I can't hide my enthusiasm with Star Trek in the title, but recently I got hooked with Nokia Nseries' WOMWorld, so they've loaned me a N95-1. Might not sound interesting since it's been out for some time, but I never got a N95-1, and just skipped a generation and got the N95-3. The first thing that crossed my mind was to try is a battery draining competition using music playback, because almost every other comparison has been blogged to death. I want to get cold, hard numbers in standardized conditions. What I'm really trying to see here is that if the 26.3% battery increase translates practically in terms of music playback.
Offline Mode, Bass Booster:
N95-1 (950 mah, BL-5F): 4 hrs 31 mins
N95-3 (1200 mah, BL-6F): 5 hrs 24 mins
N95-3 (950 mah, BL-5F): 4 hrs 12 mins
Offline Mode, Default Equalizer:
N95-1 (950 mah, BL-5F): 4 hrs 41 min
N95-3 (1200 mah, BL-6F): 5 hrs 23 min
N95-3 (950 mah, BL-5F): 4 hrs 8 min
Offline Mode, Bass Booster, Headphones:
N95-1 (950 mah, BL-5F): 6 hrs 7 min
N95-3 (1200 mah, BL-6F): 7 hrs 13 min
Offline Mode, Default Equalizer, Headphones:
N95-1 (950 mah, BL-5F): 6 hrs 17 min
N95-3 (1200 mah, BL-6F): 7 hrs 8 min
Network: AT&T, Bass Booster
N95-1: (950 mah, BL-5F): 4 hrs 15 min
N95-3: (1200 mah, BL-6F): 5 hrs 21 min
N95-3: (950 mah, BL-5F): 4 hrs 7 min
Network: AT&T, Default Equalizer
N95-1: (950 mah, BL-5F): 4 hrs 13 min
N95-3: (1200 mah, BL-6F): 5 hrs 8 min
N95-3: (950 mah, BL-5F): 4 hrs 2 min
Network: AT&T, Bass Booster, Headphones
N95-1: (950 mah, BL-5F): 5 hrs 41 min
N95-3: (1200 mah, BL-6F): 7 hrs 1 min
Network: AT&T, Default Equalizer, Headphones
N95-1: (950 mah, BL-5F): 5 hrs 57 min
N95-3: (1200 mah, BL-6F): 7 hrs 0 min
I decided to also put the BL-5F battery in the N95-3 to standardize the test as much as possible.
I couldn't test the BL-6F inside the N95-1 because it doesn't fit, even without the battery cover.
Conclusions:
All percentiles are averages:
1. The N95-3 lasts 19.3% longer than the N95-1 in battery draining using the music player.
2. a. Using the headphones in N95-1 increases battery life by 36.1% comparing to not using them.
b. Using the headphones in N95-3 increases battery life by 33.2% comparing to not using them.
3. a. Using the default equalizer in N95-1 increases battery life by 2.6% comparing to bass booster.
b. Using the default equalizer in N95-3 increases battery life by -1.4% comparing to bass booster.
(meaning that bass booster in N95-3 is 1.4% more efficient than default equalizer.)
4. a. Switching to offline mode in N95-1 increases battery life by 7.6% comparing to general mode.
b. Switching to offline mode in N95-3 increases battery life by 2.6% comparing to general mode.
5. The most interesting part of these experiments show that using the same battery (BL-5F) in to both handsets, the N95-3 drains 6.5% faster than the N95-1. This shows that although the battery power has increased by 26.3%, it doesn't mean that the N95-3 (BL-6F) will last 26.3% longer.
Overall
So if the battery power has increased by 26.3%, but in music playback it shows that the N95-3 (BL-6F,1200 mah) lasts 19.3% longer than the N95-1 (BL-5F, 950 mah), therefore 7.0% is missing. Conclusion 5 shows that the N95-3 hardware drains 6.5% more than the N95-1 hardware, which is really close to that 7.0% figure calculated.
Disclaimer:
As with any experiment, there is always bound to be error. Take the results here with ± 5% error.
(for example in conclusion 1, 19.3% x 0.05 = 0.96% , therefore the result should be considered between 18.3% and 20.3%)
Misc.
The settings for the N95-1 (BL-5F) and N95-3 (BL-6F):
Headphones: Sony MDR-V150
Music Player Volume: 100%
Repeat: All
Visualization: Album Art (if present for a particular song)
Music : 69 songs @ 5hrs 29 min
light sensor: 50%
Power Saver Time-out: 60 secs
Light Time-out: 10 secs
Active Standby: On
Theme: Nokia Nseries3
Network Mode: GSM only
Operator Selection: Automatic
Memory Card Used: Sandisk Microsd 2GB
Screen Saver: Date and Time
Bluetooth: Off
No other apps running simultaneously.
Any other settings are default settings to the best of my knowledge.
Photo Credit: Physlink
Cool!
I've also made some similar benchmarks but only with the N95-1 to find out what effects the different sound transport modes (A2DP vs. wired headphones) and equalizer / stereo widening settings have on the battery life. This was of particular interest because Windows Mobile users know well enough A2DP requires a lot of CPU time on WinMo, resulting in far faster battery drain.
See http://www.howardforums.com/showthread.php?t=1273334 for the results.
Posted by: Menneisyys | November 26, 2007 at 12:41 AM
BTW, the slightly lower battery life on the N95-3 may be caused by the somewhat bigger RAM.
Some devices (for example, the Pocket Loox 720) aren't affected by the additional battery usage caused by the bigger RAM; some other (for example, back in 2000...2002, the iPAQ 36xx series expanded to 128M) exhibited a very strong battery life degradation because of the additional RAM needing constant refreshing. With the latter, the difference was some two-fold (!), which clearly shows engineering deficincies.
All in all, the 5% difference (if it's only caused by the additional 64M RAM) isn't that bad IMHO.
Posted by: Menneisyys | November 26, 2007 at 03:18 AM
Yes the more RAM is a ripe candidate for the additional power drain.
Now that the new firmware for the N95-1 is out, I'll try to revise and hopefully add new benchmarks for the new firmware. I have sent the phone back, so I'll need to get a hold of another one.
Posted by: Amir | November 26, 2007 at 05:30 PM
It would be interesting to see what would have happen with internet radio using gprs/edge/3g when headphones are plugged :)
Posted by: CanRaps | November 27, 2007 at 12:37 AM
Now that the new 20.0.015 firmware for the N95-1 is out, which according to NOKIA offers 30% better music playback time - it would be interesting to see the test performed with the new firmware as well.
Theoretically it should bring the N95-1 on par with the N95-3 even with the smaller battery
Posted by: N95 user | November 27, 2007 at 05:08 AM
@ CanRaps: Internet Radio is more tricky than using the media player, because there are times when for no reason the connection would disconnect, and that would ruin the process.
Also with these tests I made sure the same music is played over and over again each time, and that's not possible with Internet radio.
3G isn't applicable for N95-1 in the states, so I couldn't do a 3G comparison.
Posted by: Amir | November 27, 2007 at 08:31 AM
@ N95 user: I'll gladly add the numbers that the new firmware puts out soon.
Posted by: Amir | November 27, 2007 at 08:52 AM
CanRaps, Amir: please see http://www.howardforums.com/showthread.php?t=1286676 for more info on my benchmarks.
Posted by: Werner Ruotsalainen | December 01, 2007 at 10:36 PM
On what is your error estimate based? Error analysis? Statistical analysis? What firmware did you use? Especially on the N95-1 (Classic) the new v20 firmware makes a difference.
In any case, the N95-8GB figures with the BL-5F seems to indicate the bigger screen might take a bigger powerdrain. Unless you switched the screen completely off?
Posted by: snoyt | December 04, 2007 at 06:05 AM
I haven't done any standard error analysis, I'd say 5% error is just an educated guess. It was not in my ability to do at least 3 tests for each data entry. The firmware used was the old V12 if I'm not mistaken, not the new one. When I published this two days later the new firmware came out. I'd love to give this another try with the new firmware on the N95-1.
Posted by: Amir | December 08, 2007 at 03:14 PM