Back

Guides

Community created guides, helpful strategies, and more.
TOPIC | Coliseum Encounter Project
1 2 ... 4 5 6 7 8 ... 45 46
@Nika, you're amazing, thank you. It makes finding the enemy so much easier. ^^
@Nika, you're amazing, thank you. It makes finding the enemy so much easier. ^^
rIxOVnO.pngdQDtyBL.pngGaCoVKV.pngnJW9Vjy.png
@KaidaHime No problem, and thanks for pointing it out!
@KaidaHime No problem, and thanks for pointing it out!
@Galeia
Pinging myself so I don't forget this exists....
@Galeia
Pinging myself so I don't forget this exists....
NPpR96f.png
@Ammoth @Coldblooded @Drezdin @Maki @nika @Sylvandyr I got the Woodland Path up to 5000 enemies in preparation for the Nature festival! The Nature % doesn't look quite as good with the higher numbers as it did originally, but it's still a nice 22%, and noticeably higher than the equal-pack-frequency based 19%.
@Ammoth @Coldblooded @Drezdin @Maki @nika @Sylvandyr I got the Woodland Path up to 5000 enemies in preparation for the Nature festival! The Nature % doesn't look quite as good with the higher numbers as it did originally, but it's still a nice 22%, and noticeably higher than the equal-pack-frequency based 19%.
@nika - Great! I think I can get some Kelp Beds done this week. It'll probably be a point of interest for flights pushing during the festival.
@nika - Great! I think I can get some Kelp Beds done this week. It'll probably be a point of interest for flights pushing during the festival.
Bonsai pixels (tofu and tea motif) by miirshroom
@Sylvandyr Cool! I think I need a pre-festival break, but that's a good idea (and maybe Golem Workshop, but I doubt a lot of people do exalt-training there...).

As for later priorities, it would be good to get Delta and Training Fields data up to 5000 before the Light festival, since they'll be relevant then.
@Sylvandyr Cool! I think I need a pre-festival break, but that's a good idea (and maybe Golem Workshop, but I doubt a lot of people do exalt-training there...).

As for later priorities, it would be good to get Delta and Training Fields data up to 5000 before the Light festival, since they'll be relevant then.
@Sylvandyr @nika
I've just added my latest Woodland Path data, but as that is up to 5000 now I'm going to concentrate on the Delta.
@Sylvandyr @nika
I've just added my latest Woodland Path data, but as that is up to 5000 now I'm going to concentrate on the Delta.
tysharina_quicksilverchild_fr_by_tysharina-d8tcy3k.png
@Tysha Great, thank you!
@Tysha Great, thank you!
@nika - By the way, I have a minor concern - not sure if you and Drezdin discussed it already, sorry! - if you're counting number of enemies, different zones have different pack densities. So, 5000 enemies in the Kelp Beds is way fewer battles than 5000 enemies in Training Fields. But packs are randomized by battle.

Do you think it's necessary to up the threshold of "enough data" for the zones that naturally have larger pack sizes on average? Is a few hundred over 1000 battles for the big pack zones enough?

Here's the estimated pack size average per zone:

Training Fields: 1 < avg < 2
Woodland Path: 2 < avg < 3
Scorched Forest: 2 < avg < 3
Sandswept Delta: 2 < avg < 3
Forgotten Cave: 2 < avg < 3
Bamboo Falls: 2 < avg < 3
Waterway: 2 < avg < 3
Arena: 2 < avg < 3
Rainsong Jungle: 3 < avg < 4
Boreal Wood: 2 < avg < 3
Harpy's Roost: 2 < avg < 3
Mire: 2 < avg < 3
Kelp Beds: 3 < avg < 4
Golem Workshop: 3 < avg < 4

Just trying to gauge at what point I can be lazy and stop recording, and what the "would be nice to have once we take care of the essentials" number is. I won't go out of my way to record past 5000 enemies if I have nothing to do in the zone, but if there are "nice to have" numbers to hit, I may as well hit those if I'm doing the battles anyway.
@nika - By the way, I have a minor concern - not sure if you and Drezdin discussed it already, sorry! - if you're counting number of enemies, different zones have different pack densities. So, 5000 enemies in the Kelp Beds is way fewer battles than 5000 enemies in Training Fields. But packs are randomized by battle.

Do you think it's necessary to up the threshold of "enough data" for the zones that naturally have larger pack sizes on average? Is a few hundred over 1000 battles for the big pack zones enough?

Here's the estimated pack size average per zone:

Training Fields: 1 < avg < 2
Woodland Path: 2 < avg < 3
Scorched Forest: 2 < avg < 3
Sandswept Delta: 2 < avg < 3
Forgotten Cave: 2 < avg < 3
Bamboo Falls: 2 < avg < 3
Waterway: 2 < avg < 3
Arena: 2 < avg < 3
Rainsong Jungle: 3 < avg < 4
Boreal Wood: 2 < avg < 3
Harpy's Roost: 2 < avg < 3
Mire: 2 < avg < 3
Kelp Beds: 3 < avg < 4
Golem Workshop: 3 < avg < 4

Just trying to gauge at what point I can be lazy and stop recording, and what the "would be nice to have once we take care of the essentials" number is. I won't go out of my way to record past 5000 enemies if I have nothing to do in the zone, but if there are "nice to have" numbers to hit, I may as well hit those if I'm doing the battles anyway.
Bonsai pixels (tofu and tea motif) by miirshroom
@Sylvandyr Hmm, I see what you mean... We're doing statistics on enemies, not on battles, so enemies are the correct primary thing to look at, but you're right that it matters what size "chunks" they come in - the statistics I put in the bottom of the MAIN tab would only really be correct if each battle was a single enemy. It's the same as the problem of doing statistics on Gathering results when they come in stacks rather than singles. Sigh.

You're right that there's an issue here, but I don't have any good idea of how to statistically deal with it, or HOW the minimum numbers should be changed. (Besides the option of just going back to trying to get accurate pack frequencies and then doing statistics on that, like we used to, but that takes massive amounts of data to get reliable numbers, since there are so many more packs than enemies.)

I think this is probably too complicated an issue with (I think) too small a final impact to be worth worrying too much about (as in, figuring out the real statistics by simulations, etc). But yeah, informally prioritizing getting the higher-level venues to say 10,000, especially once we have 5,000 everywhere, wouldn't be a bad idea.
@Sylvandyr Hmm, I see what you mean... We're doing statistics on enemies, not on battles, so enemies are the correct primary thing to look at, but you're right that it matters what size "chunks" they come in - the statistics I put in the bottom of the MAIN tab would only really be correct if each battle was a single enemy. It's the same as the problem of doing statistics on Gathering results when they come in stacks rather than singles. Sigh.

You're right that there's an issue here, but I don't have any good idea of how to statistically deal with it, or HOW the minimum numbers should be changed. (Besides the option of just going back to trying to get accurate pack frequencies and then doing statistics on that, like we used to, but that takes massive amounts of data to get reliable numbers, since there are so many more packs than enemies.)

I think this is probably too complicated an issue with (I think) too small a final impact to be worth worrying too much about (as in, figuring out the real statistics by simulations, etc). But yeah, informally prioritizing getting the higher-level venues to say 10,000, especially once we have 5,000 everywhere, wouldn't be a bad idea.
1 2 ... 4 5 6 7 8 ... 45 46