Review and Measurements of E1DA 9038S BAL Portable DAC & Amp

[wiz2596]

Waiting for 3.5mm version

 

[JustAnandaDourEyedDude]

Recently received my #9038S G3, and it is superfine. Ran a test for upwards of an hour, using FLAC files on my Win10 PC ripped from CDs. Compared sound of #9038S with sound of Topping D90 + Monolith THX AAA 887. Used Final A8000 earphones with a 4.4mm Pentaconn balanced cable. Used good adapters for 4.4mm on each of the two amps' outputs (XLR and 2.5mm). Would pause and replay the track, while switching A8000 cable plug from one adapter to the other, and using Win10 audio control dropdown to select between Topping and E1DA devices. Used world jazz female vocal tracks primarily, along with some solo piano (well-recorded Bourree I & II and Gigue from Bach English Suite No. 2). Used a 15 second section of a favorite track of female jazz vocal to try and match volume level between the two systems. First set #9038S volume to optimal level for the song. Then adjusted Monolith volume to match (close to noon) (D90 always at -15dB). Sighted test with old ears.

To my ears, both systems sounded the same, very beautiful, more beautiful and detailed and real-sounding than with any other systems or DAPs I have used. I could adjust the THX 887's volume very slightly up or down to make either system sound a little clearer and more detailed than the other. Based on this, I think it unlikely that I could tell them apart in a double blind test. Perhaps younger keener ears could tell them apart.

I think the #9038D will be the ultimate for most music lovers who either stream music primarily through their mobile phone or carry their music collection on their mobile phone. IVX is too modest.

 

[IVX]

Guys, you should thank ES9038 designer(but not his moronic managers!), I just accurately delivered the DACs quality up to your headphones

 

[IVX]

in case you know who designed it

Wiki: In 2001 ESS acquired a small Kelowna design company (SAS) run by Martin Mallinson and continues R&D operations in Kelowna. The Kelowna R&D Center developed the Sabre range of DAC and ADC products that are used in many audio systems and cell phones.
https://www.youtube.com/user/MartinMallinson
Let's make kind flashmob and subscribe to his little youtube channel Sofar there just 19 subscribers!

 

[maxxevv]

Wiki: In 2001 ESS acquired a small Kelowna design company (SAS) run by Martin Mallinson and continues R&D operations in Kelowna. The Kelowna R&D Center developed the Sabre range of DAC and ADC products that are used in many audio systems and cell phones.
https://www.youtube.com/user/MartinMallinson
Let's make kind flashmob and subscribe to his little youtube channel Sofar there just 19 subscribers!

I have been to Kelowna, BC some 20 years ago. Gorgeous greenery !

 

[minimalist]

Wiki: In 2001 ESS acquired a small Kelowna design company (SAS) run by Martin Mallinson and continues R&D operations in Kelowna. The Kelowna R&D Center developed the Sabre range of DAC and ADC products that are used in many audio systems and cell phones.
https://www.youtube.com/user/MartinMallinson
Let's make kind flashmob and subscribe to his little youtube channel Sofar there just 19 subscribers!

Two more quite entertaining talks by Martin Mallison, revealing some details of Sabre DAC implementation:

 

[oliveoilthansnakeoil]

I just saw 'susumu 3000 edition' on aliexpress. Does it have some 'audible' differences from normal 9038S?

 

[IVX]

oliveoilthansnakeoil
I just an engineer, let wait for real user's feedback.

 

[IVX]

I want to ask you guys regarding 9038D, the single-ended 3.5mm jack version of 9038S. That model is ready to be released for a few months already, however, I still can't improve LTE immunity of that DAC mostly because of I/V stage and unbalanced topology. At the same time, I've got an impressive distortions figure with 9038D. For example, my AP SYS-2522 reads THD+N 0.000012..0.000014%(SINAD 117..118db) i.e. 1.4PPM maximum at no load, and 1.7PPM 32ohm loaded at 260mW. So, today I need to decide, should I release that DAC as is or I have to keep improving LTE buzzing immunity up to the end of time this universe? Please let me know your opinion about that.
LTE aka 4G/5G(3G much "softer") mobile internet uses a very wide frequency band and quite high power for RX/TX.
What does mean? In case if you gonna use 9038D with high-sensitive IEMs(>105..110db/mW why such stuff needs any HPA at all??) and play Tidal online via mobile interface(especially if you place 9038D at the backside of your phone!) you may notice some "static" buzzing. The internet via WiFi or LTE but with normal headphones(dynamic or planars, open-back or closed whatever, IEMs 90..100db/mW) no problem at all.

 

[Veri]

I want to ask you guys regarding 9038D, the single-ended 3.5mm jack version of 9038S. That model is ready to be released for a few months already, however, I still can't improve LTE immunity of that DAC mostly because of I/V stage and unbalanced topology. At the same time, I've got an impressive distortions figure with 9038D. For example, my AP SYS-2522 reads THD+N 0.000012..0.000014%(SINAD 117..118db) i.e. 1.4PPM maximum at no load, and 1.7PPM 32ohm loaded at 260mW. So, today I need to decide, should I release that DAC as is or I have to keep improving LTE buzzing immunity up to the end of time this universe? Please let me know your opinion about that.

How long you already worked on LTE buzzing problem? If you searched already long time for a solution, maybe it is not feasible in this small size? Perhaps just state a warning about it, that customers know? I think a lot of people are still waiting for 9038D and are eager to try it

 

[Tks]

I want to ask you guys regarding 9038D, the single-ended 3.5mm jack version of 9038S. That model is ready to be released for a few months already, however, I still can't improve LTE immunity of that DAC mostly because of I/V stage and unbalanced topology. At the same time, I've got an impressive distortions figure with 9038D. For example, my AP SYS-2522 reads THD+N 0.000012..0.000014%(SINAD 117..118db) i.e. 1.4PPM maximum at no load, and 1.7PPM 32ohm loaded at 260mW. So, today I need to decide, should I release that DAC as is or I have to keep improving LTE buzzing immunity up to the end of time this universe? Please let me know your opinion about that.
LTE aka 4G/5G(3G much "softer") mobile internet uses a very wide frequency band and quite high power for RX/TX.
What does mean? In case if you gonna use 9038D with high-sensitive IEMs(>105..110db/mW why such stuff needs any HPA at all??) and play Tidal online via mobile interface(especially if you place 9038D at the backside of your phone!) you may notice some "static" buzzing. The internet via WiFi or LTE but with normal headphones(dynamic or planars, open-back or closed whatever, IEMs 90..100db/mW) no problem at all.

Depends.. What exactly can you change that would get rid of the buzzing? Would you have to lower overall max-output power? If so, I think that's a good trade-off if it maintains that insane SINAD. I can listen to an iPhone doggle and still get my ears blasted to pieces, not really sure what sort of headphones people use that are saying volume is too low (I honestly would wager these sorts of people have hearing damage). 3G is being phased out, so I would avoid designs relying and hoping people use that (also phones don't really have toggles for this stuff all the time anyway). If you can shield it from AT LEAST 5Ghz/2.4Ghz WiFi, then I think you should release now. But if WiFi is also going to be a problem, I'd do whatever it takes to at least solve that issue.

Basically put, as long as the device performance is even 1-2dB SINAD better than the E1DA first measured by Amir all those months ago - you will still have a killer product that others can't match anyway. AND you're showing progress by providing better performance than the first device you released. I think that's what many normal and fair people care about. There's no need to chase 118db SINAD, only to suffer from horrible LTE buzzing and things like that. Don't drop the ball on an important aspect of usability, for the sake of spec-sheet boasting. Your ED1A products already blast others to pieces, no need to blast yourself by pushing harder than you need to at the cost of something else.

IF by some chance you can't immunize against this LTE or WiFi at all no matter what you do. Then go balanced, and build the device so people can use a small adapter so they can turn the device into unbalanced without having to do wierd grounding mods or whatever. I don't think you even need to make an unbalanced version of the E1DA if only unbalanced conversion were simple for example.

But at the end of the day, you're smarter than 99% of us here. You know best the technical limits of what you can do. We can only speculate from warm chairs..

 

[maxxevv]

I want to ask you guys regarding 9038D, the single-ended 3.5mm jack version of 9038S. That model is ready to be released for a few months already, however, I still can't improve LTE immunity of that DAC mostly because of I/V stage and unbalanced topology. At the same time, I've got an impressive distortions figure with 9038D. For example, my AP SYS-2522 reads THD+N 0.000012..0.000014%(SINAD 117..118db) i.e. 1.4PPM maximum at no load, and 1.7PPM 32ohm loaded at 260mW. So, today I need to decide, should I release that DAC as is or I have to keep improving LTE buzzing immunity up to the end of time this universe? Please let me know your opinion about that.
LTE aka 4G/5G(3G much "softer") mobile internet uses a very wide frequency band and quite high power for RX/TX.
What does mean? In case if you gonna use 9038D with high-sensitive IEMs(>105..110db/mW why such stuff needs any HPA at all??) and play Tidal online via mobile interface(especially if you place 9038D at the backside of your phone!) you may notice some "static" buzzing. The internet via WiFi or LTE but with normal headphones(dynamic or planars, open-back or closed whatever, IEMs 90..100db/mW) no problem at all.

Is it from the wires or from the circuit board?
If its from the board, perhaps can look at shielding in the form of high frequency shield meshes. The are pretty inexpensive in raw cloth mesh form.

Some use it to shield smart cards and RFID chips from unauthorized tapping. Perhaps you can explore that as they are known to be able to blank out phone signals.

 

[rubendub]

How long you already worked on LTE buzzing problem? If you searched already long time for a solution, maybe it is not feasible in this small size? Perhaps just state a warning about it, that customers know? I think a lot of people are still waiting for 9038D and are eager to try it

I agree completely

 

[staticV3]

EMI on the 9038D is still greatly improved compared to the 9038S G1 and G2, just not quite on the level of the G3.

 

[Veri]

EMI on the 9038D is still greatly improved compared to the 9038S G1 and G2, just not quite on the level of the G3.

It's a no-brainer then. Release it! E1DA 9038D (v1)

 

[mnemonix]

Two more quite entertaining talks by Martin Mallison, revealing some details of Sabre DAC implementation:

Fascinating, and some really interesting stuff from about 22 minutes in about some people being able to hear things that are unmeasurable and should be inaudible in the performance of different DACS. I'd be interested to know whether anything he states here challenges any of the thinking on ASR. He aludes to but doesn't explicitly state the use of blind testing and appears to be giving scientific explanations for what he describes but I'm not technically qualified enough to know for sure.

 

[Jimbob54]

I want to ask you guys regarding 9038D, the single-ended 3.5mm jack version of 9038S. That model is ready to be released for a few months already, however, I still can't improve LTE immunity of that DAC mostly because of I/V stage and unbalanced topology. At the same time, I've got an impressive distortions figure with 9038D. For example, my AP SYS-2522 reads THD+N 0.000012..0.000014%(SINAD 117..118db) i.e. 1.4PPM maximum at no load, and 1.7PPM 32ohm loaded at 260mW. So, today I need to decide, should I release that DAC as is or I have to keep improving LTE buzzing immunity up to the end of time this universe? Please let me know your opinion about that.
LTE aka 4G/5G(3G much "softer") mobile internet uses a very wide frequency band and quite high power for RX/TX.
What does mean? In case if you gonna use 9038D with high-sensitive IEMs(>105..110db/mW why such stuff needs any HPA at all??) and play Tidal online via mobile interface(especially if you place 9038D at the backside of your phone!) you may notice some "static" buzzing. The internet via WiFi or LTE but with normal headphones(dynamic or planars, open-back or closed whatever, IEMs 90..100db/mW) no problem at all.

I'd mostly use via wifi anyway- release it- and be clear what return policy is if you get EMI big problems.

 

[Cat Music]

I want to ask you guys regarding 9038D, the single-ended 3.5mm jack version of 9038S. That model is ready to be released for a few months already, however, I still can't improve LTE immunity of that DAC mostly because of I/V stage and unbalanced topology. At the same time, I've got an impressive distortions figure with 9038D. For example, my AP SYS-2522 reads THD+N 0.000012..0.000014%(SINAD 117..118db) i.e. 1.4PPM maximum at no load, and 1.7PPM 32ohm loaded at 260mW. So, today I need to decide, should I release that DAC as is or I have to keep improving LTE buzzing immunity up to the end of time this universe? Please let me know your opinion about that.
LTE aka 4G/5G(3G much "softer") mobile internet uses a very wide frequency band and quite high power for RX/TX.
What does mean? In case if you gonna use 9038D with high-sensitive IEMs(>105..110db/mW why such stuff needs any HPA at all??) and play Tidal online via mobile interface(especially if you place 9038D at the backside of your phone!) you may notice some "static" buzzing. The internet via WiFi or LTE but with normal headphones(dynamic or planars, open-back or closed whatever, IEMs 90..100db/mW) no problem at all.

I just want to make sure the 9038D doesn't have a problem with the IMR R2 Aten regarding buzz. it is right?

 

[IVX]

Cat Music
depends on your individual sensitivity as well.
Jimbob54
I remember a ridiculous case when we returned money for 9038SG3 to Korea because the user didn't find the way to control the volume on his Linux computer. He did start Aliexpress dispute and Aliexpress officer decided that it is my problem. Another user from Europe told me that he did solve Linux volume control "problem" by manual editing some sys-config file.

 

[Deleted member 18627]

oliveoilthansnakeoil
I just an engineer, let wait for real user's feedback.

I am really looking forward to hearing the difference. Although truth be told I am shocked how awesome the G3 is. I am using it as my main DAC just now in front of my SP200. I sold my iFi Zen Dac and Topping D50S as frankly I do not need them and the G3 is so compact I have velcro'd it and the SP200 to the bottom of my desk to save space.

I am am long time lurker on these forums but just joined to say thanks for this awesome product.

From an engineering point of view, would you expect the changes on the susumu version to become more apparent under load or when used as balanced DAC?

I am going to use one for the desk setup and the other for my lazing around and walking setup and would be interested to hear your thoughts and expectations.