Head-Fi.org › Forums › Equipment Forums › Portable Source Gear › The iBasso DX50 Thread - Latest firmware: 1.9.5 - June 30, 2016
New Posts  All Forums:Forum Nav:

The iBasso DX50 Thread - Latest firmware: 1.9.5 - June 30, 2016 - Page 950

post #14236 of 18489
Quote:
Originally Posted by fonsocm View Post

 

Hi. 

 

I was testing Mega Sound Explosion Second Edition and it is very good, great SQ, in my opinion, much detail, clear and round. But, in my opinion, if you proved the stock 1.5.0, it looks some artificial, too rounded.

 

Now, I installed Sound Unlocked 1.2.3 with lossless only and high irr, it sounds very, very good; great clarity and soundstage (I hope it be the correct term) without that sense of artificiality of Mega Sound. It is a very good firm.

 

The main problem is my SD card only has 32 Gb.

 

I am using a Shure 535 RLTD without dac/amp in LO.

 

Why do you say it is a bad proof? Only curiosity.

 

+1

 

I installed v1.2.3 last night, and it sounds great, with the exact same settings as you.  Listening to Led Zeppelin - Mothership (24-bit vinyl rip), and it sounds very clear.  You can really hear Plant's vocal range in some of those tracks.

post #14237 of 18489
Quote:
Originally Posted by nihontoman View Post
 


could you please give some additional info? I installed WASAPI plugin for foobar and played with buffer in foobar, but will it affect the whole audio? also, what output should I use? thre are "normal" outputs and then there are wasapi event and wasapi push. what do I choose?

If you use "normal" - you bypass the WASAPI, so you should choose "event"(e.g. pull) which is the best option, or if it is not working correctly - use "push". 

 

I personally prefer ASIO4ALL with "high process priority", "64 bit driver" set in foobar preferences ; and in ASIO4ALL advanced settings window only "allow pull mode (wave rt)" checked with buffer offset = 4ms and ASIO Buffer Size = 72 samples. These buffer settings should work equally good for anything from mp3 to 24/192 flac.

 

Cheers!


Edited by Yanec - 7/15/14 at 11:35am
post #14238 of 18489
Quote:
Originally Posted by CoiL View Post
 

One more thing - my recommendation with Sound Unlocked 1.2.2 is to turn off irr bandwidth (for me seems to make whole spectrum too harsh sounding),

It actually depends on the phones you are using - with my VSonic GR07 MKII I totally agree - they are very harsh with irr bandwidth set to on, but the Sony MH1c or Creative EP-830 really shine with irr bandwidth set to ON. The HiFiMan ER-400 are somewhere in between - I would personally use them with irr set to ON.

 

By the way using high gain on the DX50 amp anything sounds harsh to me, but with mid gain it all sounds just right - to my taste way better than low or high.

 

Cheers!


Edited by Yanec - 7/15/14 at 11:50am
post #14239 of 18489
Quote:
Originally Posted by olddude View Post
 

I just tried the above firmware using, again, Aja.  It's not much different from stock 1.5.0, soundstage didn't seem any wider.  What I did notice was a sort of lack of clarity in the middle of the soundstage.  Almost like the music was coming through a (very) thin sheet of fabric.  The edges seemed fine (slightly enhanced, perhaps) but the middle (imagine left ear/middle of head/right ear) seemed not as pristine.  I reinstalled 1.5.0 and that unclarity was gone.  

 

Again, I'm using a JDS Labs C5 amp and LO and that will change things compared to just listening through the HO of the DX50.  I really, from my point of view, think iBasso nailed it with this FW.


Thanks for sharing your impressions - it is good to have someone as experienced commenting on these!

 

Cheers!

post #14240 of 18489

Yanec, actually for PLAYBACK only it is recommended to set ASIO buffer size 512 and above. Low buffer and low latency is needed only for recording/mixing purposes. Setting buffer size higher doesn`t affect SQ, only avoids possible jitter/distortion/clicks/pauses because larger payload to/from RAM (like data on queque waiting to be played instead taking it straight and getting possible lag in data stream). I`ve had some few cases where I had to set larger buffer size to play hi-res lossless without problems.

 

More into topic again. I`ve been doing intensive comparing between irr bandwidth off vs. on and what I`ve found is - with OFF setting soundstage moves further away and has more depth towards front and it`s noticeably larger with more air/decay/reverb. Separation is still good. With irr ON soundstage is much closer and despite everything being clearer and with littlebit better instrument separation I personally find it degrading soundstage size and spaceful imaging. Maybe it is just my taste and fit for Fidelio X1 but I prefer SU1.2.3 with irr bandwith OFF + lossless only + DF filter slow roll-off + DX50 gain set to mid.

If someone bothers to compare, then additional opinion would be appreciated.

 

Btw, someone should add Dmitry`s Sound Unlocked 1.2.3 and link to first post: https://github.com/dm1try/dx50_sound_unlocked/blob/master/README.md


Edited by CoiL - 7/15/14 at 12:10pm
post #14241 of 18489
Quote:
Originally Posted by CoiL View Post
 

Yanec, actually for PLAYBACK only it is recommended to set ASIO buffer size 512 and above. Low buffer and low latency is needed only for recording/mixing purposes. Setting buffer size higher doesn`t affect SQ, only avoids possible jitter/distortion/clicks/pauses because larger payload to/from RAM (like data on queque waiting to be played instead taking it straight and getting possible lag in data stream). I`ve had some few cases where I had to set larger buffer size to play hi-res lossless without problems.

 

More into topic again. I`ve been doing intensive comparing between irr bandwidth off vs. on and what I`ve found is - with OFF setting soundstage moves further away and has more depth towards front and it`s noticeably larger with more air/decay/reverb. Separation is still good. With irr ON soundstage is much closer and despite everything being clearer and with littlebit better instrument separation I personally find it degrading soundstage size and spaceful imaging. Maybe it is just my taste and fit for Fidelio X1 but I prefer SU1.2.3 with irr bandwith OFF + lossless only + DF filter slow roll-off + DX50 gain set to mid.

If someone bothers to compare, then additional opinion would be appreciated.

 

Btw, someone should add Dmitry`s Sound Unlocked 1.2.3 and link to first post: https://github.com/dm1try/dx50_sound_unlocked/blob/master/README.md

On my PC ASIO buffers of 512 and above are fine for some formats, but introduce "clicks" to the sound of other formats. The exact buffer parameters that I mentioned are those that work without any problems on my side. May be there is no universal setting, but my settings work fine on my computers. 

 

I am amused that you find such substantial differences between irr OFF and irr ON. My personal experience is that irr OFF is more airy and analytical while being just a bit rougher and more sibilant, but I honestly felt no difference in the soundstage and the instrument placement and separation. May be it is my hearing being less sensitive(after so many concerts and parties LOL) than yours or it really depends that much on the phones.

 

+1 to have Sound Unlocked home page in the first post - it is deserving it as most puristic and neutral FW - more neutral than Rockbox and DOC's "equalized/tweaked" versions.

 

Cheers!


Edited by Yanec - 7/15/14 at 1:00pm
post #14242 of 18489

After using firmware 1.5.0 for a while (since its release) I do have to say that the gapless playback isn't as good as it was in 1.2.8.  Seems to have some random issues.  Sometimes I hear a pop or tick between .flacs in live performances, sometimes a track doesn't start at full volume but fades in.  A little bit like the crossfade bug in a previous firmware, but it isn't crossfading.  Just fades in the start of a track.  

 

A majority of the time the gapless is OK, but randomly it does either fade a song in that shouldn't be faded in or add a pop sound.  Experienced none of this in 1.2.8.  

 

I hope someday they release a firmware that is 100% consistent, reliable and feature rich.  

post #14243 of 18489
Quote:
Originally Posted by fonsocm View Post
Why do you say it is a bad proof? Only curiosity.

I have no skills in sound engineering. I need to know that we on the right way, I need a feedback.

Anyway, I have other engineering skills and it's kinda strange to me that developers use software tweaks to do soundstage tighter, hide sound details, etc.

then people use AMP to do soundstage wider, reveal sound details, etc. What do you think?)

 

 

Quote:
Originally Posted by CoiL View Post
 

 

One more thing - my recommendation with Sound Unlocked 1.2.2 is to turn off irr bandwidth (for me seems to make whole spectrum too harsh sounding),

turn on lossless only and use slow roll-off DF.

I thinks, depends on the headphones that why this one is option. 

 

Quote:
Originally Posted by Yanec View Post
 

It actually depends on the phones you are using - with my VSonic GR07 MKII I totally agree - they are very harsh with irr bandwidth set to on, but the Sony MH1c or Creative EP-830 really shine with irr bandwidth set to ON. The HiFiMan ER-400 are somewhere in between - I would personally use them with irr set to ON.

 

By the way using high gain on the DX50 amp anything sounds harsh to me, but with mid gain it all sounds just right - to my taste way better than low or high.

 

Cheers!

 

Totally agree :) BTW, some impressions about settings and used phones can be added to the github FW page.

 

Quote:

Originally Posted by CoiL View Post
 

Btw, someone should add Dmitry`s Sound Unlocked 1.2.3 and link to first post: https://github.com/dm1try/dx50_sound_unlocked/blob/master/README.md

 

And maybe someone can help to do FW description more friendly :)) Thx. 

post #14244 of 18489
Quote:
Originally Posted by Yanec View Post
 
Quote:
Originally Posted by CoiL View Post

 

More into topic again. I`ve been doing intensive comparing between irr bandwidth off vs. on and what I`ve found is - with OFF setting soundstage moves further away and has more depth towards front and it`s noticeably larger with more air/decay/reverb. Separation is still good. With irr ON soundstage is much closer and despite everything being clearer and with littlebit better instrument separation I personally find it degrading soundstage size and spaceful imaging. Maybe it is just my taste and fit for Fidelio X1 but I prefer SU1.2.3 with irr bandwith OFF + lossless only + DF filter slow roll-off + DX50 gain set to mid.

If someone bothers to compare, then additional opinion would be appreciated.

 

I am amused that you find such substantial differences between irr OFF and irr ON. My personal experience is that irr OFF is more airy and analytical while being just a bit rougher and more sibilant, but I honestly felt no difference in the soundstage and the instrument placement and separation. May be it is my hearing being less sensitive(after so many concerts and parties LOL) than yours or it really depends that much on the phones.

I just did blind test 3 times to my girlfriend and she also likes irr off better because it has more natural and airy live-concert-like imaging and soundstage... and depth - her words ;)


Edited by CoiL - 7/15/14 at 2:00pm
post #14245 of 18489
Quote:
 

 

Totally agree :) BTW, some impressions about settings and used phones can be added to the github FW page.

 

 

 

And maybe someone can help to do FW description more friendly :)) Thx. 

I'll be glad to be of any assistance, although English is not my native language as you probably already guessed. :)

post #14246 of 18489
Quote:
Originally Posted by CoiL View Post
 

I just did blind test 3 times to my girlfriend and she also likes irr off better because it has more natural and airy live-concert-like imaging and soundstage... and depth - her words ;)


Fair enough - I guess it's the earphones or my hearing as I said. :) I'll try to steal some time tomorrow for some more comprehensive comparing.

 

Cheers!

post #14247 of 18489
Quote:
Originally Posted by CoiL View Post
 

I just did blind test 3 times to my girlfriend and she also likes irr off better because it has more natural and airy live-concert-like imaging and soundstage... and depth - her words ;)

She's right... Zeppelin's Immigrant Song sounds more airy.

post #14248 of 18489
Quote:
Originally Posted by CoiL View Post
 

Actually it is not hard at all to setup. And the sound quality might improve noticeably depending on which DAC/amp being used. Directsound under windows and pulseaudio under linux for example are poor quality imho if You are looking for quality sound and are using dedicated dac/amp since that data/audio goes through OS internal kernel mixer and it affects sound in bad way. For example jumping from directsound to ASIO I had noticeably quality improvement with Aune T1, more detailed, better separation and soundstage etc. But this is only my experience with my rig. Many say they hear no difference. Media player classic should be also quite easy to setup but I don`t get the point if You are already using foobar. What concerns Youtube - I don`t really care about its sound since You can`t get high sample/bps audio from there anyway. For me personally, easiest is using ASIO4ALL. Here is a pic sample how it`s set up: http://cdn.head-fi.org/3/35/35c8322c_AuneT1_ASIOWAOPsettings.jpeg

 

Will test different plugin options tonight. Was too tired yesterday.

 

One more thing - my recommendation with Sound Unlocked 1.2.2 is to turn off irr bandwidth (for me seems to make whole spectrum too harsh sounding),

turn on lossless only and use slow roll-off DF.

So far it has been best FW regarding to SQ. Tested with Fidelio X1, Piston v2.1 and modified HD-681 and was comparing with Aune T1 (detailed setup on profile).

MSESE didn`t impress me at all, even stock 1.5.0 sounded better to me and with my setup. All I would like from SU1.2.2 is that gapless mode would work with lossless only but it`s minor thing that doesn`t bother me. When I get bored will try more FW`s but atm seems I`m going to stay on SU1.2.2 - highly recommend it to others! Thanks for that awsome FW Dmitry ;)

 

I don't care about media player classic or youtube audio quality, I DO care about video and audio sync which apparently needs additional syncing... can I just "tell" wasapi (or whatever is better to use) to just increase/decrease buffer when DX50 is being used for everything not just foobar? it would be much easier to just get that audio half a second earlier so it isn't late any more...

Quote:
Originally Posted by Yanec View Post
 

If you use "normal" - you bypass the WASAPI, so you should choose "event"(e.g. pull) which is the best option, or if it is not working correctly - use "push". 

 

I personally prefer ASIO4ALL with "high process priority", "64 bit driver" set in foobar preferences ; and in ASIO4ALL advanced settings window only "allow pull mode (wave rt)" checked with buffer offset = 4ms and ASIO Buffer Size = 72 samples. These buffer settings should work equally good for anything from mp3 to 24/192 flac.

 

Cheers!

 

the problem I encountered is that when I use WASAPI (be it event or push) sometimes audio just stops working for youtube and also media player classic (O know right, messing with utput settings in foobar shouldn't be affecting other programs, but appearently, it is...)

 

also, there is this problem - when using wasapi, I can't set bitrate to 24, as it won't play anything in foobar and if I set it at 16 - it doesn't play anything over 16/44.1 I hgave lot's of stuff that's high bitrate and everything works when I choose just "normal" ibasso mango player" as output in foobar...

 

Quote:
Originally Posted by CoiL View Post
 

Yanec, actually for PLAYBACK only it is recommended to set ASIO buffer size 512 and above. Low buffer and low latency is needed only for recording/mixing purposes. Setting buffer size higher doesn`t affect SQ, only avoids possible jitter/distortion/clicks/pauses because larger payload to/from RAM (like data on queque waiting to be played instead taking it straight and getting possible lag in data stream). I`ve had some few cases where I had to set larger buffer size to play hi-res lossless without problems.

 

More into topic again. I`ve been doing intensive comparing between irr bandwidth off vs. on and what I`ve found is - with OFF setting soundstage moves further away and has more depth towards front and it`s noticeably larger with more air/decay/reverb. Separation is still good. With irr ON soundstage is much closer and despite everything being clearer and with littlebit better instrument separation I personally find it degrading soundstage size and spaceful imaging. Maybe it is just my taste and fit for Fidelio X1 but I prefer SU1.2.3 with irr bandwith OFF + lossless only + DF filter slow roll-off + DX50 gain set to mid.

If someone bothers to compare, then additional opinion would be appreciated.

 

Btw, someone should add Dmitry`s Sound Unlocked 1.2.3 and link to first post: https://github.com/dm1try/dx50_sound_unlocked/blob/master/README.md

 

that's good to know. if I ever manage to make wasapi work with 24 bit stuff for DX50 :D

 

 

 

anyways, I just want to use DX50 as my DAC nonstop, without having to turn to inbuilt PC audio card for videos, because DX50 lags the sound about half a second and watching videos in this manner is quite frustrating...

post #14249 of 18489

That`s why I like using ASIO4ALL - I only adjust it once to desired buffer and that`s it. Now only thing You`ll have to do with other programs is to setup their output to ASIO output/plugin and every time that program starts it will automatically run ASIO4ALL at background. I haven`t tested WASAPI and its plugins so much to tell if it lets to use it same way like ASIO4ALL.

With KSP plugin I had problems playing some 24-bit hi-res files but it is much easier to setup with Winamp for example. But I still went with ASIO4ALL because it hasn`t had any problems with hi-res files and I can easily adjust buffers & latency when I do recording/mixing in Cubase. Only problem I have with ASIO4ALL is that when I have winamp playing, it won`t play additional sounds simultaneously for example with Youtube. If I close winamp, then it`s ok. Don`t know about WASAPI 24-bit/16-bit setup with different sample rate but You should set Your audio properly in windows mixer/playback settings and set to "Allow applications to take exclusive control of this device", then there should be no conflict between 16/24-bit. Actually I think that`s why winamp+ASIO4ALL won`t let Youtube to play sounds simultaneously because one is in exclusive mode with 24-bit but Youtube/browser is 16bit. Oeh... long story :D

Anyway, haven`t had lag/sync problems yet with all my other systems + ASIO4ALL. About DX50 + ASIO4ALL @ USB-DAC mode... I have to test yet because I`ve been doing mostly FW comparing. Will try to test tonight after work. Will let You know if I have any lag/sync problems using ASIO.

 

 

About IRR Bandwith OFF vs. ON ...I think it is a really good option for DX50 users - analytical listeners will probably prefer ON, while smoother and airier "tube" sound fans will probably choose OFF. Of course it is only my personal subjective view. It also depends on music, taste, cans etc. which mode to select. Most of the time I seem to prefer irr off but there are some recordings that benefit from irr on and its more analytical sound.

Quote:
Originally Posted by CoiL View Post
 

Found one 64GB uSD card from my country web-shop and it costs only 29€ - ADATA Premier Micro SDXC UHS-I 64GB (Video Full HD) +SDHC Adapter

Manufacturer link: http://www.adata.com/index.php?action=product_feature&cid=7&piid=203

Should be good but we`ll see, going to get one.


This card is approved by me - works flawless and is pretty fast. Cheap and good ;)

 


Edited by CoiL - 7/16/14 at 1:04am
post #14250 of 18489
Quote:
Originally Posted by medmitry View Post
 

I have no skills in sound engineering. I need to know that we on the right way, I need a feedback.

Anyway, I have other engineering skills and it's kinda strange to me that developers use software tweaks to do soundstage tighter, hide sound details, etc.

then people use AMP to do soundstage wider, reveal sound details, etc. What do you think?)

 

 

I prefer don't express my opinion in public about this topic. ;)

 

The DAC/AMP are the fashion device.

 

Great firm, great job.

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: 1.9.5 - June 30, 2016