With the next Vengeance test coming up in December, we wanted to give everyone an update on the Vengeance testing and where we’re headed with Cyrodiil. We recognize that you all have questions about next steps and what these tests mean for the future of Cyrodiil, and we’re here to answer some of those questions. We also want to give you as much context and info as possible, in the name of transparency.

The Goals
To recap where we started and where we’ve been with the Vengeance tests, earlier this year we set out to try some new things with Cyrodiil to address the following specific goals:

• Significantly increase the player cap in a Cyrodiil campaign so that campaigns feel lively, full, and there’s lots of action to enjoy.
  • Cyrodiil was originally designed to support 900 total concurrent players (300 per alliance.) We have not been able to support those target numbers in the current non-Vengeance Cyrodiil campaigns. We are able to hit 900 total concurrent players with the Vengeance ruleset.
• Reduce frustrating latency and related game performance issues in Cyrodiil, especially during large, sustained battles. And in turn, increase the fun and enjoyment.
• Through testing, determine if overall Ability complexity is the main cause of server stress and game performance issues in Cyrodiil, especially during mass battles and in high-population campaigns.
  • We did test other potential causes of server stress and game performance issues in Cyrodiil during the Vengeance tests, such as item sets and procs, consumable items, siege weapons, quests, vendors, etc. Ability complexity was our primary suspected main cause factor, though.

Test Summaries
Since March, we have held three Vengeance tests. The next one will begin on December 3, 2025.

For the first test, which was on the PC EU and NA servers, we introduced the Vengeance ruleset including normalized character stats, attributes, consumables, and Vengeance-specific weapons for all classes. We also disabled things like the Champion System, all quests, item sets, and access to banks, vendors, and crafting nodes.

The specific goal of this test was to give us a starting point – to reduce things to the most basic level so we could evaluate the findings and begin adding things back in. At the end of this test, we saw the most players ever in Cyrodiil at one time in a single campaign, the largest sustained battles we’ve ever had in Cyrodiil, and the best overall game performance we’ve ever seen in Cyrodiil. By all accounts, we were on the right track based on our goals.

Survey results were also the most positive following this test compared to subsequent ones, with many participants saying they enjoyed the higher population, improved game performance/lower latency, and overall experience. The aligned Golden Pursuit was also noted as a positive. The loss of unique class/character builds and customization was (and continues to be) the biggest negative point.

For the second test, we welcomed the live console EU and NA server communities. The Vengeance ruleset was largely the same as the first test, with the addition of Vengeance versions of skills from all Weapon skill lines and the Assault and Support Skill lines to the available class templates.

The specific goal of this test was to build upon the baseline we set during the first test, slowly begin introducing more skill lines and abilities, and evaluate the results. At the end of this test, we saw similar results with better game performance, lower latency, higher population, and larger sustained battles than possible in other Cyrodiil campaigns.

This test overlapped with the Zeal of Zenithar event, which we recognize not everyone enjoyed.

Survey results for this test were similar in sentiment, leaning positive. Over 80% of participants rated the added skill lines favorably (“OK”, “Good”, or “Great”), sharing appreciation for the variety and balance, while also noting that there’s room to improve.

For the third test, we layered upon what we introduced in the two previous tests, and added in meatbag catapults, as well as performance-tailored skill lines for the Fighters Guild and Mages Guild, plus an armor skill line with active abilities for light, medium, and heavy armor. We also introduced certain progression and cosmetic systems into Vengeance, including Skyshards, mount selection, titles, and achievements.

Similar to the second test, our goal was to introduce more things that players enjoy and expect in a Cyrodiil campaign, and monitor the impact on latency, game performance, and overall experience.

We did not run a Golden Pursuit during this Vengeance test, and saw disappointment about that. We also ran this test during the Undaunted Celebration, which some players understandably noted they had higher interest in participating in versus this test. Even though population in Vengeance was lower during this test, it still performed well and participant numbers were still higher than our typical population caps in a normal Cyrodiil campaign.

Survey results for this test were more mixed than the results following the first two and we saw less positivity overall. More survey participants noted fatigue with the tests and a desire to be able to play their own unique characters and builds in Cyrodiil, and frustration that the test ran during another in-game event.


Test Learnings
Throughout the first three tests, we learned with certainty that in order to deliver a performant Cyrodiil, to support a large-scale PvP zone with mass-scale battles, the abilities, procs, passives, etc. must be lighter versions of the ones that exist in the rest of the game.

The set of four graphs below illustrate the differences in population as well as the server frames per second between the Vengeance Cyrodiil campaign and non-Vengeance Cyrodiil campaigns. (On the left of each graph is the non-Vengeance campaigns and on the right are the Vengeance campaigns, on the PC EU and PC NA servers.)



Next Steps
For the test in December, we will introduce Vengeance-specific Perks and Loadout systems for character/class templates. These systems are designed to give you a bit more variety over your builds compared to what was available in previous Vengeance tests. Specifically, you will have more control over your stats with four pre-build stats packages called “Loadouts” and “Perks” are passives that give extra combat effectiveness and bonuses to your characters. The intent with these is to give characters a boost that is comparable to a single 5-piece item set that is purely passive, like Julianos.

We will also be adding a Vengeance-specific inventory, which will store all your Vengeance items. During this test, you will also be able to collect regular items while in the Vengeance campaign – those items will be placed into your regular inventory. Many systems that were previously turned off in Vengeance will also be turned back on including quests, vendors, and leaderboards. Scattershot and Oil Catapults will also join the available options for siege weapons, and Keep Recall Stones and Channeled Repair Kits will also be added.

We’ll share more detailed notes ahead of the December test. We’ll monitor the impact of these additional systems on latency and game performance, as we have during prior tests.

Lastly, the Gray Host campaign (as it is now) will be up during the second half of this Vengeance test and will monitor server performance for both campaigns. This comparison will allow us valuable side-by-side data. This will be our final “adding new things” test where we compare the game performance of Vengeance with what a campaign looks like with all the systems turned on.

We have another test or two planned for next year, for the sake of evaluating healing versus damage concerns. We’ll share more about those next year.

So where does this put us, and where do we as a dev team realistically think we’re headed for the future of Cyrodiil? With the caveat that the December test still needs to happen, we see two realistic paths forward:

• Scenario 1: We would open a Vengeance Cyrodiil campaign sometime next year with a special ruleset based on the previous and upcoming tests, and leave Grey Host open as it is now.
• Scenario 2: We would close all existing Cyrodiil campaigns and open one or two Vengeance ruleset campaigns sometime next year.

As a dev team, Scenario 1 is the one we strongly prefer and is the most likely. We want you to have a choice between playing in Vengeance or Gray Host, and would closely monitor both campaign populations to help inform any additional actions we should take moving forward.

We recognize that some of our players would prefer there be no changes to their characters, effectively how they are in the Gray Host Cyrodiil campaign, while enjoying the higher population and reduced latency/game performance issues of the Vengeance campaign. This is not something that will be possible. Based on what we have learned from the tests so far, we can offer one or the other, not both, and we want to be transparent about that.

Lastly, we do want to share a few early bits about some things that we are working on for PvP. You’ll hear a little more about these in January. For players who wish to enjoy a PvP experience that is more like Grey Host, we are in the early stages of working on a mid-size PvP space. It will be smaller than Cyrodiil but larger than our largest Battlegrounds, and offer enough room for 3-sided keep warfare and an overall similar experience to Cyrodiil. The goal with that will be to allow players to have their full suite of abilities, unique builds, equipment, etc. just like in Gray Host. We are also working on a PvP progression system that we’re excited to tell you a little more about early next year.

Thank you all for your continued feedback and support. Your participation in the Vengeance tests and related surveys has been greatly appreciated. We’d like to remind everyone that when we send out surveys where it’s important we are able to verify that participants played the content, we have to send out the surveys via email. Please consider opting in to these emails if you haven’t already, so you may be included in future survey sends. Thank you!