4.6 update and upgrade sound quality ?

one member didn’t like the 4.6 update with his nds. He went back to 4.4.   But some days after he updated again to 4.6 and found the update better than first 4.6 update. He kept the 4.6 finally.

David252 posted:

My NDS is one from 2012 when it was first released and I believe it is probably something to do with older NDS machines. It would be nice to see if there is a correlation between the age and issues.

Interesting David, my NDS is also 2012 vintage... hmmm...

As I mentioned in another posting, I'm over the hurdle of "bad sound" with my NSD4.6 -- now trying to embrace a rather 'different' sound signature.

enjoy

ken

David252 posted:

I am sorry but I think people read comments by other forum members and take them as gospel. I do not think that the changes to the functionality that is stated in the release notes (multiroom) would have such an effect on the sound.

I have writen software for over 20 years and as a developer the coding for the functionality and sound should be separate and I cannot believe would not be.

When you load the new firmware it is obvious that it is split up and programmed in three stages.

 

...

Unfortunately the sonic effect of changes of firmware isn't confined to the changes within the processing of data interaction of firmware modules through the data interfaces.  There is also the effect on sound through the interaction of the noise profiles and particularly switching noise modulating power supply lines and ground planes.  So even if the resulting data stream is identical but the processing is carried out in a different order the noise pattern in the supply lines and the ground planes will be different, this has the capacity to influence the DAC and the subsequent analogue stages, thus changing the sound, even though the data themselves are still identical.

I also have substantial experience of software design and code and also of simultaneously working in mixed mode (analogue and digital) electronic systems.

My NDS is also 2012, but I have found 4.6 a positive step. I don’t think this is a age of model issue. Still no idea what it is, I thought speakers,  but that hypothesis unravelled quickly.

my nds is from 2014 and found the update positive.

Huge posted:

David252 posted:

I am sorry but I think people read comments by other forum members and take them as gospel. I do not think that the changes to the functionality that is stated in the release notes (multiroom) would have such an effect on the sound.

I have writen software for over 20 years and as a developer the coding for the functionality and sound should be separate and I cannot believe would not be.

When you load the new firmware it is obvious that it is split up and programmed in three stages.

 

...

Unfortunately the sonic effect of changes of firmware isn't confined to the changes within the processing of data interaction of firmware modules through the data interfaces.  There is also the effect on sound through the interaction of the noise profiles and particularly switching noise modulating power supply lines and ground planes.  So even if the resulting data stream is identical but the processing is carried out in a different order the noise pattern in the supply lines and the ground planes will be different, this has the capacity to influence the DAC and the subsequent analogue stages, thus changing the sound, even though the data themselves are still identical.

exactly, the changes heard are exactly from the interaction of noise profiles from the execution and the execution timing of the  code with respect to specific hardware functions. The actual DSP algorithm that controls the response of the reconstruction filter hasn’t  changed in years... probably since launch, but it’s code execution timing has... it demonstrates why decoupling and noise management is so crucial in Naim digital devices and why there is even more focus on it on the ND555 and the new Uniti series.

The bit that I found interesting in talking with a certain Naim senior engineer was the extent the Naim team believe they can tune the code execution to affect a sonic signature in a reasonably deterministic way, but sometimes it does catch them out.. hardly surprising by the nature of what they are actually doing.

Simon

Simon-in-Suffolk posted:

Huge posted:

David252 posted:

I am sorry but I think people read comments by other forum members and take them as gospel. I do not think that the changes to the functionality that is stated in the release notes (multiroom) would have such an effect on the sound.

I have writen software for over 20 years and as a developer the coding for the functionality and sound should be separate and I cannot believe would not be.

When you load the new firmware it is obvious that it is split up and programmed in three stages.

 

...

Unfortunately the sonic effect of changes of firmware isn't confined to the changes within the processing of data interaction of firmware modules through the data interfaces.  There is also the effect on sound through the interaction of the noise profiles and particularly switching noise modulating power supply lines and ground planes.  So even if the resulting data stream is identical but the processing is carried out in a different order the noise pattern in the supply lines and the ground planes will be different, this has the capacity to influence the DAC and the subsequent analogue stages, thus changing the sound, even though the data themselves are still identical.

exactly, the changes heard are exactly from the interaction of noise profiles from the execution and the execution timing of the  code with respect to specific hardware functions. The actual DSP algorithm that controls the response of the reconstruction filter hasn’t  changed in years... probably since launch, but it’s code execution timing has... it demonstrates why decoupling and noise management is so crucial in Naim digital devices and why there is even more focus on it on the ND555 and the new Uniti series.

The bit that I found interesting in talking with a certain Naim senior engineer was the extent the Naim team believe they can tune the code execution to affect a sonic signature in a reasonably deterministic way, but sometimes it does catch them out.. hardly surprising by the nature of what they are actually doing.

Simon

 

I can fully appreciate this, but in essence we could be saying that firmware that does virtually nothing but perhaps play a digital input stream might actually sound worse than firmware that has a load of bells and whistles and wakes you up each day with your favourite radio choice!

Really shows the importance of isolating control software and electronics from the DSP/audio output sections if any kind of consistency can be maintained going forwards and even then perhaps not possible.

Alley Cat posted:

Simon-in-Suffolk posted:

Huge posted:

David252 posted:

I am sorry but I think people read comments by other forum members and take them as gospel. I do not think that the changes to the functionality that is stated in the release notes (multiroom) would have such an effect on the sound.

I have writen software for over 20 years and as a developer the coding for the functionality and sound should be separate and I cannot believe would not be.

When you load the new firmware it is obvious that it is split up and programmed in three stages.

 

...

Unfortunately the sonic effect of changes of firmware isn't confined to the changes within the processing of data interaction of firmware modules through the data interfaces.  There is also the effect on sound through the interaction of the noise profiles and particularly switching noise modulating power supply lines and ground planes.  So even if the resulting data stream is identical but the processing is carried out in a different order the noise pattern in the supply lines and the ground planes will be different, this has the capacity to influence the DAC and the subsequent analogue stages, thus changing the sound, even though the data themselves are still identical.

exactly, the changes heard are exactly from the interaction of noise profiles from the execution and the execution timing of the  code with respect to specific hardware functions. The actual DSP algorithm that controls the response of the reconstruction filter hasn’t  changed in years... probably since launch, but it’s code execution timing has... it demonstrates why decoupling and noise management is so crucial in Naim digital devices and why there is even more focus on it on the ND555 and the new Uniti series.

The bit that I found interesting in talking with a certain Naim senior engineer was the extent the Naim team believe they can tune the code execution to affect a sonic signature in a reasonably deterministic way, but sometimes it does catch them out.. hardly surprising by the nature of what they are actually doing.

Simon

 

I can fully appreciate this, but in essence we could be saying that firmware that does virtually nothing but perhaps play a digital input stream might actually sound worse than firmware that has a load of bells and whistles and wakes you up each day with your favourite radio choice!

 

Really shows the importance of isolating control software and electronics from the DSP/audio output sections if any kind of consistency can be maintained going forwards and even then perhaps not possible.

Makes me glad I have a relatively simple CDX2 digital transport into nDAC+PS555DR. I thought getting the network right was a pain, but it is under ones own control. Horrified to think the sonic signature can change with new firmware.  

Phil

Alley Cat posted: 

I can fully appreciate this, but in essence we could be saying that firmware that does virtually nothing but perhaps play a digital input stream might actually sound worse than firmware that has a load of bells and whistles and wakes you up each day with your favourite radio choice!

 

Really shows the importance of isolating control software and electronics from the DSP/audio output sections if any kind of consistency can be maintained going forwards and even then perhaps not possible.

Trouble is, that separation is almost impossible to achieve without complete separation of all the components of the system into different physical boxes; and that necessitates introducing compromised transport mechanisms between the boxes - e.g. using USB and/or S/Pdif (and with a lesser degree of sonic compromise - Ethernet).  In which case the problem becomes the limitation of the transport mechanisms between the boxes and you've just traded one problem for another that's just as bad (and may actually be worse).

The only answer would be to separate all the boxes and use a proprietary high bandwidth synchronous interface using a discrete clock line (the optimal location of this clock is something that, currently, I'm not at liberty to disclose) .  Even then the issues of propagation delay and jitter control becomes very difficult to handle.

Filipe posted:

Makes me glad I have a relatively simple CDX2 digital transport into nDAC+PS555DR. I thought getting the network right was a pain, but it is under ones own control. Horrified to think the sonic signature can change with new firmware.  

Phil

Then don't update the nDAC firmware!  

Huge posted:

Alley Cat posted: 

I can fully appreciate this, but in essence we could be saying that firmware that does virtually nothing but perhaps play a digital input stream might actually sound worse than firmware that has a load of bells and whistles and wakes you up each day with your favourite radio choice!

 

Really shows the importance of isolating control software and electronics from the DSP/audio output sections if any kind of consistency can be maintained going forwards and even then perhaps not possible.

Trouble is, that separation is almost impossible to achieve without complete separation of all the components of the system into different physical boxes; and that necessitates introducing compromised transport mechanisms between the boxes - e.g. using USB and/or S/Pdif (and with a lesser degree of sonic compromise - Ethernet).  In which case the problem becomes the limitation of the transport mechanisms between the boxes and you've just traded one problem for another that's just as bad (and may actually be worse).

The only answer would be to separate all the boxes and use a proprietary high bandwidth synchronous interface using a discrete clock line (the optimal location of this clock is something that, currently, I'm not at liberty to disclose) .  Even then the issues of propagation delay and jitter control becomes very difficult to handle.

Hmm you telling us all to upgrade to the NDS555 

Huge posted:

Filipe posted:

Makes me glad I have a relatively simple CDX2 digital transport into nDAC+PS555DR. I thought getting the network right was a pain, but it is under ones own control. Horrified to think the sonic signature can change with new firmware.  

Phil

Then don't update the nDAC firmware!  

It would be a shame. 4.11.8 is really good. ????

Chag -

Chag... posted:

Huge posted:

Filipe posted:

Makes me glad I have a relatively simple CDX2 digital transport into nDAC+PS555DR. I thought getting the network right was a pain, but it is under ones own control. Horrified to think the sonic signature can change with new firmware.  

Phil

Then don't update the nDAC firmware!  

It would be a shame. 4.11.8 is really good. ????

Chag -

I’ve asked earlier whether the nDAC firmware has been updated because I have not been notified. 

Regarding spdif, I can see that timing jitter can have random effects as all noise should be. The change in sonic signature does not seem to be a random effect though.

My observation on the DC1 digital interconnect is that it makes the music less enjoyable. I have a very nicely sounding Chord Clearway Digital IC. This does not sound like a timing problems though. My cheap AQ Forrest optical makes the tv sound exceptionally good.

Glad others think that maybe separates could be better.

Phil

Well all I can say is to repeat my 16 year old sons comment when he first heard the system after the update.

"Whats wrong with the hifi it sounds crap"

Out of the mouth of babes!

This was after resets and all suggested remedies.

We sat and listened to many of his favourate artists and his frown never left his face. As soon as I reinstalled 4.4 he smiled again.

I will stick with 4.4 until naim convince me to try again

David, I truly don’t understand why your system sounds so much worse with 4.6, when mine sounds so much better. I hope this can be sorted out somehow.

Yes, I too am still dumbfounded by the huge disparity in opinion of 4.6, particularly on the NDS. Unusually, views are diametrically opposed, with no one sitting on the fence. This does imply that the clear disparity in 4.6 SQ is less likely to be a room effect or interactions with different speakers IMHO. The vintage of the NDS also appears to be unrelated to 4.6 SQ.

I assume by the silence emanating from Salisbury (if silence can indeed emanate) that they too are perplexed, unless they are still looking in to this strange phenomenon. It would be nice however if one of the good folk at Naim could at least give some feedback.

I can only imagine how irritated I would be if the SQ died on my NDS, Naim's premium streamer (yes, the ND555 is not with us yet).

David252 posted:

All I can say is no! no! no!

I have an NDS with two 555dr power supplies and have today updated to version 4.6 and after hours of listening have had to go back to version 4.4.

I tried the factory reset but even though it did improve things a bit it was not enough.

What is going on this is not an improvement it sounds hollow the bass is tighter the lower frequencies are passable but the mid-range is missing. To add insult to injury the upper frequencies are not satisfying.

If this is the way the naim sound is going this will be the end of the line for me. 

I now can not have the other multiroom features unless i have the new sound. You should separate the functional update from the DSP update.

I am at the other end of the Naim spectrum to you with a Uniti2 but I am getting exactly the same effect - a deeper bass, wider soundstage but a 'hollowed out' midrange. It has gone from having 'PRaT' to just sounding 'HiFi' and uninvolving. The mid-range sounds as if someone is muffling it in comparison to the rest - it almost sounds a bit like 'noise-cancelling' is happening in this region.

Now I know my hearing is not as good as it used to be, and the change may just be showing this up, but I no longer listen to the system loud as it just sounds really uncomfortable - like my ears are straining to find the midrange. Closest I can describe this effect to is that of standing in an anechoic chamber where your ears ramp up the sensitivity to try and hear things. Very disconcerting...

I'd revert back to 4.4 but I can't be bothered with the hassle at the moment - in the meantime the Uniti2 is just being used as a really expensive soundbar for the TV where this effect is less noticeable. I'll revert to 4.4 in the Summer when I have a few spare hours to play with it again unless Naim come up with a solution to the problem in the interim.

I have been on NDS 4.6 for more than a week now and have had he NDS playing a random selection playlist from my collection of streamed music over the past few days while I worked. This was an attempt to let the music 'wash over me' sort of. 

the nastiness that I experienced the first time I updated is definitely gone -- so that's good -- I was able to have long listening sessions without wincing in discomfort.

I can definitely pick up  enhanced clarity -- some additional musical lines in the mix -- and this is very apparent for choral music for example.

the presentation seems more 'animated' - you could say 'brightly lit'. it is definitely very 'impressive' but feels like something new that I have to learn to like and live with long term. Old age? Personal preferences, I guess... I found myself missing what i was used to with 4.4!!!

just as a last check, I will revert to 4.4 again one more time next week --and see what I feel like after that. Unlikely I will bore you with further reports.

I will be playing a lot of vinyl this weekend --- for a break. NAT01 sounds familiarly fab in my office as I work.

enjoy

ken

Hello, i am new here. I have a Naim 172xs and updatet to 4.6. Play with Spotify and Tidal My experience is that the resolution and soundstage changed better but i am dissapointed about the bass. It is too tight for me. This is not what i am used about Naim. Before the update it was much more musical /swing analog sound. Naim please bring the bass back

Strange thing is that 4.6 with Spotify is worse (thinner/less swing etc) but with Tidal i'm hearing a very good upgrade. 

Penarth Blues posted:

David252 posted:

All I can say is no! no! no!

I have an NDS with two 555dr power supplies and have today updated to version 4.6 and after hours of listening have had to go back to version 4.4.

I tried the factory reset but even though it did improve things a bit it was not enough.

What is going on this is not an improvement it sounds hollow the bass is tighter the lower frequencies are passable but the mid-range is missing. To add insult to injury the upper frequencies are not satisfying.

If this is the way the naim sound is going this will be the end of the line for me. 

I now can not have the other multiroom features unless i have the new sound. You should separate the functional update from the DSP update.

I am at the other end of the Naim spectrum to you with a Uniti2 but I am getting exactly the same effect - a deeper bass, wider soundstage but a 'hollowed out' midrange. It has gone from having 'PRaT' to just sounding 'HiFi' and uninvolving. The mid-range sounds as if someone is muffling it in comparison to the rest - it almost sounds a bit like 'noise-cancelling' is happening in this region.

Now I know my hearing is not as good as it used to be, and the change may just be showing this up, but I no longer listen to the system loud as it just sounds really uncomfortable - like my ears are straining to find the midrange. Closest I can describe this effect to is that of standing in an anechoic chamber where your ears ramp up the sensitivity to try and hear things. Very disconcerting...

I'd revert back to 4.4 but I can't be bothered with the hassle at the moment - in the meantime the Uniti2 is just being used as a really expensive soundbar for the TV where this effect is less noticeable. I'll revert to 4.4 in the Summer when I have a few spare hours to play with it again unless Naim come up with a solution to the problem in the interim.

I talked to my dealeer yesterday to see what his opinion was but he could not help me as ne did not have an NDS any more. He did say he would ask around his customers to see what they think.

I can not understand that there has not be any kind of response from naim.

I have liked naim for a long time because the music is involving and not HIFI like. So I hope the resulting upgrade on my NDS is no intensional. 

I have decided to try the 4.6 again next week.

Someone scrolling through this post for the first time and not reading all 500 posts would get the impression that 4.6 has been a major flop in terms of sound quality. I think the reality is that most people, myself included, are very happy with the update / upgrade but as is usually the case it’s the users with negative experiences that shout loudest. Nothing wrong with that, just the way it is. 

Hopefully this does put naim off releasing future DSP updates.  Perhaps a switchable DSP selector would be the safest bet.

I would love to provide feedback to naim via an official online survey. 

wrc posted:

Someone scrolling through this post for the first time and not reading all 500 posts would get the impression that 4.6 has been a major flop in terms of sound quality. I think the reality is that most people, myself included, are very happy with the update / upgrade but as is usually the case it’s the users with negative experiences that shout loudest. Nothing wrong with that, just the way it is. 

Hopefully this does put naim off releasing future DSP updates.  Perhaps a switchable DSP selector would be the safest bet.

I would love to provide feedback to naim via an official online survey. 

Yes but the problem I have is that naim do not mention anything about the DSP in the release notes, they only talk about functionality. It would not be so bad if they had said this will change the sonic character of your player, so you could make an informed decision whether to install it.

wrc posted:

Someone scrolling through this post for the first time and not reading all 500 posts would get the impression that 4.6 has been a major flop in terms of sound quality. I think the reality is that most people, myself included, are very happy with the update / upgrade but as is usually the case it’s the users with negative experiences that shout loudest. Nothing wrong with that, just the way it is. 

Hopefully this does put naim off releasing future DSP updates.  Perhaps a switchable DSP selector would be the safest bet.

I would love to provide feedback to naim via an official online survey. 

Just reread my post and I noticed a typo.  It should read:

hopefully this does NOT put naim off releasing future DSP updates!

David252 posted:

wrc posted:

Someone scrolling through this post for the first time and not reading all 500 posts would get the impression that 4.6 has been a major flop in terms of sound quality. I think the reality is that most people, myself included, are very happy with the update / upgrade but as is usually the case it’s the users with negative experiences that shout loudest. Nothing wrong with that, just the way it is. 

Hopefully this does put naim off releasing future DSP updates.  Perhaps a switchable DSP selector would be the safest bet.

I would love to provide feedback to naim via an official online survey. 

Yes but the problem I have is that naim do not mention anything about the DSP in the release notes, they only talk about functionality. It would not be so bad if they had said this will change the sonic character of your player, so you could make an informed decision whether to install it.

I really don't know why you have been disappointed with firmware v4.6. The more music I listen to since the update, the better I think it is.