Back to Article

  • NeatOman - Monday, June 02, 2014 - link

    Now I'm going to sit here waiting for the iPhone and Nexus 5/LG G2 audio test :) Reply
  • NeatOman - Monday, June 02, 2014 - link

    Sorry!!!! I looked back and it seems to have been in the initial test (LG G2 and Nexus 5). Reply
  • cheinonen - Monday, June 02, 2014 - link

    Since i updated this a bit I will retest my iPhone 5 soon. Reply
  • erikiksaz - Monday, June 02, 2014 - link

    Any hopes of the Moto X being tested? Reply
  • Lazy_Geek - Wednesday, June 18, 2014 - link has some numbers for your interest Reply
  • tigmd99 - Monday, June 02, 2014 - link

    Test iphone 5S please!!! Reply
  • bcg27 - Monday, June 02, 2014 - link

    Really good to see some more results using the Audio Precision. Any chance of seeing some tests of independent amplifiers or D/A converters? Reply
  • cheinonen - Monday, June 02, 2014 - link

    If the opportunity comes up then sure, but I don't know what the demand for that is. I'd be interested to see how well they compare to full on desktop DACs and Amps. I've been measuring a few of those recently. Reply
  • msackman - Tuesday, June 03, 2014 - link

    I've just received a CEntrance HiFi-M8 and can easily drive it from my Sony Z1 Compact. I would love to see a comparison. Obviously one expects that dropping that much on a external DAC and headphone amp would make a massive improvement to these figures, but it would be worth seeing that measured, as much as possible. Reply
  • MadMan007 - Monday, June 02, 2014 - link

    It's quite sad how bad the audio system in phones is, especially as they have largely replaced dedicated media players. I have no doubt that there have been lots of mp3 players or PMPs with much better audio systems over the years, so why can't phone manufacturers make something decent? Reply
  • lessthanjoey - Monday, June 02, 2014 - link

    Wait, what? What about these imperfections is even audible? Reply
  • zShowtimez - Friday, June 06, 2014 - link

    Plug your phone into a decent mid to high end speaker set up. Its so easy to hear the difference. It's part of the reason i went with my htc one over a galaxy s4

    Similar issue with onboard audio vs a discreet card imo
  • Streamlined - Monday, June 02, 2014 - link

    Well, it's official. According to hard data and not just opinion, the HTC One kicks the Galaxy S5's butt. If there is one group in the tech world that purchases inferior hardware due to marketing hype, it's Samsung fans. Reply
  • hero4hire - Tuesday, June 03, 2014 - link

    But just Samsung fans... No other group in the entire world. Nope. Just the Samsung fans. Not a single purchase is bought elsewhere due to hype. Previous comment is not baiting. There is only HTC and Samsung and the best marketing in history is the most beloved and most popular Samsung company. It's official that nobody would try to invoke some other company to compare to the best marketing that is Samsung. No arguments here, Streamlined is 100% wanting to keep us focused on HTC & Samsung. Reply
  • theduckofdeath - Wednesday, June 04, 2014 - link

    Yup, because the camera in the HTC phone is a technical marvel. :) Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    It actually is. Wait until you take a picture in an almost dark room and it comes out perfect. Or if you are in a pitch black area and you take a picture with the flash and it is 100% clear. or you take a picture in a low light resturant and its clean. Or a pic at a concert and its got mid range lighting at best and its clear aswell! I have found the camera to be great in almost all lighting scenarios. Now it doesn't take as beautiful pictures as my EVO 4G LTE did in bright lighting, but its pretty close, and considering how many other scenarios the camera is useful in, the pro's far outweigh the cons. Reply
  • theduckofdeath - Wednesday, June 11, 2014 - link

    It's an extremely limited camera, made for one of the least common usage scenarios. For everything else, it's far below average. Reply
  • The1Metallian - Friday, June 06, 2014 - link

    I have a GS3 and I'm ready for an upgrade. I would buy the HTC M8 in a heartbeat, but the camera broke the deal. I don't expect the camera to be DSLR quality but at least up to par with any other smartphone in the market. I want to know that wherever I am I will be able to take a good picture if needed. The Sony Experia Z2 won't be subsidized by any provider. What am I left with? (don't say iPhone) If the LG G3 doesn't deliver when released, I'm getting the GS5. Samsung competitors need to get a clue Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    The camera is actually not bad at all. I'm not sure why everyone hates on it, but it takes great pictures in basically every lighting. And the quality is more than enough for any type of social sharing... Reply
  • gevorg - Tuesday, June 03, 2014 - link

    What about jitter? Reply
  • cheinonen - Tuesday, June 03, 2014 - link

    Measuring jitter would require these to have a discrete line out. Any jitter present off the DAC is being masked by the noise from the amplifier and cannot be reliably measured. Jitter would likely be falling around -110dB or below, but that would be swallowed up by the noise floor here. Reply
  • edzieba - Tuesday, June 03, 2014 - link

    What /about/ jitter? Unless your equipment is seriously faulty, jitter is a non-issue. It's the "this is TOTALLY why you need to spend more on our magic cables than on your equipment!" for the digial age.

    You can feed SPDIF over a coathanger ( and experience no jitter issues (or any other issues).

    When outputting an analog signal (i.e. the amplified headphone out), jitter is a total non-issue.
  • althaz - Tuesday, June 03, 2014 - link

    I'd also love to see a test of the Lumia Icon or the 1020. I haven't heard them, but the 920 at least doesn't sound as good as the iPhone or the HTC you've tested above. Wouldn't switch OS for a better sound, but would buy a newer phone. Reply
  • willis936 - Tuesday, June 03, 2014 - link

    One thing that I think is important here that is rarely measured is the amount of noise from the radios that gets rectified in the amp. We've all heard the characteristic tick tick tick from when we laid our phone next to the stereo. Some phones are better than others in terms of not having audio affected by the phone's radios. Reply
  • name99 - Tuesday, June 03, 2014 - link

    That was a feature of GSM and the particular way it timesliced. It didn't occur with CDMA devices, and it shouldn't occur with any modern phone that's using WCDMA, HSPA, or LTE. Reply
  • DoctorG - Tuesday, June 03, 2014 - link

    I get radio noise all the time with my CDMA Galaxy Nexus (and Palm Pre before that.) Reply
  • willis936 - Tuesday, June 03, 2014 - link

    I get it loud and clear from my razr (XT912) when I put it near headphone amp or stereo that my computer uses. I'm also moving to an HTC One (M7 Verizon) and it too has the issue. I haven't tried turning off CDMA altogether but since I have LTE coverage and the noise only comes when there's data usage I'm going to conclude it's the LTE. They can do clever shielding and filtering tricks to keep it from affecting the amp inside the phone. You literally can't do anything besides keep it in your pocket to avoid the noise in other amps. A faraday cage at 700 MHz is nearly air tight. Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    On my M8 I haven't noticed the audio being affected by the radio. Actually all 3 of my HTC's (HTC EVO 4G, EVO 4G LTE, One M8) haven't had this issue. They do have a bit of feedback if plugged into the charger and the battery percentage is high. But other than that their sound has been pretty pure, and better from generation to generation! Reply
  • Ammar666 - Tuesday, June 03, 2014 - link

    Which Galaxy S5 did you test? The Qualcomm one or the Exynos one. Because the Exynos one comes with a Wolfson DAC I think. Reply
  • theduckofdeath - Wednesday, June 04, 2014 - link

    I would have guessed they tested the common Snapdragon based one. But then again, that one uses the same DAC as HTC as far as I know which would mean that either the test is flawed or Samsung screwed up the software on the GS5. I remember it was the complete opposite with last years generation, where the GS4 had much cleaner sound than the old "One", even though those two also used the same DAC. (according to measurements done by GMS Arena)

    Personally, I'm not a big fan of these type of pure technical test to find the best sounding device. You need some sort of human tap-in, like an extensive blind-test, to get a real life reference as to where they really stand. Especially when you use hyperbole to the extreme like Mr. Heinonen is.
  • cheinonen - Wednesday, June 04, 2014 - link

    It doesn't matter if they're using the same DAC. They provide no line-out functionality so you're subject to the limitations of the amplifier driving the headphone jack. In this case, the amplifier that HTC has chosen to use is far more powerful, with less noise and crosstalk, than the one that Samsung chose. It's impossible to isolate a single component, like the DAC, and expect everything that uses it to have identical performance. It's part of a system and everything else in the system has to perform at a level for it to function right.

    For the subjective testing, I'll leave that to Anand and Joshua who are spending far more hands-on time with the phones than I am. I'll run the numbers and provide a straight analysis of them.
  • theduckofdeath - Thursday, June 05, 2014 - link

    The reason they need subjective listening is because they insist on using hyperbole to differentiate devices. Based on data output from a metering device.

    Here's an example why actual human testing is needed:

    Lambasting something as horribly bad without actually making a real life comparison is a typical nerd error. "The numbers are superior, therefore it must sound superior!". The human ear is really limited in what sounds it can pick up. It's not nearly as accurate as Chris appears to think as very few adults, especially those who frequently listen to headphone music, can hear anything even near 20kHz.
  • synaesthetic - Saturday, June 07, 2014 - link

    Sighted listening produces biased results. Measurements are verifiable, demonstrable and repeatable. Subjective listening tests are just that, subjective and thus useless to anyone but the person doing the listening.

    This is why I appreciate actual data. I can look at a frequency response graph for a headphone and compare it to another headphone I know I like, and I can get an idea how the other headphone sounds without having to decide which subjective listener's ears I'm going to trust.
  • theduckofdeath - Wednesday, June 11, 2014 - link

    The point I made is, measuring things that are probably not even audible to a human being ,and use that as a review for audio quality is pretty pointless. Then you've really missed the whole point of being a product review site for devices intended for regular human use. :) Reply
  • apertotes - Tuesday, June 10, 2014 - link

    but you did not answer. Which Galaxy S5 did you test? Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    The M8 sounds alot better than the GS5. Its been determined so many different times. I own an M8 aswell and I have to say the sound quality is like none other. Reply
  • edzieba - Tuesday, June 03, 2014 - link

    I'd like to see Output Impedance measurements. If that heads north of 2ohm, then that's a good red flag for things having gone seriously wrong. Reply
  • Rezurecta - Tuesday, June 03, 2014 - link

    No it is not. Where did you get this misinformation? Reply
  • synaesthetic - Saturday, June 07, 2014 - link

    Output impedance is THE single most important factor for how a given driver will sound with a given analog output. When the 1/8th Rule is violated, the driver will lose some or all of its electrical dampening. This usually tends to kill bass and warp the rest of the headphone's frequency response, making it sound abnormal.

    The fact that most manufacturers, reviewers and the like never bother to check output impedance is very, very annoying and makes headphone selection needlessly complicated. Very low-impedance headphones, such as Ultimate Ears IEMs, will not sound "normal" out of a source with an output impedance over 1.6 ohms.

    Properly designed amplifier sections should have an output impedance as close to zero as possible so that they work properly with all headphones.
  • Jimster480 - Tuesday, June 10, 2014 - link

    Well my M8 has worked fine with 5 different headphones I have tried it with. 3 Different sets of IEM's (Sony Earbuds, Skullcandy IEM's, and Bose IEM's). And a Sony over the ear aswell as a Panasonic RP-HT360 and it sounds very clear and rich on every set. No set is too loud nor does the sound reproduction sound incorrect or flawed. Reply
  • CSMR - Tuesday, June 03, 2014 - link

    I would also like to see output impedance.
    Especially as the tests now use dummy loads, which will mask the effect of high output impedence on frequency response.
  • Doroga - Tuesday, June 03, 2014 - link

    Yes, would like that one for both phones and motherboards reviews (at least 1 midrange board per brand to get an idea). Reply
  • pelms - Tuesday, June 03, 2014 - link

    Very interesting. I'm hoping phone manufacturers will come to treat audio quality with the attention they (and the press) pay to camera quality.
    It would be interesting to test the iPod Classic and Touch to get a baseline for comparison.
  • Jimster480 - Tuesday, June 10, 2014 - link

    Well I have a first Gen iPod touch with a Wolfson and I have to say there is a night and day difference to the M8. It sounds better with all forms of sound. Its alot clearer and the sound stage is alot deeper. Reply
  • Anakha - Tuesday, June 03, 2014 - link

    Nice article! :) One question, do you feel like voiding your warranty and converting the M8 to the Harmon/Kardon version? The details for the conversion can be found here: I am curious to see if this would make any difference? Reply
  • jk1 - Tuesday, June 03, 2014 - link

    thank you for this article. i would really like to see an evaluation of audio become part of every anandtech smartphone review - both as a music source and for telephone. these phones are routinely used as music sources, but that function is typically neglected in favor of exhaustive analysis of the cameras.

    if you do decide to do evaluation of dedicated portable audio sources, let me nominate the ibasso devices - dx50 and dx90, as well as the fiio's- x3 and x5. these receive the most discussion over at head-fi, so i'm sure there would be interest.

    btw- although i check anandtech's smartphone page regularly, i actually came to this article through a link posted in a thread at head-fi, a thread devoted to the htc one m8 as an audio source. i'm sure more audio reviews will generate more links, there and elsewhere.
  • Jodiuh - Tuesday, June 03, 2014 - link

    Thank you AT, I knew this already, but wanted confirmation. Can this thing power some AKG Q701's? Reply
  • synaesthetic - Saturday, June 07, 2014 - link

    Not very well. Reply
  • Zoomer - Tuesday, June 03, 2014 - link

    Testing more phones will be helpful! One M7, Nexus 5/7/10, Xperias, the Apples. Reply
  • The1Metallian - Tuesday, June 03, 2014 - link

    Thanks for the article! Now - is there a test of the quality through Bluetooth? I hear all my music on my GS3 through the sound system in my car via Bluetooth and I've noticed that music sounds better when coming from my wife's iPhone. I am now evaluating a phone upgrade and music sound quality is high in my consideration, but always staying with Android. Reply
  • bigboxes - Tuesday, June 03, 2014 - link

    I never use the headphone out. I use Bluetooth to connect my Note 3 to my Alpine and the sound is great. The only issue I noticed is an occasional tick in the sound which I assume is when I receive data (texts, e-mails, etc.). Not sure how to eliminate that. If anyone has a suggestion, please let me know. Reply
  • The1Metallian - Tuesday, June 03, 2014 - link

    @bigboxes I never have any noise due to notifications. I use a sound profile manager so when I start the music I choose the profile that has notifications muted, media almost all the way up, phone half way up (for emergencies!).

    I must add that the poor quality is more evident depending on the type of music being played. When the music is mostly guitars, bass, drums and vocals (metal) I don't notice it, but if keyboards come up front then it sounds kinda garbled. The same with sustained cymbals.

    You have a Note 3. That device is newer than the GS3. Maybe it has better bluetooth?
  • bigboxes - Wednesday, June 04, 2014 - link

    Probably. However, I came from an HTC 4G EVO LTE. The Bluetooth on that model sucked. I RMA'd a Jawbone headset thinking it was bad. When I got the Note 3 I realized it was just the HTC handset and not the Jawbone. It could be the Bluetooth or it could be the antennas or both. I'm not AnandTech where I have labs and access to different equipment. I can tell you that the audio over Bluetooth sounds amazing. Of course, the Alpine could have something to do with that. :) Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    I had an EVO 4G LTE and I never had any issues with the bluetooth! My father has one too and his bluetooth never acts up. it could have been your phone specifically! Reply
  • cheinonen - Tuesday, June 03, 2014 - link

    Bluetooth testing is something I plan to add in the future. I need to decide what I want do with it, but likely something similar to this. There is also a legibility test for talking that can be done over Bluetooth which will be interesting to see. Hopefully Bluetooth doesn't have issues with USB control as analog does, but I will test both ways to be sure. Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    I think that the sound via bluetooth would depend on the DAC of the device on the receiving side? I'm pretty sure bluetooth is a digital signal and therefore the receiving device would have to convert it. Reply
  • haukionkannel - Tuesday, June 03, 2014 - link


    It would be nice to see some pro portable flack player compared to these phones.
    What is the best audio phone and are they even near or a far, far away distance compared to high class portable audio players or real hifi DACs?
  • cheinonen - Tuesday, June 03, 2014 - link

    If you want comparisons to some really high end gear, I've been testing the Auralic Vega DAC ($3,500) and the Auralic Taurus headphone amplifier ($1,900). Since the headphone output here is a combination of a DAC and headphone amp, these numbers aren't exactly comparable.

    The Vega has 0.000345% THD+N, a frequency deviation of 0.002 dB, Dynamic Range of 125 dB, and Crosstalk of -130 dB.

    The Taurus does over 5V into 15 Ohm, and nearly 20V into 33 Ohm, 150 Ohm, and 330 Ohm. It has 0.0004% THD+N, deviation of 0.007dB, 124 dB of Dynamic Range, and -70 dB of Crosstalk. IMD, which isn't measured on the phones, is also below -100 dB at maximum gain.
  • haukionkannel - Thursday, June 05, 2014 - link

    Thanks to this info! But the comparison to real portable devices would be more important. I am in situation where normal phone does not produce good enough sound compared separate mobile music player. I am just wondering if we are getting near the situation where phone is good enough compared to dedicate high-end mobile music player? Reply
  • cheinonen - Thursday, June 05, 2014 - link

    I don't really use any portable DACs and amps, either for work or for my daily life. If I get my hands on one I will measure it, but it's an area I don't really venture into. Reply
  • Ktracho - Tuesday, June 10, 2014 - link

    Check out the Chord's Hugo DAC/amp. It's a small device with built in rechargeable battery, but has great sound quality. Reply
  • willis936 - Thursday, June 05, 2014 - link

    Are you using those to measure fluctuations in gravity? Reply
  • estarkey7 - Wednesday, June 04, 2014 - link

    Why isn't there any test for High Definition Audio? This is where the difference really matters. I would love to see the true frequency range of my LG G Flex on the Sprints HD network! Reply
  • orb242 - Wednesday, June 04, 2014 - link

    My HTC One M8 does distort certain audio at full volume... turning the volume down helps Reply
  • cheinonen - Wednesday, June 04, 2014 - link

    I only measured the extreme distortion with BoomSound enabled at maximum volume. It also would have some clipping/distortion when using USB control, as I discussed, but the regular volume control stops before that level for me. Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    I have only found this to be the case if using boomsound. I play it at max volume alot of the time without boomsound on and I don't seem to have an issue. I have a high end audio system in my car aswell, so I tend to notice artifacts. Reply
  • Vishalaestro - Thursday, June 05, 2014 - link

    Still both phones doesn't even come close to the legendary Nokia n91 Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    What? Reply
  • Cman775 - Friday, June 06, 2014 - link

    Could you test the Xperia Z2? From what I have heard from other sources it doesn't pump the same amount of power as the HTC M8. It does output a cleaner and more detailed sound signature (what sources claim). Reply
  • Jimster480 - Tuesday, June 10, 2014 - link

    Where? Reply
  • synaesthetic - Saturday, June 07, 2014 - link

  • synaesthetic - Saturday, June 07, 2014 - link

    The 20Hz drop is probably due to DC filtering caps on the output. Reply
  • NonSequitor - Monday, June 09, 2014 - link

    What is the repeatability of these measurements? The tables include many digits of precision, but if the measurement isn't that repeatable it's really quite misleading. I also suspect you are dropping trailing zeros. You might want to put ranges on those numbers and remove irrelevant digits. Reply
  • beck2050 - Wednesday, June 11, 2014 - link

    I can't even turn my Samsung Galaxy Note 3 with headphones all the way up so I'm not sure what this article is claiming. Reply
  • Nucleuscore - Wednesday, June 18, 2014 - link

    Hi Chris

    how where your settings inside the APx software? I have made measurements with a HTC One, LG Nexus 5, Samsung Galaxy S4 and Sony Xperia Z. The measurement was loaded with 33 Ohm. My results were almost equal to yours, but the wattage was much smaller.

    Sorry for the bad english

  • Haftarun8 - Wednesday, June 25, 2014 - link

    This is great! It's hard to find actual audio reviews of mobile headphone outputs with accurate test data. It'd be interesting to compare the quality via bluetooth streaming as well to objectively see the differences. PLEASE do more of these with more phones (in particular the up and coming Galaxy Note 4). Reply
  • 2kfire - Saturday, July 05, 2014 - link

    I didn't see the built-in speaker loudness tests in the original reviews so I figured they'd be here, but I don't see them here either.
    Any reason for skipping those tests for these phones? I'm interested to know if the S5 is competitive now, and if the M8 is noticeably louder than the M7. I know I could go to a retailer and check vs. my M7, but it'd be hard to tell with all the noise...
  • ayork - Friday, August 08, 2014 - link

    Did you try a test while the phone is downloading data? When the 4G radio is working the phone's sound is so bad I am tempted to never stream audio again. Reply

Log in

Don't have an account? Sign up now