CCLP3:
#147: 512 (+20, bc, tyJBL, Avalanche)
L#147: 503 L (+17, b+7, Avalanche)
Was off to bed when posting, but promised to edit in some comments:
This was quite a thing. I've been lucky to have worked on three intense levels this fall and all of them have produced what I was hoping for. Well, in this case we fell one move short of the next second in MS, but did improve the other side of the fence.
I remember working on Avalanche over six years ago, and I did get the records back then. I wasn't the first to attempt it, but I still managed a significant jump in the score (+49s). There are two main solutions to the level and it's not easy to gauge which 'direction' is quicker. Even pieguy expressed to me that he wondered which direction I chose, but didn't actually want to hear the answer before he had a chance to try it himself. Which he did half a year later, and beat my score by a good margin (+9s). I asked him if he had any idea what I might have been missing (though we didn't share routes; I was just wondering in general if he knew there was some trick I missed), and he only answered that there's probably more seconds available in the final room. I accepted that guess but I couldn't believe it was only that, because missing nine seconds in one room would be a lot, and I had put a lot of work into each transition. I had actually only worked on one of the solutions and it turns out I went the wrong way. (It turns out I was missing about three seconds in the final room for a best known solution in that direction of 495, which equals the score I managed to put together in the other direction before we started sharing routes.)
A couple of years later, J.B. came around to the level. I should ask about his side of the story, but here's how I remember it. It wasn't easy to score 500+ so he was talking to pieguy about it. Maybe he got to like 505 on his own (which was b+2 at the time), but pieguy helped him out with the ending. He had worked out a very efficient final room and shared that. That, along with finding small improvements elsewhere gave JB the current record of 512. He wanted to know if more was possible so he shared the route with me. We made an unusual deal, where if I found even a small improvement in the route (not necessarily worth the next second), we'd report the score as a joint bold. I looked at the solution, but JB's work seemed solid and I told him he was good to go. I didn't take the time to execute the score myself right away, so it took three years for this confirm. I had been both a bit afraid of the boosts in the final room and I wanted to have a chance for a fresh look at the level. I did find [3] more this time around, but we still haven't cracked 513.
I took a good stab at Lynx though, first finding b+2 and then b+4 and being pretty happy about that. This was all based on the MS route, and I concentrated my efforts in squeezing out as many block slaps as I could, mainly in the final room. I told JB I'd share whatever I found with him, and we struck a new deal that if he found anything more that I missed, this could still be a joint bold. So I took a final pass at the level and did find improvements that gave me three more seconds for a 503 score.
Before I get into what we'll do next, here's a clarification of where we stand now. The MS route was put together by JB with a little bit of help from pieguy. (I assume pieguy also has the route.) JB shared this solution with me three years ago, and I finally executed it. I put more effort into the Lynx transfer of this route, and along the way I found [3] more that also helps in MS but falls [1] short of a new record. JB will soon confirm the Lynx score. If he finds any improvements (before the year is out) this score may yet be a joint record. We hope to some day score 513.
We also want to help others get to the finish line on this one, but want to find the right way to do that. We talked about setting a threshold, so if you score a certain amount by yourself, we'd share the bold route and you wouldn't have to eek out everything by yourself. We didn't settle on anything, but 490, 500, and 505 could all work, depending on the difficulty desired. Another way would be to release just the method for the final room and let you do the rest. Or to post hints on what to do. Or to just release the full solution publicly so everyone can see the cool stuff happen. These all have merits, and some of the reasons aren't fully clear if you don't know the route(s). Right now, we're still working this out. For me, the reasons to let you do some of the work comes down to letting you enjoy that moment of discovery. There's more to this level than just pushing blocks optimally in each transition. (Why were the first records so much slower?) Try it yourself!
Overall, I've had a blast with this level. I hope you can enjoy it too. Some day I will record the 'long solution' with some commentary. And maybe get that one last move from somewhere...
-Miika
#147: 512 (+20, bc, tyJBL, Avalanche)
L#147: 503 L (+17, b+7, Avalanche)
Was off to bed when posting, but promised to edit in some comments:
This was quite a thing. I've been lucky to have worked on three intense levels this fall and all of them have produced what I was hoping for. Well, in this case we fell one move short of the next second in MS, but did improve the other side of the fence.
I remember working on Avalanche over six years ago, and I did get the records back then. I wasn't the first to attempt it, but I still managed a significant jump in the score (+49s). There are two main solutions to the level and it's not easy to gauge which 'direction' is quicker. Even pieguy expressed to me that he wondered which direction I chose, but didn't actually want to hear the answer before he had a chance to try it himself. Which he did half a year later, and beat my score by a good margin (+9s). I asked him if he had any idea what I might have been missing (though we didn't share routes; I was just wondering in general if he knew there was some trick I missed), and he only answered that there's probably more seconds available in the final room. I accepted that guess but I couldn't believe it was only that, because missing nine seconds in one room would be a lot, and I had put a lot of work into each transition. I had actually only worked on one of the solutions and it turns out I went the wrong way. (It turns out I was missing about three seconds in the final room for a best known solution in that direction of 495, which equals the score I managed to put together in the other direction before we started sharing routes.)
A couple of years later, J.B. came around to the level. I should ask about his side of the story, but here's how I remember it. It wasn't easy to score 500+ so he was talking to pieguy about it. Maybe he got to like 505 on his own (which was b+2 at the time), but pieguy helped him out with the ending. He had worked out a very efficient final room and shared that. That, along with finding small improvements elsewhere gave JB the current record of 512. He wanted to know if more was possible so he shared the route with me. We made an unusual deal, where if I found even a small improvement in the route (not necessarily worth the next second), we'd report the score as a joint bold. I looked at the solution, but JB's work seemed solid and I told him he was good to go. I didn't take the time to execute the score myself right away, so it took three years for this confirm. I had been both a bit afraid of the boosts in the final room and I wanted to have a chance for a fresh look at the level. I did find [3] more this time around, but we still haven't cracked 513.
I took a good stab at Lynx though, first finding b+2 and then b+4 and being pretty happy about that. This was all based on the MS route, and I concentrated my efforts in squeezing out as many block slaps as I could, mainly in the final room. I told JB I'd share whatever I found with him, and we struck a new deal that if he found anything more that I missed, this could still be a joint bold. So I took a final pass at the level and did find improvements that gave me three more seconds for a 503 score.
Before I get into what we'll do next, here's a clarification of where we stand now. The MS route was put together by JB with a little bit of help from pieguy. (I assume pieguy also has the route.) JB shared this solution with me three years ago, and I finally executed it. I put more effort into the Lynx transfer of this route, and along the way I found [3] more that also helps in MS but falls [1] short of a new record. JB will soon confirm the Lynx score. If he finds any improvements (before the year is out) this score may yet be a joint record. We hope to some day score 513.
We also want to help others get to the finish line on this one, but want to find the right way to do that. We talked about setting a threshold, so if you score a certain amount by yourself, we'd share the bold route and you wouldn't have to eek out everything by yourself. We didn't settle on anything, but 490, 500, and 505 could all work, depending on the difficulty desired. Another way would be to release just the method for the final room and let you do the rest. Or to post hints on what to do. Or to just release the full solution publicly so everyone can see the cool stuff happen. These all have merits, and some of the reasons aren't fully clear if you don't know the route(s). Right now, we're still working this out. For me, the reasons to let you do some of the work comes down to letting you enjoy that moment of discovery. There's more to this level than just pushing blocks optimally in each transition. (Why were the first records so much slower?) Try it yourself!
Overall, I've had a blast with this level. I hope you can enjoy it too. Some day I will record the 'long solution' with some commentary. And maybe get that one last move from somewhere...
-Miika