Ruien wrote:
But it seems that this is the right approach. Also, if Portal didn't lose momentum randomly (supposed to be air resistance?), this would work.
Ha, nevermind. You just jump after it, falling and then placing the portal below you right as the third fizzler is deactivated.:
win
||
2012-02-02_00008.jpg
Ok! Congratulations Ruien! Finally solved after 16 days!*, 14 updates and more than 100 posts to the forum.
Your ending is actually a bit cooler than the intention to be honest. As you'll see in my video, you're not actually required to jump through the portals but cudos for doing it that way. 
Also massive respect to Jonatan for pretty much describing the intended part-two solution as well as the jump-through alternative. I would love to know why he didn't post a solution by now?! See below for the in-depth achievement list.
The creation process for my second map Encode has been a fascinating experience for me so I thought I would write a short article about it while I was waiting for someone to solve it. I'm still quite happy with my first map Kerplunk, but Encode is really on a different level in terms of difficulty, complexity and in the amount of interaction that was generated with players. Almost every inch of this map has been sculpted not just by me, but by the players who found so many clever and creative unintended solutions. I've made a list below of the players who solved the various pieces of the puzzle.
First of all, the video. Please excuse the cheese!
Official solution
ZlauZNHgBHo
Encode began as about twenty pages of scribbled drawings. It wasn't really made with Hammer, but with pencil, paper and an eraser. The idea was for the player to "encode" a series of laser commands, by placing cubes on the floor and having them funneled through a laser to "execute" the commands. The simplest idea was then for these commands to open fizzlers in sequence and allow a cube to be lifted to the ceiling by the same funnel. Having prepared the room, the player would simply stand on a button and admire the resulting machinery. As I found out, any time you allow 5 reflector cubes into the map, the possibilities for unintended solutions blossom almost out of control. Many updates were required to constrain various exploits and attempt to leave only the one intended and elegant solution which would explain the name of the map. The Lift was only intended as part one. The idea was that this should be quite achievable and that players might believe this to be the end of the map, but then be faced with another difficult section in order to exit.
The crux of the design was that, having painstakingly achieved the Lift, the only way to then exit would be to repeat the same feat again but using a completely different method. To then camouflage this solution with more-likely seeming alternatives that would turn out to be impossible. This was meant to be the root of all evil in this map.
The upstairs fizzler near the exit was originally button-less. Note that the button is not used in the intended solution. It is pure camouflage. Although it can be used to avoid re-shooting the floor portals it should mainly serve to tempt the player into thinking that three cubes must be taken through the door. Of course, ninja players managed to actually do this but those solutions were eventually shut down leaving players to face the fact that a cube must somehow be brought to the upper level without passing through the door. Xtreger and Jonatan both brought brilliant unintended solutions to this problem but reluctantly I had to close them off as well since I felt they would make the "true" solution too hard to find.
I hope you have enjoyed Encode. It turned out to be much harder than I had anticipated and solving it was really a community effort. Well done to everyone who worked on it!
Achievements: following are credits for the players who first described the intended pieces of the puzzle.
Xtreger: first and foremost for breaking the map so many times and helping me make it so much better (and glassy!) for players who started later.
marKiu: for noticing that the grate was angled and would be needed to open the door.
PCdoc: for noticing that stacking 4 cubes in the funnel allows you to portal the laser and funnel together
PCdoc: for sketching the lift: funneling a cube through the fizzlers and onto the angled grate with the remaining cubes providing the laser to open the door.
Xtreger: noted that two of the fizzler sensors are at the same shooting angle which is critical.
Xtreger: for the first use of a laser "encoded" command. He placed one cube in front of the funnel and spaced it so its command would open the first fizzler at the same time as another cube was lifting through. So close to seeing the fully encoded lift!
Jonatan: for outlining much of the part two solution: "The only purpose I can find of a floor portal of a certain direction is to have a cube bounce through two floor portals and deactivating fizzlers". 
KennKong: for being first to describe the idea of multiple encoded commands using his own terminology: "there's only one way to get the laser on cubes in the fizzler stack (call then dests), and that's to have the cubes pass through the laser in the funnel (call them sources)"
Xtreger narrowed this plan down even further, saying "I strongly feel that only one cube is to be placed beneath the fizzlers, facing the mid and top receptors, and somehow get some other cube to activate the bottom-most receptor". Bang on!
KennKong: correctly executed the Lift after Ruien suggested hitting the first sensor via portal. This was a huge breakthrough and allowed everyone to start focusing on the exit. Well done KennKong!
Xtreger was the first to demonstrate how to funnel the lifted cube and drop it through floor portals (although others may have done this, they didn't describe it).
At a time when the others were starting to doubt the bouncing approach, Jonatan expanded on his early idea for the finishing move: "Is it possible having a bouncing cube hit all three receptors going up through the fizzlers and catch it at the top?... maybe there is enough time to run from the arrow near the big fizzler?". Yes there is!
Ruien correctly guessed that the two arrows are used to create the angle between portals needed to rotate the cube, hitting the first sensor while falling and the other two sensors while rising and that a "calibration" would be needed first.
Jonatan then described how the final portal, run and catch could be timed.
At this point there were really no mysteries left to solve. It was just a matter of working out the angles and executing it.
Bizzarely, it would take more than one more day before anyone implemented a version of Jonatan's plan.
Unfortunate that I did not have time to test my idea for part 2 but it seems I was right then.
What I think is weird is why the cube did not lose more height than it did. You drop it in your video at 1:34 and pick it up at 2:55. That's a good 81 seconds it has been bouncing and it ends up at the same height. This is why I never really believed this could be possible. In my experience cubes lose height much faster.
Not only the cube physics but the unorthodox receptor mechanics and hidden fizzlers in grates. In your defense you did try to explain the behavior of the receptors somewhat, nevertheless I think it is best to keep all mechanics as standard as possible.
In any case I enjoyed the map a lot. Now go design map #3!
Nonetheless, it's all moot to me. Without a mouse, I just can't turn and shoot quickly and accurately enough to get the yoyo started. I won't rate the map until I can pull it off myself, but I am going to keep my promise to ding it for tight timing. I won't downgrade it just because it requires a mouse (not your fault I handicap myself), but you could remove the tight timing issue just by triggering the door closed when a cube enters.
Even so, the puzzle design is brilliant. It's as evil as Seven Gates in the way it makes you solve the same puzzle in two different ways, but even Jonatan must be impressed by the fact that it does it with the cubes in motion.
grayarea wrote:
I should have had an entrance puzzle which demonstrates any new mechanics in a simple and obvious way.
Again though, the sensor mechanics made the whole map concept possible and I tried to be crystal clear about that.
Map #3... I'm not sure I can go through this again!
This is officially my most favorite custom map yet, and it has nothing to do with whatever I did or anything - just on the basis of the superb puzzle.
So please make more maps, even if you do so after some break (in fact that would be better). Or a better suggestion - you can collaborate with jonatan to make a difficult test 
KennKong, one of the things I liked about the design is that it's not clear how many cubes you're meant to get through the door. For me, the two cube solution is a bit harder to execute than one. Anyway I warned you in the description. It's really hard to make a map that both you and Xtreger are going to like...
grayarea wrote:
Just to finish this one off, here is a short video showing two alternative (and more difficult) methods.
This is a great video!
Yeah, this is exactly how I did it, both with the two cubes through the door and by jumping down at the end. It requires precision but we have quicksaves, and after knowing what to do it's just a matter of execution at that point.
So, while this way is harder, keeping it as an alternate solution adds to the mystique of this map. Jonatan suggested and I agreed that "jumping after the cube" would be very cool if it could be pulled off- and it can!
Edit:... or quite possibly Jonatan and I have broken PCdoc's brain with our latest maps?!