It does raise my eyebrows as to the cause??? Very curious
-
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 DAC3
- Thread starter amirm
- Start date
Are you saying it measured well with their equipment?
We need to find out why , don’t we ?
Yes he is.
Huh, strange..
Does he still have the device or it has been returned to the owner?
If the device is ok and it doesn't measure well with Amirs equipment there may be something wrong with his measurement device or with the procedure.
These issues are all at vanishing low levels , they are kinda academic but I’d like us to know why this has occurred.There’s clearly some kind of external variable as of now unknown and I find that unsatisfactory.
At the same time let’s not throw out the baby with the bath water.
Or the issue corrected itself (albeit possibly temporarily) from the shocks of transit. Second Law usually makes this work the other way, but it's certainly possible that a marginal joint or socket connection was rattled enough to make a difference.
OP
- Thread Starter
- #429
Well, half of it we can explain. The older AP analyzer they use can only go up to 32 K points. My FFTs graphs for jitter use 256k point. This gives them much higher resolution to find narrow jitter spikes.
What is puzzling is why the sample they gave me didn't have that problem. Not that they couldn't see the problem in Michael's.
Well, they couldn't say the Dac was defective..
Maybe the Amir sample was their single crown TDA1541..
I hope this is not going to become a trend, especially for the manufactures that send samples to Amir or other reviewers.
A single product modded with high quality parts inside, and the rest cheap junks.
Maybe the Amir sample was their single crown TDA1541..
I hope this is not going to become a trend, especially for the manufactures that send samples to Amir or other reviewers.
A single product modded with high quality parts inside, and the rest cheap junks.
I don’t doubt benchmarks integrity one bit so let’s not try and pour scorn on them. They have acted impeccably throughout all these measurement trails and tribulations.Well, they couldn't say the Dac was defective..
Maybe the Amir sample was their single crown TDA1541..
I hope this is not going to become a trend, especially for the manufactures that send samples to Amir or other reviewers.
A single product modded with high quality parts inside, and the rest cheap junks.
What do you mean by dynamic range? THD+Noise is around 110 dB from the first post; noise floor (unaveraged/no processing gain) around 120 dB from another post. If you use large FFTs as Amir did at one point the noise floor drops to ~-180 dBFS with processing gain.
The Dac3 has a variable analogue output.
Keith
Keith
- Joined
- Feb 23, 2016
- Messages
- 20,696
- Likes
- 37,432
The AES 17 standard for measuring dynamic range is to use a -60 db 1 khz tone. Notch out the tone, and see how much noise is left vs 0 dbFS. This is what Amir has been using of late with the newer AP he has.
Benchmark claims unweighted SNR vs 27 dbu output of 126 db, but I'm not sure of how they measure this in detail.
Looking at Amir's graphs and adapting for FFT gain, I'd say using a full scale signal the noise floor is maybe -110 or -112 db for the 24 khz band. The noise floor may drop a bit at lower signal levels or not. So reading between the lines the AES17 dynamic range is probably not less than 110 db and could be a few db more. Plus probably another 3 or 4 db if A-wtd. So between 110 to 120 db is a good guess.
I don't know if you have a DAC 3 or looking to get one. How much digital attenuation is a lot (repeating Ray's question)? Also remember the DAC 3 has a fair range of adjustment for output levels. Chances are it can be adjusted so you don't use very much excessive digital attenuation. Purite' beat me to it.
- Joined
- Feb 13, 2018
- Messages
- 574
- Likes
- 720
I am in the same situation as rajapruk there: multiway active system with digital EQ and volume control.
Matching the analog output level of the DAC with the sensitivity of the amp is mandatory of course.
But even when this is the case, combining a very low noise floor and a comfortable maximum level at the listening position is not an easy task and requires a high DNR figure, especially when considering the EQ needed for a compression driver on a constant directivity horn (typically 10dB to 15dB of midband attenuation, impacting the maximum SPL but of course not the noise floor).
As it happens, the peak sensitivity of a compression driver on a CD horn also coincides with the peak sensitivity of the ear (around 2 to 3kHz), making things even more complicated. When considering noise floor audibility, A-weighting is of course more meaningful that unweighted.
Matching the analog output level of the DAC with the sensitivity of the amp is mandatory of course.
But even when this is the case, combining a very low noise floor and a comfortable maximum level at the listening position is not an easy task and requires a high DNR figure, especially when considering the EQ needed for a compression driver on a constant directivity horn (typically 10dB to 15dB of midband attenuation, impacting the maximum SPL but of course not the noise floor).
As it happens, the peak sensitivity of a compression driver on a CD horn also coincides with the peak sensitivity of the ear (around 2 to 3kHz), making things even more complicated. When considering noise floor audibility, A-weighting is of course more meaningful that unweighted.
Last edited:
I do not have a dac3. But I am looking for a dac to buy with high dynamic range.
The Orchard Audio Gala Dac looks very promising on this parameter. So I wanted to know how good the state-of-the-art dac3 is in this regard.
The Orchard Audio Gala Dac looks very promising on this parameter. So I wanted to know how good the state-of-the-art dac3 is in this regard.
Similar threads
- Poll