Cooking methodology

The Cooking clusterhas two moving parts: ingredient density (which depends on the ingredient and how you measured it) and cup volume (which depends on which country’s convention you’re using). This page explains where both numbers come from.

Ingredient density data sources

Our density table covers 36 common ingredients. Sources, in order of authority for each ingredient type:

• USDA FoodData Central — the canonical authoritative source for flour, sugar, salt, oil, water, and most pantry staples. We use the unprocessed-form density (e.g., AP flour rather than self-rising).
• King Arthur Baking — the practical reference for baking ingredients (cocoa powder, baking soda, yeast, leavening agents). Their numbers come from extensive in-house testing.
• BBC Good Food — the cross-check for UK- conventional ingredients (golden syrup, treacle, double cream) where US sources are inconsistent or absent.

Where sources disagree (which happens for compressible ingredients like flour), we use the median of the available figures. The disagreement is itself meaningful — it’s the reason every serious cookbook now publishes in grams.

The cup volume question

Even after you fix the ingredient, “cup” means different volumes depending on the recipe’s country of origin:

• US customary cup: 236.59 ml (8 US fluid oz)
• US legal cup (on most US measuring cups): 240 ml
• UK metric cup: 250 ml
• UK imperial cup (rare, historical): 284 ml
• Australian cup: 250 ml
• Japanese cup: 200 ml

Our converter defaults to the US customary cup because that’s the dominant convention in recipes searched in English. The cup-type dropdown lets you override; see our US vs UK cup comparison for the full breakdown.

The packing problem

Dry ingredients vary in density depending on how the user scooped them. Flour can pack 25% denser than its un-tapped, unsifted state. Brown sugar is often measured “packed” (compressed firmly into the cup) rather than spooned. Powdered sugar is often sifted before measuring.

Our density values assume the “spoon and level” method — gently spoon the ingredient into the measuring cup without compacting, then level the top with a straight edge. This is the convention King Arthur Baking documents and most modern US cookbooks follow.

Output accuracy: within ±5% for the typical ingredient and measurement method, which is well inside the tolerance most recipes can absorb. For precision baking (souffles, perfect sourdough hydration), measure on a scale — our converter is a starting point, not a substitute.

Algorithm details: the four-step conversion

Every cup-to-grams conversion runs through the same deterministic pipeline; the regional cup, the ingredient density, and the requested target unit are the three independent inputs.

1. Resolve cup volume to millilitres. Look up the user-selected cup type in the regional cup table (US customary 236.588 mL, US legal 240 mL, UK metric 250 mL, Australian 250 mL, Japanese 200 mL, UK imperial 284.131 mL).
2. Resolve ingredient density. Look up the ingredient slug in the 36-entry density table (g/mL). For ingredients with multiple published values (flour, cocoa, oats) we store the median across USDA + King Arthur + BBC Good Food.
3. Compute mass. mass_g = volume_mL × density_g_per_mL × cup_count. All arithmetic in IEEE-754 doubles; the relative error of the floating-point step is < 10⁻¹⁵, well below the ±5% density uncertainty.
4. Convert to the target unit. Grams → ounces uses the international avoirdupois ounce (28.3495 g); grams → pounds uses 453.592 g. Both factors are exact by definition under NIST SP 811.

Oven-temperature conversion is independent of the density path and uses °C = (°F − 32) × 5/9 as the canonical formula. Gas Mark uses the lookup table below with linear interpolation between marks for non-standard inputs.

Sources & references

Density values are reconciled across USDA FoodData Central, the King Arthur Ingredient Weight Chart, and the BBC Good Food cooks’ reference. Where the sources diverge we use the median and surface the spread in our recipe density data study. Cup-volume definitions follow NIST Handbook 44 (US legal cup) and national metrology bureau publications for UK/AU/JP equivalents. See the Sources & references block at the foot of this page for full citations.

Assumptions & limitations

• Spoon-and-level packing only. Dip-and-sweep packs flour ~15-25% denser. If your recipe assumes dip-and-sweep (Joy of Cooking, ATK), our gram outputs for flour and cocoa will under-state by that margin.
• Room-temperature ingredients.Cold butter, refrigerated milk, and frozen berries all have slightly higher densities than the room-temperature values we store. The error is <2% for liquids, <5% for cold solids.
• No humidity or altitude adjustment. Flour absorbs 1-3% moisture in humid kitchens, which inflates the volume per gram. The density table is calibrated to ~40% relative humidity at sea level.
• Brand variance within an ingredient. European-milled flours are typically 5-10% denser than US AP flour. Almond flour density varies 20% between coarse and superfine grinds.
• No sifted vs unsifted distinction. The stored density assumes unsifted. Sifted weighs ~10% less per cup.
• Liquid measures assume 20 °C. Hot stock measured by volume is ~3% lower in mass than the same volume cold (ISO 4787 reference temperature).

Oven temperature math

See the broader units methodology for the Celsius ↔ Fahrenheit conversion. Gas Mark uses the canonical British table (Mark 1 = 275°F = 135°C, then 25°F per mark up to Mark 9 = 475°F). For non-standard temperatures, the calculator interpolates linearly between the two adjacent Gas Mark steps.