[AKSA]

John,

Difficult to put a price on it just yet as we are still finalizing chip and pcb prices.  Probably somewhere between $US800 and $950.  PLEASE don't hold me to it, there might be a meltdown in a chip factory somewhere, who knows.....

Cheers,

Hugh

[EchiDna]

I still don't understand why it can't just be read from disk and buffered....

I do just that with my media server actually pretty easy to implement...

[Tinker]

I still don't understand why it can't just be read from disk and buffered....

It can be, and considering that drives can be made whisper quiet, good buffering should take one close to theoretical limit of performance for that medium. It took me ages to figure out why this straightforward fix so often failed to deliver the results it promises.

There are some important practical considerations to address when using such buffers, however, and these will be the subject of the promised whitepaper parte the seconde. As it is most servo control/DAC chips include a small buffer anyhow. The failure of this to fully fix the problem is mostly down to the way it has been implemented and component interactions.

More later.

T.

[tg3]

The interconnection will be standard SPDIF; no additional interlinking of clocks will be necessary.

Does an I2S interconnect make any sense? I2S seem to be preferred if you roll your own transport and embed the DAC in the same box.

DIY CD transport using Philips CD-PRO2

[AKSA]

tg3,

I'll leave this one for Ben......!

Hugh

[Tinker]

The interconnection will be standard SPDIF; no additional interlinking of clocks will be necessary.

Does an I2S interconnect make any sense? I2S seem to be preferred if you roll your own transport and embed the DAC in the same box.

DIY CD transport using Philips CD-PRO2

Hi tg3, it does make sense, but...

Reading the Philips literature suggests that I2S was never intended to be used as a between-box format. As you rightly say, it is a good within-box system, and is inherantly lower-jitter than SPDIF. It IS the method, by which all the parts of the AKSADAC talk to eachother.

If you just suck the I2S out of the box there is information like sample rate, emphasis, etc. which has to be supplied separately, although in the case of CD most of the relevant values can be safely assumed. Bit depth, is not a real issue with I2S, as it was deliberately designed to deal with different word lengths fairly gracefully. Plus I2S needs 3 wires, not one... S/PDIF embeds all the information and a clock in a data stream so that totally alien devices can talk to eachother.  If you roll your own one-box jobbie, no problem with I2S. In fact it is preferred in that instance, but there are other issues like power supply modulation.

I notice the CD-PRO2 reference. The CD-PRO2 is a great transport (possibly the best!), although teac's VRDS developments deserve watching.  The 10501 board on the 'PRO2 has an I2S out and also a 16.9344MHz clock. DAISy, however, advise against trying to inject your own external clock, and having looked at the application I can understand why. The I2S is designed for in-box linkage and the 'PRO has a loopback port for developing these applications.

Incidentally, did anyone else notice that the CD-PRO2 output is reverse phase, or did I misread the TDA1305 spec?


I suppose this is an oblique way of saying, S/PDIF could be improved upon, but it is the only solid, widespread consumer standard. With judicious treatment exceptional performance can be obtained from S/PDIF linked systems. Hence the obvious choice for the most flexible, biggest bang-for-the-buck DAC that will work without drama on just about any tranport. The AKSADAC could, in principle,  be modded (as would the CD player need to be!) for I2S, but would not realise any worthwhile performance gain and could potentially cause a lot of headaches.

Any help?

T.

[tg3]

Thanks, Tinker.

Would it be too expensive to leave an I2S tap on the board for the (few) who might want to use it? I am considering adding the AKSADAC to the PRO2.



Best regards,

[SamL]

Hi Tinker,
 I'm sure the $1200 price tag of that 1GB flash will be dropping soon too.  Specifically, the question would be how soon is soon?
quote]

Not sure which currency you refering to as Transcend 1GB JetFlash-A USB Flash Drive is selling at NZ$556. Transcend 2GB JetFlash-A USB Flash Drive is NZ$1300.
It should be cheaper on other country.
http://www.pp.co.nz/

Sam

[EchiDna]

thats very cheap! those same drives are more in S$ here!!

*shocked*

[Tinker]

Hi Tinker,
 I'm sure the $1200 price tag of that 1GB flash will be dropping soon too.  Specifically, the question would be how soon is soon?
quote]

Not sure which currency you refering to as Transcend 1GB JetFlash-A USB Flash Drive is selling at NZ$556. Transcend 2GB JetFlash-A USB Flash Drive is NZ$1300.
It should be cheaper on other country.
http://www.pp.co.nz/

Sam

Things do move in leaps and bounds!

The first 128MB cost about $120 this time last year. I just got one recently (August) for $65, but the quote for 1GB at $1280 was about March. This is almost worth doing at this point! The JetFlash is cool. The open format is fairly flexible. It shouldn't be too hard to integrate one with a microprocessor... *sigh* Another programming task... CD drive ATAPI. Flash, USB-1. Hmn...

T.

[Oz_Audio]

Hi Ben

It would be good to have the DAKSA with the capability of normal transport and the facility to have the Flash Card as well.  Then the owner could buy the card at a later time to keep the price as low as possible, but still be able to use all the benefits when required.

It would also be good the have a comparison between the 2 delivery systems with all other things being equal.

Mark

[Tinker]

Hi Ben

It would be good to have the DAKSA with the capability of normal transport and the facility to have the Flash Card as well.  Then the owner could buy the card at a later time to keep the price as low as possible, but still be able to use all the benefits when required.

It would also be good the have a comparison between the 2 delivery systems with all other things being equal.

The flash comment was a  reference was an ongoing project with a uber- geek buddy of mine who does a lot of DSP. Such a system would be a chewing-gum and string affair at this point, even if we ever build it. Plus, if it were a kit,  the hardware and software required would treble the cost of the kit. I'll let you know if it ever does happen and we can shoot it out against a moving transport. Incidentally I have heard a similar comparison in a recording studio using a portable RAM recorder vs hard-disk system. I couldn't pick it, but then the jitter was super-low because of top-quality synching.

Be assured that the AKSADAC will deliver astronomically low jitter nonetheless.

T.

[SamL]

thats very cheap! those same drives are more in S$ here!!

The price NZ$556 include the 12.5% GST. If you order from oversea your get another $69.50 off. Never expect cheaper price in NZ but I am not complaining.  

BTW Tinker, any reason not going directly to USB-2 since this is for the future?

Cheers,
Sam

[Tinker]

thats very cheap! those same drives are more in S$ here!!

The price NZ$556 include the 12.5% GST. If you order from oversea your get another $69.50 off. Never expect cheaper price in NZ but I am not complaining.  

BTW Tinker, any reason not going directly to USB-2 since this is for the future?

This is cheap indeed! Another six months and...

No reason not to use USB-2, certainly not an audio one. USB-1 always was a touch slow at certain things... If you were going to develop this application yourself, I would go for USB-2. Mostly it would depend on how easy it would be to get your hands on the right software to drive the thing. Developing drivers like that from scratch is very time-consuming.

T.

[Agisthos]

Could we hook up something like a computer cdrom as the transport, commected to a ram/hardrive and purely software processing?

I seem to recall that Redgum cd players use just a cdrom transport. Cheap but you lose the remote features.

[Raj]

Hi Hugh & Tinker,

How important is wiriing from transport to dac in terms of induced jitter - does using solid core silver wire from transport to dac help in reduction?

Also there's been some hoo-haa about using green or blue led's shining over the surface of a disc, often with insufficient evidence of improvement, are the arguments for using led's valid at all from a technical stance?


Thanks
Raja

[Tinker]

Hi Hugh & Tinker,

How important is wiriing from transport to dac in terms of induced jitter - does using solid core silver wire from transport to dac help in reduction?

Also there's been some hoo-haa about using green or blue led's shining over the surface of a disc, often with insufficient evidence of improvement, are the arguments for using led's valid at all from a technical stance?

Wiring is important, but silver vs copper is not really an issue. The characteristic impedence is the only thing that really matters. It should be 75Ohm for S/PDIF and 110 for AES/EBU. This basically constrains the way that pulses get deformed in their travels from A to B.

Not sure about the "shining." The laser diodes in CDs and DVDs have particular colours so that their wavelengths match some important property pf the pits to be read. Blue lasers can read tighter pitched tracks than red ones. This is not really a problem: CDs have a set pitch (about 1.7 microns from memory) and the tracking stability has more to do with the optical pickup arrangment, e.g. those fancy 3-beam systems trick software.

T.

[Tinker]

Could we hook up something like a computer cdrom as the transport, commected to a ram/hardrive and purely software processing?

I seem to recall that Redgum cd players use just a cdrom transport. Cheap but you lose the remote features.

This has been tried and works quite well. The main issue is finding a CD-ROM that has a real S/PDIF out. There are quite a few. All you need to do then is create a stand-alone ATAPI interface driver. Once you have this, you can build on a remote. From a DIY viewpoint, you are probably better off buying a fully functional CDM-12 module (power supply on board!) or the CD-PRO2 (you need to build a power supply with four different voltages) already mentioned. It would be much quicker and cheaper!

IMHO using a harddrive pretty much scotches the advantage of this approach since a disk access must go via a disc controller, bus, bus router etc. At this point you virtually need an O/S to make things talk to eachother. I think a practical "non-mechanical" transport would have to involve more static memory elements, like FLASH, although I am willing to countenance almost any idea if the reasoning behind it is good (goodness  knows I've built enough weird bits of junk in my time!)

T.

[AKSA]

Folks,

New update in NEWS section of the Aspen website up today.......

http://www.aksaonline.com/news.html

Cheers,

Hugh

[Mark_Walsh]

Dear Tinker and others,

Could you please explain the difference between "standard" audio cable (RCA to RCA), Video 75 ohm interconnect cable (RCA to RCA), and digital cable to me?  Is it 75 ohm per metre or per mile, a maximum of 75 ohm, a minimum of 75 ohm?

If there is a dispoal bin of "high quality, name brand" VIDEO cables looking nice and sleek with colourful covers and gold and satin connectors , would these work as satisfactorily, better or worse than AUDIO cable for an AUDIO interconnect application?

Regards,
MW

For those of you interested in silver wire interconnects, these people have BARE 250 micron and 300 micron fine silver (99.99%) wire available at very affordable prices:

http://www.aemetal.com.au/precious_fine_silver_round_wire.asp