MDAC First Listen (part 00100110)

[janerik]

Jan,

As usual the review is full of errors - for starters the CD servo is NOT a DVD IDE drive but a CD slot loader with custom designed servo board whose Mechanism we originally sourced for the Lakewest CD transport project and Dominik had great fun designing the servo PCB - likewise "Custom control software written by Creek" is fictional. The idea was to make the servo solution available for other designs, and we designed it into Creeks player.

But they are VERY nice units

John,

A bit harsh... Mike Creek informed me that the Evo50CD is "based on a DVD drive", in his reply to my query about the player when I was reviewing it :

Hello Jan-Erik,

Thanks for asking about the EVOLUTION 50CD.

The transport for the new player is based on a DVD drive. DVD drives tend to take longer to read the table of contents, because they have to check for what data is on the disc much more than a simple CD transport would. In other words, is it a CD, CD-R, DVD, data or audio file, etc.

Creek developed its own custom servo board for the EVO 50CD, which enables the transport to pass files through to the internal system micro, thereby allowing for future software upgrades to be made using a CD-R.

As is usual for this type of drive and in fact common to the previous Creek CD players that used IDE - DVD drives, there is a digital buffer on the output that does provide a few seconds of memory storage, so that the signal is being clocked from a solid state source and not immediately from the disc.

A pure CD transport and servo, which normally has no memory, is therefore prone to skip if knocked, even though it would have physical suspension to try to avoid it.

This is not the reason why it takes a few seconds to read the TOC. Reading the whole disc to memory, if that is what you were referring to, would take a lot longer and serve no useful purpose in this product.

Best regards,

Mike Creek
Managing Director

Creek Audio Ltd - Epos Ltd

 

[JohnW]

Jan,

Oh now I see the cause of the confusion

We offered the solution to Mike so he had no direct involvement with the CD servo section, no misinformation was intentional from Mike.

We spent a decent amount of time trying to source a decent Mechanism and CD servo chipset solution direct from the manufactures in Asia.

The Mechanism is a High quality metal body automotive slot loader designed for operation in harsh conditions which comes without a preferred OPU from the vendor so we where able to choose from a selection of OPU's that best suited our needs. We selected a combined DVD / CD OPU as the CD market is dying and there would be no guaranteed longterm future for a CD only OPU. The DVD Wavelength Laser Diode is not used (not connected), only the CD Laser section is used.

The servo chipset is a CD only chipset, Dominik spent a significant amount of time with the CD servo software design house to optimise the Servo coefficients for our selected OPU / Mechanism - they where pretty clueless after disbanding the original Taiwanese based engineers and transferring the operation to China as is so typical thesedays on cost grounds.

Sourcing a CD only servo chip set that featured a Memory buffer and is Bit accurate was harder then we thought but the end solution works well - but it required a huge effort. Poor Dominik spent much time on the road frequently visiting the various vendors in China to problem solve there own work!

Dominik basically designed the Servo board by himself, we discuses the basics & I just glanced over the design at the end and tidied up a few small areas.

Sorry for the confused information, but only the OPU is a combined DVD / CD unit with no IDE interface involved, the servo board was designed for our Lakewest transport that never materialized.

I understand its hard for a reviewer to "break down" and filter manufactures "marketing speak"... as a designer I'm very sensitive to the technicalities of a review - there was so much written "incorrectly" in the press about our CD/CDQ./ MDAC designs that made our eyes roll - but I truly I understand you don't have an easy job!

I'm always happy to proof read any technicalities and maybe we can offer further insight into our designs - direct from the designers

 

[JohnW]

Jan,

The onboard memory is large enough to hold about 10 seconds of audio once filled, the disk is initially spun at higher speed to quickly fill the buffer, then the speed is reduced to a more normal rate once the buffer manger is happy. The Servo chipset will attempt re-read any troublesome areas on the disk to try and capture as much "error free" data as possible, although in reality uncorrectable CD errors are very uncommon even without the memory buffer re-read feature.

With "test software" its possible to unload and reload the CD without interrupting the audio playback thanks to the Memory buffer feature Its still unnatural to Dominik and myself to be able to physically stop the disk spinning and yet the audio continues to playback uninterrupted...

 

[janerik]

I understand its hard for a reviewer to "break down" and filter manufactures "marketing speak"... as a designer I'm very sensitive to the technicalities of a review - there was so much written "incorrectly" in the press about our CD/CDQ./ MDAC designs that made our eyes roll - but I truly I understand you don't have an easy job!

I'm always happy to proof read any technicalities and maybe we can offer further insight into our designs - direct from the designers

Hi John,

Thanks very much for your detailed reply ; all is now clear. I share your sensitivity to technical accuracy and will take you up on your offer to "proof-reading any technicalities". That's very generous of you !

Anyway, the design decisions that went into the Evo 50CD certainly show their merits ; the player has reconnected me with CD replay in a most addictive way.

Best regards,

Jan

 

[ChrisPa]

Chris,

A pair of PCM4202's as these are true 1Bit Modulators = true "DSD" output.

The dual complementary "Mono" single bit data outputs will be converted to PCM via the FPGA when required. We can offer various ADC filtering options for PCM.

The Master clock rate will be selectable, in Digital Pre-amplifier mode we will operate at a slightly higher clock rate then the rated Max, but well within the typical silicon tolerances of +/- 20%.

Thanks. That rings bells now. I think you've told us that before

Apart from dual ADCs helping channel separation, do the dual complimentary channels in each ADC help in the same sort of way as the paralleling of DAC channels? I'm not sure how I can see that that would work with a DSD steam

 

[JohnW]

Apart from dual ADCs helping channel separation, do the dual complimentary channels in each ADC help in the same sort of way as the paralleling of DAC channels? I'm not sure how I can see that that would work with a DSD steam

For PCM conversion we can make use of the dual complementary channels - again same for the Digital Pre-Amplifier mode. DSD has to be unprocessed so will only use one channel - although there is nothing to stop us archiving both "mono'ed" pairs of channels (so x4 DSD streams total).

There is also some benefit at die level on the ADC to have symmetrical data - this helps reduce PSU modulation effects etc.

 

[ChrisPa]

... designed for our Lakewest transport that never materialized.

I hope a derivative may appear eventually

 

[JohnW]

I hope a derivative may appear eventually

The problem is that to insure quality I'm trying to manufacture in Europe which means small scale "production" here in Czech Rep. - basically I'd like to be able to test each unit myself (like the MDAC2 / VFET amplifiers).

MDAC2 will be produced in limited numbers - and the VFET amplifiers in VERY limited numbers.

The problem is that the CD transport requires a lot of time and investment, yet would make little profit for the time spent. We might only sell say 100pcs in a year if we made GBP150 to GBP200 profit it would not cover the development and tooling costs etc.

Heres a picture of our prototype CD MTRN transport -sadly it ended up looking like a Car radio head unit!

 

[Steven Toy]

Would the Creek Evo 50 CD be up to the task of acting as a transport for the MDAC/MDAC2?

Would it need to be clock locked?

I'm tempted to buy one because the transport seems excellent and the build quality is exceptional at its price point.

PS: The 'equipment rack' looks a bit French!

 

[cstanpfm]

May want to consider the Audiolab M-CDT?

It is also a slot in transport.

 

[Steven Toy]

May want to consider the Audiolab M-CDT?

It is also a slot in transport.

I would like to use products designed and endorsed by Lakewest that are not simply 'copies.'

The Creek is a known and tried product. It is almost certainly a superior design (imho.)

I also want to give AudioWorks my business in this regard as I tried it at their shop using a mixture of my own and their equipment and ancillaries to considerable satisfaction.

Ancillaries included the venerable MusicWorks ReVo equipment rack with PEEK spars and fixing screws used in my system at home. These work incredibly well with the Creek as well as my Rega Apollo R CD used as a transport and MDAC L3 Toy.

The speakers used were the Tannoy Prestige Turnberry GRs - a more-than-slightly improved version of the Turnberry SEs that I use at home.

 

[ChrisPa]

The problem is that the CD transport requires a lot of time and investment, yet would make little profit for the time spent. We might only sell say 100pcs in a year if we made GBP150 to GBP200 profit it would not cover the development and tooling costs etc.

Heres a picture of our prototype CD MTRN transport -sadly it ended up looking like a Car radio head unit!

Could the drive and associated control and user interface software be sourced from creek?

I realise that doesn't provide a "complete transport" - the supporting electronics, and particularly casework.

I sort of assume that the electronics could be reused from the creek. However, as it would be a new board, that would be you manually laying it out. Is this one case where an autorouter will do a very good job on its own?

Which would possibly leave casework

 

[JohnW]

Could the drive and associated control and user interface software be sourced from creek?

I realise that doesn't provide a "complete transport" - the supporting electronics, and particularly casework.

I sort of assume that the electronics could be reused from the creek. However, as it would be a new board, that would be you manually laying it out. Is this one case where an autorouter will do a very good job on its own?

Which would possibly leave casework

Chris,

The Creek software is just a derivative of the Lakewest build - Dominik wrote the software. The OLED is the same as used on MDAC.

We can buy the CD Mechanism directly from the manufacturer as the solution is of our own design.

We own the servo board design and software so no problem there, the servo PCB is mounted below the Mechanism itself, so its a self contained unit.

https://dl.dropboxusercontent.com/u/86116171/CD Servo.jpg

We would need the Front panel, PSU PCB & Digital output "buffer / Reclock" PCB.

The Clock circuit on the Servo board is designed to be injection locked, which was my main input to the design. When externally injection locked the crystal acts as a High Q filter rather then as an oscillation element.

The problem is the chassis - I'd rather spend the time on sourcing the Chassis for the MDAC2 / TDAC etc then very low sales volume of M-TRN.

Why not buy the CD transport from AudioLab and I'll add the clock-lock PCB.

 

[timow]

The problem is the chassis - I'd rather spend the time on sourcing the Chassis for the MDAC2 / TDAC etc then very low sales volume of M-TRN.

I wouldn’t like to repeat my earlier question. Why not include CD slot loader to MDAC2/TDAC full width case?

 

[JohnW]

Hi John,

Thanks very much for your detailed reply ; all is now clear. I share your sensitivity to technical accuracy and will take you up on your offer to "proof-reading any technicalities". That's very generous of you !

Anyway, the design decisions that went into the Evo 50CD certainly show their merits ; the player has reconnected me with CD replay in a most addictive way.

Best regards,

Jan

Jan,

Thank you for your kind words about the Creek CD player - its good to see such positive reviews

I'm always just a PM away if you need any questions answered

 

[JohnW]

I wouldn’t like to repeat my earlier question. Why not include CD slot loader to MDAC2/TDAC full width case?

I'm not sure how much space there will be in the TDAC chassis - but it should be possible for the MDAC2 Chassis - but with design time and the "At Cost" costs would add about GBP250 to GBP300 extra to the price of the chassis.

It depends how many people would choose the MDAC2 full width Chassis with internal slot loader to see if its financially feasible. I'm not sure many people would pay an extra GBP300 for the CD slot loader.

Its expensive because I'd have to amortise say 2 months extra work over such a small number of units.

Its a good idea to consider, as I can add the CD servo interface connector to the MDAC2 Digital PCB.

 

[ChrisPa]

I sort of assumed that it was lakewest design & IP ;-)

 

[Obi1]

300GBP is in my opinion a small cost to get a CD player built into the full width chassis.
I would definately walk down that path if it is realised.

 

[Steven Toy]

I'm not sure how much space there will be in the TDAC chassis - but it should be possible for the MDAC2 Chassis - but with design time and the "At Cost" costs would add about GBP250 to GBP300 to the price of the chassis.

It depends how many people would choose the MDAC2 full width Chassis with internal slot loader to see if its financially feasible. I'm not sure many people would pay an extra GBP300 for the CD slot loader.

Its expensive because I'd have to amortise say 2 months extra work over such a small number of units.

Its a good idea to consider, as I can add the CD servo interface connector to the MDAC2 Digital PCB.

A one - box Lakewest CD player would be awesome!

 

[Stunsworth]

I've no interest in an integrated CD drive.