Quote:



I've tried both the HD650 and HD600 off the STX and they go down to as deep as I can hear at around 20hz. Are you sure you're using the amped output and not the line-outs? The graph you posted is nothing like what I hear everyday through my setup.

Basically,the two most important downsides of the hd 600 for me are the following:
1)If you don't find the right amp-source-cables compinations,then the mids(mainly the vocals) can sound a bit resseced and distant,and the bass could sound not so deep and well controlled(a bit rolled of,like you don't hear the lowest frequences well).
2)the new earpads issue

As this thread has discussion about the STX amping the HD600, and HD650 are of similar impedance, I want to ask something about them.

Discalmer: I have not deep knowledge of volatages, power output or impedance, so pardon if my question is stupid.

OK, so I am thinking about getting the M³ amp, which is known to be able to power the HD650 (or HD600 for that matter) pretty adequately. But before dedicating my finances towards it, I made a little research on the technical side of the amp.

First of all, the HD650 manual says the maximal power output they can take is 500mW. But as far as I know, listeting on normal volumes never goes even remotely close to that power requirement. So let's presume ~100mW would power them to adequate to high listening volumes.

The M³ amplifier with OPA604 opamps (bigger voltage swing than most) outputs 1.8Wrms @ 33Ω, which (if I am correct) should be 180mW @ 330Ω (the Senns). The σ11 power supply should be set to max of 40V (according to AMB's website), so the voltage swing is +-20V at maximum. I am not sure, does the voltage swing alter as the power output with different listening levels?

Then again, I checked the specs of the Texas Instruments 6120A2, which is the Xonar Essence STX's headphone amp chip. It outputs 80mW @ 600Ω, so 160mW @ 300Ω. The voltage swing is +-12V. All other specs, such as THD+N and dynamic range are very good.

With this in mind, I am very surprised to see such close specs as power outputs from a very good amp and a soundcard (even though considered of very high quality). The main argument around this forum for the desktop amps is the power supply and power output. Is the M³ then worth the extra cash for such improvement as power output? Or does the voltage swing also play a big role? I know that measurements are not everything, and I am willing to buy an amp for my Senns to unravel their potential, but as it seems tube amps (such as WA6 in the price range) offer tremendous improvement in the output power department. I am leaning toward SS however, so are there other SS amps with higher power output in the price range?

What more experimente head-fiers are claiming is that the power is not a problem for STX, but rather other specs, such as the TI 6120A2 itself not being the best one, the output full-scale voltage (2 Vrms (5.65 Vp-p)) maybe a little to low for high impedance headphones and so on. Power is definitively not a problem.

lao

This seems the more likely cause, as the voltage outputs of few SS amps are double and above this. Thanks for clearing this one for me.

Quote:



For this price range check out Audio-Gd Compass or the C2C. Massive amount of power! But I think M3 should be a solid buy from all the reviews I've read.

Quote:



Coming from my friend, a live sound engineer as well as big-time computer enthusiast, PC power is pretty much the cleanest thing next to batteries. The analog section of the Essence STX is powered straight from the PSU. So I agree that power is not an issue.

Perhaps exploring op-amp rolling options will also get rid of the bass roll-off (I'm using stock opamps, which are pretty good on this card).

I would venture that the problem comes from digital volume attenuation. At 180mw @ 300 ohms I'm ready to bet that your volume control is set way below 50%.
The problem is that every 6 dB of attenuation the resolution is reduced by 1 bit. With 24-bit DACs and 16-bit content, 8-bits of resolution can be "clipped" before affecting the signal. So with the STX, 48 dB of attenuation won't impact the signal but after that you can say good bye to bass impact among other things.

Try increasing the volume and see how it affects the general balance. If the bass becomes more preeminent (proportionally) that's likely it.

Quote:



wow, what does the voltage control (volume control) of the analog signal at the card's output has to do with the resolution of the digital signal at the card's input? The volume control only control the power sent to the headphone plug, it does not change neither the frequency neither the resolution of the primary digital signal! About the bass present, ofc that on higher volumes you will fill it more present since you need more voltage/power to move the driver with a greater amplitude to create low frequency waves (bass sound): with more power/voltage you move the driver more easily, thus producing a better bass sound when required. And one more thing, if the max power output of the card is 180mW @ 300ohm of load, it means that the 180mW will only be released to the headphone when the volume control is set to it's maximum. (And at this power the hearing experience may be not so pleasant anymore).

lao

Is there a knob to control the volume on the back of the card?
Because if you're using the Windows and/or foobar volume sliders to adjust the volume, it's the digital signal that is affected in the way I explained before getting decoded by the DAC.

What you described is analog volume control and unless Xonars are unlike all the other sound cards on the market, it's not done on the card. On the other hand amps offer analog volume control.

Finally you're right about volume and bass but try that simple test if you have an amp:
1) Windows/Audio app volume control to the max, volume controlled by the amp
2) Amp volume to the max, volume controlled by Windows/Audio app
At equal volumes, there is a lot of detail lost in the second case: it makes high-pitched sounds disappear and the bass looses its definition and extension.

Quote:



actually you sould not use the Windows volume to control the output signal. I do not have this card, but i'm pretty sure that you can configure the your hardware along with Windows in order to not change the final volume through the Windows volume set. What I mean is you bypass any Windows treatment regarding the audio bitstream and let the card make all the processing, including D/A and analog signal amplification. The card do not have a physical knob, but it does have a volume control on its driver/software and it is there where you should control the volume, not on Windows. What you need to realise is that any digital source which is stored in the pc (or coming from outside in a digital way) should be only processed by the STX and Windows should not interfere in the output volume since you are not using any software or hardware controlled by any other piece of software native on the OS, but rather an separate hardware controlled by a separate driver.

Quote:



I've already done a test similar to this one. In fact, what happens here is that when Windows (or any other software) is able to control the volume of the source, it somehow process the digital samples and alter its level creating some distorsion to the signal. If you cannot get rid of software sound control, what you got to do is let the software set to the volume level equivalent to 0 dB (in mac os x it's the highest volume level, maybe for windows too), thus changing the volume at the amp only (in the case of the stx, use the knob from it's software/driver). The good thing of using the amp to alter the volume is that it will not change the signal in the digital domain, but only in the analog domain presented at the amp input, so you don't lose resolution when attenuating and amplifying it.

Windows vista seems to change the digital samples from integer to floating point values, so that when you devise and after multiply you lose less or none resolution. But, there's other thing. For this test you describe, you got to turn off any software sound enhancement tool such as equalizer, surround emulators or other things like that. If you do that you will feel much less difference. Anyway, the best thing to do is never change the volume of the sound on the digital domain, ever.

lao

Quote:



I don't think it's mainly an amping issue, the HD600 simply doesn't have much deep bass. I've taken a measurement that clearly show there is a rolloff from 70 Hz on:



This is not so much the case with the HD650 (driven by the same amp):



Maybe there are amps out there that are able to fill out the lack of deep bass a bit, but in my experience this is an inherent characteristic of the HD600.

Quote:

sfmatt said:

I would venture that the problem comes from digital volume attenuation. At 180mw @ 300 ohms I'm ready to bet that your volume control is set way below 50%.
The problem is that every 6 dB of attenuation the resolution is reduced by 1 bit. With 24-bit DACs and 16-bit content, 8-bits of resolution can be "clipped" before affecting the signal. So with the STX, 48 dB of attenuation won't impact the signal but after that you can say good bye to bass impact among other things.

Try increasing the volume and see how it affects the general balance. If the bass becomes more preeminent (proportionally) that's likely it.

Click to expand...

At - 48db or lower you will not have bass impact anyway regardless of type of volume control. This is due to the ear response ,not loss of bass due to bit dropping. The bass is still there just reduced in volume as is all frequencies at that volume in fact exacly the same as all other frequencies. The digital volume control with 24 bit resolution has truely balance output both in the frequency domain & in terms of channel balance, in fact much better than any analog taper volume at low volumes on the channel balance.
 

Quote:

taso89 said:

@QQQ: The early roll-off, starting at around 90-100Hz results in fairly poor bass reproduction, hence the lack of adequate bass. This is based upon my experiences with calibrated home theater set-ups, as well as of course having measured the frequency response of my HD600. Headroom's graph (http://graphs.headphone.com/graphCom...=0&graphID=573) seems to have a big mid-bass hump and their measurements show a more 'shallow' roll-off, whereas for me the the low end is the problem, it's still audible but is significantly quieter in both measurements and my ear. When compensating with a software equalizer the response was much more flat but I don't find this to be an adequate solution..

@scytheavatar: The HD600 has a reputation for being dark, though not as dark as its HD650 brother.

EDIT: Here are my measurements:

After some more parametric EQ modifications I think I've gotten the HD600 to a place I like now

Eliminated that one large peak at ~3KhZ that makes me twitch sometimes and put some minor boosts in the lower end (2-3dB) to ease the roll-off.

A bit of a happy ending I suppose, and a suggestion for anyone looking to softmod their cans a little bit, not to be afraid to try this themselves.

Click to expand...

Who needs bass below 50 Hz?  This is much ado about nothing.

These are the worst graphs I have ever seen. 20 dB recession at 6kHz???? And both graphs show recessed mids, which is the opposite of how a 600 sounds. If your 600 sounds dark, you have the wrong source and or amp. The 600 is very musical and lively without being bright. With one particular tube I have, my 580 becomes a full bodied, in your face Grado.