Fosi Audio DS2 Portable DAC & Amp Review

[aka]

You’ll need an OTG cable. Here are two that I have that will work for your needs:

https://www.amazon.com/gp/aw/d/B09KLXYQDH?psc=1&ref=ppx_pop_mob_b_asin_title

Amazon.com: SHANLING L3 USB-C to L OTG DAC Adapter Cable, for iPhone iPad iOS Devices, for USB Headphone Amps Amplifiers UA2, UA3, UA5, UP5, H7, 4.72 Inch Pure Silver Cable : Electronics

Amazon.com: SHANLING L3 USB-C to L OTG DAC Adapter Cable, for iPhone iPad iOS Devices, for USB Headphone Amps Amplifiers UA2, UA3, UA5, UP5, H7, 4.72 Inch Pure Silver Cable : Electronics

www.amazon.com

Thank you both, @PetZoundz and @stefanovesa !

 

[dananski]

And I just found this now comes with a much nicer, soft 15cm usb-c cable. (That's 15cm including connectors at the end; 10cm of flexible bit.)

 

[BorisG]

Hi – is this still recommended? Or are there better options available today?

 

[Viper780]

It is 1 year old and there aren't any new improvements.
It is a very generic dual CS43131 implementation with extra Output Buffers. There are several similar portable DAC + AMP in the same 50-60USD price range.

If you get it at a good price point, don't need PEQ, have headphones that require bit more power (specific with low impedance) and want balanced headphone out than I'll still would buy it.

If not there are cheaper, smaller or more versatile options out there

 

[BorisG]

It is 1 year old and there aren't any new improvements.
It is a very generic dual CS43131 implementation with extra Output Buffers. There are several similar portable DAC + AMP in the same 50-60USD price range.

If you get it at a good price point, don't need PEQ, have headphones that require bit more power (specific with low impedance) and want balanced headphone out than I'll still would buy it.

If not there are cheaper, smaller or more versatile options out there

Thank you! I had a good price on it with same-day delivery (handy just before a long trip starting tomorrow), so went for it.

 

[db602]

Fade in effect *FIX/SOLUTION*

Ive had this dac/amp for about a year now and I really like it except for the fade in effect it has. I use it on my windows laptop and it always has this fade in where sometimes it can last up to 3 seconds to fully start the sound, and also it has that panned fade in where one side starts before the other side. Recently for about two months I stopped using it cuz of this fade in effect and just switched to the Shanling UA mini for my IEMs which has a very brief fade in, and the apple dongle for my speakers that has no fade in.

Just today I searched up which dac/amps dont have fade in effects in googles AI and it brought up a program called "sound keeper" for windows. I installed it and set it as a startup program aswell. I tested it with my drum kit samples since most of them are less than a second long and theres no fade in anymore. If anyone wants to fix the fade in issue just google sound keeper and download/install it. You just run the x64 file and it runs in the background, you can disable it in task manager if you want.

 

[danoond]

I recently acquired Fosi DS2 and now facing a problem with sound playback in Windows 11. The sound drops out occasionally when watching videos, listening to music, playing videogames etc. It recovers automatically within about a second with an annoying fade-in effect. I still haven't found a proper solution to this problem, but I have a few observations:
1) The lowest stability is observed when the device's playback quality is set to 32bit/384kHz in Windows 11 audio settings. Setting it to 24bit/192kHz eliminates virtually all dropouts, clicks, and other audio artifacts (using the standard Microsoft driver). I haven't yet tried other settings, but the difference in audio quality is very noticeable, no clicks or dropouts.
2) Using the driver from the Fosi's website (Savi Bravo-HD), stability is worse than using the standard Microsoft driver.
3) I tried the Sound Keeper utility recommended by @db602 above. This eliminated the fade-in effect when starting audio playback (you can test this in Windows settings in System>Sound>Properties>Output settings>Format>Test, w/o Sound Keeper both left and right channels have fade-in when test sound is played), but audio dropouts still persist at 32bit/384kHz.
4) I tried setting the power plan to maximum performance, disabling USB power management, and disabling unused audio devices – dropouts still persist.
5) Tried different USB Type-C cables and connected the amplifier to the motherboard port – with the quality set to 32-bit/384kHz, the audio still drops out periodically.
After reading this thread, it seems to me that this issue affects a large number of DS2 users on Windows 11. Has anyone found a solution or can offer advice regarding audio dropouts during playback? I would like to be able to use 32bit/384khz for DS2 because I work in DAWs from time to time.

 

[DSJR]

Why do you want 32/384? I mean, our adult hearing is barely out to 15khz for most people under forty and our 'dynamic range' is barely 50dB, if on a sliding scale to a degree...

Just stick to 24/96, which is still way better than we need i suspect

What resolution do other pros use when mixing or mastering?

 

[danoond]

Why do you want 32/384? I mean, our adult hearing is barely out to 15khz for most people under forty and our 'dynamic range' is barely 50dB, if on a sliding scale to a degree...

Just stick to 24/96, which is still way better than we need i suspect

What resolution do other pros use when mixing or mastering?

Thanks for your reply!

I don't really know if I *need* 32/384, it's more like "i paid for the whole thing, i want to use the whole thing" kinda stuff As for the DAWs, I haven't found any clear recommendations about bit depth and sampling frequency. The consensus as far as I understand is 24 bit 48-96 kHz for the same reasons you've mentioned, or to match rate and depth of the source material and/or DAW internal settings.

I just can't quite understand why DS2 doesn't want to work properly at 32/384. Had no issues with my friend's KA17 and SK02 on my machine. Integrated ALC4082 also seems to work fine at 32/192. I ran LatencyMon, all bars are green, no high latency spotted, system seems to be suitable...

I also have Nahimic Audio installed, I've disabled and deleted all audio optimization settings for connected devices, but maybe it's background processes are affecting the sound quality of DS2. I'll try uninstalling it completely, but because KA17 and SK02 were working fine with Nahimic installed I don't think it will change much...

 

[Labjr]

Why do you want 32/384? I mean, our adult hearing is barely out to 15khz for most people under forty and our 'dynamic range' is barely 50dB, if on a sliding scale to a degree...

Just stick to 24/96, which is still way better than we need i suspect

What resolution do other pros use when mixing or mastering?

Maybe he has a pack of dogs?

 

[danoond]

A small update on my Fosi DS2 in Windows 11

Yesterday I've been using DS2 all day playing vidogames and watching YouTube videos/listening to music on Spotify app. Playback quality was set to 24/192, the standard Microsoft driver was used, DS2 was connected to the USB-C port on the back of my motherboard (X870 Steel Legend) using Miniware 240W Type-C to Type-C cable. Haven't had any dropouts or clicking noises on my Zero:2 IEMs. USB-C cable that was included with my DS2 (newer one, same as in @dananski post) also worked perfectly fine. I've reenabled all other sound devices in Windows, reinstalled Nahimic, in other words I returned the system to the state it was in when I first connected the DS2 to the PC. No dropouts or clicks.

I tried setting 32/384 again - dropouts every 5 to 15 minutes, and the sound quality is somehow worse: bass became less massive and clear and the sibilants became much louder and began to grate on the ear. Then I changed the settings back to 24/192 and connected the DS2 to the USB-С port on the front panel - dropouts came back! So it looks like both the USB-C port the DS2 is plugged in and playback quality settings in Windows may affect the stability of the device...

Another interesting thing I've observed is that I was having dropouts almost every time my 8BitDo Ultimate 2.4 gamepad was connected or disconnected from the PC and the Steam was running. When I closed Steam completeley and manually turned on/off the controller 20 times or so there were no dropouts whatsoever! The gamepad is connected to the PC wirelessly using the included 2.4G USB dongle. All system sounds and notifications are enabled. I've set "Do nothing" when Windows detect communications activity in the communications tab of the sound settings in Windows and rebooted the PC as was recommended in Steam Discussions, but that didn't solve the issue.

I'll try to experiment some more and will post here if I find any other solutions. But at least now I can use DS2 in Win11 without any dropouts. The fade-in or ramp-up effect is still present the beginning of the sound playback, but because I have something playing in my IEMs almost all the time this isn't a problem for me. Hope this info will be helpful

 

[Rise of Seren]

@danoond , try increasing the priority to high with something like https://github.com/spddl/GoInterruptPolicy . It seems like without it set to high, other things like high network/IO activity can take priority instead and cause issues.

 

[kandamrgam]

According to the measurements made by RAA, the DS2 has no "Cirrus hump" by the definition given by my review. However, I strongly suspect that it must exhibit the distortion at the initial rising part of the test signal shown in Part II of my review, just like the JCally JM20 or JM20 Max.

Why do you suspect so? This is what Fosi told me when asked about DRE setting:

The DRE register is hidden and disabled by default. The DS2 software does not enable this feature.

Can hump happen despite DRE being disabled? On an unrelated note, I wonder how DS2 achieved those stellar measurements figures in Amir's review without DRE enabled.

 

[jkim]

Why do you suspect so? This is what Fosi told me when asked about DRE setting:

The DRE register is hidden and disabled by default. The DS2 software does not enable this feature.

Can hump happen despite DRE being disabled? On an unrelated note, I wonder how DS2 achieved those stellar measurements figures in Amir's review without DRE enabled.

I am sure the Fosi Audio representative did not give correct information. The statement should've been:

The DRE register is hidden and enabled by default. The DS2 firmware does not disable this feature.

Most critically, the Fosi Audio DS2 with CS43131 cannot achieve that high, measured dynamic range without DRE enabled. Simply impossible as we know what CS43131 is capable of with no DRE.

 

[fws]

If the question was about "DRE setting" the answer might refer not to how DRE is set, but availability of "the setting" in the UI, so for user to switch DRE. Like the switch in FiiO's app.
So perhaps it would be "The DRE register switch is hidden and it's visibility is disabled by default. The DS2 software does not enable DRE switch feature"

 

[SPD_RCR]

So, I am really enjoying this Fosi DS2 on my MacBook Pro (old Intel i9) via Qobuz and Radio Paradise. Am I missing anything that is available at twice the price? I've got a Schiit tube setup on another computer and wonder if the Apos X Geshelli Merlin or some Topping/Fosi (DX1 or ZH3) or some other desktop setup is worth trying. How do "we" stop dabbling in components and simply enjoy "our" music? The Schiit and DS2 are both really good and cost-effective.

 

[danoond]

Final update on Fosi DS2 playback in Windows 11.
Now finally everything is working perfectly fine, 32bit/384khz without any dropouts, random fade-ins in the middle of the game or song, clicking or noise or any other problems.

TL;DR: ASMedia SATA3 driver wasn't installed, 30k ACPI.sys DPC calls in 30 seconds. Installed the driver - no dropouts on any settings.

Since my last post I've tried many different configuration changes on my PC: disabled USB host controllers, reinstalled Bravo-HD drivers, disabled Ethernet/Bluetooth/Wi-Fi in BIOS, deleted all unused devices with drivers in Device Manager, cleared my system, edited power plan settings (even hidden ones using PowerSettingsExplorer), debloated Win11 and sooo many more. Nothing helped me achieve uninterrupted playback on DS2.

Yesterday I decided to run LatencyMon again and inspect every tab and every piece of information this beautiful software provides. And what I found was ACPI.sys was generating ~30k DPC calls in 30 seconds. Keep in mind that all of this was happening on a pretty modern PC (Ryzen 7700, 32Gb CL30 6000MHz RAM, X870 motherboard, almost empty 990 Pro SSD, 80 Plus Platinum PSU plugged in a pretty good voltage stabilizer and so on) with a fresh Win11 install.

Immediately after this I've completely uninstalled chipset and GPU drivers using Revo Uninstaller + DDU + AMD Cleanup Utility under safe mode and installed newest stable drivers from AMD website. But to my surprise ACPI.sys was still making too many DPC calls... Then I decided to install/update all motherboard drivers from ASRock (my MB's manufacturer) website. I've installed every driver available except SATA driver since I don't have any SATA devices. Nothing changed, ACPI.sys was flooding my PC with DPCs...

After another 4 hours of searching, I returned to drivers and decided to install SATA drivers. The only driver available was "[Beta] ASMedia SATA3 driver ver:3.2.3.0000". Without rebooting PC after installation I immediately launched LatencyMon. A SINGLE DPC CALL FROM ACPI.SYS when I ran LatencyMon for 30 minutes...

Today I have Bravo-HD driver installed, 32bit/384khz set in Windows for DS2 and I haven't had a single dropout or click for the whole day. I still have that fade-in effect when something starts playing in my IEMs after the system was idling for some time, but I think it may now be solved by using Sound Keeper v1.3.5. At least now I have it running for 15 or so minutes while I'm writing this post and listening to music on Spotify and I had zero dropouts or fade-ins.

Sorry for these huge posts I'm writing here, but I really wanted to document my attempts to solve playback problems with DS2 on Win11. I hope that someone who has similar problems with sound playback on DS2 in Windows will find a solution to their problem in the steps that I described in my messages

 

[perceptionist]

A big thank you to Amir for the review.

The DS2 is one of our products as we make our entry into the high-performance DAC market. The DAC market is filled with many outstanding companies with which everyone is familiar. We will humbly learn from these companies as we continue to grow, and we hope to follow in their footsteps by bringing Fosi Audio’s unique touch to DACs for audiophiles worldwide. In addition to portable DACs, we’ll soon be introducing a desktop DAC as well.

To show our appreciation for the support from ASR members, we’re offering a special discount code for the DS2. You can use this code to get 20% off your DS2 purchase!
ASRDS220

link:
https://fosiaudio.com/products/fosi-audio-ds2-2024-dac-headphone-amplifier
https://www.amazon.com/dp/B0CTHN2QB3
https://www.amazon.de/dp/B0BJ66KJ2B
https://www.amazon.com.au/dp/B0BJ66KJ2B
https://www.ebay.com/itm/296311456320

https://m.facebook.com/groups/fosiaudio/permalink/1307015754223754/?

Rumor says DS3 is incoming with these specs. Will it be available this year?

Model: DS3
• DAC Chip: ESS ES9039Q2M
• USB Audio Interface: XMOS Beyond Power
• Max Sampling Rate: PCM 32bit/768kHz, DSD512
• USB Audio Class: UAC1.0 / UAC2.0 dual mode
• Outputs: 3.5mm single-ended, 4.4mm balanced
• Output Level: 2Vrms (SE), 4Vrms (BAL)

 

[OK1]

Stereo Balance Issue - This is the main issue I have had.

Spoiler: Stereo Balance Issue - Details

The only issue I have faced has been the need to occasionally detach and reattach the USB C to USB C cable, which I use to connect to my Windows computer, to resolve a strange imbalance in the stereo field, that appears to creep in over time, e.g after several hours of use. I am basing this observation on my own hearing/perception, not having measured the output with any tools. I have no clue if this is a fault with my own specimen of this device, or a general fault with the dongle model, cos I own only one example of it. It may be also an imbalance in my listening devices, or my own hearing. In truth I suspect that I do have a hearing imbalance, cos this seems to be consistent across IEMs and headphones, but I have not had my ears tested, to fully confirm if its just a volume balance or a frequency response difference in my hearing between left and right, or a combination of both aspects of hearing anomalies - frequency response and volume. It could be that my IEM - the Artti T10 also has a stereo imbalance, so really hard unless one measures, to know what's causing the imbalance in stereo - my ears, IEM, or DAC/Headphone dongle. or a bit from all of them !!

In software, I adjust the balance of volume to compensate for the imbalance I am hearing. I recalibrate this about once or twice a day, cos sometimes the insertion depth of the Artti T10 IEM, seems to alter this balance, and the T10 has some kind of dual port that "pops" occasionally, also changing the stereo balance possibly.

I wonder if the dual DAC CS43131 in the Sonata BHD is a possible cause of the stereo imbalance! who knows. One of these days, I'll spare some cash to invest in a single CS43131 DAC, and see. The audible audo quality should be identical, between a single and dual CS43131 DAC, even though I can expect that the dual CS43131 based DAC will measure better

It transpires that the stereo channel imbalance I was hearing in the TempoTec Sonata BHD, was more likely caused by one of these :

1. My own hearing, which may not be 100% identical between the two ears. In my opinion the close proximity of listening through devices like IEMs, which are so close to the eardrum, may highlight any differences even further. When we do eye tests for spectacles, it is not a bid deal to have each eye having some difference from the other, when sight is measured. There is no reason why we should not expect a similar deviation in hearing. This could be at the ear drum, or due to a difference in the dimensions of each ear canal.

2. Cable - i.e poorly made with poor Quality Control.

3. The transducers themselves(headphones in my case), also due to flaws in materials or during manufacture.

4. User error, inserting one earpiece slightly deeper than another.

 

[OK1]

If there be any other concern, it would be the stability of the provided ASIO driver, which I no longer use, cos every few days it would occasionally crash my DAW (Digital Audio Workstation) which I use for audio mixing. - Reaper. I now consistently use the Sonata BHD in WASAPI Exclusive Mode, and the laptop runs reliably for several weeks, and through being put into sleep mode several times a day, and with WASAPI, crashes of the DAW are very rare.

In comparison, I have had even better stability with EMU devices running ASIO, such as the EMU 0404 USB, which would run for months with not a single crash., but these are more prosumer/professional DAC's which also cost more than the Sonata HD, when I acquired them years ago., so not surprised about this.

My point being the CS 43131 implementation, if faulty, should be a manufacturer/design issue. So whatever is the issue with the Fosi D2 Dac, is not likely to be inherent in the DAC itself, but how Fosi has implemented their DAC, and maybe also their drivers.

From another source I understand that the drivers provided by Tempotec for ASIO are old. Pretty sure they are developed by a 3rd Party and this may have been the cause for their instability - On that note though, I can imagine most owners of the dongle DACs, will not be using ASIO on Windows., so NOT an issue affecting many.

The issue may also have been caused by the laptop I was running it on at the time, which is in need of a proper reinstall. Sadly this is a Windows thing, over time, the computer gets gunked up with all manner of updates, apps, helper apps, and there is no neat way of getting rid of the garbage. Much easier to reinstall.

On a more capable desktop, issues with DAW crashing, have not been as frequent., so maybe my gunked up laptop, was simply struggling to keep up.