Author Topic: What is the the maximum release rate used to solve the following levels? (Read 13520 times)

Minim · « **on:** August 10, 2009, 09:43:35 AM »

This challenge was intended as the sequel to the 99 release rate challenge thread.

So, out of all the levels which are unsolvable with a 99 release rate, what's the maximum release rate used to solve these levels? I originally intented this for Lemmini as I claimed it had "a more accurate release rate than the DOS version and the Lemmix clone".

Anyway, these are the following levels which can't be solved with a 99 release rate, updated with new records.

Tricky 18: 87
Taxing 23: 91 (89 for 100%)
Taxing 28: 63 (3 for 87%)
Mayhem 10: 89 (97% result unconfirmed)
Wild 8: 89
Wicked 15: 77 (5 for 96%)
Havoc 5: 29
Havoc 9: 35
Havoc 12: 69
Frost 14: 53 (7 for 90%)
Hail 5: 91
Blitz 2: 97

ccexplore · « **Reply #1 on:** August 12, 2009, 08:17:31 PM »

I don't think you should lock the RR 99 thread. There's still the matter that some people may want to report 100% for some of the levels, plus there are other things people may want to comment on. Just let it go cold naturally like any other threads.

One thing I want to comment on for the RR 99 thread is, I would be a little wary if any of the results reported had only been done on Lemmini and the solution has timing-critical parts. The problem is I don't know how "trustworthy" the timing of various things are in Lemmini. I know all the results ClamSpammer and I have reported on that thread are DOS Lemmings/Lemmix, so it'd be good for other people to comment on which levels they've reported results on using only Lemmini, or confirmed that all their results were done on Lemmix/DOS Lemmings. Personally I would've kept "Lemmini-only" levels on a separate list, but regardless, if there are levels that somehow only work on Lemmini they definitely should be called out as such.

ccexplore · « **Reply #2 on:** August 12, 2009, 08:23:34 PM »

Quote from: minimac94 on 2009-08-10 03:43:35

This is the point where I prefer you use Lemmini, because this clone has a more accurate release rate than the DOS version and the Lemmix clone.

I don't get it, what's so "inaccurate" about the release rate of DOS and Lemmix?

The release rates cover all possible spacing between 2 lemmings from 4 pixels to 53 pixels, and works absolutely consistently (no elements of randomness and such).

I would push for keeping separate lists for Lemmini because the timing of various things may differ from DOS/Lemmix, and as you know, timing can be a factor when you need to deal with high RRs. Not to mention there might be glitches specific to Lemmini that doesn't work on any other versions of Lemmings or even clones thereof.

ccexplore · « **Reply #3 on:** August 12, 2009, 11:50:33 PM »

Tricky 18 can be done at RR 87: by placing the 2 builders strategically, they can add enough falling to the second lemming that overtakes the digger, just barely in time for the pit to become deep enough to trap him. See attached.

The best I can do with Wild 8 is RR 89. And Blitz 2 can be done at RR 97 using that same method that "almost but didn't work" on RR 99.

Proxima · « **Reply #4 on:** August 12, 2009, 11:57:19 PM »

Havoc 5 can be done at RR 29, thanks to a blocker/miner trick ccexplore used for saving 100% on Wicked 6.

Minim · « **Reply #5 on:** August 13, 2009, 06:49:04 AM »

Quote from: ccexplore on 2009-08-12 14:23:34

Quote from: minimac94 on 2009-08-10 03:43:35
This is the point where I prefer you use Lemmini, because this clone has a more accurate release rate than the DOS version and the Lemmix clone.

I don't get it, what's so "inaccurate" about the release rate of DOS and Lemmix?

I'll give you an example then. A 98 release rate has a 4-pixel distance which is equivalent to RR 99 in the DOS version and the Lemmix clone. Because the levels in Lemmini are basically doubled in size, The pixel spread of the lemmings is wider, (From 8 pixels to 106 pixels) making Lemmini the more accurate release rate. Oh dear.

I've just forgotten you can't use Lemmini for some levels, especially the Holiday lemmings levels.

namida · « **Reply #6 on:** August 13, 2009, 07:45:20 AM »

No, it doesn't make it more accurate. It makes it LESS accurate, because it's using it's own mechanics, not the mechanics of the proper Lemmings game.

Minim · « **Reply #7 on:** August 13, 2009, 07:58:30 AM »

Cool! A lot of odd number results going on here.

So I'll do this. If anyone (Including me) have any even results (Because they have Lemmini), I'll post them in and then I'll put in brackets the ones which are possible in DOS then.

ccexplore · « **Reply #8 on:** August 13, 2009, 09:08:16 AM »

Quote from: minimac94 on 2009-08-13 00:49:04

I'll give you an example then. A 98 release rate has a 4-pixel distance which is equivalent to RR 99 in the DOS version and the Lemmix clone. Because the levels in Lemmini are basically doubled in size, The pixel spread of the lemmings is wider, (From 8 pixels to 106 pixels) making Lemmini the more accurate release rate.

Well, it's not really a matter of "accuracy" then, just that since Lemmini has double the graphics resolution, there's a reason for it to actually differentiate even and odd RRs. Whereas DOS Lemmings have a lower graphics resolution, therefore it has no reason to differentiate between RR 98 and 99 because you can't do 3.5 pixels anyway.

Minim · « **Reply #9 on:** August 13, 2009, 08:36:53 PM »

Solved Taxing 28 with 72% at RR 18 using Lemmini. That was tough, but I'm sure it's improvable.

LemSteven · « **Reply #10 on:** August 14, 2009, 04:51:55 AM »

Quote from: minimac94 on 2009-08-13 14:36:53

Solved Taxing 28 with 72% at RR 18 using Lemmini. That was tough, but I'm sure it's improvable.

Try 70% with RR 63! Although I wouldn't be surprised if 71% is possible at the same rate. The key lies in a climber glitch that is pretty well known around here by now. This same glitch is used in the 70/80 solution to the level.

I used DOS, btw.

Minim · « **Reply #11 on:** August 14, 2009, 07:34:24 AM »

Mayhem 10 is possible with RR 50 plus one lemming. RR 52 is possible minus 1 lemming, try RR 51 because I can't be bothered to do this level anymore.

Dullstar · « **Reply #12 on:** August 15, 2009, 02:03:17 AM »

I may try to get a video of a strange lemmini glitch.

Minim · « **Reply #13 on:** August 15, 2009, 05:55:54 AM »

Cool!

I can't wait to see it. I can't do many glitches on Lemmini.

Anyway, I've put in all the results as 1 for all levels for a start (Hail 5 set to 20 as the minimum release rate) and I don't want to see them anymore. I know you can all do well on these levels I mentioned. Get going! (I might be so harsh, I may do some more Original and Oh No more lemmings myself)

Clam · « **Reply #14 on:** August 15, 2009, 08:15:26 AM »

I don't know how you came up with 77 for Taxing 23, but I managed to do this at 91 by making a 7-builder stack.

I tried Wicked 15 as well, and got up to 71 without losing too many lemmings to splats.

Proxima · « **Reply #15 on:** August 15, 2009, 09:33:05 PM »

The maximum RR for Havoc 9 is 35.

For Havoc 12 I'm fairly sure it's 15. (EDIT: to clarify, I have completed it at 15 so you can put this on the list. I'm just not absolutely certain whether it's the best possible for the level, just that it is the maximum with the method I was using.)

Dullstar · « **Reply #16 on:** August 16, 2009, 01:34:52 AM »

Most accurate RR: Amiga

That's the original version, after all.

Minim · « **Reply #17 on:** August 16, 2009, 05:59:02 AM »

The only "1" left is Frost 14 (Hmm... It's going to be over so quickly

) and also Hail 5 need to be done (Because the level's release rate is set to a minimum of 20)

Minim · « **Reply #18 on:** August 20, 2009, 09:19:29 AM »

Quote from: minimac94 on 2009-08-15 23:59:02

The only "1" left is Frost 14 (Hmm... It's going to be over so quickly )

Or slowly, because it wasn't bothered to be done yet.

Dullstar · « **Reply #19 on:** August 21, 2009, 04:54:34 AM »

All you have to do to activate the glitch is run the game... it's a major glitch that makes the game really hard to play.

Minim · « **Reply #20 on:** August 21, 2009, 07:18:14 AM »

Quote from: Dullstar on 2009-08-20 22:54:34

All you have to do to activate the glitch is run the game... it's a major glitch that makes the game really hard to play.

Dullstar, what glitch? You're going slightly off topic here and my project is almost finished. Why don't you do Frost 14 and Hail 5.

Clam · « **Reply #21 on:** August 21, 2009, 08:38:10 AM »

I had a go at Hail 5. 91 is definitely possible, and 93 looks like it might be as well, although I can't be sure of that right now.

Minim · « **Reply #22 on:** August 22, 2009, 09:54:14 AM »

My sequel project is nearly finished!

Even after Frost 14 has been done I still won't close the topic, I will need to check some levels to see if they're beaten by using Lemmini. (For Frost 14, Hail 5 and Blitz 2 it's very hard to use it on Lemmini, but it's still possible to do so)

Dullstar · « **Reply #23 on:** August 23, 2009, 07:52:08 PM »

Use Amiga lemmings for the most accurate results... remember, don't keep forgetting that the original version of lemmings was the Amiga version if I'm not mistaken.

Minim · « **Reply #24 on:** August 23, 2009, 09:35:54 PM »

Yes, I know that Amiga is the default version. Now will someone do Frost 14 (Sorry to be so harsh again, but this is the only level left with a release rate 1 and I still think it can be challenged).

Clam · « **Reply #25 on:** August 23, 2009, 10:32:55 PM »

All right, I'll do it.

I can get the required 80% on this level up to a release rate of 53.

Dullstar · « **Reply #26 on:** August 24, 2009, 03:48:05 AM »

Quote from: minimac94 on 2009-08-23 15:35:54

Yes, I know that Amiga is the default version.

That's kind of like when I accidentally said "5 word time limit!"

Clam · « **Reply #27 on:** August 24, 2009, 07:01:11 AM »

Quote from: Proxima on 2009-08-15 15:33:05

The maximum RR for Havoc 9 is 35.
For Havoc 12 I'm fairly sure it's 15. (EDIT: to clarify, I have completed it at 15 so you can put this on the list. I'm just not absolutely certain whether it's the best possible for the level, just that it is the maximum with the method I was using.)

Did you fix the entrances before playing this level? Anyway, I'm sure this can be improved - I completed it at RR17, and it looks like there's a bit of room to move yet.

Minim · « **Reply #28 on:** September 04, 2009, 04:43:25 PM »

Quote from: ccexplore on 2009-08-12 14:17:31

I don't think you should lock the RR 99 thread. There's still the matter that some people may want to report 100% for some of the levels, plus there are other things people may want to comment on. Just let it go cold naturally like any other threads.

Alright then. You don't have to say that anymore because I now agree with you. I assume that everyone was disappointed that I locked the RR 99 thread.

I was thinking that in this topic it was either a "yes" or a "no" on each level, and I was itching to press the lock button after they were done.

Anyway, I might also do perfect scores on these levels as well; And I also need to check if Mayhem 10 is doable with a 51 release rate because I haven't done that yet.

ccexplore · « **Reply #29 on:** September 04, 2009, 07:13:48 PM »

Um...thanks, although for the record, I don't think I've complained about you locking threads more than one or two times, since that one thread was I think the only thread you actually ever locked.

I don't know when I'll go back to revisit that thread, since there are all these new things going on right now that's taking up my attention.

Dullstar · « **Reply #30 on:** September 07, 2009, 05:20:56 PM »

Thanks for unlocking that thread.

Minim · « **Reply #31 on:** September 07, 2009, 07:22:26 PM »

You mean my RR 99 thread didn't you? Don't you want to put in some 100% results or something?

Otherwise, you're welcome for nothing.

Dullstar · « **Reply #32 on:** September 07, 2009, 07:30:04 PM »

Well, I will check a little for them. Of course, sometimes they take some difficulty to do the 100% solutions.

Minim · « **Reply #33 on:** April 15, 2020, 07:42:17 AM »

This thread hasn't been updated for a long while (10 years?!

), but like I said on another thread, it's nice to go back onto your old challenges, especially when an update (albeit quite an old one too) adds a feature which will help you but happens not to change the DOS mechanics. This challenge happens to be one of the blessings thanks to Lemmix's addition of frame control (Which wasn't added before until NeoLemmix implemented this feature). This really eases the solution's execution and because of that, it deserves a revival.

I made quite a few improvements, mostly for maximum %. I've bolded each of them. Most of the replays attached in a bundle will work with the old Lemmix Players currently on the NeoLemmix Website, and hopefully the new one too. I also decided to include Hail 5's 91RR result as it is fiddly to execute. The only one I'm stuck on is Mayhem 10's 97% challenge, though I improved on the 66% requirement significantly here as well. Frost 14 7RR is similar to the 90% solution I've pulled off yesterday.

Taxing 23: 91 (89 for 100%)
Taxing 28: 63 (3 for 87%)
Mayhem 10: 89
Wicked 15: 77 (5 for 96%)
Havoc 12: 69
Frost 14: 53 (7 for 90%)

My favourite level has to be Havoc 12, despite the fact there's only builders available. It's also the level with the best improvement, up from RR17. Hints below.

Spoiler (click to show/hide)

namida · « **Reply #34 on:** April 15, 2020, 06:56:32 PM »

Lemmix had framestepping before NeoLemmix even existed. NeoLemmix has added a few "smart" skips (like "go to next shrugger" or "go back to last action"), rather than just purely "jump back / forwards X frames" and the ability to customize the durations, and it has much better performance when frameskipping, but the existing old versions of Lemmix already had it too.

For reasons of replay compatibility, as well as that the new Lemmix might not be 100% accurate to DOS, I suggest sticking to the old one for challenges (until EricLang confirms that he's reproduced all known features that his existing code was missing, and someone with good knowledge of game mechanics such as ccexplore verifies this to be true).