Even so, I think 25 minutes at 1:50 is somewhat longer than optimum for HP5+. I think if you were to do a test where you develope a whole roll of the same subject and exposure, removing a few frames from the developer at, say, 3 minute intervals after 15 minutes, you might see that the contrast had reached its maximum at around 15 and the overall level increased after that. I'm thinking of a set of curves Phil Davis published some years ago in Photo Techniques, maybe while it was still Darkroom and Creative Camera Techniques I can't find my copy right now. The contrast vs development time for Rodinal flattened out fairly soon. After that point, any change in density would give more fog.