Serious Question: How can DAC's have a SOUND SIGNATURE if they measure as transparent? Are that many confused?

[raif71]

I have never really understood the ‘trying to get digital to sound like vinyl’, that’s the last thing I desire.
Keith

Just the benefits of digital but with the sound of vinyl.

 

[JSmith]

A correction.

You have to make sure the DAC's are level matched to within 0.1dB if your going to compare them. No one's having a go at you... it's just the way it is. Members trying to say otherwise will likely be met with some derision here (with fair reason). Even a slightly larger difference will result in one being slightly louder which is often perceived as "different".

If you feel the device is defective as the audio simply sounds "broken", send it back for a refund or swap for a new one then. I hope you purchased from a local seller though...


JSmith

 

[Palladium]

I have never really understood the ‘trying to get digital to sound like vinyl’, that’s the last thing I desire.
Keith

also "Studio monitors are bad because they are too revealing" is one of the most self-unaware thing audiophiles ever uttered.

 

[MaxwellsEq]

Yeah, overall the Modi's fine, just a little soft in general and in bass transients when compared to vinyl. I was just hoping to find something to make those small improvements. Listening to the Topping is like looking at a television at Best Buy where they have it turned up to "flame mode".

The Modi+ has a flat response and has excellent linearity so can not sound soft if operated properly. It's weakness is likely to be with (low levels of) intermodulation distortion which rarely has a softening effect id it's audible.

Vinyl has distortion and resonance at high frequencies, so is useless as a reference for comparison.

 

[signalpath]

These are really low noise floors, much lower than rooms.

Two thoughts.

1) With headphones, ambient noise can be at or below the threshold of hearing (-8dBSPL). Do the psychoacoustic math.

2) I know of one upscale home entertainment install that uses a Makara subwoofer (+143dBSPL) -- more commonly used in studios and live. In a 32-bit audio delivery system, a 143dBSPL peak (audio full-scale) defines an acoustic dynamic range of 151dB. In this signal path, -105dBFS is equivalent to +38dBSPL. Audible? You tell me.

Will most people care about any of this? No. But I do. Our DAC architecture in-development achieves >170dB dynamic range, with 40nVrms broadband noise. This is what I do all day Guys like Paul Kanevsky make my work that much better.

 

[SIY]

-105dBFS is equivalent to +38dBSPL.

You're mixing units.

 

[AnalogSteph]

1) With headphones, ambient noise can be at or below the threshold of hearing (-8dBSPL). Do the psychoacoustic math.

Headphones are actually a fair bit less critical than speakers... uncorrelated white noise disappears somewhere around or a tad under +20 dB SPL, getting lost among our ears' internal racket (blood flow and such; it's even worse for in-ears). The threshold for speaker playback is lower only because those aren't literally strapped to our head, so a location of sound is possible by evaluating correlation between ears while potentially using head movements as well... AFAIK it's about +4 dB SPL.

Hearing threshold for white noise?

So this one might get me the stupid question of the week award or something... but anyway: We all know hearing threshold curves, which are taken using narrow bandwidth tones AFAIK. What do they mean for minimum detectable white noise level though? How low typically is it? Are there any studies...

www.audiosciencereview.com

2) I know of one upscale home entertainment install that uses a Makara subwoofer (+143dBSPL) -- more commonly used in studios and live. In a 32-bit audio delivery system, a 143dBSPL peak (audio full-scale) defines an acoustic dynamic range of 151dB. In this signal path, -105dBFS is equivalent to +38dBSPL. Audible? You tell me.

It's still a subwoofer. It would generally sport a lowpass somewhere. Say we've got an effective bandwidth of 200 Hz to be generous, then noise would already be down by 20 dB over the full 20 kHz. The way our hearing threshold is going up down there (not to mention how hard it is to get rid of low-frequency ambient noise), I doubt what remains would be audible.

Mind you, we've had a guy cursing his Genelec subwoofer here because the driver's breakup modes at a few kHz made internal Class D amplifier noise rather audible. He ultimately got a revised plate amp from Genelec, then the pillow in front of the driver could go.

THREAD DELETED

I can't delete the thread because of some forum rules, so I'm just deleting the content

www.audiosciencereview.com

This illustrates one point - it's all fine and dandy if the DAC is capable of such a dynamic range, but what if the limit is the power amplifier(s) instead? The kind of brute typically used for subs often is a bit rustic in the dynamic range department and might top out at around 105 dB(A).

BTW, a 170 dB dynamic range for an audio DAC is insane - any analog stage you'd want to get that through would have to sport mic preamp level input noise and almost +/- 70 V power supply rails at the same time, even when only asking for unity gain (which isn't the most useful).

 

[signalpath]

You're mixing units.

We can reasonably correlate an electrical dBFS to an acoustic dBSPL, in a fixed test environment.

 

[signalpath]

Headphones are actually a fair bit less critical than speakers... uncorrelated white noise disappears somewhere around or a tad under +20 dB SPL, getting lost among our ears' internal racket (blood flow and such; it's even worse for in-ears). The threshold for speaker playback is lower only because those aren't literally strapped to our head, so a location of sound is possible by evaluating correlation between ears while potentially using head movements as well... AFAIK it's about +4 dB SPL.

Hearing threshold for white noise?

So this one might get me the stupid question of the week award or something... but anyway: We all know hearing threshold curves, which are taken using narrow bandwidth tones AFAIK. What do they mean for minimum detectable white noise level though? How low typically is it? Are there any studies...

www.audiosciencereview.com

It's still a subwoofer. It would generally sport a lowpass somewhere. Say we've got an effective bandwidth of 200 Hz to be generous, then noise would already be down by 20 dB over the full 20 kHz. The way our hearing threshold is going up down there (not to mention how hard it is to get rid of low-frequency ambient noise), I doubt what remains would be audible.

Mind you, we've had a guy cursing his Genelec subwoofer here because the driver's breakup modes at a few kHz made internal Class D amplifier noise rather audible. He ultimately got a revised plate amp from Genelec, then the pillow in front of the driver could go.

THREAD DELETED

I can't delete the thread because of some forum rules, so I'm just deleting the content

www.audiosciencereview.com

This illustrates one point - it's all fine and dandy if the DAC is capable of such a dynamic range, but what if the limit is the power amplifier(s) instead? The kind of brute typically used for subs often is a bit rustic in the dynamic range department and might top out at around 105 dB(A).

BTW, a 170 dB dynamic range for an audio DAC is insane - any analog stage you'd want to get that through would have to sport mic preamp level input noise and almost +/- 70 V power supply rails at the same time, even when only asking for unity gain (which isn't the most useful).

Audiometric testing from long ago (using headphones) found the threshold of hearing to be -8dBSPL at 3kHz. There may be blood flow effects at play, but we can hear program / tone well under a noise floor (see Fielder AES papers from the 1980s).

Dynamic range numbers are agnostic to freq range. Subwoofer freqs at 143dBSPL will be under 50Hz. Otherwise, we damage hearing. A 30Hz signal at +143dBSPL and a 3kHz signal at -8dBSPL defines an acoustic dynamic range of 151dB. This defines a -105dBFS signal at (roughly) 38dBSPL. (yes, I understand that it's not an exact correlation and will vary somewhat due to electrical-to-acoustic conversion factors).

170dB DAC dynamic range is not "insane". All the dynamic range improvement is on the low-side, the noise-side. There is nothing insane about lowering the noise-floor of an audio system. The more the better, I say. Do the math. A DAC with 40nVrms of broadband / unweighted self-noise (-146dBu) and a micamp / delivery system of +29dBu is 175dB dynamic range. A mic amp with +/-18V rails can achieve +29dBu. This is a standard pro operating level. No need for "+/-70V" rails, though many of our products use +/-25V for our Class-A discrete JFET amplifiers.

 

[Ze Frog]

Personally I'm at the enlightened stage in my audio journey these days. If something is shown to be well engineered by actual reviews that we get provided here, sounds good to my ears, well then that's perfectly fine to me and I see no need to chase the next thing anymore. I've sold and given away all of my gear awaiting an upcoming move, then I'll be purchasing some new equipment that will be my final system unless it doesn't last as long as my remaining year's, job done.

Not sure of people's ages here for the majority, but back when I first started out arguably really with what's on offer these days especially if you can look past a brand name and today you can literally get a system that is in many instances far and away better than back in the day. Until recent years the last Hi-fi heydey was the 70's in my opinion, but even the greats from than can be beat for mere pennies these days. Admittedly you don't get the over engineered casework of the 70's era, but then really who needs or wants all those buttons and respective circuits in the chain these days! Plus, I think a large percentage of so called audiophiles etc really can't genuinely hear the difference between certain things in the way the industry likes to convince them such as in the streaming market. Truth is 99% of people are likely buying into something that it's a perceived improvement because it's more expensive or because reviews tell them it's better. A large majority would never pass ABX testing no matter how they like to convince themselves they are gifted with some special attributes in perception. Don't get me wrong, there's definitely people that can, but I really struggle to believe every member of every audio forum has the ability that is likely Uber rare and likely a product of working in an environment whereby they likely have to have highly trained hearing abilities that are in use everyday, not casual music listening Joe Blogs who's brought a $10,000 DAC and ever since has been granted super abilities placing him in a magical tier of human super beings. If I buy some expensive running shoes, it doesn't make me a world class sprinter, lol, or imply I'm suddenly a qualified sports scientist that should be training Olympians no matter how expensive or fancy they may be.

 

[Miss_Sissy]

You're mixing units.

I would bet dollars to pesos that he's not.

 

[solderdude]

A fellow on YouTube compared a bunch of DACs sound,quality. He actually ran audio thru all these DACs (high end, cheap, in-between) 10 times to multiply any audible coloration by 10.

No audible difference. That does not mean that he doesn't have a favorite, of course, but not due to audible sound.

Would have been fun if he had also tested the Topping E2X2

 

[Apolloseylormoon]

Hi, I got my DX9 a month ago. Before that, there were different DACs from topping like e50/l50, d90se, from combines 2/1 matrix mini I pro3. After100 hours of burning the sound of the DX9 is excellent, at the level of $1000 DACs. The headphone gain level is also good. Before DX9 there was an D90se/A90d and it is not much inferior in volume margin. It is great for hifimen Arya steals magnets. If you take into account the functionality, convenience of 2/1, beautiful, unusual design, and the already rare ak4499eq + pleasant remote control, then this is a great deal. I didn't have any problems with any device from topping, everything was done well. It doesn't make sense to describe the sound of the dx9, it's better to listen it before buying, but I can add for myself that it's worth the money.

 

[antcollinet]

170dB DAC dynamic range is not "insane". All the dynamic range improvement is on the low-side, the noise-side

Except it reaches the limit of what is possible with the analogue stage of the dac, which tops out around 22 bits (130dB). So more in the digital realm becomes pretty pointless.

And this is quite apart from the pointlessness of spending money to make the noise “even more“ inaudible.

 

[Deleted member 72099]

I have never really understood the ‘trying to get digital to sound like vinyl’, that’s the last thing I desire.
Keith

We have a bit of a misunderstanding here. The point for me is not to get digital to sound like vinyl. But, since I can't get the actual musicians here in the room with me, the closest available reference I have is the vinyl that was hopefully/presumably made from the same master tape. When I do comparisons to vinyl with the Modi, they are generally fairly close. The E50 completely changes the character of the recording, adding a pronounced upper midrange peak. The E50 gives the illusion of increased "clarity" due to the brightness, but in direct comparison to the Modi and the vinyl it loses detail because it's lost in the glare of the midrange. Much more importantly both the Modi and the vinyl give much more realistic reproduction of a live instrument. I play both electric and acoustic guitar. Good recordings of each sound much closer with the Modi and on vinyl. On the E50, suddenly every guitar has a very pronounced high E string. That's not a coincidence.

 

[Purité Audio]

Vinyl playback has so many forms of distortion it simply can’t be used for comparison.
Domestic audio reproduction isn’t really about trying to re create a ‘live’ sound but accurately reproducing the recording.
Keith

 

[MaxwellsEq]

Vinyl playback has so many forms of distortion it simply can’t be used for comparison.
Domestic audio reproduction isn’t really about trying to re create a ‘live’ sound but accurately reproducing the recording.
Keith

+1

 

[MaxwellsEq]

We have a bit of a misunderstanding here. The point for me is not to get digital to sound like vinyl. But, since I can't get the actual musicians here in the room with me, the closest available reference I have is the vinyl that was hopefully/presumably made from the same master tape. When I do comparisons to vinyl with the Modi, they are generally fairly close. The E50 completely changes the character of the recording, adding a pronounced upper midrange peak. The E50 gives the illusion of increased "clarity" due to the brightness, but in direct comparison to the Modi and the vinyl it loses detail because it's lost in the glare of the midrange. Much more importantly both the Modi and the vinyl give much more realistic reproduction of a live instrument. I play both electric and acoustic guitar. Good recordings of each sound much closer with the Modi and on vinyl. On the E50, suddenly every guitar has a very pronounced high E string. That's not a coincidence.

Master tape has to be heavily modified to cut LPs. The bass below 50 Hz is absolutely nothing like the original recording. The top end is a bunch of distortion and resonance. It's a terrible reference to compare anything else to.

 

[Deleted member 72099]

The Modi+ has a flat response and has excellent linearity so can not sound soft if operated properly. It's weakness is likely to be with (low levels of) intermodulation distortion which rarely has a softening effect id it's audible.

Vinyl has distortion and resonance at high frequencies, so is useless as a reference for comparison.

You're missing the point. First, no question vinyl is imperfect. The point of the test however is to determine what is closest to the original master recording (i.e. the defacto frame of reference). By comparing a vinyl and CD, presumably sourced from the same master tape, that's as close as you can get, barring getting the musicians in the room exactly duplicating their original performance. As far as the Modi sounding a bit soft in comparison. Of course, that's possible if the cartridge is a bit bright, which is certainly possible. The point is, with the Modi to vinyl comparison, we're talking subtleties. With the e50 we're talking glaring differences, not only against the frame of reference but of actual instruments. Regardless of measurements, the E50 is either inaccurate in general, or my particular example is defective, which was the point of my original question (is my E50 typical or defective). If the sound of my E50 is typical, it is a flawed product regardless of test results.

 

[Purité Audio]

I would just forget about using vinyl for any meaningful comparison.
The only artefact we have is the file, if you want live then go to a concert,
Keith