Head-Fi.org › Forums › Equipment Forums › Portable Source Gear › The iBasso DX50 Thread - Latest firmware: 1.5.0
New Posts  All Forums:Forum Nav:

The iBasso DX50 Thread - Latest firmware: 1.5.0 - Page 660

post #9886 of 14482
Quote:
Originally Posted by Fluvance View Post

When I go to All Music, or change the sort mode in All Music, it takes about 10 seconds to load. Every time. It's almost three times longer than my old 32 gig card.
Rule of thumb: the larger the flash chip, the slower the read performance. It has to do with how blocks of flash cells are arranged. So, you have twice as much music on a flash chip that's measurably slower and the load times are longer? That's to be expected.
post #9887 of 14482

post #9888 of 14482
Quote:
Originally Posted by ratinox View Post


Rule of thumb: the larger the flash chip, the slower the read performance. It has to do with how blocks of flash cells are arranged. So, you have twice as much music on a flash chip that's measurably slower and the load times are longer? That's to be expected.


I'm just curious as to why the loading is even necessary. By the looks of it, the player is building the list every time you navigate to All Music, or change the sorting options. Wouldn't the RAM be big enough to hold the built list? I'm assuming that would speed it up, without taking too many system resources. I don't have a large amount of programming experience, so I'm only guessing here.

post #9889 of 14482
Quote:
Originally Posted by babs View Post

With lossless material (FLAC...) and with the EQ off, the only purpose of the CODEC is to decode the original samples of the music, as-is, and send that data to the WM8740, I see no need to alter these data before sending it to the DAC.

Well, no. It's not that simple, because...
Quote:
All the DSP processing is internal to the WM8740; Filtering, oversampling, volume control, Sigma-Delta...

... this statement is untrue. Volume control in particular happens in software prior to the DAC stage. I don't have a full list of what happens where but volume control most definitely happens prior to the DAC. You can confirm this by using the coaxial output and listening to what happens when you adjust the volume.
Quote:
I personally made some fast measurements on my DX50 (1.2.7), with a scope and a multimeter ( maximum level, volume = 255, phone out, no load)
I use pure sine signal from WAV files, at maximum level (0dBfs)

Frequency response is ruling flat from 20Hz to 20kHz, maybe -0.1dB at 20Hz and -.2dB at 20kHz
(with a load, low frequency response will be affected, but its not the point here)

Personally, i didn't notice any sonic difference between firmware versions, but i think its very difficult to validate that without owning two DX50

At the same time, I also check the EQ function:

Frequency response is also ruling flat

THANK YOU for this. This is precisely what I've been hoping someone would do (I'd have done it myself but I don't have the instruments).

You can install an older firmware version. Simply follow the upgrade instructions using whichever update.img file you want to install.
Quote:
But... there is a Bug:
In the 1.2.7 firmware, Ibasso push the maximum boost to +12dB, but keep the same pre-attenuation of 6dB !! leading to risk of clipping

Not just risk. It's certain to clip unless you reduce the master volume by another 6dB. This is definitely something that needs to be brought to iBasso's attention. Thanks for finding it.
post #9890 of 14482
Quote:
Originally Posted by ratinox View Post.
Not just risk. It's certain to clip unless you reduce the master volume by another 6dB. This is definitely something that needs to be brought to iBasso's attention. Thanks for finding it.

Didn't one step on the EQ equal 0.5dB? So you still only have 6dB to work with, but in ½-dB increments?

post #9891 of 14482
Thread Starter 
Quote:
Originally Posted by Sorensiim View Post



Same here, fun to watch the wrestling going on with ze firmwares.
I watch with my headphones on while great music is playing. popcorn.gif
post #9892 of 14482
Quote:
Originally Posted by musicheaven View Post


Same here, fun to watch the wrestling going on with ze firmwares.
I watch with my headphones on while great music is playing. popcorn.gif


Getting old though...

post #9893 of 14482
Quote:
Originally Posted by Sorensiim View Post

Didn't one step on the EQ equal 0.5dB? So you still only have 6dB to work with, but in ½-dB increments?

The steps are still 0.5dB but v1.2.7 has 24 steps, twice as many as v1.2.6 and earlier offering a 12dB range per band compared to the 6dB range of earlier versions. The EQ system is subtractive so the master volume needs to be attenuated to prevent clipping. The bug is that v1.2.7 appears to be using the v1.2.6 and earlier 6dB attenuation instead of the requisite 12dB.

A simple fix if you're using the EQ is to reduce the master volume by 6 or 7 steps (ie, from 255 to 249 or 248). This should give you the full EQ range without clipping at the volume ceiling.
post #9894 of 14482
Thread Starter 
Quote:
Originally Posted by bearFNF View Post
 


Getting old though...

 

Yep it is!

 

post #9895 of 14482
Quote:
Originally Posted by Fluvance View Post

I'm just curious as to why the loading is even necessary.
The list is generated from the scan database that is now stored on the media itself. Whenever you change play order or whatever it needs to go to that database to build the play list.

Why not store this in RAM? That's simple: scale. Rockchip RK2926 has 512MB total RAM. Extrapolating out to the 2TB theoretical capacity of microSDXC, my music database would grow to anywhere from 100MB to 200MB. That just won't fit in RC2926's memory space along with the OS, the player software, the file buffers and DAC buffers.
post #9896 of 14482
Quote:
Originally Posted by ratinox View Post


The steps are still 0.5dB but v1.2.7 has 24 steps, twice as many as v1.2.6 and earlier offering a 12dB range per band compared to the 6dB range of earlier versions. The EQ system is subtractive so the master volume needs to be attenuated to prevent clipping. The bug is that v1.2.7 appears to be using the v1.2.6 and earlier 6dB attenuation instead of the requisite 12dB.

A simple fix if you're using the EQ is to reduce the master volume by 6 or 7 steps (ie, from 255 to 249 or 248). This should give you the full EQ range without clipping at the volume ceiling.

But wouldn't that be 6dB in 1.2.7 to get to 0 (12 db in actual 0.5dB steps, instead of the nominal 1dB steps), and then you go down from there another 6 dB (with no clipping in any case, as you would be going down)?

post #9897 of 14482
I also vote for a 1.2.2 sound signature (with the functionality of 1.2.7).

I also like the sound of 1.1.6 a lot. I use two DX50's; one loaded with 1.2.2; one with 1.1.6. I've removed all the FLAC file tags and shortened file/folder names and everything works OK; you just have to live with the long and sometimes flaky SDHC scans. It's worth it for the (to my ears) better sound presentation.
post #9898 of 14482
Quote:
Originally Posted by geagle View Post

But wouldn't that be 6dB in 1.2.7 to get to 0 (12 db in actual 0.5dB steps, instead of the nominal 1dB steps), and then you go down from there another 6 dB (with no clipping in any case, as you would be going down)?

When you activate the EQ system in v1.2.6 the master volume is automatically reduced by 6 steps allowing for a 12-step range without clipping. The bug in v1.2.7 appears to be that the master volume is reduced by 6 steps when it should be reduced by 12 steps. In order to compensate you just reduce the volume by another 6 steps. This will provide you with 12 steps from +0 to +12 without exceeding the numerical volume or clipping ceiling (volume 255).
post #9899 of 14482

I somehow still preffer the 1.2.2 firmware. not because of better sound or whatever you guys imagine you are hearing :D, but because, IIRC, the multi disk albums were organized correctly there. iBasso should really improve that.

post #9900 of 14482
Quote:
Originally Posted by nihontoman View Post

I somehow still preffer the 1.2.2 firmware. not because of better sound or whatever you guys imagine you are hearing biggrin.gif , but because, IIRC, the multi disk albums were organized correctly there. iBasso should really improve that.

No, they weren't. The "x of y" and "x/y" track/disc tag formats are not part of any accepted standards or conventions outside of certain proprietary tools like iTunes and dBpoweramp. DX50 has never supported them directly. The reason that what you say worked actually worked is when DX50 registered those tags as "null" which forced it to rely on file names for sorting. Most of those tag bugs have been fixed.

The solution is to tag your music correctly. That is, number the tracks sequentially within entire albums.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Portable Source Gear
Head-Fi.org › Forums › Equipment Forums › Portable Source Gear › The iBasso DX50 Thread - Latest firmware: 1.5.0