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

[IVX]

I did update my wife's iphone to ios17 to investigate the problem.
PS: And yes, after iphone was updated to ios17 it stopped working with 9038S/D
however, Comtrue gave me the latest FW code, and I see that ios17 works with that.
So the solution is found but I need some time to adapt my custom functions to that new FW source.

 

[IVX]

The descriptors are very different, I keep working on them.

 

[IVX]

I want to offer ios17 ready FW with no HID function, there is no Tweak9038 app support. As I understand, ios17 users don't care because the app is only for Android. Also I switched off Standby, mute, and PTL functions.

 

[IVX]

During a long time R&D work around Cosmos DAC project I suddenly, and truly by the way, got 9039S proto with pretty phenomenal specs, especially loaded THD+N. As you can see this is pretty much the same 9038S but ES9039Q2M based on + composite opamps(opa1612->ad8397) output. The Dynamic Range is 130db(A), THD+N@1kHz -125db 0dbfs if no load, and -123db 32ohm 0dfs(3.45Vrms), -120db at 16ohm -.5dbfs(>660mW). The THD was optimized for the no-load condition, and THD+N -120db at 16ohm is clearly due to composite opamp giant loop gain.
I don't think if such a monster would be really popular among simple users, they would be happy with CS431** DACs $10-50 with no extra current draw and no POP-noise.. However, I can order a small quantity of 9039S for HP enthusiasts. Let me pls know your opinion.

 

[MC_RME]

Well, I for one have no clue what you talk about.

 

[ZolaIII]

@IVX your R&D makes a difference but please don't develop next gen E1D2 stereo DAC. Try making a proper multichannel external unbalanced one instead as there is a hole regarding specific product offerings on the market. There are Motu interfaces and Creative internal sound cards based on (older) ESS DAC's but I am confident you can do better and for less money. Anyway the CS43131 will continue to be dominant over Q2M offering (simply better OPAMP, less power hungry, really cheap and still leading EMI rejection rate). At least I think that would work better, best regards and have a nice time.

 

[Jimbob54]

During a long time R&D work around Cosmos DAC project I suddenly, and truly by the way, got 9039S proto with pretty phenomenal specs, especially loaded THD+N. As you can see this is pretty much the same 9038S but ES9039Q2M based on + composite opamps(opa1612->ad8397) output. The Dynamic Range is 130db(A), THD+N@1kHz -125db 0dbfs if no load, and -123db 32ohm 0dfs(3.45Vrms), -120db at 16ohm -.5dbfs(>660mW). The THD was optimized for the no-load condition, and THD+N -120db at 16ohm is clearly due to composite opamp giant loop gain.
I don't think if such a monster would be really popular among simple users, they would be happy with CS431** DACs $10-50 with no extra current draw and no POP-noise.. However, I can order a small quantity of 9039S for HP enthusiasts. Let me pls know your opinion.
View attachment 336216
View attachment 336217

View attachment 336218

If you make one- I will buy it. Love me some SotA dongle.

 

[staticV3]

I can order a small quantity of 9039S for HP enthusiasts. Let me pls know your opinion.

I want one

It'd be the ultimate pocket monster for inefficient planars on the go.

 

[JeffJ]

More power, please!

 

[IVX]

One more interesting point regarding ES9039Q2M, I found an undocumented "NOS mode" there. 12kHz and 1kHz look like that:

 

[IVX]

SMPTE 60/7000 4:1 AP shows -116db as the best point, REW <-120db, pls note this is the sum of ADC+DAC IMD.

and another channel:

 

[IVX]

High-frequency performance is also spectacular. Actually, I refused the idea of using 9039Pro in Cosmos DAC due to poor 10kHz performance, I didn't see the H3 < -110db..
I've no idea why 9039Q2M version easily gave me -135db for 1/10 of the 9039Pro cost but I like it

 

[IVX]

Regarding the jitter, 9039Q2M operates in the sync mode which helps to reduce low-frequency side-bands. IMO, 12k tone looks pretty much ideal

 

[CaféNoir]

This is pushing the limits! I would buy that.

 

[IVX]

BTW, the IMD plot above, yellow one, tends to look like a hump at -42db, however, the red plot has no such thing. The reason is different capacitors for the modulator's HF content wipe-out.

 

[IVX]

ESS holds a monopoly on portable high-performance DACs today. If AKM has a comparable top 4499 DAC in their portfolio, the portable DAC market has simply nothing against Q2Ms, whatever 9038 or 9039. CS431** looks too cheap with all their tricks to deceive Dynamic Range test, and truly fake 192-384k sampling support. Rohm? I don't think so. Who else? I think we'll see some new names in that market soon, or just R.I.P. audio DACs race.

Is the CS43198 a same fake?

Hi! Everybody knows Cirrus Logic DACs CS43131 and CS43198. Once I noticed a discrepancy between the sound of my CS43131 and its measurements. After some research, I came to the conclusion that its brilliant results are fake. I have not any source with CS43198, so it would be interesting if...

www.diyaudio.com

 

[Makahl]

Looking good! I'd be curious to see a poll about whether users prefer 2.5mm or 4.4mm, as 2.5mm seems rarer in 2023 products such as dongles and daps. But I'd imagine that would add another layer of complexity to the current design/case.

 

[Dolor]

During a long time R&D work around Cosmos DAC project I suddenly, and truly by the way, got 9039S proto with pretty phenomenal specs, especially loaded THD+N. As you can see this is pretty much the same 9038S but ES9039Q2M based on + composite opamps(opa1612->ad8397) output. The Dynamic Range is 130db(A), THD+N@1kHz -125db 0dbfs if no load, and -123db 32ohm 0dfs(3.45Vrms), -120db at 16ohm -.5dbfs(>660mW). The THD was optimized for the no-load condition, and THD+N -120db at 16ohm is clearly due to composite opamp giant loop gain.
I don't think if such a monster would be really popular among simple users, they would be happy with CS431** DACs $10-50 with no extra current draw and no POP-noise.. However, I can order a small quantity of 9039S for HP enthusiasts. Let me pls know your opinion.
View attachment 336216
View attachment 336217

View attachment 336218

I would like to have one too. :0)

 

[Heinzfi]

Looking good! I'd be curious to see a poll about whether users prefer 2.5mm or 4.4mm, as 2.5mm seems rarer in 2023 products such as dongles and daps. But I'd imagine that would add another layer of complexity to the current design/case.

I'd vote for 4.4mm, especially if the intention is for headphone users. My HD 660S2 comes with a 4.4mm cable and using it with an adapter wouldn't be compelling for me.

 

[Dolor]

2,5mm is smaller and lighter.