It is without doubt that the music on the CD #9121-2 (Prokofiev: The Complete Works for Piano Vol. 1)on the Concert Artist label, which purports to be played by Joyce Hatto, is actually music taken from CD's recorded by Oleg Marshev on the Danacord Label.
Track no. 1 on the "Joyce Hatto" disk is Prokofiev's Piano Sonata No. 1 in F minor, Op. 1, a one movement (Allegro) piece. It is identical to track no. 1 of the recording by Oleg Marshev on Danacord 392, Sergei Prokofiev: Piano Music Vol. 2.
Snippets from the end of the piece, lasting about 17 seconds, are compared below via screen captures of waveforms displayed by the sound editing software, CoolEdit2000. The first is Hatto's, the second is Marshev's. The time durations of the viewed selections are identical, at 17.50 seconds each, so the images may be compared.
The next image is an XOR of the two images above — it is the the image data that is in one but not in the other. It is clear that the only discernible difference between them is the amplitude of the sound selections. The waves stop and end at virtually identical times. This is an entirely interpretive part of the music; there is no way two humans will play it so identically. (Not even the same human.) And by the way, the rest of the piece is identical, as well. This isn't the best way (even optically) to conclusively prove these are the same performances, but it works for me.
Perhaps it is even more convincing to the the XOR difference after first colorizing tone of the images. The next image is Hatto's, colorized.
The one that follows is the colorized Hatto XOR'd with the original Marshev. To me it is overwhelmingly obvious.
For an important distinction, here is a wave form of the same duration from a recording of Boris Berman (Chandos 9017, Prokofiev: Vol. 5 of the complete piano music), folowed by the Berman image XOR'd with the colorized Hatto:
In Berman's case, the 17.500 seconds used in the Hatto-Marshev comparison is just shy of the duration of Berman's ending (when started from the appropriate position — in each of these cases, the sounds are synchronized at 0.065 seconds before the first distinguishable chord in the sequence of chords I selected for the comparison).
Another easy comparison just involves stacking the images atop one another. Here are Hatto, Marshev and Berman at once:
If seeing isn't believing, here are MP3's of the endings shown above:
Hatto Marshev Berman