This study examines the factors affecting the results of non-destructive testing of high-strength concrete performed on cubes and on cylinders and examines the processing of calibration relations. Tests were performed with both a type N and a type L Schmidt impact hammer (with a standard impact energy of 2.205 Nm and 0.735 Nm respectively). The assessed factors were internal stress in a specimen and the shape of the impact area. Test specimens were loaded by a force corresponding to the stress in specimen 0%, 10%, 20%, 30%, and 50% from the expected compressive strength. Rebound numbers of the unloaded test specimens were significantly lower than those of the loaded specimens. Therefore, calibration relations and/or correction coefficients processed by measurements of unloaded specimens can be assessed as unsuitable. To process calibration relations, we recommend exerting internal stress in amounts of 15% to 20% of the expected compressive strength of the tested HSC samples. During the determination of the effect of the shape of the test area on the cylindrical test specimen, it was assumed that the rebound numbers on the plane and the round test area would be the same. However, the test results revealed that the rebound numbers in the differently shaped test areas were different. For Schmidt impact hammer type N, the rebound numbers in the round test area were lower by 0.7 units on average, and for Schmidt impact hammer type L, the rebound numbers in the round test area were lower by 1.7 units on average compared to the plane test area rebound numbers.