settings in a blueprint, but I've noticed this hasn't had much attention yet.

To be honest I think the reason this hasn't been discussed is because we all hate sprint. In fact, in one poll it was 52-1 against sprint. So I think everyone just assumed that it will go away eventually.

[Sir_Brizz]

I'm more concerned with the cooldown. I lowered that to 0.25 and it is quite good that way. I don't want momentum to be stored, dodging should come with a penalty. It looks like Steve moved this to 0.3 from 0.35, so at least they are seeing the same thing Being able to dodge sooner makes the movement feel a lot better.

That said, I think the current DSLF is way too low. I would rather see it at 0.3 or 0.4 than 0.1. That just slows you down way too much.

And I still hate sprint. Not going to lie. Even turning off the delay and increasing the acceleration feels like **** and sprint should be taken out.

[TotesAmaze]

That said, I think the current DSLF is way too low. I would rather see it at 0.3 or 0.4 than 0.1. That just slows you down way too much.

Yeah thats what I'm talking about. Its default value is much too muddy I think.

And I still hate sprint. Not going to lie. Even turning off the delay and increasing the acceleration feels like **** and sprint should be taken out.

Yeah, hate is such a strong word...I reserve that for when I'm driving . I don't hate sprint but to be convinced I need to see it solving the problem it is supposed to. Having a good fast movement speed like UT99 will work fine, we know that. I want to see some versions of it that feel like a build up of extra speed straight after a dodge or jump. The current defaults feel weird to me...after 2 seconds (why 2 seconds exactly?) you begin slowly building speed to a point where you're going too fast through the level. There is too much time where you are accelerating...I think it feels much better as a 1-2 second duration build up of speed which kicks in instantly.

Yup I thought around 0.25 for cooldown was good too. I think a limit of 2-3 walldodges will work better than having it infinite.

[d33p_Sky]

Are you kidding? Ghosts is the most played COD, and the next one will be the most played again. If anything, that stuff is in response to Titanfall, and COD and Titanfall are essentially the same game with slightly different mechanics (which makes sense since the basics of both games were developed by the exact same people).

I'm interested to see how this stuff actually works in multiplayer. They talk about a bunch of stuff in that video that honestly would suck in multiplayer.

Out of curiousity two questions. Do you play alot of COD? Are you a fan of that series opposed to the Battlefield Series?

[Sir_Brizz]

Yeah thats what I'm talking about. Its default value is much too muddy I think.

Yes, I agree.

Yeah, hate is such a strong word...I reserve that for when I'm driving . I don't hate sprint but to be convinced I need to see it solving the problem it is supposed to. Having a good fast movement speed like UT99 will work fine, we know that. I want to see some versions of it that feel like a build up of extra speed straight after a dodge or jump. The current defaults feel weird to me...after 2 seconds (why 2 seconds exactly?) you begin slowly building speed to a point where you're going too fast through the level. There is too much time where you are accelerating...I think it feels much better as a 1-2 second duration build up of speed which kicks in instantly.

Well, maybe hate is too strong, but I do really dislike it. I've tried playing around with the values and I can't find a place where it feels at all good to me. The current settings are too "magical" (as I've said before, you can't even really tell it is happening), and anything that makes you speed up faster just makes you feel like you should be running that speed all the time (IMO). I kind of agree with TWD here that movement should just meet in the middle at 1100. 900 feels BAD.

Out of curiousity two questions. Do you play alot of COD? Are you a fan of that series opposed to the Battlefield Series?

I don't know if I play COD a lot, since I haven't played it at all in over 6 months. But I do prefer COD over Battlefield and I think COD is perfectly fine fun.

[garuDa]

It seems like I misunderstood the Dodge landing speed factor
The explanation says Horizontal speed reduction on dodge landing so I thought the bigger the number the more speed reduction.
In fact i want to keep the same momentum for smoother gameplay so it has to be 1.0 instead of 0.0

[Gilanguar]

I think its reducing your current speed down to a percentage so 0.1 would be 10% of your dodge speed speed is kept on landing. I might be way off though.

[Sir_Brizz]

I think its reducing your current speed down to a percentage so 0.1 would be 10% of your dodge speed speed is kept on landing. I might be way off though.

It's a factor, so yes. I'm not sure if it's based on dodge speed or run speed, though.

[Gilanguar]

Pretty sure it's total speed. Since your air speed when dodging is a function of dodge speed + run speed.

[Sir_Brizz]

Pretty sure it's total speed. Since your air speed when dodging is a function of dodge speed + run speed.

You're probably right. It would be trivial to make a Blueprint that output that. Maybe I'll do it later just to see.

[TotesAmaze]

This has taken me a few evenings and most of today, trawling through different combinations of movement settings on all the test maps until I was finally happy with my results, so I hope somebody looks at it. (Such is my UT fandom) . Please DL the blueprint and give it a try.

BLUEPRINT: https://www.dropbox.com/s/bcdron75r4...Character2.zip

I can't believe how hard it is to settle with the values but I'll explain my reasoning and hopefully the video should show the movement is fast, engaging and scales very well with the levels. Dodge-jump and double jump are not enabled but can be for testing, I gave them a really good shot but just found it detracted from the gameplay and helped to invalidate the level design. Lets leave the dodge-jump argument out for now. I like it in 2004 and I voted for it in the poll on these forums, but from my testing I've changed my mind. However I think the boost-dodge and wall-dodging can bring that feeling of freedom of movement and air-time to the game that dodge-jump does while not being able to do it anywhere, so all hope is not lost d-jump fans!

Main changes and reasons why:

Jump Height - Increased to 800 - (or 815 for more comfortable clearance). Without extra height and control of the double jump (plus higher gravity), i believe this value needs to be substantial enough to jump the small crates and ledges in the levels. If I can't jump those, why have a jump? I can now jump onto those small crates with weapons or armour on in deck (which I've always wondered why you can't) as well as some other smaller ledges you'll see in the video. However the jump isn't too high that I bump my head or overshoot jumping targets or remain in the air so long I'm overly vulnerable.

Accelleration - 7250, Decelleration 2000 - These settings brought me the closest to a good strafe responsiveness without feeling overly twitchy or heavy.

Gravity 2.2 - Worked best with the levels scale, so I wasn't smashing my head when dodging down from platforms to an archway. It just felt good at this value.

Movement Speed - 920. Without sprint this would feel too slow, however this speed combined with the sprint settings made for a fast but controlled navigation through the levels.

Sprint Speed - 1020, Sprint Acceleration 1000. - After testing acceleration I found I needed a higher value to simulate my character gaining speed after a quick turn, or landing from a dodge or jump. This value worked well so I quickly got up to full sprint speed after any slowdown. However the dodge spamming player will still effectively be going at 920 when not dodging. It was hard getting sprint to values I liked, but here you don't notice it as it feels intuitive. I know this is unreal but it makes sense that a sudden change of direction or landing from a fall wouldn't let you keep your top speed.

Dodge Landing Speed Factor - 0.3 - One of the most important changes (imo UT needs to move away from the crushing deceleration after a dodge) and why I left the sprint in. At 0.3 there is a much smoother transition from dodge landing to running, making the movement flow better. The dodgers penalty is transferred to being at 920 movement speed down from 1020. I also found this helped level navigation as having a slower max speed just after a dodge helped me navigate the environment better and plan my next dodge/jump/walldodge.

I think I will now retire from movement tweaking for a while and get drawing. I've said my bit, hope it helps.

EDIT: On the subject of dodge-jump: Settings of 400 for the jump apex setting and 525 for Z velocity seemed ok for me. I know I said I wasn't considering it but at this stage I think changing your mind is a good thing . Movement wise it makes the wall dodge or boost dodge more redundant in some areas which is why I wanted to just focus on wall dodge and UT99 movement, but this version of dodge-jump is less broken than UT2004 and better fits with the level scale. What I'm worried about is the boost dodge being nerfed (because it doesn't really make sense to gain the most height from wall-dodging at the start of your jump compared to the apex) despite it being fun to move around the level with. It gives a feeling of the dodge-jump while being constrained to the walls makes it mesh better with the level design imo.

[Sir_Brizz]

This has taken me a few evenings and most of today, trawling through different combinations of movement settings on all the test maps until I was finally happy with my results, so I hope somebody looks at it. (Such is my UT fandom) . Please DL the blueprint and give it a try.

BLUEPRINT: https://www.dropbox.com/s/bcdron75r4...Character2.zip

I can't believe how hard it is to settle with the values but I'll explain my reasoning and hopefully the video should show the movement is fast, engaging and scales very well with the levels. Dodge-jump and double jump are not enabled but can be for testing, I gave them a really good shot but just found it detracted from the gameplay and helped to invalidate the level design. Lets leave the dodge-jump argument out for now. I like it in 2004 and I voted for it in the poll on these forums, but from my testing I've changed my mind. However I think the boost-dodge and wall-dodging can bring that feeling of freedom of movement and air-time to the game that dodge-jump does while not being able to do it anywhere, so all hope is not lost d-jump fans!

Main changes and reasons why:

Jump Height - Increased to 800. Without extra height and control of the double jump (plus higher gravity), i believe this value needs to be substantial enough to jump the small crates and ledges in the levels. If I can't jump those, why have a jump? I can now jump onto those small crates with weapons or armour on in deck (which I've always wondered why you can't) as well as some other smaller ledges you'll see in the video. However the jump isn't too high that I bump my head or overshoot jumping targets or remain in the air so long I'm overly vulnerable.

Accelleration - 7250, Decelleration 2000 - These settings brought me the closest to a good strafe responsiveness without feeling overly twitchy or heavy.

Gravity 2.2 - Worked best with the levels scale, so I wasn't smashing my head when dodging down from platforms to an archway. It just felt good at this value.

Movement Speed - 920. Without sprint this would feel too slow, however this speed combined with the sprint settings made for a fast but controlled navigation through the levels.

Sprint Speed - 1020, Sprint Acceleration 1000. - After testing acceleration I found I needed a higher value to simulate my character gaining speed after a quick turn, or landing from a dodge or jump. This value worked well so I quickly got up to full sprint speed after any slowdown. However the dodge spamming player will still effectively be going at 920 when not dodging. It was hard getting sprint to values I liked, but here you don't notice it as it feels intuitive. I know this is unreal but it makes sense that a sudden change of direction or landing from a fall wouldn't let you keep your top speed.

Dodge Landing Speed Factor - 0.3 - One of the most important changes (imo UT needs to move away from the crushing deceleration after a dodge) and why I left the sprint in. At 0.3 there is a much smoother transition from dodge landing to running, making the movement flow better. The dodgers penalty is transferred to being at 920 movement speed down from 1020. I also found this helped level navigation as having a slower max speed just after a dodge helped me navigate the environment better and plan my next dodge/jump/walldodge.

I think I will now retire from movement tweaking for a while and get drawing. I've said my bit, hope it helps.

Wow, I really like this one. Good job. My feedback on it would be that the sprint acceleration you put in just lends credence to speeding up movement and removing sprint. I changed the movement speed to 1100 on your prototype and it felt REALLY good. I also dropped the dodge cooldown to 0.25 again. I really think that is a sweet spot for dodge chaining. I also think perhaps the dodge speed landing factor could be 0.4 but that's entirely preference I think.

The two problems I still have that maybe can be tweaked is that the acceleration still has too much of an affect on air control. If you dodge sideways from standing still and immediately hold forward, you completely change direction. Can't have that. The other problem is that strafe dynamics still feel a little muddy. That's why people have been pushing acceleration so far, although you can diminish it with higher deceleration.

[TheWhiteDragon]

The air control vs accel/decel thing is really irritating me. It would be better if we got rid of the aircontrol variable and had a completely different set of accel/decel/speed variables for when you are in the air.