Ok, let's try this again.
First, yes, I have listened to a variety of FLAC codecs in "double-blind" tests. There are audible, reproducible differences, even between version of the same codec.
This sounds illogical - but only if you don't think the problem through all the way. Yes, ALL lossless codecs are indeed LOSSLESS. This is not the issue. If you simply use the codec to encode and decode audio and look at the resulting data, they will all give you the same data. However, data is by far not the only issue in regards to quality of sound. There is also "jitter".
Jitter is the difference in timing between the samples as they are transmitted to the DAC or digital to analog conversion step (hardware/software). If one version of the codec is less efficient in its processing and clocking, it can introduce a significant amount of jitter, thus degrading the sound. Remember audio is "data over time". Getting the data right is only one step in the process.
Additionally, the load placed on the processor can also increase the noise in the system. If there is too much noise it needs to be dealt with or it make it harder to determine the signal, and thus jitter can be introduced as well. While this accounts for a much smaller effect on jitter than the processing, it is measurable.
As for sound-quality between different lossless codecs, yes, double-blind listening studies have been done over multiple people and actual measuring of noise and jitter has been done and accounted for.
Now, with that out of the way, the question. Since Windows 10 now includes a native FLAC codec, is there any way in JRiver MC 20 to have JRiver use this codec instead of its native codec? We would like to run these same tests on the Windows 10 codec and see if there is any appreciable difference between the two. Keeping the rest of the processing environment the same (JRiver) is key to understanding if the differences (if any) are due to the codec itself or some other factors.
Thank You