I really love the official trade site,
The best improvement would be to have the filters automatically filtered depending on the base or unique item selected.
It is kind of silly that we see all possible outcomes across all items even when narrowing the search.
Theoretically this is a simple fix, considering they probably have tables out there in SQL that they should be able to narrow quite easily, i know it is easier said than done but this would be the best QOL improvement
So what you're saying is make it so if you plug in a Loreweave for example, the only affixes which you can filter would be those found on a Loreweave, and not the pages of affixes which would be completely irrelevant to a Loreweave. I really, really like this suggestion. It would make it so much less tedious to search for jewels if you could plug in a Searching Eye Jewel and only be able to choose from affixes that can actually roll on said jewel.
for now, you can get around this by searching for the unique item, and clicking the icon below the image that looks like a magnifying glass over some lines. it puts all the item mods (without values) onto the stat filters. It does take a few more clicks though, and isn't that useful for non-unique items, so a faster implementation of it would still be nice.
Yes this is exactly what I am saying. If they could do this it would be such a huge QOL fix.
As long as you don't die
Is this build HC viable?
50% chance to surive any damage, either it hits or not, simple math.
Poet's Pen APS Breakpoints
In the past its been suggested that the optimal amount of APS for Poet's Pen (ie highest APS you can have without skipping casts, therefore the most procs/sec) is equal to the pen's cooldown plus 1/60th of a second to account for server tick rate shenanigans. In my experience this is completely incorrect, here's my theory:
When trying to tweak my maximize APS without skipping casts today I realized that the optimal amount of APS with 23% cooldown speed (belt + gloves) was the exact same amount as with 12% cooldown speed (just belt). Now if you merely had to add 1/60th of a sec to the cooldown then we should see a different APS cap at each different amount of cooldown speed, but that's clearly not what I was seeing here. So the next question was how much CD speed do I need for it to actually do something? At 26% nothing changed, but at 27% something had clearly changed and I was able to raise my APS again. Not a small amount actually, a massive amount: I was able to raise my APS from 4.32 to 5.04. This pretty clearly indicates some sort of CD breakpointing effect.
After a lot of mucking about in google spreadsheet I think I have a working explanation: While the server has a higher tickrate for other things, it progresses cooldowns in intervals of 1/30th of a second. Furthermore, the server doesn't allow for perfect chaining of cooldowns; you can't cast a spell whose cooldown just finished on the same tick. This means that you need to calculate your cooldown, add some small value, then round up to the nearest multiple of 1/30th to get your real cooldown. This produces APS and CDR breakpoints like so:CDR APS 0%-7% 3.750 APS 8%-26% 4.286 APS 27%-51% 5.000 APS 52%+ 6.000 APS
Now if you fiddle with your APS around these breakpoints you'll notice that you're actually able to go slightly above those numbers without skipping casts. In my experience I was able to do 3.769992, 4.32432, and 5.04504 APS respectively without skipping casts. My theory for this is that your "real" attack time gets rounded to the nearest 1/nth of a second where n is the server's tickrate. With the data I have it seems like n is either 210 or 240, not sure which but it also doesn't really matter. Those were the only (reasonable) valid integers where those APS values I just listed would round down to fit into the cooldown while also having 1% extra IAS round up to no longer fit the cooldown. I think this also means that, for example, that any APS values between 4.95 and 5.05 are an identical attack rate and therefore an identical amount of procs/sec.
So all that considered would produce APS caps that look like this (assuming n=240):CDR APS 0%-7% 3.721-3.780 APS 8%-26% 4.248-4.324 APS 27%-51% 4.948-5.053 APS 52%+ 5.926-6.076 APS
So if all this is correct, to figure out how much APS you need you find the row in the above chart that corresponds to your cooldown speed, then just get any APS value within that range (they should all be the same procs/sec).
I'm curious if anyone can provide any evidence of these breakpoints & caps being incorrect, if so let me know.
PS: The in-game character sheet seems to do some weird rounding as its different than what I would get if I manually calculated the APS by more than 0.01 sometimes, so I'd recommend using PoB's APS number for that.
I was always noticing skipped casts after I go higher than 3.75 aps without CDR.
P.S. An actually useful post? Appreciate your work OP, thanks.
Do remember tailwind fucks with all this as it isn't sheet APS so you have to throw in one more calc there!!!
Your video just tells the same thing OP did, except with less precision, math, and without a table.
Keep in mind that your internet/ping/predictive mode can also cause missed casts. Unless you have a very stable, low latency connection you will be missing some casts. It doesn't take much to get slightly out of sync.
If you're just coming into the PoE scene you have no idea how much hard work this guy has put in. He grinded out his viewerbase during the HC elitist era, all his success is well deserved as he's now one of the most popular streamers for our game.
mathil's a really articulate guy, i guess a lot of streamers are just good at talking in general, in order to engage well with their audience if they want to be successful
In havocs interview, Kripp was also credited for getting his attention towards PoE. I guess that OJ chugging ex nolifer really deserves his special thanks in the game credits.
And a mathil interview without any pics seems like a waste
The subreddit has changed a lot, it used to be much more toxic with SC players looked down upon with disdain. SC and HC had separate league mechanics as well, if you were a SC streamer you were lucky if you could get a hundred viewers. The merging of challenge leagues and rise of SC streamers has honestly reduced the toxicity and made the community more accepting.
What is the 'Mathil Effect'?
Something that started out innocently enough but grew to be somewhat of a cancer.
All hail our lord and savior....Tencent!
Indeed. This was an internal initiative that has been going on for several years.
Thats basically what I got from that. GGG did nothing about botting for years and Tencent cleaned that shit up.
Granted its probably in preparation for p2w features on the Chinese servers but its a perk for us.
It doesn't necessarily mean it actually was Tencent though. Basically any time something changes good or bad now, people are gonna say it's Tencent, when it could just be something that would happen anyways.
Slap that on the back of a playing card deck.
Amazing job! The uniformity soothes my OCD.
*slaps on back of card deck *
Just activated the act 2 boss Vaal Oversoul guys.
Looks like I have some time to kill, how you guys doing??
i usually finish 2-3 techies games before i can atack
Did you already make any plans for the time it takes for the door to open after you defeat him?
submitted 37 minutes ago
Door to act 3 should be open any time now.
not nice to see a fucking techies picker, though...
Banana plants grow up to 25 feet high, and their leaves can grow to be 9 feet long and 2 feet wide. Their roots can be hundreds of years old.
I'm a Bot bleep bloop |  Unsubscribe | 🍌
I'll offer you 36c, you profit 35c.
Good bot lmfao
Woah, you can 3D print metals now? Very neat! But how big is it? Over a kilogram seems big!
Probably hotfixed tomorrow. FeelsBadMan
The Lost Maps prophecy finally has a use
If you just do a couple t15 maos instead you can get elder as quickly, my average is around 2 maps when farming uber elder, I have done over 120 kills this league.
In before "delete this" posts