How to Find Optimal Climb Speeds for (Almost) Every Airplane

Currently, I am in the process of creating a chart of optimal climb speeds for most planes, but seeing as that will take a while, I would like to share how to do it yourself.

First, navigate to the War Thunder Datamine (link provided below), and locate the search box with the prompt “Go to file”. Type “performance/%name of aircraft%” and look for a .blkx file with the path "aces.vromfs.bin_u/gamedata/flightmodels/performance/%name of aircraft%.blkx. For example, if I wanted to determine such values for the Yak-1, I would search “performance/yak-1_early.blkx”.

The file will contain 3 values for each kilometer of altitude for your plane. “altSpecs0” corresponds to 0 m in altitude; likewise “altSpecs5000” corresponds to 5000 m. The values we are looking for is “optSpeed”. For the Yak 1, at 0 m, the plane’s optimal climb speed is 69.2861 m/s (or 249.43 km/h). So, according to this, for maximum altitude gain in the shortest time in the Yak 1 below 1000 m, you will want to climb at an IAS of approx. 249 km/h.

Be aware though, for some reason, some (very few) planes don’t have a performance file corresponding to themselves (or I may be blind).

And that is all there is to it. (though my explanation is probably subpar)

But you may be asking, how do I remember these values for the plane I’m using. Well, what I would do is setup War Thunder Real Time Information (WTRTI) to indicate if your below, at, or above your optimal IAS for climbing at a given altitude. I would go into depth and explain how to do this, but I am lazy. (and in my mind it seems simple to setup)

War Thunder Datamine


You might want to talk to this guy. He seemed to put a lot of effort into data like this and seems to be willing to help out people with similar intentions

1 Like

Thanks for the suggestion, I actually use his MEC data/settings a lot.


Right? His work with the P-38L made it fun again for me, especially with a HOTAS and binds for MEC.

Climb data really would help. I think an acceleration reading of any type - ideally separate - makes energy control easier. So long as we have the data on an airframe model, I feel that calculated performance might be close enough to test by. Then it’s pushing for accuracy, which means more datamining; lots of work if it’s not made autonomous. I think the last part, automating it, would make the chart generate close-enough that you can turn into near-perfect data.

1 Like