Because of the time traveling mechanics of this speedgame (changing the system date midruns), there is quite a gap between the different platforms (DS, DSi and 3DS console families). Since system menuing is actual, optimizable inputs done by the runner (unlike loading times), I think just removing the periods of console menuing would go against the spirit of speedrunning. Removing the cost of the maneuver would definitely alter the run. That's why when I set this board up I wrote in the game rules that the IGT field would be used to hold a normalized time to minimize that inequality.

To best reflect player performance in normalized time, one would have to factor in everything that could affect the length of time traveling (number of days scrolled, if the SD card slows things down...). But that would be a lot of effort for something that will in the end not be quite accurate.

Here is a compromise that would essentially remove the menuing times while keeping the cost of time traveling and incentivizing runners to go fast during menuing :

• Figure out good, permissive upper thresholds for each of the three consoles
• For each submission (roughly) time all time travel segments of the run
• If a segment is below its console-specific threshold, then remove its time and replace it with the upper threshold of the fastest platform (DS)
• If it is above its console treshold, do the same thing but add the margin above the treshold

That means that the normalized time ("IGT") of a good DS run would be slower than its real time.

The upper treshold could be adapted on a case-by-case basis if a system has significantly slower/faster boot times i.e.

Comparison of the time travel segments* of NEREMIS' personal best (2DS) and mine (DS) 1st travel: 1:00.00, 0:21.00 2nd travel: 0:56.90, 0:22.13 3rd travel: 1:04.33, 0:21.56 4th travel: 0:59.73, 0:21.23 5th travel: 1:01.33, none *timing of the segments starting at the screen turning black at poweroff/sleep and ending on the the EA copyright text appearing on game launch Menuing was okay in both runs afaict. I suggest tresholds of ~8 seconds above the average of these segments :

• 1:10 for 3DS family consoles
• 0:30 for DS family consoles I don't have any data for DSi yet.

I never properly gathered data about game lag and actual loading times. It never appeared obvious to me whether a console was worse than the others in those regards so I'll ignore these for now.