|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
67.60.98.130
In Reply to: RE: cMP - the open source high-end Memory Player posted by cics on December 30, 2007 at 05:42:01
This has been a long-time coming (delayed by a back injury, a bout with the flu, another with an upper-respiratory infection, and just a lot of work), but it's up and running now and soon to go to it's long-suffering owner.
It's a Twisted Pear Buffalo32 board mounted to a modified Juli@ digital section. The Buff32 is powered by TP's Placid shunt regulator boards and the modified Juli@ has it's own supply for the 5v and 3.3v which are regulated by Dexa drop-in boards.
One thing I wanted to do with this is keep it as a one-box system, both for convenience and to keep the supply and signal lines short, especially the I2S signal lines between the Juli@ digital section and the Buff32 input. Originally the power transformers were going to go into a separate box, but careful placement allowed them to fit next to where the owner has installed a hybrid-linear supply running a PicoPSU.
Sound (after only a few hours of burn-in) is very good, definitely in the same class as the hyper-tweaked AK4399-based DAC I've been experimenting with since early last summer. As it sits now, I wouldn't change one for the other... both are very detailed, dynamic, and musical. But of course, the Buff32 should improve as it breaks-in, so I may be singing a different tune a week from now. Also, I have a few more 'tweaks' I plan to implement before shipping it to it's owner.
And of course, my question for Theob... when you were first implementing your Buff32, what were the symptoms that eventually led to you adding the damping resistors on the I2S lines? Was it the recurring noise, sound quality, lack of signal lock, or something else? I ask because I'm not hearing anything in my implementation that leads me to believe that I need to add them here... which was what I was trying to achieve with the <1/2" I2S lines. Thanks!
More impressions as it breaks in.
Everything matters!
Follow Ups:
Hi Gregg welcome back, hope you are feeling well again!
I read the Wackyterbacky thread in DIY Audio and corresponded with him. I tried I2S for the heck of it (always looking for easy tweaks) and it worked well so I took the next step of damping resistors (22 ohms) and it improved sonics yet again. More importantly ...no more metallics. Its been 2 months at least. So I highly recommend it.
Other than the metallics I was happy with I2S and even spdif for that matter. Damping resistors significantly improved imaging/dynamics... no more of that sloshing around excessive ambience (which many may like and I did too) but when I heard what the damping resistors revealed I was hooked: bigger, deeper, wider sound stage with more separation of images within the soundfield.
Theo,Thanks for the welcome and the well-wishes... I'm not quite over the flu and the back, but both are getting better & going in the right direction.
Also thanks for the info on how you got to adding the series resistors in the I2S signal lines. That all makes sense and provides a data point for whether I should do the same for this Buffalo32 implementation. I have to say I'm leaning towards not adding additional resistance.
Why? First, the Buff32 already has 22 ohm resistors in series with the data lines.
Second, the Buff32-on-top-of-Juli@ implementation means that the connections are actually shorter than the equivalent connections for the stock Juli@. For the connection from the Juli@ digital section board to the Buff32 board, it is shorter than the conductor length of the Juli@'s inter-board connectors. And the path length of the traces on the Buff32 board from where the lines enter the board to where they hit the DAC is shorter than the same lines on the Juli@ analog section.
Then third, I did not hear the type of sonic degradation with this implementation that I heard with the AK4398 and AK4399 DACs when I used them with no resistors on the I2S lines (muffled and veiled sounding).
Of course, the only way to really know for sure is to try them, which I might still do.
Again, thanks!
Greg in Mississippi
P.S. It is sounding better as it runs in... perhaps a hair more detail now than the AK4399 setup I'm using. Its also very listenable, if ultimately not as convincing and musical as the AK4399 setup. But, it's only been breaking in for 8 hours or so... more to come!
Everything matters!
Edits: 01/31/10
Wackyterbacky indicates that Steve Nugent recommends removing the 22 ohm resistors on board. I never tried for fear of goofing something up on the Buf32s board. Apparently Wacky also did not remove these either although that could be a sonic opportunity. Anyway I am very happy with the I2s resistors on both ends.
Theo,
Thanks for the info.
I'd agree that removing the 22R resistors on the Buff32 board would be tricky unless one is familiar with SMD soldering. OTOH, they really just need to be bypassed which is much easier... It still takes a small-tipped iron and small-gauge solder tho. I do think that would be a performance increase in your setup and Wackyterbacky's (although I'm not sure what they look like in his DAC).
But as I understand it, the setup I'm using on this implementation is more like the original Juli@ analog board's connection to the Juli@ digital board... Except that my connections are slightly shorter. And the size and placement of the resistors on the Buff32 board are pretty close in location and value to that of the Juli@ analog board.
OTOH, the longer cable connections you and Wackyterbacky are using are a different world and require different techniques to make them work right... Hence the source-end signal resistors and removal of the destination-end resistors.
Greg in Mississippi
Everything matters!
Yes I forgot you have short i2s wires so yes you are probably there already!
Happy listening!
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: