Abaddons Mouth
Adding precision to pets is bugged
Abaddons Mouth
Interesting and concerning.
Kudos to you for looking into this. Did you happen to record the tests? If so, can you please share?
I think several of us would do well to do the same and look at it as well and record it. The more attention it gets, the more likely it is to be fixed … and with all the code changes coming, now could be the ideal time for this to get changed too … there is already evidence of ArenaNet looking at pet stats.
Try your best to not make mistakes, but, when you do make mistakes, learn from them.
Better yourself.
I didn’t record it on video, the only record I have left is the notes I made when I counted the crits. I used a online tally counter where I could bind hotkeys to increasing one of the 2 numbers.
Abaddons Mouth
Thanks for that testing, I’m surprised nobody has done that before! I would say the best thing to do is record 100 hits from each pet and post the video in the bug forum, speeding it up may be a good idea
This is concerning indeed, I wonder if precision is not the only bugged stat!?
Wasn’t there a thread not too long ago about the condition damage from Expertise Training not applying to pet condition damage? Precision is probably not the only bugged stat at the moment.
Will update once Path of Fire releases.
Oh ya, the pet precision is kittenED, I think I’ve had my ravens at like 90% crit chance and they only crit like 4/10 times consistently. Why did I never bring it up?…. /shrug
Am I good?… I’m good.
Ferocity trait seems to be bugged too.
Herald of Ventari
Please go ahead and report these things in the bugs section.
Wasn’t there a thread not too long ago about the condition damage from Expertise Training not applying to pet condition damage? Precision is probably not the only bugged stat at the moment.
condi dmg from expertise training works. But the trait doesn’t always trigger immediately. Sometimes with salamander drake I try initiate with fire breath and it wont trigger trait till half way though fire breath and sometimes after its done channeling.
Wasn’t there a thread not too long ago about the condition damage from Expertise Training not applying to pet condition damage? Precision is probably not the only bugged stat at the moment.
condi dmg from expertise training works. But the trait doesn’t always trigger immediately. Sometimes with salamander drake I try initiate with fire breath and it wont trigger trait till half way though fire breath and sometimes after its done channeling.
Thats coz you need to be in combat for the trait to trigger. Salamander will have hit at least a few times before you enter combat.
Wasn’t there a thread not too long ago about the condition damage from Expertise Training not applying to pet condition damage? Precision is probably not the only bugged stat at the moment.
condi dmg from expertise training works. But the trait doesn’t always trigger immediately. Sometimes with salamander drake I try initiate with fire breath and it wont trigger trait till half way though fire breath and sometimes after its done channeling.
Thats coz you need to be in combat for the trait to trigger. Salamander will have hit at least a few times before you enter combat.
That simply isn’t true.
I can be ooc and send pet with f1 and it will get +condi dmg instantly and recall it with f3 and it will return and still have +condi dmg while neither of us have entered combat. I don’t know why that even happens, maybe it was a qol change anet discretely put in? Shame it doesn’t happen reliably with f2.
Stupid that they thought the pet should need to be in combat for the trait to take effect anyway.
A request for science!?
Your test was performed with two drastically different pets, it may be worth testing each individual pet, but most important is to ensure your results are repeatable. For this, I used a jungle stalker attacking the indestructible golem with all precision boosting traits and a second time without any precision boosting traits. For these two trials I found that:
Cat, base:
Crit: 67
Nor: 111
rate: 37.6%
expected: 58.0% 2061 precision
Cat, 30 empathy + spotter:
Crit: 131
Nor: 70
rate: 65.1%
expected: 79.5% 2511 precision
For the cat boosted by 30 empathy, you’ll just have to trust my judgement, but the attacks definitely seemed to have more power behind them, so our pets or at least our cats do seem to scale with power.
As you can see, in both of these trials, neither result seems to match what we would expect. However, the results of our two experiments do appear to agree with one another, if we assume there is a +-5% range of uncertainty. So I stipulate that perhaps the conclusion to draw is not that our pets do not scale with traits, but that their actual crit rate is below what we would expect for a player with the same precision. In all three trials, the measured crit rate was 15~20% less than what would be expected, the range of which is similar to the difference between our two experimental results.
Or the indestructible golem is bad for testing crit rates, I’ll definitely be looking into this topic more later.
“There, it’s dead and it’s never coming back!” – Famous last words
(edited by HotHit.6783)
Stupid that they thought the pet should need to be in combat for the trait to take effect anyway.
This. I don’t know why those bonuses couldn’t just be applied 100% of the time. It’s not like there’s an advantage when both the player and the pet is out of combat.
Will update once Path of Fire releases.
Stupid that they thought the pet should need to be in combat for the trait to take effect anyway.
This. I don’t know why those bonuses couldn’t just be applied 100% of the time. It’s not like there’s an advantage when both the player and the pet is out of combat.
They probably didn’t want pets being set passive all the time as aoe buff/debuff bots.
Stupid that they thought the pet should need to be in combat for the trait to take effect anyway.
This. I don’t know why those bonuses couldn’t just be applied 100% of the time. It’s not like there’s an advantage when both the player and the pet is out of combat.
They probably didn’t want pets being set passive all the time as aoe buff/debuff bots.
Sadly, you have to use them that way some of the time, others, you want the buff before combat so it doesn’t have to stand around channeling a shout for 3s, not doing damage. Also, why make a trait like Expertise Training NOT take effect until you get into combat? Makes no sense.
One less thing to compute if it were just on all the time.
Also, why make a trait like Expertise Training NOT take effect until you get into combat? Makes no sense.
None of these “combat only” traits make any sense.
Not even furious grip and tailwind. They got internal cooldowns matching the cooldown of weapon swapping in combat, so it’s not like you can spam them anyway. Engies get acces to perma swiftness when equiping a kit, with no combat requirement, so why is the ranger limited? Same thing with protect on dodging compared to thieves and their might and swiftness on dodge. It doesn’t seem very well thought out at all, and besides, removing the requirement would grant the ranger some welcomed mobility outside of leaps.
I think the pet traits where made like this because you previously could take advantage of F2 recharges on pet swap back in the days when using Concentration Training to stack tons of boons, but that isn’t even a thing anymore. No wonder why people get the feeling that the ranger has been overlooked as far as balancing goes for a long time when they don’t even care to revert changes that doesn’t make any sense anymore.
(edited by Lazze.9870)
I’m right there with you Lazze. Wondered this myself for quite some time. There is a lack of consistency that I hope ArenaNet addresses … either with:
- consistency between same mechanics
- compensation for the disparities in the mechanics to balance out those differences
Try your best to not make mistakes, but, when you do make mistakes, learn from them.
Better yourself.
I’m right there with you Lazze. Wondered this myself for quite some time. There is a lack of consistency that I hope ArenaNet addresses … either with:
- consistency between same mechanics
- compensation for the disparities in the mechanics to balance out those differences
Took them two years to adress the inconsistency of ranger signets compared to other professions. I got no high hopes for them to adress the remaining inconsistencies in the near future.
Not sure if they’re even aware of them at this point. They specifically said the old ranger signets where a bit unfair compared to other professions.. Well, guess what Anet, you got a bunch of ranger traits suffering from the same thing.
Edit: when I come to think about it, tailwind and furious grip didn’t use to have ICD. I bet the in combat requirements are still there for the same reason it’s still there for pet traits – they don’t pay attention to the actually changes which make these limitations not only redundant but even unbalanced. And they actually wonder why people think they hate rangers?
(edited by Lazze.9870)
I imagine it isn’t hate but oversight. They look to be trying to gain some more consistencies across classes. From a programmer’s perspective, this is good as it means you don’t have a completely separate chunk of code supporting one class’s Signets and another for another class’s Signets and then yet another for a 3rd class’s signets.
I think some things got rushed when the game was released so they didn’t have the opportunity to do this then. On top of that, they have a nice parent company that the rumor-mills say has been pretty tight with the leash, so that can hinder getting around to the code-quality fixes that need to take place … not to mention having X resources and needing Y resources for new things like an entirely new Specialization System, Living Story, bug fixes, expansion, etc.
Programming large projects isn’t as easy as many people seem to assume it is. Managing them is just as big of a challenge.
Try your best to not make mistakes, but, when you do make mistakes, learn from them.
Better yourself.
I imagine it isn’t hate but oversight.
Same. I just find it funny.
I find some funny. Others I look at and think … wow, some thought but not enough …
Example:
At launch:
- Could run through Mesmer Temporal Curtain to stack Swiftness
- Could run through Mesmer Veil to stack stealth
- Sitting in Guardian Staff Symbol only refreshed duration, did not stack it.
After launch
- Can’t run through Mesmer Temporal Curtain to stack Swiftness
- Can’t run through Mesmer Veil to stack Stealth
- Can sit in Guardian Staff Symbol to stack Swiftness
… moved in opposite directions …
Try your best to not make mistakes, but, when you do make mistakes, learn from them.
Better yourself.
All the pet traits used to be active out of combat, it led to issues. For example, you could pre stack up to like 2 minutes of boons in only a few seconds. In addressing this they further prevented this type of behavior by changing all pet related things to be in combat only.
Am I good?… I’m good.
All the pet traits used to be active out of combat, it led to issues. For example, you could pre stack up to like 2 minutes of boons in only a few seconds. In addressing this they further prevented this type of behavior by changing all pet related things to be in combat only.
The problem is that they didn’t do this for everyone. Some other classes can still do this with the right traits and/or stats and just set a macro to cast whatever every X seconds while they eat. Then they come back with several minutes of whatever boon(s) they were able to stack.
Try your best to not make mistakes, but, when you do make mistakes, learn from them.
Better yourself.
All the pet traits used to be active out of combat, it led to issues. For example, you could pre stack up to like 2 minutes of boons in only a few seconds. In addressing this they further prevented this type of behavior by changing all pet related things to be in combat only.
Still doesn’t make sense for the other pet traits such as Pet Prowess, Expertise Training, Malicious Training, and Compassion Training. The second a pet attacks or inflicts a condition, it’s in combat – no need for it to apply seconds later. It’s just redundant.
As for the boon duration, the only pet that could take advantage of pre-stacking before combat was Jungle Stalker, but even with the trait and F2 reduced cd, it couldn’t even keep 100% uptime due to the 3 second cast.
Will update once Path of Fire releases.
As for the boon duration, the only pet that could take advantage of pre-stacking before combat was Jungle Stalker, but even with the trait and F2 reduced cd, it couldn’t even keep 100% uptime due to the 3 second cast.
The in combat requirement was added at a time when you could play around with pet swapping and refresh pet skills (perhaps it worked with mighty swap aswell, not sure). You used to be able to stack up a stupid amount of boons with certain pets. Hence our earlier discussion about Anet not reverting these kind of changes when they’re no longer relevant.
If I remember correctly, I believe it was the Siamoth with the consume plasma. Players would use it, reset the F2, repeat.
Try your best to not make mistakes, but, when you do make mistakes, learn from them.
Better yourself.
That was also back in the day when we could also equip the same pet twice. I recall that was changed not too long after launch.
Will update once Path of Fire releases.
(edited by Wondrouswall.7169)
I hope this post gets recognized by the forum mods, I would very much appreciate an answer regarding this issue.
For this, I used a jungle stalker attacking the indestructible golem with all precision boosting traits and a second time without any precision boosting traits. For these two trials I found that:
Cat, base:
Crit: 67
Nor: 111
rate: 37.6%
expected: 58.0% 2061 precisionCat, 30 empathy + spotter:
Crit: 131
Nor: 70
rate: 65.1%
expected: 79.5% 2511 precisionFor the cat boosted by 30 empathy, you’ll just have to trust my judgement, but the attacks definitely seemed to have more power behind them, so our pets or at least our cats do seem to scale with power.
As you can see, in both of these trials, neither result seems to match what we would expect. However, the results of our two experiments do appear to agree with one another, if we assume there is a +-5% range of uncertainty.
For a quick, rough guide, uncertainty can be calculated as sqrt(1/n) where n is the number of samples. Since your samples were n=178 and n=201, your uncertainties are +/- 7.5% and +/- 7%.
The exact value is 1.96 * sqrt (p*(1-p)/n) for a 95% confidence interval. That will correct for cases of extremely common or rare events (p nearly 0 or 1). But most of the time sqrt(1/n) gets you close enough.
I’m right there with you Lazze. Wondered this myself for quite some time. There is a lack of consistency that I hope ArenaNet addresses … either with:
- consistency between same mechanics
- compensation for the disparities in the mechanics to balance out those differences
That’s been a long-time issue with ranger. It was almost like someone high up on ArenaNet’s staff really disliked rangers, and kept crippling ranger skills. So you had stupidly obvious imbalances early on like signet actives affecting all classes, except rangers who needed to take a GM trait to get the effect of their signet actives. Or warrior banners providing better effects for more people, with more uptime, and better mobility than ranger spirits. (Actually in that case I think it was more warriors being the favored child for a long time.)