BOINC app for Android is out!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Fardringle

Diamond Member
Oct 23, 2000
9,197
763
126
It probably isn't great for the health of the battery, Philippart. I definitely don't plan to run it long term. Right now I'm trying to get results on a set of four work units (for four cpu cores) from each project that has an Android app just to see how they run and what credits they get.
 

2is

Diamond Member
Apr 8, 2012
4,281
131
106
I did some experimenting and figured out my power issue. I have been using the Nexus 7 charger with a longer USB extension cord since the 3 foot cord that came with it just wasn't long enough most of the time. However, I installed the Battery Monitor Widget app and discovered that while connected to the charger with the 6 foot extension cord, my Nexus pulls between 300 and 500 mA from the charger (and uses about 1000-1200 mA while running BOINC), but when it is connected to the charger with only the short cord that came with it, the Nexus pulls between 1100 and 1400 mA (enough to counteract the drain caused by running BOINC). I didn't expect that a longer cord would have any effect, but apparently either the charger or the Nexus 7 (or both) don't like it for some reason.

The short cord also charges the tablet a lot faster while it is idle, so I guess I need to remember only to use the USB extension when I actually need it, and just plug in the short cable any time I'm running BOINC or when I want the Nexus to charge faster.

Unfortunately, the Einstein@home tasks started out running quickly (about 5% in the first 20 minutes) they have slowed down considerably since then and look like they will take about 20 hours to complete, just as the Asteroids@home tasks did.

It's not that the devices don't "like" the longer cord. It's a physical limitation of the wires. They are extremely thin and can only carry a finite amount of current. When you lengthen these cables, you're increasing resistance which means more loss and less current is available at the end of the cable. My Apple devices are the same way. Not so much the iPhone because it only uses 5 watts and most cables with extensions can accommodate that. But on my iPad there's a huge difference in charge times. Heck, there's even a difference between a cheap 3rd party cable and the one that came with the iPad even if they're both the same length
 

Fardringle

Diamond Member
Oct 23, 2000
9,197
763
126
OK, my test for Seti@Home Beta is done.

Apparently there are two different types of work units on this project. I had two tasks take 17.75 hours each to finish (average 37.9 points) and two that took 34.9 hours (88.8 points). Not only is it a bit odd to have such large differences in times, but it's also only an average of 2.5 points per hour per CPU core for a new "last place" position in my list..


The only other two projects that I know of that have Android apps are OProject@home, which doesn't have any work available, and Yoyo@home, which apparently requires a lot of memory because the client says there is work available but my device doesn't have enough memory.

Edit: I just checked the BOINC event log and Yoyo is actually saying that it requires 2098.08MB free memory, which means that they aren't going to get much participation from Android devices. Are there any at all that have more than 2.5GB of RAM? (at least .5 for the OS and apps, plus 2GB for Yoyo)
 
Last edited: