What's a "reasonable" PC audio hardware? Options

[Gelatinous Blob]

I'm sure this question has been asked many times before, but I can't find any answer.

I'm not really an audiophile, but I do care about sound quality to some extent. My problem is that I don't want (and probably don't need) high-end equipment that costs thousands of dollars, but I'm not satisfied with the most basic stuff either - and I don't really know where the "middle ground" is for me. I'm listening to music only on headphones connected to my PC (the files are either FLACs or 320 kpbs MP3s) and I don't want to change that. Right now I'm using Philips SBC HP890 headphones connected directly to a Creative X-Fi Xtreme Audio sound card (actually, it's a rebranded Audigy SE). I don't listen to jazz or classical music, only "popular" stuff (mostly indie rock). I feel satisfied with my current setup, but I've never heard anything better, so I can't compare. My question is - is that setup sufficient for someone like me? Will buying slightly more expensive stuff make a difference that even an untrained ear can hear? For example, I think that audiophiles are using a headphone amp between the PC and the headphones - will that make a big difference?

Thanks in advance for all your suggestions

[chelgrian]

Right now I'm using Philips SBC HP890 headphones connected directly to a Creative X-Fi Xtreme Audio sound card (actually, it's a rebranded Audigy SE).

If you switch off all the Crystalizer processing and any other processing on the creative card or being done by its drivers then you probably aren't going to notice any difference with a better card.

As for headphones I suspect you would be able to hear a difference between those Philips ones and the "HiFi' end of Sennheiser's range. I find that at the same price open backed designs tend to sound better than closed back designs. Although of course, like any sound reproduction device, it's down to personal preference and saying which is "better" is utterly subjective.

Unfortunately it's very difficult to do blind listening tests on headphones as you tend to be able to feel which ones you've got on

Most headphone outputs on most cards will competently drive most headphones, in my limited experience headphone amps only make a difference on a small number of difficult to drive specialist designs.

[Fandango]

HD Audio on-board has improved a lot when it comes to sound quality. When your motherboard only has AC97 sound, I would rather buy an extra sound card, but if it's a fairly new motherboard with HD Audio, I would stick with the onboard sound. HD audio has got digital out, too.

[Kitsuned]

Just keep all audio effects off on Creative cards. The 0's and 1's are otherwise all produced the same, so its the speakers that make more a difference in the end. Can't go wrong with FLAC files or high bitrate mp3s.

If you're happy, why change? I think you've done all you need to be happy with the music you have.

[Gelatinous Blob]

If you switch off all the Crystalizer processing and any other processing on the creative card or being done by its drivers then you probably aren't going to notice any difference with a better card.

I'm using this card under Linux, and since all the DSP stuff on the X-Fi Xtreme Audio is done by the drivers, I'm pretty sure that it's all switched off. I'm not really able to do blind listening tests here, but I think that the card sounds as good under Linux (and ALSA drivers) as it does under Windows using Creative's software.

As for headphones I suspect you would be able to hear a difference between those Philips ones and the "HiFi' end of Sennheiser's range. I find that at the same price open backed designs tend to sound better than closed back designs. Although of course, like any sound reproduction device, it's down to personal preference and saying which is "better" is utterly subjective.

My old headphones are dying anyway, so this is a perfect opportunity to check out something better I don't think I can afford Sennheiser's HD 650 or 600, but 595 is doable and 555 is ok, so I'll check out some reviews

Thank you for all your answers - that's really helpful

@Fandango: the X-Fi is not a built-in card, but it's rather on the "budget" end of the spectrum, so I was wondering if the sound quality is not affected by that

@Kitsuned: I'm happy, but I don't know how much more happy I can be I know that sound cards and headphones can be much much more expensive than the stuff I'm using and I was wondering if the difference in sound is as noticeable as the difference in price Plus, I need new headphones anyway

[Eldo]

You might also consider something like the Alessandro MS1s, at $100, a lot of people like them. They are slightly tuned Grados. One problem is that they aren't the most comfortable headphones, they're ok, but comfort is not a key part of their design. They would probably work pretty well for rock if comfort isn't a key part of your decision.

If you happen to know someone with decent headphones, trying them out would be your best bet, the Grado VS Sennheiser sound is definitely different, and its hard to decide what your preferences are based on internet reviews

And yes, diminishing returns goes into effect very quickly for audio equipment. You're probably best off spending your money on the speakers/headphones end of things rather than soundcards/DACs/amps.

This post has been edited by Eldo: Sep 7 2008, 18:04

[buktore]

My recommend is to stay away from Grado/Alessandro.

They sound awesome... but apart from the sound (and the look, I actually like it), everything about it are horrible.

This post has been edited by buktore: Sep 7 2008, 23:35

[WmAx]

By far, the most accurate phone for $100(or even $200) range is the Sony MDR-7506. It also has best build quality - even compared to phones costing 5x the amount. The only part subject to wear, in a practical sense, is the ear pad set. These pads last on average, about 2-3 years of moderate to heavy use. They cost about 20 including shipping to replace. I have heard of rare occasions where the pads failed within a month. But this would only be due to the phone being OLD stock. I have found that the pads do not keep - the chemical make up in them breaks down after a few years, regardless if they have been used. If you did happen to get old stock(say, 4 or 5 year old phone), the rest of the phone would be perfect, but the ear pads would likely deteriorate/flake apart very rapidly.

Now, will you prefer this accuracy? I don't know. It's a tool, really, and was never really intended for music listening in an enjoyment capacity. Some people hate it, even. But for the price range, it is a superb phone for listening IME. For pro use - there is really no better phone at any price that I know of, overall, for live tracking/monitoring on live recording location.

-Chris

This post has been edited by WmAx: Sep 8 2008, 00:03

[Gelatinous Blob]

Thanks for all your answers and suggestions I'll try to test some headphones from the range you have suggested and see what I like

[honestguv]

> I feel satisfied with my current setup, [...]
> is that setup sufficient for someone like me?

Obviously yes if you are satisfied with it.

> Will buying slightly more expensive stuff make a difference that even an
> untrained ear can hear?

Yes if wisely purchased. You can buy expensive stuff that will sound worse.

> For example, I think that audiophiles are using a headphone amp between the PC
> and the headphones - will that make a big difference?

Probably.

The analogue headphone amplifier circuitry in audio cards is often very cheap and suffers from a high output impedance among other things. This causes problems for headphones with an impedance that varies strongly with frequency which is typical of high quality headphones. For example, the Sennheiser HD 580/600/650 will have an obviously deficient woofly bass response when plugged into the built-in headphone port of a typical portable PC compared to the headphone port of a typical hi-fi amplifier.

Cheaper headphones tend to have a fairly flat impedance and do not suffer as badly. You can see impedance curves for common headphones here:

http://www.headphone.com/

I do not know how poor/good the headphone amplifier circuitry is in your sound card (anyone?) but I would guess the chances are that it will not drive a Sennheiser 595 without obvious deficiencies should you compare it against a headphone circuit with a low output impedance typical of those found with most hi-fi amplifiers, some powered speakers, some external sound cards and most dedicated headphone amplifiers. Dedicated headphone amplifiers are usually poor value for money though.

> I don't think I can afford Sennheiser's HD 650 or 600, but 595 is doable and
> 555 is ok, so I'll check out some reviews

I would suggest ignoring the reviews. Headphones vary greatly in sound from the laid back sound typical of Sennheisers to the forward sound typical of Grados. This can be easily heard by a casual listen at the headphones rack in a department or electrical store. Comfort is also important and headphones that put pressure on the ears are rarely comfortable for any length of time.

[Gelatinous Blob]

@honestguv: Thanks a lot for that info It really changes things a bit I did some research on the whole headphone amp issue and I just realised that my sound card has no headphone jack. I'm simply connecting my headphones in the place of the speakers. Is that a problem? Do I have to buy a headphone amp if I don't have a dedicated headphone jack?

The guys at HeadRoom claim that even Grado SR-80 will benefit from an amp, but they actually manufacture and sell these things, so I imagine they're a litte biased. On the other hand - I hate the idea of buying good headphones and not realising their full potential, so now I really don't know what to do. I'll try to test as many models as I can, but I won't buy anything until I solve the amp issue

[Slipstreem]

If the output is designed to drive loudspeakers then it has an inherently very low output impedance. This is exactly what you need in order to reduce the effect of the varying load impedance with frequency of the headphones as the ideal drive impedance is zero.

However, there's a possible fly in the ointment. Is it really a loudspeaker output or is it an output designed to run powered loudspeakers which have an inherently higher input impedance than passive loudspeakers. If it's the latter, then the inclusion of a buffering amplifier may be of benefit. In either case, the following still applies...

There are still other criteria to take into consideration such as audible noise and the frequency response of the loudspeaker output, but the addition of a further stage of amplification or buffering between this output and the headphones is likely to do little if anything to improve either.

Having said all of this, it's largely based around speculation and the only way to tell what's going on for sure is to dive in there with an oscilloscope to measure the frequency response and noise level of the output in question under the load in question.

Cheers, Slipstreem.

[Gelatinous Blob]

Is it really a loudspeaker output or is it an output designed to run powered loudspeakers which have an inherently higher input impedance than passive loudspeakers. If it's the latter, then the inclusion of a buffering amplifier may be of benefit.

It's a typical PC sound card, and AFAIK typical PC speakers are powered, so I would guess that it's not a real loudspeaker output.

There are still other criteria to take into consideration such as audible noise and the frequency response of the loudspeaker output, but the addition of a further stage of amplification or buffering between this output and the headphones is likely to do little if anything to improve either.

Having said all of this, it's largely based around speculation and the only way to tell what's going on for sure is to dive in there with an oscilloscope to measure the frequency response and noise level of the output in question under the load in question.

I don't even know what an oscilloscope is, so that's probably too much trouble for me Hovewer, I just found something suspicious: Creative published the official specs (including frequency response and signal-to-noise levels) for X-Fi Xtreme Music and other X-Fi cards, but not for Xtreme Audio. Looks like they have something to hide, doesn't it?

Having said that, I can see now that if I want to buy something more expensive, I'll have to spend some time (probably several weeks) carefully choosing the right headphones. I don't really have that much time now, so I'll probably buy something cheap (like KOSS UR-40) as a temporary replacement for my Phillipses. I'm still interested in opinions and suggestions concerning sound cards and headphone amps though

Edit: is there any sound card, internal or external, that's known to produce output good enough (in terms of frequency response and noise level) for decent headphones? I'm fed up with Creative and their shady tactics

This post has been edited by Gelatinous Blob: Sep 9 2008, 13:31

[Slipstreem]

Edit: is there any sound card, internal or external, that's known to produce output good enough (in terms of frequency response and noise level) for decent headphones?

There are plenty not carrying the Creative badge that are good enough and even far better. It all depends how a person defines "good enough".

Personally, I've always found the C-Media chipset-based cards to be more than sufficient in all respects. Take a look HERE for the soundcard I currently use for driving Sennheiser HD447 headphones from the 'speaker' output and a separate DAC via the optical digital for loudspeaker listening. They're not cheap though.

Cheers, Slipstreem.

[uart]

Edit: is there any sound card, internal or external, that's known to produce output good enough (in terms of frequency response and noise level) for decent headphones? I'm fed up with Creative and their shady tactics

I think high frequency response is the least likely problem. Low frequency response may be an issue with lesser cards. Very high quality coupling capacitors are not cheap and cutting cost here can lead to degraded low frequency response. I've found a telling tests here is simply to look at the harmonic distortion near the low end, say a 50Hz tone for example. Low quality cards very often show a relatively high amount of harmonic distortion at the very low end, even when the THD is totally fine when measured at a higher frequency. This is a classic example of how a manufacturer can easily hide their shortcomings by choosing how they measure the specs. THD at 1kHz may be exelent while THD at 50Hz may be terrible. I've found the loopback tests in RMAA (rightmark audio analyser - freeware) to be a very good way of looking for this.

[Gelatinous Blob]

Today I had a chance to test a pair of Sennheiser HD 555s and I have to say that I'm not all that satisfied with my current setup any more First I tried out a pair of 515s but I was not impressed - the sound quality was certainly good, but not markedly better than my current Philipses. But then I tried the 555s and OMG! The difference was very noticeable Of course I'll still have to test several models to choose the right one, but at least now I have a proof that it's worth to look for something better. A pair of 595 or good Grados + an amp will ruin me, but I feel that I won't mind

@uart: RMAA needs a decent microphone I suppose? Unfortunately I don't have anything like that. I'll search the web though - maybe someone already tested my card. Creative cards are rather popular after all

[odigg]

A pair of 595 or good Grados + an amp will ruin me, but I feel that I won't mind

I'm going to call BS on headroom's recommendation that the Grado needs a dedicated amp. I purchased an SR225 and Total Bithead from them. I compared the headphones plugged straight into my motherboards onboard sound versus plugging the Grados into the Total Bithead and plugging the line in of the Total Bithead into my motherboards sound. I heard absolutely no difference. I tried the same routine with an Denon D2000 and an Audio Technica AD900. Again, no difference whatsoever.

I could be deaf, but I don't think I'm that deaf. If there was a difference, it was subtle.

I respect headroom as a company, but a lot of their recommendations are based on subjective judgments. In the headphone world everything including IEMs need a dedicated amp.

This post has been edited by odigg: Sep 18 2008, 20:22

[pdq]

These days designing an excellent speaker amplifier is hardly rocket science, but designing an excellent headphone amplifier is kindergarten stuff.

[chelgrian]

These days designing an excellent speaker amplifier is hardly rocket science, but designing an excellent headphone amplifier is kindergarten stuff.

You do have to work within your constraints though. Certainly on some portable devices with very inefficient headphones you run into current limitations.

You can also run into problems, even in expensive bits of kit, with the wide variation of impedances. For example I've consistently found that the headphone outputs on Yamaha M7CL digital mixing desks is very noisy using 70 Ohm Sennheiser HD25s, however they are perfectly quiet when the same desk is used with a pair of 250 Ohm Beyerdynamic DT150.

[Martel]

Today I had a chance to test a pair of Sennheiser HD 555s and I have to say that I'm not all that satisfied with my current setup any more First I tried out a pair of 515s but I was not impressed - the sound quality was certainly good, but not markedly better than my current Philipses. But then I tried the 555s and OMG! The difference was very noticeable Of course I'll still have to test several models to choose the right one, but at least now I have a proof that it's worth to look for something better. A pair of 595 or good Grados + an amp will ruin me, but I feel that I won't mind

@uart: RMAA needs a decent microphone I suppose? Unfortunately I don't have anything like that. I'll search the web though - maybe someone already tested my card. Creative cards are rather popular after all

I really don't see much benefit of having an external amp other than max. volume. The signal coming from the card will be limited by card's frequency response/distortion/noise and the external amp is not likely to improve that (it can always make it worse, though).
Some headphones have fairly high impedance (400+ ohm) but this limits only the maximum volume. In fact, the higher the impedance the less it strains the amplifier. If you have headphones with high enough sensitivity (110+db), you need not fear that the conventional cards will not be able to "feed" them.

[Gelatinous Blob]

I really don't see much benefit of having an external amp other than max. volume.

You mean that you actually tested something like Sennheiser 555 or 595 on a PC with and without an external amp and you hear no difference? Or just theoretically?
Remember that I don't even have a headphone jack, only the line-out that is designed to work with powered PC speakers.

Some headphones have fairly high impedance (400+ ohm) but this limits only the maximum volume.

Are you sure about that? I've read many times that headphones that are not driven correctly will sound worse, not only quieter.

If you have headphones with high enough sensitivity (110+db), you need not fear that the conventional cards will not be able to "feed" them.

After I select my new headphones (which won't happen soon...), I'll think about some way to test whether they are working at their full capacity without a headphone amp

[Slipstreem]

Some headphones have fairly high impedance (400+ ohm) but this limits only the maximum volume.

Are you sure about that? I've read many times that headphones that are not driven correctly will sound worse, not only quieter.

Yes. The only requirement of an amplifier that matters in relation to the headphone impedance is that the output impedance of the amplifier should be significantly lower than that of the headphones. In a theoretically perfect world, the output impedance of the amplifier will be zero.

Deliberately designing an amplifier to have an impedance that matches the headphones is a ridiculous and unnecessary task. It wastes half of the power in the output devices before it even gets to the headphones for starters.

An amplifier with a voltage gain of 1 and a high uniform output impedance is electrically the equivalent of a resistor. You can buy a couple of resistors for a few pence and put them in series with your headphones if you like, but they'll either sound exactly the same or slightly worse. They're extremely unlikely to sound any better.

Cheers, Slipstreem.

This post has been edited by Slipstreem: Sep 24 2008, 16:02

[odigg]

You mean that you actually tested something like Sennheiser 555 or 595 on a PC with and without an external amp and you hear no difference? Or just theoretically?

I have heard many low impedance high(ish) sensitivity headphones (Grado SR125, Grado SR225, Denon AH-D2000/D5000) that many people claim are absolutely horrible without a dedicated amp when using a sound card stereo out. In my experience a dedicated headphone amp offered little to no change. I do not even have a special sound card, just a standard motherboard stereo out.

Are you sure about that? I've read many times that headphones that are not driven correctly will sound worse, not only quieter.

You are aware that many of the people making these claims are also people who claim that there there a "night and day" difference in "sound quality" between a $2000 CD player a $500 CD player? Even if you assume that the $2000 player is doing something the $500 player cannot, it is nearly impossible to believe the $500 is doing something so poorly there is a night and day difference between it and a $2000 player.

I will concede that with some particular high impedance low sensitivity headphones improve with a dedicated amp, although I have not undergone enough blind testing to make this statement with any confidence. The improvement, however, is incremental.

This post has been edited by odigg: Sep 24 2008, 20:46

[CSMR]

I wouldn't attach headphones to integrated sound for regular listening.
Integrated sound to amp, if there are no problems, can be good.

The problems with not having an amp are:
-Volume: no volume control so you have to use digital volume control (very bad!)
-Amplification: Lower output impedence (so better frequency response). Measure the frequency response from consumer headphone outs when you attach a headphone. It can change significantly and that's bad. Also lower noise, there can be a (faintly) audible noise floor on integrated audio.

You might not have enough volume from integrated sound but that is unlikely and if it is the case you should listen to quieter music to avoid damaging your ears.

This post has been edited by CSMR: Sep 25 2008, 05:13

[uart]

@uart: RMAA needs a decent microphone I suppose? Unfortunately I don't have anything like that. I'll search the web though - maybe someone already tested my card. Creative cards are rather popular after all

Sorry Blob I didn't notice this before so I didn't reply earlier.

No you just need to use a straight cable loopback (line-out to line-in) to test the soundcard. If you get low noise and good linearity (low THD and IMD) then you can be pretty well assured that both the line-in (recording) and line-out (playback) on your sound card are in great shape. It's a very easy and also a very telling test IMHO.

This post has been edited by uart: Sep 25 2008, 19:26