As a devoted admirer of Aarvo Pärt, this music was awesome!
A couple of announcements in a light of the upcoming patch:
- All Phys retal builds are going to be exluded from the sheet.
- All Physical builds are going to be exluded from the sheet unless their performance in 220.127.116.11 (final version) is presented.
- The table for 18.104.22.168 is going to be frozen (the same way we did it for 22.214.171.124) once the patch hits.
3.1. No further changes in criteria will ever affect it.
- The rules (i.e. no pharma in CR) are going to be added to the OP. As for now they’re not written properly (and not placed anywhere).
- Point distribution for CR and SR are going to be rescaled and overall lowered.
Given the content of p. 3 I encourage everyone to submit builds that are not in the table as of now, because we havent been adding any in recent 2 months unless asked directly.
DW AA Vitality Reaper >126.96.36.199-188.8.131.52< (DW Melee Vitality non-set)The Tender Reaper. Consistent SR 75-76[SR 90] (%100 can kill Ravager of Minds/Flesh but I can’t since I choosed ‘‘souls’’ by mistake.)
Voldrak Cadence Witchblade(probably doesn’t rely on converted phys that much) >184.108.40.206< Fate of the Cadence, 2H Physical Cadence Witchblade. SR 90/Ravager of Minds/Mogdrogen
DW Fire EoR Templar >220.127.116.11<Melee Templar, wait it's Channeler! DW Fire EoR Templar. SR 85ish
S&B Vitality Oppressor(just for variety) >18.104.22.168< Oppressing Darkness. S&B Vitality Oppressor(not retaliation) 75-76/Ravager of Flesh
DW Melee Fire Sentinel >22.214.171.124< Fire Walk With Me! DW Melee Sentinel. SR 85(ish)
Fire FW Warlord >126.96.36.199< Fire Forcewave Warlord. Shar'Zul's Fireworks. SR 85 [No Greens]
S&B Vitality/Bleed Warder(got some buff in 188.8.131.52) >184.108.40.206< Warding Beyond the Meme... Vitality/Bleeding S&B Blood Knight Warder. SR 85ish
DW Melee Aether Sorc(For variety) >220.127.116.11< The Taken Sorceress, DW Melee Aether FS-SR 85
I don’t have any CR times, never tested those builds and no intention to test until I got a better gaming computer. Maybe someone adjust them for their playstyle and test&share it.
Naked CR: -
Celestials: Ravager, Callagadra, Crate
May I ask why it has to be with extra spawns?
EDIT: Clearly, I’ve been out of touch of the game for quite some time, but < 7min as the lowest bar for baked Cr + Extra spawn seems harsh?
IIRC from my good ol’ piloting days, I would be very satisfied with anything less than 8:30; with 9minutes being acceptable.
Also, if you want to truly make an objective scoring system, I’m of the opinion that taking the average of the fastest “Insert # here” runs should be used, and not just the fastest run.
Further, the consistency of the build should also be factored in, especially for naked Cr runs, with there being a minimum cutoff needed for naked specs. (I personally consider anything less than ~70-80% consistency in naked Cr a bust)
Lastly, regarding SR, should there be a punitive score for the number of deaths?
I’m guessing because that’s how dmt does it and people followed. But I’m not too sure. I wasn’t active when grimsheet happened.
I’m of the same opinion on the scoring system, especially regaring the uselessness of extra spawn, but i was in the minority. And changing the system requires a lot of work, and we barely have motivation to test things.
Also, how’ve you been?
You are alive!!!
Welcome back spanks!
Busy busy busy. There’s a lot more work to do now as a junior doctor because of covid - hence the hiatus.
But let’s continue this on PM so we don’t derail this thread.
It’s good to be back (in the limited capacity that I’m back in)
TL;DR - Love the idea @grey-maybe has, but am absolutely opposed to the practice of using the single best clear time as an objective measure. Simple solution to the issue would to get people to start doing triple reset Cr runs.
If you’re deadset on sticking to the way things are currently done, then in my experience, naked Cr normally takes about 30-40% longer than buffed bannered Cr. The faster the buffed/bannered clear time, the larger the gap grows.
A great example would be my PB blademaster - fastest Cr time by plasmo was ~3:55, fastest naked time by me was ~5:40
To make this reply relevant to the thread, I just want to add several things:
- Why I don’t like the practice of posting single best runs or using that as an objective measure
- The caveat that I could very well be misunderstanding what OP has in mind for this compendium as I’ve always theorycrafted with my own agenda in mind.
- Why I made naked Cr “my thing”
I was a little curious as to why < 5:30 was the lowest bar for normal Cr runs, but I took a quick look around the forum yesterday and was absolutely blown away by the clear speeds posted.
Despite of also being guilty of doing so from time to time, I personally really dislike the habit of posting only the fastest clear time.
I mean it’s fine if you’re doing so to comepete for a speedrun record, but it makes no sense to use it as a metric for a specs peformance. (I mean I wouldn’t trust any research paper which presents only a single data point).
There’s also the additional disadvantages of:
- Unfairly getting specs nerfed.
- Encouraging reckless piloting (after all, you only need to succeed once).
It was already bad enough to have this be common practice when posting a build, but to make a dedicated compendium centered around it?
No offence @grey-maybe (because I DO think that what you’re doing is a fantastic idea), but the skeptic in me wonders if the entries in this compendium are simply going to be a poor man’s version of @ArchHeretic’s Fastest cleaners in the world
Then again, take what I say with a huge pinch of salt.
Of all the veteran Cr-dedicated theorycrafters here, I probably have the lowest standards insofar as build performance goes probably because I favor different things when it comes to theorycrafting.
After all, I’ve always prioritized 2 things over performance:
- The creative and the unorthodox
- Player accessibility - i.e. never understood the point of writing a guide if the people following it can’t pilot the spec successfully in end game challenges
Which is why I made naked Cr my thing.
The rationale is that if I can complete Cr with no buffs/banners in X minutes, then surely most players will be able to do so with the help of tributes.
For the record, I am aware of this.
Apologies if I’m being rude, for I only wish to be direct.
However, to put the onus on theorycrafters/build posters to make the aforementioned change is to ignore the fact that this is your compendium. You get to decide what the submission criteria should be the same way you get to decide the format of the submission form.
We are not dead set on anything but “simply getting people to do that” won’t work as we don’t have such miraculous power.
Grim Sheet isn’t just a leaderboard with specific rules, it doubles as the balance research project. To serve that purpose it needs to encompass as much useful data as possible. We didn’t invent a ruleset, we simply took what people had already been doing and quantified it into scores. Sure the input data could had been better, but if only a handful of players complies with the more strict standard it would be a loss for us in the end.
I see. That’s a very sensible approach. I stand corrected.
But perhaps “average timing” and “run consistency” can be applied to solely naked Cr. Given how few players do that challenge, I would assume naked cr would be the easiest place to make this change.
I also think it’s the challenge which needs that change the most.
Because the randomness of mutators/spawns have a far more severe impact when playing naked. That 2-3s increase in kill time on a nem in buffed/bannered, can increase to 6-8s in naked. Which in turn may force you to kite and further prolong the run.
Also, one shots are much more common in naked cr.
If nothing else, these would be my proposed changes to the scoring system for naked (though my knowledge may very well be outdated):
- Make extra spawn optional, and maybe give it 0.5 (?) points if a player wishes to roll with it.
- If you’re starting Cr @ <5:30, then I would suggest starting Naked Cr @ 7:30ish (maybe 7:45)
- Have it scale accordingly as normal Cr.
Currently a sub 6 naked Cr run is worth 2 measly points! The only build of mine which could do a sub-6 naked run was the PB blademaster, and it gets a whopping 8.2 points for its sub 4 run in normal Cr.
Further, I believe there’s only 1 other build in OP’s table which is capable of doing a sub 6 for naked (and it’s a retal build).
The amount of score awarded for naked CR is like that because it is a bonus category that not many people do. Same for super bosses.
Currently we have a situations when one player might focus exclusively on CR but hate deep SR, and another might focus exclusively on SR and hate CR. It’s very common.
As a result their builds are somewhat undervalued compared to people who play both. It isn’t unusual for one or two builds secure a good position for the class in a tier list simply because their authors bothered to have all the checkerboxes.
If we add a few other categories that could let a build earn just as much it would further skew the scores in favor of select authors who don’t mind to play every mode.
Because just naked is not that much different from buffed ones? It doesnt tell more about survivability/efficient damage ratio compared to buffed or naked extra spawn, being something inbetween the two.
thats very outdated.
Sheet has descriptive function, not prescriptive. Besides, N numbers of runs are not all that different from picking the fastest one because:
a) its still the pilot who picks runs for the statistic, therefore they are the one in control of data they present.
b) for most builds and for most runs it will be “fastest run +constant”. Because the ranking is built around “comparing”, it wouldnt change a thing.
No way to prove other than taking pilots words for truth.
Yes, i should come to their houses and force them to do so. Just as simple as it is.
With how fast people burn out grinding the same build in CR (especially when the build is cancerous to play) that’s the only way to make them do “triple reset” (if by that you mean 12 runs, if you mean just 4 runs than they already do it afaik).
thats outdated as well. The difference between buffed run and naked extra spawn for most builds is around a minute.
Most people either post multiple CR runs or at least vocal about them.
People (pilots) wont grind absolute piece of glass to death just to get that one run, because burn out i’ve mentioned recently. Build with ~50% succes rate is considered glass af.
Thats unironically kinda offensive.
Because creativity is so easy to meausure by a liner and put in rankings.
The main problem is GD is a very complicated (and still constantly changing) system with not nearly enough participants to gather the data. Also of those who are still playing you can never imagine to force people to gather more data because lets all be honest SR and CR possess very limited content to be a fully fledged replayable endgame. The main content of this game for tryhards is making builds.
So we just have to make do with bloated results of select builds where authors really went all out to grind their points. But the positive data that GrimSheet provides is the class power. With so many build entries the class data somewhat evens out. You can look at the table and see what class and dmg type underperform.
You misunderstand me, friend.
I’m merely suggesting having a video recording of 3 (or 4) consecutive runs be part of the submission criteria.
Really wasn’t trying to be a dick mate.
Nothing but love (and unhelpful suggestions) from me. <3
Wasn’t suggesting trying to measure this
Just trying to explain why I may be missing the point completely.
That’s so surprising. Maybe one day I’ll have another go at it myself.
Out of curiosity, when you say most builds…are you talking about builds posted by players like DMT? Or most Cr builds listed in the table?
Not trying to be argumentative (or passive aggressive) here, but a quick glance at the table doesn’t seem to reflect that. Am I missing something?
I mean, while timings have gradually gone down over the years, my (very possibly outdated) experience playing both naked and normal Cr has shown that 30-40% correlation to be rather consistent.
Oh hey, noticed I’m not on there, here ya go:
Patch Version: 18.104.22.168
CR timing: -
Naked CR: -
Retested a first physical build: [22.214.171.124-126.96.36.199] Siegebreaker Paladin: All Celestials, SR 100 (vid)
Not only does it holds up the conversion rework, it now clears SR 100
Can’t wait to see how new builds perform with the balance changes!
Finally killed Callagadra on my Dark One Cabalist which was essentially unchanged since my last post. And it only took me a year! No video though since I honestly didn
t think Id be able to do it. Also it was just me running around her for 40 minutes.
Grimtools: Cabalist, Level 100 (GD 188.8.131.52) - Grim Dawn Build Calculator
Indeed, I have been looking over this sheet and am unsure whether it is worth it to submit some of the builds I have made and posted. I don’t play CR as I don’t like time-chasing, I only do SR. I am sure a good pilot could do well in Crucible as well with them, but that is not me.