CUDA GPU Occupancy Calculator を使って占有率等をみるときにActive Thread Block per Multi Processor が高いほうがいい
と聞くのですが、以下の二つの状況だとどちらがいいでしょうか?GTX480を用いています。

1.
Enter your resource usage:
Threads Per Block 256
Registers Per Thread 32
Shared Memory Per Block (bytes) 3072
GPU Occupancy Data is displayed here and in the graphs:
Active Threads per Multiprocessor 1024
Active Warps per Multiprocessor 32
Active Thread Blocks per Multiprocessor 4
Occupancy of each Multiprocessor 67%

2.
Enter your resource usage:
Threads Per Block 1024
Registers Per Thread 32
Shared Memory Per Block (bytes) 12288
GPU Occupancy Data is displayed here and in the graphs:
Active Threads per Multiprocessor 1024
Active Warps per Multiprocessor 32
Active Thread Blocks per Multiprocessor 1
Occupancy of each Multiprocessor 67%

1と2の違いはThread Per Blockを四倍にしたかどうかです。
つまりSM内のThread Block数が多いほうが単純にいいのか、BLOCK内のThread数が多いほうがいいのか、
あるいはどちらも同じなのか教えてください。