MP3 or AAC for iPod Batterylife? Options

[sinspawn]

Do you know which import-setting would give me the most playback time? AAC 128 kbps or MP3 160 kbps? I'm using iTunes 4.6 and the 4th generation 40 GB iPod.

[idioteque]

Do you know which import-setting would give me the most playback time? AAC 128 kbps or MP3 160 kbps? I'm using iTunes 4.6 and the 4th generation 40 GB iPod.

My perception was that disk reads are by far the biggest battery user, as compared to processing for either MP3 or AAC. Also, typically the lower the bitrate the less CPU cycles are used for decoding, although that's difficult to apply here since you are comparing AAC and MP3 and I don't know how the complexity of these implementations differ on the iPod. My educated guess is that AAC 128 will use less battery.

[Gabriel]

On an IPod, 128kbps AAC and 160kbps MP3 will give you a very similar battery life.

[rpop]

I haven't noticed any significant difference in battery life between those two formats.

[QuantumKnot]

I have noticed that 128 kbps AAC tends to drain the battery more than 128 kbps MP3.

This post has been edited by QuantumKnot: Oct 15 2004, 12:23

[AgentMil]

I heard that listening to the iPod drains the batteries more than just leaving it off... only kidding...

On my brothers and sisters when listening to tunes imported via CD using iTunes on its default setting which is 128kbps seem to give roughly about 5 days usage with usage limited to 2 to 2.5 hours a day with 50/50 split between listening to full albums and the rest on shuffle mode. Hope this helps... I would say using MP3 would increase battery time... due to its nature of being such a highly developed codec.

Regards

[fairyliquidizer]

On an IPod, 128kbps AAC and 160kbps MP3 will give you a very similar battery life.

This is the official apple position. If you dont mind ripping with a seperate program I'd suggest 160kbps AAC with LAME. I was too lazy and use 128kbps AAC. The downside is that this makes me reluctant to move to a player that doensn't play AAC. The upside is if I had used 160 MP3 I would have filled my iPod by now :-(

[ErikS]

On an IPod, 128kbps AAC and 160kbps MP3 will give you a very similar battery life.

This is the official apple position. [...]

So where have you read this?

[QuantumKnot]

Interesting thing I learned from the Linux on iPod forums, in this thread is that the 4G iPod dynamically changes the clock speed of the Portal Player chip. The highest frequency is 80 MHz but it is usually lower in order to save battery power.

I just created the file _show_speed in iPod_Control/device and had a look at the speed when playing 128 kbps AAC and MP3. When playing AAC, my iPod is hovering around 28 to 32 MHz while 128 kbps MP3 is around 26 to 27 MHz.

[AgentMil]

Amazing... guess that almost proves that MP3 uses less power then? Unless there are other things that factor in as well.


Regards

[Busemann]

What is the CPU usage for VBR mp3's? one would think it is a little higher..

[Jojo]

What is the CPU usage for VBR mp3's? one would think it is a little higher..

I can answer this Nope, it's pretty much the same. I've tested CBR 192 mp3 against LAME --preset standard (~200kbps). It changes all the time and ranges from 41-43 (same as CBR)...

This post has been edited by Jojo: Dec 26 2004, 22:07

[Busemann]

What is the CPU usage for VBR mp3's? one would think it is a little higher..

I can answer this Nope, it's pretty much the same. I've tested CBR 192 mp3 against LAME --preset standard (~200kbps). It changes all the time and ranges from 41-43 (same as CBR)...

Ah, so it is shows 26-27mhz for 128kbps & 41-43mhz for 192kbps.. In another thread you experienced 1 hour less battery time when playing AAC, which is on average one or two mhz higher than mp3. Doesn't that mean 192kbps should drain the battery several hours faster than 128kbps as the mhz difference is so big?

[QuantumKnot]

What is the CPU usage for VBR mp3's? one would think it is a little higher..

I can answer this Nope, it's pretty much the same. I've tested CBR 192 mp3 against LAME --preset standard (~200kbps). It changes all the time and ranges from 41-43 (same as CBR)...

That sounds a bit high. I've played CBR 192 kbps on my 4G iPod and it's about the same (26-27). I believe I even tested with a 320 kbps MP3 and got something similar.

[Jojo]

Ah, so it is shows 26-27mhz for 128kbps & 41-43mhz for 192kbps.. In another thread you experienced 1 hour less battery time when playing AAC, which is on average one or two mhz higher than mp3. Doesn't that mean 192kbps should drain the battery several hours faster than 128kbps as the mhz difference is so big?

I've never said that 128kbps mp3s show 26-27! On 128kbps files, I get the same as for 320kbps files or VBR files or whatever. In terms of decoding the bitrate doesn't matter but rather the format!

I'm not sure if the number shown is in mhz...since 4G iPods and minis use different processors...but it would explain why a 4G iPod plays much longer than an iPod mini.

This post has been edited by Jojo: Dec 27 2004, 17:11

[QuantumKnot]

oh yeah, I forgot you had an ipod mini which seems to show a higher number than the 4G ipod.

[Cygnus X1]

oh yeah, I forgot you had an ipod mini which seems to show a higher number than the 4G ipod.

I can verify that as well: 128kbps AAC files generally show 36-40Mhz on my mini. Interestingly enough, my old ~220-240kbps AAC files encoded with PsyTel's "extreme" profile show similar values, usually staying around 38 or 40. So, if the numbers are indeed accurate, this might suggest that larger AAC files drain the battery faster not because of CPU load so much as more frequent HDD access. In any case, I haven't noticed a huge difference in battery life when using 128kbps AAC as opposed to my regular encodings (224kbps AAC and up): they both suck

[QuantumKnot]

I can verify that as well: 128kbps AAC files generally show 36-40Mhz on my mini.

What about a 128 kbps MP3?

[Cygnus X1]

I can verify that as well: 128kbps AAC files generally show 36-40Mhz on my mini.

What about a 128 kbps MP3?

A quick test with a 128kbps CBR MP3 file showed values ranging from 33-36, averaging around 34-35.

Other interesting tidbits: while idle (such as in a menu), the CPU value is a constant 20. During the first 3 seconds of a song (presumably, while it is being loaded into the RAM), the value is a constant 80.

[Busemann]

Just out of curiosity, how do you know it measures actual CPU usage?

[Jojo]

I can verify that as well: 128kbps AAC files generally show 36-40Mhz on my mini.

What about a 128 kbps MP3?

as said many times before, the bitrate doesn't matter...just the format. So a CBR 128kbps mp3 shows the same value like a 250kbps VBR file. The same applies to AAC.

Just out of curiosity, how do you know it measures actual CPU usage?

now that is interesting. You doubt your own words: http://www.hydrogenaudio.org/forums/index....ndpost&p=261688

I guess your next question would be 'how do you know that a higher value indicates more battery usage'...I mean, what else is it supposed to show...considering that it uses 20 when idle and a much higher value when playing mp3's and even higher when playing AAC and that it is connected to battery life? How do you know that it does not indicate CPU usage?

This post has been edited by Jojo: Jan 11 2005, 19:51

[Busemann]

hey take it easy jojo, it was just a question.. who knows if it is an estimate or *actual* usage..


jeez

[Jojo]

hey take it easy jojo, it was just a question.. who knows if it is an estimate or *actual* usage..

sorry, but sometimes I think you are just looking for any reason to 'protect' AAC and try to make facts sound like some random guess.

I don't think it's an estimate...maybe to some degree since it doesn't show 20,00001 or something. But what does it matter? And besides that, it seems pretty accurate since it often shifts around while playing a song...so it's never absolutely constant...

And of course, no one can answer you that except for the person who implemented that feature...but even if he would read this he'll most likely not tell us

This post has been edited by Jojo: Jan 11 2005, 19:59

[QuantumKnot]

as said many times before, the bitrate doesn't matter...just the format. So a CBR 128kbps mp3 shows the same value like a 250kbps VBR file. The same applies to AAC.

uhh, I wasn't asking about the bitrate but rather, what MP3s were using in comparison with AAC. Cygnus X1 only told me the one for AAC but not MP3 on the iPod mini and I'm getting the impression that they both show the same speed.

On the 4G iPod, MP3s us 27-29 while AAC uses 30-33. Is there this difference between the formats observed on the ipod mini (36-40)?

This post has been edited by QuantumKnot: Jan 12 2005, 01:10