Why does it matter? The dac will be the limiting factor in distortion measurements and in noise at most volume range.And that's why I want to see it! Now I want to send my dac in to be connected to this.
-
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!
Review and Measurements of Benchmark HPA4 Headphone Amp/Pre
- Thread starter amirm
- Start date
No BS, we try to avoid stepping in that.
The HPA4/LA4 can improve the system SNR at all volume settings when added to a DAC3 to AHB2 system. Obviously it cannot improve the SNR of the DAC3 itself. It just optimizes the operating point of the D/A converter.
What I failed to point out is that the output noise of the LA4 is less than the equivalent input noise of the very quiet AHB2. The 1.9 uV output noise level of the LA4 is 134 dB lower than the 9.75 Vrms needed to reach the full output of the AHB2 amplifier. Furthermore, this noise level decreases when the LA4 volume is turned down below unity gain. This means that the LA4 will never be the limiting factor in the output noise of the system. The output noise will be determined by the noise of the amplifier, the noise of the D/A converter, or a combination of the two. System output noise is never determined by the LA4 at any volume setting below unity gain.
The stepped attenuator in the LA4 is a very low impedance design. Many of the resistors are less than 1 Ohm. This means that the attenuator must be actively buffered, but it can be shown that this provides many advantages including avoiding the usual rise in noise at mid volume.
Jason Liao (Oppo Digital VP) posted that the UPD-205 (and 105) volume maintained resolution above 70 (100 is max).
My 2-channel connection is the Oppo UDP-205 -> LA4 -> AHB2 (low gain).
I often listen a low to moderate levels so I assume that the LA4 provides two benefits in this configuration: improved signal to noise driving the AHB2s and improved resolution at lower levels. Is this correct?
- Rich
My 2-channel connection is the Oppo UDP-205 -> LA4 -> AHB2 (low gain).
I often listen a low to moderate levels so I assume that the LA4 provides two benefits in this configuration: improved signal to noise driving the AHB2s and improved resolution at lower levels. Is this correct?
- Rich
This is much clearer. But this is common knowledge. It's simply the basics of analog preamp. (tho many preamps are very bad design) The previous post sounds like it's a "new" thing and none other have made it or stuff like that. That's not the case.
I wasn't talking about in a complete system including ahb2 but a dac with preamp. I didn't mention ahb2 in my post. So that's that. I pointed out the key word was "at most volume settings". As you stated here it's consistent.
Yeah I was paid. Like 100M per post.
I just don't like seeing misleading stuff on here. I have to make sure it's clear.
No it bothers me. So I replied. Has anything to do with you? Huh? Sorry. I'm not replying after this. Ignored.
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:
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.
Thanks for the performance comparison. But it should be clear that it's compared to no preamp. (which is not the use case for most people) It's more realistic to compare to some other implementation or use conditions.(not saying to have to make such comparisons to demonstrate)
RME adi2 dac/pro has automated step attenuation to improve SNR at lower range. It's a combination of digital volume control and analog, which seem to be pretty ideal as an implementation.
I don't like to say it, but preamp shouldn't be hard to design. It doesn't need to drive anything more than the cable. For ideal performance it doesn't need more than a modded objective 2. (no xlr of course). It's just most designers out there like coloration from preamp so they chose discrete design with significant amount of noise and distortion. But you get the idea. I do appreciate the implementation of the relay resistor array for the perfect balanced and consistent low noise. For what it is, it's awesome. Just that for the sake of SNR alone, distortion even, it's achievable at very low price point. Simply: opamp buffer in, 1k pot, opamp buffer out. And it will achieve even less than 0.5uV. Some design will still have power supply noise but it's possible to make better design cheaply or to use battery. For benchmark system with dac3 and ahb2, any preamp with inherent noise under 2.4uV(which is the equivalent input noise of AHB2) will not cause bottleneck in the SNR performance. The performance of DAC3 is not in the equation, it will be attenuated by the preamp anyway, so unless there isn't a preamp to attenuation the noise, there won't be difference from the dac.
My summary: The preamp in LA4/HPA4 has superb performance that's good enough to not cause limitation in the best systems. And it does provide large improvement over pure digital control. It may not give benefits over some alternative solutions.
neutralguy
Member
- Joined
- Apr 16, 2019
- Messages
- 68
- Likes
- 157
Can you elaborate on what causes this noise for passive attenuators? And is this the reason why the DAC3 attenuates its DAC output digitally instead of using its analog potentiometer?
First one, passive attenuator is essentially resistor divider. Mid point is where with the constant sum of the two resistors and the largest parallel resistance. And resistance = noise, hence it the noisiest.
Second one, it's a dac, so it's not conventional to have a pot in a dac. And since it's a dac it can always use digital volume control. And no, the mid point won't be an issue in any circuit. 10k pot is essentially better than 99% of the circuit. 1k pot is going to have 0.3uV noise only so I have not seen any active component can have such low noise. Best is to have a couple of ad797 or opa1611 in parallel. But that's too much of an over kill.
One of the best and economical way to implement this is like adi2dac and element x which switch a relay every 20db of attenuation or so. And it's fully digital controlled. This preserve most SNR and won't have very complicated circuit like the one in LA4. Relays and precision matched resistors in that amount are really expensive.
- Joined
- Oct 1, 2018
- Messages
- 826
- Likes
- 1,226
This is astonishing. Never heard of a preamp actually improving your system. I might have to buy one of these.
Wait. Do you work for Benchmark? I feel like that should be disclosed.
Last edited:
https://audiosciencereview.com/forum/index.php?members/john_siau.2646/ open about tab
btw an analog attenuator improving SNR over digital volume is very very basic of electronics. John successfully tricked another one to think this is unusual...
- Joined
- Oct 1, 2018
- Messages
- 826
- Likes
- 1,226
I wish I understood this stuff. I heard it was common knowledge that using a 'normal' preamp circuit is better then using digital control. Something about digital control decreasing resolution for every step of volume you go down, I think. But I've also heard that properly implemented, a digital volume control can be indistinguishable and sound better, and actually improve the sound.
So correct me if I'm wrong, John does work for Benchmark?
Tricked implies deceit.
It's so much easier to be an expert critic than it is to fund, design, and deliver cutting edge products.
- Rich
Last edited:
If I wasn’t an old bloke with more than enough kit and an amp I like I would have bought the Benchmark kit after seeing the Stereophile article. I don’t use this site as purchasing advice but a place to read worthwhile (mainly) discussions about audio.
Engineers often like to focus solely on technology and the result can be a technical monument.
THX had the feed forward technology but it took a partnership with Benchmark to make it real.
Excellent products come from excellent engineers and management. It is not possible to do so consistently without both.
- Rich
No. I am saying that a company that consistently produced SOA products results from a fully competent team including management.
You may have noticed that I am from MA. The Patriots do not consistently win because of Brady and other talented players. They win because of the management, coaches, and players. When their management changes, then so will the team. Nest pas.
- Rich
Good lord. Could you possibly be any more clueless? Please step away from your keyboard.
Let's please be a bit more respectful to Mr Siau, he's large part of the engineering excellence for one of the most respected high performance audio electronics manufacturers in the world. He's definitely not the marketing guy, though seems to of drawn the short straw in having to deal with the online audio forums lol
http://www.aes.org/aes/john-siau
It's a absolute privilege to have him and indeed other knowledgeable industry folks participate here so please treat them with the utmost respect and courtesy.
http://www.aes.org/aes/john-siau
It's a absolute privilege to have him and indeed other knowledgeable industry folks participate here so please treat them with the utmost respect and courtesy.
Similar threads
- Poll
- Poll
- Poll
