Deep Mode shows your phrase for seventeen milliseconds, once every seven seconds. Both numbers were argued over, so this is the record of why they won.
Seventeen milliseconds is one frame on a common 60-hertz display. Any shorter and the operating system cannot reliably draw it; much longer and the flash crosses into something you consciously read, which is a different experience — that is what the ordinary motions are for. One frame is the shortest presentation a real desktop can honestly promise, so that is the number. On faster monitors we hold the duration rather than the frame count, because the promise is about time, not hardware.
Perception research has studied presentations in this range for decades: stimuli this brief typically slip beneath conscious identification, yet the visual system registers them — work by researchers like Sid Kouider and Stanislas Dehaene has traced how much processing happens in that silent window. A single frame sits right at the edge of awareness, which is exactly where Deep Mode is meant to live: present enough to count, quiet enough to never interrupt.
Seven seconds is the humbler number. Flash too often and the flicker itself becomes noticeable as rhythm, which defeats the point. Too rarely and the daily count collapses. Seven seconds yields several hundred quiet repetitions in a working day while staying beneath the threshold where testers reported noticing a pattern.
One number we did not negotiate: safety. Deep Mode is exactly why SOMOTE shows a photosensitivity warning before anything flashes, and why the app will not begin until it has been read and acknowledged. If you or anyone who shares your screen has a history of photosensitive seizures, please read that warning and talk to a doctor first.
Everything else — typeface, size, opacity — Deep Mode inherits from your settings. It is the same words, delivered in the quietest way we know how to build.