PonPonWeiWei

This is all largely correct. Gods would have a primary stat that care about, and this determines most of their scaling, very similarly to Smite 1. Strength additionally is generally associated with how to make a gods Basics deal the most damage.

However, these can be used to scale different things, and some abilities may favor one scaling type over another if the god is Hybrid. Chaac's ult hits hardest with Intelligence, and he gains bonus Healing and Protections from his 2 in our current testing; so intelligence is used for a variety of things on him.

Each god will be a case by case basis on if and how they leverage having 2 potential scaling stats.

Damage typing is still Physical and Magical. We are messing around with gods not dealing strictly one type of damage type.

So for example, Ymir's Ice abilities may deal Magical Damage but smacking someone with his Ice Club will deal Physical Damage. What is scales off of (Strength of Intelligence) is technically separate from what type of damage it deals.

In a lot of cases your intuition from Smite 1 will hold. Anubis largely deals Magical Damage, and Largely scales off Intelligence; but you might get some surprises like Mummify having a damage component that scales off Strength, and deals Physical damage.

The flexibility to change damage typing from the scaling stat though is one of the reasons we went with a name change to make it clear the stats don't work exactly like Smite 1.

You can see for yourself :

https://normal.casualsmite.com/gods/persephone?mode=conquest ...

I mentioned this on twitter but her old 2 did have a problem in that your ability to hit it ended up being quite heavily dependent on your opponent's skill level and not your own. Once enemies knew a quick stutter step maneuver this was not much you could do to make them connect.

Most top level players were mostly forgoing the plants and using the buds to detonate, especially off flourish as more reliable parts of their damage output. With the newer plants you have more control to let these connect or use them in ways to cut off opponents and get them to move where you want them to. They are still dodgable but you have more ability as the Persephone to get them to connect.

I do appreciate that Persephone fundamentally is a different goddess now; many of the tricks and mechanics she once used have been adjusted in ways that have similar theme but function differently from moment to moment. Ideally, we would just balance to help her out but at her core her design mad...

I feel like boiling it down to 'successful passive' carries a lot of baggage that the linked post goes more into detail discussing. We do acknowledge that it has been a sore spot on her rework but the direction we took was one we felt we needed to make and at the very least has not harmed her rework from being successful.

SInce her rework we have discussed her reworked passive a LOT and I would still hold the following :

1. Allowing her to fight at any level after death is likely to introduce frustration that would lead to similar problems that the rework was aiming to resolve.
2. The clearest way to make this more unique/strong would be to add non-combat functionality after death (more than just placing plants). The most common of these is for her to be a living ward but I think it serves as a good example of after-death breaking fundamental promises players have with SMITE and why we had to rework her in the first place. Killing an enemy should offer yo...

Yep :) Definitely not useless and very much intended!

Her first basic attack has a super low prefire such that if you want to cancel into it it is really rapid. Similar feel to Nemesis using her shield to trigger a very quick 2 autos.

Yep, we wanted to ensure that portal was only used in places where it was clearly her ultimate or in context with an animation (dying, recalling) that could provide the context it wasn't a gameplay element.

For entering/exiting the wall we wanted it to be visually distinct so that was clear for enemies to know her intentions and that if they see a portal in that scenario they know it is dangerous. Functionally she also doesn't use the portals to 'travel' so the theme of her becoming ghostly might have been lost if they were used.

We did test this one actually. At the time we looked to see how Anhur impale interacted and Odin also get stunned in that scenario. Since that was how something similar already interacted we figured it was best to maintain already existing behavior.

Ill try around those areas. If you do get a clip of the interaction I would love to see it.

Do you have any specific walls this is occurring on?

Now you're cooking with slows.

If it is bigger it should probably get it's knockup back for theme reasons. Although we did remove that; probably can't bring it back.

How about a slow?

Sure, and I would be OK if this is just something we disagree on.

To me the 1 being random is about playing around the hand you are dealt with the occasional 'high roll' with the drawback being a specific lack of consistency. This isn't the only version of RNG in her kit but it is the more extreme version.

To some people this is going to be a complete deathblow to the desire to play her. I think this is OK. Assault is somewhat similar. Some people like the randomness and others don't and won't touch the mode. We even mitigate the RNG some with healer rules but if you add too many more I think you remove a lot of what is appealing about the mode. (TL;DR here : It's ok if the RNG alienates some players. There is also probably an 'optimal' RNG amount to maximize how many people enjoy her).

Here would be the two ways in which I think the OK goes to Not OK.

1. If her popularity suffered so much due to this. She isn't the most popular but she also...

Possibly, but it also helps us push to explore territory that can provide unique gameplay. With pure focus on Gameplay it can be easy to not explore more dangerous ideas or just immediately dismiss them; but often you would miss out on the insight of both what options to explore once you commit to an idea but future lessons for gods.

3/4 of your examples I would say show this pretty well works. Pushing how "disorienting" we can make his horn was a relatively low risk exercise on our part that wouldn't break the characters theme if it was not there and had some potential payoff that is could have bolstered his theme some and potentially provided unique gameplay. It was tested as a known higher risk aspect but it was a pretty easy thing to try out, iterate on, and discard if it didn't feel like an add.

Tsukuyomi's Basic Attacks is similar. It had the potential to double down on his theme and reinforce his precise and calculating demeanor and the pointedness of the ab...

This is actually neat. Often when working on gods we are more specific to them and may or may not fall into patterns without realizing this so this was a good opportunity to gut check some of the decisions made on those gods. While if I could go back and change some decisions I have made through various gods developments (I think any designer on any game feels similarly, lol); I am unsure if I would change much about these slows.

I was curious to go back and get specific numbers but I know in general SMITE is a pretty slow heavy game. Looking at the total god roster of 116, 89 have a slow to some capacity (76.7%). Removing all of these gives a historical 71.2%. It serves as a reward inside a kit to follow up on further aggression without being so disruptive as to remove control from players and in general creates good fair moments of tension so I am not surprised that this is the case. It is also why Winged Blade has always been (and likely always will be) a really good ite...

Text is not updated due to how this was implemented. I did hop in this morning into a Custom Conquest to confirm it functions as intended. Local matches (such as Solo Jungle Practice) may not show it working either.

That system sounds like it takes a different approach; which is fine but it is a tradeoff of being explicitly slower. You could make our system roughly match that by essentially using our confidence metric as a multiplier on MMR. At 0% confidence we assume you are 1500; then multiply by 0 to get you 0. You play and win; your confidence goes to like 10% and your MMR goes to 1600. You functionally have 160 MMR.

This is a common approach with systems like this but again; it is a tradeoff. While it lessens the pull of bringing down the median MMR it isn't actually ignored and it will explicitly take longer to test and validate a players MMR. We tend to find that these types of changes seem more detrimental in a game like SMITE where there is a lot of noise when it comes to measuring player skill. (9 other factors on your team than say a 1v1 for example).

In the short term it would have the effect you want.

In the long term it would not. If every new player started at 0 over time 0 would become the new mid-point.

The tradeoff being made here is mostly that this is the quickest way to place people accurately. You can do things to make it less accurate (artificially set players MMR, force restrictions on who they can play against) but these make the system take longer to find their true MMR, and this extra time further compounds into less accurate matchmaking in the long term. We make some of these compromises (there are soft restrictions in place to prevent new new players from facing veterans) but Matchmaking systems are quite delicate as they rely on assumptions and breaking those assumptions can lead to unexpected behavior and wonkiness in the long run.

For what it is worth the general goal for Starter Upgrades internally was to shoot for about 110% the power of other T3s. They ask players to lock themselves into potential build options very early on and we felt it was important for these to always feel like a + so that decision feels good. It's why we recently just announced buffs for basically all the Jungle Starters as they didn't meet that goal.

As for this post; a bit of math.

Let's use Ra's 1 as an example since its numbers are nice :

335 + 100% Magical Power. In the above build we have 725 before we add either Sacrificial or Obsidian Shard. We also have 10% Penetration from Rod Of Tahuti.

Let's assume 3 targets. 100 Magical Prots. 200 Magical Prots. 300 Magical Prots.

Sac 100 : (335 + 840) * (100 / (100 + (100 * .9))) * 1.15 = 711 damage.

Sac 200 : (335 + 840) * (100/ (100 + (200 * .9))) * 1.15 = 482 damage.

Sac 300 : (335 + 840) * (100/ (100 + (300 * .9))) * 1.15 = ...

Those are 'local' matches. As in they run on your machine and don't connect to a separate server to play.

If you had someone in your party and went into Jungle Practice, or played Conquest Custom, etc; you won't see this issue since those are 'networked' matches.