NoMaDhOoK
-
Posts
72 -
Joined
-
Last visited
Content Type
Profiles
Forums
Gallery
Store
Downloads
Blogs
Events
Posts posted by NoMaDhOoK
-
-
On 2/21/2020 at 4:38 PM, Boar said:
Ha, I didn`t get initailly what you mean here, but I think note on github cleared it up for me.
So in essence sometimes v.low probability maximal damage would not be shown, right? Something like fleetmaster with venomfang blade who, at least in theory could cause 24 mortal wounds, but since probability is low it never shows up because you would need not thousands, but often milions or more simulations?
Yes, I probably didn't explain it very well in the release notes (it's a little hard to summarise in a bullet point or two). I'll take a little time this weekend to take some before and after screenshots, with detailed explanations, and post it here and on the subreddit as a means to explain why this seemingly small change actually makes quite a large impact on the quality of the results (this was a 2.0.0 release after all).
On 2/21/2020 at 4:38 PM, Boar said:Also would it be possible to add export feature (to csv) to some of derivative stats? For example from window with average damage per save. So you could export something like table below. I mean, you obviously can copy and paste like I just did but maybe it could be useful. Though I have to admit, as I am typing this I am starting to doubt usefulness of such feature.
I do actually like this idea. I was already planning on overhauling the PDF export page to give users the ability to customise exactly what gets put in it. But now I think it would be better to make a more generic "Export" page, which lets you pick between CSV, or PDF (or other formats in the future) and customise what you want exported (with CSV obviously having more restrictions)
Edit: Here is the Github issue for the new Export screen planned in for v2.2.0 (with Google Drive Integration)
-
1
-
-
damonhook released this 11 minutes ago
Additions
- Users can now toggle profile and target modifiers using a toggle switch (fixes #26).
UI
- Remove Beta Tag from non beta features
- Small UI overhaul for Desktop (Mostly related to the drawer)
- Add Warcry Statshammer links and buttons (fixes: #23)
-
Change SVG Text Object -> Path (fixes #25)
- Use SVGIcon instead of loading through img src
-
Make the Left navigation bar responsive based on the following criteria (fixes #34)
- Large: Show the full drawer
- Medium: Show a Rail but let users open the drawer
- Small: Hide the rail, let users open the drawer
-
Add an Average damage graph to the
Metricstab in theSimulationsscreen for easier reference
API
-
Use a calculated population max (The actual max) rather than basing it on the max value of the simulated sample (fixes: #30)
- This has had the effect of vastly improving the accuracy of the simulated metrics.
- Also use the population mean, rather than the sample mean
- Remove Median as the way the data is distributed, it is usually extremely close to mean anyway
- Calculate frequency table on the fly, rather than creating a large results array first (fixes #32)
-
Completely refactor the response structure of the
/api/simulateand/api/simulate/saveendpoints for better visibility and extensibility. - Added Sentry to API to capture errors (Fixes: #39)
Notes
- This UI overhaul was more focused on navigation, the next one will be more about layout
- Though it may not look it, this release had a huge amount of behind the scenes changes
- Since the 2.0.0 branch, I have been using Github projects to greater effect, this means you can watch the project for a specific version to see the currently planned items, and their progress
-
3
-
Feature Additions
-
Add an import screen that allows users to select a number of fighters to import (fixes #4)
- For now it is just the Season 1 Warbands (Going to add more very soon: tasks)
- NOTE: The preset fighters only contain the weapons. no runemarks, art, wounds, toughness, movement, or points will be added. This is not a replacement for cards/ToC
-
Allow users to customise the toughness range used for results (fixes #14) @GuitaRasmus
- Configure the min/max toughness to generate data for. Leave as Auto to let the API decide
-
How the Auto range is determined:
-
(MIN STR - 1) <---> (MAX STR + 1)
-
UI
-
Added Snackbar notifications (fixes: #7)
-
Deleting Fighter
-
Undo: will re-add the fighter at the deleted position
-
-
Deleting Profile
-
Undo: will re-add the Profile at the deleted position for said fighter
-
-
Error fetching stats
-
Retrywill re-attempt the API request
-
- Importing fighter success
-
Deleting Fighter
- Added AoS Statshammer as a Social Button (Footer and About page)
-
Change SVG Text Object -> Path (fixes #16)
- Use SVGIcon instead of loading through img src
-
1
-
Add an import screen that allows users to select a number of fighters to import (fixes #4)
-
Double Post, sorry
-
1 hour ago, GuitaRasmus said:
Thanks for the reply. However, thinking about it some more, it would still be great to have the ability to compare two different stats against the same toughness - that would be my primary purpose for the tool. I wanted to compare a Slaughterprist with axe vs. the one with Hackblade, and at first they seemed very similar, because I didn't notice the damage was vs. T5 vs. T4.
👍I have the configurable toughness ranges planned in for the next release (0.3.0)
For the moment, if you put both fighters in, it will compare them at equal toughness levels.
It shouldn't be too long and I'll be sure to tag you as soon as the feature is released.
-
UI
- The Left Navigation Bar on Desktop has been revamped and will no longer scroll with the content.
API
-
Instead of generating data per toughness, instead build
lt,eq,gtproperties. Then map that to the different toughness values (Fixes: #13)- This should have no difference in the final result, however, should improve performance greatly
- If you try and generate data for attacks > 8, switch to a simulation of dice rolls rather than generating every permutation (Fixes: #13)
- Increased max attacks from 8 -> 15
- Added Sentry to API to capture errors (Fixes: #11)
Notes
Here is a table of the number of permutations as the attacks grow:
Attacks Permutations 2 36 4 1 296 6 46 656 8 1 679 616 10 60 466 176 12 2 176 782 336 Because of this it becomes too expensive to get population permutations for attacks > 8. Hence why it switches over to doing
1 500 000simulations instead. It should be accurate enough as the probabilities are limited to 2 dec. places on the UI anyway -
1 hour ago, GuitaRasmus said:
Hahaha, thanks - boy, do I feel stupid for not realising that. 🤣 Again, thanks for a great tool - it's very helpful.
It's no problem 👍, if anything it highlights that I should maybe add a little info dialog to the stats screen explaining how the T-range is determined (similar to how AoS Statshammer has an info dialog for how its simulations work).
I can see that it may be confusing as to why the T-range changes depending on the fighters you have (as nowhere do I state how it's done).
-
9 hours ago, GuitaRasmus said:
I just found this, and it's a very cool tool - thank you for putting this up!
A small question though - it would be cool to be able to see all the stats vs. varying toughness values - for example, when putting in a str. 6 weapon, I can only see the stats vs. Toughness 5 and up - it would be cool to see it vs. T 4 and below.Thank you for the feedback.
Some info on how the T-range is decided: Since Warcry only cares about Strength <, =, or > Toughness. A S6 weapon will do the same damage to a T4 as it does to a T5. So the way the API decides on the toughness ranges is: lowest STR - 1 <----> highest STR + 1.
Solution: However, what I could do is provide an option for the user to provide the toughness ranges they want or use the auto generated ones. Quick mockup here:
I've added this as a feature request on Github: https://github.com/damonhook/warcry-statshammer/issues/14
-
Warcry Statshammer
A new sister project (Warcry Statshammer) has been released
URL: https://warcry-statshammer.herokuapp.com/
TGA Topic:
NOTE: This does not mean that development has stopped for AoS Statshammer. AoS Statshammer will remain my main project, but Warcry Statshammer allows me to experiment and test new features in a much simpler environment. I can then take those learnings over to AoS Statshammer
-
1
-
1
-
-
Warcry Statshammer
A tool for calculating and comparing damage outputs for Warhammer Warcry Fighters (also referred to as Mathhammer).
The production deploy of the tool is located at: https://warcry-statshammer.herokuapp.com/
Features
-
Average Damage Stats
- Gather the average damage that each unit is expected to do against various toughness values.
- This average is a calculated value (not simulated)
-
Probabilities
- Generates Discrete and Cumulative Probabilities
-
These values are calculated (not simulated)
- This is done by gathering every permutation of the dice roll(s) and uses that to get a population probability graph
- Dark Theme
- Persisitent data between refreshes
-
And more
- Keep an eye on the Github Releases, or the SubReddit to get notified of new additions
Roadmap Features
- Choose from a preset list of fighters
- Export PDF
-
Google Drive integration
- Save / Load Fighters
Related Projects
-
AoS Statshammer
- Provides similar functionality to this, but for Warhammer Age of Sigmar
-
5
-
2
-
Average Damage Stats
-
Note:
I have upgraded the AoS Statshammer server to a paid tier, this should hopefully improve the response time of the simulations, and over the weekend I am going to do decide on a higher max limit for the customisable simulations slider.
Don't worry, the paid tier is still extremely cheap ($7 pm) so I have no intention of looking at monitisisation plans.
As for why I have decided to upgrade the server, I needed the free tier server for a new (sister) project that will be launching very very soon. I will obviously still continue to work on AoS Statshammer, and this will remain my primary project (I still have a lot of requested features to get to).
Here is a quick on-the-nose, terribly cheesy, totally unnecessary teaser for this new project:
-
2
-
-
15 hours ago, Arcian said:
I saw somewhere on their facebook (i'll try and track a screenshot down later, i'm at work) that OBR during the first round of the siege get +1 to their saves.
I'm sure you can imagine the salt that flowed through the comments.Yup that's their "atacker" siege trait. While strong, I am kinda meh on it. You are just delaying combat by 1 round (because nobody in their right mind will get into combat t1 with 2+ save mortek).
It will give you a bit of a buffer against T1 shooting, which is really handy, and may be really good depending how the game mode plays (gives you time to move up and set up)
For me I think I'd have the most fun with SL. Some tasty cav charges with +1 save on first round sounds fun.
I'm not sure on the other general siege rules though, so will have to wait and see the impact it will have.
-
1
-
2
-
-
Additions
-
You can now customise how many simulations (per save) are performed
- A higher number of simulations will give you more accurate results, however, will increase the load times
-
Added a lot more detail to the
Aboutpage
UI
-
Added
Releasesto the Footer and Drawer - Added the first draft of the new Logo (time to get off the react logo)
New Project: coming-soon
-
3
-
You can now customise how many simulations (per save) are performed
-
15 minutes ago, Scurvydog said:
I also find it a shame nothing influences battleline options in the book, like other books such as KO or Ogors, where either a specific general or subfaction changes battleline options. Ivory host would if nothing else create some different lists if it could make stalkers, immortis and/or morghasts battleline.
Yeah, it would also help to set petrifex apart. Change the +1 save to wound chars of 4+ (as was mentioned earlier), then make Stalkers battleline in PE. This would solidify PE as the "Elite" Subfaction.
Maybe Immortis Guard BL in MP, as the subfaction focuses around heroes, so making the hero shield battleline fits quite well thematically.
-
11 minutes ago, Scurvydog said:
This is not true though as the rage is only in effect in the combat phase:
Ahh you are right, I missed that (going to edit my comment now). Yeah I agree that I hate the way rage is "activated" it is completely out of the OBR players control (which is why I changed it up in my suggestion)
-
My idea for Ivory Host (along with descriptions as to why). This is a pipe-dream though because something this heavy would never come in an FAQ:
- Change +1 to hit -> + 1 to wound (we already have too many sources for + to hit)
-
Change the entire allegiance to be something along the lines of:
-
At the start of the combat phase, choose whether your army is calm or enraged, if they are calm nothing happens, if they are enraged, get + 1 to wound but -1 to save (maybe even +1 to hist and +1 to wound, you'll see why below).
- This also means that you are only weaker in the combat phase, rather than making you more vulnerable in all phases, when you get no benefit
- This also means you get to control when to get he bonus, rather than relying on getting a unit injured
-
Change the CA from remove the -1 save detriment to "give +1 to attacks characteristic to an enraged" unit
- Allowing you to circumvent the -1 save penalty means that the bonus can never be made to be worth it, otherwise when you use the CA, the unit becomes too powerful. This removing the circumvent CA means that we can get a stronger benefit
- Since you also get more control over enraged vs calm, you don't need the CA (though you get no benefit from being calm)
-
At the start of the combat phase, choose whether your army is calm or enraged, if they are calm nothing happens, if they are enraged, get + 1 to wound but -1 to save (maybe even +1 to hist and +1 to wound, you'll see why below).
-
2
-
On 1/28/2020 at 8:58 AM, Saxon said:
I've seen nothing but the opposite of this statement!!!!
From what i have seen people are getting very tired of specific PE lists and want them to get a points increase so that they see something different.
Yes, people want point increases because of PE lists, however, many units are overcosted in other legions (not Mortek of course, they could probably go up to 140 and still be fine). So in order to decrease the other unit costs in other legions, PE needs to be tweaked so that it does not make PE any stronger (Nerf PE, then look at points afterwards). Again obviously not talking about Mortek guard point decrease, that would be silly
On 1/28/2020 at 10:32 AM, Phasteon said:Praetorians - generate RDP on 5+ instead of -1 bravery
Petrifex - reduce rend by 1 instead of +1 save
Ivory Host - +1 to wound instead of +1 to hit.
- I sort of prefer +1 save to units with a wound characteristic of 3+ (so no mortek). This solidifies PE as the Elite subfaction
- Ivory host needs a rework,
OBR is already weak against shooting, and -1 save makes that even worse.Considering how low of a wound count OBR has, becoming fragile is really dangerous. My idea for Ivory host is too heavy for an FAQ though. (I'll post a comment below). There is just no case where I think +1 to hit is worth the -1 save -
2 hours ago, firtahl said:
How'd it go? Looks like the doom and gloom about OBR wasn't reflected in the top 8 with just 1 OBR player. I'm curious if you saw a trend of OBR players loosing to specific strategies, battle plans, or if every army was geared in one way or another with OBR in mind.
Honestly I think it's a bit of all of the above. The long text below is my opinion, I have no where near the experience needed to say I know 100% the answer
Due to the fact that, while powereful, OBR is quite an honest army. It is strong through sheer staying power, rather than through shenanigans like teleporting, fight first/last, fight twice, etc. IMO this leads to the following:
- You can build against them without hurting your list when facing other armies (it never hurts to have rend/MW).
- They seem like a bit of a "gatekeeper army". If you don't know what you're doing, or you don't equip yourself properly, they will crush you with very little effort. But if you have the tools, they're quite manageable (almost like gotrek in a way)
- Their low model count (and low total wound count), mixed with a medium speed means there are some scenarios that they definitely struggle with. (the mid speed is probably why the top list was quite cav heavy, and OBR cav is very fast)
Petrifex still needs a tweak, even just for the fact that some of OBR units need a little points decrease for other legions (not mortek), but you can't decrease their points with PE in its current state without risking tipping it over the edge.
People love extremes, either something is super underpowered, or completely unbeatable. When in reality it's somewhere in between.
-
3
-
-
I don't mind too much. And I can definitely understand that while it limits our freedom, it gives GW more freedom to create more unique subfactions that may not work / be broken if we could mix.
Maybe somewhere in between could work. Where each subfaction could have 3 traits and 3 Artifacts to choose from (that differs between subfactions) . So still somewhat restrictive with a little customization
-
Note: I had some issues with deploying 1.0.0 that meant I had to put the site into maintenance mode for a few minutes.This is resolved and was due to a 3rd party library
Fixes
-
Crash on load due to 3rd party library
-
A library used
react-swipeable-viewsreleased an update (patch version) so bad that it caused anyone using said library to crash on launch... - I have pinned the library for now, and I am going to look into alternatives
-
A library used
-
Crash on load due to 3rd party library
-
Additions
-
Added
Undoaction to delete Unit / Profile -
Added
Rerunbutton toSimulationsscreen
UI
-
Changed primary colour in the
light themeto better match thedark themeone -
Renamed
Advanced Stats->Simulationsto better reflect that they are sample probabilities -
Added Return button and # simulations to
SimulationsScreen -
Added an info dialog to the
Simulationsscreen explaining the 3 tabs -
Renamed
Singletab ->Discrete -
Renamed
Tablestab ->Metrics -
Added
Infopage that gives a small brief on the tools goal -
Added
StatsandInfoto the Bottom Navigation on Apps - Added AppBar to PDF page
-
Lighten info notification in
dark modefor better contrast
Fixes
-
Fix table scrolling in
Simulationsscreen when on mobile
Architecture
-
Moved the entire codebase from
Javascriptover toTypescript
-
2
-
Added
-
As usual people seem to boil down 1 overpowered legion (PE with Mortek Guard blobs) to be that the whole army is overpowered.
Someone not wanting to play against petrifex elite is understandable, but if they refuse to play the whole army (with other legions), they're just jumping on the hate train and aren't worth playing against anyway
-
1
-
1
-
-
Yes they can. Unless the spell specifies that it only affects a certain keyword (one they don't have). Then, since mercenaries don't get your allegiance keyword, they are not valid target
-
Note: A quick update on what's been happening over the last week:
I have been taking some time to do some project maintenance (tests, docs, storybook, etc.) and have decided to convert the entire code base over to Typescript.
I am also adding in various QoL updates (e.g: an undo on the deleted unit/profile snackbar, tooltip, info page, etc. ).
It is near completion, and once it is out, I will return to tackling new / proposed features.
-
1
-
What would you like for AoS 3
in Age of Sigmar Discussions
Posted · Edited by NoMaDhOoK
Pipe dream: I actually hope that shooting becomes alternating activations (like combat phase), as I think that will allow GW to nerf shooting to a more balanced state.
IMO: A major issue with shooting heavy lists in AoS is that either you get the double turn, can shoot twice in a row and decimate, or you get double turned, spend 2 turns doing absolutely nothing, and get decimated.
This somewhat forces shooting armies into trying to force the double turn in order to win. That massive disparity makes shooting pretty much impossible to balance (either way too good, or really bad).
Having both armies be able to shoot in both turns means that it can be balanced (and reduced) in a more controlled environment (maybe incorporate some of the other suggested shooting changes).
---
This assumes that the double turn stays (though personally I wish it would go away)