Surround Sound via Headsets - Why not implement in software?

I've been looking into surround sound headphones/headsets over the last few days and noticed that most of them are based on Dolby Headphone technology. There a small handful of arguably superior ones that actually have more than one speaker per ear (versus virtualizing them as Dolby does) but to me this seems like overkill; you only have 2 ears after all right? Anyway that's not really the main question I wanted to bring up...

Dolby goes out of its way to say that their technology "works with any set of headphones", but this seems to be a statement more aimed at potential licensees then the consumer. The fact is that I need to shell out for specially made headsets (or one of like 4 random netbooks) if I actually want to experience DH (Dolby Headphone).

What I wonder is why DH has to be implemented in some ridiculous brick on my headphone cord instead of a piece of software running on my computer. I have done some googling and it appears there is actually software that does essentially the same thing as DH; and of course that means you can supply your own ordinary headphones. It may or may not be the best option but you can check out the one that came up first in the rankings here.

So please tell me folks, what am I missing out on, if anything, by grabbing a regular headset and using software instead of DH?

BTW software will always get bonus points from me because I can easily upgrade it; versus the hardware alternative of buying a whole new headset every time there is a new feature or a bug.

 

There are a lot of headphones that promise virtual surround but I don't even think half of them use dolby headphone. Make sure you read the full description because headsets like the Razer Megalodon and the steel series 5hv2 claim to offer virtual surround sound but mention nothing about dolby headphone. That's because they use their own algorithm which may be better or worse than Dolby Headphone. If you really want to use Dolby Headphone with any set of headphones, you can purchase a Astro mixamp or a sound card that outputs dolby headphone.

My theory as to why they don't offer dolby headphone in software is because many motherboards with crappy audio sometimes have hiss and static when listening to audio. If you were able to purchase the dolby headphone software, you would be experiencing the effect but the static and hiss will still be there. This is part of the reason why most people get external sound cards anyway. Also, with the exception of a few headsets, dolby headphone is implemented in hardware so your CPU does not have to work at all to process audio. You probably won't notice a difference if you have a good CPU anyway but that is something to think about.

 

Yeah DH is more specific than I really intended. Just substitute it for any product that uses virtual surround implemented in hardware.

I have an HP ENVY 17; my CPU should be more than capable of the added workload. I'm not as sure about the onboard audio quality but lets just say I don't experience the static hiss. Is there any reason not to go the cheaper route of software + (nice) stereo headphones?

 

Does anyone know of something similar to the Astro MixAmp that virtualizes 7.1 instead of 5.1 surround? I have a laptop so unfortunately I can't just buy a soundcard.

What I'd like to do is purchase and compare it to a software solution, since there doesn't seem to be much information on the subject.

 

According to the Astro Mixamp manual found at http://www.astrogaming.com/support/download/5/ , The astro mixamp will actually function as a 7.1 dolby device. I guess they normally don't mention this because most people buy it for their 360 and the only 360 game that I know of that outputs 7.1 is prototype. IMO, you probably won't notice a difference between virtual 5.1 and virtual 7.1

 

Well, I'd imagine that a certain degree of tuning is required for a particular phone to get the optimal median HRTF (since we all perceive differently, HRTF implementations are based on a 'median pick' of potential human perception) on it.

I'm not absolutely sure, but aren't all Dolby implementations in software? In that it's built into the drivers from each virtual-surround vendor?

 

TL;DR version of post: All surround sound through headphones is software-based. Dolby headphone mode truly works with any set of headphones for that reason. "Gaming 5.1 surround" headphones are a gimmick. Surround sound through headphones depends on driver or game support.



Long version:

To understand how surround sound works in headphones, you need to first
understand how directional sound works in general.

The way that we, as humans, perceive directional sound is that sound wave will hit one ear before it hits the other. The sound may also echo off of other surfaces in the room (e.g. walls, tables, ceilings, floor, ear lobe, etc) before it hits our ears. Our brains processes that difference in timing + echo as directional sound. Surround speakers in a home theatre environment work well, because they have the advantage of using timing + echo to present positional sound to the listener.

When you get into headphones, it gets trickier. With headphones, the sound is getting piped directly into your ears. You do not get differences in timing or echo. The only way to get surround sound with headphones is to use some kind of software algorithm. That algorithm basically "calculates" the timing + echo, and estimates what your left and right ears would typically hear if there were true positional sound sources in a physical room. That software algorithm may be a piece of software that you install, or it may be built in to some device. But no matter what, there is some kind of software that is being used to process.

There are many names for this type of software algorithm - Dolby Headphone Mode, surround downmixing, virtual surround, etc. But they all do the same thing - they use software to calculate what your left and right ears SHOULD hear in a true 5.1 surround speaker setup, and pipes it through to your headphones. When Dolby says that their Dolby Headphones mixer "works with any set of headphones", they are not exaggerating the truth. It really does, because Dolby Headphones mode is really just a piece of software that happens to be baked into an external hardware device.

Now, when you see things like "surround gaming headphones" or "5.1 gaming headphones", they are all a gimmick. They all use some kind of software-based surround virtualization. Some "gaming 5.1 surround" headphones them even actually have multiple sound drivers in the headphone cups. But once you understand how the human brain perceives surround sound, you can quickly see that there is no way that having two sound drivers in a single headphone cup that are less than an inch apart can somehow emulate the true timing + echo of a true 5.1 surround sound speaker setup (with one exception, which I'll discuss later). They all use some kind of software.

So, once you know all of that, you quickly see that the best way to get surround sound on headphones is to get yourself a good software algorithm, and a nice set of high-quality headphones that emphasize clarity in the mid and high frequencies more than anything else (like Etymotic canalphones).

The best way to get virtual surround through headphones is at the driver level. Windows 7 does a decent job of this, with its Virtual Surround setting. It doesn't quite have all of the options as a Dolby Headphone algorithm, which lets you do things like choose room size, wall materials of your "virtual room", etc (and thereby alter the way it performs the sound calculations), but it works. And most importantly, it works universally across all Windows apps. You can set your Windows Sound device to use Virtual Surround, and then set your games to use 5.1 sound output. OpenAL in Windows will intercept the 5.1 signal, and downmix it into 2-channel headphones for you.

The other way to get directional sound is if a game supports it directly. Valve Source engine games do a great job at this. I don't recall seeing any other games that do a spectacular job at this, though. You're probably better off sticking with the Windows Virtual Surround setting.



** Exception - Psyko Audio Labs 5.1 Headphones - they are the one exception that I mentioned earlier to "surround gaming headphones" that actually work. They mount 5 speakers on the headband of the headphones themselves, and simply have audio channels that lead into the ear cups. When a directional sound is played, it actually plays on one of the 5 speakers, which pipes through the sound channels into the ear cups. That design allows them to achieve some sort of timing + echo in hardware, to go along with their software. From what I have read, the surround sound works pretty well. The major drawbacks of those headphones is the high cost ($300), and the fact that they get uncomfortable rather quickly.

 

Thanks for the very informative post. This seems to back up the feeling that I was getting that implementing the algorithm in firmware (headphone brick) vs software (PC) really doesn't add to the experience, with the possible exception of having the benefit of an external sound card.

It's puzzling to me though that Dolby doesn't realize that. Certainly they have more to gain from releasing Dolby Headphone drivers for all the major operating systems then implementing their product in only certain headphones... Unless CPU power and crappy onboard audio is really that much of a hindrance, which I'm not sure I believe.

On a semi-unrelated note my top pick for an upcoming headphone purchase right now is a pair of Sennheiser PX 200-IIs. The 100-II seems to have gotten the best reviews for the form and size I'm looking for, but I'm not usually listening in optimal environments for open headphones. As far as I can tell the only difference from 100-200 is the open vs closed design so they seem like the right choice. And at only $60 I almost can't go wrong. =) Any words of wisdom?

 

Correct. There is very little difference with a hardware-based implementation, with the exception that you essentially get an external sound card. Doing that eliminates a lot of the static and noise you get from electromagnetic interference (EMI), which is a problem with any onboard sound solution.

I disagree. Dolby certification is considered to be a top-end certification on sound quality. By limiting who can use the Dolby name, they get to preserve the perceived high-end value of their sound technology. They limit the use of the Dolby name on the PC side by allowing only hardware partners to use the technology. Their business model is a high profit margin, low volume business.

If they released drivers for all major operating systems, then they essentially become a commodity. They dilute the value of the Dolby brand name. And they lose control of their product, once it becomes software. It becomes copyable, and it becomes wildly distributable without any money going back to Dolby. Typically, mass distribution is when you want high volume, low profit margin to capture market share. This is not what Dolby does.

It is the same business model as Apple - Apple has absolutely no interest in gaining large market share. They have a tightly controlled environment on both their Mac computer side with OS X, and their iPod/iPhone side with iOS and iTunes. Because of that, they get to sell expensive high-end products that makes them a large profit on each unit sold.

I'm not a huge fan of small over-the-ear headphones. For portability, I stick with canalphones (Etymotic ER-4P at the high-end, or Sennheiser CX-300B at the low-end). I do own a pair of full-sized headphones (Audio Technica ATH-A900) that I use for my desktop gaming rig, although those might be out of your price range. You can always look into their smaller brother, the Audio Technica ATH-A500's... great sounding closed-design full sized headphones that sound great without a headphone amp.

If you specifically like the Sennheiser sound (as do I), then you wouldn't go wrong with the Sennheiser HD555's. Fantastic headphones for the price. And the models being sold in 2010 are all 50 ohm models, so they can be easily driven without a headphone amp.

 

Wouldn't the actually audible distortion of the Etys in the mid-high upwards throw off such efforts? Just curious. Also, DH is about presence as well as surround. Etys have zero presence, due to a total lack of bass (especially from a 'theater' perspective)...?

 

What do you mean by audible distortion on etys?

 

You're not aware of the distortion on the Etys especially noticeable from the mid-high upwards? Like a very fast version of someone treading on chip packets mixed in with the sound? I couldn't actually believe people rated them so highly given the issues in 'overdriving' the single armature driver over the frequency range / frequency bias that the ER-4P/S addresses.

 

Advice is appreciated, but I'm not too willing to budge on the form factor. Huge over-ear headphones are great but I wouldn't want to lug them around or, frankly, look like a tool wearing them in public. On the other end of the spectrum in-ear stuff tends to bug me during extended use. "Regular" size headphones are a good tradeoff.

I should throw in that this will be my first pair of headphones that weren't free (and most likely crap quality) with something or other. Never too late to appreciate better sound though right?

 

Between my wife, my mother-in-law, and I, we have owned 6 pairs of Etymotic headphones over the past 6 years (2x Etymotic ER-4P and 4x Etymotic ER-6i). I have heard plenty of distortion that I hadn't heard before - but that was because the Etymotic's had such high clarity, that they revealed sound artifacts in 128kbps MP3's that I have never heard before with cheaper sound hardware. I have never heard any audible distortion before that was because of the headphones.

I did some research online about this, and the most common reasons for distortion were:
- Distortion in the source recording material.
- Distortion from poor quality / low powered amplifiers, like on portable music players.
- Distortion from dirty filters on the Etys. Time to change them.

Most people narrow down the cause of distortion by changing source material and source audio gear. Those that still have problems change the filters on the Etys. Of course, it is always possible that you somehow have a defective pair of headphones that needs to be RMA'ed.

And I absolutely don't think that Etymotic's have a problem with presence. The bass is there - it *sounds* like bass, it doesn't *feel* like thumpy bass. Not for everyone, but definitely ok in my book considering the exeptional audio clarity, soundstage, and instrument separation that you get. I have been using my pair of Etymotic ER-4P's for 6 years for music on various music players, movies on my laptops that I've owned (both disc-based DVD's and file-based movies), and gaming on desktop and laptop computers. Out of all of the portable headphones that I have owned (Shure E3C, Shure E5C, Audio Technica ATH-CKM55, Bang & Olufsen A8, Klipsch Image S2, various $30-$50 earphones from Altec Lansing & Sennheiser), and all of the portable headphone amplifiers I have owned, the unamped Etymotic ER-4P's remain my absolute favorite headphones for an immersive experience.

 

I knew this wasn't the case (i.e. distortion in the material) prior to suspecting the Etys - and this is like from 7 years ago. And I can hear artifacts in 128K MP3's with $5 earphones, let alone Etys - it's not that level of audible issue. What I do find though is a lot of people believe they can hear artifacts more clearly with Etys, when it's actually the case that they aren't able to make out the artifacts in full-response audio even with much cheaper gear - The Ety's lack of lower frequencies concentrates perception on the mid/treble and makes it easier to pick up obvious artifacts in the treble and upper midrange - some of the most obvious points for aliasing of lower-bitrate material.

Issues with the source material is not the problem as far as what I said is concerned - it's the distortion induced into the sound by the armature driver. Which is why I've never rated Ety's as particularly 'high fidelity' despite the patina of it. Dual or multi-driver Shures on the other hand sound duller - because they're using the drivers more within their capabilities and haven't put in high-treble drivers for crossover and/or cost reasons, but have higher source fidelity due to severely curtailed distortion. Given the choice, I prefer using them as they're truer to the original material, and also because the distortion actually gets on my nerves after a while.

You've got me curious though as to whether Etys (or other in-ears) are an effective virtual surround platform. There's more HRTF work to be done if I'm not mistaken with a pair of in-ears...?

 

I strongly disagree with the idea that the "lack of lower frequencies concentrates perception on the mid/treble" as the reason that the Ety's have more detail. If your theory of "lack of bass concetrates perception in mid/high" were true, then you would achieve the same effect if you took a regular pair of headphones and EQ'ed out the lower frequencies. Or, if you took a pair of Ety's and EQ'ed the bass frequencies up higher. The fact that Ety's reveal a huge amount of detail from music is not a simple mind trick.

If you ever listen to a pair of Etymotics, you can clearly tell that they reveal more detail in music, especially if you have songs that have any black space that should be silent. With Etymotic, you can actually hear the artists counting under their breath to themselves, or licking their lips in preparation to sing. Tracks like "Fleetwood Mac - The Chain" or "The Beatles - Strawberry Fields Forever" are where I first noticed this effect. Some people hate this, some people love it.

But I think the discussion is getting academic and moot at this point. It is clear that you don't like the Etymotic sound, and prefer the Shure sound. Personally, I don't like the Shure sound because it gives me a headache after 30 minutes, and sounds too flat to me (a characteristic I know many people prize about Shures, for their neutral sound). But having different opinions on preferences is the very reason why there are so many options in headphone manufacturers.

As for IEM's being a better virtual surround platform - it is simply because IEM's block external sound, giving you a much quieter listening environment to hear the nuances of surround sound.

There is not more HRTF work to be done with IEM's. From the perspective of timing + echo for surround effects, the only difference between full-sized headphones and IEM's is that full-size headphones get their sound distorted by the natural curves of the outer ear. However, every IEM designer will alter the frequency response of the IEM's to match this. From a headphone hardware perspective, there is no difference in surround effects... just that IEM's are more likely to give you in a quiet listening environment.

 

I encourage you to try it - Use a 'known' headphone with a documented frequency response curve. I'd suggest something like the DT 880 or HD650 as they can go either way, or a DT 770 for a slightly more uncouth but nevertheless effective demo as it has a bowl-curve-like response.

In fact, it's worth noting just how many factors which can cause us to write about what we hear in a different way *are* actually mind tricks. And why, to have a usable baseline for discussion ad not just an e-peen waving contest, you should develop methodologies to either eliminate or mitigate as much of those factors as possible.

It's just that in an IEM, we're (and have to be, really) prepared to make more excuses for the product - some of us more than others it has to be said - in terms of the mobile compromise.

The drivers are the limiting issue, so the Ety's don't respond very well to EQing up the bass - nevertheless, it's possible to a degree, but with corresponding impact on the rest of the sound. Once again, try it.

As for preference of one thing or another, it's merely a matter of compromise for me as neither quite delivers what I want but in different ways. Sound-wise, I can hear the distortion spikes the Ety's impart which is curtailed in Shures (although it's still there, but much less audible), and I also thought that the tip selection at the time was more comfortable with the Shures, despite the slightly lower isolation of the shallower foam tips. This was before I identified increasing ear irritation with both however, and moved to customs for multi-hour travelling use. These days the - yes, Shures - are used purely for commutes, since I feel that the Comply-style tips are a better comfort/isolation balance and for 'quick-extract' wearing, the Ety olives are no advantage over the Shures.

I'm not absolutely positive that ultimate isolation contributes to HRTF viability, although common sense would seem to suggest it. There are for example some very viable HRTF systems based on open or semi-open headphones.

 

What about the Turtle Beach surround sound headphones, more specifically something like the HPA2 or the discontinued AK-R8?

 

I haven't used them myself. The general consensus from the reviews seem to have positive reviews about them and their surround capabilities.

But my personal preference is to own a high-quality set of 2-channel stereo headphones. Surround headphones typically use cheaper/lower-quality individual sound drivers in the ear cups, are most effective in situations that have surround support (only games, movies), are less portable, and are harder to resell due to their single-purpose product design. Regular headphones are more portable, can be used in multiple situations, can be more portable depending on design, and are much more easily resold. YMMV.

 

I was actually just pointing out that the Psyko Audio Labs weren't the sole exception. I enjoy mine, but I'm both probably not a "real" audiophile, and I don't think I've ever had any "serious" headphones to compare them to.

 

The only reason that I mentioned the Psyko Audio Labs headphones was because they were actually using physics to achieve the timing + echo effect. They use a design where a single speaker emanates a single sound that is heard by your two ears at two different times. There is surround processing algorithm in place, but it uses a combination of software + physics to achieve the "timing" aspect of surround effects.

The design of the Turtle Beach headphones, and many other "surround" headphones, don't really use physics. Each ear cup has its own independent set of speakers. The "timing" aspect of surround sound is entirely done through software. And I say that if you're going to do everything through software anyway, why bother hooking up that software to a proprietary set of headphones? Why not bring your own?