00:00:00 --- log: started forth/15.10.03 00:00:10 --- quit: xyh (Remote host closed the connection) 00:31:47 --- join: DoYouKnow (~DoYouKnow@unaffiliated/doyouknow) joined #forth 00:44:27 http://www.greenarraychips.com/home/documents/greg/PB001-100503-GA144-1-10.pdf 00:44:33 sounds like a good thing for an oscilloscope 00:44:37 or other DSP device 00:44:40 maybe a spec analyzer 00:45:37 or for computing huge FFTs 00:45:50 to measure the accuracy of natural or man-made oscillator 00:46:03 +s 00:49:04 using distributed memory of course 00:49:05 bbl 00:56:21 --- quit: DoYouKnow (Quit: Leaving) 01:32:03 --- join: xyh (~xyh@14.20.4.255) joined #forth 01:39:47 --- quit: xyh (Ping timeout: 268 seconds) 01:47:08 --- join: xyh (~xyh@14.20.4.255) joined #forth 01:57:36 --- quit: xyh (Remote host closed the connection) 01:58:26 --- join: xyh (~xyh@14.20.4.255) joined #forth 02:25:08 --- quit: xyh (Ping timeout: 240 seconds) 02:31:40 --- join: xyh (~xyh@14.20.4.255) joined #forth 02:43:54 --- join: xyh- (~xyh@14.20.4.255) joined #forth 02:48:10 --- quit: xyh (Ping timeout: 255 seconds) 02:55:00 --- quit: xyh- (Remote host closed the connection) 02:55:53 --- quit: rprimus (Ping timeout: 256 seconds) 02:56:39 --- join: rprimus (~micro@a9.lence.net) joined #forth 02:57:03 --- nick: rprimus -> Guest8745 03:07:24 --- join: xyh (~xyh@14.20.4.255) joined #forth 03:08:55 --- quit: segher (Ping timeout: 256 seconds) 03:09:46 --- join: segher (segher@bombadil.infradead.org) joined #forth 03:41:30 --- quit: xyh (Remote host closed the connection) 03:47:06 --- join: xyh (~xyh@14.20.4.255) joined #forth 04:10:32 --- quit: xyh (Remote host closed the connection) 04:55:12 --- join: xyh (~xyh@14.20.4.255) joined #forth 05:02:21 --- quit: xyh (Ping timeout: 264 seconds) 05:22:44 --- join: xyh (~xyh@14.20.4.255) joined #forth 05:29:11 --- quit: xyh (Ping timeout: 246 seconds) 05:30:50 --- nick: Guest8745 -> rprimus 05:37:40 --- join: Zarutian (~zarutian@168-110-22-46.fiber.hringdu.is) joined #forth 05:58:25 --- join: nighty-_ (~nighty@hokuriku.rural-networks.com) joined #forth 06:01:32 --- quit: proteusguy (Read error: Connection reset by peer) 06:01:33 --- quit: proteusguy_ (Read error: Connection reset by peer) 06:10:54 --- join: kumul (~mool@adsl-64-237-237-112.prtc.net) joined #forth 06:12:12 --- join: xyh (~xyh@14.20.4.255) joined #forth 06:18:02 --- join: proteusguy (~proteusgu@ppp-110-168-230-101.revip5.asianet.co.th) joined #forth 06:18:02 --- mode: ChanServ set +v proteusguy 06:19:12 --- join: proteusguy_ (~proteusgu@ppp-110-168-230-101.revip5.asianet.co.th) joined #forth 06:24:10 --- quit: xyh (Ping timeout: 255 seconds) 06:31:11 --- join: xyh (~xyh@14.20.4.255) joined #forth 06:38:01 --- quit: xyh (Ping timeout: 256 seconds) 07:02:54 --- quit: kumul (Read error: Connection reset by peer) 07:22:02 --- join: mnemnion (~mnemnion@2601:400:8001:14c9:f899:c60e:4c8a:d538) joined #forth 07:36:43 --- join: BitPuffin|osx (~isak@199.72.208.46.dyn.plus.net) joined #forth 07:37:03 hmm I should probably maybe ask on the mailing list about my paralellism question the other day 07:39:26 * impomatic wonders why Euroforth is so expensive :-( 07:46:23 --- join: kumul (~mool@adsl-64-237-237-176.prtc.net) joined #forth 08:09:16 --- quit: karswell (Remote host closed the connection) 08:10:27 --- join: karswell (~user@61.255.199.146.dyn.plus.net) joined #forth 08:26:29 --- quit: karswell (Read error: Connection reset by peer) 08:26:46 --- join: xyh (~xyh@183.14.254.181) joined #forth 08:27:43 --- join: karswell (~user@61.255.199.146.dyn.plus.net) joined #forth 08:43:19 BitPuffin|osx: what was your question? 08:43:47 Zarutian: how to do get paralellism with gforth, ie something like native os threads with mutexes and so on 08:44:14 which I think would essentially involve having thread-local dictionary memory so that you don't have allot in different threads fighting each other 08:44:29 BitPuffin|osx: cant help your there. I usually run forth on bare metal or extremely small emulators. 08:44:30 but at the same time what does that then imply for word definitions, it gets kinda weird I feel 08:44:48 Zarutian: ah okay :) I was thinking about using forth for game development 08:45:08 BitPuffin|osx: tell you what. Multi threading in the same memory space is a quackmire you dont want to step into. 08:46:04 Zarutian: yup that's why I was thinking you need thread local dictionary memory 08:46:19 though I think shared memory is sometimes essential 08:46:23 just that you gotta be careful :P 08:46:33 (I do multiprogramming/threading though in Forth. Just save the stacks and PC to memory and restore from another such block) 08:46:55 why is shared memory sometimes essential? where does that assumption originate? 08:47:12 well it could be performance etc 08:47:26 you can't afford message passing sometimes 08:47:33 why is shared memory more performant than the converse option? 08:48:13 depends on how the message passing is impemented. If it is done via message buffer segments that only have linear pointers to them then it is extremely cheap 08:48:36 hmm well say you want a certain update to propagate to all the different threads, then sending a message to all of them and have them update their thing might have too much overhead in some cases 08:48:40 basically it's entirely situational 08:48:44 and sometimes it's cheap enough 08:48:47 but in something like games 08:48:49 if you do it (in systems that have virtual memory) through page ownership transfers then it is also cheap. 08:48:52 say you're making a VR game 08:48:59 where the ideal framerate will pretty much be 120hz 08:49:13 then you have 8.3ms to do shitloads of work :P 08:49:35 (not counting the fact that everything doesn't need to happen every frame) 08:49:51 you dont do that shitloads of work on one cpu and memory. The von Neuman memory access bottlenet will kill your performance 08:50:07 (if you do it that way that is) 08:50:13 could be, anyway 08:50:21 my point is that yes a lot of times you can and should avoid it 08:50:24 but there's always edge cases 08:50:52 so I'm not arguing that there aren't better ways, I'm just saying that different problems require different solutions 08:51:01 I recommend that you look into FlowBased Programming even though you wont split the work into many modules. 08:52:20 VR stuff usually mean that you have some scene graph that is rendered from the point of view of one or two virtual cameras, yes? 08:52:46 Zarutian: yup 08:53:07 you have to render the same scene twice at a higher framerate, nice surprise for the devs out there :D 08:53:31 so you got some GPU that is rendering (eather traditionally or via ray tracing) what those two cameras see. 08:54:30 (that they dont just use two GPU cards to do that beats me) 08:56:17 so you pretty much have at end of that pipeline two 'screens'. Before that you have a GPU per screen. Before those two you have some device or memory that keeps track of the info in the scene graph that pertains to rendering it, yes? 08:57:05 (shapes, meshes, materials, graphic textures and all that jazz) 08:57:45 yeah usually there's a gpu involved 08:58:02 only one? 08:58:03 though I'm starting to get interested in reviving software rending but I*m weird 08:58:29 I think the reason for not using two GPUs is that it's hard to share the state between them maybe, though actually maybe not 08:58:41 just that you get twice the overhead to upload the data to the gpu 08:58:42 GPU is software rendering only using an extremely paraell processing units. 08:58:44 althouuuuuuuuuuugh 08:58:53 some GPUs actually pretty much contain two GPUs 08:59:04 and act as though they're one 08:59:18 not quite the same with how you'd program it if you had two gpus 08:59:32 the reason we only use one GPU is because consumers will generally not have two 08:59:34 what is the overhead of uploading data to the gpu? Stop thinking about the traditional PC or console architecture. 08:59:48 and if they do they will have them set up in like SLI or something 09:00:09 Oh, right, again the usual Forth programmer's bullshit about "quagmires." 09:00:21 The whole world uses threads without major problems. 09:00:41 ASau: have you played Robo Rally? Have you debugged a race condition? 09:00:50 Yes, I debugged race conditions. 09:01:10 A lot of Java programmers use threads without major problems. 09:01:10 ASau: and how did you solve it? added another mutex or? 09:01:25 Zarutian: oh I dunno off the top of my head, but I believe it's slower than reading from ram probably 09:01:28 In some cases I added locks. 09:01:37 (cache missed reads I mean) 09:01:41 In some cases I rewrote using non-blocking sync. 09:01:49 ASau: and you see how 'fast' those Java programs are. Usually each object has a 'monitor' or is a mutex reagion. 09:01:52 In some cases I rewrote using sync variables. 09:02:01 --- quit: asagk (Ping timeout: 240 seconds) 09:02:04 Yes, I see how fast those Java programmers are. 09:02:09 BitPuffin|osx: still thinking of the traditional PC enviroment. 09:02:13 Some of them beat C and C++ in performance. 09:02:26 Zarutian: yes because that's what the people who we ship games to have 09:02:54 BitPuffin|osx: aah, okay. So the VR headset would only be like a diffrent display. 09:03:00 correct 09:03:06 you have GearVR 09:03:11 which is hooked up to a samsung phone 09:03:39 and then the other occulus thingy 09:03:46 which is just like a screen for a pc 09:03:48 HTC Vive 09:03:48 BitPuffin|osx: well you will get into the queasyness problems then. 09:03:49 same thing 09:04:14 Zarutian: hmm? queasyness problems? 09:05:28 BitPuffin|osx: see something in the corner of you eye. You lock on it and rotate your head to get a better look. I think the occulus people solved it by rendering much more and dragging the viewport around. 09:06:21 ah you mean that it would be like blurry in the courners etc 09:06:22 BitPuffin|osx: change of depth is also often what causes the quasyness if the hardware cant keep up. 09:06:29 didn't they solve it by tracking eye movements? 09:06:36 or hmm I'm not sure what they did 09:07:20 ASau: to me C or C++ is extremely slow both in performance and incrimental developement. 09:07:50 ASau: hard to get the 'pottery wheel' feedback feeling with something that takes a while to compile. 09:08:51 If you have problems with performance in C or C++, you just are poor programmer. 09:08:54 --- quit: karswell (Read error: Connection reset by peer) 09:09:06 you can write really slow C or C++ 09:09:12 but you can also write quite fast 09:09:13 You can. 09:09:23 But that's another problem. 09:09:30 if you work with the compiler and figure out when it will actually do the optimizations you desire etc 09:09:37 If you have problems with fast development in C++, you just don't know C++ well enough. 09:09:51 BitPuffin|osx: works so long as you never update or upgrade you compiler. 09:09:59 well to some degree it could also be attractive to have a language that encourages things that are faster and makes those things easier 09:10:07 --- join: karswell (~user@61.255.199.146.dyn.plus.net) joined #forth 09:10:14 ASau: nobody knows C++ well enough and cannot. (It is too damn big) 09:10:28 so to some degree you can blame bad language design, but that doesn't mean it's impossible to write fast C or C++ code 09:10:46 BitPuffin|osx: it is just very laborous. 09:11:06 pretty sure Bjarne has said many times the same thing Zarutian 09:11:09 said 09:11:59 but then again, I am doing embedded stuff and electronics manifacturing. 09:12:14 cool 09:12:21 though not surprised to see that in a forth channel :) 09:14:16 it is amazing what one can do with something like just romlogic (a rom whose address lines are tied to sensor inputs and the output of an state register. The data lines of it are then tied to ouput lines and the input of the state register) 09:14:19 either way I'd like to be able to use gforth for toying with some game stuff with threads etc 09:14:36 though I can't actually ship anything in gforth because license 09:14:50 --- join: asagk (~asagk@i59F6D4CF.versanet.de) joined #forth 09:14:56 BitPuffin|osx: by all means ask the mailing list and google gforth multithreading. 09:15:12 I've googled a little bit and searched the mailing list archive 09:15:25 there was a little bit about some library for cooperative threads I think 09:15:32 BitPuffin|osx: oh, the whole 'software license' shit rears its ugly head. 09:15:41 :) 09:16:05 but, I can probably not too slowly whip up my own llvm-based public-domain licensed implementation or something when I need to ship something 09:16:15 so I think toying with something in gforth is pretty ok for now I believe 09:16:34 it seems like a relatively competent implementation other than its license :D 09:16:58 I wish they would have put something like the link exception in though like gcc 09:17:00 BitPuffin|osx: here is a thing. You can sell software even though it is under GPL. You only need to give them copy of the source code used. 09:17:16 I'm fully aware of that 09:17:35 but not everyone is comfartable with that 09:18:10 which used to be the case with all software because if you cant see and fix the software you buy then you cannot be sure that it has something malicious in it. 09:18:46 s/that it has/that it has not/ 09:19:15 sure 09:19:25 I mean I'm all for open source code 09:19:37 which opens one another whole box of topics I could go on and on about: computer security or more spefic access control for program instances in computer systems. 09:20:21 I'm pretty much an advocate of getting rid of copyright and everything be public domain etc 09:20:29 so don't misunderstand where I'm coming from :P 09:20:33 * Zarutian finds the "I trust every program I run with access to everything in my user account!" concept which most OSes are build on today rather ridicilus. 09:20:54 yeah it's quite funny with OSs 09:21:09 I mean to some degree the model makes sense on servers 09:21:11 but at home it's like 09:21:21 every program can delete all my user files and pictures and documents 09:21:28 look at capros and its priors if you want to know something about it. 09:21:32 but at least they can't delete my programs xD 09:21:33 big whoooooooooop 09:21:59 because if some program deletes the kernel image that would be so much more worse if they deleted all of my personal data 09:22:05 totally operating system totally 09:22:12 * Zarutian wonders where the heck the seperation between designation and authority (access) originated. 09:46:13 --- quit: xyh (Remote host closed the connection) 09:57:42 --- quit: Quozl` (Quit: No Ping reply in 180 seconds.) 10:01:52 --- quit: darkf (Quit: Leaving) 10:09:09 --- part: beretta left #forth 10:09:44 --- join: beretta (~beretta@cpe-65-185-42-203.columbus.res.rr.com) joined #forth 10:36:24 --- quit: kumul (Ping timeout: 265 seconds) 11:40:30 --- join: bedah (~bedah@dyndsl-037-138-118-183.ewe-ip-backbone.de) joined #forth 11:48:04 --- join: lynx`` (~lynx@104.254.90.243) joined #forth 11:53:27 --- part: lynx`` left #forth 11:53:53 --- quit: beretta (Ping timeout: 252 seconds) 12:13:47 --- join: beretta (~beretta@cpe-65-185-42-203.columbus.res.rr.com) joined #forth 12:30:36 --- join: Mat4 (~claude@ip5b40b95e.dynamic.kabel-deutschland.de) joined #forth 12:31:09 good day 12:41:00 Haaais. -ω-... 12:44:35 whatever this means... 13:02:14 Basically "hello". -ω-. 13:03:27 oh, hi 13:04:47 --- quit: pointfree (Ping timeout: 265 seconds) 13:14:58 Hais. -ω-. 13:56:57 --- quit: Mat4 (Quit: Leaving) 14:52:40 --- quit: true-grue (Read error: Connection reset by peer) 15:36:10 --- quit: Zarutian (Quit: Zarutian) 15:39:55 --- quit: nighty-_ (Quit: Disappears in a puff of smoke) 16:12:00 --- quit: karswell (Read error: Connection reset by peer) 16:13:13 --- join: karswell (~user@61.255.199.146.dyn.plus.net) joined #forth 16:35:58 --- join: daowee (daowee@m90-141-251-170.cust.tele2.se) joined #forth 16:49:28 --- quit: bedah (Quit: extreme quitting :D) 17:38:35 --- quit: mnemnion (Remote host closed the connection) 17:47:49 --- join: xyh (~xyh@183.37.2.31) joined #forth 17:52:01 --- quit: karswell (Read error: Connection reset by peer) 17:53:13 --- join: karswell (~user@61.255.199.146.dyn.plus.net) joined #forth 18:00:45 --- join: proteusguy__ (~proteusgu@ppp-110-168-230-86.revip5.asianet.co.th) joined #forth 18:01:45 --- quit: proteusguy_ (Ping timeout: 244 seconds) 18:01:45 --- quit: proteusguy (Ping timeout: 264 seconds) 18:13:36 --- join: proteusguy (~proteusgu@ppp-110-168-230-86.revip5.asianet.co.th) joined #forth 18:13:36 --- mode: ChanServ set +v proteusguy 18:39:08 --- join: mnemnion (~mnemnion@2601:400:8001:14c9:f899:c60e:4c8a:d538) joined #forth 18:41:14 --- quit: xyh (Remote host closed the connection) 18:44:02 --- quit: mnemnion (Ping timeout: 246 seconds) 19:02:42 --- join: xyh (~xyh@183.37.2.31) joined #forth 19:14:14 --- quit: xyh (Remote host closed the connection) 19:14:17 --- join: Shark8 (~Thunderbi@75-161-74-239.albq.qwest.net) joined #forth 19:31:30 --- quit: karswell (Read error: Connection reset by peer) 19:32:42 --- join: karswell (~user@61.255.199.146.dyn.plus.net) joined #forth 19:37:28 --- join: darkf (~darkf___@unaffiliated/darkf) joined #forth 19:59:40 --- join: mnemnion (~mnemnion@c-68-40-58-223.hsd1.mi.comcast.net) joined #forth 20:09:07 --- quit: karswell (Remote host closed the connection) 20:10:14 --- join: karswell (~user@61.255.199.146.dyn.plus.net) joined #forth 20:31:47 Keshl: is it a language-specific greeting? I remember seeing it occasionally used (w), and I'm also on a (english though) channel where it's the most common greeting form, but its origin was lost :) 21:10:52 --- quit: mnemnion (Remote host closed the connection) 21:22:25 --- join: xyh (~xyh@183.37.2.31) joined #forth 21:34:05 What, "hais"? That's standard lolspeak, oωo. http://icanhas.cheezburger.com/ -- Really. Not joking. Actual origin! As for the owofaces, that's because I like the ":3" emote turned sideways and have an over-addiction to the face. <ω<' 21:43:48 --- quit: xyh (Remote host closed the connection) 21:45:15 --- quit: nisstyre (Ping timeout: 256 seconds) 21:52:28 Keshl: thanks... did that already exist on IRC ~15 years ago though? heh 22:00:47 Not sure.. But it does now, so there. -ω-. The emote its self comes from another culture entirely, though. While that /has/ been around for over fifteen years, and on IRC for at least ten or so, it's on another network. So there. -ω-. 22:01:17 And if need be I *promise* I can make this on-topic. <ω<' 22:09:11 heh.. not that my question was on-topic :)) 22:09:15 thanks again 22:09:22 Ish welcomes. -ω-. 22:14:16 --- join: nisstyre (~yourstrul@li611-52.members.linode.com) joined #forth 22:41:32 --- quit: scoofy (Ping timeout: 240 seconds) 23:35:39 --- quit: Keshl (Quit: Konversation terminated!) 23:59:59 --- log: ended forth/15.10.03