00:00:00 --- log: started forth/16.06.20 00:06:24 --- join: nighty (~nighty@d246113.ppp.asahi-net.or.jp) joined #forth 00:06:44 --- join: wa5qjh (~Thunderbi@203.111.224.46) joined #forth 00:34:21 --- quit: nal (Quit: WeeChat 1.4) 00:51:54 --- quit: wa5qjh (Remote host closed the connection) 01:19:30 --- join: true-grue (~true-grue@176.14.216.104) joined #forth 01:29:52 --- quit: joneshf-laptop (Ping timeout: 240 seconds) 01:57:43 --- quit: karswell (Remote host closed the connection) 01:58:56 --- join: karswell (~user@31.185.131.182) joined #forth 02:07:55 --- quit: mnemnion (Remote host closed the connection) 02:46:17 --- quit: ASau (Ping timeout: 244 seconds) 02:49:30 --- quit: nighty (Quit: Disappears in a puff of smoke) 04:12:31 --- join: nighty (~nighty@s229123.ppp.asahi-net.or.jp) joined #forth 04:36:00 --- join: joneshf-laptop (~joneshf@c-73-220-86-28.hsd1.ca.comcast.net) joined #forth 05:51:19 --- join: proteusguy (~proteusgu@183.88.69.94) joined #forth 05:51:19 --- mode: ChanServ set +v proteusguy 05:55:16 --- join: timeslice (~photon@1.132.96.3) joined #forth 06:07:53 --- join: guna (~user@h-201-187.a328.priv.bahnhof.se) joined #forth 06:20:25 where is mat4? 06:20:42 i havnt saw him for serveral days 07:05:04 --- part: timeslice left #forth 07:27:14 dys: Impressive stuff! 07:27:18 --- nick: backer_ -> backer 07:28:23 I don't understand the vitriol against forth here. 07:28:59 But I also don't understand the cultish/religious ferver over it, either. It's just a tool - good for some uses, not for others. 07:29:32 backer: it's been too quiet to even hear any "vitriol against forth" 07:29:34 :-D 07:29:44 backer: basically, Forth sucks 07:29:54 backer: the problem is that everything else sucks worse 07:31:02 backer: you want to bring up examples where it is good. ;) 07:37:30 at least Forth doesn't abstract away the underlying machine so much 07:37:32 not like C 07:38:11 Oh, really? 07:38:16 What a bullshit. 07:39:17 Tell us then, where is the second machine-assisted stack on i386? 07:39:42 What constructs does Forth provide to access registers? 07:44:36 * yunfan maybe backer was running his app on GA144 :D 07:45:27 yunfan: that would be funny argument. 07:46:18 So, if I'd run my code on Java CPU, I'd have all rights to claim that Java is over all, is it like that? 07:46:25 i just dont know why cant they bring some board like rasperberrypi 07:46:36 the single board need $450 07:46:40 Or, if I dig Smalltalk CPU out of oblivion, that would be Smalltalk, right? 07:46:41 its too expensive 07:46:47 DGASAU: What I don't understand is why people like yourself engage in the religious warfare over programming languages. Do you feel that you are contributing something to the world by trying to prove to newcomers that using Forth is a waste of time? 07:47:34 backer: yes, I have convinced some people not to waste their time on Forth. After some consideration they agreed. 07:50:53 DGASAU: So you basically hover around the #forth channel, day in and day out, waiting for opportunities to impart this wisdom? 07:51:06 backer: you're beginning to catch on 07:51:09 welcome :-) 07:52:15 backer: you overstate time consumption. 07:52:39 any time is wasted on using Forth, but no time fighting it is wasted. 07:52:44 yes, we understand. 07:52:57 backer: Besides, I don't need to look for arguments against Forth, they are all pretty clear. 07:53:01 dzho: He sounds like a guy who packs a lot of "knowledge" but isn't making much use of his energy. Or maybe he doesn't have that much to give, anyway. 07:53:28 backer: I have written it off as a quesiton of personal constitution. 07:53:43 there's a word on the tip of my tongue to describe it, but I can't quite get it. 07:53:51 So, I'll settle for an image: Eeyore 07:54:12 DGASAU: If that's the case, why waste your time trying to dissuade others from using Forth? 07:55:43 dzho: At least Eeyore is lovable. 07:55:59 backer: I do it only as a diversion from the useful work. 07:56:23 backer: so is DGASAU 07:56:27 Contrary to what is claimed, I'm not here all around o'clock. 07:56:45 Besides, you can see the quality of pro-Forth arguments here: 07:56:57 backer: with the right approach, you can see how cute his little hobby is 07:57:18 whenever I point to fundamental problems in Forth, others start the usual religious bullshit or name calling. 07:57:57 because using terms like "religious bullshit" isn't tantamount to name calling. 07:58:00 Gotcha. 07:58:12 I'll just be over here in my robe and wizard hat 07:58:13 backer: the problem is that everything else sucks worse 07:58:26 all software sucks 07:58:42 Yes, yet some software does suck worse. 07:58:53 DGASAU: They are probably responding more to your own religious bullshit and vitriol than responding in defense of the language. 07:58:56 on a context dependent basis, yes 07:58:56 It is possible to find examples that compete with Forth. 07:59:11 Yet in modern world those are pretty rare. 07:59:11 I personnaly prefer brainf*ck 08:00:04 I've used Forth enough to know there are some problems in using it. It doesn't mean I have to take a religious stance on it and make that part of my identity. 08:00:09 backer: I'll save us some time, here: DGASAU likes lua better than forth, so if you want to curry favor you can extol its virtues here, even though this isn't #lua 08:00:38 dzho: I agree that all software sucks! :-) 08:02:35 --- quit: proteusguy (Ping timeout: 258 seconds) 08:05:17 < DGASAU> backer: I do it only as a diversion from the useful work. 08:06:19 dzho: never really saw the point of lua 08:08:22 --- join: proteusguy (~proteusgu@183.88.78.90) joined #forth 08:08:22 --- mode: ChanServ set +v proteusguy 08:08:31 it's like Python but with crazy syntax, isn't it? 08:08:54 Ha-ha-ha! 08:09:40 gordonjcp: lua is simple and embeddable. That's what makes it appealing, generally. In lua, everything is a table. 08:10:06 For the record, I'll say why I am attracted to Forth. In my small, closed embedded projects, I like to have absolute control over the source of every instruction. Using Forth allows me to have that without reliance on gcc, binutils, etc. 08:10:52 And the latter is bad because?.. 08:11:38 See, I can use it without religion for or against it. Just because it works for me, in that regard (complete control), doesn't mean I'm stuck with Forth. 08:11:39 backer: similar to my use of forth. Embedded stuff where I enjoy having a simple toolchain and I like having an interactive environment. 08:13:23 DGASAU: It's not bad. But it separates one from the actual emitted code. It's not common to find a real bad bug in GCC, but I have had a few instances in the past where dealing with the complication of GCC made things more difficult. In one case, it was a silicon bug that needed an extensive workaround. Making it work in GCC was a truly hair-pulling experience. 08:13:25 I use the GNU C toolchain for some projects, but there is an appeal to owning and understanding the entire system. 08:13:25 Gracana: okay, so what would it be useful for? 08:14:17 backer: it is even less uncommon to find a real bug in your own code generator, unless the latter is very very simplistic. 08:15:35 A channel such as this, in my opinion, should be a haven for people who are choosing to use the language for whatever purpose they desire. You are begin with the assumption that any use of Forth is misguided, and then you proceed for force your views on others without understanding where they are coming from. 08:15:52 If you are going to be that arrogant, make it constructive. 08:15:55 gordonjcp: a very common use case for lua is as a scripting system for games. Game is written in C++ or somesuch, lua is easily embedded in the engine and used to implement all the fast-changing/often-tweaked level/character/vehicle/event design and coordination stuff 08:16:32 that's what I meant by "embedded," sorry if that was unclear. 08:17:02 (different from my meaning of "embedded" when talking about forth, hah :X) 08:17:17 (And I think it's fine for someone to be haughty, provided that they are using their rare talent to advance the field.) 08:17:42 backer: that's not an assumption, it is the result of analys of a number of use cases. 08:18:22 backer: even in your case, if you think over the reasons why you use Forth, you'll find out most likely that it is for two reasons, most likely. 08:18:46 backer: of which only one is more or less valid, the other one is you being "control freak." 08:19:34 backer, what sort of embedded projects are you working on? 08:21:57 Gracana: okay, so lua is good for very large systems with essentially infinite RAM and storage and incredibly fast processors 08:22:19 Pf! 08:22:40 Lua is faster than Forth while operating with similar memory foot-print. 08:23:50 DGASAU: Well, the control freak thing is not entirely false. But again, you are mistaking prejudice with insight. Another reason is that I want it to be extensible without need for cross-compiling. The target has 64-128kB SRAM. 08:24:39 "Extensible without need for cross-compiling" doesn't entail Forth at all. 08:25:38 Gracana: digital audio playback and recording with minimal processing 08:26:21 Oh, huh. What sort of hardware are you using? 08:26:26 DGASAU: cool! So how do I get started porting Lua to my target system? I'm really keen to replace Forth with Lua here 08:26:55 gordonjcp: it may already be ported 08:27:15 DGASAU: What alternatives do you suggest? Target is 128-192kB SRAM, max. 08:27:16 ??check out elua 08:27:45 Gracana: pretty sure it's not 08:28:07 128 kB is more than enough to run Pascal IDE. 08:28:24 Just for your information. 08:28:26 Gracana: Right now, PIC32MX series, using SPI as I2S interface. 08:28:43 DGASAU: great 08:29:07 DGASAU: Pascal?! Interpreted or compiled? And supporting inline assembly? 08:29:08 DGASAU: my target system is an ATtiny 45, it's got 256b of RAM and 4k of flash 08:29:16 backer: both. 08:29:20 It looks like you've got virtually no experience with CPUs that are considered embedded now. 08:30:00 Oh how do you like the PIC32MX? I like the idea of it (nice 32 bit mips microcontroller), but boy the errata for those devices is enormous. 08:30:25 gordonjcp: that is probably not enough for elua 08:30:49 almost certainly 08:30:51 DGASAU: Please, enlighted me. Also, can you point me to an implementation of Pascal? 08:31:28 Turbo Pascal 3 did exactly that. 08:31:48 Gracana: The MZ series has been plagued with a lot of bugs, yes. The MX series has had a lot of kinks ironed out. 08:32:17 gordonjcp: and Forth fits those restrictions because?.. 08:32:43 DGASAU: I'm not sure what you're asking me. 08:32:51 Yes, Forth fits on that system. 08:33:08 Ah that's gotta be it backer, I've mostly looked at the MZ stuff. 08:33:10 Sure, you can cram Forth into 4K, but that's pretty tight, and you could do that a lot better with saner language. 08:33:12 obviously I cross-assemble it 08:33:20 DGASAU: like what? 08:33:42 Anything that has native code optimizer. 08:34:13 08:34:28 DGASAU: Turbo Pascal? And so, my case in point. You have nothing constructive to offer and no practical examples. 08:34:30 so I'd still have to write everything in assembler 08:34:52 I don't see what advantage there would be in that 08:34:59 C is just a macro assembler after all 08:35:21 backer: my point is that Turbo Pascal 3 did that like 30 years ago, so you can definitely find something saner than Forth. 08:35:29 Or you can implement it yourself. 08:35:36 I wish. Modern C compilers do a hell of a lot more than that. 08:35:37 DGASAU: And why is Turbo Pascal saner than Forth? 08:35:51 DGASAU: or Lua, for that matter 08:36:33 Gracana: I don't know what's going on with MZ, but I don't need those capabilities quite yet. I imagine I will jump ship to ARM down the road... 08:36:36 Because it has consistent support for infix notation, access to variable length data, implements subroutine calls, offers integration with well-known code generation algorithms and a lot more. 08:37:14 Essentially, 40 years of knowledge from around the world. 08:37:50 backer: that's what I did. I'm not a huge fan of ARM, but every time I stray away, I come back to ARM with renewed appreciation for what that architecture/ecosystem offers. 08:39:31 DGASAU: subroutine calls as in library calls? 08:39:40 Atmel stuff in particular. ST is popular but their docs aren't the greatest and their hardware abstraction library will make you cry. It's just that ugly. 08:39:49 backer: inlining is possible too. 08:40:26 Besides, sane implementation of local varibles matters too. 08:40:44 DGASAU: inlining is not possible in Forth? 08:41:00 Like not wasting full machine word for an octet variable. 08:41:10 backer: try implementing that. ;) 08:44:07 If you do manage it, don't forget to publish. Solving halting problem will definitely bring you some prestigious prize. 08:46:32 DGASAU: So you could implement a 32-bit Pascal in less than 64kB that supports variable length data with a compiler that can pack the data, with minimal padding, while at the same time not relying on unaligned accesses for all data? 08:47:19 If you have short operation codes, sure, you can. 08:47:44 So why is this not possible in Forth? 08:47:52 Try it. 08:48:09 Since you already know, enlighten me. 08:49:00 The main obstacle is very loose syntactical structure with absolutely no attributes. 08:49:45 A variable entry in the dictionary can return any address which does not need to be aligned. Why can the programmer not refer to byte or short length variables in the code dictionary? 08:50:08 When you bolt the latter to your Forth, you get something that reminds the same Pascal, only in a weird syntax and looking very messy. 08:50:25 Again, try doing that. 08:50:30 Why, because C! and C@ are used instead of ! and @? 08:50:48 Try doing that for local variables. 08:51:00 Try implementing indexing with all that. 08:51:09 I bet that in your Pascal implementation (try doing that), you'll just end up padding up to cell length when the next variable is of a different length. 08:51:23 Not at all. 08:51:49 How will you avoid that? 08:52:13 (I can think of a way, I just want to see what you're thinking. :-) 08:52:13 In Pascal, I am allowed to reorder local variables any way I like _and_, which is important, I have syntax and semantics helping me arranging that. 08:53:26 It is possible to do that in Forth, only in a very very messy way. 08:53:54 Ok, so you'll be able to do this while maintaining the compiler and full tools on the target, including a parser, on a 32-bit, 128kB target, without using an interpreter or tokens and compiling to native code? 08:54:00 Effectively, implementing half-assed Pascal in Forth syntax _and_ monomorphized. 08:54:32 Yes, if you have octet-sized operation codes, it is pretty much possible. 08:54:40 Turbo Pascal did exactly that 30 years ago. 08:54:53 On an 8-bit target. 08:55:10 It fit into 32K also. 08:55:17 Perhaps even less. 08:58:44 Maybe you could consider writing such an IDE for modern microcontrollers. I would like to see a modern example of what you're talking about (though I am partial to C). 08:59:31 Do you really think that implementing proper parser for unambiguous context-free grammar takes a lot of space? 09:00:39 C is worse because its grammar is ambiguous. 09:01:14 Gracana: re: HAL, I have completely circumvented using any of the vendor's tools with the exception of using the PIC USB HAL. 09:01:18 Or even, perhaps, (mildly) context-sensitive. 09:02:26 backer: even if you're context freak, learn classical compiler technology. 09:02:35 Perhaps, you'll learn a lot useful stuff. 09:02:41 DGASAU: I haven't done it, but no, I don't think it takes a lot of space. But I don't see how it would help my needs in my particular situation. 09:03:00 You'll get a lot more maintainable code. 09:03:12 BTW, have you ever seen Small-C? 09:03:35 I don't remember if it has MIPS or ARM targets, but it is pretty small to understand. 09:05:25 DGASAU: Yes. I have used it, including a port to MIPS. And I decided that pulling together a hodge-podge of crippled tools is far less elegant than taking the time to write a subroutine-threaded, inlining forth of my own. 09:06:25 If you know how to implement code generator, you can as well learn implementing proper CFG parser. 09:06:26 DGASAU: Have you seen RetroBSD? It has taken a whole team of developers many months to pull together this said group of tools. I don't know if they ever really got it working, either. I gave up on it. 09:06:44 I'm not interested in 2BSD. 09:07:16 DGASAU: I'm talking about Small-C and associated tools to make an IDE, not the BSD aspect. 09:07:37 If the system has MMU, then it is better to implement pmap and use NetBSD. 09:07:54 You are a master of employing the non-sequitur! 09:09:07 I consider striving to carry on-board development tools a waste of time. 09:09:08 I would be interested in revisiting some form of a bare-bones Small-C or TinyCC-based system. But that would only further delay the embedded work I am doing. 09:10:02 In recent decade I haven't met any case where this would pay off. 09:10:16 If you count small systems. 09:11:44 Oh, so our entire discussion is pointless. 09:12:28 Even 68K Palm (Palm III? Or was Palm II the last one?) was powerful enough to run on-board C and, IIRC, even Java IDEs. 09:13:37 Yes, it is mostly pointless, because you haven't convinced me that you do need on-board tools there even, let alone Forth. 09:14:22 DGASAU: is that why you're here, to be convinced? I thought it was to convince other people. 09:14:24 In any case, if you have 32K of memory _and_ short operation codes, you are able to implement IDE for conventional programming language. 09:15:00 "I consider" <-- this at least acknowledges that it is a question of opinion, of preference, of taste. 09:15:57 dzho: for start, he poses problem in terms of, roughly, "I'm control freak, hence I want on-board tools at any price." 09:16:37 you know, I'm pretty sure your involvement in this topic isn't dictated by his preferences or how he states the problem. 09:16:55 you're consistently on the side of "don't use forth" regardless of how the question is posed. 09:17:15 it's hard to take your participation as being in good faith 09:17:23 Yes, even in the way he poses the question, Forth is suboptimal solution. 09:17:38 it doesn't matter what he says, though: that would be your conclusion. 09:17:49 because it fits your preferences 09:17:49 DGASAU: In all of your opinions here, I have yet to hear any sound argument for not using Forth. 09:18:33 but instead of just settling with "these are my preferences" you seek to convert your preferences to some hard and immutable truth. 09:18:46 perhaps you aren't so convinced as you say you are. 09:18:53 DGASAU: Maybe I don't absolutely need on-target development. I do need it to be extensible for the end user, and your questioning of that requirement is only a distraction if you can't clearly show why it's wrong to use Forth in this case. 09:19:13 * DGASAU sighs. 09:19:16 backer: once again. 09:19:29 DGASAU: It's like you can't clearly state why it is incorrect to use it, so you look to non-sequiturs and distractions. 09:19:32 Postfix-only syntax is very strong reason not to use Forth.. 09:19:40 again, a preference 09:19:51 dzho: exactly 09:19:52 Monomorphic semantics is, again, very strong reason not to use Forth. 09:20:06 DGASAU: why can you not accept that your preferences are your preferences, and leave it at that? 09:20:10 Open stack is very strong reason not to use Forth. 09:20:13 DGASAU: Is it? I've written both postfix and infix syntax. I have no problems with either. 09:20:24 backer: yes, it is. 09:20:38 duck season, rabbit season 09:20:41 we can do this all day 09:20:45 DGASAU: Again, you're just rehashing the same opinions and are not making a clear argument. 09:20:57 backer: these are not opinions. 09:20:59 These are facts. 09:21:02 hehehee 09:21:11 "lugubrious" is the word I was looking for. 09:21:26 They are admitted even by your fucking Forth lovers like Anton Ertl and Bernd Paysan. ;) 09:21:35 ooooh 09:21:51 Oh, right, stack-based VM is another reason not to use Forth. 09:21:52 dzho: It's getting to him. ;-) 09:21:58 See their paper. 09:22:06 and yet, they love forth 09:22:12 how does one account for that? 09:22:26 these are opinions, and preferences 09:22:26 Yes, because they have invested a lot of their time into it. ;) 09:22:30 I've read their papers. They're not saying that it's a matter of fact that infix notation is superior to postfix. 09:22:36 Again, these are not opinions and not preferences. 09:22:51 because how one chooses how to spend time, how one enjoys any given process, is a personal preference. 09:22:58 Those "opinions" are reasons why VM for Lua 5 is not stack-based. 09:23:20 Read corresponding report for better understanding why. ;) 09:23:28 where is the falsifiability with regard to whether someone wants an on-board environment? 09:23:35 cite? 09:24:02 people are not the same, and will choose different tools. 09:24:24 this is not so hard a concept for most people to grasp. 09:24:48 That's argument that is strong anti-Sapir-Wharf. :) 09:25:20 even noting that someone has invested time in a particular tool set doesn't invalidate this as a motive for making a certain choice, unless one has a time machine or a fountain of youth 09:25:29 DGASAU: Sure, it makes sense to use a register machine model when running on an actual register machine. Android uses Dalvik. So, succinctly stated, what "facts" do you derive from that? 09:25:50 dzho: that's argument of "I've done it for so long that I don't want to learn anything new." 09:25:56 so? 09:26:04 That fits profile of Ertl and Paysan who are quite old now. 09:26:09 so? 09:26:21 It doesn't mean that their preference is optimal. 09:26:25 will your campaign remove the fact that forth ever existed? no. 09:26:39 In fact, since the whole world has moved away from that, you have strong evidence that their way is suboptimal. 09:26:57 DGASAU: are you familiar with the term "Pareto"? 09:27:12 Yes. 09:27:25 Forth is on the opposite branch though. 09:27:55 how about the term "Gish gallop" 09:28:26 this discussion is a victory for DGASAU because, whatever its other merits, the time someone spends on it is time they don't spend writing forth. 09:28:46 Do you have explanation for reasons why almost the whole embedded world turned away from Forth in mid-80s? ;) 09:29:06 dzho: That's the perfect term for this situation. :-) 09:29:31 DGASAU: if almost the whole embedded world turned away from it, what is your point here? 09:29:42 kill the last Jedi 09:30:07 excuse me, I'm going to go work on my light saber :-) 09:30:12 DGASAU: probably the same reason they turned away from interpreted languages like BASIC, and Lua 09:30:35 DGASAU: and even crappy compiled languages like Pascal 09:30:49 Are you serious about Lua? 09:30:49 but hey, stick to your toy languages if you want 09:30:58 Somehow that's not what you can observe. 09:31:10 okay, so what's *one* serious application of Lua? 09:31:37 Khm! 09:31:44 What is one serious application of Forth? 09:31:49 lift controllersa 09:31:52 -a 09:32:11 you'd be amazed how many of them run something remarkably like Forth 09:32:17 Which one? The one you did? 09:32:29 the ones I repair, certainly 09:32:53 these days they tend to be based around a hoofing great IO board and an embedded PC 09:33:03 so of course they spend most of their time not working properly 09:33:24 masses of SCADA stuff in the oil industry runs Forth 09:33:38 anyway I asked you first 09:34:24 Are those systems modern or are we talking about holdouts from early eighties? 09:35:49 early 2010s, but bear in mind they have to interoperate with older systems 09:36:23 and they can't just be replaced with stuff crufted together in the hipster language of the week 09:36:55 DGASAU: So if some language or technology falls out of favor, that is proof that it is technically inferior? 09:36:58 So, essentially, a lock-in situation. 09:37:13 Following that argument, MUMPS is pretty nice language that has its use. 09:38:06 backer: you need to push really hard in order for technology to stay, if it doesn't provide benefits. 09:38:26 DGASAU: You avoided answering the question. 09:38:42 Yes, it is pretty strong evidence that it is inferior. 09:39:13 DGASAU: you still haven't answered the question 09:39:20 DGASAU: can you name any serious application of Lua? 09:39:20 I did. 09:39:29 Snort, nginx. 09:39:40 okay, and a serious application? 09:39:49 nginx is pretty serious. 09:39:54 a toy webserver? 09:39:57 Toy? 09:40:10 Write any competing "toy". ;) 09:40:26 why bother? 09:40:31 apache already exists 09:40:34 To prove that it isn't a toy. 09:40:53 Why nginx pushed Apache aside then? ;) 09:41:05 has it? 09:41:14 Yes, it has done exactly that. 09:41:27 extraordinary claims require extraordinary proof 09:41:37 Nginx is written in C. 09:41:55 Nearly everywhere where there is high throughput demand, nginx pushed Apache aside. 09:42:17 backer: and it uses Lua for configurable parts. 09:42:40 is that why it's got such fucked-up flakey configuration syntax? 09:42:58 Oh, really? 09:43:07 Why don't you implement something better and offer for review? 09:43:16 because it's not something that interests me 09:43:19 apache already exists 09:43:25 If you provide better solutions, sysadmins will be grateful. 09:43:31 Apache doesn't sustain the load. 09:43:47 so you claim 09:43:54 That's why nginx pushed Apache aside a decade ago. 09:44:01 but you also claim that Pascal is a serious useful language 09:44:08 so I'm not inclined to take your word for it 09:44:09 gordonjcp: that's not my claim. 09:44:12 That's fact. 09:44:21 sure 09:44:26 whatever 09:44:31 You just lack knowledge. 09:44:42 Which demonstrates my point exactly: 09:44:47 your toy webserver written in Visual BASIC or whatever is the fastest thing out there, yup 09:44:51 Forth lovers just lack general knowledge. 09:45:09 yeah, whatever 09:45:29 away and play with your minecraft webserver or whatever hell it is 09:45:32 This "toy webserver" serves high throughput configurations. 09:45:38 no-one cares 09:45:53 Oh, srsly? 09:45:58 Ha-ha-ha! 09:46:20 Try come out to some sysadmin channel and try to tell these news there. :) 09:46:32 DGASAU: couldn't give a toss about sysadmin channels 09:46:38 DGASAU: I have a social life 09:47:13 Sure, it is better not to come where you will be laughed at and called moron. :) 09:47:19 one that doesn't involve fedora-wearing shitwizards drivelling on endlessly about their web-over-iscsi varnish puppet configuration 09:47:52 DGASAU: Are you on IRC mainly to talk in the #forth channel? I see that's the only channel in your list. 09:48:38 backer: no idea what you see, I see a number of other channels. 09:59:20 http://blogs.bu.edu/pbokulic/2013/11/18/gish-gallop-fallacy-of-the-day/ 09:59:34 "The term “Gish Gallop” was coined by Eugenie Scott of the National Center for Science Education. The phrase refers to a debate tactic that was a favorite of Duane Gish, a young-Earth creationist who was also a highly skilled debater." 09:59:49 "The Gish Gallop is the tactic of snowing your opponent under a mountain of supposed “pieces of evidence” or “problem cases” and claiming that the opponent’s inability to respond to this pile of evidence shows that your side is right. This tactic counts as a fallacy because its effectiveness doesn’t depend on presenting arguments that are right or even well-supported. Quantity is offered as a sub 09:59:55 stitute for quality." 10:01:53 :D 10:02:16 5 is "a mountain" for Forth lovers. :D 10:02:46 except, see, this is the fundamental asymmetry you face here, DGASAU: 10:02:58 No one gives a shit if we convince you to use forth. That's not why we're here. 10:03:08 You, on the other hand, seek to convince us *not* to use it. 10:03:44 And yet he regards anyone who argues with him a religious zealot! 10:03:50 It wouldn't surprise me to learn that you have some sort of financial interest in Forth's success. It's pretty clear that the worst thing that can happen to anything in the social realm is apathy. 10:04:07 Yet, here you are, providing ample and recurring opportunity for people to show anything but apathy. 10:04:08 dzho: here you assume that I'm going to convince you personally. 10:04:10 No, I'm not. 10:04:24 no, you seek to convince #forth, which is even more pathetic. 10:04:34 No, I don't. 10:04:50 There's no convincing him, dzho. He's basically a deeply repressed robot. If he was doing anything of significance, he wouldn't be here on a daily basis, arguing with us mere mortals. His best days are probably behind him, sadly. 10:04:51 oh, right then. 10:05:24 backer: "daily basis" as in "looking from time to time into mostly silent channel"? 10:05:29 let's talk about the optimality of trolling a channel instead of dutifully cranking out that sweet, efficient Turbo Pascal code. 10:05:34 Yeah, "daily basis" indeed. :D 10:05:47 backer: he's right. that's why I said "recurring". 10:06:36 backer: for daily (or at least, nominally daily) forth hate, see Skuzzy 10:06:39 DGASAU: I'm not on here often, but I scan the logs to see that you jump on every opportunity to spread your drivel. So, yeah, you're obviously checking in on a daily basis. 10:06:40 --- join: mnemnion (~mnemnion@2601:643:8103:f9d0:dd3f:eca4:498e:979a) joined #forth 10:07:12 That's exactly the quality of argument here: calling facts "a mere opinion", "checking silent channel" a "daily basis" and so on. 10:07:26 s/logs to see/logs and see 10:07:27 backer: ever used modern IRC client? 10:07:48 you know, one written in Turbo Pascal? 10:07:58 In any language. 10:08:10 any language but forth amirite 10:09:03 Perhaps, if you'd look around, you would notice that modern IRC clients have indication of activity in some channel. ;) 10:10:13 * dzho notices that oftc didn't have a forth channel 10:15:50 DGASAU: Thanks. My client does have that feature. Thank you for trying to enlighen me regarding the latest IRC and programming language technology. (I'll counter your gratuitous condescension with a little sarcasm!) 10:17:05 DGASAU: The point is that, whether by interrupt or by polling, you regularly expend energy feeding your ego in the name of enlightening others. 10:17:39 this leaves the question as to whether DGASAU uses his modern, non-Forth IRC client to watch like a hawk for activity in this channel 10:18:15 DGASAU: But, in fact, you have nothing to offer to the community but your tired talking points. 10:19:26 backer: au contraire, his diligent, nay some might say obsessive, attention to this channel assures that it does not meet the apathy to which he claims, at least on the surface, it deserves. 10:19:52 DGASAU: It seems on the surface that you are knowledgeable about languages and compiler technology. But, when pressed, you aren't able to substantiate your claims with anything of constructive value. You point to Forth's lack of popularity as proof of your claims. That makes me wonder if you really possess the understanding that you claim you have, and, if you do, whether or not you can make anything of it. 10:20:12 DGASAU is one of the best friends this channel ever had (assuming of course one only wants to address Forth-existential questions instead of, say, comparative questions of usage, implementation, etc) 10:20:18 I'm not here to explain clear arguments to those who aren't able to understand them. 10:20:33 you seem confused as to your purpose here, DGASAU 10:21:06 DGASAU: Don't flatter yourself. It's software, not theoretical physics. You're not as smart as you think you are. 10:21:28 backer: you wanted to know reasons why not to use Forth, I presented them. 10:21:37 tbh theoretical physics isn't that difficult to understand 10:21:38 backer: you call them "a mere opinion" 10:21:47 That's fine with me. 10:21:48 DGASAU: so, if you don't use Forth, why are you here? 10:21:59 backer: see, it's all your fault 10:22:02 But it puts you into the same cohort as creationists and other believers. 10:22:02 you encouraged him 10:22:20 dzho: I think it's fun to expose the guy for who he is. He's not able to reflect on his shortcomings. 10:22:24 They too ask for reasons only to reject them based on their beliefs. 10:23:12 dzho: I know. :-) 10:23:16 I think we've got a ways to go before the stratigraphic record has anything to say about forth. 10:23:34 dzho: if the case plastics of the Jupiter Ace are anything to go by, yes 10:23:41 And yes, one of reasons not to use Forth is that you are not smart enough to recover meaning from postfix notation, when expression becomes complex enough. 10:23:43 heh 10:23:50 Nobody is. 10:24:10 postfix notation is how we work out arithmetic in our heads 10:24:15 DGASAU: I didn't counter that the idea that Forth has shortcomings. Having used it, I'm aware of some of them. I'm countering that you are offering anything useful or constructive, and I'm countering what you consider to be "fact." You haven't proven anything, even remotely. 10:25:08 Pf! 10:25:13 I've proven it well enough. 10:25:24 It's your problem that you have decided to reject it. 10:25:34 --- quit: dys (Ping timeout: 258 seconds) 10:25:36 See the note about creationists above. 10:26:13 DGASAU: so you're a creationist now? 10:26:17 okay then 10:26:23 each to their own, I suppose 10:29:20 gordonjcp: exploiting fallacies now? sounds like you're out of your normal arguments. ;) 10:29:26 gondonjcp: It's just a low blow because we called him out on the gish gallop thing. It refers to the tactics of a well-known creationist, so he's trying to turn it around on me. Without success. 10:30:27 DGASAU: Since you have spent many hours arguing that any use of Forth is futile, I suggest you write a paper on it and present a complete argument. I think it would be very interesting. 10:31:00 DGASAU: Keep in mind here, you are the one who has something to prove. You're the one with an axe to grind. We're not making such definitive claims, we're only asking you to support your own. 10:32:00 DGASAU: Perhaps after publishing it you can provide a link to be added to the channel topic. It would save you a lot of effort in the future. 10:32:08 That's another case of tactics that creationists like: 10:32:27 make your opponent to do heavy lifting to prove arguments that are pretty clear already. 10:32:57 Haha, turning it around again. What heavy lifting? You've already written all about it. Now you just need to put it together and put your name on it. And make it public. 10:33:28 Yes, I have written it already. 10:33:35 So go and read what is written above. 10:33:48 Again, if you've failed to understand it the first time. 10:33:51 It should actually save you a lot of time in the future. 10:34:24 No, because if you had to write it as a defensible, somewhat academic paper, you would have to defend it against minds greater than yours. You can't be the big man in the small pond, then. 10:36:00 I understood all of your points, but none of them were sufficient to prove that using Forth is a waste of time. Again, I challenge you to write a definitive argument and put your name on it. 10:36:25 That's exactly what I'm saying: you're nudging me to do it in heavy lifting style, 10:36:30 playing video games is a waste of time too 10:36:36 Why you would even be trying to prove (yes, on a daily basis - over the course of YEARS!) that using any language is a waste of time is beyond me. 10:36:37 and that's only to prove what is pretty clear already. 10:36:42 but it's fun, so why not do it anyway? 10:37:02 DGASAU: If it's pretty clear, why is it heavy lifting? Again, YOU are the one making the argument, and repeatedly so. Not me. 10:37:26 (Awaiting non-sequitur or further circular reasoning...) 10:37:35 backer: time to wave /ignore I suspect 10:37:50 --- join: real-grue (~true-grue@176.14.216.104) joined #forth 10:40:18 gordonjcp: I will, I just want to rub his nose in it first. Every time I check the backlog on this channel its filled with his robotic nonsense, so I thought it might be fun to see if I can get him to extend himself beyond the usual platitudes and provide real food for thought. 10:40:34 --- quit: true-grue (Ping timeout: 250 seconds) 10:40:53 backer: aren't there any ops around? 10:41:11 --- join: bedah (~bedah@dyndsl-091-248-078-166.ewe-ip-backbone.de) joined #forth 10:42:04 gordonjcp: I don't know. If I was an op, I wouldn't ban him. He's still entitled to his opinion, and we're free to ignore him. I'm just looking to see if there's any substance - a coherent argument of any kind - behind his front. 10:43:42 gordonjcp: Apparently, because postfix expressions become increasingly difficult to read with length, Forth has no practical use regardless of the situation. That's as much substance as I've been able to glean from him. 10:44:46 Anyway, that's my allotment of IRC usage for the month, so I'll be on my way. I'll let DGASAU have the last word. He needs the glory a lot more than I do. 10:54:02 backer: I just ignored him and the other obnoxious kid on here 10:54:55 sorry guys i'm not a channel op any longer. just logged and saw 10:55:42 backer: so please do stick around, it's nice to see people who are doing interesting things; that's what I joined the channel for, not for Internet Arguments. 10:57:48 backer: because arguments are pretty clear, if you understand the topic to some extent. Yet you suggest to prove them in a heavy-lifting way. 10:58:13 Perhaps, you have never written a paper and don't understand how involved it is. 10:58:59 Again, it is exactly how creationists do that: 10:59:30 "Oh, prove that," they say to arguments that are rehashed many times already and even entered daily practice in medicine. 11:00:30 DGASAU: please stop. stop arguing. stop telling people they're wrong. stop being right. stop showing people they're wrong. etc. 11:00:35 please leave the channel. 11:00:41 if i had op i'd kick you 11:00:48 this has gone on long enough 11:00:59 this is a channel for people interested in learning about, and trying, and playing with forth. 11:01:09 not people talking about how it's not X or not Y 11:01:11 please leave. 11:01:24 Why don't _you_ leave the channel? ;) 11:01:34 your presense here is not constructive, helpful, friendly, or desired by most people 11:01:48 it is unreasonable for us to have newcomers ignore you 11:01:50 Are you afraid that you lose your belief? 11:02:20 i don't have beliefs to lose except that i like forth for that it is and like to think most people here are helpful, nice people 11:02:43 for what* it is 11:15:32 --- mode: ChanServ set +o bluekelp 11:15:41 mwaahaha! 11:16:10 DGASAU: if you agree to stop trolling people and stop convincing them not to try forth you can stay 11:16:15 otherwise you should/will leave 11:16:28 let people try. let them fail. help them learn. do not be a jerk. 11:27:26 DGASAU is a most active user here! Do you realize that the channel will be totally silent without him? :) 11:33:32 maybe other users would become active :P 11:33:39 :) 11:50:39 yes - several people have asked for him to be removed in the past - for that reason. he stifles many conversations. 11:51:01 which is unfortunately because he does know a lot about forth - though be uses his skills "for evil" :) 12:00:09 --- quit: DGASAU (Ping timeout: 244 seconds) 12:04:43 --- join: DGASAU (~user@lmpc.drb.insel.de) joined #forth 12:06:01 --- join: Zarutian (~zarutian@168-110-22-46.fiber.hringdu.is) joined #forth 12:19:33 --- quit: phadthai (Ping timeout: 246 seconds) 12:23:53 --- join: true-grue (~true-grue@176.14.216.104) joined #forth 12:27:55 --- quit: real-grue (Ping timeout: 240 seconds) 12:45:46 --- join: phadthai (mmondor@ginseng.pulsar-zone.net) joined #forth 12:52:58 --- join: dys (~dys@ip-109-44-2-114.web.vodafone.de) joined #forth 12:54:55 --- quit: phadthai (Ping timeout: 240 seconds) 13:57:31 --- join: vsg1990 (~vsg1990@pool-173-64-14-42.bflony.fios.verizon.net) joined #forth 14:00:53 bluekelp++ 14:06:12 indeed 14:11:27 I wonder what would be a good way to simulate this? http://www.bradrodriguez.com/papers/piscedu2.htm 14:23:50 --- join: ASau (~user@netbsd/developers/asau) joined #forth 14:26:01 --- mode: ChanServ set +o proteusguy 14:36:54 --- join: wa5qjh (~Thunderbi@121.54.58.148) joined #forth 14:38:59 --- quit: wa5qjh (Read error: Connection reset by peer) 14:41:23 --- join: wa5qjh (~Thunderbi@121.54.58.148) joined #forth 14:45:29 whee gordonjcp, the PISC 14:54:56 Huh, I never really noticed until now that it's not terribly well documented. Well, you have the schematic so you could do a mechanical conversion to verilog, by modeling the chips and connecting them as shown. 14:55:53 The details on the page are interesting to read, but not great from the perspective of someone trying to re-implement it. 15:09:53 --- quit: nighty (Quit: Disappears in a puff of smoke) 15:11:08 Gracana: yeah 15:11:35 Gracana: turns out that Farnell actually stock 74LS181s 15:17:39 --- quit: true-grue (Read error: Connection reset by peer) 15:19:19 I have a lot of 7400 series stuff but I don't have the ALU chips :/ 15:20:32 Gracana: it's the register chips that are unobtanium these days 15:20:50 the actual circuit diagram is simple enough, I could knock that up in an afternoon 15:21:16 maybe I can make a kind of daughterboard for the registers 15:21:33 they're basically a bunch of quad latches 15:22:19 what do you have for prototyping on? 15:24:27 I recently found that wire wrap stuff is expensive. Sockets and nice tools (like an electric or squeeze-type wire wrap gun) are very expensive. 15:24:43 It's just not a common method anymore. 15:24:51 yeah 15:25:03 it's easy enough to just etch PCBs though, particularly single-sided 15:26:10 www.ebay.com/itm/231925633779 and boards like this, man I wish I had that to work with :) 15:26:20 yeah, luckily etching is a lot easier now 15:27:14 and surface mount is actually pretty easy to solder. Discretes are a bit tricky but drag soldering SOIC is super easy 15:27:30 and you don't have to drill holes! 15:27:45 yeah 15:27:51 I mean I have a proper PCB drill press 15:27:53 but FTN 15:28:05 fiddly, fills my kitchen with fibreglass dust 15:28:24 yeah, it's an additional messy step 15:28:30 oh, hm, what do you use for etching? 15:28:40 my friend showed me an etchant I hadn't seen before and it is amazing 15:28:47 it's uh.. cupric chloride I think? 15:28:56 yesss 15:29:40 apparently it's not as nasty as ferric chloride, and it works really fast 15:31:04 ferric chloride isn't particularly nasty 15:31:13 if you get it on you, it'll stain your skin, that's about it 15:31:47 http://i.imgur.com/8N6pV1y.jpg <- SMD chip in a synth I repaired 15:32:01 that one was a bit of a twat because it's a particularly fragile board 15:34:04 for etching I just use ferric chloride in a bubble tank 15:34:17 What did you use, hot plate and heat gun? Looks pretty nice. 15:34:28 it's pretty cheap, pretty clean, doesn't stink the place out, and when it's worn out you just dump it into a bucket of plaster 15:34:34 nope 15:34:36 just a soldering iron 15:34:48 can't use a hotplate, too much stuff on the back of the board 15:35:08 I've got a hot air station but mostly use it for shrinking heatshrink 15:35:08 hot plate for pre-heating, I mean 15:35:13 no need 15:35:22 and again, there's stuff on the back of the board 15:35:23 How did you remove such a big chip with just a soldering iron? 15:35:28 I didn't 15:35:33 I removed it with a scalpel 15:35:47 what you do is, you *gently* cut along the top edge of the pins 15:35:50 Oh, okay. I do a similar thing. 15:35:55 then work the point in underneath and cut up 15:36:04 then with your iron and some braid, wick all the legs off the pads 15:36:11 yeah 15:36:20 then drop the chip on, tack two corners, bead of flux, drag solder 15:36:21 boom 15:36:31 took about as long to write that as it did to do it 15:36:51 --- quit: Zarutian (Quit: Zarutian) 15:38:12 mhmm 15:40:28 --- quit: bedah (Remote host closed the connection) 15:40:40 Gracana: my day job involves shit like this -> http://i.imgur.com/9N4zHkq.jpg 15:42:07 What's your job? 15:44:25 I work for a communications company 15:44:54 I do industrial controls. Designing systems with PLCs and drives and pneumatics and junk. I wish I were on the micro side rather than the macro side.. I feel like there's a lot more interesting work there. And there's a lot to be done, so much of this stuff is just terribly implemented. 15:45:02 so, I spend a lot of time designing radio systems for places, and a lot of time repairing stuff on the bench, and quite a lot of time climbing up tall buildings to fit or repair stuff 15:45:14 oh neato 15:45:33 and as a sideline I repair electronic stuff 15:45:54 weird bits of machinery and things 15:47:51 I have some friends who are way into ham radio. I'm licensed as well, but RF is crazy stuff. I would like to build a software defined radio at some point but there's a lot to learn. 15:48:49 eh, there's not that much to it 15:48:59 which bit are you stuck on? 15:50:04 Gracana: about all you need to know is how if you mix two signals, you get the sum, the difference and the two original frequencies 15:50:27 Gracana: you can think of the SDR front-end as being a pair of direct-conversion receivers, with their local oscillators 90 degrees apart 15:51:19 I mean, I can build antennas just fine and I understand basic AC signal stuff fine, but designing an analog frontend and working with in phase / quadrature components and .... I learn more every time I dive back into it, but still. It's a big domain and I get over my head quickly. 15:51:46 okay 15:51:48 got a bike? 15:52:01 yar 15:52:19 turn it upside down like you were going to take a wheel off 15:52:41 turn the front wheel so that the tyre valve is on the right hand side, at the 3 o'clock position 15:52:45 that's your zero point 15:53:07 your valve is dead level with the wheel spindle 15:53:17 it's also exactly 1 diameter to the right of the wheel spindle 15:53:45 as you rotate the wheel anticlockwise, the valve gets higher as you measure vertically, and further left as you measure horizontally 15:54:05 Yes. The basics of phasors I understand. 15:54:06 once you've turned it 90 degrees it's now exactly above the spindle, it's 1 high and 0 right 15:54:09 okay cool 15:54:17 so you've constructed a sine and a cosine, then? 15:54:21 you get that bit 15:54:25 yeah 15:54:34 damn, that's the bit most people get stuck at 15:54:35 :-D 15:54:57 okay, so you know how a direct-conversion receiver works? 15:56:02 you've got an incoming signal at say 7001kHz 15:56:31 I've done some reading but this would be a good thing for you to explain to me, so go ahead. :) 15:56:36 if you mix in a local oscillator at 7000kHz, you get the sum - 14.001MHz, the difference, 1kHz and the two original frequencies 15:56:43 you can't hear much above 20kHz 15:56:59 I'm 42, I've got pretty good hearing, I'm damn sure I can't hear much above about 18kHz these days 15:57:05 mhmm 15:57:14 so obviously the 7MHz and 14MHz components are right out 15:57:20 but you'll hear your wanted signal at 1kHz 15:57:36 7.001MHz - 7.000MHz = 1kHz difference 15:58:01 this is a dead easy circuit to build, it's an oscillator and some sort of mixer (dual gate mosfet is good here) 15:58:32 the downside is, say you want to hear a signal at 7011kHz, you stick in 7010kHz 15:58:36 you get a 1kHz beat note 15:58:37 good 15:58:44 someone else fires up on 7009kHz 15:58:48 you get a 1kHz beat not 15:58:50 *note 15:58:52 damn 15:58:56 they sound the same 15:59:11 okay this is where it gets tricky 15:59:37 you've translated the radio signals down by 7010kHz 15:59:55 your wanted signal is at 1kHz, 0kHz (DC) is 0kHz 16:00:10 does that mean that the unwanted signal at 7009kHz is now at -1kHz then? 16:00:20 how can it be -1kHz? 16:00:23 this is where you need to know the phase relationship, right? 16:00:26 the I and Q part 16:00:28 bingo 16:00:51 if you have two local oscillators both at 7010kHz but 90 degrees apart 16:00:54 and two receivers 16:01:13 you can determine that the +1kHz signal is leading and the -1kHz signal is lagging 16:02:00 now what you need to do is make them cancel out 16:02:44 I'm a bit shaky on the maths behind this, but if you use a Hilbert transform it rotates the phase of an incoming signal by 90 degrees 16:03:03 hmmm! 16:03:14 so if you transform one leg and then add the two, you end up with the "left hand" signals cancelling and "right hand" signals adding 16:03:28 ahhh 16:03:33 or you subtract, and the "right hand" signals cancel 16:03:53 now the other bit is, you need to tune the actual SDR 16:04:00 to pull the bit of band you want to receive in 16:04:11 yeah, this stuff I do recall reading. Don't remember the hilbert transform part, but yeah. 16:04:15 guess what, that's another local oscillator and mixer but that's digitally generated 16:04:49 you take a unit vector and multiply it by a vector that describes the 16:04:55 damn, lost the word for it 16:05:05 the phase angle per sample 16:05:36 so you end up with a complex value that represents an IQ sine wave oscillator 16:05:56 then you multiply that with your complex IQ signal, which is the mathematical counterpart of a four-quadrant mixer 16:06:19 except it's maths, so it's always perfectly balanced and doesn't leak carrier all over 16:06:40 https://github.com/gordonjcp/lysdr 16:07:24 ! 16:08:15 http://gordonjcp.github.io/lysdr/ 16:08:22 oh, my webserver is borked 16:08:35 no test file for you 16:08:47 nginx really is great, dgsau was right 16:09:11 I have some SDR test files kicking around somewhere but they might not be the right format 16:09:26 they just need to be I/Q 16:09:31 this is super cool though 16:09:39 if they're Q/I then it'll tune backwards, it's no biggie 16:09:48 USB and LSB will be backwards 16:09:53 the way you were describing things earlier (being fuzzy on the math etc) I didn't expect that you'd have written something like this, haha 16:09:58 you'll probably cope with that 16:10:40 well 16:10:46 performance isn't exactly stunning 16:10:54 it's a simple proof-of-concept really 16:11:23 I'm quite pleased with how it does the waterfall scrolling in two blits without having to move more than one line of pixels 16:12:35 ah do you move a window around a larger buffer? 16:13:00 kind of 16:13:13 so say you've got a pixmap 100px high 16:13:28 you want to draw a new line of pixels at the bottom and scroll the rest up 16:14:08 oh I guess you don't need a larger buffer 16:14:09 the obvious to do it is to memmove() from the second line of pixels to the bottom line of pixels up one line, right? 16:14:37 and then blit, to draw that pixbuf to the screen 16:14:46 don't do that 16:14:57 treat the pixbuf as a ring buffer 16:15:03 yeah 16:15:16 draw the new line below the last one you drew 16:15:33 then blit from the next line to the bottom of the buffer, to the top of the drawable 16:16:06 then from the top of the pixbuf to the line you just drew, blit that to the bottom part of the drawable 16:16:44 shunting pixels around the pixbuf is done by the CPU but drawing the pixbuf to the drawable is done by the GPU 16:17:08 even on profoundly shitty hardware it can usually manage to hardware accelerate a flat blit 16:19:08 which version of gtk did you use for this? 16:20:43 --- part: Pastaf left #forth 16:20:48 I did some dabbling with stuff like this, no SDR really but played with some discrete signals math, and I used SDL because gtk3 seemed to make drawing pixels to the window an enormous pain in the ass. 16:22:01 Gtk2 16:22:14 it should be relatively easy to port to Gtk3 16:22:27 much of it was written with converting to 3 in mind 16:23:29 ah here we go, sdr_waterfall_expose 16:24:16 oh and waterfall_update below that 16:25:28 so you make a cairo image surface and write to it and paint to it, and that is reasonably performant 16:26:01 Eeeenteresting. I expected that to work poorly. :) 16:26:26 well 16:26:38 it can easily exceed the rate at which I stuff blocks of samples into it 16:36:17 good enough 16:37:50 I need to go buy some groceries and make dinner, but I've saved your github page and I'll give your SDR program a shot later. Nice chatting with you! 16:38:33 no worries 16:38:38 it might be fiddly to get going 16:38:43 I'm off to bed since it's nearly 1am here 16:39:09 Okay, goodnight then. I'll let you know how/if it goes. 16:43:52 --- quit: ASau (Ping timeout: 240 seconds) 16:49:03 --- join: ASau (~user@netbsd/developers/asau) joined #forth 16:56:43 --- join: nighty (~nighty@d246113.ppp.asahi-net.or.jp) joined #forth 16:57:06 --- quit: ASau (Ping timeout: 260 seconds) 17:14:25 --- join: ASau (~user@netbsd/developers/asau) joined #forth 19:14:05 --- join: vsg1990_ (~vsg1990@pool-173-64-14-42.bflony.fios.verizon.net) joined #forth 19:15:50 --- quit: vsg1990 (Ping timeout: 260 seconds) 19:39:50 --- join: phadthai (mmondor@ginseng.pulsar-zone.net) joined #forth 20:43:34 --- join: dram1 (~Thunderbi@112.65.46.78) joined #forth 20:43:38 --- part: dram1 left #forth 20:43:44 --- join: dram1 (~Thunderbi@112.65.46.78) joined #forth 20:55:40 --- join: xyh (~celtic-bo@111.50.69.121) joined #forth 21:25:11 --- quit: phadthai (Ping timeout: 252 seconds) 21:25:58 --- join: phadthai (mmondor@ginseng.pulsar-zone.net) joined #forth 22:24:26 --- part: xyh left #forth 23:59:59 --- log: ended forth/16.06.20