Ascension is the Sakoto fork of the original plugin known as hot or not. It has been entirely reworked to serve as a comprehensive stash entity rating and tier system that scales and adapts to your database in real time. This fork sets out to further expand and develop core features of the original while maintaining a consistent, fair, and stable rating system.
Additional functionality has been added surrounding mobile support, quality of life, optimization, complex matchmaking, database integrity, new IU experience, and much more.
-
- Should I start over or keep my ratings?
- How long will it take before my performers are all out of F tier?
- Why S-Tier is hard to climb?
- Why do I see unrated performers so much?
- Why am I not seeing performers that graduated ranks?
- Why am I not seeing enough of X tier?
- How to I backup and restore my database?
- How can I check the system is working correctly?
- How do I start over?
- How can I support development?
-
Ascension features 3 modes with more to come to better enhance the entity rating experience.
-
The user must select the better performer card to allocate scoring. As a performer climbs in score, they climb in tier further increasing their competition. There is 5 tiers an entity can belong to exclusively at any time.
- S-Tier - Score of 85 or above
- A-Tier - Score of 70 or above
- B-Tier - Score of 55 or above
- C-Tier - Score of 40 or above
- D-Tier - Score of 25 or above
- F-Tier - Score of 24 or below
Performers are allowed to battle within a 15 point window of their score. This allows for cross tier matchmaking and maintaining competitive integrity. During the head to head mode, a 10% chance event can happen where a cross tier match is forced allowing performers to battle as wide as a 20 point gap. S-Tier performers are limited B-Tiers and above when it comes to the cross-tier match event to maintain competitive integrity and to prevent score padding.
-
Placement mode allows a performer to battle up tiers to find their place in your DB. This can accelerate priming your database and giving performers a VIP pass out of low tier hell. When selecting this mode you are presented with the option of 6 different performers to choose from. Furthermore, launching a Ascension on a performer page will start the plugin in placement mode, switching to champion mode will maintain this login.
-
A mode where winner stays on. If a performer loses their match up, they are knocked out and the next performer takes over. As mentioned in placement mode, you are presented with an option of 6 different performers. If Ascension is launched on a performer page, you can switch to champion mode and it will start champion mode with the performer selected.
-
Matchmaking has been completely reworked to adapt to the user in real time while maintaining match variety, integrity, and fairness. The logic behind this system uses math and real time variables and metrics to further calculate scoring, match selection, and frequency. This is a fairly complex topic that I have elaborated on in the Ascension White Paper.
-
A new scoring system has been implemented over the static scoring system that previously existed in the original hot or not plugin. It rewards upsets, calculates scoring based on tiers, understands match tier and point differences, and understands a performers 'experience' in the system.
All performers start with a K-Factor of 32 to give them their best ability to move between tiers. As their match count grows, their k-factor will gracefully and gradually slide to a K-Factor of 16. The 50% reduction is to help 'settle' performers into their tiers. At 35 total matches, the K-Factor has become 16 for any performer marking them a 'veteran'
-
A weight system has been added to maintain fair match distribution that is calculated off the 'last match' time stamp. Weight distribution is on a scale of .10 to 6.00. A higher rating value is used for newcomers and the unrated. Once a performer has been rated they enter a blackout phase. The performer will exist in this state 30 minutes with a weight of 0.
During this time the performer cannot be selected by the matchmaking system. Once the blackout has expired their weight begins to recharge over a 12 hour period. This process continues until either every performer is blacked out, or 90% of the database has a low weight in which case the system resets every performers weight back to 1.
-
New and unrated performers get significant weight bonus while respecting recency until they hit 10 matches. By 10 matches, a performer is considered integrated and they will no longer get boosted weights. This is to maintain a fair match distribution and to account for scaling for already established databases.
The system will try to catch up an entity as well if their match count has significantly fallen behind the rest of the database (20%) and will boost their weight further protecting match distribution. Once that entity is caught up, normal weight rules are followed.
-
A Tier system was added to not only better visualize your entities, but to classify where they stand on the grand scale of your database. Entities can battle across tiers to maintain their status and can expect many challenges climbing and maintaining. With the Dynamic K-Factor Entities can not only see a reduction in point scoring due to match count but also due to their tier. This is to prevent Entities climbing to tiers they dont belong in and keeping tiers exclusive.
In tiers such as S and A, users can expect a 40% reduction in scoring for S Tier performers and a 20% reduction in A-Tier performers. This stacks with the Dynamic K-Factor.
-
Modifiers add to the complexity of the system further boosting or nerfing performers in real with proper context taken into account.
-
Penalty Modifiers
- 0.6x 1st appearance
- 0.3x 2nd Appearance
- 0.1x 3rd Appearance
- 0.7x Top 10% match count
- Complete exclusion (Blackout) for recently selected; 30 mins
-
Booster Modifiers
- Recency(Full Charge): Max 1.0
- Low Match Boost: Max 2.0
- Tier Boost: Max 2.0 (Only during tier focused matches)
- Distribution Boost: Max 1.5 (Bottom 10%)
-
The UI has been reimagined and is better streamlined on mobile for the on the go user. Changes have been made on how information is displayed and consumed.
-
The leaderboard is now interactive and integrated with the tier system. Users can now better visualize their entities and sort accordingly.
-
The battle cards have been reworked to display more stashDB related information to give you a better assessment of the rating. Some stats have been omitted to prevent bias however users can expect to see the following:
- Now displays sex next to name (emoji) | Scene Count, Gallery Count, Image Count.
- Tier
- Country
- Height in ft and cm
- Measurements
- Fake Tits
- Tags (truncated after 3, can be expanded)
- Point cards show points and scoring with decimal
On Mobile Ascension now uses a deck of cards swipe carousel feature. Users can swipe between cards before making their decision, no more consistently scrolling up or down. Its all right in front of you.
-
The performer ledger (and scoring) now reflects score updates with a decimal value ie 55 = 5.5. Tier changes. Other improvements have been made to further to cleanup minor visual bugs that could occur when using the ledger.
-
A Battle Log has been added to the UI so you can further see what is happening with the plugin mechanically rather than opening the console menu. This gives users more information to assist with debug and overall understanding on how selection\scoring is done.
The battle log features a direct stream of the console logs to the UI here's what you can expect to see:
- Seed selection
- Performer weight
- Selected Seeds total match count
- Performer ID
- Selected Challenger
- Win\Loss result and scoring per said result
- Skip logging
- Undo Match logging
- Tier Selection logging
- Error reporting
The logs are color coded to make information easily glanceable. A trash can (clear) button, hide logs, and close battle log button also exists within the container. On Mobile, the battle log displays under all other containers keeping it out of the users way and only available should the user need the information.**
-
Lots of visual animations have been prodded to the cards on mobile and desktop to give the app a more put together look. Users can expect to see Tier change notifications, and other improved UI elements on mobile and desktop
New tasks have been added for users to protect databases. Users can now prime, snapshot, reset, and restore their databases at the click of a button. This allows users to test beta builds or even the production build with no long term impact on their stash database and without having to backup the stash database as a whole.
-
Randomly assigns all performers a rating between 1 (F-Tier) up to 40 (C-Tier) honoring correct distribution.
-
Deletes all Performer Acension history from custom fields.
-
Resets all Performer ratings to 0.
-
Writes all performer ratings, match history, and ledger to a json file. Example: '[[Date]-[Time]] - Ascension Database Snapshot.json'
-
Looks for for most recent snapshot and restores database.
There are 3 database types
Unprimed
- Your ratings start from 0, you have no match history or ratings.
Sub-primed
- You have ratings or match history from a previous version of hot or not (or elsewhere)
Primed
- Your database started from unprimed and is now fully fleshed out across tiers.
If start with sub-primed DB, you could benefit skipping the F tier grind if you delete your performer match history however it is important to consider that your tiers may be lobsided or inflated and could possibly put some performers at a disadvantage over time due to k-factor scaling. If you keep your match history, then performers with a k-factor of 16 will have less of an advantage than those closer to 32.
You can use the Primer task to jumpstart your database if you do not want to start from 0.
If you started from 0 it can take a quite a bit of time before your performers fill the other tiers on a reasonable bell curve. This is by design. A healthy tier system is going to filter more and more performers as they hit tier walls and bounce back to where they need to be. The scoring is meant to make your selections meaningful. Depending on your performer count, this could take a very long time.
However, there are accelerators that can help your performers climb. For example the cross tier match up will pair performers minimum 20+ points higher at a 10% chance. If they win, in most cases they can skip a tier. Tier Focus events further give performers meaningful matches allowing them to score higher vs lower tier performers due to closer point matching. This begins to domino and grow exponentially as more and more performers pass the 1.7 rating wall. You will start to see your database accelerate as your S and A tiers break out.
Another option is to use the other modes. Placement mode and Champion mode with performers at least in C in D tier can help accelerate getting performers out of F tier. In Head to Head mode with the underdog bonus, performers can see points as high as 2.3.
This is by design. S-Tiers are meant to be S-Tiers and their matches are meant to reflect that. They cannot battle anyone below B tier and face a significant point reduction at their level. If an S tier is in fact an S-Tier this really shouldn't be an issue.
Keep in mind when your database starts from 0 unrated performers take overwhelming priority to give everyone a fair baseline and match count. Over time once your habits of play and scheduling kick in and you've got a baseline.
If your performers have just graduated they probably have a low weight. It is important to remember higher weights will always trump lower weights as you play. Over time, their weights recharge towards 1.0.
It takes time for your tiers to fill out enough to where the matchmaking can satisfy the requirement (minimum 20 entities), you also have to consider whether weight is counting against the performers as well. The database will scale over time with your input. You will see more and more cross tier matches, more and more different tier matches, and more and more competitive matches.
Use the snapshot task in your plugin tasks menu. It will create a json file in your Ascension plugin folder with a date. When running the import, it will look for the most recently created snapshot and import automatically.
The browser console menu features an extremely robust logging system for debug. Users can use this to further gauge if matchmaking is working as intended. Most logs are also forwarded to the battle log stream for review as well if you do not want to use the console menu.
Go to Settings -> Tasks -> Scroll down to the Ascension plugin tasks. From here you can wipe match history and ratings.
Number 1 I will NEVER and certainly have NO PLANS to paywall any of my work or contributions to the plugins I have touched nor will I ever. However, I spend a significant amount of my time trying to perfect these different plugins for you guys while I try to survive in the world. Selflessly. I have a buy me a coffee link where you can send me a few pennies for my time if you'd like to help. Its certainly not a requirement, but it does help.. alot.