Skills

Physics capstone: a simulation or data paper

10 weeks · 2 milestones

An 8–15 page paper (or documented notebook) on a topic beyond the core: a simulation of a real system or an analysis of open data, with the full error analysis, then a recorded 10-minute talk. The last step of the Varsity Curriculum: Physics, read by a blind review for Honors.

Part of: Physics · Proves for: UG, Career

Milestone map

2 milestones

Write the capstone paper

8–10 weeks (about 12 hrs/week)

Pick a topic beyond the core and write an 8–15 page paper in LaTeX, or a documented notebook, in a public repo. It is either a simulation of a real system or an analysis of open data (for example LIGO, CERN or NASA), and it shows the full error analysis. This is what the Honors blind review reads.

Proof required

Link the public repo with the paper (PDF and LaTeX source, or the notebook), the code and data or the data's source, and the error analysis.

What gets checked

The question is beyond the core courses and stated at the start.
The error analysis covers statistical and systematic uncertainty.
Someone else could rerun every figure from the repo.

Common mistakes

A summary of other people's work with no result of your own.
A result with no error analysis.

Powstik Guide

An 8–15 page paper (or a documented notebook) with code, data and a full error analysis.

Steps

Choose a real system or an open dataset.
Write the question and what you expect.
Build the simulation or analysis in a public repo.
Work out every uncertainty.
Write the paper: question, method, result, uncertainty, what it means.

Template

Paper
Title:
Question:
Method (simulation or data, and why):
Result ± uncertainty:
Systematic checks:
What it means and its limits:

What gets sent back

No error analysis.
Figures that the repo cannot reproduce.

Resources

Foundationstart here

What a verifier looks for

Is the physics right?
The uncertainty is worked out from the data or runs (spread of repeats, instrument limits, fit errors or a range of inputs), not guessed, and it is shown on the result.
Do the code and data reproduce the figures?

You'll sign in first, then come straight back here.

Give a 10-minute talk on it

1–2 weeks

Record a 10-minute talk on your capstone for people who know physics, then take their questions.

Proof required

Link the recording and the slides, and say who the audience was (for example a study group or a physics club).

What gets checked

The talk fits in 10 minutes.
The main result is stated with its uncertainty.
Questions are answered honestly, including what you don't know.

Common mistakes

Reading the paper aloud.
Avoiding the hard questions.

Powstik Guide

A recorded 10-minute talk with questions.

Steps

Plan one slide per idea.
Rehearse to 10 minutes.
Invite people who know physics.
Record the talk and the questions.

Template

Talk plan
One-line result:
Three slides that get there:
Questions I expect:

What gets sent back

No recording.
No questions taken.

Resources

What a verifier looks for

Is the result stated with its uncertainty?
Were questions answered honestly?
Does the talk match the paper?

You'll sign in first, then come straight back here.

Part of

Physics Curriculum: Capstone →

Related outcomes

All outcomes

Skills

Physics capstone: a simulation or data paper

10 weeks · 2 milestones

Part of: Physics · Proves for: UG, Career

Milestone map

2 milestones

Write the capstone paper

8–10 weeks (about 12 hrs/week)

Proof required

Link the public repo with the paper (PDF and LaTeX source, or the notebook), the code and data or the data's source, and the error analysis.

What gets checked

The question is beyond the core courses and stated at the start.
The error analysis covers statistical and systematic uncertainty.
Someone else could rerun every figure from the repo.

Common mistakes

A summary of other people's work with no result of your own.
A result with no error analysis.

Powstik Guide

An 8–15 page paper (or a documented notebook) with code, data and a full error analysis.

Steps

Choose a real system or an open dataset.
Write the question and what you expect.
Build the simulation or analysis in a public repo.
Work out every uncertainty.
Write the paper: question, method, result, uncertainty, what it means.

Template

Paper
Title:
Question:
Method (simulation or data, and why):
Result ± uncertainty:
Systematic checks:
What it means and its limits:

What gets sent back

No error analysis.
Figures that the repo cannot reproduce.

Resources

Foundationstart here

What a verifier looks for

Is the physics right?
The uncertainty is worked out from the data or runs (spread of repeats, instrument limits, fit errors or a range of inputs), not guessed, and it is shown on the result.
Do the code and data reproduce the figures?

You'll sign in first, then come straight back here.

Give a 10-minute talk on it

1–2 weeks

Record a 10-minute talk on your capstone for people who know physics, then take their questions.

Proof required

Link the recording and the slides, and say who the audience was (for example a study group or a physics club).

What gets checked

The talk fits in 10 minutes.
The main result is stated with its uncertainty.
Questions are answered honestly, including what you don't know.

Common mistakes

Reading the paper aloud.
Avoiding the hard questions.

Powstik Guide

A recorded 10-minute talk with questions.

Steps

Plan one slide per idea.
Rehearse to 10 minutes.
Invite people who know physics.
Record the talk and the questions.

Template

Talk plan
One-line result:
Three slides that get there:
Questions I expect:

What gets sent back

No recording.
No questions taken.

Resources

What a verifier looks for

Is the result stated with its uncertainty?
Were questions answered honestly?
Does the talk match the paper?

You'll sign in first, then come straight back here.

Part of

Physics Curriculum: Capstone →