Experienced Forum User, Skilled player (1841)
Joined: 4/20/2005
Posts: 2160
Location: Norrköping, Sweden
I made a Pitfall TAS, partly to try out TASing on BizHawk, but also simply because I felt like it could be fun. I aimed for highest score, 114 000, and no deaths, while getting an as good (in game) time as possible. I finish the game with 1:34 left on the timer. The game is quite slow and not very exciting in my opinion... It is very easy (and quite fun) to TAS, so I think it's quite a good game for beginners to try out on. Everything in this game is based on global timers, so no fancy features like RNG in this game. And this game is also too cool to have an ending, so it just stops after you get the last treasure. I made maps, planned the route in detail, dissected all the object timers and such, so I had quite fun making this. Here's the movie file: http://tasvideos.org/userfiles/info/2039595497723066 I would be interested in hearing if anyone else has TASed this game, and if so, how their TAS compares to mine. I'm having serious doubts about this game quality when it comes to getting published... I think you can watch this in 200% speed (or even higher) without missing anything. Though I think (and hope) that some of you might enjoy this run purely for nostalgia's sake, and perhaps if you are interested the fastest route to get a perfect score in this game. Oh, and any encoder is very welcome to encode this run. Enjoy!
Experienced Forum User, Skilled player (1639)
Joined: 6/11/2006
Posts: 818
Location: Arboga, Sweden
I too have looked into this game. I find it incredibly boring and it makes a more or less useless TAS, despite this run being a really really amazing technical effort. I gave up watching after the first long stretch (some almost 8000 frames) but I suspect there might be a second one, longer going on. Despite this being a true technical effort, this is just a horrible game for TASing.
Warp wrote:
omg lol this is so fake!!!1 the nes cant produce music like this!
Experienced Forum User, Experienced player (925)
Joined: 9/29/2008
Posts: 964
Location: 🇫🇷 France
Over at Atari age, Thomas Jentzsch posted a modified emulator that plays "perfect" Pitfall and ends with the timer at 1:42 (see this thread). So your time can definitely be improved.
Marzo Junior
Editor, Experienced Forum User, Skilled player (1845)
Joined: 6/15/2005
Posts: 3192
I've made an "imperfect" run of Pitfall. The goal is to end the game by collecting all 32 treasures, allowing for deaths and point loss (<_<). http://tasvideos.org/userfiles/info/2154325866925250 Input file is 18m18s long, and 1:46 remaining on the clock. I intentionally aim for a nice-looking score at the end, though not the perfect 114000 score; if you think about it for a moment, you should know which score I aimed for. :) Some observations (please see http://pitfallharry.tripod.com/MapRoom/PitfallMap.html for a Pitfall map): - Taking the tunnel from screen 38 to 26 in the map above and going right for the last treasure is faster than staying above ground. It is about 8 seconds faster. I assume this accounts for the difference between 1:34 (Randil's run) and the 1:42 mentioned in marzojr's post. Dying twice accounts for the 4-second difference between 1:42 and 1:46, since the clock is stopped during death. - Even though, after getting the first treasure at screen 7, it takes fewer rooms to go left and take the tunnels at 1 and 251 instead of the tunnel at 12, for some reason it is slower. I've checked this. I have not tested the forward route from 1 to 239 or the backward route from 1/255 to 7 (though I believe this is slower) so who knows what will happen. - Unfortunately, swinging vines, crocodiles, and shapeshifting holes seem to be all on a global timer (the game clock, I guess). This makes it very hard for small improvements to matter. - I intentionally die twice. However, because the global timer is now shifted from Randil's run (with respect to input time), it introduces a huge variance in apparent time difference (anywhere from near-equal to almost 10 seconds faster) throughout the run. Near the end at screen 38, input times are nearly equal. - Rolling logs do not run on a global timer (only a local one, when you enter the screen) so it is possible to change their position relative to, say, the swinging vine, if necessary. - When bypassing a ladder, it is slightly faster to ladderboost by pressing down to mount the ladder, then dismounting at the top, instead of jumping over it. Mounting the ladder shifts x-position a bit. - Vineboosting may be used in some cases. I have not totally figured it out. Sometimes the shift in x-position is to the left, and sometimes it is to the right. - Address 0x61 is x-position. Important x-positions are 82 (when to ladderboost when going right to left), 108 (standing on right edge of hole without falling in), 44 (standing on left edge of hole without falling in), 62 (standing on left edge of left crocodile's head without falling in when its mouth is open), 51 (when to release vine to clear hole going right to left), and 99 (when to release vine to clear hole going left to right; may depend on even/odd frames).
Editor, Experienced Forum User, Skilled player (1845)
Joined: 6/15/2005
Posts: 3192
Imperfect run of the forward route (18m22s, 1:37 remaining) http://tasvideos.org/userfiles/info/2308650477446130 Add important x-position of 68 (when to ladderboost when going left to right)
Joined: 7/18/2011
Posts: 2
Here is a "perfect" run that ends with 1:42 left on the clock http://tasvideos.org/userfiles/info/52403100870488761
Post subject: Pitfall WIP
Experienced Forum User, Expert player (2128)
Joined: 11/26/2015
Posts: 180
Location: Australia
I had a go at this game a little while ago. I was going to do a deathless run than use that knowledge to figure out which deaths are the fastest and do a max score run with deaths. But the global timers annoyed me so I dropped it. I don't think I found anything that FractalFusion hadn't found. I haven't had time to compare everyone's times but I want to upload this in case there are any possible improvements. Here's the bk2. It's for Bizhawk 2.0.1: http://tasvideos.org/userfiles/info/52413181064048164 It's good to see you working on this, pozzum. I hope having my work (and the work of others) as reference can help a bit. I'm glad to see this game getting some TASing attention and I look forward to seeing more.
Active player, Experienced Forum User, Judge (266)
Joined: 10/16/2017
Posts: 442
Location: France
I think the most interresting TASes for this game are a no branch TAS and a maximum score TAS. pozzum beats by 8.015 seconds Randil's TAS for the max score. FractalFusion doesn't have a max score since he collects all treasures. then I need to compare The8bitbeast's and pozzum's TASes.