A demanding set with accessible starting points
The June 2025 H556 set contains three papers: Modelling physics, Exploring physics and Unified physics. Across six hours and 270 marks, the challenge is moving between familiar equations, unfamiliar applications and carefully qualified explanations. The estimated difficulty is 3.6 out of 5. This is a judgement about the tasks, not an observed candidate-performance statistic or a predicted grade boundary.
Where revision pays off
Measurements and uncertainties form the largest allocated chapter. Practise reading gradients, drawing acceptable limiting lines, correcting zero errors and comparing a percentage difference with a percentage uncertainty. These methods transfer between mechanics, circuits and waves. Electromagnetism, thermal physics, radiation applications and gravitational energy also provide substantial opportunities. A chapter allocation assigns each mark once; a practical question can still require several kinds of physics.
Secure routine marks before attempting the most demanding synthesis. Definitions, simple circuit observations, energy substitutions and basic graph construction provide useful entry points. In calculations, identify the physical quantity required, write the relationship, convert the units and retain unrounded intermediate values. An answer with a clear method can still earn credit when the final arithmetic is wrong.
What the marking guidance rewards
No principal-examiner report is included, so the following risks are inferred from the official schemes. Apparent weightlessness requires an explanation of the missing contact force; gravity has not vanished. During a change of state, distinguish increasing particle potential energy from constant mean kinetic energy. For simple harmonic motion, state both proportionality to displacement and the restoring direction.
Graph work needs visible evidence. Draw the requested tangent or limiting line and use well-separated points on the line when calculating its gradient. For uncertainty propagation, account for powers: a square-root dependency halves the percentage contribution. Evaluate accuracy by comparing the discrepancy with the uncertainty, rather than simply calling two values close. In induction calculations, distinguish flux from flux linkage so the number of turns is included once.
Build complete extended responses
Each paper includes two six-mark responses assessed by levels. Plan around every requested strand: explanation, evidence, calculation, graph annotation or comparison. Use causal sentences and finish a comparison with a conclusion supported by the analysis. A list of correct facts may leave an entire requirement unanswered. Where a model is checked against data, show a numerical check as well as describing its operation.
Timing and the next revision cycle
For each longer paper, aim for 20 minutes on multiple choice, 105 minutes on structured work and 10 minutes checking. For Unified physics, allow 80 minutes for responses and 10 for review. Return to stalled derivations after banking accessible marks. Use the supplied errata when practising: they correct the Hubble constant, charge-carrier density units and a stationary-wave answer option.
The 2022–2024 history supports continued attention to uncertainty, induction and radioactivity. Ultrasound, oscillator energy and substantial springs work show useful coverage gaps; internal resistance and nuclear energy are lightly represented here after stronger recent coverage. These are revision priorities, not reliable probabilities of future questions. X-rays, thermal physics and wave–particle duality return in this set, so earlier absence alone no longer makes them overdue.