Recently I stumbled upon on a different method of grading the programs in programming — where instead of the time limit, instruction limit is used (the amount of assembler code instructions CPU has done). That was the first time I've stumbled upon it and have never before seen system like it.

Hence, I am very interested in hearing opinions about what are the advantages and disadvantages of this method. From a quick overview, there is a massive advantage because amount of instructions isn't dependable on the performance of CPU. Again, from a quick overview I couldn't see any massive disadvantages, but because I've never worked with it before, I would love to hear the opinions of community. Maybe someone has already dealt with this approach.

Thanks in advance!