jerky's Feed

07-20-2021 at 08:30 PM
Rate this Entry

More Dirty Adventures with Dig Dug

Dig Dug Rev2 romset progress report:

Average score up. Pumping down ;)

More to come...

  1. redelf's Avatar

    How you can remember all those different patterns is impressive. Add another 50k or so with some dirt digging and this would be record pace easy.

    Thanksjerky thanked this post
  2. jerky's Avatar

    With Dig Dug "patterns" the player is usually adjusting things on the fly across that repeating route to pump group most of the monsters together for the first rock drop. Then after getting the fruit you have to group up the last two monsters for the final rock drop usually using a similar route each time for specific rounds. This is due to random monster behavior like Frygar flaming, and the high resolution of potential game positions the player can exist in. For example it takes about 12 taps of the controller to dig through one square of dirt. The timing of pumping up monsters is also hard repeat. That makes it difficult to get the exact same result each time which leads to many slight variations of how each "pattern" can work out...or not.

    The Round 1-31 starting patterns aren't that hard to get down. You even can repeat a few of them with slight variations here and there for 2-3 times like Rounds 12/16/20, 13/17, 14/18, 15/19, 23/27/31, 24/28, 25/29 and 26/30. After that you can get away with using the same starting route for many of the repeating Round 32-35 boards.

    Once I get to the repeating Rounds 32-35+, I am breaking with the traditional way of handling Round 33+ by going for the first rock drop to the left all the time. I've found that the starting rock drop to the right can be much more consistent and safer (for me anyway). You also don't have to worry about squishing all the monsters with that first rock drop to the left except for the one at the top left of the screen...that then runs off screen ending the level before you can get the 8000 point Pineapple.

    Unfortunately there isn't a one size fits all solution I have found to make this work up to Round 137+ when you can finally use the same pattern over and over until the kill screen (again Rev 2 has the slow down bug fix). This has led me to use slight variations of the right rock drop starting pattern for these Round 33+ boards that for now I am keeping track of by glancing at my mobile phone where I have screenshots of them to point me in the right starting direction. Again like earlier Rounds there are pattern starts that repeat in pairs or groups of 3 or more occasionally, so it's less than 27 different ones until Round 137+ when you need to remember just one until the kill screen at round 256.

    As for the other repeating rounds...Round 32+ is pretty much always the same until the KS. Round 34+ stays mostly the same other than a special pattern I use for Round 110 that has odd monster behavior from time to time I choose to avoid. But at Round 138 on the Rev 2 romset all the early versions of popular "notch" pattern for the 34's breaks down and I switch to a single new one I came up with to get through this as best I can. The tricky repeating Round 35 board can be handled with a few similar patterns that change slightly at a couple key points as the monsters speed up. I've decided for now to accept a slightly less average score on these rounds because most of the 10k first rock drop patterns I have figured out have a very high risk versus reward factor. Like Round 33+ above you also need to use a bunch of different starting patterns to get those to work.

    With regards to world record pace...The Rev 1 WR scores are comparable to this romset until the slowdown bug kicks in around Round 126. At Round 136 the monsters freeze up and after that they begin to move super slow until around 149. From Round 150+ they slowly increase speed at a more reasonable rate until the kill screen at 256. The Rev 2 romset I am working on is the "fixed" one that got rid of the slowdown issues but kept the kill screen. For this version the monsters speed up as the programmers intending until Round 136-139. At that point you can use the same 4 starting patterns over and over until the KS.

    So it's a bit of an apples and oranges situation after Round 136 for score comparisons between Rev 1 and Rev 2. For Rev 1 games that means many more patterns and or grouping strategies to deal with/execute until the KS (and much more time required by hours), while a Rev 2 game can use the same 4 patterns until the end from that point.

    That said on a very good day I could potentially score about 1,035,860 by the end of Round 54 just after the first million point rollover, and about 2,052,570 by the end of Round 103 using the patterns I have in front of me now. That puts me slightly ahead by about 23k and 45k respectively versus the formidable Donald Hayes 5 man TGTS performance of 5,147,610 points. This is theoretical versus reality but that recent 1,032,780 point playback I shared above isn't far off that pace through Round 54 (again without digging out the whole screen every round which takes many hours longer to do).

    I am actually in the process of doing another save state score test run this weekend so I will see how high I could potentially get going this route I am on now. I need to know what my possible maximum scores are at each million point rollover are so I can decide whether to ditch a scoring attempt or not in the future.

    More to come...

    Likesredelf liked this post
    Updated 07-24-2021 at 07:59 AM by jerky
Join us