ska wrote:YetiRevolution wrote:I think the lag is just caused by a slow harddrive. Like Eddy said, it has to write your new PBs to the .sol file, so if your harddrive is failing/slow that takes a long time. Not much to do about it.I think until macro retries with a blank sol, we should factor in level transition lag into his final time. I think he lost a good 10-15 seconds to lag.
Without an accurate way to time the lag I dunno how we'd factor it in to his time :/ The speedrun.com leaderboards are RTA (real time attack) which is usually meant to include all time in the game, except load times that can be accurately accounted for. That said I'm sure he can improve the time with a blank .sol so it probably won't matter for long.
I think it would be really cool if we had a tool that actually counted the in-game frames, but I don't know of a way to do that which accounts for deaths.
NReality has a built in frame counter in the top-left corner; you could just use that to add up all the frames. No gold calculations required (as it's a speedrun). If the player dies, just take a still frame of the frame count on that death and add it to the tally.This would completely eliminate any discrepancies to do with lag/how fast you can mash the space bar.
This is tricky, if you don't count the time inbetween the runs, you can just exploit this rule and take a break, or even practise the level in a different computer, which would be technically valid. I know I know, this is very unlikely, but the reason RTA rules are applied is because you need to have exploitproof rules in official comps, and RTA is the most secure method to ensure so.