Battery benchmark on X5 under Rockbox(with graphs)

[Disa]

Hi all! Yesterday I've decided to test my X5 battery running under Rockbox. I've charged it to 100% level, set to repeat one album that contains 22 tracks in 256 CBR MP3, set volume to 3dB level. Then I didn't touch the device until it switches itself off.
I used battery_benchmark plugin from Rockbox. Here you can see resulting curves based on battery_benchmark measurements - red for true dependence and blue for approximated dependence. R^2 is coefficient of approximation reliability.

Large picture: http://pelotki.net/picture_library/battery_large.gif

As you can see, battery life under this conditions is approximately 390 minutes(6 hours and 30 minutes).
I've used Rockbox version 060413-2039 CVS.
And what do you think about it?
I think it's nice result for 10%-complete power handling=)

//Please sorry for my bad English, I'm from Russia and I have a problem with correct operating with tenses=)

[ya1950]

That's very, very interesting. I'm getting much less than that with no enhancements, no crossfade, and no EQ. Or at least the battery is showing 0 charge after about 2 1/2 - 3 hours play time with the value set to 1300 - it never actually shut down before I recharged. Maybe the battery has a greater charge than we were lead to believe?

[larryschild]

tnx a lot for the test...
6.5hours is quiet a good time... battery life was a main concern for me... I might end up trying rockbox next weekend

[Martinp]

Are you using the latest CVS builds. I think there are some updates to power handling, but I am not sure, seeing that I no longer have my X5 :-(

[Disa]

ya1950 Hmm, only 3 hours? Strange...Can you test your battery like me and çùûå you results in this thread?

eXity Latest build with updated power handling was released after I've already tested my battery=)
I think I'll repeat this benchmark under new build tonight.

[ya1950]

@ Disa - tell me how to enable the test and I'll give it a try. Of course it will have to wait for overnight when I'm not listening . I just updated to the latest daily build so maybe it will make a difference. I think I've been using builds earlier than the one you tested but I'm not sure.

There were some posts with results similar to mine but earlier this month. Maybe the battery has been optimized since?

http://forums.rockbox.org/index.php?topic=3387.0

[Disa]

ya1950 I think it'll be better to upgrade to CVS build, because it has latest changes.
My method:
1. Make playlist with 20-25 tracks and set Repeat to All
2. Turn off the player, insert AC cable, turn off highlight.
3. After full charging turn on the player, go to Browse Plugins then select battery_bench. When you selected it info screen appears. Short-press joystick to exit.
4. Go to /root and make sure that file battery_bench.txt exists.
5. Start to play playlist that you made in step 1. Set volume level on 3dB
6. Do not touch the device until it switches itself off

Do not open battery_bench.txt and do not start any other plugins.
After player's cut off charge it a little then copy battery_bench.txt to PC. All information you need exists in this file.
More info about this plugin you can read at http://www.rockbox.org/twiki/bin/vie...tteryBenchmark

[ya1950]

OK thanks. I installed the latest CVS build, created the playlist, and I'm charging. Only thing is that for some reason 3 DB is blasting with my setup. I need to use a -41 DB to get a comfortable level. Not sure why this is. Anyway I'm going to set to a comfortable listening level and the start the test. I'll let you know the results.

[Disa]

Ok, I'll be waiting for results!=)
And about volume level - I've used 3dB with standart headphones, and it was a little loud for me. What headphones you using?
//I think it's better to set volume a little louder than comfortable level so as make sure that you have some capacity reserve(because in real life you using highlight, pause, rewind, menu etc)
////sorry if it's difficult to understand what I'm talking about=)

[ya1950]

No problem at all understanding you - your English is great! I'm using Shure IEM E4G which probably accounts for the differences. The earphones go directly into the ear canal so I guess the volume needs to be lower that with other earphones.

[Disa]

Thanks a lot=)
Oh, Shure e4, cool=) Now it is clear to me why you need to use only -41db level. But for test I recommend to set it to -30dB or so.
//Have you already started your test?

[ya1950]

Test started - let's see what happens

[HiMDforMe]

i'll probably run this test tonight, i have the L-version so we can see how it compares. i agree 3db is really loud. i usally am at -20db to -30db

[Disa]

Hmm, it'll be very interesting to see results for X5L.

i agree 3db is really loud. i usally am at -20db to -30db

Once again - it depends on your headphones.

[afruff23]

Isn't 1e-12 the calculator's definition of approaching zero? Also, where in your results are 390 minutes? Your graph shows that the battery in this test lasted for about 360 minutes until it reaced 10%. Did you not let it run until it reached 0%? And you just approximated?