[jj69]

The thing is iBasso have repeatedly stated that they dont change the sound between fw versions

 
I've always suspected that it is Rockchip (manufacturer of the SOC processor) and not iBasso that is behind the sound changes.  My understanding is that firmware development comes along with purchase of the SOC,  so any firmware changes iBasso wants to make are most likely referred to Rockchip for implementation. 
 
Why would iBasso put so much effort into using the best hardware components in their DAPs, only to intentionally color the sound with DSP processing tricks?  If that's the kind of DAP they wanted to build, why even bother with expensive components like Sabre DACs?  The cheapest reasonably well designed mass market components + DSP would likely yield the same result. 
 
As for exactly why the sound changes (and quite frankly, has always been plagued with colorations IMO), I've always felt it was due more to incompetence on Rockchip's part rather than anything else.  Whether they're using "immature sound libraries" as some have speculated, or intentional DSP processing to create a specific sound signature, isn't really relevant.  I can see no logical reason why anyone would make changes to an existing sound library, nor do I see any reason why anyone would use DSP processing in a specialty DAP.  The whole thing remains a mystery to me. 

 

[goodvibes]

My guess is that you would be totally wrong. I would wager you any amount that there are no 'DSP processing tricks' beyond what I will mention here, The codecs have likely never changed and there is no DSP for sound tailoring. That would show up in testing which the DX90 does extremely well. What iBasso is likely playing with are things like buffers and resource allotment to get it more optimized. The sonic differences aren't that great but noticeable simply because it's a very revealing device. FW updates on my much less revealing clip or fuze had even greater sonic differences and none approached the sound of any of the DX90 iterations for me.
 
Folks need to stop thinking that if you can hear a difference that it will show up in specs or that something untoward was done to achieve it. Assuming that they are cheating the sound and complaining about it is very unfair to the maker.

 

[HiFlight]

My guess is that you would be totally wrong. I would wager you any amount that there are no 'DSP processing tricks' beyond what I will mention here, The codecs have likely never changed and there is no DSP for sound tailoring. That would show up in testing which the DX90 does extremely well. What iBasso is likely playing with are things like buffers and resource allotment to get it more optimized. The sonic differences aren't that great but noticeable simply because it's a very revealing device. FW updates on my much less revealing clip or fuze had even greater sonic differences and none approached the sound of any of the DX90 iterations for me.

Folks need to stop thinking that if you can hear a difference that it will show up in specs or that something untoward was done to achieve it. Assuming that they are cheating the sound and complaining about it is very unfair to the maker.

Very well said!!!

 

[jj69]

  My guess is that you would be totally wrong. I would wager you any amount that there are no 'DSP processing tricks' beyond what I will mention here, The codecs have likely never changed and there is no DSP for sound tailoring. That would show up in testing which the DX90 does extremely well. What iBasso is likely playing with are things like buffers and resource allotment to get it more optimized. The sonic differences aren't that great but noticeable simply because it's a very revealing device. FW updates on my much less revealing clip or fuze had even greater sonic differences and none approached the sound of any of the DX90 iterations for me.
 
Folks need to stop thinking that if you can hear a difference that it will show up in specs or that something untoward was done to achieve it. Assuming that they are cheating the sound and complaining about it is very unfair to the maker.

 
Well, then we'll have to agree to disagree.  The sonic differences beteween firmware releases are significant ones to my ears, and they're even more obvious if you listen to the differences in the DX50's firmware versions.  If the sonic differences I'm hearing are simply the result of buffer settings, then the entire device must have some serious design problems.  Personally, I don't believe that is the case at all.  If you listen to the DX50 running Rockbox, all of that device's irritating colorations immediately dissolve.  I can only hope we'll some day get to hear the DX90 under the same conditions. 
 
I have also never said the sonic differences would show up in "specs" or "testing."  In fact, it doesn't suprise me in the least that they don't - but that certainly doesn't mean those sonic differeces don't exist. 

 

[headwhacker]

I looked at each FW packages and I can see differences with sound fx libraries (/etc/audio_effects.conf) being loaded in the system.img.

2.1.0 does not seem to load anything from the soundfx lib folder because it is empty. While 2.0.5 has 5 libraries and all are loaded.

However, I followed what Doc2008 did with his unlocked modded fw for DX90 by commenting all lines in audio_effects.conf and deleted audio_policy.conf. The sound changed enough to be noticed. I detected a mid bass bump which make bass sound boomy than the stock 2.1.0. The mids a little bit recessed which give me the impression of a V sound.

Whether it's iBasso or Rockchip who is responsible, I am convinced each FW version is different and has influence with the sound.

I think the stock player without any enhancement has a default V sound.

Just my 2 cents worth..

 

[SilverEars]

I looked at each FW packages and I can see differences with sound fx libraries (/etc/audio_effects.conf) being loaded in the system.img.

2.1.0 does not seem to load anything from the soundfx lib folder because it is empty. While 2.0.5 has 5 libraries and all are loaded.

However, I followed what Doc2008 did with his unlocked modded fw for DX90 by commenting all lines in audio_effects.conf and deleted audio_policy.conf. The sound changed enough to be noticed. I detected a mid bass bump which make bass sound boomy than the stock 2.1.0. The mids a little bit recessed which give me the impression of a V sound.

Whether it's iBasso or Rockchip who is responsible, I am convinced each FW version is different and has influence with the sound.

I think the stock player without any enhancement has a default V sound.

Just my 2 cents worth..

I think I said U shaped when I got it, not V. Maybe I meant to say V shaped, I dunno. To me, the treble and bass lift was very apparent.  It wasn't a warm signature that's for sure.  The way the treble was done, it sounded cold to me.

 

[headwhacker]

But I believe they made a very good balance with the stock 2.1.0.

 

[SilverEars]

But I believe they made a very good balance with the stock 2.1.0.

Did the LO sig change also?  By the way, I got rid of my Vorzuge, O2 still tops it.    I think I was over hyped with the Vorzuge, it sounded tad brighter, but not more detail and not as wide as O2.

 

[ttt123]

   
Well, then we'll have to agree to disagree.  The sonic differences beteween firmware releases are significant ones to my ears, and they're even more obvious if you listen to the differences in the DX50's firmware versions.  If the sonic differences I'm hearing are simply the result of buffer settings, then the entire device must have some serious design problems.  Personally, I don't believe that is the case at all.  If you listen to the DX50 running Rockbox, all of that device's irritating colorations immediately dissolve.  I can only hope we'll some day get to hear the DX90 under the same conditions. 
 
I have also never said the sonic differences would show up in "specs" or "testing."  In fact, it doesn't suprise me in the least that they don't - but that certainly doesn't mean those sonic differeces don't exist. 

From the perspective of high end audio, where everything makes a difference, from reseating the RCA plug to remove oxide buildup, applying a cleaner, like Caig Pro-Gold to plugs, and all metal to metal contact surfaces. I highly recommend this for earphone headphone contacts, plugs, etc., BTW.  This is the cheapest and simplest improvement you can do, with major effects on the SQ, due to removing oxide buildup, and providing a good metal to metal contact.  Should be done periodically.  This Caig Pro Gold, and a German product called Cramolin (which I cannot find anymore) are the only contact cleaners I found that  really work, and over an extended period of time.  If you've never tried them, you'll be amazed at the difference a clean contact can provide. 
 
People should keep in mind that discussion and feedback is a civil medium, preferably between people with the same interests.  When done positively, the result is beneficial to everyone.  Keep in mind that both sides do have the same interest, and keep the discussion positive.  Neither side owes anything to the other, and when there is no benefit, either side can walk away.  For us users, we all vote with our wallets.  Find something that is better, and also a good value, and we will move on to the next "best value".
 
For me, the DX90 is a blast, as it is offering a SQ that is rivaling DAPS costing 3-6 times more.  Yes, the DX90  is not perfect, but it is pretty good for the money, and it is also making music more enjoyable and affordable.  For that, I have a lot of patience and forgiveness for the negatives that are present, which I am choosing to live with.  Everything is about balance, whether the positives override the negatives.  This is true of your work, your mate, your hobby.
 
Case in point, each fw version has a different SS (Sound Signature).  You can look on that as a positive or a negative.  On the positive side, you can pick whichever one you like, and that has synergy with your equipment, and use that one.  Isn't it nice that you have a choice? 
 
After all, there  is  absolutely no way to make 100% of the people happy, given this super finicky audience.  Every choice has tradeoffs, for the Manufacturer as well as the user.  That is  the name of the hobby,  and I am sure that we each want our own choice.  And deservedly so.  I want my biases, no matter whether other people believe in them or not. 
High end audio is by it's nature imperfect, as we happily change batteries, cables, IEMs, DAPs, Amps. 
 
Look around at your non-audiophile friends, who are not afflicted with this disease. They are supremely happy with their Radio Shack setup, while you are agonizing on DAPS, amps, silver cables, versus silver plated,   BA versus Dynamic, hybrids, multi-driver IEMS at ridiculous prices that could be enough for a down payment on a car.
 
Just remember that you could choose to be like them, happy with a simple setup, with no urge to upgrade/change, or be constantly evaluating the SS/SQ after every change of hw or fw.
 
But admit it.  You are happy with the state of the art, and not having perfection, which you would find boring. 
 
So have a beer ( or a couple), enjoy the music, and look forward to what you can change next. 
 
Me, I'm thinking about paralleling two batteries to see what it sounds like, and whether it is better than the single Japan/China battery (of which I have 3 !), as soon as I am back home in about 2 weeks. 
 
But I am already happy with the way it sounds now, on 2.1, with a Japan/China battery, so I get to satisfy my further curiosity, with nothing to lose.  The DX90, paired with an IE800, I am finding pretty enjoyable.  More than any previous setup I have had before, so for that, I am happy with iBasso, and the DX90.

 

[headwhacker]

Did the LO sig change also?  By the way, I got rid of my Vorzuge, O2 still tops it.    I think I was over hyped with the Vorzuge, it sounded tad brighter, but not more detail and not as wide as O2.

 


I have not tried the LO yet. Only the HO using Roxanne, JH16 and SR325e. Will try the LO with O2 when I get the chance.

 

[jamato8]

  Me, I'm thinking about paralleling two batteries to see what it sounds like, and whether it is better than the single Japan/China battery (of which I have 3 !), as soon as I am back home in about 2 weeks. 
 
But I am already happy with the way it sounds now, on 2.1, with a Japan/China battery, so I get to satisfy my further curiosity, with nothing to lose.  The DX90, paired with an IE800, I am finding pretty enjoyable.  More than any previous setup I have had before, so for that, I am happy with iBasso, and the DX90.

Having paralleled two of the Samsung Korean made batteries, to my ear, while the Samsung is better than anything else I have tried, two are even better. 

 

[coolmingli]

I just got my DX90 today, after one hour of listen I found the sound too open, very airly, spacious, almost too mellow, bass could be a little tighter, while I like wide stage and depth, but the DX90 sound more like thru a surround sound processor, on the good side, the high is very smooth and extended, very crisp and excellent clarity, the mid layers is also very good, but I cant stand the vocal coming from top of my head and far away, i like the in your face type mid vocal, I think the DX90 will be nice with acoustic, jazz... music, so so for songs with vocal.

 

[nmatheis]

coolmingli: Which firmware is on your DX90?

 

[headwhacker]

I just got my DX90 today, after one hour of listen I found the sound too open, very airly, spacious, almost too mellow, bass could be a little tighter, while I like wide stage and depth, but the DX90 sound more like thru a surround sound processor, on the good side, the high is very smooth and extended, very crisp and excellent clarity, the mid layers is also very good, but I cant stand the vocal coming from top of my head and far away, i like the in your face type mid vocal, I think the DX90 will be nice with acoustic, jazz... music, so so for songs with vocal.

 


Try FW 2.1.0 with any Grado phones. I have a 325e and the mid is nothing like you described. In fact, the best mid I heard so far. The mids doesn't sound too distant with Roxanne as well.

 

[coolmingli]

FW is 2.0 right out of the box, I just downloaded the 2.1 but don't know how to upgrade, can someone enlighten me?