Cayin Mini-CD MKII CD Player Review

[pdb]

I also own a Linn Mimik (II) unfortunately not working anymore - and a Beogram CDX bought new in 1986 for around 900 dollars (5,995 dkk) The CDX has the rare 14 bit Philips TDA1540 dac chip and a CDM-1 transport (diecast zinc) It also has a feature not mentioned in the manual - it closes the lid after 30 minutes of inactivity (got a minor shock the first time ) Memory lane...

 

[amirm]

OK, the review is fully updated with the correct test CD. Using that, and running a bunch more tests, it seems that the bitstream out of the Cayin S/PDIF is toggling its low order bit causing distortion even in an external DAC! Check it out and let me know if you think there is another explanation for it.

I also ran more tests including S/PDIF output of the LINN which is perfect.

 

[amirm]

Sorry @amirm for the harsh words.

Oh it is fine. The player team had no signal processing experience. Their sister team, the signal processing/codec group, did. But it was hard for them to get the WMP team to accept their help in these regards.

 

[restorer-john]

@amirm Did you work on Windows Media Centre too? That software is in my opinion the most beautiful product to ever have come out of MS. Way before its time.

 

[Blumlein 88]

OK, the review is fully updated with the correct test CD. Using that, and running a bunch more tests, it seems that the bitstream out of the Cayin S/PDIF is toggling its low order bit causing distortion even in an external DAC! Check it out and let me know if you think there is another explanation for it.

I also ran more tests including S/PDIF output of the LINN which is perfect.

So, having found this anomalous result, it might make sense to digitally record the digital output and see how it differs from the file on the test CD. Finding a toggled bit should be fairly easy.

 

[amirm]

@amirm Did you work on Windows Media Centre too? That software is in my opinion the most beautiful product to ever have come out of MS. Way before its time.

That was the sister division to mine. We provided all the underpinnings, they provided the UI, DVR, etc. for "10 foot experience." Company politics caused their demise. Our then CEO Steve Ballmer (incorrectly) thought that the Xbox group is where all hardware dependent technologies need to go and moved that group out of Windows. A new SVP from Office took over Windows around the same time and his policy was that he didn't want to have any dependency on a group that was not under his control. In short order these two actions caused the demise of the MCE group even though it was generating good bit of revenue for us. I also moved to Xbox group but quickly realized there was zero interest in what we were doing there and so left. My group got dissolved into rest of Windows producing nothing new 15 years later!

 

[amirm]

So, having found this anomalous result, it might make sense to digitally record the digital output and see how it differs from the file on the test CD. Finding a toggled bit should be fairly easy.

I leave this as an exercise for the company. The comparison to LINN clearly shows this to be a bug they need to investigate.

 

[restorer-john]

That was the sister division to mine. We provided all the underpinnings, they provided the UI, DVR, etc. for "10 foot experience." Company politics caused their demise. Our then CEO Steve Ballmer (incorrectly) thought that the Xbox group is where all hardware dependent technologies need to go and moved that group out of Windows. A new SVP from Office took over Windows around the same time and his policy was that he didn't want to have any dependency on a group that was not under his control. In short order these two actions caused the demise of the MCE group even though it was generating good bit of revenue for us. I also moved to Xbox group but quickly realized there was zero interest in what we were doing there and so left. My group got dissolved into rest of Windows producing nothing new 15 years later!

Fascinating. That software could have launched MS onto every flatscreen TV if they'd pursued it. The DVR function with up to 3/4? DTV tuners was incredible at the time. I used two tuners and a Dell ZinoHD HTPC +NAS with MC for years. Could have streamlined and embedded Windows/MCE into flatscreens instead the mess of proprietary systems that came along with early smart TVs. There would be no Android TV now.

It was just sad to see it hung out to dry with no development or bug fixes. It was clear that the hardware stars aligned and the depth of talent at the time produced something amazing. Most people didn't even know it even existed on their 'puters.

 

[Rja4000]

it seems that the bitstream out of the Cayin S/PDIF is toggling its low order bit causing distortion even in an external DAC!

A test I wanted to do but I didn't have time yet:

Add the 16 bits 44.1kHz bit-perfect test file from RME to the test disc and run it through the SPDIF out to a RME ADI-2 Pro.

I suppose that should trigger the bit-perfect confirmation message.

EDIT:
WMP doesn't accept to add the file to track list.
Revenge against @MC_RME words above ?
Well, most probably, its because the track is too short, at just 4s duration.
I'll have to open Wavelab and make it longer, I guess...
@MC_RME That could be useful to make this very file a bit longer...

 

[Mnyb]

Voted poor ! It's actually very rare to have a dysfunctional spdiff output , most transports just work . So this is indeed a bad design error .

 

[Palladium]

Voted poor ! It's actually very rare to have a dysfunctional spdiff output , most transports just work . So this is indeed a bad design error .

It baffles the mind how this even happens when even $10 100mbps network switches are bulletproof at data integrity.

 

[Mulder]

Here is a pro Tascam CD player I tested a while back: https://www.audiosciencereview.com/...tascam-bd-mp1-review-cd-blu-ray-player.26969/

That's 10 dB better than Cayin. It was not expensive either at $499.

Could the harmonics at the digital output in this case also be supposed to be an effect of the same test-CD issue?

 

[Nicolas]

A test I wanted to do but I didn't have time yet:

Add the 16 bits 44.1kHz bit-perfect test file from RME to the test disc and run it through the SPDIF out to a RME ADI-2 Pro.

I suppose that should trigger the bit-perfect confirmation message.
View attachment 304428

EDIT:
WMP doesn't accept to add the file to track list.
Revenge against @MC_RME words above ?
Well, most probably, its because the track is too short, at just 4s duration.
I'll have to open Wavelab and make it longer, I guess...
@MC_RME That could be useful to make this very file a bit longer...

Interesting, I wasn't aware of that feature on RME DACs, thanks.
Reading this thread i was wondering about the output of my own CD transport (which to me is essentially a black box : no idea if it is a capable unit or not).
I happen to own an adi-2 DAC and this test is easy to try, so I will. Now I just have to find a blank cd...

Edit : Well it is done. Had to make the test signal longer to make the cd readable by the transport, but it worked.
Not sure about the value of this, but let it be known to all interested parties that the RME 44.1/16 test signal was successfully detected by an adi-2 DAC when played back from a cambridge audio CXC (optical out).

 

[MC_RME]

A test I wanted to do but I didn't have time yet:

Add the 16 bits 44.1kHz bit-perfect test file from RME to the test disc and run it through the SPDIF out to a RME ADI-2 Pro.

I suppose that should trigger the bit-perfect confirmation message.
View attachment 304428

EDIT:
WMP doesn't accept to add the file to track list.
Revenge against @MC_RME words above ?
Well, most probably, its because the track is too short, at just 4s duration.
I'll have to open Wavelab and make it longer, I guess...
@MC_RME That could be useful to make this very file a bit longer...

Yes, it most probably is too short. The first version was one second and various software did not want to loop it, so I used a wave editor to make it 4 seconds...and that's really easy, just copy it multiple times...

 

[Rja4000]

Yes, it most probably is too short. The first version was one second and various software did not want to loop it, so I used a wave editor to make it 4 seconds...and that's really easy, just copy it multiple times...

I guess something like

sox "441_16_adi2pro_bittest 4s.wav" "441_16_adi2pro_bittest 4s.wav" "441_16_adi2pro_bittest 4s.wav" "441_16_adi2pro_bittest 12s.wav"

would do the trick...

 

[Herbert]

Voted poor ! It's actually very rare to have a dysfunctional spdiff output , most transports just work . So this is indeed a bad design error .

I think Philips also spoiled it a looong time ago. When I remember the story correct, they somehow put the spdif after the oversampling filter on certain "all in one" ICs, early nineties?. Stil being 44.1, it was 0.5db lower. I did modify some 2nd gen. Players with Sony chipsets (CX23035) with SPDF-Out, normally it is derived from the demodulated EFM signal...

 

[amirm]

Could the harmonics at the digital output in this case also be supposed to be an effect of the same test-CD issue?

Yes it would have impacted that test as well unfortunately.

 

[Rja4000]

I guess something like

sox "441_16_adi2pro_bittest 4s.wav" "441_16_adi2pro_bittest 4s.wav" "441_16_adi2pro_bittest 4s.wav" "441_16_adi2pro_bittest 12s.wav"

would do the trick...

It works

With both the NAD M50 and the Philips CD604, from SPDIF output (RCA in this case)

PS: I just see @Nicolas has done it before (see above).

 

[Palladium]

I think Philips also spoiled it a looong time ago. When I remember the story correct, they somehow put the spdif after the oversampling filter on certain "all in one" ICs, early nineties?. Stil being 44.1, it was 0.5db lower. I did modify some 2nd gen. Players with Sony chipsets (CX23035) with SPDF-Out, normally it is derived from the demodulated EFM signal...

I like how all audiophile myths can be traced back to gross incompetence.

 

[amirm]

I recall a blind test of CD players in 1980s where one sounded different. They found out it had a stuck bit or something like that.