Upgrade Evaluation
Ву Wheredoyoukeepthechocolate?
Upgrade evaluation is one of the most important skills in the game, but it’s also one of the hardest to master. The TLDR is to take the rarest upgrade when in doubt about what is best. However, there are many situations where it’s better to take a lower rarity upgrade based on current context.
As a general rule, I tend to favor upgrades that increase damage output above most other things. If you get overwhelmed because you can’t kill the enemies fast enough, nothing else really matters. Let’s examine some general principles of upgrade evaluation.
Fire Rate vs. Reload Speed
Most projectile weapons in the game will alternate between firing and reloading. When evaluating which is more valuable, consider how much time it takes to empty your clip vs. how much time it takes to reload. This can be easily calculated by dividing the clip size, minus 1, by the fire rate per second. For instance, if a weapon has a clip size of 22 and fires 7 times per second, it will fire for 3 seconds before it needs to reload. Subtracting 1 from the clip size is necessary for accurate calculation because the first bullet fires the instant the reload speed cooldown finishes. Let’s say that the gun also takes 1 second to reload. That will give us a 3/1 ratio of firing to reloading. For such a weapon, reload speed upgrades are only one third as valuable as fire rate upgrades.
Most weapons will significantly favor one or the other. For instance, fire rate is nearly useless on the jury rigged boomstick, and reload speed is terrible on the zhukov. If a weapon is exactly 50/50, I’d favor reload speed as a tiebreaker since having less downtime is helpful. Often, it will be correct to take a lower rarity upgrade depending on which weapon you’re considering. In almost all cases, however, damage upgrades will be better than an equivalent amount of fire rate or reload speed.
Additive vs. Multiplicative
One of the most important aspects of upgrade evaluation is knowing which bonuses are additive with each other and which are multiplicative. There are five buckets for bonuses: skill, overclock, artifact, meta, and mutator. Bonuses are additive within buckets and multiplicative across buckets. If a bonus is listed with blue text, it will go in the skill category. Any other category will be listed with yellow text. A complete breakdown of this can be found in the detailed weapon stats screen during a run.
What do additive and multiplicative actually mean in practice? Let’s say we have a weapon that deals 100 damage per shot as a baseline without any bonuses of any type. If we were to take two legendary (+50%) damage upgrades for it back to back, it would double the damage per shot. The equation would look something like this: 100 x (1 + .5 + .5) = 200. However, if we were to take an artifact that increased the damage of that weapon by 50% in place of one of those weapon upgrades, the two bonuses would multiply. Since 1.5 x 1.5 = 2.25 the weapon would actually end up dealing 225. From this, we can see how valuable getting bonuses from multiple sources is. As a general rule, yellow text is quite a bit better than the same upgrade in blue text.
When trying to evaluate upgrades from different categories, I find it most helpful to think of it in terms of increase to overall damage output. For instance, let’s take that 100 damage per shot weapon from earlier. At the start of a run, a common damage upgrade will give an additive +10% to the damage of the weapon. Since this is our only weapon at the start, that will result in a 10% increase to overall damage output (let’s ignore weapon levels and progress toward overclocks for now).
Later on in the run, let’s say we’ve taken enough upgrades to damage that we already have +100% and we’re offered that same common damage upgrade again. Since we already have +100% in the skill bucket for that weapon, adding 10% onto that now only represents a 5% increase to the overall damage from that weapon.
From this, we can see one of the most fundamental principles of this game: bonuses in a particular bucket make future bonuses in that bucket less attractive and the degree to which that is the case is directly dependent on how large a bonus we already have in that bucket. In theory, the way to get optimal increases to your damage output would be to increase the skill, meta, overclock, and artifact buckets equally. In practice, it’s far easier to increase the skill bucket than any other, so we should always consider that when evaluating our options.
How Valuable is a Weapon Level?
This is one of the most frequent evaluation dilemmas and it’s probably the hardest one to figure out. Do I take a tag upgrade that applies to three of my weapons in order to get more immediate power, or do I take the uncommon weapon upgrade in the hopes that the overclock will provide more of a payoff later?
Weapon levels are only as strong as the overclocks they unlock. Therefore, the value of weapon levels is dependent on the strength of the overclock pool for that weapon. I consider a weapon level to be worth one sixth of the average value of the overclocks you might be offered.
For example, let’s say there’s a weapon with three balanced overclocks. One that increases damage by 100%, one by 50%, and one by 25%. You know you’ll be offered two of them and will always take the higher value one. This means that you’ll take the +100% overclock 2/3rds of the time, the +50% one 1/3rd of the time, and you won’t ever take the +25% one. Thus, the average value of an overclock for this weapon will be +83.3% damage (since this is 2/3rds of the way between 50% and 100%). Since a weapon level is worth 1/6th of this, that means I’d evaluate a level for this weapon as being equal to +13.9% to the overall damage output of the weapon. It’s just that the bonus is delayed until I actually unlock the overclock.
Unstable overclocks are usually extremely powerful and will often double the DPS of that weapon. Because of this, it’s often correct to take weaker upgrades in order to push a powerful weapon to level 18. Later on in a run, it’s also worth considering how likely you are to even reach the next overclock for a particular gun. If you don’t think you’ll get there, then weapon levels are useless.
It’s also worth considering how powerful a weapon is to start with. Often, it’s optimal to pick a strong weapon early, push it to level 18 as soon as possible, and let it carry your run. In such cases, levels in your primary weapon are far more valuable than levels in your secondary weapons. Also, support weapons often don’t need overclocks, or only need minimal overclocks, to serve their purpose. For example, the cryo cannon can do its job pretty well at level 1, so pushing for weapon levels is not usually a high priority for it. If you’ve taken the sidearm overclock for the stubby or subata, then it’s fine to leave those weapons at level 6 while you invest elsewhere.
Thoughts on Each Upgrade Category
Because 15 gold is worth about half a level, I’m going to try, where possible, to sketch out where the break point is between “begrudgingly takeable” and “just reroll.” The advice on when to reroll is oriented toward regular 5 stage dives. If you’re playing on weapon or biome mastery, be sure to take the shorter or longer dive length into consideration. As a general rule, a weapon level in a weapon that you care about upgrading is, by itself, worth enough to justify not rerolling, even if the stat buff that comes with it is negligible.
Armor: Armor blocks a percentage of incoming damage and the amount blocked decreases with each stage. With most or all of the meta upgrades completed, 10 armor will provide an effective HP increase of 15-20%. Thus, armor has a synergistic relationship with HP improvements. If you have a small lifebar, armor won’t provide as much of a benefit, but it can be quite strong if your max HP is high. Uncommon armor upgrades are pretty terrible, rare ones are decent, epic and legendary are quite good. When given a choice between armor and an equivalent rarity HP upgrade, I think HP usually has the edge because it will usually result in more healing from health regen over time.
Damage: One of the best upgrades in the game. Reroll at uncommon, but everything else is premium. It’s worth noting that general damage upgrades also increase status effect damage from electricity, fire, and acid, but weapon specific and tag based damage upgrades do not unless they specifically apply to that element. On weapons that alternate between firing and reloading, damage upgrades are usually significantly more valuable than an equivalent amount of fire rate or reload speed.
Experience: Great in the first stage, mediocre in the second stage, and generally untakeable after that. I like to think in terms of how long it will take to pay back the level I spent getting the upgrade. Let’s say you give up an average upgrade, like an uncommon weapon upgrade, in favor of a +10% xp upgrade. In order to provide more value than what you gave up, the xp upgrade needs enough time to earn enough xp for a full extra level. Only after that point will you be better off than if you’d just taken the average upgrade. If that payoff won’t happen, or will happen too late, it’s not worth taking.
It’s worth noting that early experience upgrades will allow you to get your build online earlier, and depending on the build, that can be crucial. XP upgrades fall off in value much faster than luck because of how much new levels dry up after about level 50. Whereas, luck continues to provide its benefit in the shop, supply drops, and in huuli hoarder crates.
- Stage 1: Always good to take.
- Stage 2: Epic and legendary are good, rare is usually takeable.
- Stage 3: Reroll rare. Epic and legendary are sometimes takeable depending on context and what else is being offered.
- Stage 4-5: Just reroll.
Fire rate: The value is completely dependent on which weapons you’re using and how much reload speed you have on them. Refer to the fire rate vs. reload speed section above.
HP: Common or uncommon upgrades are pretty weak. Rare or higher are quite good. More health allows you to be more aggressive with using your health as a resource to save time or secure objectives. Note that passive health regen is based on how much life you’ve lost, not on the percentage of life lost. Thus, a larger health bar will generally result in more health regen. Try to maintain a large enough health pool that a few hits won’t result in instant death, but don’t neglect damage output. It’s very difficult to avoid all damage, but if you’re getting hit constantly, more health probably won’t help as much as more damage will.
Luck: Phenomenal early, but falls off as the run goes on. Luck improves the rarity of all upgrades you see, including the ones in the shop and the artifacts in the supply drop, but its value is hard to quantify objectively. My sense is that luck is extremely strong as long as there’s enough run left for the improvements to pay off. 50 luck will keep you from seeing any common upgrades and 100 will keep you from seeing any uncommon ones.
Actually getting to 50 luck is uncommon, but even getting an extra 10 can make a huge difference. This isn’t like XP upgrades where the improvements are basically linear. In many cases, improving the rarity of an upgrade by one tier improves the output by 50% and sometimes it even doubles it. If a luck upgrade increases the rarity of even a few upgrades that you take later in the run, it’s probably already paid for itself. Certain upgrades don’t appear below certain rarities. If these are particularly valuable to you (+1 beams, +1 drones, piercing, etc.) then luck goes up in value. Also, if you already have lots of luck, then rerolling becomes more attractive since the average value of an upgrade is higher than normal.
If I could pick one upgrade to always see at level 1, it would be a +1 drone/+1 beam. Second would be 10 luck. It really is that powerful.
- Stage 1: Always happily take.
- Stage 2: Basically always take.
- Stage 3: Still take epic. Rare is context dependent, but still probably worth taking.
- Stage 4: Maybe take epic, but probably reroll.
- Stage 5: Just reroll.
Pickup Radius: The main value of pickup radius is really in the amount of time it saves you. Less backtracking is required and you can spend more of your focus on other things because you don’t have to think about whether or not to go back. Once you have most of the screen covered, further upgrades are a low priority.
- Uncommon: Just reroll.
- Rare: Good in stage 1-2, context dependent in stage 3, reroll in stage 4-5.
- Epic/Legendary: Great in stages 1-3, probably reroll after that.
Reload Speed: Like fire rate, the value of reload speed is completely dependent on which weapons you’re using and, if applicable, how much fire rate you have on them. There are quite a few weapons that reload non-stop. For those weapons, reload speed is usually equal to damage in regard to increasing DPS. Often, it’s slightly better since more reload speed reduces downtime, allowing you to deal damage more consistently and avoid awkward moments of downtime.
Crit Chance/Crit Damage: One of the most powerful sources of multiplicative damage in the game. With the right artifacts, it’s quite possible for most of your damage output to be from crit bonuses. Calculating what percentage of your damage output is from critting and how valuable an improvement is can be quite complicated, especially since status effect damage doesn’t crit. As a general rule, crit chance is quite a bit better than crit damage and both become more valuable later in the run. Crit chance and crit damage have major synergy with each other. Improvements to one make the other more valuable. This can be a little bit slow to get off the ground, but the payoff is there in spades. When the meta upgrades are maxed, the baseline crit chance is 17% and the damage when you crit is 198%. In this case, an uncommon crit chance upgrade will provide about a 5% improvement to overall DPS (assuming no elemental weapons). This is quite strong for an uncommon upgrade, especially in the late game. Don’t sleep on crit. This will all be covered in more detail later.
Status Effect Damage: Quite strong if you’re running a build that relies heavily on status effect damage (SED), but pretty weak outside of that. Quite important for the interrogator driller. Some elemental weapons don’t actually apply that many stacks and do most of their damage directly, others do most of their damage from SED. The value of SED will depend on how many weapons it applies to and their ratio of direct damage to SED.
Lifetime: Excellent for most turrets, but pretty weak for drones and most beam weapons. Becomes especially good with the disposable tech overclock.
Potency: Generally pretty good. Will be explained in more detail in the potency section.
Explosion radius: Quite good. A 10% increase to radius will result in about a 21% increase in the area covered by the explosion. Multiple upgrades stack quite effectively. Hitting half the screen with one volley of grenades is a good way to make sure no little enemies ever get near you.
Weapon Range: Similar to explosion radius. Improving the range of weapons allows them to hit more enemies at once and to start targeting them sooner. Quite powerful.
Piercing: Usually quite good. Most projectile weapons start with 4 piercing, so increasing that by 1 can increase damage output by as much as 25% in the right situations. Plasma weapons interact differently with piercing. More details on that later.
Move speed: Quite strong and does more for you than you’d think. For instance, +4% move speed might not seem like much, but here’s a non-exhaustive list of things it does for you. It lets you reach objectives faster, makes it easier to pop exploders without taking damage, makes it easier to slip through gaps in enemy formations, makes kiting enemies easier, and probably most importantly, it lets you outrun enemies. Even a marginal improvement to movement speed will pay disproportionate dividends when you’re trying to put some distance between you and a swarm. This goes double when you’re trying to farm XP at the end of a level and the enemies are getting faster and faster.
Mining Speed: Weak at uncommon, solid at rare, great at epic or above. Goes up quite a bit in the salt flats. Mining speed is quite a nice quality of life improvement, but the most important threshold is: can you mine fast enough to escape from enemies that are closing in at close range? If you have the meta upgrades maxed out, the answer is basically already yes. Mining speed is helpful for finishing off veins of minerals and maneuvering around the map, but it’s not a super high priority.+1 beams/+1 Drones: Probably the most valuable non-overclock, non-artifact upgrade in the game. They’re only available at legendary rarity, but if you get them, they can carry a run. Luck goes up in value when you have weapons that can get these upgrades.
