Fuze battery life(just MP3)

MP3 players Sansa Fuze
1

@ ewelot wrote:
Interesting discussion. Anyone out there who has some numbers on battery life with and without external memory card (e.g. playing mp3 only)?

Not me…my card is always in. :smiley:

Perfect! You are the one I’m looking for because I don’t own a card. :smileyvery-happy:

I get 25hrs of mp3 playback (96kbps, volume level 75% on “Normal” scale, backlight 50% for a total of about 20min). I’m curious to see your numbers soon …

Thing is, I have lots of different file types, and bitrates…so I don’t know how scientific a comparison would be. I do have a few WMA VBR and Ogg Vorbis files that are close to your bitrate there, but those codecs use more battery power to decode than MP3. The vast majority of my files are much higher bitrate…my 4GB +8GB card has 2,224 songs on it. I’ll try to remember to keep track of the time the next time I use the Fuze, but it’ll take a few days to get through it. You might have to remind me though.  :stuck_out_tongue:

Only a suggestion: pick a single mp3 file and let the fuze play it back repeatedly over night or some other time you are not using it - if there is any. :wink: 

Well, I found one album that is 160kbps, CBR MP3 ( the lowest bitrate MP3 that I have) . The Fuze backlight timer, 10 seconds…brightness is about 33%… volume, about 66% into 60-ohm 'phones. The Fuze is repeating that album (The Doors-Waiting For the Sun) which happens to be on the external card (a SanDisk 8GB class 2 microsdhc ) . When I crash out tonight, which will be about 04:00 , I will  make a personal note of battery level and time elapsed, and shut it down. I’ll continue on with it tomorrow.

P.S. After 4 hours it was at 84% battery, although it’s well documented that the battery meter is not exactly linear. Test will continue tomorrow.

@ewelot wrote:

@marvin_martian wrote:

…

P.S. After 4 hours it was at 84% battery, although it’s well documented that the battery meter is not exactly linear. …

Instead of “well documented” I suspect that it rather has been vaguely guessed - or somehow felt. At least with the latest firmware there is no measurable deviation from linearity. Even the scatter of the individual points matches with PERFECT linearity and the fact that the battery power is given at steps of about 5-6 percent. This has probably not been taken into account by a few users.

Happy listening to another few hours of The Doors …    (Yeah, I like this one a lot)

@marvin_martian wrote:

@ewelot wrote:

@marvin_martian wrote:

…

P.S. After 4 hours it was at 84% battery, although it’s well documented that the battery meter is not exactly linear. …

Instead of “well documented” I suspect that it rather has been vaguely guessed - or somehow felt. At least with the latest firmware there is no measurable deviation from linearity. Even the scatter of the individual points matches with PERFECT linearity and the fact that the battery power is given at steps of about 5-6 percent. This has probably not been taken into account by a few users.

Happy listening to another few hours of The Doors …    (Yeah, I like this one a lot)

 

P.S.: I didn’t succeed to insert my battery power plot, grrrrh. Any advise (file formats)? To whom may I address this issue?

Message Edited by ewelot on 05-14-2009 04:50 PM

I’m not listening to it…just letting it run. I’ll look at itafter another 4 hours have gone by, and see where it’s at.You may have a point about this particular firmware , but in the past, both Fuze and Clip have not run down in a very linear fashion…I recently did a FLAC test with my Clip here which illustrates what I meant.

 

P.S. After an additional 4 hours, the tally stands at… 63% after 8 hours.

Message Edited by Marvin_Martian on 05-14-2009 02:05 PM

Thx for the reference. And there is the plot of the battery running down on my Fuze (mp3, 96kbps, volume 75%, backlight 50% for a total of 20min):

 

timn brought up a valid point that this could be in it’s own thread, so here we go!

 

2

@ewelot wrote:

@marvin_martian wrote:

…

P.S. After 4 hours it was at 84% battery, although it’s well documented that the battery meter is not exactly linear. …

Instead of “well documented” I suspect that it rather has been vaguely guessed - or somehow felt. At least with the latest firmware there is no measurable deviation from linearity. Even the scatter of the individual points matches with PERFECT linearity and the fact that the battery power is given at steps of about 5-6 percent. This has probably not been taken into account by a few users.

This has been stated as fact by Sansafix more than once. And of anybody here, I suspect he would be in the best postion to know. :wink:

3

The tally stood at… 63% after 8 hours.

Now, after an additional 5 hours, it dropped from 42% to 36% as I was shutting it down. So 36% after 13 hours.  I never normally run it this low, but in the interest of science, I’ll continue on tomorrow,  although once the battery meter reads 0% the little blue beauty is getting hooked up to the wall charger.   I don’t know what to expect for a final number, but I’ll cheerfully live with 20 hours instead of ewelot’s 25…because I couldn’t live with 96kbps MP3.

4

@tapeworm wrote:

@ewelot wrote:

@marvin_martian wrote:

…

P.S. After 4 hours it was at 84% battery, although it’s well documented that the battery meter is not exactly linear. …

Instead of “well documented” I suspect that it rather has been vaguely guessed - or somehow felt. At least with the latest firmware there is no measurable deviation from linearity. Even the scatter of the individual points matches with PERFECT linearity and the fact that the battery power is given at steps of about 5-6 percent. This has probably not been taken into account by a few users.

This has been stated as fact by Sansafix more than once. And of anybody here, I suspect he would be in the best postion to know. :wink:

Yes, I agree on that. Maybe I better should say, that with the current firmware I can’t spot any deviation from linearity (using my setup). The normal user who only occasionally checks battery power would not be able to draw a definite conclusion because of the limited accuracy of the indicated power levels. These are not continuous values but decrease in steps of 5 or 6 percent.

Moreover, other things like switching on/off the fuze, changing backlight level, changing file types etc. might have a higher impact on normal operation. In short: users shouldn’t care too much. Anyway they are watching the battery icon in most cases - but that’s another story I’m not getting into. :wink:

What I’d like to draw more attention is battery life (yeah, now on the title!) instead of (non)linear power meter. And even then I’m not too scientific and wouldn’t mind of any differences less than about 20% in fact.

5

Marvin:

I love the chart! I believe you said you are using a memory card, it would be interesting to see a similar chart running from internal memory.

I think the suspicion of many people is that the power level would decrease linearly, but slower from internal memory, suggesting that power consumption is greater for external memory. My guess would be that there is a difference due to the extra ‘circuitry’ used with the external memory.

care to do that and chart them side-by-side?

personally if I get 20-30% less play time from external memory I’m fine with that just for the convienience.

My 2GB Fuze + 8GB chip (=10GB)  was slightly cheaper than even a 4GB Fuze when I bought them, and about the same as the 2GB Fuze alone when not on sale.

6

In earlier firmware versions the battery meter was not exactly linear, even with the same codec.  We characterized the battery capacity better vs. time and made some adjustments to the algorithm…

After making subsequent revisions to the firmware its nice to see that the battery meter linearity with playtime has been confirmed by users.

Although imperfect and subject to some errors,  the battery gauge should provide a decent indication of the remaining playtime.

Keep in mind that playtime varies depending on the content on the player, its bitrates and codecs.

So for example if you started out playing mp3s and saw the battery meter at a certain level,  then started playing OGG or FLAC content at high bitrates,  the battery would drain faster and make the indicator seem less linear.

If you stick with a certain codec and bitrate,  the meter should work very well for estimating the remaining play time.

Cheers,

SansaFix :slight_smile:

Message Edited by sansafix on 05-15-2009 09:57 AM

7

@timn wrote:

Marvin:

I love the chart! I believe you said you are using a memory card, it would be interesting to see a similar chart running from internal memory.

 

I think the suspicion of many people is that the power level would decrease linearly, but slower from internal memory, suggesting that power consumption is greater for external memory. My guess would be that there is a difference due to the extra ‘circuitry’ used with the external memory.

 

care to do that and chart them side-by-side?

 

personally if I get 20-30% less play time from external memory I’m fine with that just for the convienience.

My 2GB Fuze + 8GB chip (=10GB)  was slightly cheaper than even a 4GB Fuze when I bought them, and about the same as the 2GB Fuze alone when not on sale.

I can’t take credit for the chart…that’s ewelot’s chart.:smiley:

8

@sansafix wrote:

In earlier firmware versions the battery meter was not exactly linear, even with the same codec.  We characterized the battery capacity better vs. time and made some adjustments to the algorithm…

 

After making subsequent revisions to the firmware its nice to see that the battery meter linearity with playtime has been confirmed by users.

 

Although imperfect and subject to some errors,  the battery gauge should provide a decent indication of the remaining playtime.

 

Keep in mind that playtime varies depending on the content on the player, its bitrates and codecs.

 

So for example if you started out playing mp3s and saw the battery meter at a certain level,  then started playing OGG or FLAC content at high bitrates,  the battery would drain faster and make the indicator seem less linear.

 

If you stick with a certain codec and bitrate,  the meter should work very well for estimating the remaining play time.

 

Cheers,

 

SansaFix :slight_smile:

 

 

Message Edited by sansafix on 05-15-2009 09:57 AM

I agree completely on the different codecs and bitrates, it is the reason I initially hesitated to do a comparison to ewelot’s results (which were obtained with 96kbps mp3 ).

I have all different codec and almost all much higher bitrates than that, which I knew would have an effect. The only similar bitrates to ewelot that I have are either WMA VBR or Ogg Vorbis, which both take a little more power to decode than mp3. So I undertook this with 160 kbps CBR mp3, which was my lowest bitrate mp3, for the sake of consistency. 

The battery life generally is not an issue for me anyways with my Fuze, as it rarely leaves my house (I have Clips for that). I also rarely run it below 60-70% before I charge, but I gave this a try out of curiosity (and had a couple days off ) .

9

At the 16 hour mark, 21% battery life. To be continued…

10

@sansafix wrote:

In earlier firmware versions the battery meter was not exactly linear, even with the same codec.  We characterized the battery capacity better vs. time and made some adjustments to the algorithm…

…

You did a very good job on that!

I took some other measurements a few weeks ago because I was curious about battery life with wav files and the impact of the backlight. If anyone is interested I could provide the related plots as well. By the way they confirm the good linearity of the battery meter.

(Edit: quoted messages truncated)

Message Edited by ewelot on 05-15-2009 10:20 PM

11

Marvin_Martian wrote: 

I agree completely on the different codecs and bitrates, it is the reason I initially hesitated to do a comparison to ewelot’s results (which were obtained with 96kbps mp3 ).

I have all different codec and almost all much higher bitrates than that, which I knew would have an effect. The only similar bitrates to ewelot that I have are either WMA VBR or Ogg Vorbis, which both take a little more power to decode than mp3. So I undertook this with 160 kbps CBR mp3, which was my lowest bitrate mp3, for the sake of consistency. 

 

The battery life generally is not an issue for me anyways with my Fuze, as it rarely leaves my house (I have Clips for that). I also rarely run it below 60-70% before I charge, but I gave this a try out of curiosity (and had a couple days off ) .

On a similar line of thinking, I’ve been wondering which codec uses the least power while providing transparency? Using LAME VBR, transparency is supposed to be about ~165 to ~190k (V4 - V2). Anyone know what sounds good (or good enough) with WMA or Ogg Vorbis, and the kind of battery life you get? Currently I’m using LAME V4 (~165k) and it sounds pretty good, though I think I can notice some compression.

Message Edited by not1975 on 05-15-2009 09:51 PM

12

@not1975 wrote:

 

On a similar line of thinking, I’ve been wondering which codec uses the least power while providing transparency? Using LAME VBR, transparency is supposed to be about ~165 to ~190k (V4 - V2). Anyone know what sounds good (or good enough) with WMA or Ogg Vorbis, and the kind of battery life you get? Currently I’m using LAME V4 (~165k) and it sounds pretty good, though I think I can notice some compression.

Message Edited by not1975 on 05-15-2009 09:51 PM

The least amount of power while still providing transparency would probably still be MP3…it’s the easiest of the 3 codecs mentioned to decode.

I have gotten great sounding files using V0 LAME MP3, Q7 Ogg Vorbis, and WMA VBR (target range 135-215kbps).

When pinched for space, the WMA VBR target range 85-145 gave me files that were roughly 40% smaller than a 192kbps CBR file, that still sounded decent.I got equivalent results with Q3 Ogg Vorbis that were even a smaller file size.

But I never did a battery test comparison with either of those codecs like I’ve done here, so I’m not sure how big the difference in battery life would be.

13

@marvin_martian wrote:
At the 16 hour mark, 21% battery life. To be continued…

at 17 hours ,15% battery life…and the meter has gone red :dizzy_face:

at 18 hours, 10%…to be continued tomorrow when I can keep a close eye on it (no way I’m running it totally dead )

Message Edited by Marvin_Martian on 05-16-2009 03:01 PM

14

So, after your battery indicator goes red , and you allow it to shut down, will the resulting deficit of energy in the battery cause the Fuze to suddenly float up from the table, start rotating, and then wormhole the entire contents of the room into a zone of singularity like the cool brass orbs used in the film, The Arrival??

For those who haven’t seen it, Charlie Sheen is a satellite TV technician who happens upon an alien infestation.  I am elated that his skills include a working knowledge of radiotelescope antenna arrays and such.  I liked the movie, especially the orbs.

I threaten my daughters with the judicious application of these devices, in cases where their rooms are liberally littered with clothes and papers.  Imagine, simply toss the orb into the room, and all is vacuumed, literally.

Your Fuze will automatically shut off safely above the wormhole threshold, ready to recharge.

On the other hand, should a small miscalibration be encountered, and your Fuze is seen slowly levitating over the desk, exit the room and allow it to perform its room-cleaning task.

Oh, and don’t forget to charge it up, unless the PC has been sucked into the void.  Perhaps, check the wall outlets, as there may be a remaining USB AC adapter dangling about.

Bob  :smileyvery-happy:

15

@marvin_martian wrote:

@marvin_martian wrote:
At the 16 hour mark, 21% battery life. To be continued…

at 17 hours ,15% battery life…and the meter has gone red :dizzy_face:

 

at 18 hours, 10%…to be continued tomorrow when I can keep a close eye on it (no way I’m running it totally dead )

Message Edited by Marvin_Martian on 05-16-2009 03:01 PM

19 hours…holding at 10%…

16

@marvin_martian wrote:

 

at 18 hours, 10%…to be continued tomorrow when I can keep a close eye on it (no way I’m running it totally dead )

Message Edited by Marvin_Martian on 05-16-2009 03:01 PM

19 hours…holding at 10%…

Holding at 10% after 20 hours :dizzy_face:

I have chosen to stop after 21 hours…it was down to 5%, and that’s good enough for me. :stuck_out_tongue:  I know I wouldn’t break 20 if I had played all the different file types on my Fuze. 

Message Edited by Marvin_Martian on 05-18-2009 12:11 AM

17

And now it’s all recharged, and reloaded…2,105 songs, with 12mb free.:smiley:

18

@marvin_martian wrote:

@marvin_martian wrote:

 

at 18 hours, 10%…to be continued tomorrow when I can keep a close eye on it (no way I’m running it totally dead )

Message Edited by Marvin_Martian on 05-16-2009 03:01 PM

19 hours…holding at 10%…

Holding at 10% after 20 hours :dizzy_face:

 

I have chosen to stop after 21 hours…it was down to 5%, and that’s good enough for me. :stuck_out_tongue:  I know I wouldn’t break 20 if I had played all the different file types on my Fuze. 

Message Edited by Marvin_Martian on 05-18-2009 12:11 AM

Well done! So the total battery life with your 160kbps mp3 on external card is about 22 hours. This is only 3 hours less then my test on low bitrate mp3 files. Given the different bitrates, different devices and slightly different test procedures I would conclude that the impact of using an external card is rather small - maybe about 10% only. That is really good news, I should buy a hugh card soon … :smiley:

I’m leaving for 2 weeks vacation. Maybe I’ll do some more testing with some ripped CD files which are of higher bitrate (of cause :wink: ) then the files of my previous test. These were streams from internet radio stations. I’ll let you know …

Message Edited by ewelot on 05-19-2009 10:31 PM

19

@ewelot wrote:

@marvin_martian wrote:

@marvin_martian wrote:

 

at 18 hours, 10%…to be continued tomorrow when I can keep a close eye on it (no way I’m running it totally dead )

Message Edited by Marvin_Martian on 05-16-2009 03:01 PM

19 hours…holding at 10%…

Holding at 10% after 20 hours :dizzy_face:

 

I have chosen to stop after 21 hours…it was down to 5%, and that’s good enough for me. :stuck_out_tongue:  I know I wouldn’t break 20 if I had played all the different file types on my Fuze. 

Message Edited by Marvin_Martian on 05-18-2009 12:11 AM

Well done! So the total battery life with your 160kbps mp3 on external card is about 22 hours. This is only 3 hours less then my test on low bitrate mp3 files. Given the different bitrates, different devices and slightly different test procedures I would conclude that the impact of using an external card is rather small - maybe about 10% only. That is really good news, I should buy a hugh card soon … :smiley:

 

I’m leaving for 2 weeks vacation. Maybe I’ll do some more testing with some ripped CD files which are of higher bitrate (of cause :wink: ) then the files of my previous test. These were streams from internet radio stations. I’ll let you know …

Message Edited by ewelot on 05-19-2009 10:31 PM

One biggie to remember is that I didn’t use the Custom EQ. That takes an additional toll on the battery too if you use it, FYI. :smiley:

20

@marvin_martian wrote:

One biggie to remember is that I didn’t use the Custom EQ. That takes an additional toll on the battery too if you use it, FYI. :smiley:

 

Is this true if you have the Custom EQ selected and all set to flat/zero?

…Adding, at a minimum your test dispells the concern that just the mere presence of an external card is a significant battery drain, which is good to know. 

Message Edited by Robisan on 05-22-2009 01:07 AM