'Analog Dinosaur' succumbs to the joys of FBA : enter Raspberry Pi 2/IQ-Audio DAC+!!

[Marco]

Anyway, back to audio...

Indeed. Guys, please don't turn this thread into a snooze-fest about computer programming lingo!

Was a busy boy yesterday, with one thing and another, so will comment later on recent posts. Glad to see you're getting stuck in Gordon

Marco.

[NRG]

The whole the point of projects like Volumio, squeezeplug, picoreplayer and others is to allow the user, who knows little or no linux, to use their Rpi as an audio player with relative ease. No doubt users will pass their verdict on the success, or otherwise, of these various projects meeting their declared aims.

I would imagine the relative merits and ease of software development in a Linux environment would be of little interest to users unless it was so difficult as to have actually held back meaningful development, which is clearly not the case.

I must admit I didn't spot any 'Linux' when installing either Volumio or SqueezePlug, both where straight forward and SqueezePlug even includes a detailed step by step guide on how to configure it, right there on the the opening screen.

I have delved under the hood to perform some tweaks that are not really necessary for normal operation on either of these two solutions but it was hardly difficult in any case....However, if you don't understand computers then I can see the reason why somebody wouldn't want to go the DIY route because if there is an issue you are a bit stuck!

[Halfway Tree]

Q) Hardware volume control
A) Right, that's a good topic and I'll try and address it simply.
Some DAC chips output at a specific fixed line level (TI PCM5102 plus many more), others have variable output, the TI PCM5122 as used on Pi-DAC+, for example. The variable output is controlled on-chip through a volume register. Write value "a" into the volume register and you get max volume "A", value "b" and you get volume "B". How this volume change is handled on-chip is deep in the device design. This volume is delivered to the Linux system through the Linux ALSA driver and the Linux ALSA utilities (alsamixer for example) will either pass this variable volume to the DAC or will emulate the volume adjustment by removing bits from the data BEFORE it gets to the DAC. For best SQ you want ONE volume control over the whole system (IMHO) and you don't want to start removing bits of real data to emulate low to high volume - you're throwign away data before it gets converted.

This is why we suggest that HW volume control is better than "emulated" software volume control.


Gordon

Gordon, firstly thanks for answering the volume control question which I'd asked early on in this thread.

I looked at the RPI+IQDac some time ago, but as I need 3 dacs as I have an active system - one for each frequency band I couldn't see a way to implement it.

Having used both digital and analogue volume controls I think it's as much about how well they are implemented than about the method. Currently I'm using a digital volume control (JRiver - 64 bit) as I think it is better than my other option of using a multi-channel analogue volume control (MSB) just before the power amps. It seems to be quieter and there is less cable clutter.

However, an RPi and multiple IQDacs would be very appealing for an active system if I could find way to implement it....

Gordon

[montesquieu]

Does it?, I'm not convinced IME there are notable differences between DACs



Yes, the streaming part is important as is functionality, however the beauty of an integrated DAC, if it is implemented well is that it eliminates the USB or SPDIF interface and associated issues…like clock recovery and (hopefully) power line noise. One reason, IMHO, the external DAC market has sprung up is the poor implementation of internal DAC's in CD players and early streamers due to cost restraints and and the need to hit a market price point…an external DAC dangles the carrot of improved sound quality but I see nn reason why equivalent sound can't be achieved with a well designed internal DAC….one fly in the ointment could be the requirement for a quality power supply.

The other benefit, for me, is less real-estate on the shelf….

The differences between DACs (abilities with hi-res file formats apart) are primarily the same as the differences between preamps, ie in the output stage. Why on earth wouldn't it be? My Audio Note DAC (2.1x upgraded to 4.1x spec) has pretty much and equivalent ouput stage to an Audio Note M5. It's a total nonsense to suggest that the op amp chips in a one-piece DAC board like this do the same thing as this or any other high-end DAC. Don't be daft.

I remember all the fun people had when DIY DAC boards from China appeared on ebay, great sound for not very much. (I ran one myself for a bit with a SRPP tube output stage). But who is still using them? It was a fad and people realised eventually that DIY fun as they were, there were all sorts of limitations when compared to 'real' kit. This looks to be exactly the same proposition, with a tiny computer attached to do the streaming bit.

With a digital signal there is nothing to be gained from an integrated streamer and DAC if all you are doing is passing data. Interface issues pale into insignificance anyway if a) you do it right as Audio Note do with buffers/transformers on SPDIF/AES-EBU input and output, or you can provide USB output to a DAC's Asynchronous USB input, and b) compared to preamp performance, see above.

As I said before the DAC element of this is a red herring in my book. The interesting part is a useful, functional replacement for the Squeezebox (which had its own DAC too, remember, though most people used it as a transport).

All good fun I'm sure but I do detect a bit of collective forum mania coming on. Sorry about the water.

[Gordon@IQ]

Montesquieu, take your point.

I think the difference here is that there is more control of the features and functionality that can be implemented without the need for big CE companies.

G

[Gordon@IQ]

However, an RPi and multiple IQDacs would be very appealing for an active system if I could find way to implement it....

It's possible to have 4 PCM5122 on the I2C bus, each could be made to be a unique sound card and then each could be EQ'd too. It would be a hack and need some specific Linux Kernel development but worth a play. I'll add it to my todo list :-)

[Marco]

Hi Tom,

As I said before the DAC element of this is a red herring in my book.

It's a total nonsense to suggest that the op amp chips in a one-piece DAC board like this do the same thing as this or any other high-end DAC. Don't be daft.

Having used and familiarised myself with the Pi/IQ-Audio DAC combination now for over a week, I couldn't disagree more!!

Not sure how you can say that, really, when others and I have already reported that this little £32 wonder has outperformed some Naim gear, the DAC inside a £6k Rega Isis CDP, and has given my (circa £5k) highly-modified Sony X-777ES/DAS-R1 combo a serious run for its money - and indeed continues to do so...

How much more of a bargain do you want, or proof that this little thing kicks serious ass??

For me, the IQ-Audio DAC+ is an integral part of the magic of this combination. The truly superb sounds it produces are a fundamental reason why I've finally embraced the joys of FBA, after many years of not finding a solution (at any price point) that satisfies my ears. Like I said before, the Pi/IQ Audio DAC+ is bonkers good for the money, and many, many multiples of money more!!

Marco.

[Marco]

It's possible to have 4 PCM5122 on the I2C bus, each could be made to be a unique sound card and then each could be EQ'd too. It would be a hack and need some specific Linux Kernel development but worth a play. I'll add it to my todo list :-)

Gordon, any idea when the Spotify error in Volumio is going to be fixed?

Duncan visited me yesterday, armed with some Linux code (or at least I think that's what it was), which he thought would fix matters, but despite keying in my Spotify username and password, ensuring that both were correct, I still can't access Spotify within the Volumio user interface (in effect so that it appears beside my NAS box and WEBRADIO, when hitting 'Browse', thereby allowing me to select it for playback).

Marco.

[NRG]

The differences between DACs (abilities with hi-res file formats apart) are primarily the same as the differences between preamps, ie in the output stage. Why on earth wouldn't it be? My Audio Note DAC (2.1x upgraded to 4.1x spec) has pretty much and equivalent ouput stage to an Audio Note M5. It's a total nonsense to suggest that the op amp chips in a one-piece DAC board like this do the same thing as this or any other high-end DAC. Don't be daft.

I remember all the fun people had when DIY DAC boards from China appeared on ebay, great sound for not very much. (I ran one myself for a bit with a SRPP tube output stage). But who is still using them? It was a fad and people realised eventually that DIY fun as they were, there were all sorts of limitations when compared to 'real' kit. This looks to be exactly the same proposition, with a tiny computer attached to do the streaming bit.

With a digital signal there is nothing to be gained from an integrated streamer and DAC if all you are doing is passing data. Interface issues pale into insignificance anyway if a) you do it right as Audio Note do with buffers/transformers on SPDIF/AES-EBU input and output, or you can provide USB output to a DAC's Asynchronous USB input, and b) compared to preamp performance, see above.

As I said before the DAC element of this is a red herring in my book. The interesting part is a useful, functional replacement for the Squeezebox (which had its own DAC too, remember, though most people used it as a transport).

All good fun I'm sure but I do detect a bit of collective forum mania coming on. Sorry about the water.

Don't be daft?! I suggest you get down of that high horse as the rarified air up there seems to be affecting your thinking!

DAC chip technology moves on at a rapid rate you need to reset your thinking from the old TDA151A days. Modern DAC chips give outstanding performance when correctly implemented and offer some great features. An integrated DAC close to the data stream utilizing the I2S bus with its separate serial line and clock has got to be better than encoding the PCM stream for SPDIF and worse USB. Why on earth wouldn't it be? Just because Audio Note have used an isolation transformer on the SPDIF interface does not make it perfect, its laughable to think that as you still need to recover the clock and you have issues dealing with ground plane noise. Utilising SPDIF / USB adds complexity and is unnecessary if you can use I2S with a well implemented DAC, it needn't cost Audio Note prices....the only issue as I mentioned before is the dependency on the power supply...

[Marco]

DAC chip technology moves on at a rapid rate you need to reset your thinking from the old TDA151A days. Modern DAC chips give outstanding performance when correctly implemented and offer some great features. An integrated DAC close to the data stream utilizing the I2S bus with its separate serial line and clock has got to be better than encoding the PCM stream for SPDIF and worse USB. Why on earth wouldn't it be?

+1 (with bells on, although the Sony DAC, featuring the use of TDA1541s, still has a very entertaining and musically valid sound of its own).

Anyway, a couple of observations, based on recent experimenting and further listening to the Pi/IQ-Audio DAC combo:

First of all, the ethernet cables that Gazjam recommended further back in this thread, shown here: http://www.ebay.co.uk/itm/1717289032...%3AMEBIDX%3AIT

...unquestionably sound significantly better than the much thinner Maplins Cat5e cable I had been using before. Technically, why that is, I don't know. I suspect it's got something to do with the shielding used in the former.

However, there is a marked difference in the sonic presentation of both ethernet cables (bearing in mind that the Duronic ones are actually £5 less than the Maplins cables I bought, so definitely not 'overpriced foo'). The Duronic cables tighten bass and open up the midrange and top-end, allowing markedly more detail through. The effect is so obvious, one can hear it instantly.

As such, I now use these cables from my NAS box to the router, and from the router to the Pi. Highly recommended, and at a bargain price! Nice one, Gary

The other thing is that I must retract my earlier statements about the IQ-Audio DAC+ sounding better with no active preamp in line. It turned out that the reason why I wasn't getting the proper results with my Croft was because I hadn't set the 'gain' high enough in Volumio, when adjusting it, before disabling the Hardware.

Once I had set the controls to 96 (the sweet spot in my system), and had successfully applied that setting (disabling the Hardware), the sound is stunningly good now through the Croft, and significantly better than using the Volumio digital interface alone to adjust listening levels, which I guess is as it should be.

I feel, however, that the process in Volumio should be explained more clearly for newbies, in terms of how to optimally partner the Pi/IQ-Audio DAC with an active preamp, as it is far from obvious, and thus mistakes are easily made! That said, things have now never sounded better

Marco.