[Jon773]

So I am on vacation and therefore have to be away from my home headfi setup for a whole week. This is quite a long time when you consider how used to listening to music at home everyday you get when 'working from home'. One of the very few benefits of the pandemic...

I was always planning to look into music servers once i had my core system sorted out. I figured it would be one of the last things on the list to consider.

Now i am planning on demo'ing servers as i think for the money the improvements could be fairly minor given where i am at the moment; but as i have learned in this hobby of ours, it can sometimes really surprise you.

At the moment i stream music in one of two ways into my Bartok.

1) Tidal > English Electric 8Switch > Bartok
2) QNAP NAS (with SSD cache and Linear PSU) > English Electric 8Switch > Bartok

The change up to the 8Switch was quite noticeable; which surprised me, given how controversial audio grade switches are.

What i am interested in learning from the community is what your experience are of changing from a NAS to using a music server from companies like Innuos, Melco, Aurender in your setups. Even better if you are feeding a Bartok. Changing to a dedicated server i would be able to cut out the 8Switch from feeding the Bartok and remove a step in the chain, which i hope would be a good thing. I would also remove the NAS, which even though its driven by a LPSU, will still be noisier than a decent server (I would expect).

 

[bluecar]

I have to say I find this perspective really strange. If you can get a serious bump in performance through a system via a major firmware upgrade, how can that not be better than buying another new box.

dCS design their systems to have a long life and be firmware upgradable; it’s a major selling point.

Other companies bring out a version 2.1 of the box a year later; which would concern me far more than a firmware release.

Chord are slow to replace their boxes, which is a positive. dCS very slow to replace a box but they do a great job with firmware; as I am sure most Rossini owners will tell you.

This can cut both ways - it depends on how and what dCS upgrade - if they can stay toe-to-toe in terms of performance over a 7 year period then they are on to a winner - most of the progress in digital has been in core processing, so it seems feasible that they can stay at the front through FPGA code upgrades. Given that dCS can effectively re-write the core processor logic for their chips, it's not beyond the bounds of possibility that they can deliver sonic step-changes with FPGA firmware upgrades.

Of course, when I hear 'forward upgrade path', I think 'fault correction mechanism', but I'm just a cynical engineer

 

[bluecar]

The change up to the 8Switch was quite noticeable; which surprised me, given how controversial audio grade switches are.

I've got the 8-port netgear switch it was based on - let's do some A/B testing when you get back off hols, and see if it's the switch, or the LPSU.... Also, you're not getting your Stellia's back...ever.

 

[Jon773]

I've got the 8-port netgear switch it was based on - let's do some A/B testing when you get back off hols, and see if it's the switch, or the LPSU.... Also, you're not getting your Stellia's back...ever.

 

[Articnoise]

So I am on vacation and therefore have to be away from my home headfi setup for a whole week. This is quite a long time when you consider how used to listening to music at home everyday you get when 'working from home'. One of the very few benefits of the pandemic...

I was always planning to look into music servers once i had my core system sorted out. I figured it would be one of the last things on the list to consider.

Now i am planning on demo'ing servers as i think for the money the improvements could be fairly minor given where i am at the moment; but as i have learned in this hobby of ours, it can sometimes really surprise you.

At the moment i stream music in one of two ways into my Bartok.

1) Tidal > English Electric 8Switch > Bartok
2) QNAP NAS (with SSD cache and Linear PSU) > English Electric 8Switch > Bartok

The change up to the 8Switch was quite noticeable; which surprised me, given how controversial audio grade switches are.

What i am interested in learning from the community is what your experience are of changing from a NAS to using a music server from companies like Innuos, Melco, Aurender in your setups. Even better if you are feeding a Bartok. Changing to a dedicated server i would be able to cut out the 8Switch from feeding the Bartok and remove a step in the chain, which i hope would be a good thing. I would also remove the NAS, which even though its driven by a LPSU, will still be noisier than a decent server (I would expect).

Switching to a dedicated server doesn’t appear to make a good audiophile switch like the Melco S100 unnecessary, not even with *Taiko Audio Extreme or *Innuos Statement.

There are many different things that can affect the sound and there are also many different ways to connect the NAS, server with the rest of the audio system. Some people think that it is better to place the NAS in another room and connect it to a switch and that it sounds much better than having and playing the files stored in the server. There are some who seem to think the opposite as well, but it may be due to the type of server.

If you already have an audio switch and NAS, I would recommend you keep them for now and try what you think sounds best when you get a dedicated server.

*I have no Bartok or Statement/Extream so please take my advice with a pinch of salt.

 

[adrianm]

Of course, when I hear 'forward upgrade path', I think 'fault correction mechanism', but I'm just a cynical engineer

Exactly. Cynical software developer here.

From what I have read about the Rossini firmware releases the move to 2.0 was a decent bump up. The dCS use FPGA which I believe are programmable...

From what i've read about it, it was mostly fixing some roon issues they had in the first place.
They are programmable, they can do whatever they want but software development is expensive. I don't do low level programming (as in hardware related) , but the principles remain the same as in high level languages, even more so. If the variables (components being one of them, though they're actually constants here) stay the same, why would my algorithms be better the second time i take a swing at the problem? Improving on what's already there is harder than starting from scratch and doing that is extremely expensive and makes no financial sense. It takes years and by that time there's already newer hardware out there. Better variables to play with.
The improvements are mostly bug fixing, which is actually good for them because it allows them to bring a product to market faster and worry about fixing minor issues later.

 

[Jon773]

Exactly. Cynical software developer here.


From what i've read about it, it was mostly fixing some roon issues they had in the first place.
They are programmable, they can do whatever they want but software development is expensive. I don't do low level programming (as in hardware related) , but the principles remain the same as in high level languages, even more so. If the variables (components being one of them, though they're actually constants here) stay the same, why would my algorithms be better the second time i take a swing at the problem? Improving on what's already there is harder than starting from scratch and doing that is extremely expensive and makes no financial sense. It takes years and by that time there's already newer hardware out there. Better variables to play with.
The improvements are mostly bug fixing, which is actually good for them because it allows them to bring a product to market faster and worry about fixing minor issues later.

Errrrm maybe read this article.... https://www.stereophile.com/content/dcs-rossini-v20-firmware-upgrade

 

[adrianm]

Errrrm maybe read this article.... https://www.stereophile.com/content/dcs-rossini-v20-firmware-upgrade

I did ,that's the one i was refering to :
" dCS's new Roon Ready module addresses that earlier deficiency. "
Bug fixing.
As for the other improvements i think it's just trickle down code from the more expensive Vivaldi range. The new maps "adapted" to Rossini.
It's great that you're getting constant upgrades, i'm just arguing that they already have the code in the more expensive ranges and it's trickling down slowly. Everyone does it though, the VW group is trickling down tech from Porsche to Audi to VW to Skoda, Nvidia and Intel are selling the same gpu more or less at vastly different price points, but with locked cores to differentiate between tiers. Essentially limiting performance.
Everyone does it for a reason, it works. People who pay for Vivaldis are essentially subsidizing your Bartoks. Which would be fine if the Bartok was a lot cheaper than what it is. But to each his own.

 

[Pastwa]

Which would be fine if the Bartok was a lot cheaper than what it is

Bartok actually is much cheaper then people think, especially if you break it down:

- good quality streamer,
- outstanding DAC section,
- very decent class A headphone module with a balanced output,

All in one elegant box solution, saving therefore desk space and what's more important, all the additional power and signal cables, which can be a substantial cost when going for a two/three box option.

For me dCS Bartok it is sonically such a complete package (realistic weight to each note, hence the music has a proper body/mass, plus the sense of space), even with the original software, that personally I wouldn't care much about any future upgrades, but if they are released, then great.

 

[endre83]

I did ,that's the one i was refering to :
" dCS's new Roon Ready module addresses that earlier deficiency. "
Bug fixing.
As for the other improvements i think it's just trickle down code from the more expensive Vivaldi range. The new maps "adapted" to Rossini.
It's great that you're getting constant upgrades, i'm just arguing that they already have the code in the more expensive ranges and it's trickling down slowly. Everyone does it though, the VW group is trickling down tech from Porsche to Audi to VW to Skoda, Nvidia and Intel are selling the same gpu more or less at vastly different price points, but with locked cores to differentiate between tiers. Essentially limiting performance.
Everyone does it for a reason, it works. People who pay for Vivaldis are essentially subsidizing your Bartoks. Which would be fine if the Bartok was a lot cheaper than what it is. But to each his own.

Do you go to all threads that you do not plan to own to tell there why I mean @adrianm you do not ask any questions from owners, have any listening experience to share etc. You are the guy driving an old Volkswagen Polo telling the guys in an Audi A6 thread that you are going to buy a VW Arteon very soon..

Your posts do not add any value to the thread, since you appeared here there is not more information to the reader imho.

 

[Jon773]

Bartok actually is much cheaper then people think, especially if you break it down:

- good quality streamer,
- outstanding DAC section,
- very decent class A headphone module with a balanced output,

All in one elegant box solution, saving therefore desk space and what's more important, all the additional power and signal cables, which can be a substantial cost when going for a two/three box option.

For me dCS Bartok it is sonically such a complete package (realistic weight to each note, hence the music has a proper body/mass, plus the sense of space), even with the original software, that personally I wouldn't care much about any future upgrades, but if they are released, then great.

and its also got a great internal clock and upsampling capability.

 

[adrianm]

Do you go to all threads that you do not plan to own to tell there why I mean @adrianm you do not ask any questions from owners, have any listening experience to share etc. You are the guy driving an old Volkswagen Polo telling the guys in an Audi A6 thread that you are going to buy a VW Arteon very soon..

Your posts do not add any value to the thread, since you appeared here there is not more information to the reader imho.

I'm actually driving a new A5 irl but the "Arteon" is here
A6's are for old men

 

[endre83]

I'm actually driving a new A5 irl but the "Arteon" is here
A6's are for old men

good reply

 

[adrianm]

I was just trying to clear up some misconception about the software part as that's my actual job, but if you consider it to be of 0 value, fair enough, i'll let you back to "audiophile" stuff.

 

[Jon773]

Only time will tell if us Bartok owners are treated to a nice firmware upgrade that is more than a few minor enhancements. We don't need one, lets be honest the Bartok is a great piece of kit, but do we 'want' one... oh yeah.

I remain positive that this is a definite possibility for the future development of the Bartok.