#milkymist IRC log for Monday, 2012-07-02

azonenberglekernel, kristianpaul: http://siliconexposed.blogspot.com/2012/07/bga-process-notes.html00:21
azonenbergThis is the same process I use with FT256 package00:22
azonenberghavent tried on FG484 but i expect it to work00:22
kristianpaulThanks !00:25
kristianpaulDo you concern about transition times when changing owen temperature?00:27
kristianpaulseems dont :)00:28
kristianpaulgotcha, visual inspection is fundamental00:29
azonenbergkristianpaul: yeah, its all human-in-the-loop feedback00:33
azonenbergnot scalable to mass production but requires very little fancy equipment00:33
azonenbergi do have the capabilities to make a real profile controller00:34
azonenbergbuilding one is on the short-term TODO list00:34
azonenbergits going to be ethernet enabled00:34
azonenbergso i can put a board in there00:34
azonenbergthen go back to my desk and dial in a profile00:34
azonenbergand it'll beep when its done00:34
kristianpaulis more like an OpenCV PID00:35
kristianpaulif you think about infrared..00:35
azonenbergLol well the real one is going to be based on a thermocouple and a relay00:35
azonenbergand some kind of closed-loop feedback loop00:35
Jialekernel: may you tell me how did you use clang&binutils? I mean, how could you make clang calling as&ld?01:04
kristianpaulJia, i tried once from here  https://raw.github.com/milkymist/milkymist-ng/master/README01:15
kristianpaulif you mean for lm32...01:15
Jiayeah, lm32, and, your help is awesome01:18
Jiakristianpaul: cmake ..; make; make install  is there anything like --prefix= ?01:48
kristianpaulshould be, but dont find it in the  CMakeLists.txt02:00
kristianpaullast time i installed clang was using with autotools :)02:01
kristianpaullekernel: fake dma? :-)02:17
kristianpauli tought imagine you said not using dma anymore?..02:17
Jiahttp://paste.ubuntu.com/1070695/ build error04:47
lekernelazonenberg: nice07:35
lekernelJia: I modified clang07:35
lekernelsee the last commits in git07:35
lekernelI don't know about your build error ... works for me... are you using clang/llvm from the milkymist github?07:36
Jialekernel: I'll read your code when I get home:)10:24
Action: Jia are confused about the lm32-gcc bug10:25
kristianpaullekernel: had you used amber?13:55
Jialekernel: your last commit is "Mico32 -> LM32", may you give me more detils?15:20
lekernelJia: there are others. this one is just to match the cpu name from binutils.18:21
lekernelkristianpaul: no18:21
lekernelwhat's Amber?18:21
kristianpaulcheck opencores18:39
kristianpaulfree of arm patents they claim so far it called my antetion18:48
kristianpaulafaik no mmu18:49
azonenbergkristianpaul: open source ARM core?20:08
azonenbergi'm sticking to mips since its a nice old architecture20:08
azonenbergall patents on mips1 have long since expired20:08
kristianpaulazonenberg: kinda yes, old arch same as your mips core20:09
azonenbergWell yeah, all the current ones are patented up the wazoo20:10
wpwrakkristianpaul: 40 MHz on a Spartan 6 45T ...20:26
wpwrakkristianpaul: (page 45 of the PDF) 80 MHz on a virtex20:27
azonenbergwpwrak: 40 MHz on an ARM core on s6?20:44
azonenbergeven my MIPS hits 80, do they have significantly better performance per clock?20:44
azonenberghalf the clock rate would need a much more efficient architecture to be competitive20:44
kristianpaulwpwrak: need to verify that without the ddr3 controller from xilinx20:46
lekernelazonenberg: can you deposit titanium films (on a KTN crystal)?20:58
azonenbergIn house, or do i hav access to it?21:00
azonenbergIn house the only film coating i can do is spin coating, i plan to build a sputtering setup but don't have the parts yet21:00
azonenbergi can however do evaporation on campus21:00
wpwrakazonenberg: they only compare it with itself. so no idea how it compares with other cores in real life. 1.0 MIPS/MHz suggests they're not ~2x faster than the rest of the pack (which they'd need to make up for the low clock speed)21:00
lekernelincluding titanium?21:00
azonenbergLet me look up the temps21:01
azonenbergI've done copper (melts 1084C boils 2562C)21:01
azonenbergtitanium is a little higher (1668 and 3287)21:01
azonenbergi've also done Cr (1907, 2671)21:02
azonenbergand while i have not personally other people use the tool for gold (1064, 2856)21:02
lekernelI'm trying to build a laser scanner/projector out of this https://www.ntt-review.jp/archive/ntttechnical.php?contents=ntr200709sp2.html21:02
azonenbergSo it should be possible to get hot enough21:02
lekernelafter reading up, it's a little more complicated than what I thought21:03
lekernelfirst those KTN crystals are expensive as hell21:03
azonenbergThe tool cannot get hot enough to do Pt (1768, 3285)21:03
azonenbergvapor pressure is too low21:03
azonenbergand hmm...21:03
lekernelthen, I read mentions of the crystal lattice orientation, but they are contradictory. so I should actually find a way to make failures less expensive. I know a crystal processing workshop in Berlin, maybe I'll order a boule and process it locally21:04
azonenberglol nice21:04
azonenbergwpwrak: 1.0 MIPS/MHz is comparable to my architecture21:05
azonenbergmaybe eve na little worse if they're talking DMIPS21:05
azonenbergand i clock twice as fast21:05
azonenbergno experience with LM32 but i'v eheard its comparable to a hair faster than mine21:05
lekerneland finally, the electrodes. seems you need more than just an electric field - you need an ohmic junction to inject some electrons into the crystal. and the way to do that is to deposit titanium on a polished face.21:05
azonenbergI see21:06
azonenbergwell, sputtering it shouldnt be too difficult21:06
lekernelso, except those three "little" problems, it seems like a cool technology :-)21:06
azonenbergi havent built a sputtering system yet, its the third on my TODO list21:06
azonenbergthe first two being a better spin coater and a 2-inch contact mask aligner21:06
mumptaia DIY foundry?21:07
azonenbergmumptai: #homecmos21:07
azonenbergits coming along slow since i'm doing it part time while in school21:07
azonenbergbut it is moving21:07
lekernelazonenberg: so, titanium sputtering sounds like something you would be capable of doing?21:08
azonenbergAm currently capable of, no. Will be at some point, yes21:08
azonenberghow long it'll take depends on a lot of factors21:08
lekernelwell, using tools on the campus. get it done, no matter how ;-)21:09
azonenbergIf i can evaporate it, i can do it easily21:09
lekernelinduction heating should be possible, no?21:09
azonenbergevaporation is done with a tungsten filament21:09
lekernelah, I see... old school21:10
azonenbergthe hard part is getting the necessary vacuum21:10
azonenberg1E-6 torr range21:10
lekerneldoes it really need to be that perfect?21:10
azonenbergSputtering can be done much hgiher21:10
lekernelor else?21:10
azonenberg1E-3 to 1E-221:10
azonenbergYou get oxidation etc21:11
azonenbergsputtering is done with an inert gas plasma at low temperature21:11
wpwrakazonenberg: (MIPS) yes, DMIPS. so it sounds like a nice achievement in terms of proof of concept. but it doesn't quite feel as if they had just obsoleted what we have.21:41
--- Tue Jul 3 201200:00

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