Upwards bhops

[pants]

Had these sitting on a txt file on my comp, wanted to delete it but not lose this info.

Basically I was looking at how bhops would behave if you jumped up and onto them (stairs basically), I started with the assumption using a std. space only jump height, the bounce table heights and some basic maths etc. Good in theory but not every height worked, but a few did.

In short:
Raw stair bhop heights I found at decent heights for stairs, giving reasonable space to time the hops (you can do 2 in a row of the same height before your next bhop height changes)
30
32
42
45

If you want up to 7 in a row  (these I found just starting from 32 units high, which felt the most comfortable)

32, 32, 31, 31, 29, 29, 28

After 7 the pattern got weird and diverged from the 2 in a row thing it had goin on (and I basically got bored and stopped):
32, 32, 31, 31, 29, 29, 28, 26...
32, 32, 31, 31, 29, 29, 28, 29...

Its all (mostly) useless, but someone might find some nice ideas to use it.

[RNC1839]

This is very interesting to me. And I think some of your numbers are wrong. BUT I haven't tested anything, as usual. So I could be very very very wrong here, but whatever.

In case anyone wants to know how to calculate the absolute heights of these things (which is easier to understand imo) just find the raw height you want then add the height of the previous (or next) raw height you want then subtract one and round down. To know how much higher one is from the next just subtract ezpz, and I'll be doing the latter for the rest of this post.

With this using 46.381244 as a seed I get a pattern of 45, 46, 47, 46, 46, 47, 46, 47, 46, 46, 47 ... Which is a lot nicer.

But choosing a number with a lower (.1) or higher (.9) decimal value will be even nicer

Seed: 25.831251
Pattern: 24, 26 x 10, 25, 26 x 11, 25, 26 x 11 ...

Seed:  42.106243
Pattern: 41, 42 x 8, 43, 42 x 8, 43 , ...

Changing the decimal just changes how often the numbers bump up that 1 unit.

Now all that was just done using the same height every single time. Changing the height every time would yield much different results. And there are many raw heights that all result in the same calculated heights so you could play around with the numbers as much as you want.


Spreadsheets are sexy:
https://docs.google.com/spreadsheets/d/1SmPKKOUx1ElbVKi0YcL_Aqq2SXzgcNA41tFAmkCShXc/edit?usp=sharing

[RNC1839]

Just realized all my numbers were off by .03125 again... But the concept still works.

But anyway I found some thing fun with crouch jump bouncehops

Seed: 70.02539
Pattern: 69, 70 x 38, 71, 70 x 38, 71 ...

Even better

Seed: 19.995117
Pattern: 18, 20 x 203, 19, 20 x 203 ...

[pants]

These were all manually tested after some initial guesswork. I just made a whole bunch of stairs in hammer and plonked away with the tas plugin to find the heights that would work past 2+ bounces.

Nice to see there is some nicer patterns that can arise though. Sweet spreadsheet.

However I think there still might be something weird past 7+ which might be somehow related to how I was slowly decreasing height as the easiest way to find the next bounce, going up always seemed harder to find. And its probably just due to rounding.

edit: stuffed 1 number in 1st post

[RNC1839]

I just tested the seed of 25.800001 and I easily got 10 bouncehops (I didn't bother making more than that)

So now I'm just wondering how the hell your method works lol. My best guess is that you aren't using the same number of ticks between each bouncehop like my method does. Which is probably the reason that you get splits at 7+.

OK yeah that's what it was you were changing the timing of your bouncehops by adding 1 tick every few bouncehops. I just ran the numbers and you can get increments of 28-34 at least very easily, which is why you were getting your numbers. I am still confused how you start with two 32's tho.

I'll be looking at this more tomorrow for sure.

[pants]

Its not so much a rigorous method as far as: go into hammer and brute force it till I found something that worked :/

In the end 4 in a row seemed plenty hard for the bonus on huh.