SMSL DL100 DAC and Headphone Amp Review

[Cataudio]

Usually I like mini filter.

Sorry what filter is this? The DAC offers FL1-5, FL5 being no filter if I understood well.

 

[Jimster480]

Sorry what filter is this? The DAC offers FL1-5, FL5 being no filter if I understood well.

Check the manual and it will explain which filter is which.

I would use F1 or F2 on this DAC. F2 is the closest to what I used on my other DAC labeled as "Mini" which is Minimum Phase Fast

 

[JIW]

Hi, I just bought this DAC and I know audio like a five year old. Would you be kind enought to tell me what filter should I use?

Filter 2 would be my general recommendation as that has extended frequency response, fast roll off and linear phase.

 

[Cataudio]

Check the manual and it will explain which filter is which.

View attachment 500876

I would use F1 or F2 on this DAC. F2 is the closest to what I used on my other DAC labeled as "Mini" which is Minimum Phase Fast

Filter 2 would be my general recommendation as that has extended frequency response, fast roll off and linear phase.

Thanks!

 

[Bob101]

Hi, I just bought this DAC and I know audio like a five year old. Would you be kind enought to tell me what filter should I use?

These filters are measurably different in pre/post-ringing and phase linearity. It's a mix of trade-offs.
Your room will mask at least approximatelty 82.6% of the ringing (My personal opinion). And more linear, or compensated phase is best on paper. But the human ear's sensitivity for phase shifting is still a topic of dabate in places where debates take place, which means: almost nobody can consistently hear it. The placement of your speakers and speakers with passive crossovers will alter the phase much more than your DAC filter setting, and no speaker review ever mentions phase linearity more than almost not. That should tell you enough. Don't be afraid to admit you can't hear a difference. I don't. For Headphones or near-field listening: then it matters. A little.
NOS, or: no filter, is for the NOS-curious... Maybe it is what you are looking for? On paper it is not advisible, but according to some it's the only true way to listen, like God intended. Again: do not be afraid to admit not hearing a difference, or worse: liking it. it adds distortion, and though distortion sounds like a dirty word: it also really is. But do not overestimate your aversion to said distortion. My first stereo probably had 10% THD+N when not playing any music at all, and man, did it sound great. Because it was mine, loud and it pissed off my parents.
- Slow means less ringing (=better) but less attenuation (=worse). It's a trade-off.
- Phase compensated is best, but comes with some latency, due to processing time. If you don't hear the audio lagging behind what you see in movies or gaming, you're ok. For music this is the default.

I'd set it to F4 and forget about it. You do you.

 

[EJasonC]

I have a 7.2 Surround Sound Processor connected via HDMI ARC to my LG OLED.
Primary source is an Apple TV 4K using the Apple TV remote to control power on/off & volume of the SSP.
Works nearly flawlessly with only an occasional handshake issue which seems to be very common.
But the configuration is very convenient and easy to use.
I would like to connect an SMSL DL100 in the same way as my surround system via HDMI ARC but for another simpler stereo 2.0 channel system.
My question is, does the DL100 function the same way, namely allowing the Apple TV remote to control power & volume of the DL100?
I've read a few comments here that seem to indicate it will work fine, but not really seeing anything very definitive especially from the DL100 manual.
Appreciate any insight from those who have a similar setup.
Thank you in advance.

 

[Cataudio]

These filters are measurably different in pre/post-ringing and phase linearity. It's a mix of trade-offs.
Your room will mask at least approximatelty 82.6% of the ringing (My personal opinion). And more linear, or compensated phase is best on paper. But the human ear's sensitivity for phase shifting is still a topic of dabate in places where debates take place, which means: almost nobody can consistently hear it. The placement of your speakers and speakers with passive crossovers will alter the phase much more than your DAC filter setting, and no speaker review ever mentions phase linearity more than almost not. That should tell you enough. Don't be afraid to admit you can't hear a difference. I don't. For Headphones or near-field listening: then it matters. A little.
NOS, or: no filter, is for the NOS-curious... Maybe it is what you are looking for? On paper it is not advisible, but according to some it's the only true way to listen, like God intended. Again: do not be afraid to admit not hearing a difference, or worse: liking it. it adds distortion, and though distortion sounds like a dirty word: it also really is. But do not overestimate your aversion to said distortion. My first stereo probably had 10% THD+N when not playing any music at all, and man, did it sound great. Because it was mine, loud and it pissed off my parents.
- Slow means less ringing (=better) but less attenuation (=worse). It's a trade-off.
- Phase compensated is best, but comes with some latency, due to processing time. If you don't hear the audio lagging behind what you see in movies or gaming, you're ok. For music this is the default.

I'd set it to F4 and forget about it. You do you.

Man thank you so much for this detailed explanation, it's very clear and helped me a long way! I'll take a moment and try to hear any difference between the filters using both speakers and headphoned. Odds are my bad hearing will miss everything...

It looks like phase compensation does not add any noticeable lag so this good!

 

[scruffy1]

oops, looks like i just "accidentally" bought one of these off the ocau forum - for 40% off and only a fortnight old

i seem to keep doing that stuff - although a clear upgrade from the stoner acoustics ud130 for bedroom duties

 

[scruffy1]

but after some appreciation it won desktop duties instead, and the alo pan am and tone 1 can get that bedroom gig

 

[Cataudio]

Had two surprises today using the headphones output extensively for the first time:
- I just switched to the headphones output and launched a movie, after 15 minutes the DAC muted; had to circle the outputs to find the sound again
- playing different songs on my Fostex T50RP MKII: at high volume (minus 20-15dB) there's a noticeable degradation of sound quality, I believe it's distortion; I know the headphones can do better and the source isn't the problem

I knew the headphones amplifier wasn't great but is the quality that low?

 

[scruffy1]

if it muted i suggest the problem might be the source muting after a period of inactivity, and if it's via usb, my suggestion would be to look at how your computer handles (perceived) inactivity
fwiw, i have connected via optical (cleaner chain, no faffing with usb drivers which are reportedly problematic on zen 2 & 3 based machines), and in my vigorous use since earlier this week, it has no time out on any output to headphones or unbalanced (or all) there is no issue such as you report
if it's balanced output, i can't comment on the situation

most of my use has been akg q701 - higher impedance, but apparently easier to drive than the fostex seem, and yes, they need to be turned up much higher than the ath ad700, but even the focal elear doesn't seem to need a great boost to sound great

it's not so much that the quality isn't great, but you need a much lower impedance to drive your phones than the available options; maybe it cuts out from overloading the amplifier at higher output, given the requirements? not competent to say if that's a real possibility here, but other contributors may be able to offer assessment of that being a cause

claude told me :

The Fostex T50RP MKII is notoriously difficult to drive despite its modest specs on paper:

• Impedance: 50Ω
• Sensitivity: ~92 dB/mW (very low for a 50Ω headphone)

With 16Ω output impedance:

• Damping factor = 50 ÷ 16 = 3.1:1 (below the ideal 8:1)
• This will result in looser bass, reduced control, and slightly altered frequency response
• You'll likely get enough volume, but sound quality won't be optimal

With 32Ω output impedance:

• Damping factor = 50 ÷ 32 = 1.6:1 (quite poor)
• Significant bass bloat and loss of control
• Noticeable degradation in sound quality
• Not recommended

The Real Problem:The T50RP's low sensitivity is the bigger issue. These planar magnetic headphones are power-hungry and really need a proper headphone amplifier with:

• Low output impedance (<2Ω ideally)
• High current output (planar magnetics need lots of current)

Sources with 16-32Ω output impedance are designed for high-impedance headphones (250-600Ω), not the T50RP.

Bottom line: Neither option is ideal, but if forced to choose, 16Ω is less bad than 32Ω. However, the T50RP really deserves a dedicated solid-state amp with <2Ω output impedance and sufficient power to unlock their potential.

so it seems you have bought a knife to a gunfight

claude continues, with my further query:

Yes, lower output impedance is definitely better for the T50RP MKII.

Why lower is better:

1. Better damping factor = tighter bass, more controlled sound, flatter frequency response
2. More accurate sound = the headphones perform closer to their design intent
3. Less coloration = output impedance won't alter the tonal balance

Ideal for T50RP MKII:

• Output impedance: <2Ω (ideally <1Ω)
• This gives you a damping factor of 25:1 or better

But there's a second requirement:

Low output impedance alone isn't enough. The T50RP also needs sufficient power/current because of its low sensitivity (92 dB/mW).

A typical phone might have 1Ω output impedance but only delivers 10-20mW, which will give you:

• Good damping (tight, controlled sound) ✓
• But insufficient volume ✗

What you actually need:A headphone amp with:

• Low output impedance (<2Ω)
• AND enough power output (100-500mW into 50Ω)

Examples of good matches:

• Schiit Magni/Heresy
• JDS Labs Atom
• Topping L30
• O2 amplifier

These typically have <0.5Ω output impedance AND plenty of power for the T50RP's inefficiency.


so it would seem the problem is not the smsl "quality", it's the inappropriate pairing

hope that's helpful for your situation

 

[Cataudio]

if it muted i suggest the problem might be the source muting after a period of inactivity, and if it's via usb, my suggestion would be to look at how your computer handles (perceived) inactivity
fwiw, i have connected via optical (cleaner chain, no faffing with usb drivers which are reportedly problematic on zen 2 & 3 based machines), and in my vigorous use since earlier this week, it has no time out on any output to headphones or unbalanced (or all) there is no issue such as you report
if it's balanced output, i can't comment on the situation

most of my use has been akg q701 - higher impedance, but apparently easier to drive than the fostex seem, and yes, they need to be turned up much higher than the ath ad700, but even the focal elear doesn't seem to need a great boost to sound great

it's not so much that the quality isn't great, but you need a much lower impedance to drive your phones than the available options; maybe it cuts out from overloading the amplifier at higher output, given the requirements? not competent to say if that's a real possibility here, but other contributors may be able to offer assessment of that being a cause

claude told me :

The Fostex T50RP MKII is notoriously difficult to drive despite its modest specs on paper:

• Impedance: 50Ω
• Sensitivity: ~92 dB/mW (very low for a 50Ω headphone)

With 16Ω output impedance:

• Damping factor = 50 ÷ 16 = 3.1:1 (below the ideal 8:1)
• This will result in looser bass, reduced control, and slightly altered frequency response
• You'll likely get enough volume, but sound quality won't be optimal

With 32Ω output impedance:

• Damping factor = 50 ÷ 32 = 1.6:1 (quite poor)
• Significant bass bloat and loss of control
• Noticeable degradation in sound quality
• Not recommended

The Real Problem:The T50RP's low sensitivity is the bigger issue. These planar magnetic headphones are power-hungry and really need a proper headphone amplifier with:

• Low output impedance (<2Ω ideally)
• High current output (planar magnetics need lots of current)

Sources with 16-32Ω output impedance are designed for high-impedance headphones (250-600Ω), not the T50RP.

Bottom line: Neither option is ideal, but if forced to choose, 16Ω is less bad than 32Ω. However, the T50RP really deserves a dedicated solid-state amp with <2Ω output impedance and sufficient power to unlock their potential.

so it seems you have bought a knife to a gunfight

claude continues, with my further query:

Yes, lower output impedance is definitely better for the T50RP MKII.

Why lower is better:

1. Better damping factor = tighter bass, more controlled sound, flatter frequency response
2. More accurate sound = the headphones perform closer to their design intent
3. Less coloration = output impedance won't alter the tonal balance

Ideal for T50RP MKII:

• Output impedance: <2Ω (ideally <1Ω)
• This gives you a damping factor of 25:1 or better

But there's a second requirement:

Low output impedance alone isn't enough. The T50RP also needs sufficient power/current because of its low sensitivity (92 dB/mW).

A typical phone might have 1Ω output impedance but only delivers 10-20mW, which will give you:

• Good damping (tight, controlled sound) ✓
• But insufficient volume ✗

What you actually need:A headphone amp with:

• Low output impedance (<2Ω)
• AND enough power output (100-500mW into 50Ω)

Examples of good matches:

• Schiit Magni/Heresy
• JDS Labs Atom
• Topping L30
• O2 amplifier

These typically have <0.5Ω output impedance AND plenty of power for the T50RP's inefficiency.


so it would seem the problem is not the smsl "quality", it's the inappropriate pairing

hope that's helpful for your situation

Brother, thank you so much for the explanation!

I'm very reassured that's not my unit that's deffective. I've got the O2 and, yes, it drives my headphones correctly. As you found out, the DL100 just doesn't have the specs needed. I'm not disappointed however because I bought the DL100 as a DAC for my speakers and very rarely use my headphones. Also they work fine a normal volume. For the price of the unit it's still great.

For the problem of the HPA output that got muted: I'm also using toslink to a W11 PC and I'm sure it's not the source. I'm using the BAL output to my speakers everyday for two weeks now and never had a problem. As told, yesterday I used HPA for the second time and this strange stop suddenly intervene. Will report if that happens again.

Thanks again for the help!

 

[Cataudio]

Sorry to flood this thread.

I'm using the DL100 for almost a month now and I've a few days ahead of me to decide if I return the unit. Can you help me?

I'm very happy with the practicality of this DAC and its integration in my living room: the design is nice and it has all the features I need, especially the screen that goes off in 5 sec. The remote range is low however but I can live with it.

The headphone outputs stops suddenly and inexplicably (see my post above, it happened two times). It's also not powerful enough for my cans and I'm in the process of buying a separate headphone amp that will represent the biggest investment in my setup, likely a VIolectric V222. This in order to drive the Audeze LCD-X that I'll buy later.

On the other side I'll soon receive two Kali Audio in-8 V2 monitors. They're not very expensive but they're known to be very capable.

I'm not a good audiophile or listener, I can easily hear the differences between two pair of headphones or speakers but the differences between two DAC are much more subtle. That's also true for amplifiers, albeit less, but I'm happy to invest in the V222 as it's powerful and also very qualitative.

So my question is: does the DL100 represents a real weak link as a DAC in the system I'm setting up or do I get very litle upgrading to a better unit?

 

[scruffy1]

the differences between (highly specced) dacs that are in the "excellent" sinad range is likely minimal - because you literally cannot hear a dac if it does its job - it's the most transparent part of the chain if it is working correctly

define "better than already excellent"...

you can buy a more expensive dac, sure, or one with a different chip (ESS, AKM, Texas industries (burr-brown) are the other big names apart from cirrus logic in the smsl), but the real reason for any other option rests more usefully with what additional features they can provide

in the case of the topping e70, there's merit in the "better" dac specifications (ESS), or even the velvet version (AKM), but my one, which feeds the htpc chain wins the prize for 12v trigger so i only need to turn on one piece of equipment and it wakes up the (nad d3020) amp downstream; but fwiw the smsl has hdmi arc, so it will work with a tv with less faffing- and it then comes down to whether a 12v trigger beats hdmi arc connectivity

the dl100 isn't a weak link for dac, but what you might consider is whether an amp as a separate item can successfully drive both your headphones and the anticipated kali speakers you mention - all of which will need adequate amplification quite independent of the analogue signal your dac can supply very nicely

so if the dac sounds okay (if anaemic for driving planars) your next step is maybe an amp that will address that deficiency AND drive passive speakers, in a single package

what that might be is something i can't really recommend, as the smsl rocks all my headphone options, and the klipsch promedia 2.1 thx monitors on (and under) my desk are active, so no need for independent amplification

 

[Cataudio]

Thank you very much for your answer Scruffy, it helps a lot.

The Kali speakers are self-powered so that's one less problem.

I don't need more feature from a DAC than the DL100 already have and I'm very happy with it so I'll stick with my unit. But instead of the V222 I might find a headphone amp that also does preamp so everything works together more easily.

 

[scruffy1]

not sure a pre-amp benefits the kali; others may offer clearer reasoning for that choice, but adding an unnecessary step in the speaker chain is superfluous i would think

wrt the headphones, a dedicated separate headphone amp would benefit, but in terms of controlling the hardware, the smsl can switch the output and if you use the rca out to the speakers, you have a balanced output on the xlr to go to a headphone amp, and gain maximal opportunity to this used matched output or simple stereo plug on any capable amp

so, a dedicated headphone amp with xlr in would be a capable choice for maximal headphone benefit using available smsl out options; can't really see how a preamp would help speakers given the smsl IS a pre-amp

hope that's food for thought

 

[tccalvin]

In theory, this box could satisfy all your needs for DACs, bluetooth receivers, headphone amplifiers and preamplifiers. You shouldn't need anything else.

In practice, it's a shame that this device likely doesn't measure as well as this review would lead you to believe because of the sound defects of the CS43131 chips and, unlike other companies like FiiO, SMSL doesn't seem interested in acknowledging/fixing the problem through firmware updates.

If it hadn't been for this annoyance, I would have easily chosen this guy over the DL200 since I like the features (HDMI ARC and full size XLR) more.

 

[scruffy1]

i'm not some major fanbois of cirrus - this is their first device i have used since i abandoned pci-e cards with them in (xonar hdav) long ago

the reported "cirrus hump" - how does that differ from the equally vilified "ess hump", and have subsequent releases based on the ess not ess-entially smoothed it to inconsequential?

is the measurement from amirm fabrication? or do you have current measurements to contest his findings ?

and perhaps more importantly, have you listened to one, or did you choose not to because you had read it was "defective" ? because the tone 1 despite the hump was a huge improvement on prior dacs i have listened to (admittedly mostly on sound cards, which is an unfair contest)

genuinely interested as to why this device is so "bad" as to my delicate cochleas it's up there with the topping e70 (ess, not velvet, which i am not purchasing) for transparency and quality of analogue output

thanks for any information, but it will still sound very acceptable capable to me regardless

 

[tccalvin]

the reported "cirrus hump" - how does that differ from the equally vilified "ess hump", and have subsequent releases based on the ess not ess-entially smoothed it to inconsequential?

It differs in the sense that you need a complex (music-like) multitone signal that varies in amplitude in order to trigger it, which is a test that is not present in Amir's reviews.

is the measurement from amirm fabrication? or do you have current measurements to contest his findings ?

Not at all. The measurements are right, they just don't tell the full story. Since I don't have measurements for this specific device, it's possible that the defects are not present in this one, but I doubt it because:

1) The issue seem related to the converter chip itself. In order to get rid of the unwanted distortion, you need to turn off the DRE (Dynamic Range Enhancement) function of the chip and the measured high SINAD numbers suggest this has not been done for this device.

2) I've owned both a DS100 and a M300SE and I've personally measured them and confirmed that they do indeed exhibit the "Cirrus hump", which leads me to believe that SMSL doesn't have any "secret sauce" at their disposal.

genuinely interested as to why this device is so "bad" as to my delicate cochleas it's up there with the topping e70

It's not that bad. It's just not as good as it seems.

thanks for any information, but it will still sound very acceptable capable to me regardless

Cirrus DACs sound just fine to me as well. I just determined that my 200€ was better spent elsewhere.

If it sounds good to your ears, then it is good. I would only recommend switching to something else if you're planning to use this DAC for professional workloads.

 

[scruffy1]

for $170aud as new, delivered, i am staying with it for now

however, the perceived extra "noise" in "song of the sisters" was with a single chip dac with an i-thing, and well over half way through that song i'm a little uncertain that the crackles might be misinterpreted hans zimmer intent, but i haven't heard anything crackle

and with all due respect to hans zimmer, i don't feel any compulsion to listen to the song again (i guess it's more appealing in a cinema as part of some grand spectacle, but as audio alone... meh! back to steely dan)