tileinfo info

ColtonD719
Posts: 631
Joined: Tue Jun 21, 2016 4:15 am
Team: Privateer
Location: Oregon
Contact:

Re: tileinfo info

Post by ColtonD719 »

I’ve been following the erode discussion in the 2018 NA SX thread and it got me wondering. Is there any difference, either in performance or the actual erosion, between say (game/server erode 1, erode depth 2) vs (game/server erode 2, erode depth 1)?

I’ve always assumed that the erode setting in the game/server and erode depth in the tilemap multiply each other, but otherwise have the same effect. Is that correct?
barrington314
Posts: 6614
Joined: Fri Feb 18, 2011 6:03 pm
Team: me
Location: Charleston, SC

Re: tileinfo info

Post by barrington314 »

jlv wrote:Instead of comments, would a "name" field that shows in the editor be better?
Yes, Please do. That would be really nice.
jlv
Site Admin
Posts: 14910
Joined: Fri Nov 02, 2007 5:39 am
Team: No Frills Racing
Contact:

Re: tileinfo info

Post by jlv »

ColtonD719 wrote:I’ve been following the erode discussion in the 2018 NA SX thread and it got me wondering. Is there any difference, either in performance or the actual erosion, between say (game/server erode 1, erode depth 2) vs (game/server erode 2, erode depth 1)?

I’ve always assumed that the erode setting in the game/server and erode depth in the tilemap multiply each other, but otherwise have the same effect. Is that correct?
Correct. It's multiplied with erode_inner_depth and erode_depth. Ideally the depth settings should be set so an erode setting of 1 is realistic.
Josh Vanderhoof
Sole Proprietor
jlv@mxsimulator.com
If you email, put "MX Simulator" in the subject to make sure it gets through my spam filter.
ColtonD719
Posts: 631
Joined: Tue Jun 21, 2016 4:15 am
Team: Privateer
Location: Oregon
Contact:

Re: tileinfo info

Post by ColtonD719 »

Makes sense. Thanks!
checkerz
Posts: 8787
Joined: Tue Apr 22, 2008 4:23 am
Team: RaGe Factory

Re: tileinfo info

Post by checkerz »

jlv wrote:The first number tells it when to use that setting. If the track is worn 1 foot below the original height, it will use the line where <depth> is 1. The second and third numbers scale erode_depth and erode_inner_depth.

You have the default correct. That will gradually reduce erode_depth and erode_inner_depth to 0 as the track reaches 1 foot underground. Should be something like dirt that gets progressively harder until becomes concrete at 1 foot deep.

If you want it to be unaffected for the first foot and then suddenly hit a hard concrete layer, try this:

erode_limit 0.0 1.0 1.0
erode_limit 0.9 1.0 1.0
erode_limit 1.0 0.0 0.0

Going the other way, if you want a hard layer that gets softer as it breaks through, maybe something like mud that's dried out on the surface, try this:

erode_limit 0.0 0.25 0.25
erode_limit 0.25 1.0 1.0
On Seattle, Stephen ran these settings:
erode_limit 0.0 1.0 1.0
erode_limit 1.0 0.0 0.0

And erode got ridiculously deep throughout the race.

That doesn't make sense to me as it should've gotten to 1 foot and basically stopped?
barrington314
Posts: 6614
Joined: Fri Feb 18, 2011 6:03 pm
Team: me
Location: Charleston, SC

Re: tileinfo info

Post by barrington314 »

checkerz wrote:
jlv wrote:The first number tells it when to use that setting. If the track is worn 1 foot below the original height, it will use the line where <depth> is 1. The second and third numbers scale erode_depth and erode_inner_depth.

You have the default correct. That will gradually reduce erode_depth and erode_inner_depth to 0 as the track reaches 1 foot underground. Should be something like dirt that gets progressively harder until becomes concrete at 1 foot deep.

If you want it to be unaffected for the first foot and then suddenly hit a hard concrete layer, try this:

erode_limit 0.0 1.0 1.0
erode_limit 0.9 1.0 1.0
erode_limit 1.0 0.0 0.0

Going the other way, if you want a hard layer that gets softer as it breaks through, maybe something like mud that's dried out on the surface, try this:

erode_limit 0.0 0.25 0.25
erode_limit 0.25 1.0 1.0
On Seattle, Stephen ran these settings:
erode_limit 0.0 1.0 1.0
erode_limit 1.0 0.0 0.0

And erode got ridiculously deep throughout the race.

That doesn't make sense to me as it should've gotten to 1 foot and basically stopped?
Most of the track was actually this:
erode_limit 0.0 1.0 1.0
erode_limit 0.4 1.0 1.0
erode_limit 0.5 0.0 0.0

Which makes your case even better yet, if it did go deeper than 1 foot. I didn't get in the server to save the terrain to see how deep any of the flat spots got. :(
jlv
Site Admin
Posts: 14910
Joined: Fri Nov 02, 2007 5:39 am
Team: No Frills Racing
Contact:

Re: tileinfo info

Post by jlv »

checkerz wrote:On Seattle, Stephen ran these settings:
erode_limit 0.0 1.0 1.0
erode_limit 1.0 0.0 0.0

And erode got ridiculously deep throughout the race.

That doesn't make sense to me as it should've gotten to 1 foot and basically stopped?
It'll go over the limit but not by a huge amount. It might get down to 1.25 or so with the limit at 1.

If you want the first little bit fast and then slow the rest of the way, try something like this:

erode_limit 0.00 1.00 1.00
erode_limit 0.25 0.25 0.25
erode_limit 1.00 0.00 0.00
Josh Vanderhoof
Sole Proprietor
jlv@mxsimulator.com
If you email, put "MX Simulator" in the subject to make sure it gets through my spam filter.
checkerz
Posts: 8787
Joined: Tue Apr 22, 2008 4:23 am
Team: RaGe Factory

Re: tileinfo info

Post by checkerz »

jlv wrote:
checkerz wrote:On Seattle, Stephen ran these settings:
erode_limit 0.0 1.0 1.0
erode_limit 1.0 0.0 0.0

And erode got ridiculously deep throughout the race.

That doesn't make sense to me as it should've gotten to 1 foot and basically stopped?
It'll go over the limit but not by a huge amount. It might get down to 1.25 or so with the limit at 1.

If you want the first little bit fast and then slow the rest of the way, try something like this:

erode_limit 0.00 1.00 1.00
erode_limit 0.25 0.25 0.25
erode_limit 1.00 0.00 0.00
By ridiculously deep, bike swallowing deep. Well beyond 1.25. With those numbers above that should've been "concrete" under.
jlv
Site Admin
Posts: 14910
Joined: Fri Nov 02, 2007 5:39 am
Team: No Frills Racing
Contact:

Re: tileinfo info

Post by jlv »

checkerz wrote:By ridiculously deep, bike swallowing deep. Well beyond 1.25. With those numbers above that should've been "concrete" under.
-1.25 can look pretty deep especially if it's +1.25 right next to it. It's important to measure so you know for sure what's going on. If you have a demo I could check it for you. Although for practical purposes if 1 is too deep, try .5 or .25.
Josh Vanderhoof
Sole Proprietor
jlv@mxsimulator.com
If you email, put "MX Simulator" in the subject to make sure it gets through my spam filter.
checkerz
Posts: 8787
Joined: Tue Apr 22, 2008 4:23 am
Team: RaGe Factory

Re: tileinfo info

Post by checkerz »

jlv wrote:Tag
JLV,

Could use your feedback.

This is the base setting we've come up with for much of Hangtown (excluding jumps/downhills) that works pretty well.

I would like it to develop similar (maybe even a touch faster initially) but any faster and it gets very chattery and we get the pixelated hard edge feel that's very unpredictable while running at 1.0. When they ran at .6 it didn't get chattery nearly as bad, but it only allowed one line to develop and didn't get rough fast enough to work properly to force new lines.

What can I do to make it form up in a similar speed without the chattery stuff?

Code: Select all

friction 1.0 0
friction 1.2 1.5
friction 1.1 15
friction 1.2 30
roll_resist 40.0 60.0 35.0
roll_resist 50.0 70.0 45.0
erode_depth 4.5
erode_lower 50.0
erode_upper 150.0
erode_displacement 1.75
erode_radius 3.5
erode_inner_depth 0.3
erode_inner_radius 1.75
erode_min_slope -0.45
erode_max_slope 0.45
erode_off_speed 10.0
erode_on_speed 15.0
erode_slip_base_scale 1.0
erode_slip_scale 1.5
erode_slip_lower 0.0
erode_slip_upper 5.0
erode_downhill_scale 0.8
erode_limit 0.00 1.00 1.00
erode_limit 1.25 0.35 0.35
erode_limit 1.75 0.15 0.15
erode_limit 2.25 0.05 0.05
jlv
Site Admin
Posts: 14910
Joined: Fri Nov 02, 2007 5:39 am
Team: No Frills Racing
Contact:

Re: tileinfo info

Post by jlv »

I'd increase the radius to make it less chattery. The larger the radius the smoother it'll be. Might have to use a smaller depth to compensate for the extra volume.
Josh Vanderhoof
Sole Proprietor
jlv@mxsimulator.com
If you email, put "MX Simulator" in the subject to make sure it gets through my spam filter.
checkerz
Posts: 8787
Joined: Tue Apr 22, 2008 4:23 am
Team: RaGe Factory

Re: tileinfo info

Post by checkerz »

jlv wrote:I'd increase the radius to make it less chattery. The larger the radius the smoother it'll be. Might have to use a smaller depth to compensate for the extra volume.
Any other thoughts on smoothing? Reason I ask is the larger radius makes even bigger, more massive "berms" and doesn't allow for as many ruts in the turns. I feel like I'm already wider than I'd like to be and I got to this number by reducing chatter as much as I could.
jlv
Site Admin
Posts: 14910
Joined: Fri Nov 02, 2007 5:39 am
Team: No Frills Racing
Contact:

Re: tileinfo info

Post by jlv »

Maybe try erode_upper 100 and erode_displacement 1. The radius is really small though. It's gonna be choppy.
Josh Vanderhoof
Sole Proprietor
jlv@mxsimulator.com
If you email, put "MX Simulator" in the subject to make sure it gets through my spam filter.
checkerz
Posts: 8787
Joined: Tue Apr 22, 2008 4:23 am
Team: RaGe Factory

Re: tileinfo info

Post by checkerz »

jlv wrote:Maybe try erode_upper 100 and erode_displacement 1. The radius is really small though. It's gonna be choppy.
Will do. To make sure I'm understanding correctly, the radius is the width of the terrain affected by the tire correct (both up and down)? So 3.5 feet being effected by an 8 inch tire seems on the big side already or am I missing something? Or just is on the small side for what the game can handle?

Sounds like I have more testing to do :?
jlv
Site Admin
Posts: 14910
Joined: Fri Nov 02, 2007 5:39 am
Team: No Frills Racing
Contact:

Re: tileinfo info

Post by jlv »

checkerz wrote:
jlv wrote:Maybe try erode_upper 100 and erode_displacement 1. The radius is really small though. It's gonna be choppy.
Will do. To make sure I'm understanding correctly, the radius is the width of the terrain affected by the tire correct (both up and down)? So 3.5 feet being effected by an 8 inch tire seems on the big side already or am I missing something? Or just is on the small side for what the game can handle?

Sounds like I have more testing to do :?
That was what I originally thought - that the only effect was the tires directly moving dirt around through roost. That's why the original roost based erode algorithm worked so poorly. I think it was actually you that straightened me out on that. There was a video that someone posted (I thought it was you) that was a fixed camera filming a berm. Every time a rider hit it you could see the entire berm move well beyond where the tires touched it. It was visibly moving several feet away from the direct contact and I'm sure it was imperceptibly affected even further than that.

So the new system tries to model it more like clay being squished around. Imagine you draw a grid on some clay and then push the center of the grid in some random direction. The radius is the max distance where the grid would be distorted by the force.
Josh Vanderhoof
Sole Proprietor
jlv@mxsimulator.com
If you email, put "MX Simulator" in the subject to make sure it gets through my spam filter.
Post Reply