Well I worked on it during my grinding, there are 3 different categories to be considered for ILs:

1. Any% : when the level is played for the first time in a new game, as fast as possible without skipping it.

2. New Game+ : when the level is replayed after having saved all the mandatory animals on your way during a previous run. This category doesn't apply to Scrat and Boss levels.

3. 100% : when the level is played for the first time with the condition of getting the star in normal levels and Boss levels, and picking the golden acorn in Scrat levels.

I specify "played for the first time" because some animals are mandatorily on your way and have to be saved during the first play. Those animals are gone afterwards, so It has to be considered for Any%.

Then we have to decide how to time the ILs. The simplest way to go is to consider only the IGT displayed at the and of Scrat levels and after clearing each normal levels. But there is no IGT for Boss levels. If we consider RTA, then it is more complicated to decide when to start and when to stop the timer.

About the optimization, only the Scrat levels or somewhat interesting because there is no cycles or RNG involved. For most normal levels, you pretty much rely on the earliest geyser cycles, some of which are related to the best possible Crash&Eddie RNG. And for Boss levels, resetting until you get slightly better RNG than the previous record isn't much appealing to me.

Lastly, if you play on emulator and use a savestate right before entering a level, resetting at the savestate make you get the same geyser cycles and the same Crash&Eddie behavior consistently. This means that you can easily setup the savestates in order to get the ideal cycles and RNG, then reset over and over while keeping those cycles and RNG. That is not interesting ILs to me, it is the method I used to optimize my gold splits.

I'll let other mods give their opinion on this analysis before adding anything.

"So, on boss-levels, if you finish before all possible coins had a chance to spawn you will still get 100% if you collected all that did spawn?"

Yes, the 100% counter will take into account only the loaded coins. By "loaded" I'm referring to all the coins that are still on the field, have been picked or have vanished. The coins that can spawn won't be loaded until they have actually spawned. On boss-levels, you can theoretically get an infinity of spawned coins, but only the coins that have been loaded will impact the 100% counter. The total score formula is something like 100 * (1 - Total unpicked loaded coins / Total loaded coins) %.

"Does the same go for normal levels? Can coins that spawn upon enemy death be ignored if said enemy is never killed?"

Yes, for normal levels, the coins that can spawn from an enemy won't be loaded until that enemy is actually killed and has released the coin. So you can just ignore as many enemies as possible and run directly to the end of the levels.

For the running levels Hero's Gauntlet and Passageways of Eternal Torment, once you took a path (either left or right), all the coins on the other path won't be loaded because the field itself isn't loaded. So you just have to pick everything on the paths you chose to take. Besides, because of the checkpoint immediately after each divided path, you cannot take the other path by death abusing.

For fairness between both PC and PSX version, we can simply ban skipping the "HERCULES" screen for the PSX version. I agree that it would be weird but it's a simple solution. That skip is possible on PSX due to a structural difference between both versions, just like the extra loads, except that the loads are removed for both versions.

"How does the coins work on boss-levels? Is 100% just all collecting all coins that have spawned or could finishing before still give you 100%?"

On Boss levels, unlike other levels, there are no permanent coins on the field. They are only coins that can spawn randomly from hitting Hades, falling rocks for Hydra, and skeletons for Medusa. Any coin that has spawned will impact the 100% counter at the end of the level. If a spawned coin isn't picked, it will vanish and then the 100% counter will decrease. So you have to pick them as quickly as possible once they spawned.

"You seem to have a pattern on Hydra to ensure all vases"

Yes the vases spawn following a very precise pattern on Hydra. Firstly, you must cut two heads, you'll have then six heads left to be cut. Secondly, the 4 vases are respectively released by cutting the 2nd, 4th, 1st and 7th head in that order. Finally, you will have the two last heads to be cut, you can cut them wherever between the vases.

"are vases just pure random on Hades?"

From what I experimented, on a normal Any% Hades level, you can get only 3 vases from hitting him. So in order to get the 4th vase, you have to death abuse after a few hits. The amount of hits that makes each vase spawn is pretty random, so we cannot rely on a minimum hit strategy. To be safe, I death abused right after getting the 2nd vase.

"we have to force runners to show the entire score screen"

Yes, we cannot just evaluate and count the coins that have been picked during the levels. For example, some coins can spawn out of screen from falling rocks on Hydra. We cannot tell if a coin spawned and vanished out of screen until the 100% counter has been shown. Another example in Centaur's Forest, if a bird is killed behind a tree and releases a coin hidden by the tree, we cannot tell with certainty that the coin has been picked or not. So the runner has to show the entire score screen. Besides, displaying your "Hero rating" which results from the 100% score must be part of "100%".

Also, on the PC version, if you skip the "HERCULES" screen, you will also skip the "100% counter" screen entirely because those screens are linked. So you have to wait for each letter to be cleared. On the PSX version however, you can skip the "HERCULES" screen and then show the 100% counter screen because they are independent screens. So the PSX version is technically faster than PC version after removing the loads.

Yo,

I post this thread to describe what a 100% category would look like. Taking into account all the collectables of the game, the 3 following conditions must be satisfied:

1. 100% Level Completion: You have to pick all the required coins, then show the "100%" screen at the end of each level. That includes picking all the permanently visible coins, and all the coins released by defeated enemies. If a coin is missed, the 100% will drop, thus the run is invalidated.

2. All Letters: You have to collect all the letters in the levels in which they appear, then show "HERCULES" in full letters at the end of the level. Missing a letter invalidates the run.

3. All Vases: You have to pick all 4 vases which contribute to creating a password, then show the password screen at the end of each level. Missing a vase invalidates the run. The trophies that give an extra life do not count as vases, missing those is fine.

If a screen is not shown in order to prove that all the conditions have been fulfilled, then the run will be invalidated.

I did runs under those conditions for the Medium difficulty on the PC version. This is what my best run looks like:

If you ever find this run interesting enough to consider creating an official 100% category, that would be cool.

Regards.

Hello, did anyone know about this strat at the last pole before the bird glitch?

Basically, Hercules uses his sword to grab the pole from the left side. It saves 0.5 to 1 second depending on how optimal the movement is. However, it is precise and will kill the run if failed.

Heya, the runs you did for both Scrat and Boss levels can be categorized as Individual Level Any% runs (no golden acorns and stars). There is no Individual Level categories created yet for this game. Since all the Scrat and Puzzle levels end with an in-game timer, we might created ILs. I don't know for Boss levels though but a consistent timing can be set. Else, a full-game Any% category has be thought in which it's not allowed to skip the Scrat and Boss level.s Since nobody has done a full run of it, the category hasn't been created. So you're free to do and post such a run, you seemed to enjoy the game that way. However, making a new category including only all Scrat/Boss levels is irrelevant because it's as if you gathered the 1st level of each world under a category. The same could be done for all 2nd, 3rd levels etc. Moreover, you'll have to finish the game in the first place before you could do a run. I let other mods give their ideas.

Hi guys, I was practicing All Letters category when this glitch happened:

I accidentally passed through the ground after only 1 jump+down. This glitch seems useful only for a TAS of All Letters, but it's pretty cool.

Hi guys, it's been almost 2 years and still no leaderboard for Ice Age 2 GBA? I know that no one is active for this game but why not creating its own leaderboard?

Hi guys, I know it's been a while but here is a quick load comparison between epsxe 1.7 and 1.8:

I synchronized a test run using epsxe 1.8 with my current Any% run (epsxe 1.7). Since there're no significant differences, I would advise runners to use the 1.8 version of epsxe just because this version is smoother than 1.7. Now that I have some more free time, I'm thinking about coming back running the 100% PSX category with epsxe 1.8 for a new deathless WR.

And happy new year.

I'll optimize my splits with epsxe 1.8 and see the differents compared to 1.7.

Zockerstu and I we used epsxe 1.7.0 for our last Any% runs. First I used epsxe 2.0.5 but the loads are a bit longer than 1.7.0 (30 seconds at least). I don't know about the other versions of epsxe and other emulators.

Here is my first 100% run. I intend to do a better one.

Hey, I finished the optimization of the GBA vesion. Here's my any% run: You can know more about how I played it in the video description.

Alright then, thanks. I'll post an any% run here. This game takes some time, so it won't be ready soon but I'm willing to do it. Cheers.

Hi guys, I'm interested in running the GBA version of this game, but there isn't a leaderboard for it. Does anyone know about the GBA version?