types of processor
Complex Instruction Set Computing | Reduced Instruction Set Computing |
|---|---|
used in laptops/desktop computers | used in smartphones and embedded systems |
complex hardware | simple hardware |
large number of complex instructions, variable lengths | smaller number of simple, standardised instructions, fixed length |
multiple machine/clock cycles per instruction | single machine cycle per instruction |
larger, more complex circuitry, requires more silicone to produce, therefore more expensive | smaller, less complex circuitry, less silicone required, typically cheaper to produce |
greater energy consumption | lower energy requirements |
can't support pipelining | can support pipelining |
compiler has to do less work | compiler has to do more work |
small code sizes, higher cycles per second | larger code sizes, lower cycles per second |
makes more efficient use of RAM | heavier use of RAM (risks bottlenecks if RAM is limited) |
multiple addressing modes available | fewer addressing modes available |
a graphics processing unit (GPU) is a a type of co-processor which (unlike CPUs) has lots of independent processors which work in parallel making it very efficient at completing repetitive tasks such as image processing and machine learning
highly specialised
made of thousands of stream processors
SIMD structure
a multi-core processor is a single chip with multiple independent cores that can complete instructions separately which results in higher performance
parallel processing is the process of dividing instructions between multiple processors or processor cores, to be executed simultaneously
Complex Instruction Set Computing | Reduced Instruction Set Computing |
|---|---|
used in laptops/desktop computers | used in smartphones and embedded systems |
complex hardware | simple hardware |
large number of complex instructions, variable lengths | smaller number of simple, standardised instructions, fixed length |
multiple machine/clock cycles per instruction | single machine cycle per instruction |
larger, more complex circuitry, requires more silicone to produce, therefore more expensive | smaller, less complex circuitry, less silicone required, typically cheaper to produce |
greater energy consumption | lower energy requirements |
can't support pipelining | can support pipelining |
compiler has to do less work | compiler has to do more work |
small code sizes, higher cycles per second | larger code sizes, lower cycles per second |
makes more efficient use of RAM | heavier use of RAM (risks bottlenecks if RAM is limited) |
multiple addressing modes available | fewer addressing modes available |
a graphics processing unit (GPU) is a a type of co-processor which (unlike CPUs) has lots of independent processors which work in parallel making it very efficient at completing repetitive tasks such as image processing and machine learning
highly specialised
made of thousands of stream processors
SIMD structure
a multi-core processor is a single chip with multiple independent cores that can complete instructions separately which results in higher performance
parallel processing is the process of dividing instructions between multiple processors or processor cores, to be executed simultaneously
