Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[phusis]

I've up until recently used MC16 via WASAPI Event Style, but the other day I tried out Kernel Streaming(w/lowest buffer setting) and found it an obvious sonic upgrade over WASAPI. The latter appears to result in a somewhat muted and center-focused presentation with restricted reverberation, whereas via Kernel the sound opens up in a more relaxed manner with better differentiation and without a shimmer of hardness heard in some string instruments(acoustic guitar) via WASAPI. ¨

What experiences have others made in this area?

[Jonahsdad]

I too prefer Kernel Streaming.

[BryanC]

I'm staying out of this one.

[phusis]

I'm staying out of this one.

It would seem not entirely This may be a controversial subject, or at least one being dependent on the varying systems it is to be implemented. Myself I'm simply going by what my ears tell me, and I know that my hearing is not controversial; so much I've learned to depend on them hearing differences, though not always what could be deemed more or less "right" or "wrong." In this specific case though my verdict is rather obvious, but I'd like to hear the oppinions of others nonetheless.

[Skogkatt]

I can't speak about kernel streaming as this isn't an option in my configuration.
However, having tried almost any other options, I have finally found peace of mind
using the following settings:

1) No resampling and "Output Format" in the DSP configuration DISABLED (not checked)
2) ASIO

I always used MC's internal resampler to convert regular CD sample rate from 44.1 to 192kHz thinking
this was the way to go based on tests made with early releases of MC15. I never changed this setting but
with time passing I realized there was something missing.

I spent a lot of time and efforts optimizing my system in many different aspects but never thought to touch resampling.

Of course I' ve tried WASAPI and WASAPI Event Style when it was released but found ASIO better sounding for my taste.
 
To my surprise and almost by chance I discovered that disabling the resampler improved the sound in terms of ambiance,
high frequencies smoothness, definition, and dynamics.

Now the soundstage is not anymore "recreated" in my room but the original place, where the recording was made,
is physically transferred in my listening room. I can't say this wasn't happening at all before the change, but it was
definitely less pronounced and somehow less real.

Going further I just tried to see what happens by unchecking the output format configuration and
I got another improvement: now everything seem to be right, so natural that I keep listening just music...  

If I have to rank what helped to improve sound in my configuration I would say (in order of importance):

1) No resampling
2) Unchecking Output Format in DSP configuration
3) A better SPIDF Coaxial Cable (electrical not optical)
4) Latest Realtek HD SW driver
5) ASIO a bit better than both regular WASAPI and WASAPI Event Style
6) Shorter SATA cable to the hard drive (10cm vs the standard 30-40cm)
7) Better PSU (PC)

I do realize that results may change using different HW/SW configurations but the sound quality
I achieved with JRiver MC16 configured this way is truly spectacular and unrivaled by any other
configuration I tried.

This is my humble experience and it would be nice to receive feedback from others about resampling and output
format.

P.S.: I do not imply in any way that MC's resampler isn't up to the task. It does its job very well
to the point I was using it for a long time with satisfaction. Resampling in real time is a tough task
and many resampling SW aren't real time. I just hear differences that make me prefer the  sound
of the original source. A possible explanation is also due to the ratio between 192kHz and 44.1kHz that
is not an integer number. It would have been nice if my sound card was supporting a sample rate of 176.4kHz
(1:4) to test if it was making any difference.  

[Vincent Kars]

Resampling is a tricky one.
Infinitewave give a nice explanation: http://src.infinitewave.ca/help.html
They also offer a comparison of several SRCs: http://src.infinitewave.ca/

Uli Brueggemann did a nice experiment.
Upsample the original 44.1 to 176 and back to 44.1
Some screenshots can be found on my website: http://thewelltemperedcomputer.com/KB/SRC.htm

Beside the quality of the SRC used, some DACs do change their sonic signature with the sample rate.
DACs might apply SRC internally. This is done with a DSP chip.
The question is how good is the SRC implemented in this chip.

Drivers do make a difference.
But it is hard to tell how it will turn out on a specific system.
It is also hard to explain why. Bit perfect is bit perfect, isn’t it.
Likewise media players.
A couple of years ago Foobar was the audiophile player of choice on Win.
Today the sentiment on audiophile forums is  switching to  MC.

[Sgucci99]

I don't want to speak on thecnical differencies, I'm only reporting my personal listening experiences which are quite similar to
the Skogkatt ones.

I prefer too to listen without resampling, with the output module disabled, and with the ASIO interface, which is by far bettter than the WASAPI and WASAPI event style.
Of course the digital SPDF output must be electrical, not optical.

I must say that the differences between the output module disabled, and enabled, leaving the sample rate at 44.1, are very very little, i'd say  indistinguishable.

I also did some experiments by resampling the file with an external sw at 192/24, not in real time, and the results, on top on the huge size increase, were similar to the configuration without resampling.

So I think the resampling itself it is not the problem, but how the algorithm is implemented, considering also that MC do it in real time.   
 

[Blaine78]

I can't speak about kernel streaming as this isn't an option in my configuration.
However, having tried almost any other options, I have finally found peace of mind
using the following settings:

1) No resampling and "Output Format" in the DSP configuration DISABLED (not checked)
2) ASIO

I have to agree on this, ASIO, and turning off 'Output Format' give music the best focus. bass seems tighter, drums and percussion have more power to them. been using WASAPI event style for while, and found it didn't sound quite right.
been listening to ASIO and output format disabled for few hours now, best I've heard from any software player!

[JimH]

been listening to ASIO and output format disabled for few hours now, best I've heard from any software player!

ASIO is what we recommend if it's available:
http://wiki.jriver.com/index.php/Audio_Output_Modes