Warhammer 40,000: Darktide – Comprehensive Review of Emperor Gifts

Results from a 2-week collection period tracking various mission metrics and the impact on Emperor Gifts.

Summary Statement

Emperor Gifts are seemingly rewarded at random. Previous consensus I garnered from reddit was that, in one week, 20 gifts seemed to be a soft-limit on EG rewards with significantly diminishing returns. Gifts also seemed to favor one “main” character with a soft-cap of 10, with other “secondary” characters receiving a soft-cap of 5. After a month of data collection, I can say that using a rotating character method there seems to be no true EG cap, as in one week of data collection I was able to get 14+ rewards per character. Furthermore, other factors (difficulty, mission lead, modifiers) may also affect EG distribution.

Introduction

There are many theories and speculations regarding what influences Emperor Gift (EG) rewards. After a good community effort, and contributions of data collection by a redditor, the conclusion was that EGs were awarded randomly with a soft-cap of 20 with diminishing returns to a possible hard-cap of 28.

One theory that arose was that one “main” character had a soft-cap of 10 gifts, and subsequent characters had soft-caps of 5, all with diminishing returns to the total hard-cap of 28. This theory followed from Serial gift collection (collecting maximum gifts from one character before proceeding to the next).

Given those findings I wanted to explore what would happen if EGs were collected in in a Parallel fashion, meaning a character would be played until an EG was rewarded, then a new character would rotate in and follow suit. The initial results were startling, as I exceeded the theorized cap and was able to collect a total of 45 EGs in one week (albeit requiring completion of over 80 missions). Given these results, I expanded on the data collection to examine several factors that may influence EGs, including: Mission Lead, Quick-Play, Difficulty, Side-Objectives, and Mission Modifiers.

Methodology

Data was collected with a defined week of Monday UTC 00:00 to Sunday UTC 23:59, and included time-period of 15JAN – 29JAN2023. Mission parameters collected were: Mission, Lead (the person who assigned the mission), Difficulty, Side-Objectives (further stratified by type and partial vs full completion), if the mission was joined via Quick-Play, and mission modifiers.

The majority of the data collection was done in a character-parallel fashion, where one character was played until they received an EG, at which point another character would be rotated in until they also collected an EG. At some points Characters were arbitrarily rotated out prior to receiving an EG. Character rotation order was also arbitrary, meaning characters did not have equal EG collection values.

Given the dates, data was collected on version 1.0.21-1.0.23. Unsure how much relevance this will continue to have since FatShark liked to do stealth-adjustments without noting them in the patch notes.

Given the amount of variables and the relatively small amount of data, no statistical analysis was completed, and data is simply anectdotally compared.

Results

The complete 2-week excel data is a little large, so I’ll just include some of the compiled results:

  • Total Missions Completed (2 weeks): 122
  • Total Gifts Rewarded (2 weeks): 60

Character Breakdown

CharNo EGEGWins
Vet141933
Psy242246
Ogr241943

Mission Lead

LeadNo EGEG
Hadron1317
Masozi48
Morrow2317
Zola2218

Difficulty

DifficultyNo EGEG
Sedition01
Uprising11
Malice3217
Heresy2941

Side Objectives

BookNo EGEG
Scriptures3541
Grims118
None1611

Mission Modifiers

ModifierNo EGEG
Low Intensity1021
High Intensity118
Pox Hounds31
Lights Out12
None3728

Limitations

Given the sheer amount of variables and permutations, the data pool is not large enough to do an in-depth statistical analysis of EG rewards per variables, rather this guide just notes observations based on data trends.

EG Data was collected in a parallel fashion, so inferences may not apply to people playing a single character, or playing a character in serial fashion before proceeding to the next.

Data only includes 3 lvl 30 characters. And data followed rotations involving all 3 characters.

Mission selection was by my choice (except when Quick-Play was chosen), which impacts the additional variables attached to missions.

Furthermore, all data was collected on a single user account, and assumes a weekly interval of Mon-Sun based off of UTC.

I’m sure there are other data limitations. Take my results with a grain of salt and do with it what you may.

Analysis / Conclusions

When it comes to Emperor Gifts, there may be more to the story than just simple RNG, however my data is limiting as noted previously. Some take-aways that I found are:

  • 1) No Caps to Emperor Gifts, at least when playing characters in a parallel fashion. The limiting factor seems to be the amount of missions. I was going to test out doing a mix of serial-parallel EG collection, but I’m getting tired of the same missions, modifiers, and bugs. Further more they patched-out borderless fullscreen upscaling, which now makes playing much less enjoyable due to FPS hits.
  • 2) Difficulty – Looks like difficulty may play a role in mission rewards, however this needs a BIG CAVEAT. My initial data collection (in which I collected 47 gifts) was excluded due continued variable refinement, and during that period I predominantly played Malice, and my Ogryn was sub 30. So there may be some other factor tied into difficulty impacting results, such as the average difficulty you’ve been playing recently or character gear level (something akin to VT2s Power Level). If difficulty does impact gift rates, then there’s definitely something influencing which difficulty will deliver what rewards.
  • 3) Characters – I was surprised to find that my Vet was getting mission rewards at a less than 2:1 Ratio (2 missions:1 reward). The remaining 2 seemed to get gifts at slightly worse than 2:1. My vet was also my first-created character and has the most missions under his belt. My Psyker and Ogryn follow respectively, and they were created respectively.
  • 4) Mission Leads – This was also surprising to me. Masozi, albeit with only an n=12, rewarded gifts at a rate much higher than 50%. Increasing the n would vastly improve the robustness on this variable. I also wonder if there’s some interplay between Mission Leads and Characters, for example, does Morrow and Zola reward Vets more frequently than other classes? No idea. But definitely worth investigating.
  • 5) Mission Modifiers – Was not expecting to see “Low Intensity” pull out a 66%-ish payout rate, and was also not expecting “High Intensity” to pay out less than “Low”. I avoided Pox Hounds as much as possible, and Lights Out entered the rotation too late to really test. Having no modifiers pays out less than 50%, but again given a relatively low n, and the amount of noise from all the other variables it’s really hard to say.
  • 6) Missions – Included more for fun than anything. The caveat again is the individual numbers are too small to really tell anything. Maybe Commsplex pays out more, maybe Excise vault pays out less. Need Moar Data.

That being said, if you’re constantly hopping between characters, you can average somewhere near a 50% Emperor gift payout rate. Maybe doing “Low Intensity” missions given out by Masozi will give you a higher rate. Maybe not. Someone else can pick up where I left off because I’m tired Boss…

Helena Stamatina
About Helena Stamatina 2689 Articles
My first game was Naughty Dog’s Crash Bandicoot (PlayStation) back in 1996. And since then gaming has been my main hobby. I turned my passion for gaming into a job by starting my first geek blog in 2009. When I’m not working on the site, I play mostly on my PlayStation. But I also love outdoor activities and especially skiing.

Be the first to comment

Leave a Reply

Your email address will not be published.


*