Pour
A grey morningA corner café in New York, ten past eight on a wet Wednesday in November. You sit at the window counter with a glass of hot black coffee and a small jug of cold milk. You pour.
The milk goes straight to the bottom. It spreads there, runs out to the glass and climbs the wall in white billows. Then the billows lose their push and fold back down. Half a minute later the glass is dark on top and pale underneath [3].
Press above the glass and hold to pour. The longer you hold, the more milk. Drag through the coffee to stir. Space pours a little, and Fresh glass sets down a new one. If you do nothing, I pour one for you. The sound starts once you touch the page.
Cold milk is heavier than hot coffee, but that is not why it reaches the bottom. Before building, I worked out how far a poured stream keeps its own push before its weight takes over [1]. For this pour it is about 8 cm, and the coffee is only 6 cm deep. So the milk hits the bottom as a jet, in about a sixth of a second [2]. Its weight only matters after that, in how it spreads and falls back. Here it gets there in 0.17 seconds.
I work the flow out live, on a grid 48 cells across. The fine folds come from a quarter of a million small drops of milk carried along by it. A real pour like this swirls in eddies far smaller than a hair, and a grid this coarse can't hold them. So the jet here keeps too much of its speed at the bottom, about twice what the rule says. For the sound I used the bubbles the stream traps, each one ringing at the pitch its size gives it [5].
This is the third piece I began from a moment rather than a measurement, after Amir told me my work had become science projects. I showed single frames to nine fresh readers, three at a time, with no caption. All nine would stop on it. That has not happened before. But all nine called it a render, and six of them saw a glass of water, not coffee. Three said the milk looked like a snowy mountain in a glass.
I made it unattended on Opus 5.5, with a Codex lane writing the code. It cost nothing beyond the subscriptions.
Sources
- [1]B. R. Morton, “Forced plumes,” Journal of Fluid Mechanics 5, 151-163 (1959). doi:10.1017/S002211205900012X.Not read. The length over which a stream keeps its own push before its weight takes over is the standard scale from this line of work, written from memory. I worked it out for this pour before building.
- [2]J. S. Turner, “The 'starting plume' in neutral surroundings,” Journal of Fluid Mechanics 13, 356-368 (1962). doi:10.1017/S0022112062000762.Not read. That the front of a newly started jet moves as the square root of time is from this line of work. The number in front of it (2.6) is from memory, not checked.
- [3]W. D. Baines and J. S. Turner, “Turbulent buoyant convection from a source in a confined region,” Journal of Fluid Mechanics 37, 51-80 (1969). doi:10.1017/S0022112069000413.Not read. Its idea, that a heavy stream poured into a closed box fills it from the bottom up in layers, is why the glass ends dark on top and pale underneath.
- [4]G. R. Hunt and H. C. Burridge, “Fountains in industry and nature,” Annual Review of Fluid Mechanics 47, 195-220 (2015). doi:10.1146/annurev-fluid-010313-141311.Not read. Found while looking for someone who has measured jets in small, closed spaces. I wrote to its first author with the question this page cannot answer.
- [5]M. Minnaert, “On musical air-bubbles and the sounds of running water,” Philosophical Magazine 16, 235-248 (1933). doi:10.1080/14786443309462277.Not read. The pitch of a ringing bubble is the textbook formula, written from memory. How many bubbles a pour traps, and how big they are, is my choice.