This topic is for the purpose of discussing #7807: SpaceColonizer's NES Rampage "1 player" in 1:27:27.59

Firstly, I want to say that I've been waiting for a full TAS of this game for a while now, and had even considered attacking it myself. Review: Syncs on FCEUX 2.2.3 (yea I should probably upgrade this sometime soon). Desyncs on BizHawk when converted to .bk2 - Even when adjusting the first few inputs to get the first stage starting correctly, desync occurs by the end of the first stage. I'd be concerned that trying to fuly re-sync this for bizhawk may be a major challenge with RNG manipulation. Full transparency, I did not watch the entire run at normal speed, I watched it completely at turbo-speed. From what I saw, the run looks pretty solid overall. Regarding optimization: Since the game limits the collapse of buildings to one-at-a-time, optimization is effectively going to be limited to getting the first building falling as fast as possible, then attacking the remainign buildings to be ready to collapse by the time the first (or previous) is done collapsing. A cursory viewing seems to demonstrate that this is accomplished on most, if not all, levels. For anyone familiar with casual play of this game, it will be immediately noticalble how little damage is taken. This aspect of the TAS alone makes it superhuman. The only improvement I can imagine would be finding a stage where it may be possible to get a different building collapsing earlier. Unfortunately, I anticipate that this would affect RNG and require full redo of everything beyond that stage. _{If I have time before a judge ultimately decides on this submission, I may look into going through stage-by-stage and seeing if I can get differnt buildings toppling earlier, but I'm pretty limited on time right now.} One other possiblity for improving the run might be adding the 2nd player. Normally speaking, a 2nd player wouldn't speed anything up in regards to toppling buildings due to the on-at-a-time building collapse rule. However, if there are stages where Player 2 can get a building toppling faster than Player 1 can due to proximity of the building to the right hand side of the screen, it would yield a faster overall TAS. This would obviously require a complete redo of the game using 2 players. OR, if it would be acceptable, use the "join a game in-progress" feature to use this run as a base and join the 2nd player in just before the first stage where the first right hand building collapse would be faster. Personnally, I think just starting with 2 players would be the better/more impressive option. I may do a more in-depth review of optimization of 1st builidings later on if I'm able. I may also consider doing a 2-player run myself someday. I would have made a temp enocde if I could have gotten a bizhawk sync, but I couldn't. And I always seem to get crappy audio when dumping temp encodes from FCEUX, and I don't have time to troubleshoot that right now.

I think I found a possible improvement. Since the speedrunning strategy mostly consists in demolishing the the first building as soon as possible, the only stage that needs full routing is the last one, in order to be able to stop inputs as soon as possible. In that regard, I think that the last three buildings could be demolished faster. Here is the route you used: And here is my idea: Basically, instead of descending to the ground level and then attacking the last two buildings, we're attacking all the three buildings in a single go. SpaceColonizer, could you try to implement this? No need to add me as a co-author.

i'll take a look after work. but i guess since we measure by last input i could just phone it in on the last level and let military guys bomb the buildings for me. could probably even do that some levels earlier so long as the HP holds out.

i'll take a look after work. but i guess since we measure by last input i could just phone it in on the last level and let military guys bomb the buildings for me. could probably even do that some levels earlier so long as the HP holds out.

Nice idea! Though I foresee that it will require some luck manipulation. Edit: I must say though, I should discuss with the staff team to see if there is any preference. I think we usually give freedom of choice to the author, however the idea from my previous post would make inputs shorter while not wasting real-time completion.

OK. Implemented your suggested routing. Cuts off 196 frames from inputs. redoing the rng manips wasn't so bad. But when looking into fastest last input possible, found something interesting. First off... if you leave too many windows undestroyed, there are too many solider guys throwing grenades, and since you're not moving they will mostly hit you. These will break you down before the dynamite guys come out. So destroyed three buildings, with the last being the big one and that seriously reduced the incoming damage. Also since the route stops me on the top of a building, I was out of reach of the cop car until a missile knocked me down, so avoided that damage for a long while. then here's where things get really interesting... somehow the game glitched so even when on the ground i was immune to the cop car! without that, lizzie would have most likely been damaged to death before the dynamite guys finished the job. it does take a LONG time for those dynamite guys to start coming though. This cuts off another 214 frames for last input. I'm conflicted. I prefer that the buildings be destroyed be player action. but that surprise "f*** the police!" glitch is interesting. I think my preference would be to submit the updated version with your improvements, then just put in the notes about the fastest last input. It's as easy as "truncate from this frame to see X Y Z". EDIT: here's new TAS: https://tasvideos.org/UserFiles/Info/638050330128198356 truncate at 315159 to see the glitch

This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [4959] NES Rampage "1 player" by SpaceColonizer in 1:27:27.59