I can only add my experience with my own Galaxy S4 :

Running 2 cores permanently with SETI for me is only possible in winter basically, right now I'm forced to limit it to 1 core due overheating issues otherwise.

Since the casing is basically plastic, placing it onto a cooler metal don't do anything for me.

Also keep in mind its display is an AMOLED which lifespan will suffer if permanently exposed to high temperatures.

Running 3 or even all 4 cores on the S4 would be downright impossible for me, short of submerging it into cold water inside a thin waterproof trashbag in a fridge or something...
These things just aren't built for permanent full CPU load.

---

I did a reset on the #2 Ematic EGQ307 7-30-14 @ 0900 Z then reinstalled BOINC/S@H.
The memory problem cleared up and this tablet is behaving normally so far.

Regarding the heat issue ... the 307 is in a plastic case and to make matters worse - on the bottom surface there is a thin foam rubber like coating that acts as insulation further trapping heat. Placing the tablet on the heat sink and holding it tightly to the heat sink with 2, #33 rubber bands seems reduce the internal heat ... enough that both 307s run OK (memory issue not withstanding).

The heat sink weighs 2 pounds 7 ounces measures 4.6 by 8.0 inches. There are 15 fins and the thing stands 1.75 inches high. It comes in a standard 4.6 inch width and you purchase it by the length (8 inches in my case). I got it from
HeatsinkUSA, LLC Store
http://www.heatsinkusa.com/ $ 31.60 USD delivered.

The problem is all of these devices are passively cooled, and the cooling ability of the device is largely based on how much the internals can spread the heat.

I have a Galaxy Note 3 and experienced the exact same thing with heat. I ran Seti@Home on it for maybe 3 days and basically just dumped it all together because of how hot the device got. That doesn't mean I don't run Seti@home on Arm devices. I have several of them in my house and figured out a few of them could run full time if I made a few tweaks.

First I have a Ouya that is crunching all the time. It is hooked up next to a HTPC that has really good ventilation. Just having good air movement next to it allows it to run full time.

Then I also have a GK802 android Mini PC. It was a experiment that was really a waste. It is a POS, and all it is good for is crunching Seti@home. I did have to take a old AMD Phenom CPU Cooler and set it on top to remove heat though.

Ironically my Nexus 10 tablet seems to churn through seti@home with no heat issues what so ever. It barely gets warm at all.

I also have a old Note 1 that is running a snapdragon S3 CPU and runs very cool as well, and is able to run full time.

My first suggestion to anyone with heat issues would be just try doing something that will improve air movement around the device. Even just setting the device propped up next to a fan will make a huge difference.

I am also curious if anyone has tried a Odroid-U3, or the Odroid-XU. I wonder if the (Odroid-U3)http://hardkernel.com/main/main.php would be a good cheap full time low powered cruncher.

I have also noticed some weird results from my devices when it comes to processing time. Ironically it seems my devices speed go as follows from lowest to highest: GK802, Ouya, Galaxy Note 1, Nexus 10

---

Re speed (good question) my HP7+ has an average RAC 0f 241.
Ematic EGQ307 # 1 has an average RAC of 195. Ematic # 2 is up to 138 now but I expect it will fall in line with Ematic # 1 in a week or so - once the moving average stabilizes. I'll try putting a fan on one of them - if I can find a spot to do this undisturbed for about 2 weeks.

Got again a WU with lots of "stackdumps unavailable" and restarts in the std_err. WU 3664664147.

Any ideas why this is happening?

---

BOINC on android seems to be a bit strange atm. Remember that I complained about not getting more then 3 or 4 tasks at once? Well, last time I connected to my Wifi I suddenly got 17 beta WUs. So I guess I will do beta exclusivly for some time now on my phone.

---

Is Seti main still sending out 7.21 for Android?

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

I have an odroid U2 running now for about one year. Running Android most of the time, have also tried Kubuntu and Lubuntu. Works fine, runs overclocked to 2GHz 7/24 but requires definitely extra cooling.
Device draws ~6 Watts out of the wall. Has no KB/Mouse/Monitor attached, is remote controlled via VNC.
Currently it is crunching for The Skynet Pogs and Simap, has reached an overall score of 381,674 credits @ pogs (rank 2984 today) and 129,931 credits @ Simap (rank 7523 today).
Has also crunched Milkyway for 6 weeks, just to show that the top 10.000 can be reached in that time.
Has also crunched the Einstein Arecibo Neon wu's for some weeks, Android and Linux if I remember correctly.
All together: No problems so far.
At nativeboinc.org a statistic over various projects can be found.

There is a new XU3 Odroid available now, should be able to run all 8 cores at the same time, has a OpenCL 1.1 full profile GPU on board. Comes with an 5V 4A PS.

But this raises another question: How to cool this device? Has anyone experience with oil-bath cooling? I mean putting the whole device into an container full of oil.

Alexander

---

Well, no idea how to cool it, but to be honest, I don't think it's worth it anyway.
I run Beta on my old Ace 2. It's oc'ed to 1050 Mhz and gives me a antutu score of 10500 points. The XU3 makes about 41000 points, so it's about 4 times as fast as my old phone. My Ace 2 currently has an RAC of 68 on Beta, so you could probably get 300 with the XU3. Maybe 400 if you can use the GPU also. That's still much less then you get with a cheap laptop.

---

Well, no idea how to cool it, but to be honest,

Putting the mobile device on something metal, like the top of a computer case will conduct heat away quite well,
conversely putting it on something like a car seat when charging from the cigarette lighter socket will make it get extra hot since the heat doesn't get conducted away.

Claggy

He asked about putting the device into oil, so I suppose just putting it on some metal won't be enough. Not sure though, I never owned something like that. My old phone doesn't require any cooling although it's overclocked by more then 30%.

---

The U2 (discontinued) came with a large heat-sink. At the beginning I tried it with a fan, but the temp of the heatsink came up to > 65 degree C.
So I made a watercooler for this device.
https://dl.dropboxusercontent.com/u/50246791/watercooled%20Odroid%201.jpg
https://dl.dropboxusercontent.com/u/50246791/watercooled%20Odroid%202.jpg

The new device has a much smaller heatsink, has 8 insted of 4 cores and I want to use the gpu. So an even better cooling is required. Heatsink + fan will not be sufficient for 7/24 @ full load, so why not dipping the whole device into a cooling liquid? The thermal resistance should be excellent low and the noise of a passive cooling is unbeatable!
But the contacts, will they survive some years?

Alexander

---

I have an odroid U2 running now for about one year. Running Android most of the time, have also tried Kubuntu and Lubuntu. Works fine, runs overclocked to 2GHz 7/24 but requires definitely extra cooling.
Device draws ~6 Watts out of the wall. Has no KB/Mouse/Monitor attached, is remote controlled via VNC.
Currently it is crunching for The Skynet Pogs and Simap, has reached an overall score of 381,674 credits @ pogs (rank 2984 today) and 129,931 credits @ Simap (rank 7523 today).
Has also crunched Milkyway for 6 weeks, just to show that the top 10.000 can be reached in that time.
Has also crunched the Einstein Arecibo Neon wu's for some weeks, Android and Linux if I remember correctly.
All together: No problems so far.
At nativeboinc.org a statistic over various projects can be found.

There is a new XU3 Odroid available now, should be able to run all 8 cores at the same time, has a OpenCL 1.1 full profile GPU on board. Comes with an 5V 4A PS.

But this raises another question: How to cool this device? Has anyone experience with oil-bath cooling? I mean putting the whole device into an container full of oil.

Alexander

Doing a google image search for odroid U2 showed many of them with I guess an optional heatsink like this. Some had fans some didn't. I would guess that is the best cooling option for that hardware?

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Doing a google image search for odroid U2 showed many of them with I guess an optional heatsink like this. Some had fans some didn't. I would guess that is the best cooling option for that hardware?

Correct, this is how I got mine. Heatsink was not optional, the fan was. But overclocking from 1.7 to 2 GHz required extra cooling.
Google for odroid cluster, you may find real freaks.

---

Doing a google image search for odroid U2 showed many of them with I guess an optional heatsink like this. Some had fans some didn't. I would guess that is the best cooling option for that hardware?

Correct, this is how I got mine. Heatsink was not optional, the fan was. But overclocking from 1.7 to 2 GHz required extra cooling.
Google for odroid cluster, you may find real freaks.

I saw their 40mm fan but I imagine a larger 60mm or 80mm fan would do a better job. Maybe even a 120mm set off to the side to blow across the whole thing. Most are 12v, but finding 5v fans, often with USB connectors, are easy enough to locate these days.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

I saw their 40mm fan but I imagine a larger 60mm or 80mm fan would do a better job. Maybe even a 120mm set off to the side to blow across the whole thing. Most are 12v, but finding 5v fans, often with USB connectors, are easy enough to locate these days.

Basically you are right, fans of all sizes and operating voltages are not the problem.
But these arm devices do have a plastic package, the heat sink has no copper core, so anything that increases the heat transport is welcome.
And a second reason: the noise ...

---

Sooo... except for the small issue I wrote about before, it works OK and all results were valid so far.

What I don't understand... now that all app versions have compleated at least 10 WUs, so the server knows how fast each app is, it decided to send me only WUs to the SLOWEST app, i.e. the app with the lowest GFLOPS. My records of compleated WUs confirm, that this app is the slowest, in particluar on "normal" AR WUs, VLARs and VHARs are about the same for all apps as far as I can say with the WUs I got untill now.

Was the idea of sending all apps to a device not to find the fastest app for this device and than send WUs only (or at least mostly) to this app?

As you can see on the application details page the two apps with the highest GFLOPS just got the standard 10 WUs each, the two slower apps much more (and the slowest app got most of course just to confirm, that the server likes it most).

---

Sooo... except for the small issue I wrote about before, it works OK and all results were valid so far.

What I don't understand... now that all app versions have compleated at least 10 WUs, so the server knows how fast each app is, it decided to send me only WUs to the SLOWEST app, i.e. the app with the lowest GFLOPS. My records of compleated WUs confirm, that this app is the slowest, in particluar on "normal" AR WUs, VLARs and VHARs are about the same for all apps as far as I can say with the WUs I got untill now.

Was the idea of sending all apps to a device not to find the fastest app for this device and than send WUs only (or at least mostly) to this app?

As you can see on the application details page the two apps with the highest GFLOPS just got the standard 10 WUs each, the two slower apps much more (and the slowest app got most of course just to confirm, that the server likes it most).

Worth to file complain on BOINC forum or alpha/dev lists cause this has nothing to do with project apps... and volunteers that will read your message here. It's question for BOINC devs and it will never be answered if posted in wrong forum.

Sooo... except for the small issue I wrote about before, it works OK and all results were valid so far.

What I don't understand... now that all app versions have compleated at least 10 WUs, so the server knows how fast each app is, it decided to send me only WUs to the SLOWEST app, i.e. the app with the lowest GFLOPS. My records of compleated WUs confirm, that this app is the slowest, in particluar on "normal" AR WUs, VLARs and VHARs are about the same for all apps as far as I can say with the WUs I got untill now.

Was the idea of sending all apps to a device not to find the fastest app for this device and than send WUs only (or at least mostly) to this app?

As you can see on the application details page the two apps with the highest GFLOPS just got the standard 10 WUs each, the two slower apps much more (and the slowest app got most of course just to confirm, that the server likes it most).

The server needs 11 completed and validated non-outlier tasks before it is considered, not 10.

Who knows if the scheduler will ever send more work for those two faster app versions.

Claggy

Don't know if this has already been asked here but can a smartphone also crunch when it's not connected to the main current? In other words, can an android crunch only on its battery?

---

rOZZ
Music
Pictures