DX340 iBasso developed Discret DAC, Easily replaceable batteries, Amp Modules! ***New firmware on 1st page. v1.03 local update. AMP16 NOS Tube amp. Review on the 1st page! 4/6/25

[ClaudiaRox]

Ye

This topic has been discussed quite a bit here.

Is your unit relatively new?

If so, drain the battery to 0. Fully charged to 100% before powering on. Run through a few of these cycles to calibrate the battery. This helped a lot for me.

What headphones/IEMs are you running and at what volume? If they are super power hungry that could explain it.

Running the DC In super gain mode will also preserve battery life somewhat.

I would recommend @Stouthart's tweaking guide but if you are only running MangoOS I don't believe this will have any impact.

Also lastly once the batteries are calibrated, if you still aren't satisfied, just keep it hooked to a power bank or other power source. Since the batteries are replaceable there is no harm in running the batteries suboptimal IMHO.

Yea my set is relatively new. I did the calibration thing like twice on the day of purchase, I'll try it again.

I'm mostly running 32ohm earbuds at mid gain about volume 30.

Yea I'm just confused at this point because I'm not using the power hungry Android... Haha

When the device is not in use, I'm also losing like 1-2% per hour.

 

[Whitigir]

Ye

Yea my set is relatively new. I did the calibration thing like twice on the day of purchase, I'll try it again.

I'm mostly running 32ohm earbuds at mid gain about volume 30.

Yea I'm just confused at this point because I'm not using the power hungry Android... Haha

When the device is not in use, I'm also losing like 1-2% per hour.

Just use the optimizer from Remko, it helps greatly

 

[gwompki]

Just use the optimizer from Remko, it helps greatly

If @ClaudiaRox is only using MangoOS the optimizer will not help correct? Or is MangoOS running on top of Android?

 

[ClaudiaRox]

anyways I just completed the Remko optimization. Initially I decided to skip it because I'm solely using MangoOS, but I'll see if there are any difference now after I do another battery 0 to 100 calibration.

 

[Peekayoh]

I use Mango exclusively and battery life is ok. I asked and Remko said there was likely no benefit to optimising.

 

[Whitigir]

I know some of you had concern about new vehicle and Aux in. I can confirm that. However, they can operate on HDD like 5TB WD black that I have stored my library on. It can also utilize just MicroSD and an adapter that I have which carries 400GB with me. This one has Bang&Olufsen speakers and between this to Harman Kardon, I prefer BO more

I tried to look for Aux in and they are no where to be found, so does the media source interface choices



I used Remko optimizer and advanced Tweak. I can say that I do prefer AndroidOS more than MangoOS nowadays but using Mango Player.

An update is that I have used this and left it 2 days of a full 48 hours under idling. My Dx340 only drained 5% leaving untouched on my desk with one tap of power button to turn on. It is not a deep sleep FYI. No wireless features on (wifi/bluetooth)

 

[Stouthart]

iBasso DX340 Debloating & Tweaking Guides

Version v4.7 is live (only code optimizations). Please check the release notes for changes and additional options.

I was about to launch an option to lower the minimum frequency for the 8 CPUs in the DX340, because - according to many - this could increase battery life. Below you can find the reason not to do this. Warning: nerd-talk ahead.

The CPUs in the DX340 are configured in 2 so-called big.LITTLE clusters: 4 energy-efficient cores for background tasks and 4 CPUs for foreground and "top" app's. iBasso and/or Qualcomm decided not to set the minimum CPU frequency value to the lowest possible number (300 MHz), but a bit higher. So I just set the minimum frequency for all cores to 300 MHz and started monitoring the actual usage.

Android/Linux provides an easy way to check the amount of time spent on each of the frequencies supported by a CPU (cat /sys/devices/system/cpu/cpu*/cpufreq/stats/time_in_state). For CPU0 (of the "LITTLE" cluster), these were the results after a period of mixed usage:

Freq    Time
300000  0
614400  151285
864000  62900
1017600 2203
1305600 1717
1420800 105
1612800 129
1804800 4342

As you can see, the CPU frequency NEVER drops to the lowest value. And most of the time is spent at 614.4 MHz, exactly the predefined value for scaling_min_freq.

For CPU7 of the "big" cluster, the results were as follows:

Freq    Time
300000  0
652800  162084
902400  17890
1056000 3700
1401600 17173
1536000 1189
1804800 1859
2016000 7531

Similar results, right? Not quite. When idle, it dropped to 652.8 MHz. When streaming audio however, the sweet spot seems to be 1401.6 MHz. And I swear that when I set this value as the minimum allowed frequency, the sound quality improves. That has to do with - most likely - the "cost" of frequency transitions (a topic for another time).

The moral of the story: just setting the minimum scaling frequency to 300 MHz is pointless. I may add an option in the future to tune the "big" cluster (CPUs 4-7) for battery life (652.8 MHz) or performance (1401.6 MHz), as the current default value of 1056 MHz does not make much sense.

If you've gotten this far... please get back to the music

 

[Monkeyk2010]

iBasso DX340 Debloating & Tweaking Guide

Version v4.7 is live (only code optimizations). Please check the release notes for changes and additional options.

I was about to launch an option to lower the minimum frequency for the 8 CPUs in the DX340, because - according to many - this could increase battery life. Below you can find the reason not to do this. Warning: nerd-talk ahead.

The CPUs in the DX340 are configured in 2 so-called big.LITTLE clusters: 4 energy-efficient cores for background tasks and 4 CPUs for foreground and "top" app's. iBasso and/or Qualcomm decided not to set the minimum CPU frequency value to the lowest possible number (300 GHz), but a bit higher. So I just set the minimum frequency for all cores to 300 GHz and started monitoring the actual usage...interesting results.

Android/Linux provides an easy way to check the amount of time spent on each of the frequencies supported by a CPU (cat /sys/devices/system/cpu/cpu*/cpufreq/stats/time_in_state). For CPU0 (of the "LITTLE" cluster), these were the results after a period of mixed usage:

Freq    Time
300000  0
614400  151285
864000  62900
1017600 2203
1305600 1717
1420800 105
1612800 129
1804800 4342

As you can see, the CPU frequency NEVER drops to the lowest value. And most of the time is spent at 614.4 GHz, exactly the predefined value for scaling_min_freq.

For CPU7 of the "big" cluster, the results were as follows:

Freq    Time
300000  0
652800  162084
902400  17890
1056000 3700
1401600 17173
1536000 1189
1804800 1859
2016000 7531

Similar results, right? Not quite. When idle, it dropped to 652.8 GHz. When streaming audio however, the sweet spot seems to be 1401.6GHz. And I swear that when I set this value as the minimum allowed frequency, the sound quality improves. That has to do with - most likely - the "cost" of frequency transitions (a topic for another time).

The moral of the story: just setting the minimum scaling frequency to 300 GHz is pointless! I may add an option in the future to tune the "big" cluster (CPUs 4-7) for battery life (652.8GHz) or performance (1401.6GHz), as the current default value of 1056 GHz does not make much sense.

If you've gotten this far... please get back to the music

Thanks for all you hard work, one thing (IT mode on) I think you mean Hz instead of ghz else we would have an ultra fast dx340 with a sweet spot of 1401ghz

Edit; ofcourse it should be mhz instead of Hz my Bad.

 

[Stouthart]

Thanks for all you hard work, one thing (IT mode on) I think you mean Hz instead of ghz else we would have an ultra fast dx340 with a sweet spot of 1401ghz

Sharp, thanks! Actually...it should be MHz (max supported frequency is ± 2 GHz = 2000 MHz). Hope I fixed them all.

 

[lairpost]

Thanks for all you hard work, one thing (IT mode on) I think you mean Hz instead of ghz else we would have an ultra fast dx340 with a sweet spot of 1401ghz

err, I think you meant MHz. Looks like the cores aren't stable at 300Mhz thus they can't be set there.

 

[Stouthart]

err, I think you meant MHz. Looks like the cores aren't stable at 300Mhz thus they can't be set there.

Exactly!

 

[pietro77]

If anyone is interested someone selling on ebay uk dx340 for £1000.

 

[Monkeyk2010]

Yup you are all right my bad mhz it is

 

[Death_Block]

If anyone is interested someone selling on ebay uk dx340 for £1000.

How come?
Very tempting….

 

[pietro77]

How come?
Very tempting….

Well, I have no idea. But few days ago I saw listings which ends with price little over £900. That makes me think maybe I should wait little bit longer and pay much less than full price. On other hand I'm enjoy dx340 for month now