So, I really enjoyed the stretches of this game where I could push the story forward, but after getting hard stuck for about five or six hours, I noticed that I quit on the game. Haven't played it for a few weeks now.
It's a good game, and people rave about the ending, so I'd like to un-giveup. Does anyone have any non-spoiler suggestions? From browsing this sub a bit before, I know that confused people are asked to upload their ship log, so here is mine.
It's so small I don't know how you can decipher it, lol. Anyway, the only "missed something" notifications I have are in Sunless City / High Energy lab, and I'm at my wits end for what to do there. Already explored Eye Shrine, found the lab, solved the angler fossil puzzle, etc. In the lab, there's the tablets, already did the black hole / white hole thing with the scout...
As I guess you can see, I'm stuck with these mysteries:
-Black Hole forge (I don't even see a way in, but also haven't made it to top floor of Hanging City).. have bounced off the side a bunch
-The lab in the core of Giant's Deep (partially spoiled, I know the jellyfish are somehow related but I haven't found the connecting clues yet so waiting)
-Dark Bramble (have found the obstructive teleports at the signal ground zeros but nothing else except lots and lots and lots of dying)

• Ash Twin Project (it's hard to find the teleport spots! the only one I know of is in Giant's Deep, which never works even when launched above atmosphere, and Ember Twin (the structure is obvious but can't find the warp spot) (not counting White Hole Station)
  
• Quantum moon / gravity cannon / Ancient Civ shuttle on the Black Hole Planet
  
• Comet (spoiled in that I know something is there)
  
• Sun station
  

-Lakebed cave (I swear I've found it - the north pole area of Ember Twin is only so big, but I wander around it and nothing happens and it fills up with sand and never gets marked complete)
Any help appreciated!!!
https://preview.redd.it/luztnxmoyu1d1.png?width=1518&format=png&auto=webp&s=a059eb43f8f647121092476c0b1080dcbd22190f

TLDR - Shiny Weapon Drop Rate is 3.85% (1/26~) base, from any source. probably. With the 'Drop Rate Buff' added on May 21, 2024, that is likely doubled, to a 7.7% (1/13~) chance.
I tracked 1k drops for drop, attunement, shiny, and perks (just to quantify how unlucky i am)

Shiny Drop Rate over time

https://i.ibb.co/pPNv1p5/image.png
As the sample size builds up, there is some fluctuation until it stabilizes between 3.5% - 4.0% Sample is obviously not large enough to be certain. But based on Bungie's historical enthusiasm for using the number 7 literally everywhere, I am drawing the conclusion they tested the initial drop rate as either 7, 7.7, or 7.77%, and found it a bit too high, so sliced it in half. With today's 'BRAVE drop buff', it appears to be a drop rate buff specifically for shiny drops, which is likely doubling it, to restore the drop rate to 7.7% (ish) (probably)
and some of the raw stats -
Some of the finer details you may be interested in:

Raw Info
Total Weapon Drops	1025
Total Shiny Weapon Drops	38
Total Drops while Attuned to Any Gun	1011
Total Drops, Dropped Gun is Attuned	493
Attune Hit Percent	48.76%
Shiny Percent	3.71%
Shiny Hit Rate	1 / 26.97
Drop is Shiny, AND is Attuned	23 / 38
Shiny=Attuned Percent	60.53%

Gun	Total Drops	Attuned Drop Hits	Attuned Drop Chances	Attune Hit Pct	Attune Hit Rate
recluse	97	0	0	0	0 / 0
hungJury	89	0	0	0	0 / 0
succession	91	0	0	0	0 / 0
edgeTransit	134	94	183	51.37%	94 / 183
fallingGuillotine	0	0	0	0	0 / 0
elsiesRifle	106	44	89	49.44%	44 / 89
mountainTop	79	24	59	40.68%	24 / 59
midnightCoup	170	121	242	50.00%	121 / 242
hammerhead	132	106	220	48.18%	106 / 220
forebearance	0	0	0	0	0 / 0
blastFurnace	63	63	151	41.72%	63 / 151
lunasHowl	64	41	67	61.19%	41 / 67​

Attunement Rate for Mountaintop, Luna, are a bit off, but the sample size is small enough to attribute it to luck. You may have noticed an oddity though - 0 drops of Falling Guillotine, or Forbearance. So lets also get into some drop mechanics / Onslaught Tips

Weapon Drop Pool

Your initial drop pool, is only the first 4 weapons available - Recluse, Hung Jury, Succession, and Edge Transit. The remaining 8 weapons cannot drop, until you FULLY complete the associated weapon quest. Specifically, when the quest makes you Attune to the weapon on step 2, for the first time, that adds the drop into your loot pool. If you just don't do the quest, don't attune, etc - you can remove guns you don't want, from your loot pool. Between the weapon unlock mechanics, the layout of the map, and the initial plan of '6 extra weapons unlock 1 per week', I believe the initial plan was 8 weeks of Onslaught, with 1 unlock per week, and that was 'buffed' before the announcement to 6 + 6 over weeks, and finally into 6 + 6 over 3 weeks.

Shiny Drop Sources

All weapon drops are equal - wave 10 boss, opening the chest w/ 10 tokens, claiming a vendor engram from Shaxx, or turning in a weekly Arcite bounty - Every drop has a chance to be shiny.
finally, if you only take one piece of information away -

Heavy Ammo Crates in Onslaught only despawn when a build phase ends, or a player enters the boss portal.

NOT the Spark / Round 6 Portal. If you have a heavy crate sitting on the ground when the Spark/Rift portal opens, Don't open it, leave it outside, burn your heavy clearing the Sparkbearers inside, then refill after.

Severe Weather Alert Service
Updated Forecast
Your Updated Forecast
BREAKING NEWS
Active weather pattern expected from the Midwest to the Southern Plains
Forecast Today & 7 Day
National Severe Weather Map
Breaking News
After prosecution rests, Trump lawyers ask judge to dismiss hush money trial
Internet search engine users share tips to disable artificial intelligence chatbots
US regulator to probe VinFast crash in California in which four died
Schumer says US Senate will try again to pass border bill
US FDA tested retail milk samples for bird flu in 17 states
Genetic profile may predict best response to weight-loss drug Wegovy
US race-neutral lung assessments to have profound effects, study finds
Food Safety Alert: E. coli Outbreak
Crabmeat recall, potential foodborne illness - Listeria monocytogenes
Noodles recalled
Helpful Links
National Severe Weather Map
Threat and Terrorism News

Rockies (15-31) @ Athletics (19-30)

First Pitch: 7:40 PM at Oakland Coliseum

Team	Starter	TV	Radio
Rockies	Cal Quantrill (3-3, 3.66 ERA)	COLR	KOA
Athletics	Aaron Brooks (0-1, 3.86 ERA)	NBCSCA	A's Cast, KIQI (ES)

MLB	Fangraphs	Baseball Savant	Reddit Stream	IRC Chat
Gameday	Game Graph	Strikezone Map	Live Comments	Libera: ##baseball

Line Score - Pre-Game

	1	2	3	4	5	6	7	8	9	R	H	E	LOB
COL										0	0	0	-
OAK										0	0	0	-

Box Score

Posted at 4:40 PM. Updates start at game time.
Remember to sort by new to keep up!

Hey everyone! Seeking 1 to 3 more players for a character driven campaign. I have 2 friends playing already but would love more to join. Check the info below.
The Order: Monsters Among Us
Adventure Link: https://startplaying.games/adventure/clw7f7tf0002hlmibhcytzllk
Session Time: Tuesdays 7 pm ET
System: D&D 5E
Platforms Used: Foundry VTT
Campaign
Cost: $30
Pitch: Home-brew Campaign set in the world of Elloria on the human-centric continent of Arisia. The Order: Monsters Among Us is set up around the guild of monster hunters who have found that there has been a influx of monsters on the continent and must use their investigation skills to track down monsters and bring them down.
Expect:

• Challenging custom and classic monsters
• Linear story through given goals that leads to character focus storytelling with proactive story elements
• This adventure walks the line between optimistic heroism and the characters' morally ambiguous dilemmas
• Access to Wiki for notes and tracking character goals

Game Primer: https://www.localdungeonmaster.com/the-order-monsters-among-us-prime

GamerLegion 🇪🇺 2-1 🇷🇺 AMKAL

Ancient: 19-17 Vertigo: 9-13 Nuke: 13-11
 
 

Map picks:

GamerLegion	MAP	AMKAL
X	Inferno
	Anubis	X
✔	Ancient
	Vertigo	✔
X	Dust2
	Mirage	X
	Nuke

 

Full Match Stats:

Team	K-D	ADR	KAST	Rating
🇪🇺 GamerLegion
🇷🇺 FL4MUS	69-60	88.7	72.0%	1.24
🇭🇺 sl3nd	53-54	75.4	69.5%	1.03
🇵🇱 Snax	49-52	72.1	69.5%	1.00
🇪🇪 aNdu	51-58	73.5	67.1%	1.00
🇷🇴 volt	48-49	62.8	76.8%	0.98
🇷🇺 AMKAL
🇷🇺 TRAVIS	66-60	91.7	72.0%	1.23
🇰🇿 ICY	54-50	67.1	74.4%	1.06
🇺🇦 Sdaim	49-54	67.0	68.3%	1.01
🇷🇺 Krad	55-56	74.2	64.6%	0.99
🇷🇺 Forester	46-51	64.3	67.1%	0.95

 

Individual Map Stats:

Map 1: Ancient

Team	T	CT	OT	Total
🇪🇺 GamerLegion	4	8	7	19
	CT	T	OT
🇷🇺 AMKAL	8	4	5	17

 

Team	K-D	ADR	KAST	Rating
🇪🇺 GamerLegion
🇪🇪 aNdu	28-25	91.1	72.2%	1.19
🇵🇱 Snax	25-22	78.1	72.2%	1.14
🇭🇺 sl3nd	25-24	72.6	69.4%	0.99
🇷🇴 volt	20-21	52.5	75.0%	0.92
🇷🇺 FL4MUS	19-29	71.8	58.3%	0.87
🇷🇺 AMKAL
🇷🇺 TRAVIS	28-27	91.0	77.8%	1.21
🇷🇺 Krad	29-23	90.6	69.4%	1.20
🇰🇿 ICY	21-21	61.1	77.8%	1.04
🇷🇺 Forester	22-22	64.4	69.4%	0.98
🇺🇦 Sdaim	18-24	63.0	63.9%	0.92

Ancient detailed stats and VOD

 

Map 2: Vertigo

Team	T	CT	Total
🇪🇺 GamerLegion	3	6	9
	CT	T
🇷🇺 AMKAL	9	4	13

 

Team	K-D	ADR	KAST	Rating
🇪🇺 GamerLegion
🇷🇺 FL4MUS	22-16	82.5	81.8%	1.39
🇷🇴 volt	13-17	74.9	72.7%	0.92
🇭🇺 sl3nd	12-17	68.4	63.6%	0.85
🇪🇪 aNdu	10-17	61.8	50.0%	0.72
🇵🇱 Snax	8-16	50.1	54.5%	0.62
🇷🇺 AMKAL
🇷🇺 TRAVIS	23-15	121.8	72.7%	1.63
🇺🇦 Sdaim	15-11	75.4	86.4%	1.31
🇷🇺 Forester	15-11	73.6	77.3%	1.24
🇰🇿 ICY	16-15	74.7	77.3%	1.07
🇷🇺 Krad	14-13	58.5	77.3%	0.99

Vertigo detailed stats and VOD

 

Map 3: Nuke

Team	T	CT	Total
🇪🇺 GamerLegion	5	8	13
	CT	T
🇷🇺 AMKAL	7	4	11

 

Team	K-D	ADR	KAST	Rating
🇪🇺 GamerLegion
🇷🇺 FL4MUS	28-15	119.6	83.3%	1.74
🇭🇺 sl3nd	16-13	86.0	75.0%	1.26
🇷🇴 volt	15-11	67.2	83.3%	1.17
🇵🇱 Snax	16-14	83.3	79.2%	1.17
🇪🇪 aNdu	13-16	57.9	75.0%	1.00
🇷🇺 AMKAL
🇰🇿 ICY	17-14	69.3	66.7%	1.09
🇷🇺 TRAVIS	15-18	65.0	62.5%	0.93
🇺🇦 Sdaim	16-19	65.4	58.3%	0.88
🇷🇺 Krad	12-20	64.0	45.8%	0.71
🇷🇺 Forester	9-18	55.7	54.2%	0.66

Nuke detailed stats and VOD

 
This thread was created by the Post-Match Team. If you want to share any feedback or have any concerns, please message u/CS2_PostMatchThreads.

I tested models and code in https://github.com/deepcam-cn/FaceQuality
I converted model to onnx :

import torch import onnx from models.model_resnet import ResNet, FaceQuality import os import argparse parser = argparse.ArgumentParser(description='PyTorch Face Quality test') parser.add_argument('--backbone', default='face_quality_model/backbone.pth', type=str, metavar='PATH', help='path to backbone model') parser.add_argument('--quality', default='face_quality_model/quality.pth', type=str, metavar='PATH', help='path to quality model') parser.add_argument('--database', default='/Users/tulpaDownloads/_FoundPersons.db', type=str, metavar='PATH', help='path to SQLite database') parser.add_argument('--cpu', dest='cpu', action='store_true', help='evaluate model on cpu') parser.add_argument('--gpu', default=0, type=int, help='index of gpu to run') def load_state_dict(model, state_dict): all_keys = {k for k in state_dict.keys()} for k in all_keys: if k.startswith('module.'): state_dict[k[7:]] = state_dict.pop(k) model_dict = model.state_dict() pretrained_dict = {k: v for k, v in state_dict.items() if k in model_dict and v.size() == model_dict[k].size()} if len(pretrained_dict) == len(model_dict): print("all params loaded") else: not_loaded_keys = {k for k in pretrained_dict.keys() if k not in model_dict.keys()} print("not loaded keys:", not_loaded_keys) model_dict.update(pretrained_dict) model.load_state_dict(model_dict) args = parser.parse_args() # Load the PyTorch models BACKBONE = ResNet(num_layers=100, feature_dim=512) QUALITY = FaceQuality(512 * 7 * 7) if os.path.isfile(args.backbone): print("Loading Backbone Checkpoint '{}'".format(args.backbone)) checkpoint = torch.load(args.backbone, map_location='cpu') load_state_dict(BACKBONE, checkpoint) if os.path.isfile(args.quality): print("Loading Quality Checkpoint '{}'".format(args.quality)) checkpoint = torch.load(args.quality, map_location='cpu') load_state_dict(QUALITY, checkpoint) # Set the models to evaluation mode BACKBONE.eval() QUALITY.eval() # Create a dummy input with the correct shape expected by the model (assuming 3 channels, 112x112 image) dummy_input = torch.randn(1, 3, 112, 112) # Adjust channels and dimensions if your model expects differently # Convert the PyTorch models to ONNX torch.onnx.export(BACKBONE, dummy_input, 'backbone.onnx', opset_version=11) # Specify opset version if needed torch.onnx.export(QUALITY, torch.randn(1, 512 * 7 * 7), 'quality.onnx', opset_version=11) print("Converted models to ONNX successfully!") But the inference code for onnx giving error : how to convert correctly and make the inference /Users/tulpaProjects/FaceQuality/onnxFaceQualityCalcFoundDb.py Traceback (most recent call last): File "/Users/tulpaProjects/FaceQuality/onnxFaceQualityCalcFoundDb.py", line 74, in  main(parser.parse_args()) File "/Users/tulpaProjects/FaceQuality/onnxFaceQualityCalcFoundDb.py", line 64, in main face_quality = get_face_quality(args.backbone, args.quality, DEVICE, left_image) File "/Users/tulpaProjects/FaceQuality/onnxFaceQualityCalcFoundDb.py", line 35, in get_face_quality quality_output = quality_session.run(None, {'input.1': backbone_output[0].reshape(1, -1)}) File "/Users/tulpaProjects/venv/lib/python3.8/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 192, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Got invalid dimensions for input: input.1 for the following indices index: 1 Got: 512 Expected: 25088 Please fix either the inputs or the model. Process finished with exit code 1 

Debit/Credit Cards Accepted No Closing Costs Cash Price: $13,500 Finance with $750 Down and 72 Payments of $225 Per Month No Credit Check, No Income Documentation, No Prepayment Penalty 

Hey everyone! Seeking 1 to 3 more players for a character driven campaign. I have 2 friends playing already but would love more to join. Check the info below.
The Order: Monsters Among Us
Adventure Link: https://startplaying.games/adventure/clw7f7tf0002hlmibhcytzllk
Session Time: Tuesdays 7 pm ET
System: D&D 5E
Platforms Used: Foundry VTT
Campaign
Cost: $30
Pitch: Home-brew Campaign set in the world of Elloria on the human-centric continent of Arisia. The Order: Monsters Among Us is set up around the guild of monster hunters who have found that there has been a influx of monsters on the continent and must use their investigation skills to track down monsters and bring them down.
Expect:

• Challenging custom and classic monsters
• Linear story through given goals that leads to character focus storytelling with proactive story elements
• This adventure walks the line between optimistic heroism and the characters' morally ambiguous dilemmas
• Access to Wiki for notes and tracking character goals

Game Primer: https://www.localdungeonmaster.com/the-order-monsters-among-us-prime

It's another favored terrain post, yay!
Joking aside, the PC warforged ranger was created and grew up in the same few square miles of forest near Waterdeep and has never left before. They took forest and swamp as their favored terrain. I've asked them if they'd like to learn the jungle terrain as it is somewhat unfamiliar with them, but pretty similar to forest. I've made a rule set and reminder sheet for this and I'd like some feedback before I run it by the player. Let me know if you spot any major errors or have suggestions. I'm planning to type it up in homebrewery, so let me know if you'd like a link after I implement any changes.
I know the suggestion is that I should just give the Ranger Jungle terrain, but the party as a whole has expressed interest in resource tracking, hex crawl shenanigans, etc. I've given them a summary document for Chult Hex Crawl rules with information for them to reference for these rules.
Jungle Favored Terrain
The DC: Terrain similar to your favored terrain can be learned with a series of successful navigation checks. The DC for the initial check will be high, but will be reduced by one or two for each successful check depending on how much the check is exceeded by. This means the DC should decrease faster as additional successful checks are made. The DC will never increase.
Navigation Checks and Lost
A Navigator must be designated at the beginning of each travel day. At the start of travel, the Navigator makes a survival check. The guide is likely used to navigating. The DC will not decrease unless the Ranger makes the check. The base DC for the navigation check varies based on terrain type.
Check modifiers:

Condition	Modifier
Lightly obscured	+5
Heavily obscured	+10
Slow travel	-2

If the Navigator fails the check by 5 or less, the party will become slightly lost. The DM rolls 1d12 and consults the tables below. All new directions are relative to the intended direction around the side of the hex. Players might still travel in the correct direction when lost.
Slightly Lost:

Result	Direction
1-5	1 hex clockwise
6-7	Intended direction
8-12	1 hex counterclockwise

If the Navigator fails the check by more than 5, the party will become significantly lost.
Significantly Lost:

Result	Direction
1-2	2 hex clockwise
3-5	1 hex clockwise
6	Opposite direction
7	Intended direction
8-10	1 hex counterclockwise
11-12	2 hex counterclockwise

When lost, the DM will conceal the current hex until a successful check is made. The DM will then reveal the current hex, but not the hexes traveled through.
Travel Pace Reminders

Pace	per day	per hour	Hex
Fast	15 miles	1.5 miles	1.5
Normal	10 miles	1 mile	1
Slow	5 mile	0.5 miles	0.5

Fast: -5 to passive perception scores, cannot forage
Slow: Can move stealthily, +5 to navigation in previously mapped hexes. If a slow pace is set, there is a chance you may advance 1 fewer hex per day (1 by canoe, none by foot).
Characters travel 2 hexes per day by canoes on a river. The rate of travel up or down the river is the same. Canoes move 1 hex per day in swamp terrain
Favored Terrain and Hex Crawl
While traveling for more than an hour in your favored terrain:
The party will be able to maintain the canoe 2 hex travel pace in swamps if the Ranger is navigating
Navigation checks are not needed for the day if the Ranger is navigating in their favored terrain. Regardless of not being able to get lost, you do not know where you are going as many locations are unknown to you.
Rangers will contribute to group's ability to notice threats (i.e. perception) during foraging, navigating, or tracking.
Cannot navigate and forage at the same time.

Playing it for a few days, it's a solid game. The gameplay reminds me of monster hunter and god of war, with loot system.
Combat is fun, but hard to master.
Music is pretty good too.
Voice acting i find pretty good too (idk about others)
Graphics are very good
Nice colors and enviroments
Only issue i have is they could have expanded the lore or explored it, the world building and story too, this games story, world building and lore had so much potential.
Also i dont mind fighting in the same map, monster hunter does it well too.
The game is a 7/10 for me
I recommend it as it has gotten really cheap. But would not buy it for full price. Good game 👍
Also what are your fav weapons, mine are:

1: Longsword

2: Greatsword

3: Dual blades

Recently downloaded a fun zombie modpack, and along with it I downloaded a HUGE city map for me and my friends to live out our zombie apocalypse fantasies.
There are chests spread around the entire city and outskirts for loot. The problem is, it's way too much to do manually.
I checked the map's webpage, and apparently there are almost 7,000 chests spread across the entire city. I spent a good couple hours filling chests with 3-4 random items before I started going crazy.
Does anyone know of a way to fill that many chests with random loot quickly and painlessly?

Twins (24-23) @ Nationals (21-25)

First Pitch: 5:45 PM at Nationals Park

Team	Starter	TV	Radio
Twins	Joe Ryan (2-3, 3.57 ERA)	BSNO	TIBN, Twins(Sp) (ES)
Nationals	Patrick Corbin (1-4, 5.59 ERA)	MASN	106.7, DC 87.7 FM (ES)

MLB	Fangraphs	Baseball Savant	Reddit Stream	IRC Chat
Gameday	Game Graph	Strikezone Map	Live Comments	Libera: ##baseball

Line Score - Middle of the 1st

	1	2	3	4	5	6	7	8	9	R	H	E	LOB
MIN	0									0	1	0	1
WSH										0	0	0	-

Box Score

WSH		AB	R	H	RBI	BB	SO	BA
SS	Abrams	0	0	0	0	0	0	.267
RF	Rosario, E	0	0	0	0	0	0	.190
LF	Winker	0	0	0	0	0	0	.232
DH	Meneses	0	0	0	0	0	0	.225
2B	García Jr., L	0	0	0	0	0	0	.281
3B	Senzel	0	0	0	0	0	0	.202
1B	Gallo	0	0	0	0	0	0	.136
C	Ruiz, K	0	0	0	0	0	0	.165
CF	Young	0	0	0	0	0	0	.270

WSH	IP	H	R	ER	BB	SO	P-S	ERA
Corbin	1.0	1	0	0	0	0	18-9	5.47

MIN		AB	R	H	RBI	BB	SO	BA
RF	Margot	1	0	0	0	0	0	.196
DH	Jeffers	1	0	0	0	0	0	.261
SS	Correa	1	0	1	0	0	0	.274
3B	Miranda	1	0	0	0	0	0	.272
CF	Buxton	0	0	0	0	0	0	.250
LF	Castro, W	0	0	0	0	0	0	.265
2B	Farmer, K	0	0	0	0	0	0	.165
1B	Santana	0	0	0	0	0	0	.201
C	Vázquez	0	0	0	0	0	0	.156

MIN	IP	H	R	ER	BB	SO	P-S	ERA
Ryan	0.0	0	0	0	0	0	-0	3.57

Highlights

Description	Length
Patrick Corbin against the Twins	0:07
Bullpen availability for Washington, May 21 vs Twins	0:07
Bullpen availability for Minnesota, May 21 vs Nationals	0:07
Bench availability for Minnesota, May 21 vs Nationals	0:07
Bench availability for Washington, May 21 vs Twins	0:07
Fielding alignment for Minnesota, May 21 vs Nationals	0:11
Fielding alignment for Washington, May 21 vs Twins	0:11
Starting lineups for Twins at Nationals - May 21, 2024	0:09
Carlos Correa triples (1) on a sharp line drive to right fielder Eddie Rosario.	0:30
Jose Miranda pops out to first baseman Joey Gallo in foul territory.	0:13

Updated at 5:53 PM.
Remember to sort by new to keep up!

It is good, just not good enough

I get what the developers are going for and commend them for their hard work and novel idea. However, there is not much meat on this bone. For the 25 USD price point the game has not met my expectations. Combat is lacking, some mechanics are straight up bad and should be removed, and the game can soft lock you repeatedly during random encounters

The 'Wordle' part

The setup of the game is simple. You type words to execute an action or attack. You find new actions or attacks by revealing words one at the time per character. The way the game helps you discover the words is through combat, which reveals 1 letter you choose, or through riddles, where the letters in the answer will apply to all of the characters speeding up the discovery proces.

Combat

The combat is based on the player typing these discovered words.

• One of the main gripes I have with combat is the cooldown mechanic. It completely removes the player's typing speed from the combat skill equation: How fast you can type only matters at the initial attacks, after that you just line up the words and hit enter after the character's cooldown expires.
• There is also not a lot of variety in attacks between characters
• Some words (like salmon or urchins) do the same thing on the same character.
• The overview of all the unlocked 'words' (i.e. attacks) mixes the important attack words with other useless 'memory' words that have no use throughout the game
• Enemies have a shield that can soft lock you. You unlock crack early on in the game for Jorro. If the shield enemies have armor you need to crack before being able to do damage AND a shield that states any of the letters that are used in the word to counter it, you can't progress and just have to wait for your party to die.
• The armor mechanic also makes heavy hitting attacks much less useful, as you have to remove the armor before being able to deal more damage. So knowing puppet, rattle or cackle (which are all, again, stacked on the same character) often does little

Exploring

The exploring gets old quickly. All areas start to look the same. Rocks, grass, a wooden door and rampart. It never really changes. The black and white visuals do little to redeem this and perhaps even add to the repetitive look of the dungeon. The rooms you will explore can be quite big. As a dungeon crawler it makes sense you want to explore these rooms, but the game does nothing to reward you for it. You can waste a lot of time navigating all the nooks and crannies of the map, expecting perhaps a chest or puzzle, but instead be rewarded with more black and white walls.

Puzzles and Riddles

The riddles are great and are a nice addition to the game. The mental switch from 'Wordle Dungeon Crawler' to solving riddles is fun and feels good. Not all riddles are equal though, some are unoriginal and you've probably heard before many times whereas others are quite clever and new. The one thing they completely messed up is that not all riddles have the correct answer.
Early on the game acknowledges this in the tombs. The riddle "What has an eye but cannot see" can be given multiple answers and the NPC will reply with "I did not expect that particular answer but it is also correct so here is your prize". You can answer blind, needle or storm and all will be true and counted as such.
Later in the game this completely goes away. For example the riddle "What do you have but never use, and others use all the time" requires the answer "name", but a) you do use your name to introduce yourself and b) In that case something akin to 'your reputation' should also fit. Or Your phone number. There are many riddles that can be given multiple answers to in this game and I was hoping, due to the core mechanic being trial and error inputting words, the game would accommodate this throughout the game.
This makes solving riddles much more frustrating and less fun. It makes it feel abitrary. It highlights the fact that the 'dungeon master' that you interact with has much more limited interactions than you'd expect at first.
The game also has a big difficulty spike at Ulara castle when it comes to riddles. There is one where you straight up need create an algebraic function and solve it. It makes the pacing feel weird and the new area feel like a brick wall which I am sure will discourage a lot of players. You can have the narrator literally give you the answer after 1 or 2 tries, but for a game about solving riddles, where is the fun in that.

The story

You are probably not buying this for the story but rather for the gameplay mechanics described above. If you mainly want a good dungeon crawler story, this is not the game for you. The characters in this story are weak and interchangeable. Your party of heroes consist of mutes. The game tells you to talk to NPCs and interact with them using action keywords like 'talk', 'use' or anything contextual. But 90% of the time you can't talk to the NPCs and the dungeon master gives you the same reply again and again. 'You will talk when I will tell you to' as you roll your eyes and move towards the next door on your map. The fun and interesting parts in the story are few and far between.

Conclusion

The game has an insane amount of potential. It just needs a lot more polish and more thought into the design elements of the game. The gameplay loop could use a review. The combat needs a complete overhaul that caters more towards typing fast while remember the words you unlocked, removing the cooldowns or making them much short will be a great start. Currently the game to me is a 4 or 5 out of 10 because of the reasons mentioned above. With some more TLC it can be a 7 or 8.