Heat-assisted-magnetic recording (HAMR) is hoped to be the future recording technique for high density storage devices. Nevertheless, there exist several realizations strategies. With a coarsegrained Landau-Lifshitz-Bloch (LLB) model we investigate in detail benefits and disadvantages of continuous and pulsed laser spot recording of shingled and conventional bit-patterned media. Additionally we compare single phase grains and bits having a bilayer structure with graded Curie temperature, consisting of a hard magnetic layer with high TC and a soft magnetic one with low TC, respectively. To describe the whole write process as realistic as possible a distribution of the grain sizes and Curie temperatures, a displacement jitter of the head and the bit positions are considered. For all these cases we calculate bit error rates of various grain patterns, temperatures and write head positions to optimize the achievable areal storage density. Within our analysis shingled HAMR with a continuous laser pulse moving over the medium reaches the best results, and thus having the highest potential to become the next generation storage device.