As a student currently taking CMSC 22200 - Computer Architecture at the University of Chicago, I want to make it clear that I have found this class highly interesting and extremely rewarding to take. The material is detailed, the readings abundant, and the labs exhaustive. Unfortunately, because we are testing out a new class structure this year, many of the assignments are being trialed for the first time here, resulting in many issues that still need to be ironed out:

• Unclear Instructions: Currently, the instructions for both labs now have not been either clear enough nor precise enough. The issues with lab 1 has already been documented by another student in a previous piazza post, so I'll mainly address the problems with lab 2. These ranged from information fragmentation (navigating the sea of information between the textbook, lectures, writeups, TAs, professor comments, Piazza), resolving inconsistencies and contradictions between these sources (the lectures and textbook use LEG v8 instructions and diagrams, but we are supposed to implement ARM v8 versions and expected to find all the differences between the two on our own - in fact most people didn’t even initially realize this distinction) and sometimes within the same source (sometimes TAs have given information that conflicted with each other or with the write up)
• Shifting Goalposts: We were given a golden simulator and told our simulators had to exactly match the golden one cycle by cycle. However, that simulator was later revealed to be faulty, rendering work that many of had previously done to be wasted and resulting in additional effort to match the new simulator which was released unfortunately close to the due date.
• Accumulative Labs: The labs build off of each other, resulting in an ultimate failure of a student if they are not up to speed. For the students that were unable to finish Lab1 and Lab2 by the due date, they are punished with having to fix their previous labs before they can even start on the new lab, which may take most of the time before even being able to execute the new lab. This results in a student being doomed for the rest of the quarter if they cannot fix one project, resulting in a cycle of failure. Other classes that have accumulative labs (e.g. CMSC 15200) give us working implementations for the next stage after each lab is done, and I think it would be of great benefit if this class did the same.
• Extreme Time Investment: Because of the aforementioned reasons, many students end up spending upwards of 30 hrs a week doing coursework for this single course. Although I am not opposed to spending more time learning when the material is inherently challenging, unfortunately many of the hours spent are used navigating the aforementioned difficulties. For students who do not manage to rise above these difficulties and fall behind, they are doubly penalized by having to invest additional time to finish previous labs.

We hope that an open and informed class discussion will proceed from this petition, and our concerns appropriately addressed.