Posted by pacificat0r 9 hours ago
All my audio equipment was on the same UPS (and therefore outlet) as my gaming PC.
The result is that any time a particularly stressful game would be open, I'd get buzzing in the speakers. (Especially if the framerate was at 360) If you ask audiophiles online they will swear up and down that a cheater plug, balanced cables, or optical isolation will fix it - that will not fix it. It's not a ground problem. It's not coming from the connection from the PC to the DAC - it's a power issue.
It seemed almost inconceivable to them that the problem was EMI from the computer making it into the equipment.
I temporarily got a double-conversion UPS (converts AC to DC to AC again) and housed the audio equipment on that instead (separate from PC) Lo-and-behold the noise was completely gone.
However, those UPS are extremely expensive, and far worse they're very loud because the fans run constantly.
So, I went with a simpler alternative. Just get a power strip and plug all the audio equipment into that on a different outlet. That reduces it massively. You can also get some strips that are designed to reduce EMI, but I haven't felt the need as of yet.
You may also be able to solve the problem with a simple common mode choke, either the clip-on type, or a toroid that you wrap the cable through a couple-few times. https://palomar-engineers.com/rfi-kits/acdc-power-line-choke...
I agree however that indiscriminately throwing ferrites at problems can be a good solution!
[Youtube] "Hitler fails radiated emissions" - Orin Laney: https://www.youtube.com/watch?v=eeo8ZZTfwZQ
Your problem of an AC power supply not sufficiently filtering out high-frequency noise from mains is exceptionally rare to the point that yes, I also don’t believe that was the correct diagnosis of your issue.
And one of those, even the cheapest ones, run for about ~$900. And they are LOUD.
How efficient could a small AC->motor->generator->AC chain be with a modest flywheel mass to provide cycle-to-cycle stability?
Could it ever make sense to put one of these after a standby UPS so the output is always filtered by the motor-generator but the UPS only has to kick in for outages?
Lifting the ground on my studio monitors absolutely fixed my noise problems. I run them off a MiniDSP 2x4HD, so other sources like EMI aren't really a factor.
The problem I have with a double conversion UPS is that it isn't an ideal sinusoidal source. It implies it is on the tin, but when you've got protected loads with PWM power delivery slamming around 1+ kilowatts, there's no way to guarantee a smooth waveform with a typical ~2500VA unit. Directly passing through to the grid could provide cleaner power under the most transient conditions.
I have all connected to the the same power circuit and with a Elektron Digitakt as audio device and have zero noise.
With audio devices powered by USB there is a lot of noise.
My audio equipment is not connected by USB. It's connected by optical (TOSLINK) to an external DAC. TOSLINK is not great, but it shows that it is not a USB noise problem.
That shielding was carrying noise from my PC, through the network switch, to my raspberry pi that I used for music streaming. Absolutely nuts.
I swapped to unshielded ethernet cables and it went away.
As for building wiring, this issue has persisted in multiple buildings.
Did you try replacing the PSU?
Do you have a PC case with a huge window? They (used to?) have grounded metal housings with only tiny openings for a reason.
well they used to be cheaper, not sure what is going with prices these day, I remember this being around $100 a decade ago
https://www.officedepot.com/a/products/146730/Tripp-Lite-500...
https://www.officedepot.com/a/products/493958/Tripp-Lite-100...
Shouldn't really happen on USB DAC, it should have enough filtering to get any interference injected by power, and enough shielding (and just being far away enough from machine) for other EMI
Just ordered a hat for my Raspberry Pi with optical out, with a plan to make that my main music streamer. Excited to see if that works out!
The trick is that your 3.5mm connector only needs to connect on the sides, so the end of the jack can be open for light to be transmitted.
This was seen pretty frequently on laptops for a while, but I think two things doomed it. One, most people just don't use optical. Two, there's nothing to advertise its existence. If you do have one of these ports, you probably don't even know you could plug an optical connector in there.
---
All the cheap boards have neither. Most of the high end boards have both
If power lines run anywhere near the sound lines, you are just asking to pick up interference whenever the computer does basically anything. It doesn't take too much of a pulse to be picked up. For a 3.5 jack, the voltage is anywhere from 0.002 to 0.5V. Even a pretty small induced voltage will be audible.
It sounds exactly like the reads on a physical hdd, which is silly because it has an SSD. Haven’t figured out what it is yet.
> SPECS THAT MATTER
> Distortion: inaudible; 100-1000x lower than any transducer (speaker or headphone) you're using > Noise: inaudible; far below a typical headphone or speaker amp
The truth is that DAC is not the problem… everything else in the analogue audio chain is. Amplifiers are messy analogue devices. Speakers and headphones are incredibly messy analogue devices. Power supplies and power conditioners are messy analogue devices. And noise is not down to any one component, but is a whole-system design problem. A particularly cool thing about power supplies is that they often create noise that will be picked up by other devices on the same circuit.
Of course, when people are buying a “DAC” they are really buying a box of some kind that also includes an amplifier, but this naming choice surely contributes to people paying attention to the wrong specs.
I then proceeded to investigate that by using REW to produce measurements of each output and the findings confirmed my hearing, the motherboard audio was outputting 5db less, plus a noise floor 15 db higher than the x-fi, resulting in 20db of extra noise (When you compensate with 5db extra at the amp). The resulting frequency response also revealed a much more aggressive high pass filter, I was actually getting another -3db@30hz compared to the x-fi confirming where the lack of "punchiness" was coming from. And then the cherry on the top was that all the extra channeles (surrdound/sub/center) had an even more aggressive highpass filter, they cheapened out on the cheapening out, probably assuming you are not gonna use more than stereo.
Here is the graph, red is x-fi outs and green is motherboard outs: https://i.imgur.com/dxoLXJO.png
Obviously this is just my anecdote, but I suspect this is very widespread.
EDIT: Oh and I haven't even talked about the line-ins! It's pretty much unusable in ALL motherboard audio I've cared to try. There is some insane noise gating and AGC going on to try and mask the fact that they use some really low bit depth ADCs with terrible dynamic range. Meanwhile I can capture pristine audio from my n64 into the x-fi no problem.
This seems like a lot but it's only 20-30db lower than whatever reference they're using.
This is the spec that really matters: THD+N: 0.0003% which is roughly -110 dB. It's very good and completely inaudible but not exceptional these days.
Good story, though.
I wonder if that will be next fad in PC building, just putting live power line graph on the screen inside
or perhaps live wire into the seat, tied into a transistor on this signal, so if performance drops enough you're sure to be alerted to it
Unrelated, but why? Querying a point in a basic quad tree takes microseconds, is there any benefit to overengineering a solved problem this way? What do you gain from this?
To be totally coherent, you have to draw the entity ID in the same order you would draw the visible color, in cases where entities could "tie" at the same depth.
The reason the game is 3D has to do with partially visible things being way easier than with isometric textures layered in the right order.
Also, now that i just grab a pixel back from the GPU, it's no overhead at all (to construct or get the data for it).
I don't agree with that sentiment. Their designs are subpar and the quality of the soldering is (maybe was) unacceptable:
https://www.audiosciencereview.com/forum/index.php?threads/h...
https://www.audiosciencereview.com/forum/index.php?threads/b...
The above review specifically goes into the problem from OP.
There's also their amplifier with a rather non-standard architecture that tries to solve a non-problem (injecting feedback in a NFB loop - I might remember wrong, if so, forgive me) which leads to it measuring double digit (!) THD if you feed it a sine wave. I'm not an experienced engineer but it is IMO a non-starter to have an amplifier try to decide what is and isn't a musical signal as part of its protection circuitry, short of detecting DC offsets or shorts (pun not intended). I'm not in the market for a 1800$ amplifier that goes bzzzt if you feed it music it disagrees with [1]
https://www.stereophile.com/content/schiit-audio-ragnarok-in...
>Noise from using USB power delivery for audio devices is common.. that’s why you can (and should) use the dedicated power input to you DAC/amp
I don't disagree with your point. However, a company designing products like these should be able to design a filter for this usecase unless you're trying to use your DAC as a a measuring device, or there is something seriously wrong with your motherboard. I honestly haven't heard of any other brand product with this problem unless it's ~20 years old and in need of repair. It doesn't cost much in the BOM, however it does cost in engineering hours/competence and QA and this is something that should have been caught by the latter.
[1] https://www.youtube.com/watch?v=NzMbY4sZvIw
Edit: I just want to add that I don't want to hate on Schiit. Honestly I'd like new audio companies to succeed and I applaud them for rejecting MQA back in the day and for not going all-in on the audiofool bullshit one sees too much of. But seeing such poor engineering and QA leaves a sour taste in my (electronics engineering) mouth. Maybe they have improved lately, I wouldn't know. I'm not really in their market anyways.
E.g. https://www.audiosciencereview.com/forum/index.php?threads/s...
> No doubt you have noticed my frequent use of terms "nice" and "excellent" and that sums up the performance of Modi+. At this price point, we don't expect objective perfection but competent engineering and that is what we have. Physically, the unit is solidly built and of course supported by an English speaking US company. For people with such preference, the Modi+ provides an excellent option. That they can stay competitive with far east audio companies is definitely a feather in their cap.
> I am going to recommend the Schiit Modi+ DAC. Great to see Schiit continue the (new) tradition of optimizing objective performance as they cater to their traditional audience.
I just know that if I handed over something with such shoddy soldering to a customer, I'd lose my job or at the very least lose soldering privileges. But I am working with things that cost slightly more than 99$ that you can't find on store shelves :)
That sounds interesting. Could you elaborate further?
Thankfully doesn't happen with an external DAC.
Sounds like the game is doing more when the cursor moves around, they're probably checking for where the cursor is, and something is making the CPU/GPU do a bunch of extra work, which finally triggers the coil whine when the PSU is more heavily used.
I've basically had the same issue with Nvidia cards since the 2080ti started doing coil whine as soon as I opened Unreal Engine. Some programs trigger different sounds, depending on how much/well they use the GPU, and I've had the literal same experience with "hovering with my mouse over element X triggers coil whine" multiple times before.
