**** BEGIN LOGGING AT Tue Mar 27 03:00:04 2018 Mar 27 03:22:11 muneeb17: Just went through your proposal, looks good to me. I was wondering if you could implement the log_printf thing a bit earlier in the timeline. Mar 27 04:23:03 Mentors, I've uploaded my final proposal. But there's still 12 hours to make edits, so I'd appreciate any last minute suggestions Mar 27 04:58:06 muneeb17: Posted this earlier, just went through your proposal, looks good to me. I was wondering if you could implement the log_printf thing a bit earlier in the timeline. Mar 27 04:58:30 yeah, sure. I'll adjust it earlier Mar 27 05:01:41 Ds2: Can you review the changes I made in my proposal. I am planning to use bela cape instead of bela mini Mar 27 05:01:42 since it has onboard speaker amps and add MAX9814 AGC microphone amplifier. Mar 27 05:01:43 https://www.elinux.org/BeagleBoard/GSoC/ModernSpeak-Spell Mar 27 05:02:21 ds2: ^ Mar 27 05:09:43 @wormo can you take a last look at my final proposal Mar 27 05:10:26 suggestions,changes would be appreciated Mar 27 05:40:34 zeekhuge: Thanks for the suggestion. Updated the proposal to add Debug functions. abhishek_: shifted the log_printf implementation Mar 27 05:42:51 jyothi[m]: when are you creating the grammar files? Mar 27 05:42:58 and how Mar 27 07:49:08 ds2 : I thought it will be better if it just had a keyword list to detect the commands since the module requires only basic commands. Using grammar files may restrict the command recognition to certain extent right? Mar 27 08:08:40 jyothi[m]: you still need to give something to pocketsphinx Mar 27 08:12:12 ds2:Ohh. Didn't know that. Will add that part. Mar 27 08:12:13 Any other changes u recommend? Mar 27 08:13:08 the .lm file I think that's it Mar 27 08:14:01 not sure what to specifically suggest but I'd be concerned about it working right (speech recog). I did some stuff with it and it was pretty sensitive to mic type/volume Mar 27 08:14:18 it is not a drop in (at least for me) Mar 27 08:14:28 Okay. Thanks:) Mar 27 08:17:03 ds2 :Ohh. Do u think it is possible to finish it in the first three weeks then? Mar 27 08:17:24 I also found many forums talking about the volume and threshold problems during the recognition Mar 27 08:17:42 I can may be extend it to one more week. I have little slack during the gaming part Mar 27 08:25:45 that I donno off hand... Mar 27 08:26:17 that's why I was asking about HW AGC...HW AGC made it a lot better for my use case Mar 27 10:01:24 Ohh. Okay. Mar 27 12:06:24 ds2: See the proposal now... https://elinux.org/Beagleboard:GSoC_2018_Proposal:Beaglebone_GPU_Offload Mar 27 12:31:42 jkridner: added the proxy server implementation in week 4 in proposal https://elinux.org/BeagleBoard/GSoC/etcher-beagleboot Mar 27 14:42:16 jkridner: Well how does it look now Mar 27 14:44:10 zeekhuge: jkridner: Just updated this draft: https://elinux.org/BeagleBoard/GSoC/sdabhi23/BeagleBoneRemoteSeismometerNode Mar 27 14:44:22 Please have a look Mar 27 14:59:55 final hour. Mar 27 15:00:56 hgonzalez: did you make either of your proposals final? Mar 27 15:01:50 sdabhi23: were you able to get Wormo and nerdboy to look? Mar 27 15:02:05 jkridner: Well did you look Mar 27 15:03:00 sdabhi23: can you give a link to a description of BOINC? Mar 27 15:03:07 sdabhi23: don't miss the deadline... best to make it final... mentors can re-open as needed for a bit. Mar 27 15:03:18 jkridner: I'm refactoring the PRU SPI/I2C master one, to focus more on the application example, but I don't have much time to go too deep on that. I proposed extending the 2016 project and the video capture application that ds2 proposed. But honestly, that last part is not really clear and in my opinion a bit too broad. Mar 27 15:03:25 Sidharth: are you the same as sdabhi23 ? Mar 27 15:03:34 sdabhi23: did ping them on the channel about the previous version of draft Mar 27 15:03:51 * jkridner[m] Mar 27 15:03:53 oh, your are GPU Offload Mar 27 15:04:05 Sidharth: did you get ds2 to look? Mar 27 15:04:23 Sidharth: some of the weeks are combined without a specific deliverable for each week. is ds2 good with that? Mar 27 15:04:33 jkridner: here it is http://boinc.berkeley.edu/ Mar 27 15:04:58 Well ds2 didn't look.. Mar 27 15:05:16 I put it up on the mailing list as well Mar 27 15:05:21 and am working on the final version also Mar 27 15:05:43 Students : you can submit the PDF proposal on the dashboard as many times as you wish, until the deadline is done. So do it now, and then you can edit and upload it again. Mar 27 15:11:13 jkridner[m]: Can you help me contact him? Mar 27 15:12:23 Sidharth: yes, I can google it, but I don't think people reading it should need to. Mar 27 15:27:05 Well, final proposals submitted Mar 27 15:33:27 jkridner: Do you have his mail id? I am editing the proposal, but still I needed some feedback from him as well Mar 27 15:39:01 I didn't get what you said. (That google line...) Mar 27 15:43:22 sorry, that should have gone to sdabhi23. Mar 27 15:43:31 with regards to BOINC Mar 27 15:44:33 jkridner: Yep. Added that in the final proposal. Mar 27 15:44:53 https://elinux.org/BeagleBoard/GSoC/fix-bugs-bonescript please consider my proposal Mar 27 15:45:21 vaishnav98_: did you make a final entry as well? Mar 27 15:45:37 Yes I did Mar 27 15:46:26 anujdeshpande[m]: have you looked it over? Mar 27 15:47:18 Arduino-for-userspace-linux has been largely superceded by create.arduino.cc Mar 27 15:47:54 they have "native" support for Arduino sketches on BeagleBone now. Mar 27 15:51:10 jkridner[m]: I didn't know about it... still adding c support seems to be useful Mar 27 15:55:03 I'm not sure how the C support is that helpful. Being able to target PRU seems more importnat. Mar 27 15:55:12 * jkridner isn't seeing your final submission... Mar 27 15:56:59 vaishnav98_: ah, see it now. Mar 27 15:57:39 I have described about adding support for ultrasound sensor HCSR04 by targeting PRU Mar 27 15:58:29 Well it was fun researching for the proposal. Look forward to the future, whether I get selected or not. I will get the GPU to work though! Mar 27 15:58:35 spiritus: did you submit a final proposal? Mar 27 15:58:44 Thank you all for your help Mar 27 15:58:49 Sidharth: ! :-D Mar 27 17:06:04 Sidharth: I don't see the point of having QT at all... make pure EGL/GL calls Mar 27 17:06:11 we are not doing graphics here Mar 27 17:22:22 hmmmm Mar 27 17:54:38 Its done ? Mar 27 17:55:22 Deadline . Mar 27 18:14:22 quiet Mar 27 23:03:57 ds2: Good point! I confused Qt as providing a interface to drivers, not the other way round. Mar 27 23:06:56 I thought that Qt with its framework provides interfaces which itself manages as well (as a part of OS) (like an included DRM) Mar 27 23:10:12 Qt includes a lot of things Mar 27 23:10:25 coding to EGL is as bare as you can get Mar 27 23:10:39 I can dig out some little bits I did before but it'll take some time to go through my old stuff Mar 27 23:25:44 BTW, what ways are there to access the GPU healessly? Mar 27 23:26:10 I was looking into this mesa thing called GBM but... Mesa drivers not for SGX Mar 27 23:26:29 DirectFB also does not support PVR2D Mar 27 23:27:45 libEGL is the key Mar 27 23:27:46 I thought virtual display driver (maybe? like VirtualBox does) Mar 27 23:28:02 none of those really needed Mar 27 23:28:28 the drivers will try to use the frame buffer Mar 27 23:34:03 ds2: What is your timezone? Mar 27 23:37:22 (I saw on mentors listing, don't bother replying..) (Didn't see it that clearly before) Mar 27 23:46:31 Sidharth: PDT (GMT-7) **** ENDING LOGGING AT Wed Mar 28 03:00:01 2018