|
Okay, you three stood up so fast after the last batch of codes that I went ahead and pushed you through - codes sent, check your PMs ;)
Also, is your design team open? I am fairly good at optimizing filters on large data tables, and I would love to take a look at what you are using for feat validation and see if I can streamline it so you get better performance, since you cited that as an issue earlier.
That is a kind offer! The algorithm for feat validation is pretty good - it likely could stand some cleaning up, but overall I doubt such a cleaning would make it much faster (if any).
The issue before was that the validation routine was firing off *way* too often; by allowing users to choose when they validate feats, and allowing the table to be interacted with while the validation is occurring, it is improved.
I believe the remaining bottleneck is simply processor speed of the server the site is on. The server is a VPS (not dedicated), and its fine for most things, but this particular application is very busy and could use a server with more oomph. As the site grows, we would very much like to look into a dedicated server, which should provide some improvement.
Since we're on the topic I will mention that I'm definitely seeing slow down as people use the application. So, normally feat validation is 20-30 seconds; but if there are a number of people logged in and making character sheets, that 20-30 second wait may grow. I don't like it either and we definitely hope a dedicated server (as well as perhaps more code tweaks) will improve it.
|
A couple of things:
For Aasimars, it appears that switching heritage is making it appear that one no longer has the Skilled trait (which is actually just modified), so that one cannot swap it for something like Truthspeaker.
Also Oracles don't appear to be able to select cure spells or inflict spells (they choose one type at level 1 and automatically gain such spells as they level, such as cure light wounds at level 1, cure moderate wounds at level 4, ect.
|
|
| 1 person marked this as a favorite. |
A couple of things:
For Aasimars, it appears that switching heritage is making it appear that one no longer has the Skilled trait (which is actually just modified), so that one cannot swap it for something like Truthspeaker.
Also Oracles don't appear to be able to select cure spells or inflict spells (they choose one type at level 1 and automatically gain such spells as they level, such as cure light wounds at level 1, cure moderate wounds at level 4, ect.
Both comments are valid and I added both items on an internal 'to-do' list. I thought we finally nailed heritages down. We made the mistake of considering the 'Skilled' and other abilities to be replaced by the heritage, which was wrong. We'll fix it - thanks! ;)
|
That said, for the moment - all characters are now added. I think Elsa is my current favorite.
Thank you ;)
Mechanically she could have been more powerful (it especially hurt to use that second trait to get Perform as a class skill), but the theme won out! Been away from the PC for a couple of days but I have a bunch more to input now and see if anything else pops out as being wonky!
Still loving the site and the speed of your improvements continues to astound, this is turning out to be an amazing resource :)
|
|
| 1 person marked this as a favorite. |
Still loving the site and the speed of your improvements continues to astound, this is turning out to be an amazing resource :)
I won't lie; I love hearing nice things. Thanks so much and really I'm just thrilled that this seems to be catching on and growing in a positive way.
Side note - racial heritages 'should be' fixed (Aasimars, Tieflings, and Dhampirs take note). They no longer replace other racial traits as that was realized to be a mistake. Existing characters of these races who previously selected heritages may need work - I'd suggest removing any heritage they have, and then re-adding it. Doing that should 'fix' the character so that their racial traits are as they should be.
Thanks!
|
|
| 1 person marked this as a favorite. |
A couple of things:
For Aasimars, it appears that switching heritage is making it appear that one no longer has the Skilled trait (which is actually just modified), so that one cannot swap it for something like Truthspeaker.
Also Oracles don't appear to be able to select cure spells or inflict spells (they choose one type at level 1 and automatically gain such spells as they level, such as cure light wounds at level 1, cure moderate wounds at level 4, ect.
Alright, fixed both. Oracles now get a new choice on their class page - Cure or Inflict.
As mentioned Aasimars and similar race characters may need to remove their heritage and then re-add it in order to 'fix' themselves. Future Aasimars should be just fine.
|
|
Okay, we are catching up with our tickets and I'm starting to think archetypes might get poked at soon. I'm inclined to start with a single class, add all the archetypes, then work out any issues before doing the remaining classes.
So - any votes for the first class to have archetypes? Clerics seem pretty popular. Or...suggestions? ;)
Side note - we've added the most basic of temporary modifications on the character sheet. On the 'Stats' tab, you can now add or decrease your ability scores to reflect game state. After changing a score, hit the 'Refresh' button to make it propagate throughout the character sheet.
|
|
The number of things changed shouldn't matter all that much once we dive into it. This is going to work by setting up an 'engine' of sorts that will handle the process whether the class only replaces a single item or several.
The work on Paladin's Oaths was a good start for archetypes. They are pretty darn similar, and I believe (cough) they are working correctly in the app. At a minimum we can approach archetypes in the same method, though we'll be looking to make it 'better'.
On a very simple level, again alternative racial traits even compare. They replace existing traits, and you can also take more than one so long as the traits they replace do not overlap.
The 'extra' work will be when an Archetype replaces a 'big' class feature as Kegdrainer's example where a Wizard can lose Arcane Bond and their School; those sorts of changes will require updates to the Wizard UI. But - still, that's not so bad.
|
|
| 1 person marked this as a favorite. |
Are codes still being sent out, I would like to try it.
I'll see about sending some new ones out soon - a week at most. Monitor PMs ;)
Are Armour Spikes a thing that's in the generator yet? Couldn't find them on the Armour or Weapons lists to purchase, but I may just be blind!
You're not blind! All is well. Anything that modifies armor (or weapons) such as enchantments or other attachables like spikes are not in yet. Tentatively we're planning on doing those all at once when we do special materials (e.g., mithril, cold iron) and other enchantments.
|
|
Paladin please.
And three days later - all PFS legal Paladin archetypes that I am aware of are now IN. Please go try them out and find things to break :) You will find the archetypes right next to oaths on the 'Paladin' tab. Please *do* expect bugs and issues - the ink the code was written in is not yet even dry.
The UI for archetypes is a bit unwieldy as each archetype has a lot of information. I'd like to change that (maybe a 'slideshow' technique like with class and race - what do you all think?). For the short term though, just getting it to work is something, as the logic has a lot of shuffling to do and did get squirrelly.
We also fixed oaths. Previously, you could only select one of them. After looking into it further, we realized that oaths are really just like archetypes - you can select more than one so long as the class features they replace do not overlap. So, now you can select as many oaths as you want so long as the class features they replace are around. Of course this made archetypes for Paladins extra-interesting as the replaced class features need to be mapped between the two.
If we could get suggestions on making this archetype page as good as we could get it, that would be appreciated. We can then make it the 'standard' UI for the other 30 odd classes.
Thanks!
|
|
Not sure if I mentioned this before, but thank you and your cohorts for doing this.
Watching form the sidelines on comments on coding issues, and playing with it for some of my PCs, is interesting and informative.
Thank you! It is a definite pleasure to be working on something that is needed and which may provide real benefit to this game that we all love. We have a long way to go but with patience, tenacity, and the support of the community - we will get there.
|
|
Might I suggest monk archetypes next? There are quite a few you can use simultaneously so that may be a good test for your system.
I think that'd be fine ;) We need to take a moment to look into a couple of other issues, but monks can be the next to get their archetypes.
Tentatively I am hoping we can start on that in a few days.
|
Might I suggest monk archetypes next? There are quite a few you can use simultaneously so that may be a good test for your system.I think that'd be fine ;) We need to take a moment to look into a couple of other issues, but monks can be the next to get their archetypes.
Tentatively I am hoping we can start on that in a few days.
Just a reminder: The Qui Gong (sp?) archetype for Monk includes a special thing, where you don't have to actually trap out the original pieces, which makes that archetype work with most (all?) other Monk archetypes, but makes it a pain to program.
|
|
Just a reminder: The Qui Gong (sp?) archetype for Monk includes a special thing, where you don't have to actually trap out the original pieces, which makes that archetype work with most (all?) other Monk archetypes, but makes it a pain to program.
Thanks for the heads up! I had not read about this archetype before though I've seen the name referenced. It is indeed unique.
For the current state of the app, this archetype is actually incredibly easy, as at level 1 it does not actually replace a darn thing. It will get more interesting when we up it to a full level 20 application.
But - even then - it won't be so bad. The logic seems pretty straight forward, if not typical. If I'm reading it correctly, a Qinggong monk can trade off their level 4 class feature 'Slow Fall' at that level for a new Ki Power (as an example); however, if they took a second (or third) archetype at level 1 that also removed Slow Fall, then this extra Ki Power selection will be denied entirely. This is just a conditional check and is easy.
If that is the gist of it - then it will take a little special code, but it'll be okay ;)
|
|
Okay, pushed up a big update!
The package which handles many of the data tables was upgraded to the most recent version, which has a different API. This means things may look and behave a little better - but it also means that there may be a couple items that were unintentionally broken. Keep an eye out and let me know.
The biggest change is the Feedback has been completely gutted and redone. The communication cycle of feedback was really bad before - now it is more like a standard forum thread where we can discuss a fix. Also, users may now close their own tickets - we will only rarely close them from now on. This is to make sure that users feel comfortable that their request was met and the ticket is only closed when they feel happy with it.
With that done, we will start looking at Monk archetypes next. It may be next weekend until we see them on the server, though. We'll do our best ;)
Once those monk archetypes are done, we'll also send out another round of codes to those who have requested them.
Thanks!
|
|
Animal Companion Skill points: Only one point is allowed to be allocated per skill, but animal companions start with two HD.
Animal Companion Tricks: Animals with 1 int start with 3+1 tricks instead of 6+1 tricks. Generator allows for 7 tricks on 1 int animals.
Fixed and fixed :)
If either still isn't quite right, let me know and we will get it corrected.
|
|
How quick could you get the vigilante class (classes? archetypes? I don't remember what the specializations are called offhand) up? Quick enough that we could tinker around with them before the playtest ends?
I really like the way you are thinking, though regrettably I don't know if we could pull this off.
But, to answer your question first - I'd say about a solid week of work.
That said - I'd worry of the value of it, given that it is only a playtest at this time (meaning - we know it will change, and so the app would have to change with it). We're working on the archetypes of the second class of about thirty total classes, and getting all the archetypes done is going to be a long haul. I'm hungry to get deep into it. With thirty classes - if we take a week to do each one, that is thirty weeks and is the better part of a year, and that is not acceptable to me. I'd like to get all archetypes done much sooner than that, but it is going to require some real focused effort.
So, my concern is that we'd have to stop that crunch in order to work this playtest. If it were later in the game, the answer would be definitely 'yes', but as it is I think we need to maintain focus and get these archetypes done.
However - when Tabletop Giant is a fully functional application, I think using it for play testing purposes would be a great use of it. A week to construct a new class isn't that bad - if there isn't much else going on. Further, being that the application is centralized, we could even create metrics based upon how people are creating characters using the application, such as how often each specialization is used, and what feats/traits/skills are most often used with the class. I would think that the game developers would adore such easily accessed data.
For now though I'm steaming ahead with Monks archetypes and I estimate completion sometime this weekend. I am also planning on taking a week of my day job's vacation time in order to start cranking harder on this. I have some big ideas what things we can do with this application once it is a full level 20 character generator and I just can't wait to get into it.
|
|
I promise to send out new codes to those who asked once we get this monk archetypes up - by the end of the weekend at the latest, I think (which begs the question - which class would you like to see have archetypes next?).
I've made numerous sites that integrated a Facebook login, and you are right that such technology is pretty easy to setup. I believe Tabletop Giant is going to go a more independent path. I realize it is a small frustration to have yet another login and password; at the same time, I've developed an apprehension at the idea of involving the code of a mega-corporation such as Facebook - especially 'big' code that handles things like user authentication. The motives of Facebook in creating such login tools are dubious and are certainly rooted in their desire to track their user's interests and activities by being able to see what sites they login to. I don't really want to be a part of that tracking.
However - if Paizo, on the other hand, wanted to talk about a universal login, then we may have something worth pursuing. :)
Related note: I also feel the same way about Google Ads. I do hereby promise that this site will never be besmirched with tacky streaming flash ads from Google or from anywhere else for that matter. Ugh.
|
|
Aww hopefully it soon I need the odd for a friend soon
Try it now - went ahead and pushed up the prep code for monk archetypes up and that fix was in it.
Note - monk archetypes aren't in yet - but soon.
sorry just want to make sure im on the list for a code. Can I get a code please when you have them.
You're good - I'll get you one, I promise :)
|
|
Monk archetypes are UP. This includes Unchained Monks - they appear to be eligible for Hamatulatsu Master, Master of Many Styles, or Weapon Adept (feel free to point out ones missing or if any of these three are not correct for any reason).
Tentatively I'm planning on Wizards next, followed by Rogues. Wizards look to be a bit more challenging, so it's a good one to dive into.
Please do try out the new Monk archetypes and let me know if anything is amiss.
Lastly I did also send out new codes to all those who asked - if you do not have a PM in your mailbox and asked for a code, please do shinkick me and I will get you one ASAP.
Thanks!