Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I see a business opportunity here in providing very expensive repackagings of compilers and libraries of "audiophile" quality. It just needs to be really, really expensive.


Make it really really expensive seems to be a really critical part of the business model for audiophile product.

I opened up Stereophile magazine on apple news recently and their was a breathless review of a $10,000 external DAC.

They had a gut shot, so I started looking at it. The whole thing was just a fancy box & power supply driving a $20 off the shelf DAC chip.

I'd go out on a limb and guess you can probably find a $100-200 DAC with the same chip in it that no one is ever going to be able to tell apart from the $10,000 one.

The $10,000 one didn't even have a particularly flashy looking enclosure!


> Make it really really expensive seems to be a really critical part of the business model for audiophile product

I took a marketing class once that talked about this effect. If you have a crappy product, there are two pricing models you can go for: sell it cheap, or make it pretty and sell it for a lot of money.

Often the latter, which is a scam, is the better option from a pure business point of view. The reason is because people perceive expensive things as being higher quality regardless of whether they really are or not, and also people are very resistant to admitting to their friends that they got ripped off -- so they'll frequently keep singing the virtues of an expensive piece of junk even when they know it has no virtues. So products can still get good "word of mouth" advertising regardless, as long as the price point is sufficiently high.


Hästens beds anyone?


you can have the same dac on different boards and on one it can work perfectly and in the other it could hizz like crazy.

there's a lot of engineering going on to reduce noise and the analog part in the DAC is greatly affected by the rest of the electronics, it's not just ones and zeros.

most people won't notice the difference of course but there's a reason professional audio equipment costs a fortune.


I think their point is that "there's a reason audio equipment costs a fortune", doesn't really explain what that is.

Therefore, it's potentially easy to tell low grade hardware and simply market it as being legendary


    I'd go out on a limb and guess you can probably find a 
    $100-200 DAC with the same chip in it that no one is 
    ever going to be able to tell apart from the $10,000 one.
Good news!

You're right, and there are a lot of inexpensive, engineering-driven "audiophile" products that do exactly what you say.

Topping, Schitt, JDS Labs, and SMSL are a few who offer outstanding products in that $100-$200 price range.

For example, here's a unit from Topping that combines a top notch DAC and headphone amplifier for $200. Objective measurements taken with a very expensive AudioPrecision analyzer: https://www.audiosciencereview.com/forum/index.php?threads/t...

$129 DAC, made in America, also very close to state of the art objective performance: https://www.audiosciencereview.com/forum/index.php?threads/s...

It's fun (and actually, important) to laugh at the audiophile lunatic fringe, but most of us don't play that way.


I have several pieces of Schitt gear. I don't have one of their external DACs though, the digital pieces I have in my setup have very nice DACs that I like. Those internal DACs have similar chips to what is in the Schitt DACs.


That's true. Also, what we actually hear is the analog output stage surrounding the DAC far more than the DAC itself. That's the part that OEMs either get right or fudge up.

A top of the line DAC is just a few bucks in quantity: https://octopart.com/ak4493eq-akm+semiconductor-88954630

But, a naive implementation (say, throwing it on a computer motherboard where the analog output stage is subject to interference hell) is maybe not going to sound great, compared to a better implementation of a lesser DAC.


If you want a really clean output on a DAC, I recall you needing a very clean power supply, with $10-100 resistors and capacitors. If you were looking for a cost driver, you may have missed it.


Really clean meaning what exactly? at what price point would a DAC be indistinguishable in a blind A/B test to a $10,000 one? I'm not trying to be pedantic I just think it's important to ask these questions.


Many $10,000 DACs are measurably inferior to a $9 Apple USB DAC. Start be crossing all those off the list.

The Apple DAC is actually insanely great, the best price/performance ratio of any bit of new audio equipment ever. They are not particularly mechanically well-made, though, and can easily be broken from repeated bending.

In a real world level-matched double-blind ABX test, $9 wins.


This is a great question and I'm glad you're asking it. Once you get up to 90-100dB SNR I think most people would say you're well into transparent territory with decent headphones. 120dB is absolute state of the art (and the limit of must measurement gear!) but the difference is academic at that point.

Here's how cheap USB-C dongles from Apple and Google stack up with objective measurements: https://www.audiosciencereview.com/forum/index.php?threads/r...

Those mini dongles are pretty good, although they're not necessarily exactly world beaters. They're DACs + amplifiers and their amplifier sections don't exactly deliver a lot of current. But for e.g. desktop use you can get a really good DAC + headphone amp combo from Schiit, JDS, Topping, or others for $200ish.


Minimal distortions, but especially one thing: less reaction to external RF noise. Remember living in the early '00s, back at a time when cellphones were barely regulated? Every time someone got a call, you could tell two seconds earlier because of the characteristic buzz noise in the radio.

Another part is less reaction to unclean power. "Thanks" to everyone and their dog using Powerline ethernet, having a clean power supply is ... difficult. You won't notice that in your audio setup unless your neighbors are all blasting the power lines, but if you're working with high-frequency stuff it's definitely noticeable.


Is the decrease in RF noise from phones due to regulation or battery life concerns?


Did you spend the most you could on every component? No? Then I could be cleaner. With a massive price increase too


Here's one that costs $129, measures great, and is made in America to boot: https://www.audiosciencereview.com/forum/index.php?threads/s...


Main use case of those $10 resistors is metrology where you measure nanoamps and microvolts.


All of our code is compiled three times by different Mac Pros in a climate-controlled environment, in a special room, 30 meters below ground, enclosed in a faraday cage, and encapsulated in an anti-radiation shield of three feet of concrete and 2 inches of lead.

The result of the compilation is recorded on custom 2-inch wide magnetic FeCr tapes, aged for two weeks, and then compared three-way in a vintage Ahmdal Corporation 470V/8 that is carefully maintained by a squad of vegan hermit celibate engineers with the highest DoD security clearances, and kept permanently in an inert atmosphere of CO2 and Argon.

If the comparison shows no errors, only then can the binaries be deployed to customers.


The digital to analog conversion in the tape recording process is going to introduce noticeable smoothing and artifacting.

You really need an all-analog process to bring through more of the original punch of the compilers and avoid making the resulting program woolly and harsh.


I worry for the engineers kept in an inert atmosphere


What? This is AUDIOPHILE code!!! are you suggesting that the developers should be allowed to *gasp* breath on the computer which is compiling the code?!? Of course we can't allow that, it would introduce impurity and artifacts in the sound. They should use BSL4 vests with a closed circuit respirator!


https://www.myhifishop.de/Devices/AQ-SWITCH-SE-Audiophile-Hi...

Apparently, this is literally a vanilla D-Link ethernet switch (which costs like €70) with various bits (literally) stuck onto it internally to make it sound better.

A bargain at €798, I'm sure.


https://www.myhifishop.de/Voodoo-Corner/AQVOX-LAN-Detoxer-RJ...

LAN Caps, that is something which you can cover your unused ethernet ports... I wonder if they lower my ping in gaming too...

I was thinking of gold plated optical cables, but this is pure snake oil...

EDIT: I think it might be joke as it is in "Voodoo Corner", but the rest of the stuff isn't so...


> What is the AQVOX formatting: Using a special process, which has incorporated our 30 years of experience in the optimization of audiophile audio components, the material receives a special formatting information. The formatted material brings about that electromagnetic frequencies are harmonized and electromagnetic pollution (electro smog) is minimized

This shouts to me to be a direct con, someone deliberately taking advantage of the brainwashed. Like psychics and cancer-curing-crystal sellers¹. They are basically one step up from the fraudsters calling to try get access to our elderly (grand)parent's savings².

----

[1] though in defense of those ripping off audiophiles, at least they are not taking advantage of the desperate.

[2] the psychics & such are not one step up, at least not a full one.


This [1] used to be known as "the most expensive hdmi cable in the world" at over $10k. The vendor has since stopped selling it and replaced the picture by one of a usb power adapter, but the reviews are still there for a good laugh. I also recommend the user submitted questions and answers about the product.

And here is the real sad part: the seller only need to sell ONE, even as a joke or for the lulz, for this scam to be worth it.

[1] https://www.amazon.com/AudioQuest-Diamond-2m-Braided-Cable/d...


Every single morning I wake up, clear the haze from my eyes, shuffle to the bathroom, and stare at my aging face in the mirror.

I run my hands over my stubbly chin and gaze deeply into my own eyes, wondering.

"Why can't I figure out some way of ripping off audiophiles?"

"Some dude sold a marker used to color in the edge of CDs to audiophiles for like $20. A $0.50 marker for $20. C'mon man think of something."

"Am I too moral, or just an idiot?"

"They're so god damned gullible. Why can't I figure something out? They're all old and rich, it's not like I would be hurting anyone..."

I then sigh and continue on with my day.


Here’s the pitch:

Sure, mainstream compilers give you the best performance. But doing all the computations at once will result in spiky performance and power draw, which of course will propagate through to your ears as noise. Our compiler rearranges the instructions to present constant load level, resulting in more constant sound flows.

In fact modern computers are too fast to play music properly, back when you were young and the computers were slower, programmers tuned their programs to maximize throughput which naturally resulted in this smooth flow of instructions. It is only recently, conveniently right outside your nostalgia window, that computers became fast enough that they could fly through the code and then stop, resulting in this digital noise.

Plus, modern CPUs change their frequencies all the time, to save power. By keeping a steady flow of instructions, we keep the CPU from going to low frequency mode. If it has ever felt like you are having trouble hearing high frequencies, it is probably because they are missing on a computational level.

As a test, we have set up this laptop. Put on these headphones—we assure you, they are high quality open back headphones. When we unplug the laptop from the wall, it will naturally go into low power mode and naturally smooth out the instruction flow. Listen, no hum. But when we plug it in, you can hear some high pitch whine in the background. That’s the peaky instruction flow.

This is actually kinda fun.


Better add a consultancy arm to "tune" the computers to remove the CPU scheduling issues, sort out the P states and C states to remove those pesky optimizations.

Maybe you can create an audiophile BIOS, as well that automatically tweaks the CPU to optimize enabled core count, turbo and voltages, etc. Of course, it's a chargeable add-on for the custom PC service, and only available on supplied hardware.

Oh, do you think you could get the audiophiles into RISC CPUs too? Find a reference board, custom brand it, and charge 10x for a purer CPU. It's reduced instructions, so there's fewer of them to disrupt the pure audio quality. You've hand tweaked the RISC instruction set to remove any that might impede the logic flow of audio processing for a purer, clearer sound.


The thing about these superscalar out-of-order processors is that they reorder the instructions behind the scenes. You can run out software on something like an Intel i7 and it will sound better than anything you’ve heard in the last 10 years, but our special in-order, RISC chip with a 4 stage pipeline, while slightly more expensive than the i7, will give you the wholistic listening experience.


Not a few of the companies operating in the esoteric high end accessory market appear to be cynically fraudulent. Apparently supported by some of the hifi press and dealers.

Considering that special oils, special shiny tapes, special rocks, special boxes of rocks, special cable lifters, and very special cables have all magically delivered an incredible improvement in imaging and the soundstage etc etc etc, it's very much a sellers' market.


Monster memcpy, distributed on gold plated USB sticks.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: