Advent of Code 2021 Day 6 (BaseA Style)

Discussion thread for day 6 of the Advent of Code - https://adventofcode.com/2021/day/6

Accepted answers

All comments

Shortest gap between * and ** today, and fastest overall solve (started at 5:15am) (excluding day 1 where I can't remember when I started, but was probably the fastest day overall).

Spoiler

Workflow:

Macro:

It was much simpler than I thought, I was imagining that we were going to head back into monitoring germinating flowers in space (I think 2018?), and that the part two logic wouldn't just be a case of iterating over more days (added interface tool after the solve), but instead that there would be some need to keep track of generations as older fish might end up with longer gestation periods or something like that. So far AoC is being kind...

bflick

Tried to do it without an iterative macro, but couldn't figure it out. Looking forward to seeing if anyone does.

Spoiler

Day 6 Solution.yxzp

NicoleJohnson

I'm sure there is a math way to solve this, and given enough time I am sure I could have googled it... but today is not that day. Iterative macro it is!

Spoiler

Full disclosure, this was my second iterative macro... the first one was REALLY slow. But while I was waiting for it to run for Day 2, I figured out a way faster way to solve for both, using one macro with a configurable limit (same as @cgoodman3 & @bflick , looks like!) The trick for me was realizing I didn't have to parse an exponentially growing number of records. I could just count the number of instances of each number, use my iterative macro to adjust them down one (and add an 8 for any 0's) and then re-count my numbers and iterate again.

And the macro...

Super stoked about how this one ended up. Only 14 tools used, and the 256 day iteration runs in less than 1 second!! I love it when an iterative macro plan comes together...

Cheers!

clmc9601

Wow, the lanternfish really did take over...

Spoiler

I started with the approach of exponentially increasing the number of rows, but that would have taken all night to finish! Here's my updated version. Looks like it's pretty similar to the first several.

Screen Shot 2021-12-05 at 11.25.11 PM.png

Screen Shot 2021-12-05 at 11.25.11 PM.png

Screen Shot 2021-12-05 at 11.25.25 PM.png

afv2688

Spoiler

Built them together on a batch macro to get both results

patrick_digan

Spoiler

Workflow:

Iterative Macro:

AkimasaKajitani

Today is reflection.

Spoiler

For star1, I made a simple iterative macro. It works well for star1.

But at star2, it takes too long time and finally this macro could not output the result. Because too many data kept me away.

I didn't become aware of a simple thing!

My latest macro is as follows.

After finishing the result, I made the macro bellow to be able to change the iterative count.

Final Workflow :

GitHub

https://github.com/AkimasaKajitani/AdventOfCode/tree/main/2021

danilang

That's a lot of lantern fish, making for a very bright ocean.

Non-macro (here you go, @bflick) solution running in 0.3 sec for the 256 days. I have many happy!!

Didn't win Tool Golf, @NicoleJohnson, but I might just have the fastest solution

Spoiler

Uses a multi-row tool that looks back up to 17 rows with special conditions to calculate the current count of 8 and 6 fish.

Dan

Day 6.yxmd

patrick_digan

@danilang Very cool! @bflick It's not baseA (and it wouldn't beat the speed of danilang), but I used the alteryx abacus to solve it without macros

Spoiler

Day6.yxmd

estherb47

Went the iterative route, and like many others, realized that generating numbers was way too slow.

Spoiler

Generating the count of each number was the way to go. When you reach zero, replace those zeros with an 8 and a 6, copying the count of the number for zero for each.

I'm sure there's a non-macro way to do this, just haven't figured it out yet.

Anyone else frustrated in removing the commas when you generate a number with Alteryx?

DataG8

I finally had to bite on the macro bait. I think @NicoleJohnson is right that there is a mathematical solve with factorials and modulus functions but I didn't get there. Then I tried generating rows, that didn't work past 30 days. My first macro left it in rows and does the 80 days fine but slowed way down about 135 iterations in. After all that, I figured it out the same way most of the others did.

Cheers, Michael

Spoiler

Solution

Macro

dsmdavid

Different version of macro-less

Spoiler

with a `fun` generate rows...

https://github.com/dsmdavid/AdventCode2021

Edited for Golf points:

Spoiler

@danilang well, this one doesn't perform that well, but it counts towards golf points 😅

leozhang2work

Spoiler

Really went nowhere with this one, have to look what others did. @NicoleJohnson's '6,8' split give me the idea to do part 1, then part 2 went nowhere again with the speed, another idea on the summarize helped on the part 2. Learnt a lot about AMP engine for the wrong reason 😂

joshbennett

Like most others, iterative macro FTW.

While it runs in less than 1 minute at ~44s, I am curious if anyone comes up with a faster solution without an iterative macro - definitely seemed the path of least resistance to me.

Spoiler

aoc_2021_06_jbennett010.yxzp

afv2688

Non iterative 0.2 seconds

Spoiler

SeanAdams

Like many of you - I iterated using a macro - but used a Batch Macro instead just to buck the trend. I find that batch macros pretending to be iterative macros are MUCH easier to debug. I want to go back and look at some of the formulaic solutions now 🙂

Spoiler

first summarized so that you only had 9 buckets, and filled in gaps to make this easier.

You'll see that I pump the fish data out to a file

Then, using a Fake Macro Input (FMI) - simulated an iterative macro. Again, the great thing about this is it's super simple to debug (unlike an Iterative Macro) because each time you run it, it's one iteration

danilang

Challenge accepted @joshbennett

.3 seconds non-macro. See my post above. @afv2688 is measuring in European time. That's why his solution is .1 second faster than mine.

Dan

Mchappell67

Like everybody else, realized that the simple, brute force method was not going to work for part 2, so had to rewrite to group the days and handle mathematically. Took entirely too long as I struggled with the addition of the 0's to the 6's. Would really like the ability to create read-write variables. Anyhow, here's my solution.

Spoiler

advent6v2.yxmd

Quick Links

This months top contributors

atcodedog05 19458

Qiu 15866

binu_acs 15708

MarqueeCrew 13708

apathetichell 13703