trioaj.blogg.se

1. #BITPERFECT NATIVE MANUALS#
2. #BITPERFECT NATIVE SERIAL#
3. #BITPERFECT NATIVE DRIVERS#
4. #BITPERFECT NATIVE SOFTWARE#
5. #BITPERFECT NATIVE PC#

In case of failures that might be related to drivers, I also try with older drivers.

#BITPERFECT NATIVE DRIVERS#

#BITPERFECT NATIVE PC#

Consumer S/PDIF is produced using another PC under Windows XP SP3, using foobar's bit-perfect playback plugin (WDM/KS).Professional S/PDIF (aka AES) is produced using Forssell MADC-2 (I connect it to S/PDIF inputs using a impedance-matching cable).Recording is done on a PC under Windows XP SP3.I am going to post my findings for every gear I test, just for reference in case other people are interested.

Once I will get it, I will perform enough testing to ensure it is bit-perfect. The thing is, I don't have yet the recording audio interface, and I was just asking for GS' experience to limit the risks of buying a faulty audio interface. I keep a note about it though, as it might be possible this way to test those kinds of distortions without being dependent on the playing audio interface (that could not be bit-perfect itself). Recording several times the same source works to detect non-deterministic (not repeatable) distortion only. But maybe I am too much paranoid and in real life with the vast majority of audio interfaces such problems will never occur ? (in this case, why some manufacturers point out that their gear has a special bit-accurracy or transparent feature ?) In those cases of distortion, recording several times might not give different results although it is not bit-perfect. I was more anxious about internal digital processing made by the audio interface (dithering, level adjustments, mixers.), or invalid drivers or softwares that might have bugs, like adding deterministic distortion (for example always setting a bit to 1 every nth sample).

#BITPERFECT NATIVE SERIAL#

EthanI understand your point concerning serial communications, I totally agree. If they match then you know your hardware is functioning properly and you can continue with confidence. My best advice is to record the same source two or three times and compare them all as you described.

With serial communications it either works or it doesn't work. I don't see how there could be subtle differences. It has been a few months now that I keep finding a huge source of knowledge and help around these threads. PS: I take advantage of my first post to greet GS, and give big thanks to the GS community. This would be a correct solution to me if confirmed. Some people advise to not use any USB/Firewire/PCI streaming interface as it alters bit-perfect recording: Just use audio interfaces writing to an internal disk or SD card, and then access the files via USB/Firewire/Card reader (like Tascam HD-P2). It seemed to me that when the input signal flew directly to the USB/Firewire/PCI interface without extra processors, I would get a bit-perfect recording.

#BITPERFECT NATIVE MANUALS#

Some gears' manuals indicate their internal signal flow. Is this just marketing effect, or do I have to conclude that other gears will get a hard time in having bit-perfect recording ? Some gears have special features to ensure bit-perfect recording: Creative has bit-accuracy mode but works only up to 96khz, and Weiss DAC202 has transparency check but is over budget for me.

Here are some experiences I have gathered around: Here are some of the interfaces I have considered so far:Īnd tons of others (I can post my entire research if some are interested: 67 audio interfaces).

#BITPERFECT NATIVE SOFTWARE#

So I have a few questions regarding this issue:ġ- What parts of the signal flow (from the PCM stream to the file) might alter the audio data ? Drivers (ASIO, WDM.) ? USB/Firewire/PCI/PCIe protocols ? Audio interface internals ? Recording software ? OS ? What have you experienced and pointed out as problems ?Ģ- What would you advise me to consider buying ? I would have thought that any interface would be ok as long as no A/D or D/A conversion altered the flow, but after googling and searching around GS, it seems that people can't get an exact bit-perfect recording of their source (failing null tests with just digital loop-backs). I have found many audio interfaces that would fit the specs, but can't find any guarantee about the bit-perfect feature. I don't care about latency (as long as I don't have dropouts). I want to record this stream in my PC with bit-perfect accuracy (getting a WAV, FLAC, RAW file). I have a digital stereo PCM stream out of AES/EBU (or S/PDIF) from my converter (24bits/192khz single-wire).