2024-09-09 00:12:38 https://www.rowetel.com/wordpress/?page_id=452 has samples you can listen to 2024-09-09 00:13:09 the above command line is correct and does work to encode and decode at 1200 bits per second. note that the third command line argument (the second `-`) is an output filename to overwrite 2024-09-09 01:49:35 xentrac: did you try opusenc? 2024-09-09 02:47:34 nmz: no, just ffmpeg -i lento.wav -c:a libopus -b:a 4800 lento.48.opus 2024-09-09 02:47:50 it refused to go lower than 5500 bits per second and sounded terrible 2024-09-09 02:48:57 somewhat surprisingly, -c:a speex (with a .mkv output file) was the same bandwidth but sounded slightly better 2024-09-09 02:49:12 like, as good as the 2400-bps LPC-10 or the 1200-bps Codec2 2024-09-09 03:06:02 hey o/ long time no see 2024-09-09 03:06:14 ACTION has momentary reliable internet 2024-09-09 03:07:03 how is it going 2024-09-09 03:10:31 hi there! not dead yet! 2024-09-09 03:10:44 I'm glad to hear, what have you been up to : ) 2024-09-09 03:10:48 I just took apart a tower-case computer my wife and I found thrown out on the sidewalk 2024-09-09 03:11:14 hoping to give it a new life? 2024-09-09 03:11:20 it's from 02008, DDR2 RAM, Pentium dual-core E5200 CPU, so I was just stripping it for parts 2024-09-09 03:11:26 "holy shit BeOS!" 2024-09-09 03:11:32 no disks 2024-09-09 03:11:35 aawh 2024-09-09 03:11:39 that's cool 2024-09-09 03:11:56 I'm hoping the 65-watt CPU heatsink will be useful for some experiments in desiccant air conditioning 2024-09-09 03:12:19 like calcium chloride type things? 2024-09-09 03:12:35 what are you trying to do? 2024-09-09 03:12:37 yeah 2024-09-09 03:12:58 to my surprise people are selling the same motherboard for US$45 on Mercado Libre, but I already threw it out (and likely damaged it in the process) 2024-09-09 03:13:10 we found two of the machines, so maybe we'll plug the other one in and see if it runs 2024-09-09 03:13:36 the most basic thing I want to do is keep cool in the summer 2024-09-09 03:14:19 calcium chloride or some variant like tachyhydrite seems like a promising route 2024-09-09 03:14:44 I think thermal energy storage is likely to be a big deal due to the renewable energy transition 2024-09-09 03:14:49 we used it for a few years in the summer, but it ended up being more trouble than it was worth 2024-09-09 03:15:00 air conditioning, or calcium chloride? 2024-09-09 03:15:09 calcium chloride 2024-09-09 03:15:17 interesting, what was your setup? 2024-09-09 03:15:23 but being in a house you'll have more options 2024-09-09 03:15:52 we'd funnel the air coming from the bow through a mesh with calcium chrloride in to capture the moisture and heat 2024-09-09 03:16:04 I also have maybe more urgency; I have some kind of thermal intolerance issue which makes me get really stupid when it gets hot 2024-09-09 03:16:05 we have these dorads(?) things that bring the air in 2024-09-09 03:16:26 the residue sludge is too much hassle for us to handle 2024-09-09 03:16:35 it's quite caustic it turns out 2024-09-09 03:16:49 I spilled it a bit and it devarnished the panel onto which it fell 2024-09-09 03:17:02 we try not to carry lye, and that sort of stuff if we can help it 2024-09-09 03:17:10 yeah, it isn't normally caustic 2024-09-09 03:17:31 actually one of the problems desiccant-cycle air conditioners have is that it can break down over time to produce hydrochloric acid 2024-09-09 03:17:59 I read one paper which mixed a little bit of metallic magnesium in with their calcium-chloride brine to keep that from happening 2024-09-09 03:18:23 I have a friend who's figured a pretty clever trick to survive tokyo heat waves 2024-09-09 03:18:27 maybe it can help you as well 2024-09-09 03:18:28 but it wouldn't be terribly surprising to buy calcium-chloride desiccant with a little calcium hydroxide in it 2024-09-09 03:18:44 and also, even if it's not caustic, it's still very corrosive to metals 2024-09-09 03:18:55 just because it's a chloride 2024-09-09 03:18:57 yeah this is tricky, but in a house, it doesn't matter much 2024-09-09 03:19:14 it can fal on concrete or watever and it won't sink you 2024-09-09 03:19:26 right! 2024-09-09 03:19:29 how did you regenerate it? blow hot air through the mesh? 2024-09-09 03:19:39 yup 2024-09-09 03:19:47 simple as that 2024-09-09 03:20:06 that was a few years ago now 2024-09-09 03:20:07 so did you like move the mesh into a separate chamber to blow the regeneration air through it? or were you blowing hot air out your bow? 2024-09-09 03:20:33 the bow brought the air in, passed through the mesh and kept the salloon cool 2024-09-09 03:20:50 was it a wood panel that you spilled it on? 2024-09-09 03:20:55 we're replace the dessicant form time to time 2024-09-09 03:20:56 yeah 2024-09-09 03:21:04 we'd* 2024-09-09 03:21:32 I have to make dinner now, but I'll be back soon, I can tell you a bit more then :) I'll be able to come around near everyday now 2024-09-09 03:21:39 okay! 2024-09-09 03:21:56 I'm delighted to have the chance to hear about practical experience with desiccant cooling 2024-09-09 03:22:44 (and not surprised to hear you didn't find it worthwhile) 2024-09-09 03:23:08 I think I'll go make some dinner myself 2024-09-09 03:55:43 that was tasty. egg salad with homemade mayonnaise, white pepper, julienned cabbage, and potassium chloride 2024-09-09 04:26:51 back 2024-09-09 04:27:48 handmade pasta with sumac zataar 2024-09-09 04:31:57 sounds wonderful 2024-09-09 04:32:32 unfortunately I have to go! but I'd love to know more about how your setup worked, how efficient it was, how you powered it, what the sludge was, etc. 2024-09-09 04:33:09 that's alright 2024-09-09 04:33:15 I'll be around tomorrow all we can talk about it 2024-09-09 04:33:23 I might have pictures around which I can dig out : ) 2024-09-09 04:33:26 have a good one o/ 2024-09-09 13:28:24 awesome! 2024-09-09 14:58:32 You know, I was taught the "law of sines" (triangle with sides A, B, C and opposite angles a, b, c satisfies A/sin(a) = B/sin(b) = C/sin(c) pretty early on. 2024-09-09 14:58:58 But no one ever bothered to teach me that the value all three of those are equal to is the diameter of a circle that has all three triangle points on it. 2024-09-09 14:59:03 That's pretty damn useful. 2024-09-09 15:23:46 TIL 2024-09-09 15:24:36 indeed, threaded interpreted languages. 2024-09-09 15:32:33 I never thought of that either, KipIngram. although I can't swear nobody ever taught it to me; I didn't do very well on the exam for that part of my geometry class because I slept through most of it 2024-09-09 15:32:57 (I think it started at 7:30 AM or something like that) 2024-09-09 15:35:06 it makes sense: the angle subtended by a chord BC is twice the angle BAC between two chords from its ends to a common point on the other end of the circle, including in the special case where that forms an isosceles triangle, at which point a line through A and the center of the circle is a perpendicular bisector of BC 2024-09-09 15:36:28 which divides BC into two equal segments, each of which forms a right triangle with part of that radius, whose angle at the center of the circle is also BAC 2024-09-09 15:37:45 I mean, it's the same size as angle BAC, because two of them added together make the angle subtended by the chord BC from the point of view of the center 2024-09-09 15:38:45 so each of those two segments has length r sin BAC, and their sum has length d sin BAC 2024-09-09 15:39:37 so by dividing the length BC by the sine of the angle BAC, you get the diameter d 2024-09-09 15:40:15 Yes, it's not mysterious once it's pointed out. Just seems like something someone would have pointed out, if they were taking the time to teach it at all. 2024-09-09 15:40:47 actually you know what? I'm sure they didn't teach me that in that geometry class 2024-09-09 15:41:02 I sometimes like to solve these little "math puzzle problems" that show up on YouTube. I'll try to just look at the thumbnail and solve it, and then I'll peek at the end of the video to check my answer. 2024-09-09 15:41:10 because it was entirely devoid of trigonometry, which was postponed until algebra II the following year 2024-09-09 15:41:11 That law of sines figures in those fairly often. 2024-09-09 15:41:25 Just this morning I did one using Descarte's "four kissing circles" theorem. 2024-09-09 15:41:31 cool! 2024-09-09 15:41:43 not familiar with it 2024-09-09 15:41:56 Yeah I liked it. The problem didn't actually have four circles - it had three that were all tangent to the x axis. 2024-09-09 15:42:05 haha 2024-09-09 15:42:08 But I figured I could regard the x axis as a circle of infinite radius. 2024-09-09 15:42:09 an infinite circle 2024-09-09 15:42:15 yes, in projective geometry 2024-09-09 15:42:36 What shows up in the theorem is the curvature, so instead of infinity I had zeroes for certain terms. 2024-09-09 15:42:42 Worked out real nice. 2024-09-09 15:43:00 that's the usual case, yeah 2024-09-09 15:43:32 you say it's not mysterious, but it still took me about eight minutes to figure it out apparently 2024-09-09 15:43:57 My favorite ones are the ones where some part of the problem isn't specified - a free parameter so to speak. The problem is usually presented (drawn) with that parameter in some oddball state, but if you look at it you can see some other case that makes the problem a lot easier to solve. 2024-09-09 15:44:14 When they don't specify a parameter and yet still expect an answer, the answer must not depend on that parameter. 2024-09-09 15:44:35 well, unless you don't have sufficient information to solve the problem ;) 2024-09-09 15:45:04 Yeah, in the real world if a problem like that "just came up," you probably wouldn't be able to know for sure that the solution was independent of that parameter. 2024-09-09 15:45:31 But when it's asked in a formal quiz environment like that, and if you assume the folks posing the problem knew what they were doing, then you can assume that. 2024-09-09 15:45:59 I had one the other day where there were *two* special cases that were easy to solve, and I was able to confirm the answer was the same in both of those. 2024-09-09 15:46:51 If I recall, it was a square of side length 3 with one of its corners at the center of a square of side length 2. And then they were at some arbitrary angle, and they wanted the area of the intersection. 2024-09-09 15:46:55 Well, you could line the squares up just right and that's an easy case. 2024-09-09 15:47:06 Or you can put them at 45 degrees, and that's easy to solve. 2024-09-09 15:47:25 Both ways you get 1/4 of the 2-square. 2024-09-09 15:47:45 And in that case you can even argue from symmetry considerations that the answer has to be constant. 2024-09-09 15:47:54 So that one was really fun. 2024-09-09 16:11:12 Why couldn't the area go up a little bit as you rotate from 45° toward 90°, then going back down to 1 (2² · ¼) as you reach 90°? Or down a little bit? 2024-09-09 16:11:40 Oh, I guess the wedge you're adding on one side is exactly congruent to the wedge you're removing on the other. 2024-09-09 16:12:48 That's a case where solving the more specific problem of a specific angle of rotation is more difficult than solving the general problem 2024-09-09 17:09:48 Right - that's exactly why I like such problems. I feel like I've "gotten away with something." 2024-09-09 17:10:23 Like I said, if you were just solving some problem in the real world you'd not really have any reason to assume that independence, unless you could prove it. 2024-09-09 17:10:23 When it's an "exam question" then you either can assume it or it's a bad question. 2024-09-09 17:22:06 The other thing you can *usually* count on in such problems is that numerical answers will turn out to be integers. Usually when mine doesn't I've made an arithmetic error somewhere. 2024-09-09 17:24:11 One fun one a few days ago had a rectangle of unknown dimension with a triangle inscribed in it. One corner of the triangle was in a corner of the rectangle, but the other two triangle corners were in arbitrary positions on the rectangle sides that didn't touch that shared corner. You were given the area of the three "corner triangles" that creates and asked for the area of the central triangle. 2024-09-09 17:24:25 That one had to be chosen just right to make the centrral area be an integer, but it was. 2024-09-09 17:24:30 heh 2024-09-09 17:25:06 Well, I guess that's not strictly true. If the rectangle had integer area and the three corner triangles did to, then the remaining central triangle would. 2024-09-09 17:26:22 The algebra on that one was nice - you start by defining a bunch of "intermediate variables." The height, the width, and the places on the sides where the triangle touches. Four variables. But then all four of those fall out of the algebra and you can get a quadratic in that central area. 2024-09-09 17:26:27 With one solution positive and the other negative, and you know to take the positive one.