A speaker with a wild impedance swing will have more phase shift and more frequency response variation between the lows and the high and this can be alleviated with a super low output impedance of lets say <0.1ohm as @solderdude commented.
-
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!
Serious Question: How can DAC's have a SOUND SIGNATURE if they measure as transparent? Are that many confused?
- Thread starter Willhelm_Scream
- Start date
From the link you referenced. "Reflections" in transmission line are only important when the length of the line is comparable to or longer than the wavelength of the electrical signal in the line.
solderdude
Grand Contributor
There actually are/were speaker cables that have a (close to) 8ohm characteristic impedance in the audible band. These require additional parts to ensure amps don't start to oscillate because of their high capacitance. Speakers rarely are 8ohm over the entire audible range.
Not that they are actually better though at power transfer.
And speaker cables can actually matter (The R should be low enough) and the C must not be too high as that could potentially destabilize some amps in certain combinations.
For DACs and passive pre-amps the L and R are not of any importance in a cable, only the capacitance is at longer lengths (or with exotic cable) as it can form a low pass that actually could roll-off at the edge of the audible range and potentially be audible.
Not that they are actually better though at power transfer.
And speaker cables can actually matter (The R should be low enough) and the C must not be too high as that could potentially destabilize some amps in certain combinations.
For DACs and passive pre-amps the L and R are not of any importance in a cable, only the capacitance is at longer lengths (or with exotic cable) as it can form a low pass that actually could roll-off at the edge of the audible range and potentially be audible.
Last edited:
C. Cook
Senior Member
- Joined
- Sep 21, 2021
- Messages
- 380
- Likes
- 198
Heh, and interestingly I found myself back here after many years.
www.electronicdesign.com
What's All This Splicing Stuff, Anyhow?
Several months ago, a reader wrote in to one of the local newspapers: "If I want to move my speakers a few feet further from my amplifiers, can I splice in a few more feet of...
www.electronicdesign.com
C. Cook
Senior Member
- Joined
- Sep 21, 2021
- Messages
- 380
- Likes
- 198
Yeah capacitance is all that I was really thinking about at the frequencies and ckts in question (in terms of reactance vs. resistance).
solderdude
Grand Contributor
One can find more nonsense about audio on the web than factual and good info.
A DAC with an output resistance of say ... 50 ohm can see 32nF (32000pF) on its output and still not affect 20kHz (has a 100kHz -3dB roll-off) under the condition that it can drive a 250ohm load at 20kHz. As most opamps do not like to drive anything as low as 600ohm we can set a safe limit at 10nF in this case. 5nF when assuming 100ohm output R.
Most cables are 100pF-250pF per meter so even 25m of cable should not be audible.
That changes when one has a passive volume control (say 10k) which at a certain attenuation can have a 2.5k output resistance. For that combination 650pF would be the max. load capacitance which could be just a few meters of length.
A DAC with an output resistance of say ... 50 ohm can see 32nF (32000pF) on its output and still not affect 20kHz (has a 100kHz -3dB roll-off) under the condition that it can drive a 250ohm load at 20kHz. As most opamps do not like to drive anything as low as 600ohm we can set a safe limit at 10nF in this case. 5nF when assuming 100ohm output R.
Most cables are 100pF-250pF per meter so even 25m of cable should not be audible.
That changes when one has a passive volume control (say 10k) which at a certain attenuation can have a 2.5k output resistance. For that combination 650pF would be the max. load capacitance which could be just a few meters of length.
Last edited:
There are no transmission line effects at audio frequencies with the cable lengths typically used. (For transmission line effects to become an issue, cable length needs to be significant compared with the electrical wavelength) (Not the wavelength in air). Even for a 20KHz signal, and assuming 1/4 wavelength is significant, you'd need a cable in excess of 3000m before transmission line issues would need to be considered.
While cables do have capacitance and inductance, this is too low (in audio applications) to audibly impact the frequency response in cable lengths used in the home.
EDIT : Ninjad by a load of other people above
%20-%20Results%20Header.png)
"High-End" DAC Blind Listening Results - PART II: Results & Analyses
A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.
archimago.blogspot.com
Hmm. Under blind testing, the Apple USB-C dongle can be distinguished from other premium DACs!
100 dB isn’t the technical threshold of audibility, but it is interesting. It’s hard to write this off as pure statistical chance given the repeatability of results in differrent subgroups.
Still, a USB-C will the rest of the budget applied to speakers makes more sense.
Not with statistical significance in the overall group though. IE a significant chance the results would occur with random guessing."High-End" DAC Blind Listening Results - PART II: Results & Analyses
A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.
Hmm. Under blind testing, the Apple USB-C dongle can be distinguished from other premium DACs!
100 dB isn’t the technical threshold of audibility, but it is interesting. It’s hard to write this off as pure statistical chance given the repeatability of results in differrent subgroups.
Still, a USB-C will the rest of the budget applied to speakers makes more sense.
Also sort of confirmed by the group who were most confident in the difference they were hearing - the Dongle scored the same as the most expensive DAC - but still not statistically significant.
The overall group of the United States cannot decide on their favorite presidential candidate. There may not be a statistically significant difference in preference between the options.
But the subgroup of people identifying as Democrats have a strong, statistically significant preference for one candidate and the subgroup of people identifying as Republican have a strong, statistically significant preference for another candidate.
It doesn’t matter if a large group of people are unable to discern differences between DACs or if the people with more money than hearing ability are unable to discern the differences. As you include everyone, you run the risk of a Type II error.
Subpopulation analysis is necessary. The sample sizes of the subpopulations are pretty reasonable. We don’t want to make judgments on small sample sizes when doing subgroups, but don’t misconstrue lack of difference in the population as the whole as meaning that there is no difference.
Mr. Haelscheir
Active Member
- Joined
- Mar 16, 2023
- Messages
- 139
- Likes
- 105
Well, I made myself watch CMA's Ideon Absolute DAC "review" and found myself rather pissed to learn it was little more than a ridiculous maxing out around a "mere" ES9038PRO while I sit here content with two of those and the linear power supply in my FiiO K9 Pro ESS.
Both the whole group and sub groups are needed.
The whole tells us that the differences are so minor or non-existent that most people would be happy with any dac. This group is not as random as the voting population, these are all people with a lot of interest in the subject (perhaps not blue or red but they all have a color), so I don't think there's any error in taking a look at the whole result.
Considering the claims and prices of the Linn kits vs a dongle this is important for the discussions.
Statistical significance in subs tell that the differences exist, I agree, the whole group result does not diminish this find. I think it's important that we don't show one without the other or it will be "no difference" vs "yes difference" headlines which would make no sense. Let's not divide this by party lines.
The whole tells us that the differences are so minor or non-existent that most people would be happy with any dac. This group is not as random as the voting population, these are all people with a lot of interest in the subject (perhaps not blue or red but they all have a color), so I don't think there's any error in taking a look at the whole result.
Considering the claims and prices of the Linn kits vs a dongle this is important for the discussions.
Statistical significance in subs tell that the differences exist, I agree, the whole group result does not diminish this find. I think it's important that we don't show one without the other or it will be "no difference" vs "yes difference" headlines which would make no sense. Let's not divide this by party lines.
- Joined
- Feb 23, 2016
- Messages
- 22,240
- Likes
- 40,733
Another thing about splitting groups up is if you split it enough ways you are likely to find some sub-group that looks significant even if results are random. A stupid easy example is simply taking a spread sheet and generating random numbers which get a 0 and 1. 1's being taken as correct answers. Do that for 1000 samples and there will be groups of answers that are far from random even though the whole thing was by definition 50-50.
Both the whole group and sub groups are needed.
The whole tells us that the differences are so minor or non-existent that most people would be happy with any dac. This group is not as random as the voting population, these are all people with a lot of interest in the subject (perhaps not blue or red but they all have a color), so I don't think there's any error in taking a look at the whole result.
Considering the claims and prices of the Linn kits vs a dongle this is important for the discussions.
Statistical significance in subs tell that the differences exist, I agree, the whole group result does not diminish this find. I think it's important that we don't show one without the other or it will be "no difference" vs "yes difference" headlines which would make no sense. Let's not divide this by party lines.
IMO no matter how results are interpreted they are not flattering for any of the expensive DACs, especially when there are cheap ones that measure better than the dongle.
Not sure about the science behind it but I encourage you to listen to two different DACs and report back. I’ve tested numerous DACs and most have some sound differences. Some do sound the same or pretty close to the same. In my case, the Weiss 204 sounds better than all my DACs under $2500. I also have the RME DAC. I like it a lot but it sounds noticeably different than the 204.
Also check out all of the discussions about the sound of AKM and ESS DAC chips. Lots of discussion about them sounding different.
You can read all about the science of it right here on this very site.
Keith
Keith
Coming from the same dealer that says the regular Eversolo A6 sounds exactly the same as the Eversolo A8 when virtually EVERYONE else in the world says they sound different… they have completely different DAC chips and implementations… if 1000 people listened to both at least 90% would say they sound different. The others may think they sound similar because they either can’t hear it (hearing isn’t able to pick up differences) or haven’t listened to enough gear to distinguish differences. And I don’t really care if Purite or anyone disagrees because I’ve listened to this gear using a double blind study, level matched, blah blah blah… Believe what you want. Your ears are the only ones missing out by believing Purite. My ears are benefitting from actually trying it out and going with the gear that makes my ears happy.
Just one question. Why it doesn't sound better than dacs for example above 10k?
I’m talking about the 204 versus the other DACs I personally own. I can’t speak for every DAC out there. Amir is not saying they sound the same. They are talking about distortion and how they measure. I’m sure a device with high distortion could sound worse than other devices with low distortion. I for one would prefer gear that measures well in ASRs test.
I don’t have any DACs above $10K so I can’t answer it. Most of my DACs are under $1000. I have some between $100-$1500. The Weiss is the best sounding DAC I own. It’s noticeably better enough to my ears that it makes me not want to use my other DACs.
Similar threads
