kristianpaul | boom ! | 00:15 |
---|---|---|
kristianpaul | i found this http://en.qi-hardware.com/wiki/Copyleft_Hardware_Style_Guide#boom | 00:15 |
kristianpaul | wolfspraul: I dont see you are concerned or interested to work/support on UI design | 00:22 |
kristianpaul | I mean nanonote UI is mosrlty comunity effort, milkymist sebastien work, but are you interested on this are? | 00:22 |
kristianpaul | I mean from the value it could generate on a device | 00:23 |
kristianpaul | Or what are your general toughts about this topic | 00:23 |
kristianpaul | no care, and let it grown by it self? :-) | 00:23 |
wolfspraul | he | 00:23 |
wolfspraul | why would I not be concerned about the UI of either one? I use them... | 00:24 |
kristianpaul | I mean, conecting people | 00:24 |
wolfspraul | but I'm overwhelmed with coding contributions, indeed | 00:24 |
wolfspraul | what do you want me to do? | 00:25 |
kristianpaul | nothing :-) just asking your toughts about UI importance for our copyleft hw | 00:25 |
wolfspraul | great interview with bunnie http://blog.makezine.com/2012/04/30/makes-exclusive-interview-with-andrew-bunnie-huang-the-end-of-chumby-new-adventures/ | 00:25 |
kristianpaul | well, perhaps asking for feedback about UI design for nanonote or milkymist from people you may know from OM times.. | 00:26 |
wolfspraul | of course UI is very important, but I think what we have is not horrible and improving | 00:26 |
kristianpaul | anyway just a radom tought | 00:26 |
kristianpaul | sure not horrible and improving (growing) a lot indeed | 00:26 |
wolfspraul | for the nanonote, I have to check those latest things from David | 00:27 |
kristianpaul | yes | 00:27 |
wolfspraul | I still want a launcher that is simpler, more scriptable. no need to follow some 'sea of icons' philosophy imo | 00:27 |
wolfspraul | users are far more flexible than one might think, if there finally is something of quality that is worth learning | 00:27 |
wolfspraul | which is rare, as always in life | 00:28 |
kristianpaul | sea of icons, got it | 00:28 |
wolfspraul | chumby no more http://www.utsandiego.com/news/2012/may/01/san-diegos-chumby-industries-calling-it-quits/ | 00:28 |
kristianpaul | yeah.. | 00:28 |
kristianpaul | i noticed from the interview the non-profit and what lead to that thinking.. | 00:29 |
wolfspraul | there's a lot of things to learn from chumby, as any great endeavor, good and bad | 00:29 |
wolfspraul | kristianpaul: so you will attempt to pull together a gps expansion board for m1? | 00:31 |
kristianpaul | users are far more flexible, yeah well, most of the first nanonote impresion from people i demo it was doint kbd brutefore to get how it woek | 00:31 |
kristianpaul | wolfspraul: yes | 00:31 |
wolfspraul | nice | 00:31 |
kristianpaul | recalling, | 00:31 |
wolfspraul | where will the sources be? we will hook it up to schhist | 00:32 |
kristianpaul | oh let me do a quick import moment | 00:33 |
wolfspraul | for the verilog2html, I didn't mean a source code browser | 00:33 |
kristianpaul | sure i know | 00:33 |
wolfspraul | I meant a graphical representation of the chip schematics, converted from verilog to drawings | 00:33 |
kristianpaul | wich kind of drawings? | 00:34 |
kristianpaul | netlist, fpga map .. what else.. | 00:35 |
wolfspraul | I had in mind building blocks, like registers, combinational blocks, etc. | 00:36 |
kristianpaul | RTL | 00:36 |
wolfspraul | no something graphical | 00:36 |
wolfspraul | provides for a nice overview of the machine I think | 00:36 |
wolfspraul | and some tools like I believe synopsys can create such views, I think (that's what I'm trying to find out) | 00:37 |
wolfspraul | maybe it's not considered interesting/worthwhile by the pros though, who knows :-) | 00:37 |
kristianpaul | i honestly find hard it to get it usefull some how, with out it provide some interaction | 00:37 |
wolfspraul | huh? | 00:38 |
wolfspraul | you mean a drawing of blocks/schematics would not be helpful? | 00:38 |
wolfspraul | don't understand | 00:38 |
kristianpaul | after covert hdl code to drawings it gets bigs, i mean the data | 00:38 |
kristianpaul | i will surelly | 00:39 |
kristianpaul | but not for all people | 00:39 |
wolfspraul | have you seen or used such drawings? | 00:39 |
wolfspraul | and how did you create them? | 00:40 |
wolfspraul | maybe it only makes sense for parts of the SoC, or maybe it's a big one one could zoom into - again: don't know | 00:40 |
wolfspraul | that's what I'm trying to find out | 00:40 |
kristianpaul | i just remenber some drawings from altera soc builder when i took some courses at college about vhdl ;) | 00:40 |
kristianpaul | it was generated froma wizard | 00:40 |
kristianpaul | and allowe me to see a overview of the soc _and_ also get to the details of every module if i wanted to | 00:41 |
wolfspraul | nice | 00:42 |
kristianpaul | i not remenber combinational blocks, guess because was soc centric | 00:43 |
kristianpaul | but do remenber the blocks and registers | 00:43 |
kristianpaul | but was not just a drawing it self, as it allow me to design soc and even run code | 00:43 |
wolfspraul | I was thinking of a graphical representation of our current milkymist soc (or parts of it) | 00:48 |
wolfspraul | I will go through some free sw tools and see whether such a feature hides somewhere | 00:49 |
wolfspraul | and if not, also ok. then read verilog and think about it :-) | 00:49 |
kristianpaul | read verilog is a good staring point | 00:51 |
kristianpaul | i recognice the code browser is not the best looking way | 00:51 |
kristianpaul | but | 00:51 |
kristianpaul | if you replace some stuff from code to blocks and provide a map, but also chance to looks/zoom/down at source code in both verilog and C may look more usefull | 00:53 |
kristianpaul | data vistualization is an interesting topic anyway ;-D | 00:53 |
kristianpaul | wpwrak: u there? | 01:23 |
kristianpaul | .gbr .gto .gbo .gts .gbs from a kicad project should not be commited right? | 01:24 |
kristianpaul | .gbl .gtl forgot too | 01:25 |
wolfspraul | looks like gerber files | 01:25 |
wolfspraul | the different layers | 01:25 |
kristianpaul | all the g.. i mentioned? | 01:26 |
wolfspraul | gto is silkscreen front | 01:26 |
wolfspraul | gbo silkscreen back | 01:26 |
wpwrak | i think it's pretty much .g* :) | 01:26 |
kristianpaul | ah goo | 01:27 |
kristianpaul | d | 01:27 |
wolfspraul | gbs: soldermask back | 01:27 |
wolfspraul | gts: soldermask front | 01:27 |
wpwrak | kristianpaul: for a moment, i tried to see if i remember a .goo ;-) | 01:30 |
kristianpaul | jaja | 01:31 |
wpwrak | this lists a few of the layers: http://projects.qi-hardware.com/index.php/p/ben-wpan/source/tree/master/atben/README-PCB | 01:32 |
kristianpaul | .dsv shall be ignore as well right? | 01:32 |
wpwrak | the directory .dsv/ ? yup | 01:33 |
kristianpaul | ahh is a directory.. | 01:33 |
kristianpaul | i took that from xue.. as i dint found a .gitignore at ben-wpan ;-) | 01:34 |
wpwrak | oh, i don't bother with .gitignore :) | 01:35 |
kristianpaul | i knew it :) | 01:36 |
kristianpaul | git is all about better practices.. | 01:36 |
kristianpaul | okay what about .bak extension ? | 01:37 |
kristianpaul | EESchema Schematic File, seems should keep | 01:37 |
wpwrak | i tend to keep a gazillion of local temp files around that i name on the spot. so a shareable gitignore would only know about half the things anyway | 01:37 |
wpwrak | keep .bak, but not in the repository :) | 01:38 |
kristianpaul | hmm? | 01:38 |
wpwrak | besides, the repository IS a backup :) | 01:38 |
kristianpaul | ahh | 01:38 |
kristianpaul | :) | 01:38 |
wpwrak | (keep .bak locally. i.e., don't actively kill them) | 01:38 |
wpwrak | (for they do no harm. but they're usually irrelevant) | 01:38 |
kristianpaul | yeah, but thats generated by some kicad module right? | 01:39 |
kristianpaul | s/module/app | 01:39 |
kristianpaul | what about .drl ? | 01:39 |
wolfspra1l | drill file | 01:40 |
kristianpaul | gerber related? | 01:40 |
kristianpaul | ok | 01:40 |
wpwrak | drill file is similar to gerber. the format differs slightly, though | 01:41 |
wpwrak | (it's called excellon) | 01:42 |
kristianpaul | so no problem delete it | 01:42 |
Action: kristianpaul just deleted also some .dcm and .dck filled by comments | 01:44 | |
kristianpaul | btw can we currently make a diff between two existent .brd files? | 01:45 |
wpwrak | not usefully | 01:46 |
kristianpaul | okay .pos guess is related as well with gerber, mill and smt | 01:46 |
wpwrak | .pos is component positions | 01:46 |
wpwrak | you need that for the SMT machine. it's an output from pcbnew | 01:47 |
kristianpaul | yes | 01:47 |
kristianpaul | that is machine generated it said too | 01:47 |
kristianpaul | hmm a custom QFN28.emp | 01:49 |
kristianpaul | that stay for now i guess | 01:49 |
kristianpaul | also all the libs | 01:49 |
kristianpaul | hmm the .net | 01:50 |
kristianpaul | i dont see that one in atusb.. | 01:50 |
wpwrak | what's .emp ? | 01:51 |
kristianpaul | because boom? | 01:51 |
wpwrak | the .net is generated | 01:51 |
kristianpaul | .emp - export of module | 01:51 |
kristianpaul | according to http://wiki.xtronics.com/index.php/Kicad | 01:51 |
kristianpaul | ok | 01:51 |
wpwrak | hmm. never saw .emp | 01:52 |
wpwrak | .sch + .cmp -> .net | 01:52 |
wpwrak | so keep the .cmp | 01:53 |
kristianpaul | already | 01:54 |
kristianpaul | what about .pho ? | 01:55 |
kristianpaul | ah more gerger | 01:55 |
wpwrak | aye | 01:55 |
wolfspra1l | yes | 01:56 |
wpwrak | you have a lot of strange files. did you go through all the "create file" options ? :) | 01:56 |
kristianpaul | nope | 01:57 |
kristianpaul | i'm going to import a project from Artyom svn repo to remove uncesary items and make a gps board for m1 | 01:57 |
wpwrak | aah, i see. | 01:58 |
wpwrak | so he just committed everything in sight | 01:58 |
wpwrak | unfortunately that's even understandable, given the huge number of files kicad spits out | 01:59 |
kristianpaul | http://code.google.com/p/gnsssdr/source/browse/#svn%2Ftrunk%2FFRONT_END_PROJECT%2FFRONT_END_SOURCE%2FKiCAD | 01:59 |
kristianpaul | yup | 01:59 |
kristianpaul | lots of dumb files, just comments | 01:59 |
kristianpaul | .csv?? | 02:00 |
kristianpaul | he | 02:00 |
kristianpaul | pretty usefull | 02:00 |
kristianpaul | i saw it before never look at it | 02:00 |
wpwrak | btw, for canned procedures, you can use this: http://projects.qi-hardware.com/index.php/p/ben-wpan/source/tree/master/makefiles/Makefile.kicad | 02:00 |
kristianpaul | provides all list of material with smd and value | 02:01 |
kristianpaul | i will keep that to help me boom | 02:01 |
kristianpaul | btw where is boom repo? | 02:02 |
wpwrak | it starts here: svn.openmoko.org/trunk/eda/boom/ | 02:03 |
kristianpaul | ahh | 02:04 |
kristianpaul | not qi | 02:04 |
kristianpaul | i was lost looking there | 02:04 |
wpwrak | the continuation on qi-hw in eda-tools/boom/ | 02:04 |
kristianpaul | ah | 02:04 |
kristianpaul | eda | 02:04 |
kristianpaul | ok | 02:04 |
wpwrak | and the rewrite is in eda-tools/b2/ but that doesn't do anything useful yet | 02:05 |
kristianpaul | cause svn.o.o seems down? | 02:05 |
wolfspraul | yeah we need to move those sources | 02:05 |
wolfspraul | asap | 02:05 |
wolfspraul | I'm sure werner has local copies | 02:05 |
wpwrak | for a real-life use case of the old boom, see ben-wpan/bom/ | 02:05 |
wpwrak | hmm. still down ? that's bad | 02:06 |
kristianpaul | and not in archive.org.. | 02:08 |
wolfspraul | we have to move over | 02:09 |
wolfspraul | asap | 02:09 |
wolfspraul | I thought werner had moved mostly everything, maybe that's the last piece remaining? hopefully... :-) | 02:09 |
kristianpaul | ah wait i have werner om local archive ;-) | 02:09 |
qi-bot | [commit] Werner Almesberger: rescue files from svn.openmoko.org (master) http://qi-hw.com/p/eda-tools/881bf33 | 02:09 |
wpwrak | here they are :) | 02:10 |
kristianpaul | phew | 02:10 |
kristianpaul | ah perfect, thats what i need to read | 02:11 |
wpwrak | workflow.fig ? :) | 02:11 |
wpwrak | at least i hope cat .svn/entries | sed '/^file/{x;n;};x;d' does list all the files that exist in svn ... | 02:14 |
kristianpaul | alredy deleted as well | 02:29 |
kristianpaul | wpwrak: old-boom/README | 02:30 |
kristianpaul | but the .fig is good too | 02:31 |
wpwrak | it's not terribly useful at the moment because you can't fetch new data from digi-key. but if you have your own inventory or copy things over manually, then you can run it on that | 02:35 |
kristianpaul | i get mad sourcing some capacitors.. | 02:35 |
kristianpaul | so wolfspraul convince me move boom first ! | 02:35 |
wpwrak | caps can be nasty :) | 02:35 |
wpwrak | hmm. the old boom will die. the new one is similar but has a number of differences | 02:36 |
qi-bot | The build was successful: http://fidelio.qi-hardware.com/~xiangfu/build-nanonote/openwrt-xburst.full_system-20120430-2002 | 02:38 |
kristianpaul | damn as soon as i activated cdc_ethernet and usbnet on the mr11u i lost ethernet networking.. | 02:41 |
wolfspraul | you bricked it? | 02:44 |
kristianpaul | i'm afraid yes.. | 02:45 |
kristianpaul | confirmed, as soon as it turn on the wifi led gets on | 02:46 |
kristianpaul | it dint happen before.. | 02:46 |
kristianpaul | oh well | 02:46 |
kristianpaul | sooner or later i must have to open it :) | 02:47 |
kristianpaul | ah, i activated and nanonote was re-pluged | 02:53 |
wolfspraul | the serial console is easy to solder to, and then you can reflash with u-boot | 02:53 |
kristianpaul | i saw the cdc with dmesh on the mr11u, and last thing i did was try ifconfig usb0 | 02:53 |
kristianpaul | it worked then.... | 02:53 |
kristianpaul | yup | 02:53 |
kristianpaul | u-boot wow, i was expecting something less elaborated | 02:55 |
GitHub27 | [m1-gps-expansion] kristianpaul created master (+1 new commit): https://github.com/kristianpaul/m1-gps-expansion/commit/f866cd2 | 02:59 |
GitHub27 | [m1-gps-expansion/master] Initial import after clean up from upstream - Cristian Paul PeƱaranda Rojas | 02:59 |
kristianpaul | wolfspraul: sources to hook it up to schhist,yes, here https://github.com/kristianpaul/m1-gps-expansion | 03:00 |
wolfspraul | nice | 03:00 |
kristianpaul | icarus copy, wow, this move faster :) | 03:17 |
kristianpaul | nas? | 03:22 |
wolfspraul | kristianpaul: you have 2 dvb-t dongles now | 14:06 |
wolfspraul | do they have the same antenna connector? | 14:06 |
kristianpaul | yes they do | 14:07 |
wolfspraul | which one is this? sma? cmx? I can't figure it out | 14:07 |
wolfspraul | there are so many RF connectors | 14:08 |
kristianpaul | hmm | 14:08 |
wolfspraul | mcx | 14:08 |
kristianpaul | yes | 14:08 |
kristianpaul | i think | 14:08 |
kristianpaul | sma no for sure | 14:08 |
wolfspraul | the german wikipedia has a long article with lots of different rf connectors | 14:09 |
wolfspraul | and this picture, which looks like the one you have I think http://de.wikipedia.org/w/index.php?title=Datei:Belling-Lee-Stecker.jpg | 14:09 |
wolfspraul | belling lee connector? | 14:09 |
wolfspraul | ok, unclear connector now. How about the antenna? | 14:10 |
kristianpaul | no | 14:10 |
wolfspraul | can you use the same antenna for different frequencies? | 14:11 |
kristianpaul | from the pic is the one in the middle or left corner? | 14:11 |
wolfspraul | it seems the osmo guys throw away all accessories including antenna before shipping out the dongle :-) | 14:11 |
kristianpaul | same atenna no way | 14:11 |
wolfspraul | what characterizes the antenna? | 14:11 |
kristianpaul | polarization | 14:11 |
kristianpaul | mostly depends on physical location of radio source you want to tune | 14:12 |
kristianpaul | if is moving or not as well.. | 14:12 |
kristianpaul | satellites moves.. so... | 14:13 |
kristianpaul | for example gps requires an active antenna | 14:13 |
wolfspraul | how many different antennas do you have and worked with? | 14:13 |
wolfspraul | and what connectors do they have? | 14:13 |
kristianpaul | well, they have a coaxial-like connector not surewich one exactly | 14:14 |
kristianpaul | _but_ from atennas i had worked, just 3, a gps active atenna with sma | 14:14 |
kristianpaul | gms atenna for receiver, sma | 14:14 |
kristianpaul | and home made yaggi atenna, i used mini coxial, wich is same as sma :-) | 14:14 |
wolfspraul | gms, you mean gsm? | 14:15 |
kristianpaul | oops yes | 14:15 |
wolfspraul | btw, about the app pricing, I am thinking about increasing the price of M1 to 799 USD starting June 1st | 14:15 |
wolfspraul | I want to make room for a lot of things, more high-tech, more power, bigger margins for really good sales partners, more upgrades, etc. | 14:16 |
kristianpaul | good | 14:17 |
wolfspraul | I don't see the need to compete in any race-to-bottom with anyone, it's better to focus on making the product more valuable, and 799 USD is still reachable, even if people have to save a little - so what. that's good discipline. | 14:17 |
kristianpaul | more value. no? :-) | 14:17 |
kristianpaul | product value* | 14:17 |
wolfspraul | I just feel it's better for the future. sales are very slow recently, so the idea is definitely not to increase immediate revenues or anything. | 14:18 |
wolfspraul | but it makes the product and future more attractive, for ourselves and others. and customers don't mind paying for quality, in my experience. if the quality is really there. | 14:18 |
wolfspraul | just came to my mind when you saw the 49 USD app price. why not. if that app is good... :-) | 14:19 |
kristianpaul | i still thinking if buy it ;-) | 14:19 |
wolfspraul | if we do this (799), I will announce on the list etc. and give people the chance to order at the old price for a while. of course. | 14:19 |
kristianpaul | thats goos | 14:19 |
kristianpaul | s/goos/good | 14:19 |
kristianpaul | btw inxluding soem MIDI stuff for 799? | 14:20 |
wolfspraul | same with nanonote to 149 USD. same idea. more updates, better price level for successor Ya, which may even end up at 199 USD. we need roooooom for better tech and better products. | 14:20 |
kristianpaul | s/inxluding/including | 14:20 |
wolfspraul | which one? | 14:20 |
wolfspraul | I definitely try to make each sale a complete sale - guaranteed happiness :-) | 14:20 |
kristianpaul | dont know, a apple appstore 49usd gift for lemur ? ;-) | 14:21 |
kristianpaul | hehe | 14:21 |
wolfspraul | as a next step I want to include a wifi solution in the form of a 703, where you can just setup essid and password on the m1, and then it works | 14:21 |
kristianpaul | 703 looks very straighforward for that | 14:21 |
wolfspraul | wifi is very important to our users, and those thingies cost 16 usd and the problem is solved | 14:21 |
kristianpaul | the only problem will be you still need wires to power the wireless stuff | 14:22 |
wolfspraul | yes and powered over usb, slightly fiddly but we solve the problem and people can enjoy the m1 so much more | 14:22 |
wolfspraul | no, 2 cables. one to usb, one to ethernet. | 14:22 |
kristianpaul | ah yeah | 14:22 |
kristianpaul | imagine ! | 14:22 |
wolfspraul | I use this all the time, works great | 14:22 |
wolfspraul | in fact it works so well, I gave up using some wifi usb dongles and switched to this solution, which is much more stable | 14:23 |
kristianpaul | stable and tiny :-9 | 14:23 |
wolfspraul | antenna performance I'm not so sure, but that's a luxury problem for m1 when the main issue now is "does it have wifi at all?" | 14:24 |
kristianpaul | heeh yeah | 14:30 |
kristianpaul | also nanonote ! | 14:30 |
kristianpaul | btw you you will handle 703n brand, as it is tplink product and you install owrt then sell whole thing as milkymist one? | 14:38 |
wolfspraul | huh? | 14:43 |
wolfspraul | it's just the wifi dongle to include in the m1 box | 14:43 |
wolfspraul | plug in, play (with m1) | 14:43 |
wolfspraul | only that this dongle is not the normal usb dongle, but has a second secret ethernet cable :-) | 14:43 |
DocScrutinizer | wolfspraul: hey, nice to hear this strategy adjustment | 14:43 |
kristianpaul | second secret :) | 14:43 |
wolfspraul | yeah I have to take werner's heat :-) | 14:44 |
wolfspraul | no, but anyway. I am always looking for simple solutions that work, and this one does. | 14:44 |
wolfspraul | then we can go for bigger and better things on m1, and keep focused on what is good for m1. while users can "just use" wifi today. | 14:44 |
DocScrutinizer | wolfspraul: how much is M1 *now*? | 14:45 |
kristianpaul | surei like the solution about wifi | 14:45 |
kristianpaul | and ethernet, hacking POE !! | 14:45 |
kristianpaul | haha | 14:45 |
kristianpaul | guess the secret is ;-) | 14:46 |
wpwrak | slow sales may feel like a slight but they actually don't hurt. better to have future customers who don't know about the product yet than have customers who want it not and find it out of stock. | 14:48 |
wpwrak | (price increase) probably necessary, considering the unit cost. not sure how flexible our customers are. we may well have a segmented customer base with a group that would never buy anything above a USD 100-200 and another group that considers everything below USD 1000 a bargain. | 14:49 |
wpwrak | (nanonote to 149) that hurts, though. the 99 was a nice "round number. | 14:50 |
kristianpaul | Or people wanting to pay for and upgrade kit | 14:50 |
kristianpaul | s/kit/pack | 14:51 |
wpwrak | "integrating" the tp-link sound scary | 14:52 |
wpwrak | i'd rather have a usb wifi dongle. there's something to be said about eliminating the linux prerequisite, but tp-link is still clumsy | 14:54 |
wpwrak | DocScrutinizer: USD 499 | 14:54 |
kristianpaul | wpwrak: scary because the wiring? | 14:55 |
wpwrak | DocScrutinizer: with dreadfully low margins. going through a distributor would basically mean that it would have to be sold (to the distributor) at manufacturing cost | 14:56 |
wpwrak | DocScrutinizer: at least the R&D budget is easy to calculate then, a nice colorless zero :) | 14:56 |
wpwrak | kristianpaul: yes. and where would it be mounted ? inside the M1 ? outside ? | 14:57 |
wpwrak | kristianpaul: would it be permanently attached ? removable ? floating freely | 14:57 |
wpwrak | for me, this means too many moving parts | 14:57 |
kristianpaul | agree but | 14:58 |
kristianpaul | is not this the problem with current or any accesory | 14:58 |
kristianpaul | like kbd mouse and webcam? | 14:58 |
wpwrak | wolfspraul: ah, for M1r4: prepare to resell spare rf keyboards. people will lose them and they may prefer to buy replacements from you instead of invoking their hunter and gatherer instincts | 14:58 |
wpwrak | kristianpaul: is it not ? camera with its own bloody power supply ? c'mon :) | 14:59 |
kristianpaul | sure it is | 14:59 |
kristianpaul | my point is we already suffer from it :-) | 15:00 |
wpwrak | and we're just about to reduce the number of moving parts where it comes to mouse and keyboard :) | 15:00 |
kristianpaul | so... there is no change if wolfgang adds a nother accesor | 15:00 |
wpwrak | i.e., from 2-4 to 1(+1) | 15:00 |
kristianpaul | yes i know and benefir from it :-) | 15:00 |
kristianpaul | for me mr11u could cause less trouble | 15:01 |
wpwrak | because we suck today doesn't mean that we have an excuse for sucking tomorrow, too | 15:01 |
kristianpaul | at least one cable to be used, at least you hack 703 for work as POE | 15:01 |
kristianpaul | s/as/by | 15:01 |
kristianpaul | wpwrak: sure not, just kidding ;-) | 15:01 |
wpwrak | does the tp-link fit inside the M1 case ? | 15:02 |
kristianpaul | may be the board | 15:02 |
wpwrak | heh. on the extension header ? ;-)) | 15:03 |
kristianpaul | at least you sit it over the fpga with case | 15:03 |
kristianpaul | lol | 15:03 |
roh | wpwrak: http://shop.8devices.com/wifi4things/carambola | 15:03 |
roh | thats basically a openwrt router on a small pcb to be embedded | 15:04 |
wpwrak | cute :) | 15:04 |
wpwrak | that looks integration-friendlier than the tp-link | 15:04 |
roh | i think that was the idea | 15:04 |
roh | they got openwrt in a github tree https://github.com/8devices/carambola and a wiki http://wiki.carambola.cc/ | 15:06 |
DocScrutinizer | wpwrak: wolfspraul: for this kind of changing pricetag, you _need_ to rename the product | 15:06 |
DocScrutinizer | call it M2 | 15:06 |
DocScrutinizer | M1i maybe | 15:06 |
kristianpaul | M1 plus :-) | 15:06 |
wpwrak | hmm. makes me wonder how hard it would be to simply use the RT3050 ... | 15:06 |
roh | my guess is their next product is citramona https://github.com/8devices/carambola/commit/3bac0167868f3c89d720de18952e9e7b7d9bf958 | 15:07 |
roh | wpwrak: more sucky to source and do rf routing for ;) | 15:07 |
kristianpaul | had you consider the effetcs of putting a 2.4Ghz devices next to a fpga? | 15:08 |
wpwrak | roh: been there. done that ;-) | 15:08 |
Action: DocScrutinizer _adores_ qi-bot sed-regex X-P | 15:08 | |
DocScrutinizer | >>"at lebyt one cable to be used, at lebyt you hack<< WTF?! | 15:09 |
Action: roh still doesnt get what internal wifi should bring | 15:09 | |
kristianpaul | roh: wifi it self ! | 15:09 |
kristianpaul | that we currently lack and miss (at least i do).. | 15:09 |
kristianpaul | for OSC will be cool, have it out of the box | 15:09 |
roh | i mean.. its a 500E dedicated purpose device. people will just connect it to the network they already have. | 15:09 |
wpwrak | kristianpaul: (RF vs. fpga) i'm kinda curious about that myself. we'll see when M1r4 is made :) | 15:10 |
roh | i'd rather like to drive the price down. maybe then i could sell some | 15:10 |
wpwrak | roh: "the network" is wifi nowadays | 15:10 |
kristianpaul | indeed | 15:11 |
roh | wpwrak: bullshit. | 15:11 |
wpwrak | roh: open your eyes, look around :) | 15:11 |
roh | wpwrak: you obviously never tried running an event using 2.4 ghz. doesnt work. | 15:11 |
DocScrutinizer | netwok problems is WLAN nowadays | 15:11 |
kristianpaul | roh: jamed? | 15:11 |
kristianpaul | jammed* | 15:11 |
roh | ever idiot and his smartphone blocking frequencies or pounding your net. so either one uses 5ghz or not. | 15:12 |
wpwrak | roh: oh, i didn't say that it scales at will. but it's the default choice. you move to more complex things as needed. | 15:12 |
roh | for devices with wires on it nonetheless.. wired ethernet is the only thing thats reliable enough for 'event' stuff. | 15:12 |
DocScrutinizer | use cable as long as there is any | 15:12 |
roh | wpwrak: from my pov: wifi is a waste of money and doesnt help me find customers. | 15:13 |
DocScrutinizer | indeed I can't see any VJ using WLAN and hoping for the stability needed for the primary pirpose of M1 | 15:13 |
wpwrak | the RT3050 is amazing. it's pretty much the entire system. doesn't even need external memory. | 15:13 |
roh | DocScrutinizer: exactly. and if the club has a network and people use it to control stuff, maybe even via wifi, they got accesspoints already and dont need any inside the mm1 | 15:14 |
DocScrutinizer | how about integrating a laserpointer to M1? At least as relevant for VJ as WLAN is | 15:14 |
roh | DocScrutinizer: huh? whatfor? | 15:15 |
DocScrutinizer | >/sarcasm> | 15:15 |
kristianpaul | DocScrutinizer: use cable , indeed | 15:15 |
roh | hrhr | 15:15 |
wpwrak | roh: to eliminate interfering WLAN transmitters :) | 15:15 |
kristianpaul | then use 433Mhz ;-) | 15:16 |
wpwrak | consider people who want to use their iGadget as a control surface. like it or not, they all communicate via wlan. | 15:16 |
wpwrak | now, as you move to increasingly serious users, you'll find better solutions | 15:17 |
DocScrutinizer | errr what? | 15:17 |
wpwrak | but wlan is the point of entry | 15:18 |
DocScrutinizer | a VJ using his *iPHone* to control M1 during a live show? HAHA | 15:18 |
wpwrak | maybe iPad. e.g., lemur moved from their own "pad" hw to becoming an app. | 15:18 |
DocScrutinizer | yeah sure | 15:19 |
wpwrak | maybe it doesn't work great. but the pad competition was strong enough to drive their own hw out of the market | 15:19 |
DocScrutinizer | and I'll control my heart pacemaker via 433MHz now | 15:19 |
DocScrutinizer | or maybe via web interface, with public IP | 15:20 |
DocScrutinizer | via dynDNS | 15:20 |
kristianpaul | he | 15:20 |
kristianpaul | ipv6 ;-) | 15:20 |
DocScrutinizer | guess why *none* of the wireless microphone sets used on stage or in TV is via WLAN | 15:22 |
wpwrak | DocScrutinizer: and accessible via your facebook account ;-) | 15:22 |
wpwrak | microphones have other delay requirements than control surfaces | 15:22 |
DocScrutinizer | they *all* use very special dedicated frequency range, usually redundant twin channel | 15:22 |
DocScrutinizer | wpwrak: nope, that's definitely not the reason | 15:22 |
wpwrak | and in the M1, you normally don't need lightning-fast responses | 15:23 |
DocScrutinizer | it's all about reliability requirements vs anticipated interference from other systems | 15:23 |
wpwrak | effects are generated by recursion anyway, so you always have a lag | 15:23 |
DocScrutinizer | WLAN is a *hell* in that regard | 15:23 |
kristianpaul | so what you suguest? | 15:24 |
DocScrutinizer | use cable | 15:24 |
DocScrutinizer | cables have very few interferences, system imanent property | 15:24 |
DocScrutinizer | compared to duh WLAN | 15:25 |
roh | i think wifi should be an option, which is not built in. | 15:25 |
wolfspraul | DocScrutinizer: renaming costs money, won't do. Why do you think it's beneficiary? | 15:25 |
DocScrutinizer | USB dongle, all fine | 15:25 |
roh | support 'access via ip' and 'ethernet' properly and wifi is nothing of your concern. | 15:26 |
wolfspraul | chinese companies change chips without changing the name (really), I change price without changing the name :-) | 15:26 |
DocScrutinizer | wolfspraul: to all I learnt so far customers simply get pissed when stuff gets more expensive | 15:26 |
roh | start adding wifi and you got fcc shit on your hands again. and all the range discussion. let somebody else debug that problems ;) | 15:26 |
wolfspraul | not here. sales are slow, people are not waiting in line. and the announcement will be in advance. | 15:27 |
DocScrutinizer | even if the new higher price was just justified and fine, they will refuse to buy at new price if they know it's been available for less | 15:27 |
kristianpaul | and milkymist still liek a private group.. now 799 for enter ;-) | 15:27 |
wolfspraul | not worried about that | 15:27 |
kristianpaul | s/like/look | 15:27 |
wolfspraul | they should have bought earlier :-) | 15:27 |
wpwrak | roh: "our patent-pending proprietary XXS-range technology protects against interferences" ;-) | 15:27 |
kristianpaul | lol | 15:28 |
DocScrutinizer | wolfspraul: at very least officially announce fade out of subsidized introductory pricing | 15:28 |
wolfspraul | yes | 15:28 |
wolfspraul | definitely | 15:29 |
wpwrak | DocScrutinizer: i think the increase would be for M1r4 and beyond. that one has a number of improvements. | 15:29 |
DocScrutinizer | wpwrak: that's why I sugeested a rename to at very least M1+ | 15:29 |
kristianpaul | current stock is it 10 -15 kits? for M1 | 15:29 |
wpwrak | heh :) | 15:29 |
wolfspraul | something like that | 15:30 |
roh | i still got 2 spare case kits left | 15:30 |
wolfspraul | sales in April: 0 | 15:30 |
roh | one in violett, one in uv-red | 15:30 |
wpwrak | i'm not opposed to a name change per se. we have few enough things that say "M1". most say "milkymist". let the ministry of truth, historical accuracy, and bureaucratic redundancy take care of the IRC logs :) | 15:30 |
wolfspraul | I won't change any name, no need | 15:30 |
DocScrutinizer | no sane company will sell a significantly improved product without changing the name to make everybody aware -esp not when this improvement is associated to an increase in price | 15:30 |
wpwrak | no change is fine with me, too :) | 15:31 |
wolfspraul | sometimes I just say 'milkymist' now for everything, which is still fine | 15:31 |
roh | as long as you change any revision or so i think its fine | 15:31 |
wpwrak | DocScrutinizer: think of it as sheik buying a "rolls royce". you have your underlings to worry about mundane things like the price :) | 15:32 |
DocScrutinizer | "Wolfgang Spraul annouced the new M1 last Thursday..." .oO(what a BS, it's not new at all) <skips article> | 15:33 |
DocScrutinizer | "Wolfgang Spraul annouced the new M1+ last Thursday..." .oO(OOOH M1 PLUS??) | 15:33 |
wpwrak | DocScrutinizer: about 99.99999% of the world population will not know about the old M1 anyway :) | 15:34 |
DocScrutinizer | :shrug: | 15:34 |
DocScrutinizer | just saying, you're free to ignore me | 15:34 |
wpwrak | i don't think it's a big deal. the device is niche enough that communication paths are short. and there's little risk of someone selling M1r3s as M1r4s | 15:36 |
wpwrak | besides, you can tell by a fairly large number of visual differences | 15:37 |
wpwrak | for identification purposes, it could be nice to also change some elements of case color. but i'm not sure there would be much enthusiasm about that idea. | 15:38 |
kristianpaul | announce will take place in ML... | 15:38 |
kristianpaul | so | 15:38 |
kristianpaul | or wolfspraul have other plans? i guess not | 15:39 |
DocScrutinizer | anyway you probably won't like headlines like "Wolfgang Spraul annouced the new pricetag for M1 last Thursday... Up by 80%" | 15:39 |
kristianpaul | lol | 15:39 |
wpwrak | DocScrutinizer: how about "wolfgang announces the new model of the Milkymist 1 series [with enhanced features. price is USD 799." | 15:40 |
kristianpaul | DocScrutinizer: get yours M1 soon at least ;-) ! | 15:40 |
wpwrak | s/features/features] | 15:40 |
DocScrutinizer | which is more in line with "Wolfgang Spraul annouced the new M1+ last Thursday..." | 15:41 |
wpwrak | "remaining stock of the previous model will be sold at USD 499." | 15:41 |
DocScrutinizer | "internally we call it M1+ | 15:41 |
DocScrutinizer | " | 15:41 |
DocScrutinizer | exactly | 15:41 |
DocScrutinizer | just s/[with enhanced features]/[with _massively_ enhanced features] | 15:42 |
wpwrak | yup. i think it's important to point out the improvement. for two reasons: 1) it provides a reason for the increase. 2) it tells you what your money is used for. most people don't understand R&D cost. | 15:42 |
wpwrak | it's not a MASSIVE enhancement. at least not on the hw side. but it's better. and it took some work. so .. :) | 15:43 |
DocScrutinizer | "since we like the name so much, the official branding remains 'Milkymist 1' while internally we use to call it M1+..." | 15:43 |
wpwrak | for the massive enhancement, i'd still like an integrated pad :) | 15:43 |
wpwrak | naw, you don't need to explain the naming | 15:43 |
wpwrak | consider BMW, series 3 | 15:44 |
wpwrak | or benz, C-class | 15:44 |
wpwrak | well-established precedent :) | 15:44 |
DocScrutinizer | anyway, I guess I somewhat made my point. Dunno if it's relevant and based on good facts | 15:46 |
DocScrutinizer | wpwrak: neither bmw nor benz ever increased pricetag of successor model by such a hurting smash | 15:47 |
DocScrutinizer | at least not unless the comlete design changed so much you'd rather ask "that's WHAT? a S500?" | 15:48 |
DocScrutinizer | of course if already the product photo makes it utterly clear there's a _huge_ difference, you won't go and compare price to a model that simply had same name by any weird incidence, 5 years ago | 15:49 |
wpwrak | you mean we should add fins ? ;-) | 15:50 |
DocScrutinizer | fins are fine ;-D | 15:51 |
DocScrutinizer | plus some fancy blinkenlights | 15:51 |
wpwrak | oh, we have these :) | 15:51 |
DocScrutinizer | duh, we already got those? good! :-) | 15:52 |
DocScrutinizer | make sure they're lit for all product photos | 15:52 |
wpwrak | LED matrix powered by a 24 channel PWM ;-) | 15:52 |
DocScrutinizer | I know ;-P | 15:52 |
wpwrak | channel separation 30 dB or better :) | 15:52 |
DocScrutinizer | wouldn't hurt the product photo would show blinkenlight matrix display "M1+" ;-) | 15:53 |
DocScrutinizer | no need to rename anything then | 15:53 |
wpwrak | naw, the leds are on the edge of the board. they don't form a physical matrix. | 15:54 |
DocScrutinizer | too bad | 15:54 |
wpwrak | of course, we could perhaps show "M1+" in morse code .... :) | 15:54 |
DocScrutinizer | hard to transport on a photo | 15:54 |
wpwrak | particularly with variable spacing between LEDs :) | 15:55 |
wpwrak | well, non-uniform | 15:55 |
Action: DocScrutinizer idly recalls he still has "decode long SMS morse beep of Nokia" on his long term ToDo | 15:55 | |
DocScrutinizer | you could add a 7*18 matrix under the transparent cover, just for the product photos. footnote: "matrix display optional" | 15:57 |
DocScrutinizer | roh: I gather some of the next days there's "linux-Tag" in Berlin? I pondered to maybe visit it and just visit you and Berlin same occasion | 15:59 |
DocScrutinizer | and c-base :-) | 15:59 |
wpwrak | heh ;-) i almost have that already: http://downloads.qi-hardware.com/people/werner/m1/leds/ledm-in-m1.jpg | 15:59 |
roh | i dont have a clue. i ignore linuxtag the last 10 years or so now | 15:59 |
DocScrutinizer | roh: ooh, well then maybe I should ignore it too | 15:59 |
roh | i dunno.. it just was to boring at one point and i never revisited | 16:00 |
DocScrutinizer | I pondered visiting goldelico/Nikolaus | 17:22 |
wpwrak | DocScrutinizer: never pictured you as an apple fan ;-) | 17:29 |
DocScrutinizer | apple fan? | 17:38 |
wpwrak | golden delicious is an apple :) | 17:50 |
whitequark | wpwrak: this? http://cdn.coolest-gadgets.com/wp-content/uploads/usb-apple-fan-novelty-gadget-0.jpg | 17:58 |
whitequark | I wonder how exactly did you fit the image of DocScrutinizer with that gadget... | 17:58 |
wpwrak | naw, very different context ;-) | 18:09 |
wpwrak | these guys: http://www.golden-delicious.de/ | 18:10 |
LunaVorax | Hello everyone! | 20:58 |
JCGsp | Hi everyone | 21:15 |
tuxbrain | woflspraul are you there? | 21:25 |
tuxbrain | wolfspraul: knock knock | 21:26 |
tuxbrain | by the way hello to every one alive on the channel :) | 21:27 |
tuxbrain | oh man , wolfspraul you will make me to configure my jabber account? | 21:28 |
JCGsp | anybody uses Debian on Nanonote? | 21:29 |
wpwrak | damn. i should have paid more attention to the movies. how to thwart a zombie attack ? | 22:22 |
Ayla | lawn mower | 22:23 |
Ayla | Peter Jackson style | 22:23 |
wpwrak | JCGsp: some people do. but it seems to involve a lot of suffering. | 22:24 |
wpwrak | JCGsp: 1) because it's slow (not sure if in general or just boot time). 2) because they represent about 70% of all cases of bricked bens. fortunately, one can recover from 2). | 22:25 |
wpwrak | Ayla: hmm. i found out that today's zombie is feeding on cat at the moment. so maybe we're safe for now. | 22:30 |
Ayla | be careful of cat zombies | 22:31 |
wpwrak | oh, with nine lives, their undead state can't be too different from their after-death states | 22:32 |
wpwrak | i'd be worried if i was a rat, though. it's bad enough that they're apex predators for most practical purposes. then add zombie ... | 22:33 |
Ayla | human zombies don't eat vegetables, they eat humans; so I believe zombie cats eat cats, and zombie rats eat rats | 22:35 |
wpwrak | yes, that would make sense. so if you're, say, attacked by a tiger, just zombify it and you're safe. | 22:38 |
tuxbrain | I didn't thought before zombify is a right solution to anything... but dude you have make me rethink this with more detail | 22:41 |
wpwrak | alas, it wouldn't work with bankers. it would only change their craving from your blood to your brains ... | 22:42 |
Ayla | wpwrak: zombifying tigers would save them from extinction | 22:45 |
Ayla | (unless those zombies can die of hunger, which occurs only on bad movies) | 22:46 |
wpwrak | Ayla: a most excellent point. i think you should bring this up with the WWF. | 22:46 |
wpwrak | in fact, carpet bombing their habitats would already be sufficient to protect them from getting killed by poachers. | 22:48 |
Ayla | good idea | 22:50 |
Ayla | let's take the best of both: | 22:51 |
Ayla | bomb the poachers, and zombify the tigers | 22:51 |
wpwrak | maybe zombify baby tigers. then teach them from early on to eat poacher brains, instead of tiger brains. | 22:52 |
qi-bot | [commit] Paul Cercueil: MIPS: JZ4740: LCD: Fixed LCD_CMD consts and completed LCD_STATE consts (jz-3.3) http://qi-hw.com/p/qi-kernel/b3353d6 | 22:53 |
qi-bot | [commit] Maarten ter Huurne: MIPS: JZ4740: SLCD: In TV-out mode, add a black line on top and bottom (jz-3.3) http://qi-hw.com/p/qi-kernel/1264672 | 22:53 |
Ayla | nice, I commited something without knowing it! | 22:53 |
mth | that commit contained only changes from your patch, so I set you as the author | 22:54 |
wpwrak | tuxbrain: well well, things here are somewhat quiet these days. nanonote is in hibernation. M1 is moving forward, though, with another round of hardware in the making | 22:54 |
JCGsp | wpwrak: I have zero problems with Debian, just that I can't start X session. Either with Ratpoison or another WM | 22:55 |
Ayla | I didn't know it's possible to assing a commit to another user | 22:55 |
wpwrak | JCGsp: does the X server log its complains in /var/log/X* ? | 22:56 |
tuxbrain | wpwrak: no need to connect to chat to feel the hybernation :P | 22:57 |
wpwrak | tuxbrain: oh, the mailing lists are very quiet. things have moved to a greater extent to IRC. | 23:00 |
JCGsp | wpwrak: Nope, file is empty | 23:01 |
tuxbrain | wpwrak: I refering to sales but yes mailing list activity is also valid :P | 23:01 |
wpwrak | JCGsp: empty is strange. there should be no file at all or a file with lots of things (the latter case, if the X server is xorg) | 23:04 |
JCGsp | wpwrak: I Think so, I'll try to reinstall X again | 23:06 |
JCGsp | wpwrak: It gave an error | 23:10 |
mth | Ayla: git commit --author='name' | 23:13 |
Ayla | and --email too? | 23:14 |
mth | name can include email addr | 23:14 |
Ayla | ok | 23:15 |
--- Thu May 3 2012 | 00:00 |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!