#milkymist IRC log for Friday, 2011-05-20

kristianpaulwow finally !00:19
kristianpaullekernel:i tought it was not acepted00:20
terpstralekernel, last names for Gaetan and Bleno?08:49
lekernelwolfspraul: I don't remember if we decided to engrave a milkymist logo on the cases?10:30
lekernelthinking more about it, I think we should have one on the top10:31
wolfspraulhe :-)10:33
wolfspraulyes I fully agree - the problem is execution10:33
wolfspraulyou and roh are probably best suited to make it happen10:33
wolfspraulno need to send samples to me or others10:33
wolfspraulif you and roh decide to go ahead with this, my part would be in two areas:10:34
wolfspraul1) I am willing to share some of the costs, but we need to keep it strictly under control, I mostly can compensate some material or small costs10:34
wolfspraul2) if we do go ahead with it, I will wait a little longer with another round of picture taking10:34
wolfsprauluntil you can send me one of the final, engraved, top sides so that the photos show the final version10:35
wolfspraulwhat do you think?10:35
lekernelwe can have a discrete logo, no need to take another round of pictures...10:36
wolfsprauldiscrete logo?10:36
wolfspraulnot 'another round' - I planned to take better pictures with all the feedback from the last shooting anyway10:37
lekernelsorry, wrong word. I mean, not prominent10:37
wolfspraulbut if an engraved top is coming, I'll wait for that first10:37
wolfspraulI'm just explaining my part in this endeavor10:37
wolfspraulthe main person to pull it off is roh10:37
lekernelthe current pictures are ok, I think. what more do you want?10:37
wolfspraulprobably needs some experiments into what looks good, since looking good is the main point10:37
wolfspraulyes they are, that's why the second round has not yet happened :-)10:38
wolfspraulI keep a 'known issues' list somewhere10:38
wolfspraulfrom dirty pcb (inside the case), to hand-written label on vga connector, to depth issues, to what not10:38
wolfspraulso if we are adding a logo on the top acrylic, that would be one more item on that list10:39
wolfspraulI think a good looking logo would be great, definitely10:41
lekernelthe cases for rc3 aren't made yet, right?10:44
wolfspraulthey are 'in the making'10:44
wolfspraulI made a down-payment, roh has already sourced some material.10:44
wolfspraulmaybe he is already working on things like this shielded metal plate, the buttons, etc.10:45
wolfspraulthe lasering hasn't happened yet10:45
wolfspraulactually, good that you are there - the one open item roh and I still had were labels for the 3 buttons10:45
wolfspraulright now only the leds have labels, the buttons are unlabeled10:46
wolfspraulI'm sure most people will think that render/on/standby are the labels for the buttons10:46
wolfsprauldo you have a proposal for the button labels?10:46
wolfspraulI was thinking about L ON R for left, on, right10:46
lekernelyeah, L ON R10:55
lekernelwe agreed on that already, no?10:55
wolfspraulforgot :-)10:59
wolfspraulbut now it's done10:59
lekernelhow's the outer box going btw?11:00
wolfsprauldon't know, some work was started but I need to take a look11:01
lekernelwho's doing it? Yi?11:03
wolfspraulI don't think the box is on the critical path11:04
wolfspraulthe one thing we need to discuss is the design of the box11:05
wolfspraulblack or white? unfortunately I know very little about retail box design, too bad. so unless someone else knows we have to guess or just do what others are doing :-)11:06
lekernelour pictures are white ...11:06
terpstralekernel, poke11:12
terpstrai need last names for Gaetan and Bleno, please :)11:12
lekernelterpstra: I answered you by mail and /query11:13
terpstrasorry =)11:13
terpstraso just 2 people?11:14
xiangfulekernel: Hi here is the patch of flash.c: http://dpaste.com/544627/11:14
xiangfubase on your comments. thanks for comments. :)11:14
lekerneltold you, traveling to Darmstadt is a major difficulty for most people, especially in the morning =]11:14
terpstrameh ;)11:15
terpstraparis is only like 6 hours from here, no?11:15
terpstra5 horus even11:15
lekerneldon't tell that to me ;-)11:15
terpstraso to be here at 10:00am, leave at 4:30am to be safe11:16
terpstrameans going to bed at 9-10pm. quite reasonable :)11:16
lekernelwe'll be departing from Metz anyway. so even closer :)11:16
terpstra2 hours!11:16
terpstraand you can go as fast as you like since it's within germany11:17
terpstraso ... like 1 hour with sufficiently fast car ;)11:17
lekernelah, those Germans :)11:17
lekernelI don't drive a car that goes this fast ;)11:17
terpstrayou cross over via Saarbruecken?11:18
lekerneldon't know... I'm just planning to use the GPS *g*11:18
terpstrai'd forgotten how all those roads in france cost money to drive on11:19
lekernelbut yeah, i'll probably go through Saarbrueken11:19
lekernelyeah it's horrible11:19
terpstradon't you people pay taxes?11:19
lekernelthose shit roads double my travel costs every time I'm driving in france11:19
lekerneloh, it's coming in Germany too11:20
lekernelyou know ... privatizing of roads11:20
wolfspraulthen you'll pass by my hometown Homburg, about 30km past Saarbruecken...11:20
lekernelright now they collect money from truck drivers only11:20
terpstrayou lie11:20
terpstrai asked my colleagues11:20
terpstrano paying for roads!11:21
terpstrajust like no speed limits11:21
lekernelstill the motorways are still being privatized (as they are in France) and those companies collect money from trucks atm11:21
wolfspraulnot yet, luckily they uber-engineered the toll collecting system so badly that they have trouble scaling it to the number of cars (it's for trucks only now)11:21
wolfspraulbut unfortunately Moore's law will help them, so I guess eventually no matter how inefficient and bloated their collecting system is, it will work for all 40 million cars in Germany :-)11:22
wolfsprauland once that's the case, I'd say at most 2 years later they will start charging11:22
terpstrameine kollege sind sehr skeptisch11:23
wolfspraullekernel: do you have any ideas or drafts/sketches for box design?11:24
wolfsprauljust so we don't do something that you will immediately dismiss :-)11:24
lekernelterpstra: there even was an article in the press (that I can't find atm but I'll email it to you when I do) about increasing accidents on the hamburg/bremen motorway due to construction works done wrong by the company to maximize their ROI11:25
terpstrai can believe that11:26
terpstrabut i don't think germans will put up with tolls on roads11:26
terpstrathere would be an uproar just like if they tried to put speed limits on all the roads11:27
lekernelwolfspraul: i'm thinking about some nokia boxes11:27
lekernelthis kind of thing11:28
wolfspraulI think we should use the chance to explain our product a little11:31
wolfspraulone rule I vaguely remember from the pros is that you always at most talk about 3 things/features on the box11:31
wolfspraulso there will be a product shot, from the ones we have already11:33
wolfspraulthen maybe some screenshots?11:33
wolfsprauland maybe an explanation of what it can do, up to 3 features11:33
wolfspraulthat was my thinking, roughly11:33
lekernelxiangfu: you can commit your patch11:37
lekernellooks good11:37
xiangfulekernel: ok.11:37
lekernelwolfspraul: yeah, screenshots, and maybe a schematics of how to use the camera11:38
xiangfulekernel: about "Remove curr_layout and input_update_keyboard_layout(), and call11:41
xiangfusysconfig_get_keyboard_layout() in get_keycode().11:41
wpwrakwolfspraul: will it be sold in shops ? if not, the box doesn't matter much11:42
xiangfuif move the update_keyboard_layout to get_keycode. that will called when any key pressed.11:42
lekernelxiangfu: ?!????11:42
lekerneljust remove that function11:42
wolfspraulwpwrak: well I hope we can get into some small specialized retail shops, definitely11:43
lekernelman, keep it simple11:43
lekernelno need to duplicate data all over the place11:43
lekernelthis is a bait for bugs11:43
wolfspraulwe need to bypass the entire tech feature-compare-mania and directly explain and sell the product to music and video lovers11:43
wolfspraulso the box is quite important, say someone takes the box into his hands, and reads what it says, looks around on the box design and tries to understand what this product might be good for11:44
wpwrakwolfspraul: thought of doing an image that shows the data flow, as i suggested a few days ago ?11:45
wolfspraulnot sure what you mean with data flow exactly11:46
wolfspraulthe key point is that this box creates real-time video output based on music input11:46
wolfspraulthat it comes with xx effects preinstalled and that you can download more works/effects from the Internet for free11:47
lekernelwolfspraul: we need to show the camera setup11:48
lekernelusing the camera brings a LOT11:48
wpwrakwolfspraul: show a "live" scene, camera, arrow from camera into the mm1. some speakers/microphone, arrow from sound into the mm1. then the generated image, showing elements of the live scene plus effects. (of course, all constructed from still images, stylized drawings, etc.)11:48
wolfspraulalright, that's #3 :-)11:48
lekernelaudio is less important11:48
wolfspraulwe can only talk about 3 things on the box11:48
wpwrakwolfspraul: it's video+audio input. did you notice how hardly anyone on /. realized that ? :)11:49
wpwrakwolfspraul: maybe show the data flows in a big picture, then have three small items: effect synthesis; A/V modulation; connectivity ?11:50
wolfspraulI think in the room/environment that m1 runs in, there is always music11:50
wolfspraulor maybe someone wants to setup a silent installation with it? visuals only?11:50
wolfspraulwell let's see. we agree on the features, need to see how we can bring it out on the box well.11:51
wpwraki think it would be good to emphasize that it can generate significant effects on its own. so it's not just a video distorter from the 1980es.11:51
wolfspraulI think in order to win a customer over, you need a simple starting point to explain what this is. that would be music for me right now, but I can think about the camera more.11:52
wpwrakah, another feature is that it can be extended/programmed. maybe merge effects and modulation into one item then :)11:53
wolfspraulthere are probably more people dealing with music than dealing with video or lights11:53
wpwrakif you sell it mainly as audio-modulated image generation, that may affect how you have to position it in relation to the competition11:54
wpwrakalso, people who don't use video now may be attracted by the possibility, even if they eventually decide it's too cumbersome for them ;-)11:55
wolfspraulI'm very clear about the 'value proposition' (ahem) of the video input and DMX features11:55
wolfspraulwe just need to see how we can bring this out on the box11:56
wolfspraulwe can only talk about 3 things on the box11:56
wolfspraulit's not about perfection there, it's about getting people interested11:56
wpwraki would show video and audio more than talk about them :)11:56
wpwraki'm not sure i explained the image i have in mind well. the one with the data flow. lemme try to sketch it ...12:06
wpwrakthere's a sucky sketch: http://downloads.qi-hardware.com/people/werner/tmp/mm1-doodle.jpg12:06
wpwrakwhat the big screen is meant to show is a distortion of the camera image plus some effects.12:09
CIA-51mtk: Xiangfu Liu master * rd265742 / (11 files in 2 dirs):12:10
CIA-51mtk: move keymap output mtk12:10
CIA-51mtk:  * remove the keymap.c12:10
CIA-51mtk:  * Remove unneeded key codes12:10
CIA-51mtk:  * add "include/ascii_keycodes.h".12:10
CIA-51mtk:  * for remove all tables (shift, altgr) out of mtk.git I have to direct get ASCII code12:10
CIA-51mtk:  * update the get_ascii: direct get ASCII from input.c - http://bit.ly/izFKRO12:10
CIA-51flickernoise: Xiangfu Liu master * r3ca7f65 / src/flash.c :12:11
CIA-51flickernoise: cleanup the webupdate code12:11
CIA-51flickernoise:  * Store files on ramdisk12:11
CIA-51flickernoise:  * merge download and flash task to one12:11
CIA-51flickernoise:  * MTK is not thread safe, call 'Messagebox' only from the GUI task.12:11
CIA-51flickernoise:  move the display_flashvalid_message() into update_progress().12:11
CIA-51flickernoise:  * Use the same ok_callback with different argument. - http://bit.ly/iFJGrG12:11
CIA-51flickernoise: Xiangfu Liu master * r2e5ea6e / src/cp.c : cp.c: hook open_flash_windows on 'Web update' button - http://bit.ly/iviNLk12:11
CIA-51flickernoise: Xiangfu Liu master * r2ab05ed / (8 files):12:11
CIA-51flickernoise: add US, German, French, keymap to input system12:11
CIA-51flickernoise:  * add three tables and keymap.c:12:11
CIA-51flickernoise:  keyb_shift_table12:11
CIA-51flickernoise:  keyb_altgr_table12:11
CIA-51flickernoise:  keyb_translation_table (a lot of NIL)12:11
CIA-51flickernoise:  * move the update keyboard layout from mtk.git to flickernoise.git - http://bit.ly/mABDl912:11
lekernelwpwrak: looks good12:12
wpwrak(effects image) perhaps something along the lines of what this guy did: http://www.youtube.com/watch?v=OSRL-GUYqtw12:12
wpwrak(around 2:45 there are some effects that should also look good in a still image)12:15
wpwrakor more bombastic after 3:30 :)12:15
wpwrakthis one is nice. very AR-ish (augmented reality) http://www.youtube.com/watch?v=KiB56qnFsEw12:43
aerisWhen AR on MM1 ? :p12:45
wpwrakindeed. let's hope soon ;-)12:46
lekernelhmm... shouldn't be too hard to add the missing bits for http://www.youtube.com/watch?v=KiB56qnFsEw to flickernoise12:46
lekernelactually this guy's coming to Europe in July... I should email him again12:47
wpwrakhe should love your little box :)12:52
CIA-51flickernoise: Xiangfu Liu master * r169dd6e / (7 files): change root folder name from /flash to /ssd - http://bit.ly/llyhEI12:55
azonenberglekernel: you around?22:47
azonenbergDid you have a chance to look at the test patterns i posted? Any suggestions?22:47
azonenbergThe goal is to characterize KOH etch rates and determine the minimum feature size i can reliably fabricate22:48
azonenbergI don't know where in the process the limiting factor will be, so far I know that in tests I've hit 5um per pixel in photoresist on copper22:48
lekernelwell, those patterns look good for this purpose... what problems do you expect?22:49
azonenbergBut I haven't done the rest of the process yet (HF wet etching through developed photoresist into tantalum pentoxide hardmask, then stripping PR and wet etching with KOH through hardmask into silicon)22:49
azonenbergWell, one thing i need test patterns for even if it works fine is characterizing etch rates22:49
azonenbergThe hardmask i'm using is Ta2O5 which is marketed as an antireflection coating for solar cells and optoelectronics22:50
azonenbergIt's never been used for this purpose AFAIK22:50
azonenbergBut i needed something that could be deposited by spin coating, was more resistant to KOH than SiO2 (which will work as a hardmask but is eaten slowly by KOH and won't survive deep through-wafer etches), and could still be patterned with HF22:51
azonenbergThat's the biggest risk AFAIK - i dont know if it'll work for this purpose22:52
azonenbergI already know i can coat and pattern the photoresist, I already know PR will survive HF (I've etched glass microscope slides with it, though adhesion was poor because i over-etched)22:52
azonenbergAnd KOH etching of Si is a well documented process22:53
azonenbergSo the hardmask is the biggest risk as of now22:53
azonenbergOnce I have patterns formed in Si, though, things get a lot less clear22:53
azonenbergMetalization etc is much less well developed on the amateur side of things22:54
azonenbergI'm in talks with a friend of mine (grad student doing her research on sputtering) and a physics professor at another university who's trying to build a homebrew sputter deposition rig22:54
azonenbergBetween the three of us, and anybody else who joins my sputter research group between now and when I get Si etching working reliably (which is the last milestone before tackling metalization), I'm hoping in ~6 months of work we'll be able to get a working DC sputtering rig22:56
lekernelphew... at electrolab (hackerspace near Paris) there are two big boxes full of vacuum equipment22:56
lekernelincluding two primary pumps and two (old) turbomolecular pumps22:56
azonenbergWorking turbopumps??22:56
lekernelthey look in good condition, but we did not test them22:57
lekernelone is missing its control electronics22:57
azonenbergIf you guys want to part with any, or design a sputtering rig with my help and let me mail you guys dies to process, that'd be great lol22:57
azonenbergI could just imagine this becoming an international hacker fab with the lithography here, metalization in Paris, and wafers being mailed back and forth for processing22:58
lekernelhmm ... #1 problem here is time22:58
lekerneland then money22:58
azonenbergSame here lol22:58
azonenbergI'm a PhD student, on a grad student's budget22:59
azonenbergWhich is why this project is moving as slowly as it is on my end lol22:59
azonenbergI finally have a few extra $$ so i'll be placing some orders shortly23:00
lekernelif we do the metalization in Paris, it's going to be super slow for multilayer anyway23:00
lekernelspeaking of turbopumps, do you know of any device to protect them in case the vacuum chamber breaks or otherwise leaks?23:01
azonenberglekernel: Yes, i know it'll be slow - but better than nothing23:01
azonenbergAnd hmm...23:02
lekernelwe have been quite reluctant to try them, because they can break so easily23:02
azonenbergI know what you mean, i've seen shredded pumps23:02
azonenbergAre you concerned about a mechanical failure or user error (flipping on the vent valve while it's still spinning)?23:02
azonenbergThe second could be partially addressed by using electronic valves and building a control system that warns the user, or outright says "no", if you try venting before the pump spins down23:03
lekernelsince we are total newbies with building vacuum stuff, I'd guess our plumbing may leak, break, etc.23:03
azonenbergBut mechanical failure that brings you from high vacuum to near atmosphere instantly?23:03
azonenbergThat's gonna be catastrophic no matter what IMO23:04
azonenbergHowever, on the plus side23:04
azonenbergThat's very rare23:04
azonenbergMuch more likely is a slow leak23:04
azonenbergWhich will be something you can detect and plug in advance23:04
azonenbergAnd which, if it starts while the chamber is pumped down, is unlikely to hit a dangerously high pressure before you can shut down the turbo23:04
azonenbergAlso, consider there are other options besides turbopumps23:05
azonenbergDiffusion pumps are almost indestructible for example23:05
lekernelwe also have a mercury (!) diffusion pump23:05
lekernelbut not in very good condition and missing its heater23:05
azonenbergI'm told oil diffusion isn't that hard to build23:05
lekernelthe mercury doesn't seem to be there either, we'd need to find some23:05
lekernelthe complete thing is very rusty too23:06
azonenbergGood luck in today's world lol23:06
lekerneland rust outgases, no?23:06
azonenbergI know a guy in california with a vial containing a few CCs but thats it23:06
azonenbergOutgasses? I don't think so23:06
azonenbergThe bigger risk IMO would be the surface trapping water23:06
azonenbergWhich would outgas23:06
azonenbergBut the vapor pressure of FeO2 is high23:07
lekernelyeah but it's hydrated afaik23:07
azonenbergOh, good point23:07
azonenbergits not just iron oxide, its the hydrate23:07
azonenbergSo yeah that would be problematic23:07
lekernelon the other hand - yeah, it's a very simple device23:08
lekernelprobably DIY-able23:08
azonenbergMy advice would be to build an oil diffusion pump from scratch and not touch the turbo until and unless you get a chamber you're confident with (having gotten down to high vac with the diffusion pump and left it in operation for several hours at a stretch)23:08
lekernelah, we have a bottle of silicon oil for diffusion pumps too23:08
azonenbergI'm told diffusion pumps have been DIYed in the past23:09
azonenbergIf you get it working i'd love to see docs23:09
azonenbergIn other news - stuff on deck for ordering within the next week: bottle of 50nm colloidal silica slurry for CMP, diamond scribe for wafer cleavage, more rubber gloves (it's not a cleanroom but i don't want TOO many fingerprints), empty 4-inch cassettes, a 4-inch single side polished N doped <100> wafer, ten <110> 1cm square single side polished undoped substrates, some die trays23:11
azonenbergtwo pounds of technical grade KOH23:11
azonenberg4 ounce bottle each of Ta2O5 and SiO2 spin-on coating solutions23:12
lekernelbtw - I can order mercury chloride (and electrolyze that)23:12
azonenbergUV safety glasses23:12
kristianpaulthere is a something else besides verilog to "run" a test bench and generated output compatible with gtkwave?23:12
lekernelit's 31E for 100 grams though23:12
azonenbergAnd a 1-watt 385nm UV LED flashlight23:12
azonenbergI'd go for oil, it's safer23:12
lekernelso I'd get chlorine gas as well as mercury :-)23:12
azonenbergI know a guy who messed with Cl2 at home23:13
lekerneldid it too, but in very small quantities23:13
azonenbergAfter a lab accident (ground glass joint popped from overpressure) he had a cough and wheeze for six months23:13
azonenbergWhy he's still alive is beyond me but past experience has shown that the guy is pretty hard to kill23:13
azonenbergbeing around 20kV laser capacitors doesn't seem to bother him much either lol23:14
lekernelfunny, I'd swear that one year ago there was pure mercury in the index too.23:15
azonenbergSuffice it to say, oil diffusion is probably a much better bet lol23:15
lekernel(in France you can order all sorts of chemicals on http://www.la-mep.com)23:15
azonenbergFrance isn't as paranoid about that stuff as the rest of the EU is with RoHS etc?23:16
lekerneloh, it is23:16
azonenbergNot to say you can't buy all kinds of nasties in the USA too23:16
azonenbergBut considering stories i hear about people unable to buy isopropanol above 50% in australia etc23:16
lekernelactually this website is supposed to be for schools and labs only, but they do little checking ;-)23:17
azonenbergFor the record i have a 16oz bottle of 99% IPA / 1% DI water around that i clean stuff with lol23:17
azonenbergAnd lol, i see23:17
lekernelsame here, from la-mep :-)23:17
azonenbergI own a now-defunct LLC that i route orders through if they are suspicious23:17
lekernelI used it for titrating free fatty acids in waste vegetable oil when I was making DIY biodiesel23:18
azonenbergMy apartment is in a commercial district, the sign on the front of the building says "TROY NEON SIGN COMPANY"23:18
azonenbergSo no problems here lol23:18
azonenbergAnd i see23:18
azonenbergI actually get a lot of good stuff from a biodiesel supplier - DudaDiesel.com23:19
azonenbergThat's where i'm ordering my KOH from23:19
azonenbergIt's all pretty pure but still technical grade, if you need trace-metal go elsewhere lol23:19
azonenbergBtw, what do you think of getting something published in a journal of electronics education or something like that after this is done?23:22
azonenbergThe focus not being on hobbyists but on schools, getting kids a taste of microfab without super-expensive equipment or access to a cleanroom23:23
azonenbergDemonstrating that your average high school lab is able to do 5um fab with little additional material23:23
lekernelazonenberg: do you think the mercury diffusion pump could work with oil instead?23:28
azonenberglekernel: You'd need to clean it out well, the mixture of the two would definitely be problematic23:29
lekernelyeah, sure, and remove rust as well23:29
lekernelbut beyond that?23:29
azonenbergAnd the heater would need to be updated for a different set temperature23:29
lekernelwe have no heater ...23:29
azonenbergSo not a problem, you need one anyway23:29
lekernelthe pump misses it23:29
azonenbergMaybe some different sized apertures for the nozzle23:29
lekernelwe only scavenged the body23:29
azonenbergBut that's about it, i suspect it will work fine23:29
lekernelthat would be cool23:31
lekernelactually this pump is a surprisingly simple design23:31
azonenbergAnd it would be significantly more robust than a turbopump23:31
lekernelprobably more DIY-friendly too23:31
lekernelI'd guess anyone with an average/good metal workshop could build one23:32
lekernelwe also have three ion pumps23:34
lekernelone is badly damaged, the others are worn but probably still ok23:34
lekernelwe do not have the magnets though23:35
lekernelthey also seem _very_ simple to build. amazing they're so expensive.23:35
azonenbergIon pumps? Not familiar with the design23:35
azonenbergcryo, diffusion, and turbo are the only three high-vac i know23:35
lekernelusually you don't need those, except for ultra high vacuum23:36
lekernelit's a tertiary pump23:36
azonenbergOh, lol23:36
azonenbergSomething you use on top of a turbo etc?23:36
azonenbergThat would explain it lol23:36
azonenbergI dont expect we'd need one any time soon23:36
--- Sat May 21 201100:00

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