Hopefully, you can keep the profile smooth enough and appropriate for the bean to have it track the curve reasonably well.
That said, one can taste a finishing temperature difference of 1C or of a 20 second difference with the same finishing temperature. See this article…
https://millcityroasters.com/blogs/coff ... nsory-cues
My wife and I put it to the test. We were roasting a washed Ethiopian that is highly rated, but I find it to be too subtle. My goal was to pull more from the bean. We roasted our nominal profile plus four others: fast (-20s), slow (=20s), high (+1C), and low (-1C). We blind cupped two of each and could definitely taste the differences. Here are our experiences:
The nominal roast was already well balanced, but not exciting.
Low was more sour with some complexity.
Fast was sour and simple.
High had some astringency and tannins.
Slow retained balance and had more flavor. Bingo!
The cool thing is that this provides a roadmap. If there is a negative flavor to dial out, go faster. Adjust the end temp for sour/bitter balance. Lower the development curve and draw it out to get more flavor.
Happy roasting!