| DocScrutinizer05 | I don't think it's useful for anybody or anything meanwhile | 00:00 |
|---|---|---|
| DocScrutinizer05 | nevertheless thank you for the thanks | 00:01 |
| norly | i think that *someone* needs to take the step | 00:03 |
| DocScrutinizer05 | maybe I could contribute useful to fixing i.MX8M power consumption issues. That's at least one topic where I seem to have knowledge that might help improve things and result into sth useful | 00:10 |
| DocScrutinizer05 | thanks to my very early focus on event driven architecture, combined with always doing hw system development | 00:11 |
| DocScrutinizer05 | or I go for doing farmer | 00:13 |
| norly | since (as you said yourself) the market is small, i think it makes sense to consider joining forces with the purism project | 00:15 |
| norly | in some way, it sounds a lot like STEP2, doesn't it? | 00:15 |
| norly | and maybe that project can then feed back into Neo900 | 00:16 |
| norly | after all, it still doesn't provide a phone with hackerbus and hardware keyboard | 00:16 |
| norly | both phones have their use cases | 00:16 |
| Joerg-Neo900 | >>joining forces with the purism project<< been tried several times, and meanwhile not on my list of things I want to do anymore | 00:17 |
| norly | oh no, i'm sad to hear it didn't work :( | 00:17 |
| Joerg-Neo900 | their CTO(?) Bob Ham never had a particularly friendly attitude towards Openmoko, Nikolaus / GolDeliCo, and me | 00:18 |
| norly | how about the pyra? is there still an exchange of knowledge with that project? | 00:19 |
| Joerg-Neo900 | let alone Purism now performingf massively sub-par on exactly the topics he used to bash us | 00:19 |
| Joerg-Neo900 | all knowledge with Pyra been exachanged already, and I think not much new knowledge came up regarding PYra | 00:20 |
| Joerg-Neo900 | it's a pity they also don't ship yet | 00:20 |
| Joerg-Neo900 | and I'm tired to follow that thread of reality on a daily basis | 00:22 |
| norly | yeah, all these projects drag on | 00:24 |
| norly | at least pyra seems to near completion. it'll be done real soon (TM), in 2 months (TM) | 00:24 |
| norly | so it's not impossible | 00:25 |
| paul_boddie | Especially since it was already said, thanks from me for your efforts, Joerg! | 00:29 |
| norly | yes, really | 00:29 |
| norly | what's the status on neo900 anyway? it's been near completion years ago it seemed | 00:30 |
| paul_boddie | The work done on Neo900 is considerable, and if only I could think of a way of moving it forward. | 00:31 |
| norly | yeah, that would be amazing | 00:32 |
| paul_boddie | norly: I think the problem was mainly getting a layout done. | 00:32 |
| paul_boddie | And I share Joerg's views on the modern world, smartphones and surveillance gadgets. | 00:33 |
| paul_boddie | That people are now happy to plant gadgets with microphones and cameras in their houses and think that such things are only enabled if they say "OK Google" or whatever. | 00:33 |
| Joerg-Neo900 | we have an insurmountable deadlock, comnsisting of lack of layoutr to show off a working prototype, lack of preorders due to missng prototype to show off, and impossible sourcing of "mech parts" like display, sliders. etc, even if we could suddenly generate a 400 more preorders. So the project is in a massive zombie state with very little hope to get revived without massive changes in design and concept | 00:35 |
| paul_boddie | It's difficult to take a different path when everyone else is happy to take surveillance money. | 00:35 |
| Joerg-Neo900 | yes | 00:36 |
| paul_boddie | I did wonder whether a different "core" board would be a starting point. Things change and even Nikolaus realised this after doing his Cortex 8 board and seeing that the PocketBeagle was practically the same. | 00:39 |
| Joerg-Neo900 | there were two opportunities to carry on with STEP2 with partners, at beginning of last year. Both evaporated into nonsensical requirements either regarding HR or limit of available funds. Since then I think there's a better futer elsewhere for me | 00:39 |
| Joerg-Neo900 | future* | 00:39 |
| paul_boddie | I guess the problem with Neo900 is the physical constraints, so it's not like the ZeroPhone where things just get stacked up and the casework is done at the end. | 00:41 |
| paul_boddie | Not that I should really say anything about that without looking more closely at it. | 00:42 |
| paul_boddie | Anyway, I'll look through the Neo900 documents again soon. I made sure to download many of them the other day. | 00:42 |
| Joerg-Neo900 | paul_boddie: yes, that's part of the problem, the case | 00:43 |
| Joerg-Neo900 | but in the end the *only* problem is funding | 00:44 |
| paul_boddie | In principle, reusing the N900 case is a good move given how expensive casework is to get made. | 00:44 |
| paul_boddie | My perception is that there's some demand for a genuinely open phone, but people get easily distracted and buy the latest shiny Samsung instead. | 00:45 |
| Joerg-Neo900 | fund was "give us the better fraction of a million and we're talking". Now that evolved into "give us a low 1 digit number of millions and we can accomplish something for sure" | 00:46 |
| paul_boddie | Then there's a demand for a secure phone, but this attracts all sorts of Jason Bourne imitators and people who don't care about the openness. | 00:46 |
| Joerg-Neo900 | paul_boddie: excellent analysis | 00:47 |
| Joerg-Neo900 | andd to the "secure" domain those like purism who think security is a hardware switch | 00:48 |
| paul_boddie | So the two groups don't really overlap. Or at least the pretend Jason Bourne group doesn't really understand the issues. | 00:48 |
| Joerg-Neo900 | add to* | 00:48 |
| paul_boddie | The real Jason Bourne group have access to all sorts of top secret stuff, I would think. ;-) | 00:48 |
| Joerg-Neo900 | there's the professional security domain where devices cost a 2l or 4k or more. Those DO understand security... sometimes | 00:49 |
| Joerg-Neo900 | think blackphone etc | 00:50 |
| bencoh | please tell me you don't really think the blackphone is a seriously "secured" phone | 00:50 |
| bencoh | I know you know better :) | 00:51 |
| Joerg-Neo900 | no, definitely not | 00:51 |
| bencoh | pfew :) | 00:51 |
| Joerg-Neo900 | they aimed at that, and failed miserably despite being highly professional | 00:51 |
| bencoh | huhu | 00:51 |
| Joerg-Neo900 | because they came with a software-only perspective and had no clue about hw | 00:52 |
| paul_boddie | There was an amusing anecdote about hardware testing on Bunnie's blog involving a power switch which didn't work in the final product. | 00:52 |
| paul_boddie | URL (animated images, CPU intensive): https://www.bunniestudios.com/blog/?p=5450 | 00:52 |
| paul_boddie | Anyway, I need to go. Keep up the good work, Joerg and company! | 00:53 |
| * Joerg-Neo900 waves | 00:54 | |
| xmn | It suck when your ahead of the trends, but then it passes you by when you can get past road blocks. Now it's trendy to say your security focus. It's also unfortunate if purism didn't see any value in working together. Would be a great question to ask the ceo on one of these podcast he seem to be on a lot these days. | 00:59 |
| Joerg-Neo900 | HAHA yeah >>Thus, in order for the factory to meet the product designer’s expectation of low labor cost with simultaneously high quality, it’s up to the product designer to come up with an automated, fool-proof test jig.<< Seen that, so true. Worthwhile read, many thanks to paul_boddie for that link | 00:59 |
| Joerg-Neo900 | haha >>However, despite using a series-elastic spring and a foam pad to avoid over-stressing the servo motor, over time, we’ve found the motor still fails, and operators have disconnected it in favor of manually pushing the button at the right time.<< shit happens | 01:07 |
| Joerg-Neo900 | I had used a massive solenoid | 01:08 |
| Joerg-Neo900 | don't ever use gears when you can get along without | 01:09 |
| Joerg-Neo900 | proper solenoid may be rated for a guaranteed 1 million cycles | 01:11 |
| Joerg-Neo900 | a gearbox needs 20 ointments during that time | 01:11 |
| Joerg-Neo900 | the motor itself might not even be tested for the number of on-off cycles it can do before MTBF | 01:15 |
| Joerg-Neo900 | >>functionality of indicator LEDs and the fan are visually inspected by the operator;<< hey Bunnie, power consumption monitoring goes a long way ;-) | 01:25 |
| Joerg-Neo900 | in case of fan, you'd guess a nifty EE might have a thermosensor near a source of heat to read out and check efficiency of fan | 01:27 |
| Joerg-Neo900 | looking at >>That’s me about 12 years ago...<< I can't help but think I did similar thing 12 years ago when in GTA02-A8 the increased size of audio decoupling capacitors ruined the awkward serial connection to modem via audio port. It shouldn't but it was my fault to not instruct the tester software guys to disable the audio amp oputput and switch it to high impedance so the probkem would never have happened. Caused a full day of no testing | 01:36 |
| Joerg-Neo900 | and massive panics at fab | 01:36 |
| Joerg-Neo900 | and ^^^ this was the true reason why we had unbearable small audio decoupling until very last GTTA02 run since the chinese colleagues knew that would happen when they get larger caps but didn't think of the fix ever | 01:40 |
| Joerg-Neo900 | hehe >> Can the 2017 badge have sex with 2015 badges? Yes.<< | 01:50 |
| Joerg-Neo900 | >> How much worse is the range on this badge than the 2015 badge? It’s hard to say at this point, but early tests indicate it’s sufficient to have blinkysex with people within casual conversation distance, but not within shouting distance. << | 01:51 |
| Joerg-Neo900 | aaaaand on topic: >> Why is the badge more expensive than 2015? The primary cost driver for the badge is the cost of the injection molded case.<< | 01:52 |
| * enyc meows | 11:43 | |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!