Computing Systems代写 | COMPSCI 313 S2 Mini-Project



The objective of this project is to evaluate the performance of different mobile computing system. Evaluating
and testing the performance of a mobile application is not as straightforward. With the completion of this project,
the students can understand what the most important factor for the performance of the mobile computing system.
Especially, the students can design the mobile computing system with the most suitable processor based on some
particular applications.

Group Report Guideline

Comparing the performance of processors in modern smart phones

• Compare a few processors based on main producers:
o Qualcomm / Samsung / Apple / MediaTek / HiSilicon
• Evaluate the performance by different evaluation software:
o AnTuTu / NenaMark / Vellamo / Basemark / GFXBench / GeekBench / SpeedTest

Usage of Evaluation Software:

Group formation:
• 1 – 5 students
• The group with n students should compare n-1 processors (minium 2 processors)

Report format:
• 12 pts, single spacing, single column
• 8 pages at MOST

Marking Criteria

A. Description of the specification of the processors.

The report should clearly describe the specification of the mobile computing systems/processors that you are
going to compare. The technical specification includes a phone model, processor model, instruction set
architecture, micro-architecture, GPU model, Fab technology, and storage and memory organization.

B. Description and explanation of the characteristics and performance of the processors.

The report should clearly describe the characteristics of the processor. Are they designed using Von Neuman
architecture? Are they RISC or CISC? Is the pipelined technique applied? Are there any parallel execution
technology such as speculative execution, branch prediction, etc.. applied? In addition, the report should also
briefly identify the major difference between the processors. Explain the difference based on the knowledge you
have learnt from the class.

C. Verification of the performance difference with evaluation software.

The report should verify your explanation with the experimental results. You can use a combination of listed
software above. This section should provide evidence to support your statement made in the previous section.