by xylon on 12/9/18, 10:04 AM with 417 comments
by 013a on 12/9/18, 4:34 PM
I'm wholly convinced that the computing world is, in general, regressing. Audio has been a huge loser in this fight. Today, many modern phones don't ship with 3.5mm jacks. If you want lossless audio, you literally can't find it, even though there's zero reason companies like Spotify couldn't stream it when available (even if it costs extra) (Spotify literally asks artists to upload the masters when they publish, they have that data and then throw it away). Many artists don't even publish physical CDs anymore, so its a game of luck if they have a website where I can buy the FLACs/ALACs. And if you want the actual files to, you know, live your life in a completely legal way, those are gone.
Its more than just audio. Watching movies sucks; you now pay full price to effectively indefinitely rent movies, and have them taken away at any time. eBooks are the same and always have been; the world's oldest technology has been coopted by companies like Amazon to increase revenue, and there's practically zero competition. Applications suck; we're puking web and electron everywhere, eating up every conceivable megabyte of memory available literally only because developers are lazy, and now you're consistently asked to pay a monthly fee to access this functionality literally only because companies are lazy. Modern operating systems suck; restricting filesystem access, exposing proprietary application APIs which fundamentally make applications unportable and thus contributing to the rise of Electron/RN.
Somehow we took systems and workflows that were amazing throughout the 90s-00s and, in the course of a decade, completely ruined them.
by notacoward on 12/9/18, 1:02 PM
I'm not saying open source is bad and everyone should prefer free software. It's just a good example of the difference between the two in practice rather than in some debate about licenses and the abstract principles behind them.
by baxtr on 12/9/18, 11:22 AM
I enjoyed reading this. The whole time I thought however: that’s not an issue on my Apple devices. Then I read:
> If anyone reads this post I'm sure loads of people will tell me that my problems are all my own making and if only I invested in an iPhone all my problems would go away. Well you know what? APPLE IS A SYMBOL OF PRETENTIOUSNESS AND IGNORANCE - YOU DO NOT EVEN KNOW HOW YOUR PHONE WORKS - I DO NOT HAVE TO PAY A TAX TO APPLE TO LISTEN TO MY MUSIC.
Well, at least it usually simply works.
by dextersgenius on 12/9/18, 1:12 PM
This is inaccurate. Neutron music player bypasses Android's Media Player APIs and talks directly to your DAC and plays music without resampling (if the DAC supports it). Never had any audio popping or explosions using Neutron, and I've tried it on 6 different devices so far without any issues (LG G3, Nexus 6P, Nextbit Robin, OnePlus 3, Note 8, OnePlus 6). My headphones are a Beyerdynamic DT880.
by mindslight on 12/9/18, 6:21 PM
You've already got a real split supply with 2x 9v batteries. If you use that instead of deriving a virtual ground, you will save 9mA of quiescent current.
How purposeful is that whole low side duplicated circuit and why? [0] It seems like since you're using batteries, hooking signal ground directly to your ground and driving the output single ended would work fine. Or if you want to work towards being able to AC-power, then a differential input op-amp topology and still drive the output single ended.
Isn't there a vibrant cottage industry of external USB DACs and headphone amplifiers and whatnot? I'm more of a receiver+speakers type of a person, but I often see newly designed stuff for headphones.
[0] Driving both sides does get you the ability to swing the output a full 36 volts. But given that your goal is to cut the signal by 11 and also that by mixing both channels you can't actually do that lest you get crosstalk in the form of clipping, I don't think this is your goal!
by gcp123 on 12/9/18, 1:59 PM
by alkonaut on 12/9/18, 4:28 PM
I know it does work.
Would I prefer to have a more open phone and computer? Sure. At the cost of having to ever tinker with it? No.
I just switched my home built raspberry + HiFiBerry with expensive speakers to a simple closed Sonos system for the exact same problem the author had with linux, audio hardware, noise. Having something that just works and I don’t have to care how is a blessing and in the future that is where I’ll put my money. Ignorance is bliss.
by rayrrr on 12/9/18, 4:26 PM
by flocial on 12/9/18, 3:32 PM
by pdimitar on 12/9/18, 2:58 PM
I know several very smart and creative hardware technicians and system administrators. I admire them. I truly do. They are more independently thinking than most other programmers I know, are much more resistant to trends and fashion in technology, and rarely complain about the tech they must use. Throw Python, Perl, PHP, bash and what not at them, they can and will curse but will get the job done every time.
In that regard, they are hard-working and resourceful. And I have a lot of respect for them.
That being said, I cannot help but notice things common among all of those that I know or used to know:
- Apple hate without any regard for the advantages it gives you.
- They use 80-100% of their leisure time to tinker and work outside their main work hours. One of them recently got divorced by his wife. She said the guy barely spent any time with her. I don't know if their relationship wasn't sour before but still, it's partially indicative.
- Obliviousness how Android is not the end-all-be-all of free software -- because it's not; Android is PARTIALLY open-source and comes with a lot of strings attached. Flash a ROM and your warranty is voided. But to them, if you can tinker with your tech then you absolutely, factually, practically, and for all intents and purposes, are using free tech. Details like warranties and hours-long sessions to make basic functions work again be damned in the process.
- Complete disregard for convenience and utter lack of respect for others' time. As mentioned above, they usually spend most of their free time working, are not stressed out or burned out, are in good health shape, and find it mysterious that people might be exhausted and would just want to lie on the couch reading a favourite book -- or napping.
---
I am somewhat saddened to say the OP's article confirms my bias towards hardware technicians. He's undoubtedly smart and good at what he does but also heavily biased and lacks perspective.
I believe such resourceful people should be more open-minded. If he's reading this, I hope he takes it as a constructive feedback from the sidelines and not bashing. All I am saying is that their attitude is not helping them being taken more seriously. If you turn a blind eye to nuances (case in point: Apple hate from his side) then you are not objective in your discussions.
by projektfu on 12/9/18, 3:37 PM
I think the underlying point is very important. It does seem like quality has been sacrificed in many products to shave a few pennies off the BOM. Why should today's portable CD players be substantially worse than the ones available in 1995? I get that they are asking $25 instead of $115. But SONY and Panasonic have left the market, leaving it to the drugstore quality ones, and they don't appear to want to make a product that plays a clear sound.
As far as circuits are concerned, I can't help but notice that a TI TPA6100A2 costs $1.11 at qty 1, and outputs stereo from a 1.6-5.5V single supply drawing a max of 50mw. This would substantially reduce the weight and power draw of his mini amp. A couple rechargable AA batteries would do.
by praptak on 12/9/18, 12:22 PM
Can anybody explain that please? What is this dynamic load whose driving is hard?
by dejaime on 12/9/18, 12:45 PM
by nycticorax on 12/9/18, 6:31 PM
http://www.schiit.com/products/fulla-1
that delivers good audio quality (to my non-audiophile ears). The one I have is not perfect: there's some static when you turn the volume knob (maybe a good design tradeoff at the price, I don't know), and the metal edges on the enclosure were sharp enough that I took it apart and filed them down a bit (others might not have been bothered by this, but I was). But there sure isn't any background hissing like there is with the built-in headphone jack on my Dell workstation.
by squarefoot on 12/9/18, 1:13 PM
About the article circuit, I 'm not that sure the 5532 is a good part for headphone amps, unless the phones have a high impedance and resistance: they're intended as preamplifiers (and very good ones) so they hardly can supply the current to fully drive phones, though I guess they can still be ok for listening to soft jazz in a quiet room:)
by ricardobeat on 12/9/18, 1:12 PM
Coincidentally I was looking for something like an ipod shuffle this week and it’s become almost impossible to find, beyond some cheap thrift-store MP3 players.
by hunta2097 on 12/9/18, 1:35 PM
Literally never heard anything like this.
by Sidnicious on 12/9/18, 12:18 PM
a. What in these files triggers the explosion of noise (and whether other people have run into this or the author would mind sharing a representative problem file).
b. Whether an output attenuator (or "resistor") would work to quiet the noise (edit: the hiss, not the explosion) without a powered circuit.
by korethr on 12/10/18, 3:13 AM
The circuit design looks weird to me. Wouldn't it be better to use a inverting setup on the opamp to get get some attenuation, and then feed the gain stages into a high-current opamp like the 4556 as a buffer to drive the headphones, or perhaps a transistor that'd be good at the job? And why are there a pair of parallel gain stages on the ground line? Why not just throw your gain stages on the signal lines, ground ground and let it be ground?
1. https://nwavguy.blogspot.com/2011/08/op-amp-measurements.htm...
by Johnny555 on 12/9/18, 5:24 PM
Is it better to pay a tax to Google and not be able to listen to your music?
(I say that as an Android user who hasn't run into this problem, though 90% of my listening is through streaming services, I still have all of my CD's sitting on a hard drive, but they are all (as far as I know) available over streaming now, so I just use that)
by aetherspawn on 12/9/18, 11:30 AM
by temp0876456 on 12/9/18, 12:18 PM
I paid (a very reasonable) $2k for a set of reference headphones and a suitable amp/DAC to drive them.
I could have spent a lot more.
It is hard to drive a set of headphones with enough SQ to make them worthwhile.
You cannot plug a set of headphones into an Android phone and expect anything good to happen. And lots of DAPs suffer from software problems as well (lots of them are Android based...).
Getting good sound from a portable device, without software issues, is really only solved at the top end of the market ($700 - $4000).
It’s a bit easier if you want to use a computer as a source but far from cheap.
by AdmiralAsshat on 12/9/18, 5:18 PM
I do have the hissing sound he's described when using my ATH-M50's with my phone, and even my laptop. My solution (albeit more expensive than his DIY one) was to buy a portable DAC/amp, specifically a Fiio E17[0]. It connects via USB/micro/C, and it's worked on every laptop and Android phone I've had since the Samsung GSII. It also used to be able to dock with a separate desktop amp[1], until my dock broke. There are plenty of other portable, Android-compatible DAC's that are not too expensive. It's a bit of a bummer to feel like you need to pay for yet-another-piece-of-gear just so you can enjoy your audiophile headphones with your audiphile-quality music on the go, but sometimes that's the price of fidelity.
[0]https://www.amazon.com/FiiO-Alpen-Portable-Headphone-Amplifi... [1]https://www.amazon.com/FiiO-Desktop-Headphone-Amplifier-Dock...
by johncolton on 12/9/18, 1:44 PM
by twtw on 12/9/18, 4:36 PM
Seems like you would want to tie the input ground to the virtual ground between the two 9Vs and then maybe buffer the output ground, but this circuit doesn't do that - it just replicates the L/R chains twice between grounds.
Also, what's the point of adding the two 9Vs and then splitting them again with the two 1k resistors? Why not just pull ground from in between the two 9Vs?
by kazinator on 12/10/18, 7:17 PM
This is because the device is silly. Firstly, there is an unnecessary voltage divider for generating a reference voltage. It is too stiff (resistors are too low-valued at 1K + 1K). This provides a reference voltage that is conveyed to high impedance destinations. 18V into 2Kohms means this consumes 9 milliamps!
Secondly, the device contains a bizarre circuit that tries to amplify the GND-GND path, using two op-amps in parallel.
The numbers check out: 9 mA for the voltage divider, plus around 8 mA x 2 IC-s (typical value from datasheet): 25 mA.
Here is how we can fix things.
1. We have two batteries! That is a true dual voltage supply! We can just take the center reference voltage between the two batteries. By doing that we lose the voltage divider entirely, and save 9mA of supply current.
2. We can lose the GND-GND amplifier, and just provide a proper end-to-end galvanic ground connection. We eliminate a pair of op-amps, and thus an entire IC chip.
We're now down to the current draw for a single NE5532.
3. Use a different op-amp. The NE5532 isn't great for driving low impedances like headphones. There are op-amps that are better suited for this, and draw less quiescent current.
by dijit on 12/9/18, 12:46 PM
Well, neither do I to be fair. I seriously dislike iTunes but it can take more input than just the store..
Regardless of the official Music player app there are third party ones, some of which are super simple (presenting a http upload on the network and allowing playback of uploaded media) or, in my case, Plex (with a plex pass for local sync)
by makecheck on 12/9/18, 2:58 PM
by fsloth on 12/9/18, 2:12 PM
by amelius on 12/9/18, 11:52 AM
by IshKebab on 12/9/18, 12:10 PM
I actually bought Google's active USB-C audio dongle to solve this for my Mi A1. It works, but I think USB-C is not a good port for headphones. It isn't robust enough - if you wiggle it around as if it were in your pocket you easily get glitches.
by justin66 on 12/9/18, 6:12 PM
Maybe it doesn't like the metadata in those tracks?
and doesn't go on to discover the answer. If my ripping and encoding routine were doing something funky that causes problems I'd really want to know about it, and it wouldn't require me to debug Android as the author implies.
It's true enough that Android playback probably shouldn't be glitching, but if the track metadata is invalid or pushing the limits of what you can do, whatever he's doing to his tracks is likely to cause problems in places other than Android playback...
by mhh__ on 12/9/18, 7:16 PM
For the price that some people spend on their audio gear, you could buy several musical instruments: A far more fulfilling (and sociable) use of too much money IMO.
by setquk on 12/9/18, 5:14 PM
Oh no wait. No hole that decent headphones fit in.
Have a 6s with a dicky headphone hole and some nice HD25’s. Not sure where to go. I feel like my time on this planet is nearly over.
by DFXLuna on 12/9/18, 5:58 PM
by patchtopic on 12/10/18, 6:47 AM
by alexkavon on 12/9/18, 4:33 PM
by solemsigne on 12/9/18, 2:08 PM
Local electronics shops are indispensable when you get a whim to make a curious solution to your particular problem. By the time the parts get to you, you just can't help but think how much easier it would have been to just buy something and just feel dumb.
by kwccoin on 12/10/18, 3:21 AM
by iOSGuy on 12/9/18, 11:21 PM
by bascule on 12/9/18, 5:11 PM
https://www.audio-technica.com/cms/headphones/6117c014c965cd...
"The ATH-DSR9BT over-ear wireless headphones employ Audio-Technica’s new Pure Digital Drive system, which allows the headphones to operate without a sound-degrading D/A converter that conventional wireless headphones rely upon. Instead, the ATH-DSR9BT utilizes Trigence Semiconductor’s Dnote chipset to receive the digital audio signal from a Bluetooth wireless transmission, process and transfer it to the driver where the digital pulses of the chipset move the voice coil and diaphragm forward and backward to create the sound waves heard by the listener."
by Azerb on 12/9/18, 8:22 PM
by peter_retief on 12/9/18, 1:17 PM
by Vanayad on 12/9/18, 8:52 PM
by snarfy on 12/9/18, 12:03 PM
by duck2 on 12/9/18, 3:41 PM
by rdescartes on 12/9/18, 2:21 PM
by worldstarPanda on 12/9/18, 2:30 PM
by fipple on 12/10/18, 5:51 AM
by sasaf5 on 12/9/18, 1:33 PM
by chemmail on 12/15/18, 9:56 AM
by edoo on 12/9/18, 3:08 PM
by g5095 on 12/9/18, 1:02 PM