• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Small speaker recommendation for desktop PC use?

I've already got G Three and miniDSP Flex Balanced, here are my REW measurements without any PEQ correction. miniDSP UMIK-1 microphone should be more accuracurate compared to the one on iPhone.

Note that bumps at 70, 140, 650, 1500Hz could be from port speaker placement, room modes, SBIR, or desk reflections as those bumps are present in other speakers measurement in the same room.

View attachment 398664View attachment 398665

And KRK GoAux4 measurement for comparison (I took those measurements last month, so the air temperature/microphone position/speaker position not 100% the same)

View attachment 398668View attachment 398672

Overlayed GoAux4 and G Three graphs:
View attachment 398677View attachment 398678
I’m impressed how detailed is the graphic, compared to the WiiM Ultra ones… thinking about returning my WiiM and not my G Threes :)
 
I've already got G Three and miniDSP Flex Balanced, here are my REW measurements without any PEQ correction. miniDSP UMIK-1 microphone should be more accuracurate compared to the one on iPhone.

Note that bumps at 70, 140, 650, 1500Hz could be from port speaker placement, room modes, SBIR, or desk reflections as those bumps are present in other speakers measurement in the same room.

View attachment 398664View attachment 398665

And KRK GoAux4 measurement for comparison (I took those measurements last month, so the air temperature/microphone position/speaker position not 100% the same)

View attachment 398668View attachment 398672

Overlayed GoAux4 and G Three graphs:
View attachment 398677View attachment 398678
By the way, which sensitivity did you choose? They sound ok with by default one, or you selected the -10 dB ?
 
If you have a balanced DAC, you will surprise how it improves the sound. Mine is not expensive (Focusrite Scarlett 2i2 4th gen) and there’s a nice Topping D50 of relatively cheap price with XLR outputs
I’m impressed how detailed is the graphic, compared to the WiiM Ultra ones… thinking about returning my WiiM and not my G Threes :)
Is that really true? ASR users does not seem to be convinced that balanced/unbalanced make significant difference in the audio quality despite balanced having advantages in picking up less noise:

Also it contradicts with miniDSP explanation that you can convert TRS balanced to RCA unbalanced with minimum detrimental effect on the audio quality.


Using good DACs makes difference? Maybe yes in some caess, but balanced vs unbalnaced makes a difference? Maybe not.

By the way, which sensitivity did you choose? They sound ok with by default one, or you selected the -10 dB ?
I turned on -10dB switch just because I don't listen that loud, and didn't think that the switch makes significant difference. The speakers sound OK to me with that.

The speakers hiss slightly even without XLR cable connected, so I suspect that the hiss is from amplifier itself.
Fortunatelly it's not very loud even at 0.7m distance, fan noises from the computer, air conditioner noise, other various noises makes it hard to hear.(although the hiss is louder than other speakers I owned, such as KRK GoAux4, KEF LSX II LT).
 
Is that really true? ASR users does not seem to be convinced that balanced/unbalanced make significant difference in the audio quality despite balanced having advantages in picking up less noise:

Also it contradicts with miniDSP explanation that you can convert TRS balanced to RCA unbalanced with minimum detrimental effect on the audio quality.


Using good DACs makes difference? Maybe yes in some caess, but balanced vs unbalnaced makes a difference? Maybe not.


I turned on -10dB switch just because I don't listen that loud, and didn't think that the switch makes significant difference. The speakers sound OK to me with that.

The speakers hiss slightly even without XLR cable connected, so I suspect that the hiss is from amplifier itself.
Fortunatelly it's not very loud even at 0.7m distance, fan noises from the computer, air conditioner noise, other various noises makes it hard to hear.(although the hiss is louder than other speakers I owned, such as KRK GoAux4, KEF LSX II LT).
Mines hiss a little also, whatever the connection.

Honestly I’m a little lost in the topic about DACs quality because in principle they are (if well made) all indistinguishable.

Probably the output voltage plays a role more important than balanced-unbalanced or DAC quality: theoretically balance or unbalance signal have nothing to do with voltages, but the practice of symmetric signals doubles the output of majority of DACs.

To compare the sound I have to reduce about 6 dBu the output of my Focusrite which implies using its attenuator.

Also the -10 dB adjustment lets more place to information on the amp, letting 4 dBu more of a 4 volts balanced DAC to increase signal strength.

Post edited: as you, I can’t listen the hiss at my listening distance which is around 1,5 meters, I have to approach to 0,5 m to hear it, which is nonsense.
The date on my G Three is 2023 so you’re right, it is B model with same drivers as 8030C
 
Probably the output voltage plays a role more important than balanced-unbalanced or DAC quality: theoretically balance or unbalance signal have nothing to do with voltages, but the practice of symmetric signals doubles the output of majority of DACs.

To compare the sound I have to reduce about 6 dBu the output of my Focusrite which implies using its attenuator.

Also the -10 dB adjustment lets more place to information on the amp, letting 4 dBu more of a 4 volts balanced DAC to increase signal strength.
Doubling the voltage and decreasing the signal level later in the chain is theoretically better, though I don't think that the difference is audible. The SPL difference when the output voltage is doubled is only 3dB.


Turning on -10dB switch is similarly theoretically better than 0dB. It's probably NOT audible either though, because DACs today have really low noise floor, decreasing the volume by 10dB on the DAC to compensate only results in a non-audible performance hit.

In any case, You should avoid subjectively testing DACs without an equipment for ABX blind testing properly. Human brain is only capable of remembering the details of sound for a few seconds. You'll lose the memory of the previous DAC's sound if you take 1 minute to switch DAC and re-connect equipments, and increases the chance of recalling false memory. Unless the difference between 2 DACs is really huge (it's usually not, at least for well made DACs like you mentioned), you'll not be able to tell the difference correctly this way.

Matching the SPL accurately is also important when comparing two equipments. Even slight increase in the SPL could make you think that one with louder volume is better.

 
Doubling the voltage and decreasing the signal level later in the chain is theoretically better, though I don't think that the difference is audible. The SPL difference when the output voltage is doubled is only 3dB.


Turning on -10dB switch is similarly theoretically better than 0dB. It's probably NOT audible either though, because DACs today have really low noise floor, decreasing the volume by 10dB on the DAC to compensate only results in a non-audible performance hit.

In any case, You should avoid subjectively testing DACs without an equipment for ABX blind testing properly. Human brain is only capable of remembering the details of sound for a few seconds. You'll lose the memory of the previous DAC's sound if you take 1 minute to switch DAC and re-connect equipments, and increases the chance of recalling false memory. Unless the difference between 2 DACs is really huge (it's usually not, at least for well made DACs like you mentioned), you'll not be able to tell the difference correctly this way.

Matching the SPL accurately is also important when comparing two equipments. Even slight increase in the SPL could make you think that one with louder volume is better.

I think the SPL changes when doubling the voltage is 6 dB per speaker… doubling the wattage is 3 dB.

The THD and SNR change along the signal chain, with each gain stage: depending on what music you’re listening to it maybe very different. Not so hot on electric instruments which carry their own distortion effects but I invite you to listen to a classical concerto…

The more information you send to your speakers, the better they sound: with their native sensitivity I’m only using 25% of a 2 volts DAC, and so with one single speaker…

In classical music I notice it way more, and think that psychoacoustic effect is towards the loudest sound, not the opposite. Also fine tuning volume is better to precise listening.

In this case is also the linearity of the monitor amplification, not only the DAC: amplification is non totally linear, and usually work better the less you amplify to get the same sound pressure.

Turning down 10 dB the Genelec, you obtain less distortion (you increase the THD on the DAC and decrease THD on the amp, which compensates in general because linearity on DACs is higher than on amps) a reproduction closer to the source, less noise, better volume control…

You only loose loudness, but both of us we have the luxe of don’t need so much

POST EDITING: I forgot to mention that this depends on the particular THD and SNR of each one of stages, the Genelec ones may not be one of the bests…

A good analogue attenuator is more useful in this case than lowing the sensitivity, Genelecs tend to work with 1 volt signal
 
Last edited:
Is that really true? ASR users does not seem to be convinced that balanced/unbalanced make significant difference in the audio quality despite balanced having advantages in picking up less noise:
Also it contradicts with miniDSP explanation that you can convert TRS balanced to RCA unbalanced with minimum detrimental effect on the audio quality.
It depends on situation and implementation.

In a domestic setting the common mode rejection of the balanced interconnect isn't usually important. The immunity from ground related noise issues is the main advantage, but assumes the balanced inputs and outputs are correctly implemented (no 'pin 1 problem'). Sadly this isn't always the case even now. Single ended interconnects are inherently at risk of ground related noise issues, but usually it works well enough.

Connecting a single ended output to a balanced input is relatively easy with some potential benefits in CMRR and ground noise immunity over a pure single ended connection. The problems come in trying to connect a balanced output to a single ended input, as it depends a lot on how the balanced output is implemented:
  • If both hot and cold outputs are ground referenced then you can usually use a TRS connector with coax to the RCA, connecting tip to the RCA pin and shield to shield, leaving the ring floating. Using TS to RCA will often work too, but shorts ring to shield. Outputs are usually built to survive this without damage, but it can increase distortion.
  • Some manufacturers explicitly advise against connecting their balanced output to a single ended input. Topping have done this, citing increased distortion. In other cases it is because neither hot nor cold output are ground referenced. This may simply not work, or have a potentially damaging DC offset.
  • Transformer outputs are less common now, but may require the RCA to connect to hot and cold rather than hot and shield (no centre tap).
  • Some balanced outputs have a much higher signal level, causing overload issues for a lot of single ended inputs.
I've probably missed some other corner cases. It's safest not to do it unless the manufacturer has given instructions on how to do it correctly.
 
I think the SPL changes when doubling the voltage is 6 dB per speaker… doubling the wattage is 3 dB.
You are correct, I confused doubling the wattage vs doubling the voltage...
I've probably missed some other corner cases. It's safest not to do it unless the manufacturer has given instructions on how to do it correctly.
It could have been a lot simpler to just buy unbalanced miniDSP Flex. Since I already bought balanced Flex I need to live with it...

As you may already know from my previous comments, I bought Hosa YPR-102 and used the only one side of the cable (the side with "TIP" label) to connect to a RCA input only subwoofer, leaving the other side of the cable (the side with "RING" label) disconnected. This is the same as the option 2 recommended by miniDSP's help page, which leaves the ring floating, if I understand correctly. miniDSP does not mention if option 2 increases distortion though.

So if I get an equipment with only balanced outputs from other manufacturers in the future I should check instructions every time? That's a lot complicated than I thought...

  • If both hot and cold outputs are ground referenced then you can usually use a TRS connector with coax to the RCA, connecting tip to the RCA pin and shield to shield, leaving the ring floating. Using TS to RCA will often work too, but shorts ring to shield. Outputs are usually built to survive this without damage, but it can increase distortion.

 
Back
Top Bottom