Erode online

[checkerz]

With basically the entire MXS community begging for this for yearsso if JLV could have implemented this, he would have a long time ago.

It can definitely be done. It'll still be fairly slow since the track modifications require a lot of stuff to be recalculated that would otherwise be pre-computed, but as slow computers become a smaller and smaller minority that becomes less of an issue.

When can we go this direction?

[Rayvenator]

With basically the entire MXS community begging for this for yearsso if JLV could have implemented this, he would have a long time ago.

It can definitely be done. It'll still be fairly slow since the track modifications require a lot of stuff to be recalculated that would otherwise be pre-computed, but as slow computers become a smaller and smaller minority that becomes less of an issue.

When can we go this direction?

Yeah at some point we have to start raising the bar on what computer is expected out of a player. We cant always dismiss mechanics because of slower computers wont run it, or we´ll get nowhere.

[SKlein]

As far as multithreading, there are some things that I could parallelize but the biggest bottleneck is I generate too much traffic on the graphics bus. I need [to] redo the terrain tessellator so it keeps more terrain data on the GPU instead of resending it every frame. It doesn't sound as sexy as multicore but it'd make a bigger difference.

Would redoing that tessellator suffice for the inclusion of online erosion?

[sethiemeboi]

As far as multithreading, there are some things that I could parallelize but the biggest bottleneck is I generate too much traffic on the graphics bus. I need [to] redo the terrain tessellator so it keeps more terrain data on the GPU instead of resending it every frame. It doesn't sound as sexy as multicore but it'd make a bigger difference.

Would redoing that tessellator suffice for the inclusion of online erosion?

[SKlein]

Would redoing that tessellator be enough of a difference for terrain deformation to work online?

[jlv]

With basically the entire MXS community begging for this for yearsso if JLV could have implemented this, he would have a long time ago.

It can definitely be done. It'll still be fairly slow since the track modifications require a lot of stuff to be recalculated that would otherwise be pre-computed, but as slow computers become a smaller and smaller minority that becomes less of an issue.

When can we go this direction?

I want to add scripting first. After that I don't know. Pretty sure that any date I promise will be way too optimistic.

As far as multithreading, there are some things that I could parallelize but the biggest bottleneck is I generate too much traffic on the graphics bus. I need [to] redo the terrain tessellator so it keeps more terrain data on the GPU instead of resending it every frame. It doesn't sound as sexy as multicore but it'd make a bigger difference.

Would redoing that tessellator suffice for the inclusion of online erosion?

Not sure if it'd speed that up or not. With a modern GPU drawing extra polygons isn't such a big deal so I could probably skip some of the stuff that tracks terrain roughness. But with the height field in graphics memory it'd have to keep uploading the modifications to the graphics card. I guess it'd be a win overall since it's not critical to have the height field be perfect so I could just upload a little bit every frame.

[SKlein]

Not sure if it'd speed that up or not. With a modern GPU drawing extra polygons isn't such a big deal so I could probably skip some of the stuff that tracks terrain roughness. But with the height field in graphics memory it'd have to keep uploading the modifications to the graphics card. I guess it'd be a win overall since it's not critical to have the height field be perfect so I could just upload a little bit every frame.

I don't know if you've explained this before or not, but is the current erode system locked into its set value (IE --erode 1.00 // 3.50, etc) at all times?
For example, if you were to coast through an SX turn would it erode at an equal amount to someone who goes through the corner with the throttle on heavy? If they are the same currently is there a way you could alter the erode system to adjust dynamically like that with bike RPMs or something along that line? I've never messed with erode so if that's already how it works the case disregard this question.

[jlv]

There are settings in the tileinfo file to let different types of terrain erode differently. "erode_lower" and "erode_upper" control the range of force that it takes to actually move the terrain.

http://forum.mxsimulator.com/viewtopic.php?f=11&t=3166

[SKlein]

From my understanding, the following two erode settings basically match my question about RPM/speed based erosion correct? I had assumed it would be a harder task to code something like that, but if not that's pretty cool.

erode_off_speed <feet per second>
erode_on_speed <feet per second>

Next question:
With erode online could it be possible to, in the case of an MX event, carry over an erode log/file from one moto to the next?
Example - Moto 1 finishes: server, saveerode (filename) ~~ Moto 2 Starts: server, loaderode (filename).
I'm not sure how practical it would be, I'm guessing a track would be destroyed after one moto alone anyways.

[mxsrider96]

You could save the track after the moto ends, package it, upload it, and tell everyone to download, but I see that causing more issues than it's worth. Would be cool though

[sethiemeboi]

From my understanding, the following two erode settings basically match my question about RPM/speed based erosion correct? I had assumed it would be a harder task to code something like that, but if not that's pretty cool.

erode_off_speed <feet per second>
erode_on_speed <feet per second>

Next question:
With erode online could it be possible to, in the case of an MX event, carry over an erode log/file from one moto to the next?
Example - Moto 1 finishes: server, saveerode (filename) ~~ Moto 2 Starts: server, loaderode (filename).
I'm not sure how practical it would be, I'm guessing a track would be destroyed after one moto alone anyways.

[jlv]

From my understanding, the following two erode settings basically match my question about RPM/speed based erosion correct? I had assumed it would be a harder task to code something like that, but if not that's pretty cool.

erode_off_speed <feet per second>
erode_on_speed <feet per second>

Next question:
With erode online could it be possible to, in the case of an MX event, carry over an erode log/file from one moto to the next?
Example - Moto 1 finishes: server, saveerode (filename) ~~ Moto 2 Starts: server, loaderode (filename).
I'm not sure how practical it would be, I'm guessing a track would be destroyed after one moto alone anyways.

The speeds are slip speeds. It's how fast the contact patch is sliding relative to the ground. The bike speed is unrelated.

I doubt I'd do the erode carryover. Probably won't even sync up late joining spectators.

[Ddavis]

I think you'd need a limit on track deformation, because if it just continued to get worse it would become unplayable for the physics of the game.

[mxsrider96]

I think you'd need a limit on track deformation, because if it just continued to get worse it would become unplayable for the physics of the game.

It's a lot to ask, but I don't think it would be too bad if people would think about the lines that will form before they go 9th gear double throttle into corners and wad their brains out

[Yamaha285]

Not promoting any other games but mx bikes erode online works great and it works with late arrival players as well. What would be different here compared to that game?

https://gyazo.com/db8d731b83e1b362efa7a80cbefd446a