Chameleon DAC listening and modifications
Apr 16, 2010 at 1:07 AM Post #796 of 1,158
Quote:

Originally Posted by rhodes54 /img/forum/go_quote.gif
I want to advice you to use the upsampling board when you want to use the chameleon with your CD67SE. Unfortunately the S/PDIF output of this player is far from clean.
I did a comparison recently between my transport and my old cd player, which is the CD67SE.



Many thanks Rhodes!..Didnt realise that the spdif from a reasonable player to be that bad.. Point noted and taken. cheers..
 
Apr 19, 2010 at 7:56 PM Post #797 of 1,158
Those scope reading look awfull. But we are in crazy land here, do those scope reading predict what those CDP's sound like?

It seems we are presented with 2 choices here to improve the spidf input.

1. get rid of the clock, but be exposed to what we see on the scope.

2. improve the clock, and/or the power supply for the clock and so tidy up those CDP input signals.

My gut feeling tells me, lose the damn clock! My vote goes for the simplified input, because too many times I have found that tidying up, also sterilizes.
 
Apr 20, 2010 at 4:57 AM Post #798 of 1,158
Quote:

Originally Posted by Wood /img/forum/go_quote.gif
Those scope reading look awfull. But we are in crazy land here, do those scope reading predict what those CDP's sound like?

It seems we are presented with 2 choices here to improve the spidf input.

1. get rid of the clock, but be exposed to what we see on the scope.

2. improve the clock, and/or the power supply for the clock and so tidy up those CDP input signals.

My gut feeling tells me, lose the damn clock! My vote goes for the simplified input, because too many times I have found that tidying up, also sterilizes.



Ideally you shouldn't need to fix a source. You should have a proper source and have only the most direct pass-through possible, like with the concept of the simple input board. Fixing a crappy source is fine when it's done right but these 'fix' circuits relying on cheap TCXOs are designed to improve a broad base of inferior sources and unfortunately throw off anyone with a superior source.
frown.gif
...

I wonder how Bill Allen is doing with his I2S mod. He's surely a lot closer to the goal. If we trust what Steve Nugent has said about properly implemented SPDIF in the past, we would get close to that I2S mod with the simple input board if it's done right. He's also basically said the Hiface is as good as his Off Ramp with a Superclock in most systems. There's tremendous potential (in that) for you guys. I imagine it would be an excellent combination for Chameleon.

EDIT: March Hiface review at 6Moons. http://www.sixmoons.com/audioreviews/m2tech/hiface.html

Last email I wrote Michael I told him there wasn't enough support here to justify production of the simple input board but it surely didn't mean others wouldn't be interested if they knew about the concept now or had a ready option later. I suggested he email his Chameleon customers and ask them if they're interested, explain the benefits, and ask what features they'd like to see in the board. That was the middle of last week and I haven't heard back from him.
 
Apr 20, 2010 at 8:32 AM Post #799 of 1,158
Quote:

Originally Posted by Wood /img/forum/go_quote.gif
Those scope reading look awfull. But we are in crazy land here, do those scope reading predict what those CDP's sound like?

It seems we are presented with 2 choices here to improve the spidf input.

1. get rid of the clock, but be exposed to what we see on the scope.

2. improve the clock, and/or the power supply for the clock and so tidy up those CDP input signals.

My gut feeling tells me, lose the damn clock! My vote goes for the simplified input, because too many times I have found that tidying up, also sterilizes.



I'm currently testing with a second tentlabs XO replacing the 12Mhz TCXO for the WM8805 and USB driver. It works fine but unfortunately something's wrong with the crystal so I have to wait for a new one to arrive before I can provide you some information on the changes in sound.
For Teradak I will perform some tests with the stock TCXO's and the Tentlabs power supply to see if the onboard TCXO's gain from a better power supply. More to come...
 
Apr 20, 2010 at 5:02 PM Post #800 of 1,158
I would also like to know what has the biggest influence clock or PSU.

Here we see Teradak has upgraded its x2. Why not then a luxury board for the Chameleon?

deluxe X2 - Windows Live
 
Apr 21, 2010 at 6:42 AM Post #801 of 1,158
Quote:

Originally Posted by Wood /img/forum/go_quote.gif
I would also like to know what has the biggest influence clock or PSU.

Here we see Teradak has upgraded its x2. Why not then a luxury board for the Chameleon?

deluxe X2 - Windows Live



OCXO upgrade. The clock PCB is someone else's product. How do you suppose that measures against the Tentlabs?

It looks like it's an OCXO reclocker with USB in and SPDIF out and the chassis is from something else. Isn't that box from the TeraDak version of the Valab? Looks like something he put together to test the OCXO, like he's trying to improve on his USB circuit. I gotta quit editing this post.

Did you guys take a good look at the revised Chameleon PCB for changes? Anything significant you can tell just by looking?
 
Apr 21, 2010 at 10:20 AM Post #802 of 1,158
It is a good question.
How do you suppose the measures against the Tentlabs?
Indeed TCXO is a little worse than the Tentlabs, but the OCXO is ....
bigsmile_face.gif


Tenlabs_Vs_TCXO.doc - Windows Live

Quote:

Originally Posted by Newk Yuler /img/forum/go_quote.gif
OCXO upgrade. The clock PCB is someone else's product. How do you suppose that measures against the Tentlabs?

It looks like it's an OCXO reclocker with USB in and SPDIF out and the chassis is from something else. Isn't that box from the TeraDak version of the Valab? Looks like something he put together to test the OCXO, like he's trying to improve on his USB circuit. I gotta quit editing this post.

Did you guys take a good look at the revised Chameleon PCB for changes? Anything significant you can tell just by looking?



 
Apr 21, 2010 at 1:44 PM Post #803 of 1,158
teraguy, looks very nice.

Can we expect to see this on a daughter board for the Chameleon?
 
Apr 22, 2010 at 6:49 AM Post #805 of 1,158
Quote:

Originally Posted by Newk Yuler /img/forum/go_quote.gif
About a month ago elsewhere here at HF.

There is a highly modded X2.

Also, if you agree slim.a is an intelligent wordsmith, read this and this.



Jkenny is selling modded Hiface now, he seems to be quite realistic with his prices considering how much work is involved.


http://www.head-fi.org/forums/f6/mod...93/index2.html

With all this recent welcome attention to cleaning up the source, be it CDP or USB, it does make one wonder how many years have been wasted refining oversampling etc in a vain attempt to correct problems originating at the source in the first place.

Will be able to report back soon on the adum4160 device soon. Its been lingering in customs far to long.
 
Apr 25, 2010 at 5:38 PM Post #806 of 1,158
Here is a picture of my Black Chameleon with Nugent's direct I2S connection. (Notice the complete elimination of the main circuit board.)

CIMG4754.JPG


There are more mods to be done yet already my Black Chameleon is the best sounding DAC I have ever heard in my system. My friend Jon from Columbus, OH came out for a few days and he is a dyed blue analog man. We listened to digital one night and LP's the next night. He had to admit on the 3rd night that the digital indeed outperforms my analog. This would be the first time I too would make that claim.

It's hard to describe exactly how direct I2S from the Pace Car USB improves the sound. While my analog still has a bit more density to the soundstage, with my new digital rig you're completely immersed inside the 3D ambient field of the recording. I suspect the harmonic tails and reflections that are normally obscured by jitter (or distortion in the analog domain) remain intact and bounce around the room presenting a closer to live experience. In the past my analog rig did a better job at capturing "the live" ... but not anymore!

The USB input to the Pace Car is an OEM version of the M2Tech device with I2S output and no clocks on board, instead it uses the Pace Car's dual FIFO clocks. In my case 22mHz & 49mHz Audiocom Super Clock 4's. I asked Nugent about adding this OEM device with "onboard" clocks inside the Chameleon and he said the issue is the M2Tech only outputs 3.3 Volts, the DAC chips inside the Chameleon are looking for 5 Volts for optimum sound. (Steve modded my Pace Car for 5 Volt output.) Perhaps there is a way to get a 5 Volt I2S bus out of the OEM M2Tech device but Nugent is not interested in building it. Maybe this guy could pull it off???
https://sites.google.com/site/hifacemods/

If the OEM M2Tech I2S card could be modded for 5 volt output I have no doubt this would offer the "lowest cost" option for eliminating interface jitter on the Chameleon or Valab. But remember with just a USB input now we're talkin' computer audio only. Of coarse if you have the cash the Pace Car USB would still be the superior choice.

I sent the information about the OEM M2Tech board to Michael in the past but once again its a demand vs price point thing. However given the superiority of direct I2S input and looking at the simplicity in the pictures above ... there is a strong argument for a M2Tech I2S/USB input only version of the Chameleon.

More Pictures of the Black Chameleon I2S. http://picasaweb.google.com/baulsaud...Gt5LCwjuj42QE#
 
Apr 25, 2010 at 11:20 PM Post #807 of 1,158
Quote:

Originally Posted by Bill Allen /img/forum/go_quote.gif
Perhaps there is a way to get a 5 Volt I2S bus out of the OEM M2Tech device but Nugent is not interested in building it. Maybe this guy could pull it off???
https://sites.google.com/site/hifacemods/



LOL! That's jkeny mentioned in the post right above your's. Cat is in Ireland.

Your mod proves the absolute potential of 16 stacked TDA1543 and it should be appreciated that someone has taken it to that potential, but it isn't practical for the vast majority of people interested in Chameleon. I mean no disrespect and I very much appreciate your examples, but who is going to feed Chameleon like you have? Something else your mod proves is that someone like Michael would benefit from developing a 16x TDA1543 DAC specifically for high end audiophiles. A design without any of the circuits for broad based applications that can degrade a high end source.

No idea how well you've been keeping up with this thread, but it's been suggested that Chameleon would benefit from a simplified and well implemented SPDIF input board (sans TCXO) to advance its potential for people that already have better sources (like a Hiface). Suggesting a DIR9001 based board with an easily user implemented option of the pins 13 and 14 mod you and Steve pioneered. Michael was interested in the idea at first but he seems to have lost interest in even that. I emailed him again a few days ago and asked for his thoughts about possibly pursuing this simplified input card and he again didn't reply. I would wait and buy a Chameleon if he would at least say he's developing the board, but he apparently won't even say yes or no.
 
Apr 26, 2010 at 12:31 AM Post #808 of 1,158
Really Bill. I've been following all your threads and every dac is the best you've ever heard. How are we to believe you?

I think you need to FedEx that puppy directly to me and I will do an unbiased independent analysis of the mod. I'll get some recording engineers to come over with some of their master recordings and we'll listen.

Shouldn't take more than a couple months, give or take a year.
beerchug.gif
 
Apr 26, 2010 at 3:26 AM Post #809 of 1,158
I just threw 700USD at one of these. I wanted to wait on Chameleon but I'm not going to wait indefinitely on a simple input board that may never materialize. If it does I may again give Chameleon the consideration the potential of its design deserves. I suggest everyone intrigued with the TDA1543 DACs should look at similarities of the Audio GD designs. Run searches here at HeadFi and Google. - Burr Brown PCM1704UK DACs (multi/24 bit). DIR9001 SPDIF input with the low jitter option applied. The DSP they use has user DIP switchable options including NOS bypass. No TCXOs. I mentioned a review by slim.a a few posts back.

Pat, I haven't attached the Cinemags to the Valab as I intended during a recent vacation but I will as time allows. I ended up spending the time at the event I took vacation to attend and also rebuilding my music server. After that my day job has been devouring my time and I'm going to be working every day for at least the next couple of weeks. When I add the Cinemags I'll mention it here and post about the sound in the Valab thread.

I'm still watching and looking forward to the mods you gentlemen develop and the changes Michael makes for Chameleon.
 
Apr 26, 2010 at 4:05 AM Post #810 of 1,158
Well, by golly. Just in...

Hi,
I considered making such a simple input board, but we have no idea on its architecture right now. Dir9001 may be an option. We are very sorry we also have no schedule for making it. The simple input indeed is a great idea for the very specific customers. Someday we will make the board from product viewpoint and re-organize the input and output interfaces.

Michael,
 

Users who are viewing this thread

Back
Top