Tuning

From Elma Wiki
Jump to: navigation, search

Although Elasto Mania is 99% about skill, having the bike act differently by changing hardware equipment or software settings can not be ignored.

History

Screenshot of internal 45, Sink

It was noticed already back in 2000 that Elasto Mania worked differently on different computers. One of the best levels to notice it, was internal #45 - Sink. This was because of Sink's first apple location had a small hole where left wheel had to be inserted. The wheel made weird small bouncy movement back and forth, some people got stuck there for a while (0,2s), while other's got out relatively fast.


When talking about a 0,2s advantage in one of the biggest höyling levels, people obviously became interested in what could cause the difference.


Later on it was noticed that changing monitor's refresh rate would affect the gameplay. The refresh rate itself was irrelevant, however when vertical synchronization was on, fps was synchronized to match the refresh rate. So, for example if you put the refresh rate to 85hz, the fps was 85. Note that the refresh rate must be set on the exact same resolution that is being used in Elasto Mania to have an effect.


Accordingly, when vsync was turned off, the fps raised to maximum possible depending on hardware. Typically this meant 200fps back in 2000.

Frames per second

So, the conclusion is that everything in tuning comes down to fps. Everything else, vsync, refresh rate, hardware, software and so on, is only indirect tuning (because they affect fps!).

Low fps (30-100)

Differences to medium/high fps:

  • Long keypress delay. - Fixed in Elma Online.
  • Faster rotation. - It is very small difference, but still a difference.
  • Wheels can enter holes that are smaller than the wheel.
  • Unpredictable wheel behaviour in "\/" shaped vertex holes. - Wheel can get sucked inside the polygon, or in another case, a high speed bounce up in the air can occur.
  • 0,02 faster than high fps on a flat 10 second level.
  • Human eye notices unsmoothness until about 60-80fps, depending on how hard you concentrate on seeing it.


It is important to note that the less fps there are the less predictable wheel behaviour is in tight holes. Also note that 30 fps is relatively a whole lot less than 100 (3,33... times)

Medium fps (100-300)

Differences to low/high fps:

  • Almost unnoticable keypress delay
  • Can still enter some holes that are smaller than the wheel, for example Shelf Life world record route.

High fps (300-999)

Differences to low/medium fps:

  • Most significant difference is in brutaling. Due to the high fps, player is able to do a "slide" brutal, which is seen in the world record styles of Uphill Battle, Hi Flyer and The Steppes.
  • Totally unable to enter holes that are smaller than the wheel - Sometimes it is seen in a battle that people can't enter hangs because they're too small. Check Level Designing Etiquette


Note that all this info only gives some direction to tuning. For example it is possible to have keypress delay for other reasons than the fps. Possible reasons are (b)elma.exe CPU priority, keyboard issues or monitor (LCD is considered bad for Elma) issues. It is impossible to say 30-100 fps does exactly one thing, when having for example 70 fps instead of 60 can make a certain trick from possible to totally impossible.

A discussion of the optimal fps for internal levels is held here: http://mopolauta.moposite.com/viewtopic.php?f=1&t=7523