• Welcome to ASR. 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!

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

@somy It's the iPhone's hardware volume I meant. G3 supports it, that is you can change volume with the phone's hardware buttons and the iOS software slider. G2 always had 100% volume on the iPhone unless you used a player that has its own volume moderation like Neutron. The D version supports the iPhone's hardware volume, too. However, it draws so much current that you have to use an additional power bank. Therefore for on the go use not practical at all.
Thanks! Have you got 9038D already? I don't quite understand why D version needs more current than S version, I thought S version provides more power, no?
 
I received my 9038D ordered from hifigo a couple of days ago. I'm very pleased so far. It's been a complete plug & play affair on everything I've tried it on (2 of my PC's and my Samsung S10 phone). I don't normally listen to music seriously with headphones but working from home has made me want some music now and then and getting a decent stereo in my home office room hasn't seemed like an option for lack of space and cost reasons. I bought a Hifiman Sundara a few weeks ago and enjoyed that a lot but I haven't had any decent source in the office. I have primarily used a cheap usb dac+amp that I bought a many years ago because the sound output in my company laptop was unacceptably bad (new one I have now isn't much better). That usb interface is audibly noisy with sensitive headphones and underpowered with the Sundara.

In contrast the 9038D is completely noise free as far as I can tell, I have had no LTE noise issues with my phone on the same desk and in a quiet room the volume I use is somewhere between 40 and 50%, full volume with the Sundara is uncomfortably loud and this is with a -7dB pre-amp level in the EQ settings that I use. And on the topic of underpowered, my S10 can barely drive the Sundara with it's built in 3.5mm jack so if I use it with the phone (not really what I had planned but great option to have available) I will have a huge upgrade in available power. I'm not going to attempt describe the sound using colorful audiophile terms but it's definitely a big improvement from the albeit poor situation prior to it and I'm really enjoying listening to music with it so thanks for bringing it to market!
 
Last edited by a moderator:
XeCutor, I just a few minutes did react for the similar post in our discord channel, so I want to repost it here to save my time ;)
------------------------------------------------------------------------------------------------------------------------------------------
my DACs aren't flawless for mechanical respect or regarding USB3 compatibility as well but what I can guarantee the serious attention to measurable parameters. About 20pcs ES9038Q2M were replaced in the 9038D batch (500pcs) just because too noisy. I just wondered if somebody else goes that way? It is not about a production mistake, all 500pcs were functional, and I'm sure no one user would notice if SNR of his 9038D isn't so good, however, I have replaced 20pcs for the better one. BTW, about 9039S assembling issue, when I last time changed the silicone for 9038S/D assembling I asked Sunny to get the most expensive one silicone which is available on our market. Now the most expensive silicone seems to let me down, and some of 9038SG3 losing its caps.. All I tried to make the best and better again but the results are different ;)
WeChat Image_20201028160224.jpg
 
XeCutor, I just a few minutes did react for the similar post in our discord channel, so I want to repost it here to save my time ;)
------------------------------------------------------------------------------------------------------------------------------------------
my DACs aren't flawless for mechanical respect or regarding USB3 compatibility as well but what I can guarantee the serious attention to measurable parameters. About 20pcs ES9038Q2M were replaced in the 9038D batch (500pcs) just because too noisy. I just wondered if somebody else goes that way? It is not about a production mistake, all 500pcs were functional, and I'm sure no one user would notice if SNR of his 9038D isn't so good, however, I have replaced 20pcs for the better one. BTW, about 9039S assembling issue, when I last time changed the silicone for 9038S/D assembling I asked Sunny to get the most expensive one silicone which is available on our market. Now the most expensive silicone seems to let me down, and some of 9038SG3 losing its caps.. All I tried to make the best and better again but the results are different ;)
View attachment 89968
Amazing :D hats off to you sir
 
Veri, are you about ES9038? They are tricky people in ESS, did you see how looks their datasheet? They guaranty pretty much nothing at all and only mention "Typical" parameters. I don't say if Cirrus better but at least they guarantee some tolerance, awful one but it guaranteed.
2020-10-28_17-09-35.png
 
Veri, are you about ES9038? They are tricky people in ESS, did you see how looks their datasheet? They guaranty pretty much nothing at all and only mention "Typical" parameters. I don't say if Cirrus better but at least they guarantee some tolerance, awful one but it guaranteed.
It would definitely explain some of the thermal problems certain 9038PRO devices suffer from !
 
Veri, are you about ES9038? They are tricky people in ESS, did you see how looks their datasheet? They guaranty pretty much nothing at all and only mention "Typical" parameters. I don't say if Cirrus better but at least they guarantee some tolerance, awful one but it guaranteed.
View attachment 89977
I'm sure you can achieve better than on the datasheet. According to some person, someone has achieved 143dB DNR. And I have achieved 140dB DNR. THD+N is also way better than -122dB, at least -124dB and upto -130dB. I'm pretty sure the worst part will perform better than that in the correct circuit. If your circuit gives typical -122dB then you can't guarantee the performance to be -122dB. Also the datasheet is based on true APx555 measured performance not calculated actual number.
So yes some 9038q2m will be more noisy than others, some 9038pro won't be as good as others. But you can pretty much always achieve better than data sheet performance with proper circuit.
 
Screenshot_20201028-190557.jpg

This is from the datasheet of cs43198. If you think about it. 6dB is the difference of 4 chips/parts in parallel. It's pretty much meaningless. But I'm also with you that it's better to have better tolerances.
But that's why we measure every single unit on APx555. So...
 
JohnYang1997, BTW APx555 even noisier than my old SYS2522, and both these analyzers simply useless if you want to test SNR or DR of such chip as cs43198(550nVRMS(A)).
Still good enough to match datasheet for ESS. Also 43198 has much much higher noise when output higher than 20dB. Higher than 5uV at 0dB.
 
brute force is the sign of a modern China )) However, I only have the idea so far, let see if anything will come out.
Sometimes brute force is what it takes, am I right?
There really isn't much to a dac. Clock, I/V(or output stage), power supply/reference and pcb routing. 43198 is a really weird/unique dac it's almost immune to anything. There's also a hidden feature of 43198, see if you can find out.
 
one thing I wish to find in cs43198/43131 is the THD compensation like 9038 has. To use our Tweak app with cs43198/43131 too ;) Without that feature, all DACs look boring to me.
We got around -140dB harmonics with 43198.
 
  • Like
Reactions: PSO
Back
Top Bottom