OK, I just did some testing. I'm uploading a dropbox link of all of the test files since I really don't know how to analyze background noise, so hopefully those of you with the knowledge and tools can do this properly. I loaded all 4 files in RX4 which has the best spectrogram view of any of my software and zoomed far in to show the first 0.090 seconds of each file. The spectrograms clearly show more background noise on the M10 files, but that's to be expected since there's another device before it in the chain. Is it 25dB worse as is claimed in the dvxuser post? Hopefully someone here will be able to tell for sure.
The visual difference you will notice in the beginning of the screenshots is probably due to the fact that I was manually trying to start both recorders at the same time, but in both tests it looks like the 70D went first.
https://www.dropbox.com/sh/zx3g53ofi9fzaz8/AACyfwf9i9UsXHEVrNEiIeSGa?dl=0Here's a description of what I did:
FP24 Tape Out (1/8") > 70D EXT 1/2 IN (1/8") > 70D Line Out (1/8") > M10 Line In
FP24 sending 1kHz tone for both tests.
TEST 1:01_70D_150427_0010S12.wav
01_M10_150427_10.wav
70D and M10 levels set so both read -21dB while being sent the tone from the FP24. The 70D could not be set to -20, it was either -19 or -21 after many attempts.
TEST 2:02_70D_150427_0012S12.wav
02_M10_150427_11.wav
70D Slate Tone set to -18dB in SLATE menu, and that tone sent to the M10. M10 level set to -18dB to match.
Sending 1kHz from FP24 > 70D > M10 then read -28dB on M10 (still -21dB on 70D whose input levels were not touched from previous test).
After performing this test, connected FP24 directly to M10 with input level left the same from TEST 2. M10 showed -25dB.
EDIT: It just occurred to me that it might also be useful to repeat the same steps with the FP24's tone turned off while recording, so the 70D and M10 are only recording silence / background noise. I just packed everything away for the night though and won't have time to do any more testing until Thursday.