| --- Log opened Sun Oct 29 00:00:53 2017 |
| 00:03 | <&jeroud> | https://usercontent.irccloud-cdn.com/file/3JNsVLJ3/Opus%20Magnum%20-%20Voltaic%2 0Coil%20(2017-10-29-00-55-35).gif |
| 00:04 | <&jerith> | I can probably cut down the area slightly. |
| 00:04 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 00:05 | <&jerith> | But that's the solution I'm most pleased with. |
| 00:07 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 00:07 | | mode/#code [+o Alek] by ChanServ |
| 00:47 | <&McMartin> | Oh, that's very nice, and it makes legit use of a 6-rotor, which is no small feat. |
| 00:47 | | * McMartin has finally worked out what 3-rotors are for and has drastically improved some of his times in chapters 1 and 2. |
| 00:49 | <&McMartin> | They let you keep your maximum input rhythm of all arms on period 4, but let you use grab-turb-turn-drop instead of grab-turn-drop-turn. |
| 00:58 | | * McMartin gets Hair Product down to 50 cycles, but dislikes what he had to do to make that happen. |
| 01:00 | <&McMartin> | (I had to make the steady-state highly irregular so that the *very first* output happened one cycle sooner. Ugliness!) |
| 01:02 | <&McMartin> | But the MoI does not have that problem, that can show off the savings a three-rotor can give. https://www.dropbox.com/s/c57n3orbb3makxa/Opus%20Magnum%20-%20Mist%20of%20Incapa citation%20%282017-10-28-15-48-41%29.gif?dl=0 |
| 01:12 | <&jerith> | I use that three-arm trick all over. |
| 01:12 | <&McMartin> | My previous uses of it never actually gained anything over two normal arms, though,and often lost something |
| 01:13 | <&McMartin> | This is the first place where I actually got speedups. |
| 01:13 | <&jerith> | I use the six-arm trick as well, but it turns out using two separate arms is usually better. |
| 01:14 | <&jerith> | The trouble with three-arms is that you often end up holding both ends of a structure and thus can't pivot it. |
| 01:15 | <&McMartin> | My solutions to date have made extremely limited use of pivots except in cost-optimizers. |
| 01:15 | <&jerith> | I'll probably keep the six-arm in that one, though. |
| 01:16 | <&McMartin> | Tracks, on the other hand, I made absurdly heavy use of. |
| 01:17 | <&jerith> | The throughput of the design is limited by calcifying the fire and I like the turning wheel aesthetic more than the pair of arms I'd replace it with. |
| 01:19 | <&jerith> | I don't use tracks much. |
| 01:19 | <&McMartin> | I'll be interested in trading solutions for Universal Solvent. |
| 01:21 | <&jerith> | I'll let you know when I've done it. |
| 01:22 | <&jerith> | (Probably tomorrow.) |
| 01:22 | <&McMartin> | Looking at my own solution, mine is very... typical of my first-cut attempts for complicated bits. |
| 01:22 | <&jerith> | How do you make double-ended structures without pivots? |
| 01:24 | <&McMartin> | rotation happens as part of transit, so if I'm not optimizing cost that's often a side effect of getting it where it needs to be. |
| 01:24 | <&McMartin> | Aggressive parallelism can sometimes assist too. |
| 01:25 | <&McMartin> | It's not that I don't *use* pivots, it's just that they very rarely show up in the final designs unless it's a very complicated structure and sometimes not then |
| 01:25 | <&McMartin> | My Curious Lipstick solution has a place that seems like it should have had a pivot but instead I just sweep a major structure directly into place with a giant turn from a length-3 arm, and it is *hilarious*. |
| 01:25 | <&jeroud> | https://usercontent.irccloud-cdn.com/file/xZNcbb6Q/Opus%20Magnum%20-%20Precision %20Machine%20Oil%20(2017-10-29-02-24-36).gif |
| 01:26 | <&jeroud> | That's a failed attempt at cost optimisation, but the pivot is typical of my first-pass solutions. |
| 01:27 | <&McMartin> | Yeah, that's extremely rare for me, though not unheard of. |
| 01:27 | <&McMartin> | My first pass would generally use a piston to extrude the linear pieces and then use an ordinary arm rotation to snap it into final place. |
| 01:28 | <&jeroud> | (Failed because I spent ten G more than you, but I can replace both arms with a single piston arm. |
| 01:28 | <&jeroud> | ) |
| 01:28 | <&McMartin> | My area solutions often pivot to make subsequent rotations restrict motion to area that's already been consumed. |
| 01:29 | <&McMartin> | An awful lot of cost optimization boils down to "one piston, one bonder, clever layout and extreme patience" |
| 01:29 | | Kindamoody is now known as Kindamoody[zZz] |
| 01:32 | <&jerith> | ... I just wrote out a 36 cycle pattern correctly the first time. |
| 01:34 | <&McMartin> | The thing about the long tedious patterns is that sigils stop being words and start being letters pretty quick, IME. |
| 01:35 | <&jerith> | I can read that in two different ways, with opposite meanings. |
| 01:36 | <&jerith> | But from my experience, I think you meant "you start working with groups (words) instead of individual sigils (letters)". |
| 01:36 | <&McMartin> | Yeah. |
| 01:37 | <&McMartin> | Especially for single-arm solutions, the amount of state you need to keep in your head is also very small so you can pretty much describe the solution as you see it happening in your own planning. |
| 01:38 | <&jerith> | In my case, they're literal letters. I invariably type them with one hand while moving the pointer with the other. |
| 01:38 | <&McMartin> | For some of the, er, less practical ones I would create dummy arms as scratch space to copy-paste too and rom. |
| 01:38 | <&McMartin> | *from |
| 01:38 | <&McMartin> | Oh yes, absolutely |
| 01:38 | <&jerith> | I keep wanting to just type without having to track with the pointer, though. |
| 01:39 | <&jerith> | I also want selective-scope repeats. |
| 01:39 | <&McMartin> | You can at least save yourself the clicks on the optio menu. |
| 01:39 | <&McMartin> | You are aware of the Ctrl-drag input, right? |
| 01:40 | <&McMartin> | It's not perfect, but it's most of what you need. |
| 01:40 | <&jerith> | Yes, that was the first option I toggled. |
| 01:40 | <&jerith> | Well, the second. The first was to tell it to use a sensible resolution instead of the actual pixels my display advertises. |
| 01:41 | <&McMartin> | 10 years and hidpi still doesn't work for shit |
| 01:42 | <&jerith> | It works much better than I ever expected it to back when I first saw it. |
| 01:43 | <&jerith> | Opus Magnum is actually the first thing I've seen that didn't Just Get It Right. |
| 01:44 | <&jerith> | (That said, I've had this machine for less than a month and am still in the painful adjustment period where I keep hitting unexpected deviations from what I'm used to.) |
| 01:52 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 01:52 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 01:26 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 01:30 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 01:30 | | mode/#code [+o Alek] by ChanServ |
| 01:53 | <&McMartin> | jerith: Now that I check: |
| 01:53 | <&McMartin> | - I do use pivots in my speed solution for Precision Machine Oil |
| 01:53 | <&McMartin> | - ... but not on Area or Cost |
| 01:53 | <&McMartin> | - I have managed to surpass your speed in Waterproof Sealant. |
| 02:09 | | celticminstrel [celticminst@Nightstar-kroq0e.dsl.bell.ca] has quit [Ping timeout: 121 seconds] |
| 02:09 | | celticminstrel [celticminst@Nightstar-d4m0po.dsl.bell.ca] has joined #code |
| 02:09 | | mode/#code [+o celticminstrel] by ChanServ |
| 02:46 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 02:46 | | mode/#code [+o mac] by ChanServ |
| 02:49 | | macdjord|slep [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 03:34 | | RchrdB [RchrdB@Nightstar-qe9.aug.187.81.IP] has quit [[NS] Quit: Leaving] |
| 03:35 | | celticminstrel [celticminst@Nightstar-d4m0po.dsl.bell.ca] has quit [Ping timeout: 121 seconds] |
| 03:37 | | celticminstrel [celticminst@Nightstar-j5322u.dsl.bell.ca] has joined #code |
| 03:37 | | mode/#code [+o celticminstrel] by ChanServ |
| 03:46 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 03:50 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 03:50 | | mode/#code [+o Alek] by ChanServ |
| 04:11 | | Vornlicious [Vorn@Nightstar-75s7vm.sub-70-211-140.myvzw.com] has joined #code |
| 04:12 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Operation timed out] |
| 04:14 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 04:30 | | Jessikat [Jessikat@Nightstar-bt5k4h.81.in-addr.arpa] has quit [[NS] Quit: Leaving] |
| 04:50 | | Vornlicious [Vorn@Nightstar-75s7vm.sub-70-211-140.myvzw.com] has quit [Ping timeout: 121 seconds] |
| 04:54 | | Vorntastic [Vorn@Nightstar-8mi86j.sub-70-211-147.myvzw.com] has joined #code |
| 06:08 | | celticminstrel is now known as celmin|sleep |
| 06:18 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 06:18 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 06:25 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 07:44 | | mac is now known as macdjord|slep |
| 08:06 | | macdjord [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 08:06 | | mode/#code [+o macdjord] by ChanServ |
| 08:08 | | macdjord|slep [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 08:43 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 08:43 | | mode/#code [+o mac] by ChanServ |
| 08:46 | | macdjord [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 09:07 | | Kindamoody[zZz] is now known as Kindamoody |
| 09:39 | | HTTP__GK_1WM_SU [DEEP-BOOK@Nightstar-jro569.static.sl-reverse.com] has joined #code |
| 09:42 | | HTTP__GK_1WM_SU [DEEP-BOOK@Nightstar-jro569.static.sl-reverse.com] has left #code [""] |
| 10:19 | | macdjord [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 10:19 | | mode/#code [+o macdjord] by ChanServ |
| 10:21 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 10:34 | | Kindamoody is now known as Kindamoody|afk |
| 11:26 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 11:29 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 11:29 | | mode/#code [+o Alek] by ChanServ |
| 13:19 | <&jeroud> | https://usercontent.irccloud-cdn.com/file/yhtLBjFn/Opus%20Magnum%20-%20Universal %20Solvent%20(2017-10-29-15-18-02).gif |
| 13:20 | | Vorntastic [Vorn@Nightstar-8mi86j.sub-70-211-147.myvzw.com] has quit [Connection closed] |
| 13:20 | | Vorntastic [Vorn@Nightstar-8mi86j.sub-70-211-147.myvzw.com] has joined #code |
| 13:20 | <&jeroud> | McMartin: My Universal Solvent ^^^ |
| 13:42 | | Vorntastic [Vorn@Nightstar-8mi86j.sub-70-211-147.myvzw.com] has quit [[NS] Quit: Bye] |
| 13:42 | | Vorntastic [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 14:25 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 14:25 | | mode/#code [+o mac] by ChanServ |
| 14:27 | | macdjord [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 15:18 | | celmin|sleep is now known as celticminstrel |
| 15:37 | | Jessikat [Jessikat@Nightstar-bt5k4h.81.in-addr.arpa] has joined #code |
| 16:06 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has quit [Ping timeout: 121 seconds] |
| 16:09 | | Alek [Alek@Nightstar-7or629.il.comcast.net] has joined #code |
| 16:09 | | mode/#code [+o Alek] by ChanServ |
| 17:13 | | macdjord [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 17:13 | | mode/#code [+o macdjord] by ChanServ |
| 17:15 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 18:03 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 18:03 | | mode/#code [+o mac] by ChanServ |
| 18:06 | | macdjord [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| --- Log closed Sun Oct 29 18:23:51 2017 |
| --- Log opened Sun Oct 29 18:24:05 2017 |
| 18:24 | | TheWatcher [chris@GlobalOperator.Nightstar.Net] has joined #code |
| 18:24 | | Irssi: #code: Total of 34 nicks [25 ops, 0 halfops, 0 voices, 9 normal] |
| 18:24 | | mode/#code [+o TheWatcher] by ChanServ |
| 18:25 | | Irssi: Join to #code was synced in 64 secs |
| 19:03 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 19:03 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 19:33 | | celticminstrel [celticminst@Nightstar-j5322u.dsl.bell.ca] has quit [Ping timeout: 121 seconds] |
| 19:33 | | celticminstrel [celticminst@Nightstar-qqdk24.dsl.bell.ca] has joined #code |
| 19:33 | | mode/#code [+o celticminstrel] by ChanServ |
| 19:45 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has quit [Ping timeout: 121 seconds] |
| 20:25 | | Reiv [NSkiwiirc@Nightstar-ih0uis.global-gateway.net.nz] has joined #code |
| 20:25 | | mode/#code [+o Reiv] by ChanServ |
| 20:26 | | Jessikat [Jessikat@Nightstar-bt5k4h.81.in-addr.arpa] has quit [Connection closed] |
| 21:17 | | macdjord|slep [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 21:17 | | mode/#code [+o macdjord|slep] by ChanServ |
| 21:20 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 21:22 | | Kindamoody|afk is now known as Kindamoody |
| 21:58 | <&McMartin> | jeroud: That is indeed quite different from my own, which exploited a larger symmetry in a more haphazard fashion... |
| 22:00 | | * [R] has no idea what's happening in that |
| 22:00 | <&McMartin> | Don't pay too much attention to it~ |
| 22:01 | <&McMartin> | jerith and I are trading our solutions for the formally hardest level in Opus Magnum |
| 22:02 | <&McMartin> | Consequently my own was sent via PM |
| 22:03 | <&jerith> | Can you PM it again to me? jeroud and I share highlights, but not PMs. |
| 22:04 | <&McMartin> | Done |
| 22:04 | <&McMartin> | This also includes an example of rotation as an alternative to pivots |
| 22:05 | <&McMartin> | The shape of the track also is abolutely necessary to prevent collisions or unwanted bonding with components for the next phase, which is kind of comical |
| 22:05 | <&McMartin> | What the heck, I'll post it, but: spoiler alert, shows a solution to a puzzle in Opus Magnum: https://www.dropbox.com/s/uqc5ve6loriozww/Opus%20Magnum%20-%20Universal%20Solven t%20-%202017-10-22-02-23-20.gif?dl=0 |
| 22:08 | <&jerith> | Which of the three axes di you beat me on? |
| 22:08 | <&jerith> | I remember being ahead of you in two. |
| 22:08 | <&jerith> | *did |
| 22:08 | <&McMartin> | Cycles, apparently. |
| 22:09 | <&McMartin> | Probably because my Wheel is doing double duty. |
| 22:09 | <&jerith> | Ah, right. |
| 22:09 | <&jerith> | Yeah, you're building three pieces in parallel. |
| 22:10 | <&jerith> | I'm building everything serially, but then I overlap part of the next cycle. |
| 22:10 | <&McMartin> | Well, s do I; if you notice, the upper right component begins construction again almost immediately after hurling its product into place for the core |
| 22:11 | <&jerith> | What I mean is that you're building three pieces independently and then synchronising the final assembly. |
| 22:12 | <&McMartin> | Right, and the final assembly is in some sense a "fourth component". |
| 22:16 | <&jerith> | I overlap less than a third of my assembly, you overlap more than half. There's no way I can compete on speed without significant redesign. |
| 22:16 | <&jerith> | But I think my solution is cleaner.~ |
| 22:17 | <&McMartin> | It totally is~ |
| 22:17 | <&McMartin> | But you were asking yesterday about how you could remove pivots from piecewise-symmetric systems. |
| 22:17 | <&McMartin> | And the answer is "by hurling it around like a pinball, as shown" |
| 22:18 | <&jerith> | Yeah. You use far more tracks and pistons than I do. |
| 22:18 | <&McMartin> | Tracks are almost never actually the optimal answer |
| 22:18 | <&McMartin> | My first cuts are full of them, but my optimized ones tend to have at most two hexes worth |
| 22:19 | <&McMartin> | (Because that's the point where it's cheaper to add another arm, or to turn an arm into a piston) |
| 22:20 | <&McMartin> | That also said: yes, I tend to start with everything as pistons just in case and then turn them into arms as I realize I didn't need the extra power |
| 22:20 | <&jerith> | Yeah. One of my chapter 1 area optimisations from this morning had a lot of complicated piston dancing on a two-hex track to keep everything within the borders. |
| 22:21 | <&jerith> | This actually shows an interesting difference in how we approach these problems. |
| 22:21 | <&McMartin> | That's why I was more interested in the first cuts than in the refined results. |
| 22:21 | <&jerith> | You're much more comfortable with translation than I am. |
| 22:22 | <&McMartin> | Yeah. To the point that incidental rotation was the thing that would trip me up the most on the way to the solution. |
| 22:22 | <&jerith> | Or rather, you're more comfortable with a richer set of motion primitives. |
| 22:22 | <&McMartin> | I blame SpaceChem, really. |
| 22:22 | <&McMartin> | The trick is that when I mess those up the solution doesn't work and no image is thus produced~ |
| 22:23 | <&jerith> | I stick to the minimum (which is why I default to fixed-length arms) unless absolutely necessary. |
| 22:23 | <&McMartin> | This is also probably why your first cuts beat my first cuts on all three metrics handily~ |
| 22:25 | <&jerith> | The primary brick in my SpaceChem wall was having to track all the global information in each puzzle. |
| 22:26 | <&jerith> | The last chapter was the only one where I really had to start decomposing my solutions into multiple parts. |
| 22:26 | <&McMartin> | Meanwhile, the guy on my flist who just wrecks everyone on speed has posted some of his |
| 22:26 | <&McMartin> | I think the Surrender Flare is the most utterly amazing one: https://imgur.com/Z33C4g2 |
| 22:27 | <&McMartin> | See also "period 4 or gtfo, punks. AT. ANY. COST." |
| 22:30 | <&jerith> | Curious Lipstick and Universal Solvent are the only two puzzles where I had to put considerable effort into the final assembly of those parts. |
| 22:33 | <&McMartin> | I had to put *more* in |
| 22:33 | <&McMartin> | But I was definitely using principles I'd built up over time to do so. |
| 22:39 | <&jerith> | Yeah. I guess for me it was more "I can see the shape of the whole thing before I start" vs "this is too big to fit in my head, so I'll focus on the bits and then figure out how to assemble them later". |
| 22:40 | <&jerith> | Which is how I ended up spending nearly half my total solve time in Curious Lipstick trying to shove a complicated shape through a narrow gap. |
| 22:41 | <&McMartin> | I think it was about a quarter for me. |
| 22:41 | <&McMartin> | Those were the only two where I was regularly in danger of bonding the final compound to the unassembled bits of the next instance. |
| 22:50 | <&McMartin> | Though my basic translation technique of "run a train past a bunch of bonders to get your weird shape" often produced unwanted internal bonds in early nonworking drafts. |
| 23:01 | | mac [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has joined #code |
| 23:01 | | mode/#code [+o mac] by ChanServ |
| 23:04 | | macdjord|slep [macdjord@Nightstar-a1fj2k.mc.videotron.ca] has quit [Ping timeout: 121 seconds] |
| 23:23 | | celmin [celticminst@Nightstar-0q5lkf.dsl.bell.ca] has joined #code |
| 23:23 | | mode/#code [+o celmin] by ChanServ |
| 23:23 | | celticminstrel [celticminst@Nightstar-qqdk24.dsl.bell.ca] has quit [NickServ (RECOVER command used by celmin)] |
| 23:23 | | celmin is now known as celticminstrel |
| 23:56 | | Vornicus [Vorn@Nightstar-1l3nul.res.rr.com] has joined #code |
| 23:56 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| --- Log closed Mon Oct 30 00:00:57 2017 |