I want to bring up something I recently discovered regarding pet resistances, because I believe it has much bigger implications for pet balance than it may initially seem.
For a very long time, everyone assumed that pets followed the same resistance cap as players: 80%. Apparently, that assumption was wrong.
Pet resistance cap has actually been 100%
On the live version, pet resistances can effectively go all the way to 100%.
The character sheet and GT may display pet resistances as if they were capped at 80%, but resistance above 80% still works. If you manage to push a pet to 100% resistance against a certain damage type, the pet can effectively become immune to that damage type.
In other words:
Displayed pet resistance cap: 80%
Actual functional pet resistance cap: 100%
I have tested this directly.
SR 80 - Skeleton Cabalist with 100% all pet res (including physical res)
Full Altar Uber Nerfatal - Skeleton Cabalist with 100% all pet res (including physical res)
GT link: Cabalist, Level 100 (GD 1.3.0.8) - Grim Dawn Build Calculator
(BTW, Conjurer is much easier to achieve 100% physical res, example. I use Cabalist just for skeletons.)
What makes this particularly interesting is that this does not appear to be a recent bug. As far as I can tell, it has existed for many years. I went back and checked some videos from much older versions of the game, and the same behavior seems to have existed at least 6–7 years ago. Diviner’s Vision Set is most representative case. If pet resistance cap is 80%, Spirits can easily die for non-pet builds. But the opposite is true, even those who do not pay attention to pets at all, Spirits only die from enemies with RR skills.
I just discovered/realized this recently. I strongly suspect that very few players were actually aware of it. And that is where the balance implications become important.
Every historical pet build has, to some extent, benefited from this bug
Pet builds have always been balanced around their actual performance in the game, not some paper numerical values. But if this bug has existed for most of Grim Dawn’s history, then historical pet balance was also indirectly built around it. Nobody needed to consciously exploit anything. Players did not even need to know the bug existed.
If a build happened to overcap a pet resistance beyond 80%, the additional resistance was silently helping the pet survive. This means that when we looked at pets such as Briarthorns, Hellhounds, Familiars, Blight Fiends, etc. and concluded that their survivability was good — sometimes extremely good — part of that survivability may have come from resistance values that everybody assumed were doing nothing. In practice, pet builds have therefore been benefiting from this bug for years without realizing it.
Skeletons are obviously the major exception here, because their survivability has always been much more fragile to begin with.
Unexpected changes
The funny part is that I had already planned to report this. I recorded several test videos yesterday and was going to make a bug report after doing a little more testing.
Then the latest playtest patch arrived. And to my surprise, this bug appears to have been fixed. Pet resistances now seem to properly stop at 80%. As far as I can tell, this change was not mentioned in the changelog. I was honestly very surprised that the issue was found internally, because it has apparently existed for so many years and is extremely difficult to notice unless you specifically design a test around resistance values above 80%.
Still, fixing the bug itself is completely reasonable. The problem is what happens when this fix is combined with the other pet survivability changes.
Fixing this bug is already a very large survivability nerf
The current playtest already contains several changes intended to reduce pet durability. But I think the resistance cap fix is much more significant than most of the individual numerical adjustments. And more importantly, these changes are not independent. The old pet balance was created under conditions where this bug existed. The game was balanced based on how pet builds actually performed, and their actual performance was unknowingly benefiting from resistances above the supposed 80% cap.
So if the bug is now fixed, that removes a hidden source of survivability that has effectively been part of the pet ecosystem for years. Adding other survivability nerfs on top of that risks overcorrecting the problem. I think this is something that needs to be considered very carefully. If the 100% resistance bug is fixed, then perhaps some of the other survivability reductions should be smaller, or in some cases additional compensation may even be necessary.
Without this bug, some boss abilities become much more dangerous to pets. One reason this matters so much is that pets cannot avoid dangerous attacks in the same way players can. A player can reposition, dodge ground effects, react to telegraphed attacks, and deliberately avoid Sunder abilities.
Pets frequently cannot.
- They stand in ground effects.
- They get hit by Sunder attacks.
- They remain in dangerous AoE abilities.
And because of that, sufficiently high resistance has historically been one of the things that allowed them to survive mechanics that a player would normally avoid rather than tank. If pets were unknowingly reaching 85%, 90%, 95%, or even 100% resistance to certain damage types, that was an enormous defensive advantage. Remove that advantage, and some abilities that previously looked harmless to pets suddenly become extremely dangerous. This is especially noticeable with certain Sunder abilities and ground effects.
I provided a detailed calculation comparison in my reply.
Pet survivability after the fix
I want to be careful here because I have not tested enough builds yet to make a definitive statement about every major pet archetype. For the stronger pets, like Briarthorn and Hellhound, the situation may still be manageable. For example, in SR 36–37, my Beastcaller Conjurer still shows that pet builds can look extremely dominant against most individual bosses. In a clean 1v1 situation, pets may take very little damage. Vine is an exception, he can easilly kill Briarthorn and Hellhound. A few others bosses can sometimes kill pets or deal heavy damage to pets in 1v1, but the impression is not that profound.
However, the situation changes significantly facing with large groups of heroes and mobs, or when combinations of Nemesis together with some heroes. I am now seeing Briarthorns and Hellhounds die with noticeable regularity in those situations.
So I do not think the answer is simply “pets are still immortal” or “pets are now too weak.” The situation is more complicated than that, and it needs broader testing. But the resistance cap change is large enough that I think the entire survivability picture needs to be reevaluated with this bug no longer present.
Skeletons are where I think the problem becomes much more obvious. They were already fragile before this change. Removing the hidden benefit from resistances cap hits them much harder than it hits naturally tanky permanent pets.
Conclusion
In the end, I am not arguing that the 100% resistance cap bug should remain in the game. It is clearly a bug and fix it makes sense.
My concern is that its historical impact should be taken into account when evaluating the rest of the pet changes. This bug has likely been contributing to pet survivability for many years. That means removing it is not just a technical correction. From a balance perspective, it is a major systemic nerf. If the goal of the playtest is to reduce pets from “nearly indestructible” to something more reasonable, then I think the resistance cap fix itself may already accomplish a large part of that goal. Other survivability nerfs should therefore be evaluated again with the corrected 80% resistance cap in place, rather than simply stacked on top of it.
I will continue testing more pet builds, but I wanted to bring this up now. Hidden changes like this not being mentioned in the changelogs are honestly very frustrating, especially when they can have such a major impact on balance and testing.
.