[LAD] Android's 10 Millisecond Problem
interesting article on Android's latency http://superpowered.com/androidaudiopathlatency/ via Hacker News news.ycombinator.com ___ Linux-audio-dev mailing list Linux-audio-dev@lists.linuxaudio.org http://lists.linuxaudio.org/listinfo/linux-audio-dev
Re: [LAD] Android's 10 Millisecond Problem
On Thu, 16 Apr 2015 09:41:11 -0600, John Hammen wrote: interesting article on Android's latency http://superpowered.com/androidaudiopathlatency/ via Hacker News news.ycombinator.com Thank you for sharing this. Assumed Android devices one day should be usable for audio, I wonder if they suffer from a known iPad issue too. It's true that latency isn't an issue when using an iPad, but battery charge could become an issue. I still need to use a stop watch, but I guess after around 5 or 6 hours and sometimes even after less hours, my iPad 2 turns off. If the iPad then is plugged to the power supply, it anyway can't be used. Loss of battery charge by the power consumption of a DAW/MIDI sequencer - Audiobus - parametric EQ chain needs more power, than the 10 W power supply does recharge the battery. IOW for Apple tablet PCs compulsory breaks interrupt creative work for hours, as soon as the battery needs to be recharged. Apple devices need much improvement too, not regarding latency, but regarding the power consumption. They are far away from being perfect. The OS might be good, but the hardware isn't. I read that people using the iPad 3 for office work, stopped using them, regarding the power supply issue. Is it possible to run an Android device by a power supply, when the battery is empty? Regards, Ralf ___ Linux-audio-dev mailing list Linux-audio-dev@lists.linuxaudio.org http://lists.linuxaudio.org/listinfo/linux-audio-dev
Re: [LAD] Android's 10 Millisecond Problem
On 04/16/2015 07:10 PM, Ralf Mardorf wrote: Is it possible to run an Android device by a power supply, when the battery is empty? nexus7 (2012) here, doesn't work, no. If you run something that smokes more battery than the speed of charging then you're doomed. These computers were really badly engineered (for people using them for hours and hours at least). Nice cash machine though. ___ Linux-audio-dev mailing list Linux-audio-dev@lists.linuxaudio.org http://lists.linuxaudio.org/listinfo/linux-audio-dev
Re: [LAD] Android's 10 Millisecond Problem
On 04/16/2015 07:10 PM, Ralf Mardorf wrote: [...] Is it possible to run an Android device by a power supply, when the battery is empty? Regards, Ralf ___ Linux-audio-dev mailing list Linux-audio-dev@lists.linuxaudio.org http://lists.linuxaudio.org/listinfo/linux-audio-dev There is no such thing as the one representative android device since android runs on many different devices, made by different manufacturers. Comparing Android to Apple is a bit of a problem here. Basically, it depends on the device and the workload. For example I can do a lot of stuff with a Galaxy S3 Phone and the power reserves of the (stock) power supply are adequate for usual tasks, even for installing updates, and demanding tasks. For a tablet, this may be entirely different, since the screen is much bigger. Cheers, -Markus ___ Linux-audio-dev mailing list Linux-audio-dev@lists.linuxaudio.org http://lists.linuxaudio.org/listinfo/linux-audio-dev
Re: [LAD] Android's 10 Millisecond Problem
On Thu, 16 Apr 2015, John Hammen wrote: interesting article on Android's latency http://superpowered.com/androidaudiopathlatency/ via Hacker News news.ycombinator.com What the article totally misses... Latency is about more than clearing audio path obsticles. Can the kernel be made lowlatency? What other proprietary globs might demand cpu/bus time with no regard for anything else? I am thinking about my experiences with my netbook (back when they made those) where the wifi made anything less than about 20ms (one way) impossible without xruns. Add bluetooth, and USB issues and who knows what else. Many of the android units are really cheap (sub $100 mark) and so must use the least expensive everything. The CPU may be yesterdays super computer but the rest is not. Lowlatency audio with artifacts is not going to help. Yes the tablet I have (and my wife's phone) have way too much latency to be useful for anything beyond remote control of mixers etc. (I wouldn't want to do live fades that way :) It will be interesting to see if Touch will be any better. (still uses the android kernel to make the most of the HW drivers.) -- Len Ovens www.ovenwerks.net ___ Linux-audio-dev mailing list Linux-audio-dev@lists.linuxaudio.org http://lists.linuxaudio.org/listinfo/linux-audio-dev
