Skip to main content
⌖ OF Intel

GUIDES

OpenFront Missile Silo Strategy: Upgrade, Second Silo, and the 9-Second Cooldown

The missile silo costs 1M per level, holds one nuke per level, and fires on a flat 90-tick (9-second) cooldown with no range limit. Learn when to upgrade, when to build a second silo, how the U-LF burst window works, and where to place the silo so it survives.

Strategy Difficulty · Advanced Published Nov 20, 2026 Updated Nov 20, 2026 Reviewed by OpenFront Intel editors #silo#nuke#nuclear#structure#cooldown#strategy#endgame

Direct answer: read the cooldown, not the level

The decision you keep re-asking is “do I upgrade this silo or build a second one,” and the answer the v0.36.5 source forces is to optimize for the reload window, not for a single high number. A silo in v0.36.5 reloads on a flat 90-tick cooldown, which is nine seconds at the configured game rate where the engine ticks ten times per second, and that window does not change with level. Upgrading a silo never makes the next launch come sooner; it only stores one more warhead. So a level-3 silo and a level-30 silo both fire again nine seconds after their front launch, but the level-30 can deliver thirty warheads before it is empty and the level-3 can deliver three. That is the whole trade, and it is the reason the endgame nuke meta splits into two distinct plays that players constantly conflate. The first is a deep single silo that delivers a large stock of warheads behind one nine-second clock. The second is a field of several shallow silos that deliver a smaller stock but keep a second and third nine-second clock running in parallel, so the combined output never fully stops. Build a second silo when you need a second reload window running alongside the first, and upgrade when you need a deeper stock of warheads behind one window. Because the silo has no firing range limit, the placement question is not “can I reach that target” but “does this tile keep the silo alive long enough to spend its stock,” and because the burst keybind staggers launches across ticks, a loaded silo turns its multi-minute output into a short, dense window you should time to a map moment rather than fire on your own convenience. This guide gives the exact numbers behind each of those moves, plus the two scenarios where they are actually decided, so you can stop treating the silo as a single number on a single bar and start treating it as a capacity-versus-rate resource with a nine-second clock attached to it. The single habit to build from this guide is to read the reload window before you spend a million gold, because the window is the constraint that no amount of gold on the same silo can shorten, and the only lever that raises your sustained rate is adding a second independent clock. Before you spend a single million, check the two in-game tells that confirm this on your own server: the silo’s reload bar, which resets on a nine-second clock that does not drain faster when you upgrade, and the launch availability, which stays locked until that clock finishes. If both match the v0.36.5 figures, the rest of this guide applies to your match without modification.

The 90-tick reload window, and why level never shortens it

The reload timer is the single most important number in this guide, and it is a constant. In the v0.36.5 source the silo enters cooldown for the value of the SiloCooldown constant, which is 90 ticks, and the cooldown is measured from the front time of the launch, the moment the earliest loaded missile actually fires. At the configured game rate, where the engine ticks ten times per second, 90 ticks is exactly nine seconds of real time. Two consequences follow from that, and both are load-bearing for how you play the structure. First, a silo that is currently loading a missile cannot fire again until those nine seconds elapse; a second launch click does not queue a faster shot, it either waits for the window to close or does nothing at all, so you cannot stack two launches to beat the clock. Second, the window is identical at every level, because the cooldown reads a constant and never scales with the silo’s level. This is the point that separates the silo from most structures in the game, where an upgrade buys a real rate, range, or power improvement. Here the upgrade buys capacity only, and nothing else. The one generated-data note that claims “upgrades reduce reload time” is stale relative to the v0.36.5 tag, and you should not plan around it. When you watch your own silo in-game, the reload bar and the launch availability reset together on that nine-second clock, and no amount of gold spent on the same silo makes that bar drain faster. The practical reading is that your sustained nuke output per silo is capped by a nine-second reload between launches, and the only way to raise the sustained rate is to run a second silo so its own nine-second clock starts while the first one is still counting down. This also means the reload window is a visible resource you can time: you can watch the enemy’s silo reload, know exactly when it will be empty and when it will re-arm, and plan your burst to land either before their re-arm or in the gap between two of their own launches. The nine-second clock is the metronome the whole endgame nuke exchange marches to, and every decision in this guide is a decision about what to do inside or against that metronome. The reason the constant matters so much is that it is the only number on the structure that the enemy can read in real time: they can watch your reload bar, they can see when you are empty, and they can time their own burst to the moment your clock re-arms, so the nine-second window is not just your constraint, it is also the enemy’s window of opportunity against you.

Capacity is level: one warhead per level at a flat 1M

The silo’s level is not a power level; it is a slot count. In the v0.36.5 source a silo’s level() equals the number of warheads it currently holds, and each launch consumes exactly one slot, so a silo at level 5 has five loaded warheads and can fire five times before it must reload. Building the first silo costs one million gold, and every upgrade that adds a warhead also costs one million gold, with no escalation by count, so a silo that has been upgraded to level 5 represents five million gold of cumulative investment in a single structure. That flat, per-level pricing is what makes the upgrade-versus-second decision clean and arithmetic. To go from level 5 to level 6 costs one million, the same one million that would buy a brand-new second silo at level 1. There is no discount for adding a slot to an existing silo, and no premium for starting a new one, and that symmetry is the entire content of the decision. The practical ceiling is the upgrade batch limit, which the engine caps at fifty upgrades per command, so a single bulk-upgrade input cannot push a silo past fifty levels in one action, though repeated inputs can go higher, which is why the very deep silos you see in endgame games are the product of many separate upgrade commands rather than one. This is why the endgame silo that some players chase, a level-20 or level-30 structure representing twenty to thirty million gold, is a real and deliberate commit, not a default. The community warning about the level-20 reflex is exactly this: the last few levels cost the same as the first, so a max-level silo is a stack of one-million increments all spent on a structure that still reloads in nine seconds. Reading it the other way, a silo at level 1 is a one-million structure that can fire one warhead and then sit idle for nine seconds, which is why a lone level-1 silo is rarely the end of a plan; it is the first brick of either a deeper silo or a second clock. The generated units data lists the silo at a flat 1M cost with no level scaling, which matches the source, and it is worth re-checking that the data you read in the in-game UI agrees with the flat one-million-per-level figure before you commit a large bank to a deep silo, because a misread cost curve is the most expensive mistake this structure can hand you.

The burst window: U-LF and staggered launches

Because a loaded silo can fire several warheads before its nine-second window reasserts itself, the real tactical shape of the silo is a burst, not a steady drip. The bulk keybind introduced in v0.33.4, the U-LF combination, lets you launch or upgrade multiple nukes at once instead of clicking a single launch per warhead, and a fix in the same release prevents two nukes from firing on the same tick, so a burst is a short sequence of launches spread across consecutive ticks rather than one instant explosion of all loaded warheads in a single frame. Read that carefully, because it changes how you think about a loaded silo. A silo at level 10 that you burst with the keybind does not vanish ten warheads in a single frame; it launches them one after another, a few ticks apart, over a short window that starts with the first launch and continues as each slot frees up, with the nine-second reload pacing the back of the sequence. The tactical meaning is that a burst is a timed event you line up against a map moment: the enemy’s SAMs are reloading, their core is exposed, a frontline just broke. Firing the burst into that window is worth far more than the same ten warheads spread across three calm minutes of play. The stagger also matters for the defender, because a burst is a series of separate, individually targetable warheads rather than one object, and a defender with SAM coverage can contest each one; a staggered burst gives the defender several distinct targets over several seconds, which is exactly why a burst is only as good as the reload gap you time it into. The stagger also has a cost for the attacker: a level-10 burst is not instant, it takes a small number of ticks to finish, and during those ticks your front warheads are committed and visible, so a defender who reacts quickly can contest the early warheads and force you to spend the later ones into a partially recovered ring. The MIRV mechanics guide goes deeper on the specific case where a single carrier drops many warheads, which compounds this burst into a dense stream that the stagger makes even harder for a defender to fully intercept, and it is worth reading before you commit to a MIRV-heavy endgame plan, because the same stagger logic that shapes a single-silo burst is what makes a MIRV stream so dangerous and so hard to fully stop.

Flight time and range: distance sets the window, not the silo

The silo itself has no firing range limit. In the v0.36.5 source there is no silo range or maximum reach in the nuke execution or the engine config, so a silo can launch a warhead at any tile regardless of how far away the target is. What range does not constrain, flight time and interception do. A warhead crosses the map at a fixed speed, and in v0.36.5 the Atom and Hydrogen warheads both fly at 10 tiles per tick, while a MIRV carrier flies at 15 and a MIRV warhead at 22. At the configured rate, ten tiles per tick is roughly one tile every 100 milliseconds, so a warhead covering 300 tiles takes about 30 ticks, around three seconds, and the faster MIRV warhead at 22 tiles per tick covers the same 300 tiles in under two seconds. The arithmetic is simple: flight time in seconds is the tile distance divided by the tiles-per-tick speed, divided by the ten-tick rate. Two decisions hang on that number. For the attacker, a shorter flight to the target means less time for the defender to reposition SAMs or move assets, so when you can choose between a nearby and a distant target, the nearby target is harder to defend against on reaction time alone. For the defender, a warhead is targetable only inside a 150-tile window of its launch point or its destination, so your SAMs can contest the last 150 tiles of the flight, and at 10 tiles per tick that is 15 ticks, about 1.5 seconds, of interceptable flight; at the 22 tiles per tick of a MIRV warhead that window is under a second. The SAM launcher guide covers how many launchers cover a given landing zone; here the point is that flight speed, not silo distance, is what decides whether that 1.5-second window is wide enough to save your core. The faster the warhead, the thinner the window and the more your defense has to be already cycled and ready when the warhead crosses into the 150-tile band. This is also why a far-away enemy silo is not a safe enemy silo: because there is no range limit, the enemy silo does not need to be close to threaten you, it only needs a flight that your SAMs cannot cover, and the extra distance the enemy buys with a far placement costs them nothing on the launch itself, only a few seconds of your intercept window. Treat distance as a flight-time tool for the attacker and a coverage requirement for the defender, and you will stop misreading the map.

Placement and survivability: why a 5M silo is a target

A high-level silo is a concentration of gold, and that makes it a deliberate target. A level-5 silo represents five million gold of investment in a single structure, and a level-20 silo represents twenty million, so an enemy that can reach your silo with its own nukes can delete a very large fraction of your economy with one launch. This is the failure mode that the community’s dispersal advice is really about, and it is the reason the macro threads keep repeating that the silo should not be the single most valuable structure on the map. Because the silo has no range limit, an enemy silo does not need to be close to threaten you; it needs only a line of flight that your SAMs cannot cover, and the warhead only needs a few seconds of unobstructed flight. The placement rule that follows is to keep your silo inside your own SAM umbrella, and the mirror of that rule is the one players forget: your silo’s own inbound flight needs to be covered as carefully as your core’s is, because the enemy will aim at your most valuable structure the same way you aim at theirs. The community tip that says to put your SAM between the enemy silo and your assets has a mirror for the attacker: you want your silo positioned so that the flight path from it to the enemy core is not a straight, SAM-free corridor. A dispersed layout, the one the macro threads recommend, blunts both directions at once, because it denies the enemy a single high-value structure to nuke and stretches your own SAM coverage so the enemy cannot nuke one tile and delete half your nuke capacity. The practical checklist is to place the silo where its own core is covered by your SAM ring, where the outbound flight to the enemy has at least one contested segment, and where the structure is not the single most valuable tile on the map, because that tile is the tile the enemy will aim at. If you must concentrate gold in one silo, put it behind a SAM ring that also watches the tiles the enemy silo would fly from, and treat the silo as the most protected asset on the map rather than the most valuable one left exposed. The cost of getting this wrong is not a lost structure, it is a lost bank, because a nuke that deletes your level-20 silo also deletes twenty million gold of investment and the sustained output it was going to provide.

Upgrade versus second silo: the arithmetic that decides

The clean arithmetic of the flat 1M-per-level cost is what makes this decision easier than it looks. Adding a slot to your existing silo and building a new level-1 silo both cost exactly one million gold, so the choice is purely about what each buys you, and there is no hidden price difference to account for. Upgrading buys you one more warhead behind the same nine-second reload window, which raises the depth of a single burst and the total warheads you can spend before reloading, but it does nothing to your sustained launch rate. A second silo buys you a second, independent nine-second clock, which raises your sustained rate, because the two clocks can overlap: while your first silo counts down from its launch, your second silo is ready to fire, so your combined output can keep coming with less dead time between warheads. The rule of thumb the numbers support is that if your bottleneck is “I run out of warheads mid-burst,” upgrade; if your bottleneck is “I cannot launch often enough to keep pressure on,” build a second silo. A concrete way to make the call is to compare your burst depth to the number of warheads the defender can intercept in one cycle: if you can already saturate the defender’s intercept capacity in a single burst, extra depth is wasted and the next one million should buy a second clock; if your burst is shallow relative to what the defender can absorb, upgrade until the burst matches the threat, then switch to the second silo. In practice, against a defender with a strong SAM ring, the sustained rate is usually the bottleneck, because each individual burst can be contested but a continuous, overlapping output is hard to intercept continuously, which pushes toward the second silo. Against an undefended or lightly defended core, a single deep silo that bursts a large stock is often enough, and the gold you would have spent on a second silo is better saved for another purpose entirely. The decision is not which is better in the abstract, it is which resource your current map state is short on, and the flat one-million cost means you can afford to test both and let the map tell you. There is no permanent commitment in either choice: you can build a second silo now and still upgrade both of them later, and you can upgrade now and add the second clock when the map forces you to. The only real penalty is time, because a second silo that is a level behind your first costs an extra million to catch up, so the earlier you add the second clock, the cheaper it is to bring it to parity with your primary.

Scenario 1: deciding in a one-versus-one with a contested core

Assume a mid-game one-versus-one. You have one level-3 silo, three million invested with three warheads loaded, and eight million gold in the bank, while the enemy has a level-6 silo, six million invested with six warheads, behind a SAM ring that covers their core. The question is where the eight million goes. If you spend it on three upgrades, you reach level 6 and match the enemy’s burst depth, but you still have one nine-second clock, so your sustained rate is unchanged and you can match their burst only if the enemy does not re-arm before you do. If you spend it on a new second silo, you reach level 1 plus keep your level-3, giving you two clocks: a combined output that can keep coming while the enemy’s single silo sits in its nine-second reload. A middle option spends two million on two upgrades, taking you to level 5, and six million on a new silo, giving you a level-5 plus a level-1 with two overlapping clocks. Against a core that is already SAM-covered, the second silo is the better spend, because your problem is not that you cannot field six warheads at once but that you cannot keep warheads coming while the enemy’s SAMs reload between your bursts. The numbers that make this concrete: your level-3 silo can burst three warheads then waits nine seconds; a level-6 enemy can burst six then waits nine seconds, so a single exchange where you burst first and the enemy bursts back leaves the enemy two warheads deeper with the same sustained rate, which is why you need the overlap. The decision is to buy the second silo now, accept a shallower single burst, and win on sustained pressure. The cost of the alternative, matching depth and losing the rate, is a level-6 silo that still reloads in nine seconds and cannot out-sustain the enemy’s reloading cycle, which is a structural disadvantage you cannot close with gold. The follow-up decision matters as much as the first one. Once you have your level-3 and level-1, the next million goes to the level-1 silo to bring it to level 2, not to the level-3, because you want the two clocks to stay balanced so the overlap keeps working; if you keep upgrading only the primary, your second silo stays a one-warhead clock that adds almost nothing to your sustained rate. After a few minutes you will have two clocks each at around level 2 to 4, which is a materially different output than one clock at level 6, and it is the shape that lets you keep pressure on the enemy’s reloading SAMs without running dry.

Scenario 2: timing a MIRV burst against a reloading SAM ring

Assume the same one-versus-one has escalated and you have saved for a MIRV. A MIRV in v0.36.5 costs twenty-five million for the first one and then fifteen million more for each additional MIRV you launch, so a two-MIRV plan costs twenty-five million plus fifteen million, forty million total. A single MIRV over a target of roughly 200,000 tiles spawns about 100 warheads, the cap being 350 warheads at roughly one per 2000 target tiles, and each of those warheads flies at 22 tiles per tick, the fastest warhead in the game, so the defender’s 1.5-second targetable window shrinks to under a second of flight per warhead. The tactical shape is that a MIRV is not one big threat you intercept; it is a dense stream of individually targetable warheads arriving faster than a SAM ring can cycle its slots, because every SAM slot also reloads on its own window and cannot fire at all a hundred warheads in one burst. The timing question is when to launch the MIRV so that the stream arrives while the enemy’s SAMs are still counting down from their last launch. You do that by observing the enemy’s SAM reload cadence, waiting for the moment their front slots are committed and reloading, and then launching the MIRV so the fastest warheads land in that gap. The cost structure is the constraint: at twenty-five million for one MIRV and forty million for two, you are committing more gold than a full level-40 silo, so the launch has to be decisive, timed to a reload gap, and aimed at a core the stream will reach before the defender can reposition. If the SAMs are fully cycled and ready when your first warheads arrive, the stream will be thinned heavily and the forty million will have bought far less than it should have, which is why the timing discipline matters more than the raw warhead count. The MIRV warhead’s inner and outer blast is 12 and 18 tiles respectively, smaller than an Atom’s 30, so a MIRV stream is a volume play, not a single-blast play, and you win by overwhelming the intercept capacity rather than by any one warhead. The other cost is the gold you commit before the launch, because a two-MIRV plan locks forty million in place before the first warhead flies, and if the timing is wrong and the SAMs are ready, that forty million is gone with the burst you aimed to be decisive. The discipline is to treat the MIRV launch as a single timed action, not a sustained action: you pick the reload gap, you launch, and you do not commit to a second MIRV until you can see that the first one actually landed in the gap, which is why a two-MIRV plan is a high-risk, high-reward play that only works when your read on the enemy’s SAM cadence is correct.

Failure modes and the counter to each

The first failure is over-committing to a single max-level silo. The counter is the arithmetic from the upgrade section: because depth and rate are different resources and a second silo is the only rate increase, a level-20 reflex that spends twenty million on depth is usually a loss against a defender who simply cannot be saturated by one window, and the fix is to stop upgrading once your burst depth matches the threat and redirect the next one million to a second clock. The second failure is placing the silo where the outbound flight is a straight, unobstructed corridor, which hands the defender a clean intercept line. The counter is to choose a tile whose flight path to the enemy core crosses at least one contested segment of your own SAM coverage, and to disperse so the enemy cannot nuke one tile and delete your whole nuke capacity. The third failure is firing a burst into a fully cycled, ready SAM ring, which wastes warheads the defender was prepared to intercept. The counter is to time the burst to a SAM reload gap, as in the MIRV scenario, rather than firing on your own convenience. The fourth failure is underestimating the enemy’s own silo as a target, assuming your own structure is too far to matter. Because there is no range limit, the counter is to treat every enemy silo as a live threat to your core and to keep your SAM ring covering your own silo’s inbound flight as well as the enemy’s. The fifth failure is misreading the generated-data note that upgrades reduce reload time and planning a “fast silo” that does not exist. The counter is to trust the flat nine-second window and to plan capacity and rate as separate decisions, which is the entire structure of this guide. Each of these failures is a money leak, and the common thread is that the player is spending gold on the wrong of the two levers, depth or rate, or on a placement that gives the enemy a clean line, and the fix in every case is to check which lever the map is actually short on before you commit the million. The common thread across all five failures is the same: the player is spending gold on a structure they cannot see clearly in their own match. The fix is not a better build order, it is a habit of checking the two in-game tells, the nine-second reload clock and the launch availability lock, before every large commit, because those two observations are the ground truth that every number in this guide is built on, and if they drift from the v0.36.5 figures the whole decision framework needs to be re-derived from what your own match is actually showing you.

Mode and map adjustments: where the numbers shift

In a team mode, the sustained-rate argument for a second silo gets stronger, because a teammate’s silo can hold part of the line while yours bursts, so the combined allied output across multiple independent nine-second clocks is the real pressure, and a single deep silo looks less valuable next to the overlap your team can field. The cost of that, though, is that an enemy team’s nukes are pooled too, so your dispersed placement must protect against multiple enemy silos aiming at your core, not just one, which pushes the SAM coverage requirement up and makes the silo’s own SAM umbrella even more important. In a four-islands or fragmented map, the no-range-limit rule cuts both ways: an enemy silo on a far island is just as threatening as one on a near one, because flight time across a large island gap is only a few seconds at ten to twenty-two tiles per tick, so you cannot rely on distance as defense and must rely on SAM coverage of every plausible landing zone. In a dense, small map the opposite holds: flights are short, targetable windows are tiny, and a MIRV burst is nearly impossible to fully intercept, which makes the attacker’s second-silo sustained pressure the dominant endgame shape and makes the defender’s SAM coverage the only real defense. The consistent rule across modes and maps is that the nine-second reload and the ten-to-twenty-two tiles-per-tick flight speeds are the constants you plan around, while the variable is how much SAM coverage you and the enemy can bring to bear on the landing zones, and the silo placement decision is really a decision about where you can afford to concentrate gold that the enemy can reach in a few seconds of unobstructed flight. The practical reading is that the larger and more fragmented the map, the more you need SAM coverage to compensate for the fact that distance no longer protects you, and the smaller and denser the map, the more a single well-placed burst can decide the game, which is why the same nine-second clock produces very different endgames on different maps. The practical takeaway is to identify which regime your current map is in before you commit gold to a nuke plan. If the map is large and fragmented, budget for extra SAM coverage to compensate for the fact that distance no longer protects you, and treat every island gap as a few seconds of unobstructed flight rather than a wall. If the map is small and dense, budget for the burst, because a single well-timed burst can end the game and your second silo is mostly a way to survive the enemy’s burst while you set up your own. The nine-second clock is the same in both cases, but the coverage you need around it is very different, which is why the same silo decision produces the wrong answer on the wrong map.

The numbers you need at a glance

ParameterAtom bombHydrogen bombMIRV carrierMIRV warhead
Cost (gold)750,000 flat5,000,000 flat25M + 15M per extran/a
Flight speed (tiles/tick)10101522
Blast magnitude (inner/outer)12 / 3080 / 100n/a12 / 18
Targetable window150 tiles from launch or destination

Every number in this table is from the v0.36.5 tag (b863d48), not from a wiki. The silo rows underneath are the two that drive every decision in this guide: the silo cost is a flat 1,000,000 per level with no escalation, and the silo reload is a flat 90 ticks, nine seconds, level-independent, so capacity and rate are two separate levers that you spend gold on independently. The two decisions that dominate endgame nuke play are capacity versus rate, where an upgrade adds capacity and a second silo adds rate, and placement versus coverage, where a high-value uncovered silo is a target and a covered one is a threat, and the nine-second reload is the constant that makes both of those decisions real. Verify the two in-game tells before you commit gold: the reload bar resetting on a nine-second clock that does not drain faster at higher levels, and the launch availability that stays locked until that clock finishes. Those two observations, checked in your own match, are the difference between trusting this guide’s numbers and re-deriving them yourself, and they are the two checks that will tell you immediately if the game version you are on has drifted from the v0.36.5 figures used throughout this guide, which is why you should re-verify them in a fresh match before you build a level-20 silo on the strength of this page alone. The table is meant to be the last thing you check before you commit gold, not the first thing you trust. The silo rows in particular, the flat 1M per level and the flat 90-tick reload, are the two numbers that drive every decision in this guide, and if your game version has changed either of them the capacity-versus-rate framework still holds but the specific dollar amounts and the nine-second window shift, so you should re-derive the upgrade-versus-second arithmetic from your own version’s constants rather than from this page. The rest of the table, the warhead costs, flight speeds, and blast magnitudes, tells you what each warhead does once it flies, which is what you need to judge whether a given burst is worth the gold it costs to load, and that is the number you check when you are deciding between an Atom-heavy burst and a MIRV-heavy burst rather than when you are deciding whether to upgrade or build a second silo.

Related content

Continue reading
OpenFront AFK Teammate Takeover: Absorb a Disconnected Ally for Zero Troop Loss

A Team-mode decision guide for what a teammate's disconnect does to your win/loss position, when absorbing their land is free, when it is a trap, and what the 30-second mark and the ranked 2v2 rules actually change.