Yes, I have noticed you put effort in to this and providing additional measurements.
-
Welcome to ASR. 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 DAC3
- Thread starter amirm
- Start date
I run my DAC2 HGC in the HT Bypass Mode without jumpered attenuation (+23dBu maximum per the manual) with no ill effects here at home.
Volume control is handled at the preamp (old Krell KCT), mostly because I managed to get the TV Cable Box remote to recognize it, and it is convenient to use.
When I do use the volume control on the DAC2, my ear notices no quality differences at any chosen level. So, it's another thing to not worry about.
Last edited:
The maximum output on the DAC2 HGC was 23 dBu. We decided to increase the calibrated output level on the DAC3 HGC to +24 dBu. We also added the ability to boost the gain by an additional 3.5 dB of boost allowing a maximum output level of +27.5 dBu at 0 dBFS. The upper 3.5 dB of boost will impinge on the 3.5 dB of headroom over 0 dBFS because the analog clip point of the converter is set at +27.5 dBu.
At the calibrated 'HT' bypass setting the headroom is 3.5 dB. Above this operating point, the headroom in the ESS is reduced. But, the DSP ahead of the ESS still has 3.5 dB of headroom above 0 dBFS.
At the calibrated 'HT' bypass setting the headroom is 3.5 dB. Above this operating point, the headroom in the ESS is reduced. But, the DSP ahead of the ESS still has 3.5 dB of headroom above 0 dBFS.
svart-hvitt
Major Contributor
- Joined
- Aug 31, 2017
- Messages
- 2,375
- Likes
- 1,267
@John_Siau , I have no need for a new DAC, or a new amplifier, or a new preamp/headphone stage, but I wish I did - so I could order one of your products.
I have one question to you which may seem «audiophile» related. It’s about clocks...
Some audiophiles spend thousands on external clocks, though measurements never back up the hypothesis that these clocks improve the audio system’s specifications. So I was surprised to see that an improved clock in RME’s DAC, which used (!?) to be @amirm ’s benchmark, also improved the DAC specifications.
(https://www.rme-audio.de/download/adi2dac_e.pdf)
So my question is: What’s your view on the clock’s importance in a DAC or ADC? And would a newer clock - like the one in RME’s design - also improve the specifications of the DAC3?
I have one question to you which may seem «audiophile» related. It’s about clocks...
Some audiophiles spend thousands on external clocks, though measurements never back up the hypothesis that these clocks improve the audio system’s specifications. So I was surprised to see that an improved clock in RME’s DAC, which used (!?) to be @amirm ’s benchmark, also improved the DAC specifications.
(https://www.rme-audio.de/download/adi2dac_e.pdf)
So my question is: What’s your view on the clock’s importance in a DAC or ADC? And would a newer clock - like the one in RME’s design - also improve the specifications of the DAC3?
jaytrinitron
Member
- Joined
- Jan 27, 2018
- Messages
- 40
- Likes
- 4
First off, I apologize for my lack of knowledge and none of this is meant to be accusatory.
If it is noise modulation or crosstalk as you say, shouldn't a DAC have the capability to reject these things? You say it is noise contaminating the measurement but the noise will still affect the single-ended output in typical use, right? In fact, wouldn't it be much higher since the analyzer is probably designed not to color the measurements too much?
And could this issue be solved by buying a pair of XLR to RCA cables?
Can a car's tire fix the potholes in the road?
OK, less snotty, some things are just beyond what the DAC itself can do, but better ones do everything they can. And all the data presented indicate the Benchmark is exemplary in that regard.
Noise modulation is a function of the underlying DAC architecture; that is, the design of the actual data converter IC, and out of Benchmark's control (other than the choice of chip). Crosstalk can happen inside the chip, on the board, etc. It's worth noting (again) that most of these things appear in rigorous testing and are present at levels far beneath what we can hear.
If the noise is external common-mode noise, or a ground loop, or other noise source that acts differently on single-ended vs. balanced signals, then using XLRs might help. Using RCAs converts back to single-ended and obviates all the differential "goodness" you had with XLRs so is usually a step backwards.
IMO - Don
OK, less snotty, some things are just beyond what the DAC itself can do, but better ones do everything they can. And all the data presented indicate the Benchmark is exemplary in that regard.
Noise modulation is a function of the underlying DAC architecture; that is, the design of the actual data converter IC, and out of Benchmark's control (other than the choice of chip). Crosstalk can happen inside the chip, on the board, etc. It's worth noting (again) that most of these things appear in rigorous testing and are present at levels far beneath what we can hear.
If the noise is external common-mode noise, or a ground loop, or other noise source that acts differently on single-ended vs. balanced signals, then using XLRs might help. Using RCAs converts back to single-ended and obviates all the differential "goodness" you had with XLRs so is usually a step backwards.
IMO - Don
jaytrinitron
Member
- Joined
- Jan 27, 2018
- Messages
- 40
- Likes
- 4
Fair enough, I appreciate the analogy. Though this means there are much cheaper equal/better measuring DACs for single-ended use.
I thought Amir remeasured the new unit and SE and balanced were comparable? Have not been keeping track... But of course there are many cheaper DACs that probably have audibly similar performance (I can't say what someone else might hear).
OP
- Thread Starter
- #389
OP
OP
- Thread Starter
- #392
BTW, I did run a test with a real headphone with the original review sample: https://www.audiosciencereview.com/...easurements-of-benchmark-dac3.3545/post-85580
No, that is one thing that didn't get resolved with the new sample:
Ah, missed that, thanks Amir.
Do you ever sleep?
Fitzcaraldo215
Major Contributor
- Joined
- Mar 4, 2016
- Messages
- 1,440
- Likes
- 647
He won't sleep until he has measured EVERYTHING!,
Good, I've been waiting for him to measure how high is up and all that Schitt...
I think he picked Sir Thomas as moderator simply to have someone on the other side of the pond to prod him awake all night long.
svart-hvitt
Major Contributor
- Joined
- Aug 31, 2017
- Messages
- 2,375
- Likes
- 1,267
Good, I've been waiting for him to measure how high is up and all that Schitt...
I think he picked Sir Thomas as moderator simply to have someone on the other side of the pond to prod him awake all night long.
No, Sir Thomas and Amir are the same guy. Amir just thought it would look too funny if people noticed him being active 24/7 365.25.
So he invented Thomas.
Last edited:
You mean load impedance? 33 and 300 or 600 ohm. One high, one low impedance.Sure. Give me one output impedance to test as otherwise, it is fair bit of work.
Last edited:
OP
- Thread Starter
- #398
Well who knew, not everybody wanted feet like yours...
I think we are a pair of one offs , humanity breaths a sigh of relief...
Similar threads
