The Scent

An applied field guide to contamination after Tindalosian contact

Mercer never saw the thing again.

He changed trains seven times. He paid cash, threw away his clothes, shaved his head, and would only sleep in rooms whose corners he could smother with pillows. He stopped using his own name and he never went home. Three weeks later a landlady in Prague found him folded into an angle far too small to hold a man.

Everyone assumed the Hound had followed him. But the trail worth studying did not end with Mercer at all. It ended with his sister, back home, who still had the letter he had written to explain why he ran.

What the Hound follows

A Hound of Tindalos does not track you. It tracks the physical wake of your having been seen. Footprints, blood, body heat, and geography only matter to it when they happen to carry or connect those records, which is why running changes where the attack eventually lands without buying you a single meter of real distance from it.

Being seen changes the witness, and the witness will not shut up about it. He remembers. He talks, writes, photographs, hides, buys plaster, learns to hate certain rooms, and scares other people into doing the same. Every one of those acts leaves behind a physical state that almost certainly would not exist in a world where the encounter never happened. The scent is the sum of those differences, minus the ones that have already become impossible to recover.

Day One’s Tindalosian Observation Problem laid out the machinery: observation ties an angular thing to ordinary matter, decoherence smears the records of it across the world, and the right geometry gives the resulting Hound field somewhere to sit. This guide is about what you do with those rules once they are in play, both in the fiction and at the table. It comes down to one uncomfortable sentence.

The Hound does not track where you went. It tracks what would not have happened if you had never seen it.

Not every butterfly smells of Tindalos

Take that literally and you are in trouble, because every event changes the future. Mercer breathes a little differently, the air moves a little differently, molecules collide in a slightly altered pattern, and eventually you have a whole universe microscopically unlike the one where he saw nothing. If a Hound could smell all of that, the scent would spread without limit and could never be scrubbed out. That is bad physics for our purposes, and it makes for an even worse game, since it hands the players a problem with no handholds.

So the Hound only answers to recoverable records: stable physical systems that still carry enough information to tell “a Hound was observed” apart from “a Hound was not.” A jostled molecule is merely different. A letter describing the encounter is a record. A wall someone rounded off with fresh plaster is nothing at all until a person works out what rounded walls are for, and then it becomes a record too.

That line between difference and record is where all the interesting play lives. It leaves room to investigate, to hide things, to sacrifice, and to make expensive mistakes, and it gives the table something they can actually get their hands on.

The Scent Track

Drop this six-level track into whatever system you run. It is not counting how many people know the truth. It is measuring how strong, how redundant, and how recoverable the surviving records are.

The track is coarse on purpose. Move it when the situation genuinely changes, not every time somebody sneezes near the evidence.

What raises the scent

• Tell someone who believes you: +1.
• Make a durable description, photograph, recording, or specimen: +1.
• Get independent corroboration: +1.
• Copy a record into another physical system: +1 when the copy actually creates redundancy, not once per trip to the photocopier.
• Publish or upload credible evidence: straight to at least 4, and to 6 if it spreads past any hope of recall.
• Repeat the original experiment: +2, and hand the Hound a brand-new primary record while you are at it.
• Investigate strange defensive architecture: +1 if the investigation correctly works out what it was for.
• Prove an apparent hoax contains real material: +2.

Lies are their own trap. A lie can bury the true record while the fight over it spawns a dozen new systems all encoding the disputed event. Confusion can raise the scent even as it wrecks any human attempt to reconstruct what happened.

What lowers it

• Destroy the only durable record before it is copied: −1.
• Induce real amnesia in the only witness: −2.
• Split fragments so no one observer can reassemble their joint meaning: −1.
• Bury one clear record under several incompatible explanations: suppress manifestation by 1 level, temporarily.
• Erase the causal reading that ties otherwise meaningless evidence together: −1 or −2.
• Eliminate every independently recoverable record: down to 0.

Do not treat death as an automatic discount. A corpse can be a far better record than a live witness ever was. How he died, which room he died in, the defensive wounds, the autopsy photographs, the investigation that follows, any of it can hold or sharpen the scent. Burning a journal works, sometimes. Other times it just leaves a conspicuous hole in an otherwise unbroken archive, and the hole is the record now. Ashes have no idea what was burned. Investigators do.

Threshold and hysteresis

The Hound should not pop into being every time the number ticks up. Make the first stable manifestation wait for Scent 3. Once it has a body, though, it does not leave until the track drops below 2. That gap has a name, hysteresis: it takes more to end a state than it took to start it.

At the table:

• Scent has to hit 3 before a Hound can form a stable body.
• Dropping from 3 to 2 stops new manifestations but does not banish the one already standing in the room.
• Dropping to 1 starts visible decay.
• Reaching 0 cuts the last thread.

That asymmetry is what makes the endgame lethal. A nearly-freed Hound is not grateful and it is not calm. It can feel the last few anchors pinning it inside a body it hates, and it goes after them harder the closer it gets to loose.

What it goes after

The Hound picks targets by how much each one contributes to holding it here. Guilt does not enter into it, and neither does whose name is on the character sheet.

Fair game includes:

• The original witness.
• Anyone who has reconstructed the encounter.
• A camera or recorder that caught it.
• An archive with reliable evidence in it.
• A server holding distributed copies.
• A lab repeating the experiment.
• A room whose geometry and records happen to prop each other up.

It will walk past an armed guard to gut a filing cabinet. It will tear through a witness to get at the recorder in his coat pocket. It will kill the investigator specifically, because her clean, coherent model of events is a stronger record than the muddled memories of everyone else in the building put together. To us it reads as a hunt. To the thing doing it, it is escape. We forced it into a fixed place, a fixed order, a body, and a self, and every one of those record-making systems is a nail it is trying to pull.

Running it fairly

“The monster finds you because quantum mechanics” is just railroading in a lab coat. The players have to be able to learn the rules and make real choices inside them, and that means giving them evidence that tells one cause from another:

• The Hound shows up at the archive, not at the witness.
• A man who genuinely forgot is left alone until a therapist digs the memory back out.
• A burned photograph stops the manifestations; its forgotten cloud backup does not.
• Two half-records are harmless apart and lethal once someone lays them side by side.
• A false confession draws nothing until it accidentally includes one true detail.
• Move a record and the manifestation site moves with it, even though the witness stayed put.

Let containment be possible, but make it cost something. The real decisions here are about knowledge. Who has to forget? Which evidence is worth keeping even though keeping it might kill you? Can you herd the scent somewhere you can defend? Is destroying the truth actually kinder than leaving the Hound to suffer? And the worst one: what do you do when the only complete record left is a person?

Techniques the desperate use

Concentration

Gather every recoverable record into one geometrically controlled room. It does not lower the scent, but it pools the source and makes the manifestation predictable. It is also, unavoidably, torture. You are forcing the thing into a body of your choosing, in a spot of your choosing, and holding it there.

Fragmentation

Break the evidence up so no single fragment can rebuild the encounter, hand the pieces to different custodians, and put procedures in place to stop anyone reassembling them. This buys you exactly as long as it takes for one determined person to collect the set.

Contradiction

Flood the record network with mutually incompatible accounts. Done well, it lowers recoverability without destroying the real evidence. Done at all, it is unstable, because a good enough analyst will spot the shared skeleton under the competing lies and end up with a cleaner record than the one you were trying to hide.

Decoy anchoring

Set a carefully built record beside a prepared corner, so the Hound localizes there instead of near your people. The catch is that the decoy has to carry genuine correlation. A wholly invented account shares no causal ancestry with the encounter and gives off no scent at all, so you are back to sacrificing real evidence, or someone who remembers.

Causal excision

Do not just destroy the records. Take apart the story that explains how they relate: alter memories, change dates, make the defensive architecture look ordinary, give every gap a boring cause. This is the whole basis of The Unremembering, and it is hard, because the operation itself throws off consequences that a later investigator can recognize as the fingerprints of a deliberate erasure.

Scenario seeds

The Backup

Every known copy of an occultist’s journal is ash, and the manifestations have not stopped. The last copy is sitting in an automated off-site backup set up years ago by a systems administrator who is now dead. Restoring it will reveal which file matters, and by restoring it the investigators strengthen that file just enough for the Hound to arrive beside the recovery terminal.

The Rounded House

A family inherits a house whose rooms were obsessively rounded off. The renovators strip the plaster back to the original corners and find no idols, no sigils, just old invoices for plaster and a clause in the deed forbidding any historical research into the property. Right now the records sit below threshold. The people most likely to push them over it are the investigators.

The False Witness

A cult has been seeding elaborate fake encounters to hide one real one. The fakes give off nothing. Line them up, though, and a single detail turns up only in the accounts written by someone with access to the genuine event. Pin down that detail and you have just built the first reliable reconstruction, which is the last thing anyone wanted.

The Archive Fire

Burning the archive dropped the scent almost to zero. Then photographs of the fire went viral, and in several of them the flames are bending around a shape standing in the building’s corners. The evidence that burned inside has been replaced, neatly, by evidence of its burning.

The Amnesiac

The only surviving witness remembers nothing and is no longer hunted. A therapist, convinced the blank stretch is hiding ordinary trauma, starts patiently rebuilding it. Every good session nudges the Scent Track up another notch.

The Decoy

Investigators wall a single authenticated record inside a containment chamber. The plan works flawlessly. The Hound manifests inside it and spends every second it has left breaking itself against the curved walls. The question stops being whether containment works. It becomes how long the people outside are willing to keep listening.

The machinery beneath the game

For anyone who wants the physics under the hood, let be an environmental fragment and the classical information you need to tell a Hound encounter apart from its absence. Fragment counts as a viable record once its Holevo information clears the bar,

A system-neutral scent score is then just a tally,

with the physical coupling of each record carrier to the angular sector and a threshold function. It is bookkeeping, not a law of nature, but it keeps three things straight that the story needs kept straight: information has to be physically encoded somewhere, not every microscopic consequence is a usable record, and the more independent records exist, the harder the event is to erase. The full derivation of all of it, environmental distinguishability, redundancy, Hound-field coupling, corner modes, lives in The Tindalosian Observation Problem. At the table, the track is just that model wearing the shape of the Hound’s trap.

trollboy

I’m trollboy. Howdy.

www.shoggoth.net/members/admin