Well, you could utilize more precise operations but, according to Wikipedia, games rarely took advantage of it:
"The Nintendo 64's central processing unit (CPU) is the NEC VR4300, a cost-reduced derivative of the 64-bit MIPS Technologies R4300i. Built by NEC on a 0.35 µm process, the VR4300 is a RISC 5-stage scalar in-order execution processor, with integrated floating point unit, internal 24 KB direct-mapped L1 cache (16KB for instructions, 8KB for data). The 4.6 million transistor CPU is cooled passively by an aluminum heatspreader that makes contact with a steel heat sink above.
Clocked at 93.75 MHz, the N64's VR4300 was the most powerful console CPU of its generation. Except for its narrower 32-bit system bus, the VR4300 retained the computational abilities of the more powerful 64-bit MIPS R4300i, though software rarely took advantage of 64-bit data precision operations. N64 game-titles generally used faster (and more compact) 32-bit data-operations, as these were sufficient to generate 3D-scene data for the console's RSP (Reality Signal Processor; see below) unit. In addition, 32-bit code executed faster and required less storage space (which was at a premium on the N64's cartridges) Though powerful, the CPU was hindered by a 250 MB/s bus to the system memory; not only that, but in order to access the RAM, the CPU had to go through the RCP (Reality Co-Processor), and could not use DMA to do so (the RCP could). This problem is further compounded by the RDRAM's very high access latency.
Emulators—such as UltraHLE and Project64—benefit from the scarcity of 64-bit operations in the game's executable-code, especially when running with a 32-bit machine architecture as a host. These emulators perform most calculations at 32-bit precision and trap the few subroutines that actually made use of 64-bit instructions."
"The Nintendo 64's central processing unit (CPU) is the NEC VR4300, a cost-reduced derivative of the 64-bit MIPS Technologies R4300i. Built by NEC on a 0.35 µm process, the VR4300 is a RISC 5-stage scalar in-order execution processor, with integrated floating point unit, internal 24 KB direct-mapped L1 cache (16KB for instructions, 8KB for data). The 4.6 million transistor CPU is cooled passively by an aluminum heatspreader that makes contact with a steel heat sink above.
Clocked at 93.75 MHz, the N64's VR4300 was the most powerful console CPU of its generation. Except for its narrower 32-bit system bus, the VR4300 retained the computational abilities of the more powerful 64-bit MIPS R4300i, though software rarely took advantage of 64-bit data precision operations. N64 game-titles generally used faster (and more compact) 32-bit data-operations, as these were sufficient to generate 3D-scene data for the console's RSP (Reality Signal Processor; see below) unit. In addition, 32-bit code executed faster and required less storage space (which was at a premium on the N64's cartridges) Though powerful, the CPU was hindered by a 250 MB/s bus to the system memory; not only that, but in order to access the RAM, the CPU had to go through the RCP (Reality Co-Processor), and could not use DMA to do so (the RCP could). This problem is further compounded by the RDRAM's very high access latency.
Emulators—such as UltraHLE and Project64—benefit from the scarcity of 64-bit operations in the game's executable-code, especially when running with a 32-bit machine architecture as a host. These emulators perform most calculations at 32-bit precision and trap the few subroutines that actually made use of 64-bit instructions."
http://en.wikipedia.org/wiki/Nintendo_64