Static and Distortion with Mid to High Frequencies on Smart Phone Speakers

KingCash

New member
So i'm posting a sample of a song I've been working on.. it sounds pretty good on all my headphones, alternative stereo speakers and my studio monitors....



the only platform i can never get this, and most of my mixes to sound good on is phone speakers.... i can understand the responses saying "oh well phone speakers dont matter"

i would normally agree full heartedly.... except for the fact that all main stream music and productions that have 1 millions + hits/views/buys have nearly flawless mixes when it comes to phone speakers...

its seriously any mid to high range frequency instrument thats not drums or vocals.... and it just takes a dump on phone speakers :(

any tips on how can get this range of frequencies to sound great on phone speakers??????

:cry: PLEASE ANYTHING CAN HELP DIRECT ME TO A THREAD WHERE THIS WAS ANSWERED. ive been searching for a solution for a little over a year now all across youtube, gearslutz, and peoples inboxes and i still cant fix this :cry:

so ill be here for a minute

thanks.
 
I've wondered if there is some sort of mixing strategy for this as well but then I actually began to stream songs on my phone from these popular artists today to compare and those songs sound a bit fuzzy in phone speakers as well.
 
some of them do a little bit but for the most part i would say %95 of that music is clean enough to (never by me of course, i need bass) but its all still just clean enough for a lot of people to enjoy on the phone. i wonder how the musician eliminated distortion? was it during the mixing or mastering process? i wonder if they mix while it plays on smart phones or they are just so good and keen to know what sounds the absolute best that it just automatically adapts to a smart phone without having to try to make any sort of slight adjustment to tailor to this awful mono speaker platform.


i would also like ask is there a brand of synths one would need to be using that just dont have this distortion at high levels of volume or compression? or can the clean output in any range of frequencies be achieved from stock synths of any daw?

thanks if you can answer any of these questions.....
 
Bump.. seems like this should be a more common topic and issue.. with almost 2 billion people owning a smart phone. Music inevitably gets played on them.

I don't have an uncommon daw (Logic Pro X) and this isn't a uncommon circumstance for me.

Let me know if you have any potential solutions, blogs, guides, videos and or threads you can direct me to. ?? Thanks.
 
(Please don't "bump;" it is against the rules/tos here at FP!!)

It will be very difficult to make any one mix translate perfectly to every format and playback system (while that certainly is the "ideal" goal). Without knowing anything about your room, your set-up/equipment, plug-ins, and processes (and whether or not you have, or don't have good ears to begin with), it will be hard to diagnose or suggest anything, but bear in mind that even the major studios, record labels, and mastering houses have different mixes for different formats (for example, a mono mix, a vinyl mix, a TV mix, and yes, "mastered for iTunes"). You do want to have a number of speakers and playback devices to check mix translation, but there's nothing that says you can't have a specific mix and master for MP3, for instance. Odds are that will sound better on your phone than the mix you've just completed with no headroom from brick-wall limiting and all that bass that you were goosing for your sub system.

It takes time and effort, and a lot of going "back to the drawing board." If you are consistently finding that you have distortion or clipping or other issues, then you are probably doing something wrong or need to tweak your room &/or system. Yes, getting it right comes at the mixing stage, and yes, you should be able to get good, useable sounds, with any good gear or programs...

GJ
 
Last edited:
thanks for your wisdom.

i always just mix and master for mp3 and iTunes.

why do you say any bass would have an effect on the iPhone speakers?. i dont use a lot of bass. i tend to tighten my bass down after i move the mix from my HS7 monitors down to my old college dorm $50 logi tech speakers with a sub. that thing always makes me remove most the bass tbh.. and how do you think the brick wall limiting would cause the distortion? the vocals are the loudest part of the mix and they are crystal clear.. its only always the mid to high frequency synthesizer instruments that cause issues.
 
I didn't say "bass will cause iPhone speaker issues." Sorry if I wasn't clear; that was a hypothetical. And, it's possible that you may hear certain "distortions," because nowadays people are bass crazy, sometimes indiscriminately so, but that doesn't translate well to speakers that aren't designed to reproduce those frequencies. Same thing with over-limiting. It can kill your mix, and cause distortions and clipping when you think it "shouldn't." All hypothetical, based on the complete lack of knowledge available on you, your music, your environment, or your process... Unfortunately, I couldn't listen to your example last night, but I will try right now and see if it leads me in any other particular direction. In general though, mix towards the format you want people to hear your music in...

GJ
 
Last edited:
OK, sounds like you have maybe over-limited. You have zero headroom, yes? If I turn the volume up on my little computer speakers, it starts clipping and dropping out every other beat; sounds ok at about half the volume.

Don't mix or master so that you are just microscopically under the "red" zone on your meters for the entire track. You need to leave a lot more room for peaks, and do less "dynamic crushing," in general, or you are setting yourself up for clipping.

GJ
 
quick side question, what brand of laptop do you have and whats a more detailed description of "dropping out ever other beat" :hmmm:

i usually leave headroom and just mix without limiting for a while but i can see how since it can become a compression tool in the sense that you chop the tops allowing the lower to come up..... i also have the Stealth Limiter by I.K. Multimedia. with 16x oversampling (which i use)... i also have options to dither at 16bit and 24bit inside my limiter. but with or without any of these options selected in the stealth i can easily have it clip about 10db max and it won't cause any bad effects for all my stereo speakers and monitors.. all that being said, in my settings in logic pro x i also have it set to 24bit recording for everything, which is the max. so i would now have to ask what option inside the limiter is better to dither with, 16 or 24? or should i even dither at all on my master track output limiter?

i shall shut off the limiter for this mix, and reduce the mix until it is naturally not clipping without any limiters.. bounce it try it out, post it on SC and see if i get any better results. my hopes aren't super high for this test cause I've heard from others and from my own experience so far that the Stealth limiter is likely the best in the plugin limiter game... (yes, even better than the Pro-L by fab filter which <----which I have used for over 1 year.)
 
OK, I'm going to skip the middle section ^^^^ for now and encourage you to try what you discussed at the end (no limiting), but also to try settings at about half of wherever you had them before. Then compare the three tracks...

As to my laptop, it's irrelevant (I was listening on a desktop with outboard speakers, and I record mostly on a standalone digital recorder). What I meant by "clipping on every other beat," I meant, literally, every other beat or pulse ("boom, bap, boom boom, bap"), the speakers couldn't handle the peaking and would glitch-fart out. "boom plggggggsshh boom blshzzz%##^^"... So, I think you need to lower the final program volume considerably. But that is of course a total guess from only listening to one track!

GJ

PS-- I always dither final product and I usually record, and always mix/master to 44.1/16 (it is still industry standard, unless you're doing video production, and you're going to wind-up bumping everything down anyway).
 
Last edited:
So i'm posting a sample of a song I've been working on.. it sounds pretty good on all my headphones, alternative stereo speakers and my studio monitors....



the only platform i can never get this, and most of my mixes to sound good on is phone speakers.... i can understand the responses saying "oh well phone speakers dont matter"

i would normally agree full heartedly.... except for the fact that all main stream music and productions that have 1 millions + hits/views/buys have nearly flawless mixes when it comes to phone speakers...

its seriously any mid to high range frequency instrument thats not drums or vocals.... and it just takes a dump on phone speakers :(

any tips on how can get this range of frequencies to sound great on phone speakers??????

:cry: PLEASE ANYTHING CAN HELP DIRECT ME TO A THREAD WHERE THIS WAS ANSWERED. ive been searching for a solution for a little over a year now all across youtube, gearslutz, and peoples inboxes and i still cant fix this :cry:

so ill be here for a minute

thanks.

My take on this is that it is a combination of acoustic issues, latency, monitoring quality and EQ skills. The combination becomes difficult to target just with some mastering moves, so hence you end up with a mid range that is out of balance within the overall frequency range that is out of balance. To deal with this, fix it during the recording. Frequency sweep enough with good enough monitoring and ensure good phase. Then ensure you have low latency in the digital domain. Then when you mix and resume to mastering, involve frequency sweeping iteratively until the master is done. Then when you have all frequencies in good balance, you can apply zero phase frequency matching at the end as icing on the cake. I also think it is good to plan what sound sources you balance when and also that you have a strategy of what kind of loudness level you prefer on each sound source. Having sound sources quiet enough in the mix is just as important as having them loud enough. And also, the volume does not have to be constant, you can use volume automation to add specific dynamic impacts in order to get the right loudness levels on the various sound sources. It is stuff like this that you need to focus on. Also ensure you are balancing with multiple D/A converter + monitor/cans combos, since some of these issues might be discovered only on specific combinations. So, you will resolve this when you add these key improvements that are addressing the issue broadly enough.

Another thing to pay attention to is the impact of balancing against multiple unique D/A amp speaker chains at the same time, it is challenging to match that with cans on a single D/A. So if you can, introduce a couple of very big mains with the right amp and D/A and then a few near fields of various sizes with the right amp and D/A to those. Then align those first and then the mix against those. It is at this point when you also have great acoustics that you start to have very good performance from your balancing moves and can apply certain monitoring techniques you can't do without it. Ideally you don't want these output chains to be close to each other, instead you want each stream of frequencies from each output chain to drift a lot compared to the other frequency streams in the other output chains, so that when you then mix against them all at the same time you can quickly spot when you have the type of mid range that does not work when it translates across speaker systems out there. Some try to simplify it down to only being very selective about what monitors/cans to combine, but I would say it is better to let the frequencies drift from the D/A stage.

It is the sum of the above and balancing/reference checking with the right cans too (using multiple open back cans at the same time can further increase visibility) that just makes the mid range work automatically, because now more of what you hear by ear when you balance, ends up on the various playback systems. It does not hurt to do proper A/B on top..

It simply just sounds better then...
 
Last edited:
so the limiter wasn't clipping much at all, maybe 3db tops on the highest snare and bass kicks. heres the new one with no clipping and no limiting

https://soundcloud.com/the_sky_arcade/zero-limiters-on-this-track

when you dither is it only always on your master limiter? where else could you even dither? and what would be the reason you chose 16bit dither instead of 24 dither? or vise versa?

i have a fair size CPU and storage, so i decide to use 88.2/24bit... 24 bit really doesn't mess with my storage or CPU plus it seems to be the new DAW standard albeit not iTunes yet.

heres a big mystery..... now, i know you have to bump everything down to 44.1/16bit for distribution. but i wonder if you can capture small bits of higher fidelity when you mix, comp and EQ on, say, 88.2 then you bounce the track out inside the project at 88.2 still.... mix, master. maybe even bounce it all out again STILL in the 88.2 project.. thus potentially, for lack of a better term, "glueing" everything together at 88.2.... AND THEN you bounce it out and "bump everything down" to 44.1/16bit i feel like the compressing and EQ you did before the final bump down might have maintained more good artifacts.. instead of just doing the bump down with all the EQ's and compressors still active and "unglued".... :berzerk:
 
, i use all synths and samples besides my voice, and my voice never really distorts if i mix it right... i don't record any external guitars or drums.. so as for phase issues i would be surprised. latency, still not sure about that. i quantize my notes to the grid and control the attack. maybe EQ and monitoring and i need to sweep more. and or maybe i just have shitty synths? the distortion was mostly coming from the ESX24 inside logic pro x. electric 80's power chord preset with a little bit of altering..

i have in-phase by waves. i dont really know about phase or mess with it that much, does it affect software synths? and how do i apply zero phase frequency matching?

i do like the dynamics of volume automation. but i try my hardest to avoid doing it a ton cause the surprise of having a track get louder without a heads up can be off putting.

your second paragraph is confusing. I'm about do a little more research on youtube about dither 24 bit vs 16bit but i have a clarette 2pre audio interface and a pair of HS7 if that pertains to the d/a you speak of.
 
i feel like your terminology is a little off base. few questions if you get back. what are cans? whats D/A and whats A/B? i feel like in that second paragraph you were trying to say. mix on multiple sets of studio monitors?

does anyone here use logic pro x? does anyone know if the ESX24 is just not a quality synthesizer and won't ever be found in a professional track? or am i just not finding a way to remove these frequencies that just ruin phone speakers? its really driving me insane. of all the places to find this solution. i feel like the internet would have been a good place to find it.
 
I. Can't. Do it anymore... Here.

King Cash, I may have some insights for you. PM me if you'd like with your questions...

GJ
 
Oh man, I'm going to answer all of this. (because I want to be kind and help out)


so the limiter wasn't clipping much at all, maybe 3db tops on the highest snare and bass kicks. heres the new one with no clipping and no limiting


If you are talking about the last limiter that's a lot. Try to make it peak somewhere around -0.3 to -0.0 dBFS, then use the last limiter to add the ceiling at say -0.4 dBFS for the mp3 print and maybe -0.2 dBFS for the CD. It's extremely mild limiting.


when you dither is it only always on your master limiter? where else could you even dither? and what would be the reason you chose 16bit dither instead of 24 dither? or vise versa?


You should not dither at all, never, just the internal processing should do that. You should instead route the signal to the hardware domain and then print the audio straight to the final playback format.


i have a fair size CPU and storage, so i decide to use 88.2/24bit... 24 bit really doesn't mess with my storage or CPU plus it seems to be the new DAW standard albeit not iTunes yet.


You should use higher sample rate than that. 384 kHz sample rate at 32-bit float point precision is very ideal, but requires a state of the art production setup. If possible use 192 kHz sample rate at 32-bit float point precision.


heres a big mystery..... now, i know you have to bump everything down to 44.1/16bit for distribution. but i wonder if you can capture small bits of higher fidelity when you mix, comp and EQ on, say, 88.2 then you bounce the track out inside the project at 88.2 still.... mix, master. maybe even bounce it all out again STILL in the 88.2 project.. thus potentially, for lack of a better term, "glueing" everything together at 88.2.... AND THEN you bounce it out and "bump everything down" to 44.1/16bit i feel like the compressing and EQ you did before the final bump down might have maintained more good artifacts.. instead of just doing the bump down with all the EQ's and compressors still active and "unglued"....


It is ideal to zero phase frequency match the signal in the DAW, then route the signal out to the hardware domain where you apply hardware processing and then from there straight to the final playback format, so if the final playback format is 192 kHz @ 24 bit, that is what you record. So in practice when you are done you are basically just recording at various quality levels to be able to distribute the music at various quality levels. So for instance you create a CD 44.1 kHz@16-bit version, you create a 96 kHz@24-bit version, a 192 kHz@24-bit version, a 384 kHz@24 bit version, a 384 kHz@24 bit version that is aimed for mp3 distribution (with a lower ceiling)...


, i use all synths and samples besides my voice, and my voice never really distorts if i mix it right... i don't record any external guitars or drums.. so as for phase issues i would be surprised. latency, still not sure about that. i quantize my notes to the grid and control the attack. maybe EQ and monitoring and i need to sweep more. and or maybe i just have shitty synths? the distortion was mostly coming from the ESX24 inside logic pro x. electric 80's power chord preset with a little bit of altering..


You have latency issues, maybe 95% have that, even the mods in here.


i have in-phase by waves. i dont really know about phase or mess with it that much, does it affect software synths? and how do i apply zero phase frequency matching?


You can learn about zero phase filtering here:


https://www.youtube.com/watch?v=uPB2gdQtfvQ


You frequency match using the zero phase filtering technique.


i do like the dynamics of volume automation. but i try my hardest to avoid doing it a ton cause the surprise of having a track get louder without a heads up can be off putting.


Yes it can, but it is not an issue when you have the right peak characteristics and do it mildly, meaning when you have worked with the compressors in a good way during recording.


your second paragraph is confusing. I'm about do a little more research on youtube about dither 24 bit vs 16bit but i have a clarette 2pre audio interface and a pair of HS7 if that pertains to the d/a you speak of.


D/A means digital-to-analog signal conversion, I kind of just meant the unit that produces the output incl. the D/A. You need multiple entire signal chains after the DAW sequencer for an optimal monitoring.


i feel like your terminology is a little off base. few questions if you get back. what are cans? whats D/A and whats A/B? i feel like in that second paragraph you were trying to say. mix on multiple sets of studio monitors?


Cans, it's just a popular term for headphones. D/A is digital-to-analog converter. A/B is when you compare version A against version B.


Yes, use multiple entire discrete output chains both in parallel and in solo when you monitor.


does anyone here use logic pro x? does anyone know if the ESX24 is just not a quality synthesizer and won't ever be found in a professional track? or am i just not finding a way to remove these frequencies that just ruin phone speakers? its really driving me insane. of all the places to find this solution. i feel like the internet would have been a good place to find it.


It's not about the DAW itself, it is the combination of the DAW sequencer, the DAW sequencer version, the DAW audio interface, the sample rate, the computing capacity and how you use the DAW that is going to determine the performance.


The particular issue you are facing with the mid range is actually dealing with the question how to make a production hit? I mean this is what makes a commercial mix "pop" into a hit, the mid range is what has the impact and that requires access. When you work like you do, you have a lot of information that is missing in the audio. You need access and currently you don't have that, because of skills and because of production/recording/gear/setup/monitoring, but you can follow my advice in this post and the previous post I wrote to improve it quite dramatically.
 
Last edited:
Much of what DarkRed loves to post is a combination of his pet opinions/theories, wishful thinking regarding everyone's access to the most expensive gear available at any one time, and a sort of esoteric audio voodoo belief system. Too much to deal with here.

But a few things that must be addressed:

* "You should never dither." This is absolutely incorrect; it is the exact opposite. _Always_ dither when you are converting from a higher sample/bit rate down to 44.1/16. But only do it once, in the final formatting process (saving your finished mix/master to a final .wav file).

* 384/24???? Who is listening at 384/24??????? In my opinion and the opinion of many audio professionals, it is ok to try and record at as high a quality as makes you feel comfortable, but at a certain point there is a law of diminishing returns. I challenge anyone to tell me (in a double-blind test) which file they are listening to, 44.1/16, or 384/24. They (including DR) will not be able to tell the difference, guaranteed.

* "Use multiple discrete output chains when you monitor..." Again, most folks at FP are dealing with a laptop and _maybe_ some KRK's or headphones. Most people do not live and work at Oceanway or Abbey Road...

!!!!!!!!

GJ
 
a lot of great information here..

that zero phase video was really awesome!! i will start trying that.



ok dark I've made a lot of sense of most everything you've provided me with, but I'm going to try to simplify these last few things and make sure I can fully understand everything you've told me:::::::


1) so you say in your very first post "you need multiple mains. a good amp and good D/A"

----->then this is what my brain reads that as " you need multiple studio monitors, and a good pre amp which CONTAINS a good D/A provided on the inside of the preamp thus completing one of these "chains" you are referring to (i keep wanting to understand the D/A as a component that is on the inside of the pre-amp, so correct me if I'm wrong)

2) also when you said "having these chains AFTER the DAW sequencer"

-----> i read that as just meaning... you need to have your pre amp connected to your DAW.

3) lastly, when you say use the chains in "parallel and solo"

-----> i read as... parallel meaning play two pairs of studio monitors at once, and solo meaning just one pair at once. A/B now coming in for solo on swapping.. (now i know this one might be a bit too simple... but the reason I'm asking this is because i had a small feeling that parallel was referring to stereo and solo was referring to mono and then doing these actions on an actual setting change inside the DAW)

and that should finalize the rest of my questions on the topic for now.

again big thanks for all your help so far, and rhythmgj i understand if you dont want to help anymore. no problem at all.. its a very obscure/annoying topic.






 
Here is a trick to make you're bassline and kick sound better on your phone or any small speaker and still maintain a solid subbass!

 
Back
Top