I heard a difference between the MP3 and wave versions, except I hated the wave version! I found it to be -worse- than the MP3. It was [cold], its musical life had been [erased] somehow. I would certainly not create a master from -this- version for pressing. It does not produce the same harmonics as the MP3 versions I downloaded (yes, plural. Long story short, I ended up comparing 4 copies of the MP3 vs. 1 of the WAV. file). I suggest something in the conditions was different when the WAV was produced, and there’s a lot lot more going on here than the FR graph or null test would indicate.
By the way, I came across this page because I was searching Google to find if people think there’s a difference between an MP3, and the converted WAV file created from that same MP3. Or more specifically, the disc you hear created from the MP3. Reason being, I just burned a disc from MP3 files, compared it to the original MP3s in Windows Media Player, and the improvement in quality from the CD is tremendous. Though I admit I did make an attempt to tweak the burn for better SQ.
There is no possible way that an MP3 converted back to WAV will sound any different to the original MP3. When you decompress an MP3 it rebuilds the WAV using the info in the MP3, resulting in a file that will sound exactly the same as the MP3. The same is true for CDs burned from MP3s. I don’t know what “tweaks” did you used, but all of the software I’ve seen that “restores WAV quality to MP3s” just use an EQ curve to accent certain frequencies, as it is mathematically impossible to restore WAV quality to MP3, as the data has physically been lost.