A small difficult puzzle for hardcore portal players.

Note: As all my maps, this may require rather advanced techniques. Also note that it is probably not controller friendly.

1.1 Bugfix and small adjustment to remove unwanted solution.
1.2 Added a fizzler.
1.3 Made tube smaller.

File Name: sp_devilstower_v1.3.zip
File Size: 1.72 MiB
Click here to download Devil's Tower

If I did it right, you weren't kidding about not being controller friendly (I use a touchpad). Very tight timing and portal positioning.

I made it to the top level before giving up. I couldn't open the door.
Speculation
There seems to be an infinitesimal window of error when setting things up to have the laser point down the center at the right time.

advanced techniques is an understatement.. lol I'm guessingfunnel climbing?

Hmm, I was going to get some work done today... oh well.

Well, I think I solved how it's supposed to be done, but I can't seem to quite pull it off.

Anyone else just lacking a couple of seconds on the timer right before reaching the top (with the laser in the correct orientation during ascent)? I'm not sure if this is due to my poor mechanics or if I'm missing a piece of the puzzle. It does seem like with perfect mechanics (and a really high mouse sensitivity) my current method would work perfectly.

I've now uploaded a new version (1.2) which will remove some of the very difficult (to perform) solutions by adding a fizzler.

Hah! I'm sure this is unintended but it's kind of fun: when I reached the second cube, I managed to throw it down to bounce through floor portals while it spins around the vertical axis. I then moved the floor portals in line with the laser coming from the first cube. In this state, the sensors are randomly hit as the cube spins and every now and then the funnel switches to blue for a while. By swimming (strafing left and right) I can stand almost still in the funnel and wait for it to change colour. This way I can move up and down the funnel as much as I want. Maybe I'll upload a video later when I'm on a better connection. It looks crazy!

jonatan wrote:
I've now uploaded a new version (1.2) which will remove some of the very difficult (to perform) solutions by adding a fizzler.

Hmm, this fizzler doesn't affect my solution at all.. but it's still pretty hard to perform, though I still feel like I'm on the right track. Can you tell me how much extra time is left over (the approximate extra time buffer) when doing it the intended way? If it's less than a second I probably have the solution and just need to work at executing it.

I don't want to exactly specify what is needed since it may give out too much information about the solution. What I can say, is that it is probably one of the more technical maps I have made. And I don't recommend anything else than a mouse for this one.

Okay, that sounds great (definitely don't spoil the solution!). I'm just doing it wrong- all I wanted to make sure is that I'm not searching for a technical solution when really it's all about pristine, impeccable execution.

I'll pick this back up tomorrow, and thanks for mapping!

Obviously the solution takes reflexes. My question is, assuming you're fast enough, will the solution necessitate many quick saves? Or is it pretty clean?

Solved it. I'd guess unintended, though. Recording now! Its taking some time to render the vid...

Intended solution does not require quicksaves but of course it might help to save time. When practiced a bit one can do a runthrough without any major errors. I will not lie though. It requires some delicate operations.

I'm very interested to see your solution. Prepared to hammer away any unintended issues..

Here's my solution. Sorry in advance for not editing more of the preparations out. Execution does require reflexes, but I can't imagine it requires any more than the intended solution. The one part of my solution that leads me to think it might be unintended is the unorthodox cube orientation.

Let me know Jonatan! Great map!

http://www.youtube.com/watch?v=ZjsFlD3BFa4

Well done! That's got to be close. Two Jonatan map solutions in a row? Impressive..

Very nice solution! Well done. It is not what I had in mind however. In particular in next version I will disallow shooting floor portals from top room. That is not intended.

Jonatan, that's not really required though. I've just successfully used HeadShot's method by shooting the floor portal just after jumping through the fizzler into the funnel. It's easier than shooting portals from the top anyway.

Hmm yes. I will need prevent that too.

Just one question: Is this possible to make funnel blue all the time?

Thanks guys! Good news, i found the intended solution! I'm going to cut the exit door open with the laser. If I just wait a little longer it should start to melt...

Yes, its possible did it, video is rendering

Should not be possible now in version 1.3 to shoot floor portals from the tractor beam. Now I hope it will provide more of a challenge.

Ok, there we go:

Sorry for missing beginning of run, I decided to record only when I figured correct position of cubes. It was done same way as HeadShot. If this is the intended one I'll record full run
nNCjnlni5nc

Well done Jacu! That is indeed the intended way.

I am sorry I was not able to give you guys more of a challenge. Well next one will be #10. I promise that one will be something special, and as challenging as I possibly can.

Intended solution:

3c86Pnx3j3c

Oh dear. I was so close. Believe it or not, before Encode, I had prototyped a similar puzzle which also involves bouncing two cubes through floor portals to continuously trigger 5 timed sensors. I guess I won't be releasing it! Mine has different angles though and I hadn't quite figured this one out yet, but I'd convinced myself it was necessary.
Looking forward to map 10!

Just for fun, a trick I found earlier which needs just two cubes on the floor and one visit to the top. I haven't recorded the rather impractical setup however!
for-fun
F2U3YRvyIcA

Awesome grayarea. I can imagine it must be quite difficult to set up that rotation...

Yeah it's pretty easy to set up slower rotations, but they're not quite good enough. To get this rapid spin I threw it against the wall from up at the arrow. Very unreliable though.

Okay jonatan, I managed to solve it by one_way_to_do_it
taking a screenshot of the floor from the funnel, putting it into photoshop, and working out the angles heuristically (drawing lines from the portal edge to the laser catchers, mirroring and rotating the lines by the relative portal angle). Here is an example:
portalpositioning.jpg
(once you have the orientation lines for a portal, you can drag the other portal around on the photoshop layer and play with them until they line up right)

It took me quite awhile to realize that it wasn't absolutely necessary to have both portals on the laser line and parallel for both cubes to get a piece of the laser.
This map was amazing - a true work of art

On a side-note, I had almost solved it a completely different way, but the timers were just a bit too short for it to work (about 1.5 seconds). However, that impossible solution has become several sections in my notebook of ideas, and maybe in the future I can apply those concepts in a map of my own.

Thanks again!

FINALLY ! ! !

Not sure if this is the intended solution, but its gotta be close . . .

Screenshot-12.jpg

I always try to make hard puzzles but i fail everytime and cannot think very clever.
How did you manage to make your puzzles really clever and hard?

i can only speak for myself but i think jonatan i using a quite similar way of creating his puzzles.
for very very hard puzzles u need something new what players havent seen before. in this map he uses two bouncing cubes in two portals and i havent seen that before. in seven gates he used the trick with the vertical lightbridge. once u have found a puzzle element like that u need to build ur map and ur puzzle around this. its a big help to draw things on a paper especially when ure using lasers. this requires a lot of imagination and creativity. then u start adding obstacles to avoid unintended solutions or to make the puzzle harder. dont focus on the look first and change it after u finished ur puzzle. another rule is the more variables (cubes, laser etc) u add to your map the harder it gets. but the more variables u have the more inintended solutions will occur. i personally dont like big maps. that means maps with 20 rooms that are all included to the same puzzle. the player gets distracted and feels lost because he needs to explore the whole map before he actually starts solving. try to minimize your map as much as u can so the player feels close to the solution.
another trick to raise the difficulty of your map is lettin the player almost solve your map but insert a twist at the end. like in this map u were able to activate the funnel in many ways. this distracts the player from the intended way because they will always clutch on their solution. in seven gates u need to solve the map two times with little changes in the second part. the first time was hard because it included many variables. but it was only a matter of time untill u figure out the solution. in the second part he hides his evil special element. but u wont find it because u are distracted by the way u solved the puzzle first. u keep on holding at your solution and dont see many obvious things. and jonatan is a genious at hiding his special element.

Just 4/5 on this one, because the execution was much harder than the puzzle. I was doing OK on this one until I tried to get to the top cube. you really don't need to read this part
I had worked out the setup (like HeadShot's) but I just couldn't hit the tight angles. The trackball on my keyboard is just too low-res to get the precision I need. Since I knew the method was good, and only my execution was stopping me, I noclipped to get the top cube.

The principle behind the third lift wasn't that hard to figure out; just do the math (2 cubes X 2 portals + 1 cube = 5 targets). But I spent hours tweaking portal positions to get them all aligned. Then I spent some more time tweaking the yoyos so they would last long enough (I would get to the top just fine, but by the time I got there, they were too out of sync for the next step).

I got to the top with the cubes in sync, but the damn funnel just wouldn't grab the cube on the laser. So it was back downstairs for some more tweaking portal positions.

I got all the cubes on their way, and then wore out my F7 key trying to grab them all. My trackball was really letting me down on this step, because it tends to turn up better than down, and left better than right.
I finally had all the cubes, and proceeded to hurt my F7 key some more trying to get to the door. I think a brief cooldown timer on that wouldn't break the map.

As brilliant as this puzzle design is, I think you tipped the scale toward execution difficulty too much. I preferred Seven Gates, where finding the solution was the hard part, not executing it. So on your next map, make the 10 on the difficulty scale come from the puzzle, please?

It seems that a yawning chasm has opened up between the hard maps like you, Chander, grayarea, DieHard34 and now marKlu, and the easy maps. I wish Miss Stabby would come back with a map like Angry Bombs that was a good balance.

@KennKong: I would suggest a different cheat if you feel you absolutely need more speed and can't get it with your trackball:
bind "1" "sv_cheats 1;host_timescale 0.1; host_framerate 300"
bind "2" "sv_cheats 1;host_timescale 0.2; host_framerate 150"
...
bind "0" "sv_cheats 1;host_timescale 1; host_framerate 0"
These will slow down the action (0 to restore). I've been using these for testing my new map and they're great for demo videos (although I may not have the framerate settings right yet).
However, to claim a clean solution to a new map I strongly believe that sv_cheats (including noclip) should never be used at any time during the process, including for practice or to gain information. What do you guys think?

KennKong wrote:
Just 4/5 on this one, because the execution was much harder than the puzzle.

I agree with this, although due to my ultra-high-precision mouse, the execution was only somewhat harder than the puzzle for me. I still spent several hours getting it right.

You can't please everyone all the time though, since I personally liked this map more than Seven Gates. I feel like the technical effort put into it was amazing. But, yeah, the meticulous execution was a downer.

KennKong wrote:
It seems that a yawning chasm has opened up between the hard maps like you, Chander, grayarea, DieHard34 and now marKlu, and the easy maps. I wish Miss Stabby would come back with a map like Angry Bombs that was a good balance.

I completely agree about the yawning chasm, and I think the sweet spot is right around 4-6 hours to solve a map and with execution completely trivial to perform. But again you can't please everyone because I think Miss Stabby's map was on the easy side and it's not even in my favorite map collection. But Devil's Tower (and grayarea's Encode) both are.

grayarea wrote:
However, to claim a clean solution to a new map I strongly believe that sv_cheats (including noclip) should never be used at any time during the process, including for practice or to gain information. What do you guys think?

Absolutely. I think this is a basic assumption made by anyone who's really solving maps for fun and challenge.

My wife helped out in this respect when I designed Engineer. If she couldn't pull off the execution immediately after I explained how to do a particular piece, I didn't include it. If she got motion sick from it that also was a sign that it needed fixing too.

Overall, though, the semi-recent parade of hard maps have been so much more interesting than the sea of mediocre easy maps released in the first 6 months after portal 2 was released, so I have to reserve most criticism. Just to have such smart and talented people making maps is simply awesome.

grayarea wrote:
However, to claim a clean solution to a new map I strongly believe that sv_cheats (including noclip) should never be used at any time during the process, including for practice or to gain information. What do you guys think?

I agree. For me, sv_cheats is surrender. How quickly I surrender depends on the apparent quality of the map. When playing Jonatan's (or your's) I know there is an elegant solution to be found, and I will struggle on for days. Then there are the maps that some create just to show their pimply-faced friends how fast their reflexes are, but the puzzle is crap. I'll cheat on those just to get them finished so I can rate them in good conscience.

But thanks for the tip on slowing the game down. That might make playing certain maps possible even with a controller.

Another amazing map Jonatan though I like it less then devilschamber. This is mainly by the fact of very difficult execution. For me puzzle can be hardcore but I would like to have a simple execution. This map succeed only as it goes on puzzle. I was close to solve it but had to look how to retrieve the third cube. Ive never imagined that you will be able to make the set up ||from the top||. I considered position of first cube but havent enough imagination that you can make it from higher level. On next part I tried to put one cube on another as also was unable to make the positioning as intended. But design, cleverness, the final nothing short of spectacular. All in all really great map 5/5. Waiting for more of your evilness