Review and Measurements of Benchmark HPA4 Headphone Amp/Pre

[DACs_Lover]

There is a lot of technical jargon on this thread regarding the HPA4/LA4 and certainly the measurements are very very impressive but the question is does it sound any better than just a DAC3 HGC and AHB2 ?

I have owned the DAC3 HGC and AHB2 along with B&W 804D3 speakers for a few years now and to be honest I have not been very happy with the performance to the extent that I was actually thinking of selling my Benchmark kit for something else. I use a music streamer as my source.

The sound is very clean and the detail is good but there is just something missing that to me makes the sound boring and not engaging. Some of my friends even described it as too clinical. I used to sit there and play a track and then midway through it jump to another and another because I was never happy with the sound I was getting.

I always knew the AHB2 was a very good amp and my suspicion fell on the digital volume control in the DAC3. Good as it is I always felt that it was a compromise in the DAC3.

After reading a few reviews I decided to take a chance with the LA4 ( I don't use headphones) to see if it will make any difference.

From the moment I heard the first few bars of the first track I knew this was something very special. That first day with the LA4 I sat there and listened to music continuously for 5 solid hours without skipping a single track. It was as if I had completely replaced my system with something much much better. I was gobsmacked and never in a million years expected to hear such a massive improvement. It is now day 5 and I'm still listening for a couple of hours each night whereas previously I would maybe listen to the system twice a month and that was for the sake of not letting it sit there collecting dust. Standard 16 bit CD music now sounds like Hi-Res 24 bit music and 24 bit is now just completely amazing.

I'm not very good at describing what I hear but here goes

The soundstage was very good to start with but has now become vast, instruments have plenty of space and timbre and I'm hearing detail and micro detail that I never knew existed. Vocals sound better, more natural with no harshness at all even on poorly recorded tracks that sounded a bit harsh without the LA4 in place.
The bass is tighter and much more rhythmic and easier to follow. When I used to turn up the volume control on the DAC3 everything used to get louder even the quiet bits but with the LA4 the quiet bits stay quiet and the loud bits come in louder so the dynamics are much much better. Most important of all the music is a joy to listen to and I'm completely lost in it. My system sounds so engaging and enjoyable,

I now realise that rather than replacing my system it was crying out for the LA4 all along and what a difference that one component has made. There is no going back to the DAC3 and AHB2 combo now.

So anybody out there wondering what difference an LA4/HPA4 will make to their system just go and buy it cos you have nothing to lose and everything to gain.


I will hopefully replace the DAC3HGC with a DAC4 without the volume pot whenever that comes out and I wish Benchmark would make a Streamer with user replaceable integral SSD storage with the same footprint as the rest of the range.

One thing for sure my LA4 and AHB2 will be with me for years to come.

Does anyone know how to upgrade the firmware on Benchmark equipment as there is no mention of that in the manual ?

I hope you all enjoyed my little review.

Fully agree with you. it sounds very very transparent.

 

[Frank Dernie]

Mr.Siau works for Benchmark. And inserting a Benchmark between DAC3/AHB2 improves the sound clearly. I use this three units combo for 2 years. It's very very transparent.

Being pedantic once the claim is to reduce noise, which is technically correct and understandable, not to improve the sound ( I don't see how it could, personally)
Being pedantic a second time it is either transparent or not (I believe it is) there can be no degree of transparency
Not trying to critiscise just dotting the i and crossing the t.

 

[stan21]

One of the things that is unusual about the LA4 preamplifier is its low noise levels. As you correctly point out, "many preamps are very bad designs". Most preamplifiers will add noise if placed between a Benchmark DAC and a Benchmark AHB2. The DAC3 is one of the lowest noise D/A converters available. To the best of my knowledge, the AHB2 is the lowest noise audio power amplifier available. Most preamplifiers are too noisy to insert between these two devices. In contrast, the LA4 will actually improve the system noise. I did some system measurements to confirm the mathematical calculations.

Here are the results, measured on our APx555b using the SNR measurement function, unweighted, with a 20 Hz to 20 kHz measurement bandwidth:
View attachment 31007
The chart begins with a gain of -2 dB because this is sufficient to drive the AHB2 to maximum output power. At this gain setting, the LA4 improves the system noise performance by 1.1 dB. Note: The raw data shows 1.064 dB, but these numbers should be rounded to the nearest 0.1 dB or 0.05 dB. For the sake of discussion, I will round to the nearest 0.1 dB.

At a moderate -6 dB gain setting (4 dB below maximum power) the noise performance is improved by 4.2 dB when the LA4 is added.

At -12 dB gain (10 dB below maximum power) the noise performance is improved by 7.4 dB.

At -21 dB gain (-19 dB below maximum power, 1 dB above 1W when driving 8-Ohm speakers) the noise performance is improved by 9.1 dB. This gain setting will produce a typical listening level with many loudspeakers.

At all gain settings below -21 dB, the LA4 improves the system noise performance by 9 to 10 dB!

Notice that the attenuator does not get noisier at mid range. Fully passive stepped attenuators and variable resistors will exhibit a bump in the noise at mid volume settings. The actively buffered LA4 stepped attenuator does not have this common problem.

Summary:

At typical listening levels, the LA4 can reduce the system noise by 7 to 10 dB. If the speakers are efficient enough to reproduce the system noise at levels above 0 dB SPL, then the LA4 can make an audible improvement in the system noise floor. This will happen if the speakers have an efficiency greater than 100 dB at 2.828 Vrms, 1 meter. This would be a gain setting of -22 dB (which I am calculating from the nearest measured point which is -21 dB). If I run this again, I will make a measurement at 2.828 Vrms (1 W into 8 Ohms).

The differences may be more noticeable with headphones or with bridged mono installations.

Hi, I don’t have a technical background, so could someone please explain the progressive loss of resolution at DAC3 volume levels right below below unity gain and lower when the LA4 is not a part of the chain? I understand that the DAC3 uses a digital attenuator, which would attenuate the signal and not the noise floor, but using a 32 bit volume control should have plenty of headroom. Reading a Benchmark white paper from John Saiu, it states “The DAC2 has a 32-bit digital volume control that feeds the array of 32-bit D/A converters. This 32-bit connection allows the use of digital volume control without causing a loss of resolution. At an attenuation of 48 dB, all bits from a 24-bit input still reach the D/A converter.” If i interpret this statement correctly, there should be no SNR loss until basically below -48db, where the digital volume would begin to have a compressing impact on resolution/SNR. In the example from @John_Siau, the loss of resolution starts from -2db below unity gain and progressively increases until we get to about 9-10db difference compared to having LA4 in the chain. Thanks!

 

[John_Siau]

Hi, I don’t have a technical background, so could someone please explain the progressive loss of resolution at DAC3 volume levels right below below unity gain and lower when the LA4 is not a part of the chain? I understand that the DAC3 uses a digital attenuator, which would attenuate the signal and not the noise floor, but using a 32 bit volume control should have plenty of headroom. Reading a Benchmark white paper from John Saiu, it states “The DAC2 has a 32-bit digital volume control that feeds the array of 32-bit D/A converters. This 32-bit connection allows the use of digital volume control without causing a loss of resolution. At an attenuation of 48 dB, all bits from a 24-bit input still reach the D/A converter.” If i interpret this statement correctly, there should be no SNR loss until basically below -48db, where the digital volume would begin to have a compressing impact on resolution/SNR. In the example from @John_Siau, the loss of resolution starts from -2db below unity gain and progressively increases until we get to about 9-10db difference compared to having LA4 in the chain. Thanks!

The DAC3 B has a 125 dB A-weighted SNR at an output level of 24 dBu (at 0 dBFS). This means that the output noise is 125 dB lower than 24 dBu which is -101 dBu. If you turn the input to the DAC3 down in the digital domain, the output level is reduced, but the noise level remains unchanged (-101 dBu). If you turn the digital volume down by 12 dB, the output level will be reduced from 24 dBu to 12 dBu which is 113 dB higher than the -101 dBu noise level. The system SNR at a digital volume setting of -12 dB is 113 dB.

If you insert an LA4, the DAC3 B will feed the LA4 at 24 dBu with a 125 dB SNR. If you turn the LA4 down by 12 dB, the output level will be 12 dBu, just as with the digital volume control. The difference is that the noise will also be attenuated by 12 dB. The noise at the output of the LA4 will be -101 dBu - 12 dBu which is -113 dBu. So, if you do the math, 12 dBu signal - (-113 dBu noise) = 125 dBu SNR. The full 125 dB SNR of the DAC3 is still delivered to the amplifier. So, at an attenuation of 12 dB, the LA4 improves the system SNR by 12 dB which is the equivalent of 2-bits of digital performance (from a noise standpoint). This entire 12 dB improvement is delivered to the speaker terminals if the AHB2 is running in mono mode. The improvement will be 9 dB in stereo mode. Lets look at how this works:

The LA4 has a 137 dB A-weighted SNR, and it maintains this SNR over a wide range of gain settings due to the use of relay-switched resistor attenuators. In a DAC3 > LA4 > AHB2 system, the LA4 is never the limiting factor when it comes to system noise. The LA4 provides an increase in system SNR because it allows the DAC to operate at full output, no matter where the system volume control is set. This improved performance is only delivered to the speaker terminals if the power amplifier (or headphone amplifier) is quiet enough so that the amplifier is not the limiting factor. The AHB2 is probably the only power amplifier quiet enough to make this claim.

The Benchmark AHB2 power amplifier has an A-weighted SNR of 132 dB in stereo mode and 135 dB in mono mode. This means that the limiting factor (in an all-Benchmark system) will be the DAC3 B until the LA4 gain is turned down below -7 dB. The system SNR will be an astonishing 125 dB at the speaker terminals, at gain settings above -7 dB. The clip point of the AHB2 will be reached when the LA4 is set to -2 dB. The AHB2 reaches full output with a +22 dBu input signal.

The A-weighted output noise of the AHB2 in stereo mode is 112 dB below 1 W into 8 Ohms (2.83 Vrms). With the DAC3 and LA4 connected, this noise level is increased by 7 dB (as seen in the prior paragraph) when the LA4 volume is set at -2 dB (the volume setting required to reach full amplifier output). Doing the math, the output noise of the system is 112-7=105 dB below 1 W at full volume. This means that if you speakers have an efficiency less than 105 dB, 1W, 1m, then the acoustic output noise produced by the speakers, will be inaudible at a distance of 1 meter from either speaker when the system is set to full volume (-2 dB on the LA4). If you turn the LA4 down below -7 dB, the output noise of the DAC3 will be lower than the noise produced by the amplifier (which is 112 dB below 1W). This means that the system output noise (at settings below -7 dB) will be inaudible even with very high efficiency horns. In most cases, you will be able to walk up to the speaker an hear absolutely no noise at point blank range (ear to the driver).

If the LA4 is removed from the system, the output noise produced by the amplifier will be entirely due to the noise floor of the DAC3. The SNR of the DAC3 at the amplifier clip point (+22 dBu) is 125 dB - 2 dB = 123 dB. This means the SNR at the output of the amplifier will also be limited to 123 dB, which is 9 dB higher than the noise produced by the amplifier itself. In this system, the noise at the output of the amplifier will be elevated 9 dB above the noise of the amplifier. Instead of 112 dB below 1 W, the noise will be 103 dB below 1 W. If your speakers have an efficiency lower than 103 dB, 1W, 1m, the noise will still be inaudible at your listening position, however, you may be able to hear some faint white noise if you put your ear up the the drivers on one of your speakers.

So, the LA4 provides a measurable improvement in the system noise performance when it is inserted between a DAC3 and an AHB2. This improvement is 2dB to 9 dB in stereo mode and 2 dB to 12 dB in mono mode. At typical listening levels, the improvement is at the upper end of these ranges (9 dB for stereo and 12 dB for mono).

Will you hear a difference? If you have very high efficiency speakers, the answer is yes. You will experience total silence at full volume when the music is not playing. You music will fade into total silence without the distraction of an electronic hiss or hum. The audible difference while the music is playing will be a function of the quality of the recording.

If you are using a lower quality DAC (SNR less than 125 dB) the improvement provided by the LA4 can be even greater.

The system-level SNR improvement is due to the optimization of the "gain staging" between the DAC and the amplifier. The LA4 provides this optimization over a very wide range of volume settings.

 

[RichB]

John_Siau, It's time for my semi-annual plea for LA4/HPA4 discrete input selection via I/R remote when not using a Benchmark DAC.

- Rich

 

[The Equalizer]

Right now I am using an old Benchmark DAC1 (jumpers set for 0 dB XLR attenuation) directly into an AHB2 (sensitivity switch in lowest position). I believe the DAC1 does not employ digital volume. Would there be any system SNR advantage to inserting an LA4 / HPA4 between this converter and amplifier?

One advantage would be better channel balance, I suspect. It's not often that I set the volume knob on the DAC1 much above 9 o'clock.

 

[stan21]

The DAC3 B has a 125 dB A-weighted SNR at an output level of 24 dBu (at 0 dBFS). This means that the output noise is 125 dB lower than 24 dBu which is -101 dBu. If you turn the input to the DAC3 down in the digital domain, the output level is reduced, but the noise level remains unchanged (-101 dBu). If you turn the digital volume down by 12 dB, the output level will be reduced from 24 dBu to 12 dBu which is 113 dB higher than the -101 dBu noise level. The system SNR at a digital volume setting of -12 dB is 113 dB.

If you insert an LA4, the DAC3 B will feed the LA4 at 24 dBu with a 125 dB SNR. If you turn the LA4 down by 12 dB, the output level will be 12 dBu, just as with the digital volume control. The difference is that the noise will also be attenuated by 12 dB. The noise at the output of the LA4 will be -101 dBu - 12 dBu which is -113 dBu. So, if you do the math, 12 dBu signal - (-113 dBu noise) = 125 dBu SNR. The full 125 dB SNR of the DAC3 is still delivered to the amplifier. So, at an attenuation of 12 dB, the LA4 improves the system SNR by 12 dB which is the equivalent of 2-bits of digital performance (from a noise standpoint). This entire 12 dB improvement is delivered to the speaker terminals if the AHB2 is running in mono mode. The improvement will be 9 dB in stereo mode. Lets look at how this works:

The LA4 has a 137 dB A-weighted SNR, and it maintains this SNR over a wide range of gain settings due to the use of relay-switched resistor attenuators. In a DAC3 > LA4 > AHB2 system, the LA4 is never the limiting factor when it comes to system noise. The LA4 provides an increase in system SNR because it allows the DAC to operate at full output, no matter where the system volume control is set. This improved performance is only delivered to the speaker terminals if the power amplifier (or headphone amplifier) is quiet enough so that the amplifier is not the limiting factor. The AHB2 is probably the only power amplifier quiet enough to make this claim.

The Benchmark AHB2 power amplifier has an A-weighted SNR of 132 dB in stereo mode and 135 dB in mono mode. This means that the limiting factor (in an all-Benchmark system) will be the DAC3 B until the LA4 gain is turned down below -7 dB. The system SNR will be an astonishing 125 dB at the speaker terminals, at gain settings above -7 dB. The clip point of the AHB2 will be reached when the LA4 is set to -2 dB. The AHB2 reaches full output with a +22 dBu input signal.

The A-weighted output noise of the AHB2 in stereo mode is 112 dB below 1 W into 8 Ohms (2.83 Vrms). With the DAC3 and LA4 connected, this noise level is increased by 7 dB (as seen in the prior paragraph) when the LA4 volume is set at -2 dB (the volume setting required to reach full amplifier output). Doing the math, the output noise of the system is 112-7=105 dB below 1 W at full volume. This means that if you speakers have an efficiency less than 105 dB, 1W, 1m, then the acoustic output noise produced by the speakers, will be inaudible at a distance of 1 meter from either speaker when the system is set to full volume (-2 dB on the LA4). If you turn the LA4 down below -7 dB, the output noise of the DAC3 will be lower than the noise produced by the amplifier (which is 112 dB below 1W). This means that the system output noise (at settings below -7 dB) will be inaudible even with very high efficiency horns. In most cases, you will be able to walk up to the speaker an hear absolutely no noise at point blank range (ear to the driver).

If the LA4 is removed from the system, the output noise produced by the amplifier will be entirely due to the noise floor of the DAC3. The SNR of the DAC3 at the amplifier clip point (+22 dBu) is 125 dB - 2 dB = 123 dB. This means the SNR at the output of the amplifier will also be limited to 123 dB, which is 9 dB higher than the noise produced by the amplifier itself. In this system, the noise at the output of the amplifier will be elevated 9 dB above the noise of the amplifier. Instead of 112 dB below 1 W, the noise will be 103 dB below 1 W. If your speakers have an efficiency lower than 103 dB, 1W, 1m, the noise will still be inaudible at your listening position, however, you may be able to hear some faint white noise if you put your ear up the the drivers on one of your speakers.

So, the LA4 provides a measurable improvement in the system noise performance when it is inserted between a DAC3 and an AHB2. This improvement is 2dB to 9 dB in stereo mode and 2 dB to 12 dB in mono mode. At typical listening levels, the improvement is at the upper end of these ranges (9 dB for stereo and 12 dB for mono).

Will you hear a difference? If you have very high efficiency speakers, the answer is yes. You will experience total silence at full volume when the music is not playing. You music will fade into total silence without the distraction of an electronic hiss or hum. The audible difference while the music is playing will be a function of the quality of the recording.

If you are using a lower quality DAC (SNR less than 125 dB) the improvement provided by the LA4 can be even greater.

The system-level SNR improvement is due to the optimization of the "gain staging" between the DAC and the amplifier. The LA4 provides this optimization over a very wide range of volume settings.

Thank you, @John_Siau! Your explanations were super helpful for me and I think for anyone interested in understanding the benefits of adding an LA4 to the system. Following your detailed explanation I put together a grid, to help those of us interested visualize the SNR relationships between DAC3, LA4 and AHB2.
There is only one area where I struggled to follow and I assume I lack some fundamental understanding of the relationship between speaker sensitivity and system output noise in relation to 1W/1M. Maybe I don't fully understand the difference between system noise and SNR at specified output. I would be very grateful if you could please clarify the underlined:
-"The A-weighted output noise of the AHB2 in stereo mode is 112 dB below 1 W into 8 Ohms (2.83 Vrms)." Wouldn't the output noise or SNR be 112 db precisely at 1W? Below that I would expect it to be lower, as in the grid?
-"With the DAC3 and LA4 connected, this noise level is increased by 7 dB (as seen in the prior paragraph) when the LA4 volume is set at -2 dB (the volume setting required to reach full amplifier output). Doing the math, the output noise of the system is 112-7=105 dB below 1 W at full volume." The reference to output noise of 105db at full volume (100W 8 ohms) and below 1W at the same time is throwing me off. I understand the 7db noise increase at -2db (full volume) limiting system SNR to 125 (132-7) due to the DAC3 SNR. At 1W, which is -22dB, SNR would be 112db (stereo) with LA4 in the chain since the DAC is not a limiting factor and 103 db without LA4, in my mind. You are saying that system noise is 105 db at 1W at full volume.
-For example, my Martin Logan Montis speakers have 91 dB/2.83 volts/meter sensitivity and they are 4 ohms. 91db would be the SPL at 1.9W/1M, at the -22db volume setting. If the Benchmark DAC3+AHB2 system can resolve 103db at that output level, which is 12db higher than my speakers' acoustic output, does that mean that I can't take full advantage of the Benchmark system's level of detail at that level or any level at that matter. Am I right in thinking that the benchmark system has 125db resolution or approximately 21bits while at that power my speakers have 111 db SPL, which limits my system to 18.5 bits, or do I have this all wrong? Because if that is the case adding additional 9 db, by inserting an LA4 would add absolutely no benefit to my system. Many people are noticing improvements by adding the LA4. I can't imagine all having speakers with 112db sensitivity at 1w/1m, if such efficient speakers even exist, to be able to take full advantage of the added SNR of the LA4. So, hopefully I am wrong in my thinking and confusing terminologies. Thank you!

 

[therobin]

Will you hear a difference? If you have very high efficiency speakers, the answer is yes. You will experience total silence at full volume when the music is not playing. You music will fade into total silence without the distraction of an electronic hiss or hum. The audible difference while the music is playing will be a function of the quality of the recording.

I been curious about this myself. I have DAC3L -> AHB2 (stereo) right now. Eventually I will add a second AHB2 for additional watts. But I was considering getting an LA4 and but never sure if there would be any audible benefit. Given a high quality/mastered source, could you provide an idea of the audible benefits of the LA4 at lower listening levels? Or more so, what output levels would the LA4 be of audible benefit?

The only analog source I have is a game console and turntable. Do you think there would be significant benefits to those over using the DAC3L (I have no idea what the console outputs, but I assume my phono preamp is putting upto 2V on the line out)? Do you have any data to share on the DAC3L analog pot vs LA4?

 

[JohnYang1997]

Thank you, @John_Siau! Your explanations were super helpful for me and I think for anyone interested in understanding the benefits of adding an LA4 to the system. Following your detailed explanation I put together a grid, to help those of us interested visualize the SNR relationships between DAC3, LA4 and AHB2.
There is only area where I struggled to follow and I assume I lack some fundamental understanding of the relationship between speaker sensitivity and system output noise in relation to 1W/1M. Maybe I don't fully understand the difference between system noise and SNR at specified output. I would be very grateful if you could please clarify the underlined:
-"The A-weighted output noise of the AHB2 in stereo mode is 112 dB below 1 W into 8 Ohms (2.83 Vrms)." Wouldn't the output noise or SNR be 112 db precisely at 1W? Below that I would expect it to be lower, as in the grid?
-"With the DAC3 and LA4 connected, this noise level is increased by 7 dB (as seen in the prior paragraph) when the LA4 volume is set at -2 dB (the volume setting required to reach full amplifier output). Doing the math, the output noise of the system is 112-7=105 dB below 1 W at full volume." The reference to output noise of 105db at full volume (100W 8 ohms) and below 1W at the same time is throwing me off. I understand the 7db noise increase at -2db (full volume) limiting system SNR to 125 (132-7) due to the DAC3 SNR. At 1W, which is -22dB, SNR would be 112db (stereo) with LA4 in the chain since the DAC is not a limiting factor and 103 db without LA4, in my mind. You are saying that system noise is 105 db at 1W at full volume.
-For example, my Martin Logan Montis speakers have 91 dB/2.83 volts/meter sensitivity and they are 4 ohms. 91db would be the SPL at 1.9W/1M, at the -22db volume setting. If the Benchmark DAC3+AHB2 system can resolve 103db at that output level, which is 12db higher than my speakers' acoustic output, does that mean that I can't take full advantage of the Benchmark system's level of detail at that level or any level at that matter. Am I right in thinking that the benchmark system has 125db resolution or approximately 21bits while at that power my speakers have 111 db SPL, which limits my system to 18.5 bits, or do I have this all wrong? Because if that is the case adding additional 9 db, by inserting an LA4 would add absolutely no benefit to my system. Many people are noticing improvements by adding the LA4. I can't imagine all having speakers with 112db sensitivity at 1w/1m, if such efficient speakers even exist, to be able to take full advantage of the added SNR of the LA4. So, hopefully I am wrong in my thinking and confusing terminologies. Thank you!
View attachment 139678

Let me point it out for you, if you like.
The fundamental understanding of the matter is that
Digital volume is before the D-A conversion,
Analogue preamp is after the D-A conversion.

Volume control after the dac attenuates noise of the DAC output as well as the signal, preserving the SNR.
Volume control before D-A will leave noise at DAC output constant.

When you get your head around this, you'd realize no matter how many bits of the digital volume control has, -1dB attenuation = 1dB loss in SNR because signal is attenuated but the noise is not.

I don't like doing all the calculations, as it distracts you. In practice, whether you can improve your system by a preamp or not is solely depending on:
Do you hear noise with your current system? (Assuming the amplifier is AHB2)

This is quite straightforward. Hope this is helpful.

 

[stan21]

Let me point it out for you, if you like.
The fundamental understanding of the matter is that
Digital volume is before the D-A conversion,
Analogue preamp is after the D-A conversion.

Volume control after the dac attenuates noise of the DAC output as well as the signal, preserving the SNR.
Volume control before D-A will leave noise at DAC output constant.

When you get your head around this, you'd realize no matter how many bits of the digital volume control has, -1dB attenuation = 1dB loss in SNR because signal is attenuated but the noise is not.

I don't like doing all the calculations, as it distracts you. In practice, whether you can improve your system by a preamp or not is solely depending on:
Do you hear noise with your current system? (Assuming the amplifier is AHB2)

This is quite straightforward. Hope this is helpful.

I can't hear absolutely any noise, nothing if I put my ear on the speaker and obviously that is because my speakers don't have the efficiency of the Benchmark system. So, the ultimate question is, is that a good thing or a bad thing? If you can't hear a thing it means that the speakers are the weakest link and can't fully resolve the capabilities of the system. Wouldn't a little bit of noise mean that you speakers are matching the capabilities of your system?

 

[JohnYang1997]

I can't hear absolutely any noise, nothing if I put my ear on the speaker and obviously that is because my speakers don't have the efficiency of the Benchmark system. So, the ultimate question is, is that a good thing or a bad thing? If you can't hear a thing it means that the speakers are the weakest link and can't fully resolve the capabilities of the system. Wouldn't a little bit of noise mean that you speakers are matching the capabilities of your system?

It's a two side thing.
In terms of speakers sensitivity:
Low sensitivity needs more power. every 3dB lower = 2x power required.
High sensitivity needs a low noise system. every 6dB sensitivity = 2x the noise.

Being in the middle is a better/easier situation that you would not face any issue. This means your electronics are transparent. Improve something else instead.

 

[DACs_Lover]

I been curious about this myself. I have DAC3L -> AHB2 (stereo) right now. Eventually I will add a second AHB2 for additional watts. But I was considering getting an LA4 and but never sure if there would be any audible benefit. Given a high quality/mastered source, could you provide an idea of the audible benefits of the LA4 at lower listening levels? Or more so, what output levels would the LA4 be of audible benefit?

The only analog source I have is a game console and turntable. Do you think there would be significant benefits to those over using the DAC3L (I have no idea what the console outputs, but I assume my phono preamp is putting upto 2V on the line out)? Do you have any data to share on the DAC3L analog pot vs LA4?

You will be shocked. I had Dac3 +AHB2 before. Adding LA4 in between the two units opens width, depth and clarity of the sound. Very very and very true. I have not listened to the DAC 3 direct connection since.

 

[John_Siau]

Thank you, @John_Siau! Your explanations were super helpful for me and I think for anyone interested in understanding the benefits of adding an LA4 to the system. Following your detailed explanation I put together a grid, to help those of us interested visualize the SNR relationships between DAC3, LA4 and AHB2.
There is only one area where I struggled to follow and I assume I lack some fundamental understanding of the relationship between speaker sensitivity and system output noise in relation to 1W/1M. Maybe I don't fully understand the difference between system noise and SNR at specified output. I would be very grateful if you could please clarify the underlined:
-"The A-weighted output noise of the AHB2 in stereo mode is 112 dB below 1 W into 8 Ohms (2.83 Vrms)." Wouldn't the output noise or SNR be 112 db precisely at 1W? Below that I would expect it to be lower, as in the grid?
-"With the DAC3 and LA4 connected, this noise level is increased by 7 dB (as seen in the prior paragraph) when the LA4 volume is set at -2 dB (the volume setting required to reach full amplifier output). Doing the math, the output noise of the system is 112-7=105 dB below 1 W at full volume." The reference to output noise of 105db at full volume (100W 8 ohms) and below 1W at the same time is throwing me off. I understand the 7db noise increase at -2db (full volume) limiting system SNR to 125 (132-7) due to the DAC3 SNR. At 1W, which is -22dB, SNR would be 112db (stereo) with LA4 in the chain since the DAC is not a limiting factor and 103 db without LA4, in my mind. You are saying that system noise is 105 db at 1W at full volume.
-For example, my Martin Logan Montis speakers have 91 dB/2.83 volts/meter sensitivity and they are 4 ohms. 91db would be the SPL at 1.9W/1M, at the -22db volume setting. If the Benchmark DAC3+AHB2 system can resolve 103db at that output level, which is 12db higher than my speakers' acoustic output, does that mean that I can't take full advantage of the Benchmark system's level of detail at that level or any level at that matter. Am I right in thinking that the benchmark system has 125db resolution or approximately 21bits while at that power my speakers have 111 db SPL, which limits my system to 18.5 bits, or do I have this all wrong? Because if that is the case adding additional 9 db, by inserting an LA4 would add absolutely no benefit to my system. Many people are noticing improvements by adding the LA4. I can't imagine all having speakers with 112db sensitivity at 1w/1m, if such efficient speakers even exist, to be able to take full advantage of the added SNR of the LA4. So, hopefully I am wrong in my thinking and confusing terminologies. Thank you!
View attachment 139678

This chart is excellent, but it is an approximation because it does not actually calculate RMS noise summation. It, and my prior verbal description, just make the approximation that the noisiest device determines the resulting system SNR. This approximation works well (error less than 1 dB) when the noise sources differ in magnitude by at least 6 dB, but is not as accurate when the noise sources have similar amplitudes. For example if two noise sources at -125 dB are summed, the resulting noise will be 3 dB higher (-122 dB). The chart above would tell you that the result is -125 dB. So, in the worst cases, the actual noise levels will be 3 dB worse than the chart would predict. For example, at -9 dBFS, the DAC3/LA4 generates noise at -125 dB relative to the signal, and the AHB2 also generates noise at -125 dB relative to the signal. The chart shows the resulting SNR at 125 dB, but this should be 122 dB. In contrast, the numbers are quite accurate at 0 dBFS since the DAC delivers a 125 dB SNR while the AHB2 delivers a 132 dB SNR. The difference between 125 and 132 is 7 dB and this means that the RMS noise summation will be within 1 dB of the estimated 125 dB SNR. If I had more time today, I would update your chart by summing the noise sources instead of picking the highest noise source. These calculations would need to be made in columns 8, 9 and 10.

 

[John_Siau]

Right now I am using an old Benchmark DAC1 (jumpers set for 0 dB XLR attenuation) directly into an AHB2 (sensitivity switch in lowest position). I believe the DAC1 does not employ digital volume. Would there be any system SNR advantage to inserting an LA4 / HPA4 between this converter and amplifier?

One advantage would be better channel balance, I suspect. It's not often that I set the volume knob on the DAC1 much above 9 o'clock.

Adding an LA4 to a DAC1>AHB2 system will provide a greater increase in performance than adding an LA4 to a DAC3>AHB2 system.

One reason for this is that the DAC1 only has a 116 dB SNR. The analog volume control on the DAC1 will deliver a 116 dB SNR between +29 dBu and +20 dBu. If you turn the DAC1 down below +20 dBu, the SNR begins to decrease. Given the fact that the AHB2 reaches full power with an input of +22 dBu, you can only turn your DAC1 down by 2 dB before you start impacting the system SNR. The insertion of an LA4 will give you a system SNR of 116 dB over about a 14 dB range instead of a 2 dB range. You can expect a 12 dB improvement at typical listening levels.

The second, and perhaps more important reason, is that the L/R balance is virtually perfect on the LA4, but it varies on the DAC1, especially at lower volume levels.

 

[DACs_Lover]

Mr.Siau, I have a set of DAC3/LA4/AHB2.
will adding another AHB2 into the system make the sound cleaner / more dynamic / less noise / better soundstage?

I listen at a moderate level (-25 on LA4) with KEF Ref1 85dB /1m

Thank you.

 

[stan21]

Mr.Siau, I have a set of DAC3/LA4/AHB2.
will adding another AHB2 into the system make the sound cleaner / more dynamic / less noise / better soundstage?

I listen at a moderate level (-25 on LA4) with KEF Ref1 85dB /1m

Thank you.

Good question, I am still trying to fully understand the relationship between speaker SPL and system resolution.

Interesting article from John on the relationship between speaker sensitivity and amplifier power. https://benchmarkmedia.com/blogs/application_notes/speaker-efficiency-and-amplifier-power

What is even more interesting to me is the choice of sensitivity for the Benchmark SMS1(discontinued) speakers, which are 85db, as a match for the AHB2. An interesting comment from John: "With any speaker, most musical details are produced with just fractions of a watt." So if the Benchmark system can resolve 115db at 2.83v with LA4 + mono AHB2, where at -115db any micro music detail gets masked by noise and there is no further musical information, but the speakers output 85db, I would think that any fine music detail related to the lowest 30db (115-85), essentially those minute voltages(watts) would be lost to heat due to speaker inefficiency. I hope John can prove me wrong, as I am inclined to think that with low efficiency speakers, any additional SNR increases to a Benchmark system of up to 12db using LA4 and going dual AHB2 would have absolutely no perceivable impact to noise (which you would have none of with 85db speakers at 103db(DAC3+Stereo AHB2)) or to micro-level details. I hope I am wrong because I am itching to buy an LA4, but I want to be convinced it will add a useful sonic benefit to my system and not a placebo impact because I know the Benchmark system now has higher measured performance.

 

[John_Siau]

Good question, I am still trying to fully understand the relationship between speaker SPL and system resolution.

Interesting article from John on the relationship between speaker sensitivity and amplifier power. https://benchmarkmedia.com/blogs/application_notes/speaker-efficiency-and-amplifier-power

What is even more interesting to me is the choice of sensitivity for the Benchmark SMS1(discontinued) speakers, which are 85db, as a match for the AHB2. An interesting comment from John: "With any speaker, most musical details are produced with just fractions of a watt." So if the Benchmark system can resolve 115db at 2.83v with LA4 + mono AHB2, where at -115db any micro music detail gets masked by noise and there is no further musical information, but the speakers output 85db, I would think that any fine music detail related to the lowest 30db (115-85), essentially those minute voltages(watts) would be lost to heat due to speaker inefficiency. I hope John can prove me wrong, as I am inclined to think that with low efficiency speakers, any additional SNR increases to a Benchmark system of up to 12db using LA4 and going dual AHB2 would have absolutely no perceivable impact to noise (which you would have none of with 85db speakers at 103db(DAC3+Stereo AHB2)) or to micro-level details. I hope I am wrong because I am itching to buy an LA4, but I want to be convinced it will add a useful sonic benefit to my system and not a placebo impact because I know the Benchmark system now has higher measured performance.

If the SNR of your playback system exceeds the peak unclipped SPL of your playback system, your system will not produce any audible noise. For a 100W amplifier, add 20 dB to the 2.83 V sensitivity of your speakers, to calculate the peak SPL. If you are running an AHB2 in bridged mono, add 26 dB to the speaker sensitivity.

If you add 20 dB to a sensitivity of 85 dB, you arrive at a peak SPL of 105 dB, which implies that you could get away with a system SNR of just 105 dB. Many power amplifiers have a SNR of about 105 dB. Obviously, the AHB2 is 27 dB quieter than it needs to be for speakers with an 85 dB sensitivity. The real advantage of the AHB2 with the Benchmark SMS1 was the low crossover distortion of the AHB2.

In bridged mono, the peak SPL is 6 dB higher and the noise is 3 dB higher, giving a SNR that is 3 dB better. If you add 26 dB to a sensitivity of 85 dB, you get a peak SPL of 111 dB. In this mode, the AHB2 has a SNR of 135 dB. This means the AHB2 is 24 dB quieter than necessary for this speaker.

We have customers that use the AHB2 with very high efficiency horns. For example, the Avantgarde Trio has a sensitivity of 109 dB at 2.83 Vrms. 109 + 20 = 129 dB SPL at 100 W. This means that you need a power amplifier with at least a 129 dB SNR relative to 100W if you want your system to be noise free. I believe the AHB2 is the only power amplifier that can achieve this.

 

[John_Siau]

Mr.Siau, I have a set of DAC3/LA4/AHB2.
will adding another AHB2 into the system make the sound cleaner / more dynamic / less noise / better soundstage?

I listen at a moderate level (-25 on LA4) with KEF Ref1 85dB /1m

Thank you.

I would love to sell you a second amplifier, but it would not be necessary. The only reason to upgrade to dual AHB2 amplifiers would be to increase your peak SPL by 6 dB. If your LA4 is set to -25 dB, your peak levels are 22 dB below the 100W output of the AHB2 in stereo mode (assuming your AHB2 sensitivity switch is set to 22 dBu (switch in down position). You peak output power is just under 1 W. The 100W/ch AHB2 gives you lots of headroom and it will be absolutely silent.

 

[Omar Cumming]

Of course at 129 dB SPL you are most likely to experience hearing loss, so the SNR is infinite!

Cheers

 

[stan21]

If the SNR of your playback system exceeds the peak unclipped SPL of your playback system, your system will not produce any audible noise. For a 100W amplifier, add 20 dB to the 2.83 V sensitivity of your speakers, to calculate the peak SPL. If you are running an AHB2 in bridged mono, add 26 dB to the speaker sensitivity.

If you add 20 dB to a sensitivity of 85 dB, you arrive at a peak SPL of 105 dB, which implies that you could get away with a system SNR of just 105 dB. Many power amplifiers have a SNR of about 105 dB. Obviously, the AHB2 is 27 dB quieter than it needs to be for speakers with an 85 dB sensitivity. The real advantage of the AHB2 with the Benchmark SMS1 was the low crossover distortion of the AHB2.

In bridged mono, the peak SPL is 6 dB higher and the noise is 3 dB higher, giving a SNR that is 3 dB better. If you add 26 dB to a sensitivity of 85 dB, you get a peak SPL of 111 dB. In this mode, the AHB2 has a SNR of 135 dB. This means the AHB2 is 24 dB quieter than necessary for this speaker.

We have customers that use the AHB2 with very high efficiency horns. For example, the Avantgarde Trio has a sensitivity of 109 dB at 2.83 Vrms. 109 + 20 = 129 dB SPL at 100 W. This means that you need a power amplifier with at least a 129 dB SNR relative to 100W if you want your system to be noise free. I believe the AHB2 is the only power amplifier that can achieve this.

Thank you very much for your prompt responses and the detailed explanations, John! You really helped me piece it all together and visualize it. So i will attempt to verbalize my thoughts as a summary.
My hat is off to you and the Benchmark engineering team for designing these ultra high-quality products, that have given me countless hours of listening joy! Thank you for making yourself available and for the effort you are putting in to answer questions and to educate audiophiles like myself through your posts and release notes on your website, and for being transparent and honest. I can’t think of any other products, at least none that i have come across, where the manuals are so informationally rich, and for a good reason - nothing but solid, top-notch engineering and charts and specs to prove it. Clearly Benchmark products are measurement driven and striving for engineering perfection. Your approach to designing audio equipment- to sound great it must measure great, reminds me of Peter Aczel from the Audio Critic who i highly respect for his no-nonsense, technical approach to evaluating equipment and educating misinformed audiophiles, inundated with purely subjective and deceptive evaluations where somehow average measuring equipment miraculously sounds proportionately good to the price tag. All of these practices only serve to confuse audiophiles and push sales of more and more expensive products leading to absurdities like $10k speaker cables, which don’t measure any better than well designed cables costing a tiny fraction, but somehow possess supernatural properties, bringing the owner sonic nirvana. It takes a lot to excite Peter Aczel about a product, but his glowing review of the Benchmark DAC1 back in 2005, really got my attention and drew me to the company. I am so glad he did. From everything i have learned, analyzing the chart i put together with your help, i see that the LA4/HPA4/AHB2 are so well engineered and far exceed the capabilities of human hearing considering practical listening levels (even at any level if you can bear the loudness), ambient room noise, and sensitivity of most speakers out there, that there is virtually no practical and appreciable sonic improvement possible using the 3 components together, that would bring further sonic benefits. Only on Audio Precision graphs. From what I see, only in the case of using DAC3 without LA4 with sensitive speakers above 93db SPL the system would theoretically start picking up low levels of THD distortion at full power with THD of -113db (good luck to your golden ears). Below -8dbFS THD is -120db. Good luck detecting any distortion at normal listening levels. Noise will be picked up with speakers of 103db SPL or above, for those that have them and can hear it in a normal room at a normal distance. I am so glad i went through this exercise. I would take engineering perfection any day even if i can’t take full advantage of it. Even thinking about the extreme case of those Avantgarde Trios with 109dB sensitivity, as long as an LA4 is inserted in the chain there will be no noise with SNR of 112-115db at 2.83v and no harmonic distortion, which would be at -113db, even if one can bear listening at those levels. Anyone who thinks they can achieve a more perfect sound by “upgrading” from here especially for those of us with speakers where any noise and distortion is buried so deep at 15-30db below the speakers‘ sensitivity are in my humble opinion chasing ghosts. There is zero noise and zero distortion generated by the system. Look at the speakers for any distortion, but it’s not coming from these. The system is basically a straight wire with gain. The only possible situation i can think of where one might hear some differences with other amps is with some extremely difficult speakers, with huge impedance and phase shifts. My speakers have impedance variances from 18ohms to 0.52ohms and phase shifts from -60 to +60 degrees, but the AHB2 drives them effortlessly. They sound fantastic to my years. I am not sure where the limitations of the AHB2 lie in that regard. I would let John fill us in on that. I know some amps can become very unstable with large shifts and not sound good. No offense to anyone, but If they claim they can hear an improvement in sound quality with a different system given the speakers are within the impedance and phase shift capabilities of the AHB2, have other issues. The only people who are excused to hear an improvement are those audio magazines shills who get advertizing revenue from manufacturers and will tell you that Benchmark sounds excellent but that $40k amp sounds even better though it most likely measures worse. My audio electronics quest is over. The only reason I would trade in my Benchmark components for upgraded versions of Benchmark would be for improved visual synergy between the components, but that’s a whole other topic that we can get into if anyone is interested.
John, I wish you and the company continued success! To everyone else, happy listening to your Benchmark systems!