#qi-hardware IRC log for Tuesday, 2011-01-11

wpwrakkristianpaul: well, i guess that proves the point that there are things much worse than hell ;-)00:48
wpwrakzrafa: oooh ... no povray in jlime. pity.08:08
zrafawpwrak: is the license of povray something okey for qi?.. it was free but no for business?08:19
zrafawpwrak: anyway it should not have problems any kind right?08:19
zrafaif so, I can upload :)08:19
wpwrakzrafa: ah, right, i forgot about that weird license. don't know really if that's considered okay or not. probably okay.08:20
wpwrak(someone really ought to write a free alternative. a non-commercial license is highly ridiculous in this day and age)08:21
zrafawpwrak: if you are okey I will upload povray 3.6.1-r2 to extra packages ;)08:54
wpwraklemme check the license09:02
czrimo povray code contains a lot of legacy contributions and the licensing stuff came kind of late into it09:08
czrwhich might explain the extra weirdness09:08
czror rather afair, not imo.09:08
wpwraki think this serves nicely as an example for how not to write your license: http://www.povray.org/distribution-license.html09:09
kristianpaulyes09:09
Action: czr nods09:10
kristianpaulin cad/cam related, you can also find thos kind of lincenses09:10
kristianpauli think openscad uses one, but not body cares..09:10
kristianpauluses one lib with no nice license09:11
wpwrak2.1: does jlime constitute "a generally recognised Distribution" ? they use the term and give examples but fail to define the term.09:11
wpwrak2.4.a directly contradicts 3.309:11
wpwrak3.2: what is "other consideration of any type" ? material favours ? gratitude ? praise ?09:13
kristianpaulgratitude?09:14
kristianpaulBtw simd instructions on amd-64-like procesor supports 8 bits data type?09:15
kristianpauli guess 16 bits should be fine, but i will bloat the sample even MORE09:15
wpwrak3.1.b makes a strong assumption about technology and its use in a specific setting (not relevant to our scenario, but dubious anyway)09:16
wpwrakthere is also no clear indication of whether distribution of pre-compiled binaries is allowed09:19
wpwrak(or i haven't found it)09:20
wpwrak4.5. is slightly ambiguous. also implies a maintenance obligation.09:22
wpwrakgenerally, i have a low opinion of people who insist on non-commercial licenses or licenses with other kinds of ideological etc. restrictions. this fine example does little to change my opinion.09:23
wpwrakalso, the distribution license may actually not apply, e.g., in case there are patches. let's see what the modification license has to say ...09:24
wpwrakinteresting, in http://www.povray.org/source-license.html, 2.2.f disallows bug fixes09:29
wpwrakat least the modification license implicitly allows distribution of binaries09:39
wpwrak3.2 of the modification license if confusing, too. does it mean that the terms of the distribution license applies in addition to the modification license or in parallel ?09:43
wpwrakhttp://en.wikipedia.org/wiki/POV-Ray has an encouraging-sounding "A complete rewrite of POV-Ray ("POV-Ray 4.0") is currently under discussion, which would use a more liberal license, most likely GPL v3.", but then the article this points to is from september 2007. more than three years and still only "under discussion" ?10:00
wpwrakhmm, http://www.renderpixie.com/ doesn't look bad. LGPL and uses renderman for scene descriptions10:17
lekernellol: I just found people selling graphite paint for the EMC ...of houses10:19
lekernelfor those "electrosensitive" people10:19
wpwraklekernel: nice business idea ;-)))10:20
wpwraklekernel: and i suppose their "feinstaub" (respirable dust) sensitive neighbours can the sue them ...10:23
lekernellol, I just got a spam from an ARM partner who invites me to a seminar to "help me understand the benefits of embedded Linux". are they trolling me or...?11:39
wpwraklekernel: i guess their oracle told them that you're not entirely convinced yet :)13:08
wpwraklekernel: so h:D is the only "official" distributor for the MM1 ? (and indeed, sharism.cc don't have it. i thought wolfgang once mentioned something like only USD 350)13:16
prpplagueanyone know if the kicad pcb file has been made available yet for the nano note?14:28
lekernelwpwrak: for now... but others are coming14:31
wpwraksome good ... some quite awful :-(  http://downloads.qi-hardware.com/people/werner/wpan/tst-cw/14:58
wpwraklekernel: (others) ah, good. the detour via europe would make it something like 40% more expensive for non-EU customers.15:00
wpwrakprpplague: afaik, there are no kicad pcb files for the nanonote. the design was done with some proprietary EDA program.15:00
lekernelwell you can get them straight from sharism as well if it makes more sense to you (just mail Wolfgang)... btw the price Wolfgang talked about is without taxes and shipping15:01
wpwrakprpplague: the kicad schematics that exist of the ben are (manual) copies of the proprietary original15:01
wpwraklekernel: (S&T) yup, that's why i wrote "non-EU".15:04
lekernelwe're totally open to new distributors, you could even open/recommend one in Argentina if you want :)15:05
wpwraklekernel: zrafa seems to know some folks who work with this kind of stuff. i guess he'll ping them. you never know what may crawl out of this.15:06
wolfspraulwpwrak: this will all improve a lot over the next few months.15:09
wolfspraulI think I can be optimistic that a number of people will help, in the US, Europe, India, maybe Japan. We see.15:10
wolfspraulit's just a lot of work and more details need to come together, but right now the hackable-devices shop is great!15:10
wpwrakyeah, looks good. i was just surprised that you "advertized" it here as being available from sharism, but it's not yet visible in the sharism shop15:13
wolfspraulit's not so easy to start it.15:15
wolfspraulif we make a real launch now, there will be 100 orders and they cannot be fulfilled.15:15
wolfspraulso it's a mix of things. I am very happy that some people start selling, and slowly the stuff trickles out.15:15
wolfspraulin parallel I can work on the case, certification, pictures, box, etc.15:15
wolfspraulthis is only the beginning :-)15:16
wolfspraulthe Ben NanoNote launch was also messy, but you weren't closely following back then yet15:16
wpwrakokay, so you're ramping up google-style :)15:17
wolfspraulit's a mix of priorities15:17
wolfspraulon one hand we are selling early boards15:17
wolfspraulon the other hand we need to start to jump into the cold water and give it to people who give feedback15:18
wpwraksure. ramping up is good. i was even surprised that you already had enough cases to offer them for sale15:22
wolfspraulwpwrak: so those tst-cw pictures show that atusd creates a lot of noise on neighboring channels, unlike atusb?15:22
wpwrakyes, there's a very ugly amount of noise15:22
wolfspraulinteresting15:22
wolfsprauldid you see this before on atusd, or was it introduced at some point?15:23
wpwrakunfortunately, i don't have the same kind of data from older atusd boards, so i don't know if it's something new i introduced or if they were always like this15:23
wpwraki can do this specific test only with the latest boards15:23
wpwrakthe older ones (with the old chip) would have needed an extra signal to enter this test mode.15:24
wpwrakadded some more data from older boards. the usb line looks pretty consistent.15:30
prpplaguewpwrak:  thanks, i thought i had read that the project had switched to kicad15:35
wpwrakprpplague: many new things are using kicad15:36
prpplagueyes indeed15:36
Action: prpplague is actually trying to migrate from altium to kicad hence his question15:37
prpplaguewpwrak: was wanting to grab a few reference schematics to have a look at15:37
prpplaguewpwrak: got any urls for other projects using kicad?15:38
wpwrakprpplague: if what you need is a first impression, perhaps ben-wpan or xue could be useful. they have schematics and a layout too. they're simpler than the ben, though. xue a bit, ben-wpan a lot.15:38
wpwrakprpplague: here's an overview of the ones with kicad schematics: http://projects.qi-hardware.com/schhist/15:39
kristianpaulcase !15:39
prpplaguewpwrak: naw, i've done three small boards so far with kicad, was looking for some more compilicated multi sheet schematics to get an idea of normal workflow15:39
wpwrakprpplague: the first seven should all also have a layout15:40
prpplaguewpwrak: thanks, i'll have a look15:40
wpwrakhmm, the smeared spectrum also shows in "real" transmissions, just not as clearly. the difference in duty cycle (atusd currenly has a faster turn-around time than atusb) makes it a bit difficult to produce things that can be easily compared. i think i need to add a local hammering mode to atusb and get rid of the repeated frame download in both cases.16:05
wpwrakwell, the hammering can wait for atusd to atusd comparison16:06
wpwrakand i've already eliminated the frame download in this mode. hah, that's planning ahead ! ;-)16:07
kristianpaul"Remember that you need to use as much information as you have => use"16:49
kristianpaulthe magnitude too. Best GNSS front ends are 4 or 8 bits. It is true16:49
kristianpaulthat there is not so much advantage to use more than 4 bits but the16:49
kristianpaulmore you have the better, especially for indoor applications.16:49
kristianpaul                                                ^^^^^16:50
kristianpaul                                          ^^^^^16:50
kristianpaulwpwrak: so, resuming more bits the better :'-)16:51
kristianpauland tought that i was a about to kill the magnitude :/16:52
wpwrakit's half a bit of extra data ;-)16:52
wpwrakdid anyone explain what to do with -0 vs. +0 ?16:53
wpwraki.e., are they both 0 or maybe -0.5 and +0.5 ?16:53
wpwrak(and -1/+1 would then be -1.5 and +1.5 respectively)16:53
wpwrakor scale according to your liking ;-)16:54
kristianpauli think it depends of the resolution of your frontend16:54
kristianpaulless meanfull will if you have more resolutions bits16:55
wpwraki mean: how to interpret the data. are "-0"/"+0" the same ? or is "-0" < "+0" ?16:55
wpwrakin the latter case, is "+1" - "+0" = "+0" - "-0" = "-0" - "-1" ?16:56
kristianpaul "-0" < "+0" sounds logic17:02
kristianpaul= i will not said, may be just an aproximate17:03
wpwrakperhaps you can ask the person(s) who gave you the advice you quoted above17:04
kristianpaulsure17:04
kristianpauli'll NO17:04
kristianpaulNOW17:04
wpwrakand if the answer is "it depends" -> sige :)17:04
wpwrakkNOW ;-)17:04
kristianpaul"it depends" -> sige = for sure 2 bits adc is not smoething you can trust that much17:04
kristianpaulwell the AGC and nother analog steps in that chip dedeserve a look17:05
kristianpauldeserve*17:05
wpwrakhuh, why would a 2 bit ADC less trustworthy when it comes to the meaning of the values ?17:05
kristianpaulless  trustworthy for -0 i mean17:06
wpwrakyes, but why ?17:06
wpwrakit's a question of definition17:06
kristianpauldont knwo let me ask this guy (fabrizio)17:06
wpwrakif -0 and +0 are supposed to be different codepoints, when saying that you can't trust the difference is the same as saying that you can't trust the distinction between, say, +0 and +1.17:07
kristianpaulno, well may be, cause +1 -1 is a signal peak, but values nears to 0..17:08
kristianpaulhmm17:09
wpwrakthe basic question is: if an ADC produces this kind of data stream, what would be the "standard" interpretation. it could be that you really have -1, 0, +1, with the sign of 0 being noise, or17:09
wpwrakit could be that there is information in the sign of 0, or it could be that this format is extremely unusual and there's no clear expectation.17:10
Action: kristianpaul copying and pasting questions from wpwrak 17:11
wpwrak:)17:14
kristianpaulsent17:15
kristianpaulgota run home, read you later !17:16
wpwrakhappy running ! :)17:16
wpwrakwell, i'll toy with my spectrum a bit later, when the grumbling of the nearby wlans has ceased ...17:24
zrafawpwrak: no povray then?19:52
wpwrakzrafa: it seems to be difficult ...19:55
zrafawpwrak: ah okey. Anyway, the packages built fine so if some time we need those no extra work needed22:48
--- Wed Jan 12 201100:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!