Musical Fidelity - M1DAC - 192 KHz input problem?


I currently have a MF A3-24 which I like very much. I have been looking for a DAC which takes 192KHz input (A3-24 is limited to 96KHz), so looked into the new M1. It was reviewed well in the March Stereophile. In the new Stereophile (June) in the Follow-up section, John Adkinson notes that with 192KHz inputs, the output rolls off pretty much the same as with 96KHz - at just over 40 KHz. I would expect it to extend out to close to 1/2 the sampling freqency. He notes that, but does not make a big deal about it. But, it seems like MF increased the input sampling freqency up to 192KHz but did not update the D to A and/or output accordingly. So, although it takes in 192KHz, it does not seem to take advantage of it.

That also brings up the question as to whether other manufactures may be doing the same thing. Everyone wants to have 192KHz performance, but it seems they also need to change the D to A to take advantage of 192KHz, not just the input. I would just assume that they were doing so. But maybe not, especially at this low to mid level price point. Or am I missing something?

Thanks to John for pointing this out.
128x128dtc
I have the same cut of music in SACD, 24/88 and 24/192 format. SACD is played in DSD mode. The 24/88 is played at 96khz sampling rate while the 24/192 is played at 192khz sampling rate. At the end, the 24/88 wins. Upsampling 24/88 to 24/192 degrades the sound quality.

There are a few caveats from the above comparison:
- The SACD could be better with a better transport.
- The 24/192 could be better if the DAC is properly implemented.
- All these are results of using Oppo-95 alone. Using other DACs or transports might give a different ranking.
I have not seen anything from MF that discusses the 96/192 issue that Atkinson noted. My old A3-24 has a setting to upsample to either 96 or 192. The M1DAC seems to upsample everthing to 192. Typical Nyquist-Shannon theory would suggest that a 192KHz digital sampling should be able to reproduce a 96K analog signal. Maybe this issue is as Dchan88 suggestes, it has something to do with available DACs. If that is the case, I again wonder if other manufacturers are doing the same thing. I am certainly not going to jump to a 192KHz DAC until I understand this futher.
DTC, I agree completely about sampling. I have been using an MHZS tubed cd player that has adjustable upsampling, I could clearly hear the difference playing cd's @ 192k. I have since abandonded the cd and have downloaded files to computer and am using an external dac. I have downloaded files that are 24/48, 24/88, 24/96, 24/176 and 24/192. I honestly don't have anything except a few cd's to compare with the high res files, but there is a clear difference playing a 24/96 vs cd. I've never compared 2 file types or sample rates against each other to comment past this. My earlier point was, do you know if MF even tested a signal past 40K to print documented output differences in frequency response within sample rate?
Dchan88 - thanks for the link. Note that the author of the original article says he does hear a positive difference in going from 96 to 192, although he offers no real detail on his sources or equipment.

On your Oppo, have you tried music that was sampled directly from the masters at both 192 and 96? Once you start doing the switching you are talking about, you are having the Oppo either upsample or downsample. My old MF A3-24 upsamples Redbook to both 96 and 192 and I do not find the 192 inferior to the 96. And 96KHz files upsample to 192 do not sound worse to me. Although, from this article I wonder if the upsampling to 192 is really getting through the D to A process.

Timlub - most people find that a well engineer 96KHz digital file does sound better than one sampled at 44KHz. This may not seem logical, but even my old ears hear the difference. But the 96KHz mastering is often done with more care than the 44KHz one, so that may be a big part of it. Lets not get into double blind studies :)

One of my concerns is just how many manufacturers are adjusting their front ends to take in 192KHz, but not updating the D to A process to pass the full bandwidth. If they are doing that because of what Benchmark and Lavry are saying, then maybe they should state that and explain why. Otherwise, they are implying that there D to A process is being done at a higher resolution than it actual is.
Do you know if they even tested beyond 48K to give results up to 96K? I have a hard time hearing up there.
This might be an interesting read for you:

http://www.soundstage.com/index.php?option=com_content&view=article&id=126:96khz-vs-192khz&catid=57:reader-feedback&Itemid=24

I have an Oppo 95 that can switch between 192khz and 96khz during playback. With 192khz, the sound is brittle and 2 dimensional.