strange frequency response of m2tech evo

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
strange frequency response of m2tech evo

anyone has done any measurements of m2tech evo usb/spdif converter? I am somewhat confused with it objective performance numbers(I am quite satisfacted with it sonic , even if quite diferent form other product)

when first tried m2tech evo sonicaly I noticed some lack of airiness in sound(less expansive soundstage) but the soundstage focus was superb as bass tightness. one day I decided to use m2tech evo for speaker measurements(it gets digital MLS signal from laptop, and feed DAC) to my surpsise i saw some top octave roll off. on white nois eI can even tell from the ears(I have designing speakers form many years) i imedialty returned to chineese made thera-linx usb/spdif converter (which measures EXACTLY like internal pc measuring equipment),
thinking my unit is defective i ordered another m2tech evo unit with more late serial number, hoping its a problem of actual unit rather then design fault. but it measured the same. later i acquired weiss int202 which also confirmd to be flat ( weiss int 202 measured exactly the same as thera-linx or internal pc measuring hardware)

i tried everything -
1)diferent DAC's,
2)rca. xlr, st outputs(didnt tried i2s as its format not compatibile with my dac's)
3) variuos power suplies- the problem stays here.

could m2tech evo designer answer here. I do not want to shock customers, but I want an answer why this interface have rolled off top octave. it evenmeasures from 1.5meter on speaker axis!!

greetings to all
R.
 

Attachments

  • m2tech evo.jpg
    m2tech evo.jpg
    35.5 KB · Views: 223
Actually, my guess is the two tested spdif transmitters differ in pre-emphasis bit in SPDIF preamble and the DAC respects this information. That would produce this behaviour.

Status of the pre-emphasis bit in SPDIF stream can be checked with any card with SPDIF input e.g. in linux, perhaps there is a way in other OSes too.
 
If you have a DAC with pre/de-emphasis indicator (no idea if any exists :) ), you can check.



yes they have. well at least one i have tried.- here excerpt from manual-

De-Emphasis LED
Some digital recordings are encoded with pre-emphasis contouring. Your DAC automatically
recognizes and processes them. An LED will indicate that the recording is being de-emphasized.

no lights will come. now I need to open the dac and valid that deempahasis led is working? :)

seriuosly , I got one sugestion to isntall m2tech young dac drivers instead original evo. I will try this..
 
no lights will come. now I need to open the dac and valid that deempahasis led is working? :)

I do not know, even that may be the case :)

seriuosly , I got one sugestion to isntall m2tech young dac drivers instead original evo. I will try this..

I do not see how drivers could change frequency characteristics of bit-perfect digital transmission, and consistently too. Provided the transmission is really bit-perfect which should be verified first.
 
No, none of my DACs have an indicator. However, I think if either M2Tech driver was not bit perfect it would have been reported by now.

Hey, this just what I have heard, and many here will rubbish that because I don't have measurements to back it up....

from a quick web search:

The M2Tech hiFace uses a proprietary driver to interact directly with a player in kernel streaming mode at a low level of CPU loading. It does so without any other drivers and allows for audio data to reach the peripheral easily and without any undesired changes to the data stream. That's the long way of saying that your computer will pass a bit-perfect 24/192 data stream without bogging down your machine or screwing with the sound.


Fran
 
No, none of my DACs have an indicator.

I do not think there could be a bit-perfection indicator. The check requires recording the stream with another device equipped with SPDIF input, and comparing with the original.

However, I think if either M2Tech driver was not bit perfect it would have been reported by now.

I don't know, very few people do this check.


Hey, this just what I have heard, and many here will rubbish that because I don't have measurements to back it up....

Come on, this is not rubbishing, but the first question which comes to mind when a different sound is reported. The most likely explanation is the actual stream is different. Every analysis always starts with the most likely cause.

The M2Tech hiFace uses a proprietary driver to interact directly with a player in kernel streaming mode at a low level of CPU loading. It does so without any other drivers and allows for audio data to reach the peripheral easily and without any undesired changes to the data stream. That's the long way of saying that your computer will pass a bit-perfect 24/192 data stream without bogging down your machine or screwing with the sound.

Honestly, this is just a general marketing speak. Every asio (or linux alsa) bit-perfect driver does that.
 
Last edited:
Sorry now, don't mean to be prickly, just that I've seen so many threads go wrong: "If you can't measure it you're a fool and you must be crazy to be hearing a difference!!" To be honest I've much reduced the time I've spent here over the couple of months or so just because of this. Anyway, look rant over, don't mean to go so off topic.....


Anyway, back on topic.... nope, I don't have the facility to record via SPDIF, which is a pity as it would answer the question alright. Can the OP do it I wonder?

Take your point RE: the marketing speak - I searched for a few minutes and that was the first thing that came up that mentioned M2Tech and drivers being bit perfect. Might need to search harder, although surely someone out there has done this work already!


Fran
 
OFC it is bit-perfect. It has been tested numerous times.
The latest reference I can remember is a detailed review in AVMentor, a greek website, with measurements and stuff (very good site btw, but it's going to be all greek to you!)...

Here is the page with the measurements: Click

You'll be able to understand the graphs, at the very least but Google Translate could help with the rest.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.