How to Grind Research Points Fast in War Thunder
How the RP formula actually works: research efficiency, the skill bonus, talisman vs premium maths, and the mistakes that waste a whole session.
Research points in War Thunder are not paid per hour played. Each battle produces a base figure from your vehicle’s own multiplier and what you did in the match, and that figure is then multiplied — or gutted by up to 95% — by a research efficiency penalty most players never open the menu to check.
Correcting that one number is worth more than another three hours of queueing.
Gaijin’s wiki states the inputs plainly: RP depends on the game mode, your performance and activity in the battle, and whether you won or lost. What it does not publish is a single closed formula.
So treat RP as a chain of multipliers and fix them one at a time, starting with the one that can cost you four fifths of a session.
The RP chain, in order
Six inputs stand between a kill and a research point, in roughly this order — your vehicle’s multiplier first, the research efficiency penalty second to last, boosters last. The order matters because the biggest loss lands late, after everything you did in the match has already been counted.
- Your vehicle’s own multiplier. Every vehicle carries an RP and SL multiplier, visible on its card in the hangar. Premium vehicles have that multiplier doubled compared with a regular vehicle.
- Performance and activity. Damage, kills, captures and time spent doing something. Gaijin has never published the activity formula, so treat it as “be useful for the length of the match” rather than a number to optimise.
- The skill bonus. A hard step function based on opponents destroyed, detailed below.
- Victory or defeat.
- Research efficiency. The rank gap between the vehicle you are playing and the vehicle you selected for research.
- Talisman, premium account and boosters.
One important structural detail: modification RP is a separate pool. Each vehicle you take into battle earns RP toward its own modifications equal to the vehicle RP awarded before research efficiency is applied. Spading a vehicle is therefore never taxed by a bad rank gap — only your tech-tree progress is.
Research efficiency is the biggest lever you control
The rank gap between the vehicle you are playing and the vehicle you selected for research decides how much of your RP survives the battle, and Gaijin’s wiki puts the worst case at −95%. The penalty is also asymmetric: researching backwards down the tree costs more than researching forwards.
| Rank gap | Researching a higher-rank vehicle | Researching a lower-rank vehicle |
|---|---|---|
| 0 | No penalty | No penalty |
| 1 | No penalty | −10% |
| 2 | −60% | −70% |
| 3 | −70% | −90% |
| 4 or more | −80% | −95% |
Grinding a rank VI target while playing a comfortable rank II vehicle you know well costs you 80% of your vehicle RP — five battles for the value of one.
Going backwards is punished harder than going forwards: using a rank V jet to finish off a rank II aircraft you skipped costs 90–95%. If you want that leftover, play something near it.
Researching the vehicle directly connected by an arrow to the one you are playing goes the other way entirely, paying 130% RP in Arcade and 110% in Realistic and Simulator. Walking the tree node by node is mechanically rewarded; skipping four nodes ahead to the thing you actually want is mechanically punished.
The skill bonus rewards a concentrated game
The skill bonus is a step function on opponents destroyed in a single battle, and Gaijin’s wiki publishes the steps outright. There is nothing between them, so the fourth kill in Air RB is worth far more than the eighth.
| Mode | +15–20% | +35–50% | +100% |
|---|---|---|---|
| Air Arcade | 3–5 opponents | 6–9 (+40%) | 10 or more |
| Air Realistic | — | 2 (+35%), 3 (+75%) | 4 or more |
| Ground AB / RB | 3–5 opponents | 6–8 (+50%) | 9 or more |
| Ground Simulator | 3–4 (+20%) | 5–7 (+50%) | 8 or more |
| Naval AB / RB | 6,500–9,999 damage (+20%) | 10,000–14,999 (+45%) | 15,000 or more |
Air Realistic has by far the cheapest ladder: two kills already pays +35%, and four maxes it. Air Arcade wants ten. That alone is a reason a Realistic session usually out-earns an Arcade session of the same length, even before mode multipliers.
Note one exclusion: the skill bonus is not counted toward convertible RP.
Which modes and BRs actually pay
Gaijin does not publish a single global “Realistic pays X times Arcade” multiplier, and anyone quoting one is guessing. What is published, and what you should actually read, is the per-vehicle card in the hangar — each vehicle lists its own RP multiplier, and those differ by mode.
Crew experience, which players often conflate with RP, runs the other way entirely — the wiki publishes a flat rate per mode.
| Mode | Crew XP per 1,000 RP |
|---|---|
| Arcade | 30 |
| Realistic | 11.3 |
| Simulator | 9 |
So if your problem is untrained crews rather than locked vehicles, Arcade is roughly three times as efficient per RP earned. Diagnose which bottleneck you actually have before picking a mode.
On battle rating: higher-rank vehicles carry larger multipliers, but they also carry larger SL repair bills — and silver lions, not RP, is what stalls most accounts around rank IV. The BR that pays best is the one where you reliably reach three or four kills, not the highest one you own.
A top-tier vehicle you die in first is worth less than a rank IV vehicle you finish matches in.
Premium account vs talisman: the actual maths
These are different products solving different problems.
| Talisman | Premium account | Premium vehicle | |
|---|---|---|---|
| Scope | One vehicle | Whole account | One vehicle |
| Affects | RP only | All earnings | Its own multipliers |
| Size | +100% | +100% | Multiplier doubled |
| Duration | Permanent | Time-limited | Permanent |
The decision rule follows from the duration column. A talisman is a bet that you will keep playing that specific vehicle for a long time — including after it is fully spaded, when its only remaining job is feeding your tech-tree research.
Talismanning something you are two nodes away from replacing is money burnt. Talisman the vehicle you already return to when you want a good game.
A premium account is a bet on your own play time, and it is the only one of the three that also doubles silver lions. Given that SL is the more common hard stop, a month of premium account during a period when you actually play beats a talisman on a vehicle you will abandon.
One honest caveat: the wiki lists both premium account and talisman as +100% but does not publish the order in which they combine. Do not budget your grind around an assumed 4× — plan around the fact that both apply and verify against your own post-battle screens.
Boosters stack badly
Boosters run from +10% to +800% for personal ones; team boosters marked “(all allies)” are 10% or 22%. The trap is running several at once, because the wiki’s own table shows same-type boosters taking a hard penalty by count:
| Boosters active | Penalty |
|---|---|
| 1 | None |
| 2 | 40% |
| 3 | 60% |
| 4 | 80% |
| 5 or more | 90% |
The penalty is applied in order of increasing bonus, not activation order — so dumping your inventory means your small boosters are the ones gutted, but you still lose a large share of the total. Run your biggest booster alone.
Two more details: boosters are counted per battle, and in a long match every 30 minutes counts as a separate battle — which is a genuine argument for using multi-battle boosters in Simulator. And boosters expire from your inventory, typically on a 3-day window, so hoarding them is itself a loss.
Squadron research points are a parallel currency
Squadron vehicles do not compete with your tech-tree grind at all. They run on activity points, which the wiki documents in full: every 200 RP you earn while in a squadron adds one point to the squadron pool, and every 3 days those points convert into squadron RP toward your selected squadron vehicle.
Each player can contribute up to 360 points per 3-day period, against a pool limited to 20,000.
The practical reading: this is free progress that accrues from the battles you were playing anyway, capped per period, and dependent on your squadron actually being active. If you are not in an active squadron, you are leaving an entire second currency on the table for zero effort.
Six mistakes that waste a session
- No research target selected. With nothing queued when a vehicle or modification completes, the vehicle RP from those battles has nowhere to go — it accrues as convertible RP, which costs Golden Eagles to release at the wiki’s rate of 45 convertible RP per GE. Set the next target before you queue up, not after.
- A four-rank research gap. The −80% and −95% rows above — the single most expensive habit in the game.
- Cherry-picking modifications. Completing all of a tier’s modifications awards vehicle RP equal to 20% of that tier’s total cost, so leaving one cheap module unresearched forfeits the bonus.
- Dumping every booster at once. Five at a time costs you 90%, as the table above shows.
- Grinding a spaded vehicle with no talisman. Once a vehicle is fully modified its modification RP pool has nothing to buy, which is the exact situation a talisman exists for — and the exact situation where its absence quietly halves your return.
- Treating activity as optional. Activity is one of only three inputs Gaijin names outright, and spawning, dying and immediately leaving keeps whatever you earned but stops it climbing for the rest of the match.
What to check before your next session
- Research target — exactly one node ahead of a vehicle you actually enjoy playing.
- Arrow connection — that target linked by an arrow to the vehicle you are taking into battle.
- Booster discipline — one running rather than five, with squadron points accruing in the background.
Thirty seconds in the tech tree is worth more than any booster in your inventory, then pick the BR where you finish matches with three or four kills.