[miceblue]

.
 

Err.... 20-27 uVrms of residual noise/hum is incredibly high --- that's within their accepted tolerances? I'd accept 15 uV and below, but my personal preference would be <5 uV.

I'm going to assume that (1) high current draw (otherwise it won't generate enough power without multiple step-up transformers and/or DC-DC converters), and (2) lack of galvanic isolation (otherwise it wouldn't be able to draw that much current) are to blame for the huge residual noise. But heck, what do I know.

Is it? I don't know these measurements well enough to know what is considered good vs bad.

 

[jonbernard]

I've heard three Geek Outs now, two 450s and a 1000. One of the 450s and the 1000 are silent, even at volumes I would never use for music, but the other 450 has noticeable background hiss even at normal volumes.
 
Unfortunately, I don't have the results of any tests with the AP for any of these.

 

[jexby]

Wow shocker, another data point that shows LH QA on all units is not equally applied.
If this happens with the Pulse there should be riots and refunds, in that order.

 

[tomscy2000]

  Is it? I don't know these measurements well enough to know what is considered good vs bad.

 
Well, I can't be sure whether that parameter actually is testing for residual noise or not, but from what I've seen, it's the only one where the units are measured on the order of microvolts (also represented in terms of dB under full scale or its corresponding percentage), so it's unlikely to be anything else.
 
Consider the "lowly" (but very good for the price) FiiO E10K, which is being sold for <$70. It has QC parameters for its line-out to have <10 uV RMS of hum/noise, while it's headphone-out should have <15 uV RMS of hum/noise. This is a similar comparison, as it is also an USB-powered device. The higher-end Resonessence Concero HP is rated at <3.4 uV (A-weighted). A battery-powered device like the Cowon P1 is measured at ~3 uV (the Calyx M is around 2 uV), while the iBasso DX50 is a bit worse at ~11 uV.
 
For me, I can hear residual noise at moderate listening volumes even for the Concero HP, but I have to really listen for it. In practical terms, it's silent. I remember hearing some very low level hiss with the DX50 and very low level hiss with the E10K.
 
It'd be interesting if you could ask JDS about the residual noise performance of your C5D (it's a simple reference in their DSIII, and they seem willing to publicize measurements), as it is similarly priced to the GO. The GO still has a more advanced design overall, but the C5D has admirable power filtering at its price point.

 

[miceblue]

  Is it? I don't know these measurements well enough to know what is considered good vs bad.

Well, I can't be sure whether that parameter actually is testing for residual noise or not, but from what I've seen, it's the only one where the units are measured on the order of microvolts (also represented in terms of dB under full scale or its corresponding percentage), so it's unlikely to be anything else.

Consider the "lowly" (but very good for the price) FiiO E10K, which is being sold for <$70. It has QC parameters for its line-out to have <10 uV RMS of hum/noise, while it's headphone-out should have <15 uV RMS of hum/noise. This is a similar comparison, as it is also an USB-powered device. The higher-end Resonessence Concero HP is rated at <3.4 uV (A-weighted). A battery-powered device like the Cowon P1 is measured at ~3 uV (the Calyx M is around 2 uV), while the iBasso DX50 is a bit worse at ~11 uV.

For me, I can hear residual noise at moderate listening volumes even for the Concero HP, but I have to really listen for it. In practical terms, it's silent. I remember hearing some very low level hiss with the DX50 and very low level hiss with the E10K.

It'd be interesting if you could ask JDS about the residual noise performance of your C5D (it's a simple reference in their DSIII, and they seem willing to publicize measurements), as it is similarly priced to the GO. The GO still has a more advanced design overall, but the C5D has admirable power filtering at its price point.

Holy cow, then that really is high on the Geek Out with a 19 μV measurement and being able to pass the inspection tests. :/

That's a good suggestion for the C5D. I'll send an e-mail to them shortly. My previous experience with their customer service has been really positive and they usually reply back < 24 hours later.

 

[miceblue]

Wow, talk about top-notch customer service! I already got a reply back.

Here's a converter:
C5 Noise floor = -105dBu = 4.355877 microvolts

Noise measurement for the C5 was recorded here:
http://blog.jdslabs.com/?p=464

C5D has a slightly higher noise floor at -103.3 dBu as opposed to the C5's -105 dBu (5.297557 μV according to this chart):
http://blog.jdslabs.com/?p=722

That's a really impressive measurement relatively speaking.

 

[Supperconductor]

So I finally got to hear my GO 1000 with my friend's Shure SE 846 IEMs. He bought some new tips just for me to try his headphones (yeah, he's a super nice guy).
 
I have to say I was expecting ***SOME*** background noise and hiss but neither of us could detect any. I have to say the SE 846's sounded magnificent. Lots of texture, detail, and slam, but never fatiguing.
 
So for those of you who are hearing some background noise from their GO devices, I say get them exchanged.

 

[miceblue]

So I finally got to hear my GO 1000 with my friend's Shure SE 846 IEMs. He bought some new tips just for me to try his headphones (yeah, he's a super nice guy).

I have to say I was expecting ***SOME*** background noise and hiss but neither of us could detect any. I have to say the SE 846's sounded magnificent. Lots of texture, detail, and slam, but never fatiguing.

So for those of you who are hearing some background noise from their GO devices, I say get them exchanged.

Again, like I've said before. I already sent my unit back and they replied back with those test results and stated my unit passed their inspection tests. With this, I can only conclude that the background hiss is in fact normal.

And as what I, and others, have stated already, the GO 1000 has less noise than the GO 450.

 

[Supperconductor]

Again, like I've said before. I already sent my unit back and they replied back with those test results and stated my unit passed their inspection tests. With this, I can only conclude that the background hiss is in fact normal.

And as what I, and others, have stated already, the GO 1000 has less noise than the GO 450.

That's a shame. With more sensitive cans like IEM's used with a GO 1000, is it possible to attenuate the signal with an in-line resistor? Or will a resistor alter the sound character of the IEM's? With the SE846 I had to use a lot of digital volume reduction which reduces overall resolution. 

 

[miceblue]

Again, like I've said before. I already sent my unit back and they replied back with those test results and stated my unit passed their inspection tests. With this, I can only conclude that the background hiss is in fact normal.


And as what I, and others, have stated already, the GO 1000 has less noise than the GO 450.

That's a shame. With more sensitive cans like IEM's used with a GO 1000, is it possible to attenuate the signal with an in-line resistor? Or will a resistor alter the sound character of the IEM's? With the SE846 I had to use a lot of digital volume reduction which reduces overall resolution. 

I'm actually not sure. I'm not too familiar with how resistors work in regards to output impedance vs headphone/load impedances if they're added on to the headphone cable or amp output. It might not be a good idea to use one if you have an extremely small load like the SE846 (~8 Ω impedance rating) since it could draw too much current; but again, I'm not 100% sure about this.
http://www.soundandvision.com/content/shure-se846-ear-headphone

The SE846 has a lower rated impedance than that of most other earphones, which range from 16 to 25 ohms; the SE846 is spec’d at 9 ohms at 1,000 Hz. Shure engineers are quick to point out that the impedance rises to 16 ohms at 60 Hz, where the current demands are higher.

 

[matbhuvi]

That's a really impressive measurement relatively speaking.

 
Unlike GO, C5D is battery powered device. That is why Shiit Wyrd is giving considerable performance gain in GO for me.

 

[zenpunk]

An usb isolator had zero impact on my go.
Adding a main powered unit to fix issues with a portable device is a bit lame and completely defeat the main purpose of the Geek Out.

 

[MikeyFresh]

An usb isolator had zero impact on my go.
Adding a main powered unit to fix issues with a portable device is a bit lame and completely defeat the main purpose of the Geek Out.

Perhaps your intended "main purpose", but not necessarily that of all others. And that's no attack on you, so don't take it as one.
 
I use my GO1000 as a line level DAC into a stereo pre/headphone amp, driving powered monitor speakers in a bedroom system.
 
Along with a Mac mini running Audirvana+ and the iFi iUSBPower, this makes for a tidy bedroom system that sits in a bay window, taking up zero floor space.
 
Granted that's my particular main purpose, and not necessarily that of most others, but therein lies the point. Not everyone will use the device in the exact same way.
 
I don't know what USB isolator you used, and I don't wish to belabor a point previously covered, but all computers are noisy in the context of acceptable typical noise levels in audio equipment. Most any USB bus powered device will benefit from either battery power or a bus power cleansing/replacement device. The issue those methods intend to fix is not that of the Geek Out nor any other DAC, it fixes the problem of dirty computer power.
 
And no, don't go listening for that as an audible noise floor difference with sensitive IEMs. The noise they address is in the ultrasonic region, it's EMI, no one ever said it would correct for audible noise floor issues unless those are directly caused by EMI's negative effect on the audio circuitry.

 

[zenpunk]

My issue is that quite a few compromises have to be made to build such a small and powerful DAC/Headphone amplifier, which are sadly pretty obvious in the case of the Geek Out.
You would likely get better performance from a compact desktop solution for less, or little more, than the price of the Geek Out +iUSBPower. It would also be a less messy solution.
Audio-GD 11.32 comes to mind but there is plenty of choice those days.
But if you are happy with your set-up then all is well. 

 

 

[matbhuvi]

An usb isolator had zero impact on my go.
Adding a main powered unit to fix issues with a portable device is a bit lame and completely defeat the main purpose of the Geek Out.

The issue is when you use the Geek out with desktop. I can't use GO with my Nexus devices. I am yet to try otg cable, y splitter and external battery power. But, that becomes a transportable rig rather than portable rig imo.

  My issue is that quite a few compromises have to be made to build such a small and powerful DAC/Headphone amplifier, which are sadly pretty obvious in the case of the Geek Out.
You would likely get better performance from a compact desktop solution for less, or little more, than the price of the Geek Out +iUSBPower. It would also be a less messy solution.
Audio-GD 11.32 comes to mind but there is plenty of choice those days.
But if you are happy with your set-up then all is well. 

 

 
Audio GD does punch well above its weight. In a pure desktop setup, i agree with you that 11.32 makes more sense than GO. I have 10ES. But, that is almost as big as my CPU taking 50% of my table space..