Jason Rupert's Answers Part 2- From the EA Forums thread - Bahnzo's Blog

Bahnzo's Blog

Jason Rupert's Answers Part 2- From the EA Forums thread

Posted on February 3, 2009 at 01:19 AM.

(0)

Part 2. Same as Part 1, Jason's answers are in BOLD.

> One thing i do think the game gets right about diving
> though is that if you dive IN FRONT of somebody, and
> then they skate into you and fall down or lose the
> puck, it is correctly a no call. In both real hockey
> and in the game i like to slide horizontally in front
> of people, either blocking the puck, or causing them
> to trip over me, which is legal since i was there
> first.

``Clipping is the act of throwing the body, from any direction, across or below the knees of an opponent.`` http://www. nhl.com/rules/rule49.html

I had that done to me in my last (real) hockey game. Was not happy about it or the fact that it went uncalled.

> I'd like to atleast see 90% of the interference
> penalties called. Right now they are almost never
> called, and if someone is called, they are not even
> called for interference, they are called for cross
> checking. I understand that the problem might be the
> defensive AI that is programmed to be quite
> aggressive, so it might be a problem to do balance
> this on this patch.

Yes, I think it would not be going out on a limb to say this will be fixed for NHL10. I would classify the rarity of interference calls as a bug. Hits in front of the net are allowed by the game now, though they could get a cross-checking or elbowing penalty if they are too extreme -- that aspect is deliberate, not part of the lack of interference bug.

> Broken sticks? Let me get back to that later.

Broken sticks show up every year on feature requests and are a popular idea internally too. They are also a very relevant part of hockey (as are dropped sticks). The problem with them is that a player without a stick can still do things on the ice, and each of those actions WITHOUT a stick is different than it is with, including skating. That means doubling most of the animation in the game, if you want to do it right. For broken sticks? Alternatively you could just have the player rush off to the bench immediately, but what if that would be unrealistic, say if that player is the only one who could defend a scoring chance? That would be a bug, possibly one that is means the broken stick feature does more harm than good.

I try to avoid features that have built-in bugs or that double the amount of animation we need for a smallish improvement. There is no shortage of things that need work AND do not have that sort of downside.

> Jason, since you are involved with the design of the
> gameplay as well as the actual gameplay, what do you
> think of the fight initiation as it is now? I mean, I
> won't say what it looks like to me, because this is a
> respectable thread, but it doesn't look like nhl. Is
> there anyway to add a button that shoves other
> players a little bit, maybe puts them in a headlock
> after the whistle, or just plain old drops your
> gloves? Maybe a little facial action? You know, the
> classic NHL rough play along the boards type thing.
> Something to submerge us a little more deeply into
> the authenticity of the game. Just wondering if that
> is possible.

This sort of follows on top of what I said in response to ChitownFan526's question about fighting. If it does not tie into the rest of gameplay fighting is just a random mini-game, so I would want it to flow visually as well, much as you describe.

Another point to mention about fighting is that the cost of fighting improvements is typically higher than a lot of other gameplay improvements because it has mechanics that are very distinct from everything else, and it is not uncommon to see real NHL games go without ANY fights, so it is definitely a peripheral aspect of the game, at least compared to thinks like skating, shooting, passing, pickups, hitting, goalies, etc. That is not to say that it isn't important, but these sorts of considerations have to go into the decisions we make about where to put our efforts.

> Why did EA decided that the Offensive defenseman
> should have only 65 defensive awareness to start
> with, when every other offensive player, even snipers
> has at least 75? Isn't that a mistake?

The base attributes for the players is something that is on the borderline of gameplay, but it was actually done by a different subteam this year. When I noticed what you observed myself I thought the same thing. It seems like a mistake, I have not followed up with anyone about it yet so I am not sure.

> On a related note, you shouldn't allow to use forward
> player models as defenseman. Again i must tell you
> that the most used player models on defense by the
> top EASHL teams are:
>
> 1- Playmaker for the extra def awareness, passing and
> agility compared with the Off Defenseman.
>
> 2- Dangler for the extra def awareness and
> agility/speed compared with the Off Defenseman.
>
> 3- Power forward for the extra slap power and puck
> control than the Defensive defenseman.

I almost always play defense in EASHL myself and used a power forward myself for a long time.

> Here is my suggestion/question i guess. I think it'd
> be great if, in the EASHL, ONLY the goalies had a
> fatigue factor. This way when you put the pressure on
> their goalie there is more incentive for them to stay
> back and play D because with eacho shot we put on
> their netminder, he gets slower to make the save, and
> slower to get back up. This would deter cherry
> picking until at least later in the games. As it
> stands now, the cherry picker basically knows his
> goalie will bail him out eventually then he can be
> set up for the breakaway pass.
>
> Anyways what do you think of that? One more
> suggestion. Why not enforce the 2 line pass rule for
> cherry pickers? People would argue that this is NOT
> AN NHL RULE...well my answer to that is...this isn't
> the NHL...IT ISTHE EASHL....You guys at ea can make
> your own rules..lol
>
> Thanks agan Jason for taking time out of your busy
> schedule for us. IT is very much appreciated.
>
> Ken

Agreed on the fact that the scoring balance is not great, this is something that is being worked on for the patch. Deeper changes like the ideas you suggest would have to happen after that. I had not thought of the mechanism of goalie fatigue as a solution (goalies have NO fatigue during a game now), it will be kept in mind when we figure out what we do to address the scoring balance going forward.

For cherry-picking: the NHL works well with no two-line pass, so my inclination would be to ask why cherry picking is not effective in the NHL and what the game is missing that allows it to be effective in the game. My first thoughts there are that it is way too easy to make the pass and that the defense are not good enough at stopping it. Fixing these would probably contribute to the gameplay in other ways besides making cherry-picking a less attractive strategy.

> If you add broken sticks please do it very rarely. I
> can see it being like injuries where there are too
> many of them and they will happen at the stupidest
> times like injuries currently do like when you are on
> a breakaway and a player gets injured way back behind
> the play and the play is whistled and your breakaway
> is ended. It seems like alot of the new features
> added each year like stick lifts for example are done
> like 100% effectiveness in your face just to make
> people aware of the new feature. Introduce new
> features so that they are just a part of the game but
> don't overpower the gameplay. Another example of
> this was wraparounds in 2005 they were way to
> effective.

Injuries are a good comparison with sticklifts. The way that injuries work now, with the flaws it has, is probably the best that broken and dropped sticks would come out if we implemented them. The problems with injuries CAN be fixed though, they are more poor design than an inherent difficulty.

Unless we discover a way around the design issues I described regarding broken sticks they will only get in the game over my dead body, so to speak.

> What I am talking about is diving or sliding onto the
> ice, and then having THEM (and / or the puck) run
> into ME.
>
> In my opinion the game does VERY well with this. If
> i screw it up and dive INTO a player, taking them
> out, there seems to be a penalty a fairly decently
> high amount of the time. When I slide down and block
> the ice in front of somebody, (i have a right to be
> there and hold my ground, just like blocking /
> charging in basketball), and then they skate into me,
> usually causing the puck to stop at my body and them
> to trip over me, a penalty is correctly NOT called.
>
> Not only is this called correctly penalty-wise, but
> it seems to be very balanced from a gameplay risk /
> reward perspective.
>
> If i time it perfectly and chose my spot well, it can
> be a very effective defensive tactic. If I miss, or
> they are alert and swerve around me, i completely
> remove myself from the play.

Ok, I get what you are saying and I think it fits in with what I do like about the diving poke check. What I don't like about it are that you can get away with taking away a breakaway chance by tripping, that you can just throw yourself in a heap *and stay down* in the way and that, combined with how difficult it can be to maneuver around a prone player, make it overly effective sometimes. None of these things are anywhere near the top of the list of biggest problems with the game, I'm just saying there are aspects of it that I don't like.

> How about the LB glitch to get up after you get hit

Barring problems during testing this will be fixed by the patch.

> Right now all I got is complaints about collision
> detection in BAP spectator mode...ie. on the
> bench...guys have invisible heads and stuff.. dont'
> know if this is considered "gameplay" or not...

It is gameplay. Basically anything that affects the physical and behavioural simulation of hockey on the ice (or bench) is gameplay, as well as controls. Stuff like rendering, cameras, crowds, audio, online servers are not. There are some gray areas on the borders too. No problem asking if you aren't sure.

This is not a problem I have noticed I'll have to keep an eye out for it.

> I don't like the cross checking when you are
> expectinga pass and you use the toggle stick... seems
> like I take penalties for this too much...don't know
> what can be done as I like the toggle stick check
> control but only when I WANT to check?

One possibility here would be to extend the "Vision Control = Do Not Hit" fix that was put in the first NHL 09 patch so that it applies everywhere. I considered it for the first patch but was afraid that it might be problematic for people who use Vision Control a lot when on defense. They would have to release it before trying to hit.

The other approach is just to put more logic in to determine the context and automatically choose which to do (shoot or hit) for you. Getting it right for all situations is tricky and different users would probably expect different things so it is probably impossible to nail something like that to everyone's satisfaction.

> I'm not sure if this was answered or not, but how
> exactly is the patch going to fix this problem? I'm
> sure it's going to prevent players from maxing out
> their stats, but what is being done to the players
> that have already maxed their stats? Will all stats
> be cleared...including both player and league stats?
> I think that is probably the best solution in my
> opinion...

It will not allow people to have higher attributes than their Card level entitles them to. People who have legit players won't have to do anything, people who have "juiced" players will need to redo them.

A decision will have to be made about what to do with stats and rating points. I don't see any solution that can be fair to everyone. It is impossible to separate all the players and teams who have played honestly and fairly the whole way from those who have used exploits. As a result either some of the teams who have been honest will lose what they've achieved, or some cheats will keep their ill-gotten gains.

My own thought is that it should be reset, but people should keep their stats and cards because forcing them to work their way back through that is not reasonable. I would be interested to hear the opinion of anyone who has an argument against that, not that it is my decision to make, but I can make sure that all sides in the matter get heard. But DO NOT REPLY IN THIS THREAD, there are already threads about resetting the standings and we are following them.

> Also in EASHL when on a PK the AI seems to dump it
> out of play frequently. Will this be addressed in the
> patch?

Yes, again barring having to pull changes back due to problems during testing. This will fix the problem for offline play too.

I'm pretty confident in the patch changes that I've acknowledged so far but some of the other changes are a bit more tenuous so I should wait until we've gotten through more testing before discussing them.

> Are injuries being addressed with the coming patch?
> I've been only playing on Pro mode since I got the
> game and even through well over 100 games played
> I've yet to ever have a guy get hurt.
>
> If they are not being addressed, is there something I
> can do with the sliders to fix it?
>
> Thank you.

Unfortunately there are no sliders that will significantly increase injuries. The problem is a complicated one because it is in part caused by the fact that the CPU really just doesn't hit that much. Human players on the other hand can easily cause several injuries in a game.

We will discuss what is and isn't in the patch when testing is further along.

I am not really supposed to talk about the plans for NHL10, but I'm sure it is ok to bend the rules a bit and say that improving the sliders is something that is *currently* in the plans for NHL10. What is there now is a bit tragic.

> Great game and I am glad to see the programmers on
> the boards themselves.
>
> I play goalie in EASHL and its very very annoying to
> have the "sticky posts" especially when I try to stop
> a one-timer? have you guys looked at this issue? if
> so, is it being looked at to address in the patch?
>
> Not a gameplay request I would like to have a
> "filter" system for EASHL so those with goalies could
> play teams with other user goalies.
>
> Thanks

I have to confess that I have only played a handful of games as a goalie and I did not work on goalies myself for NHL09 at all until the patch. Can you explain the problem a bit more for me? This was first mentioned to me a few days ago and when I tried it I didn't really see the problem.

We are at the stage where we are not adding to the patch, but just testing and making necessary adjustments to the changes that are already in, so it is unlikely that anything would get done for this in the patch. We can investigate it for NHL10 though.

> I apologize if this has been mentioned earlier in
> this long thread and if so, you're welcome to ignore
> this message. I want to ask about the 3rd period
> "comeback" logic we all suspect is in the game. I
> realize that Littman has denied it and explains that
> - as per real life - a downed team will try and work
> harder in the 3rd period in an effort to score goals
> and tie/win.

Yep, I'd love to bury some myths about this stuff and get to the bottom of whatever real issues there might be.

First off, there is absolutely no CPU attribute cheating as a mechanism for comeback logic. That might be a bit too bold of a statement to make since I have not read every single line of code in the game, or even in gameplay, but I have worked on it for over three years now and I feel that I can say with that with a very high degree of confidence.

The CPU does change its strategy settings, but only using the same settings that are available to the user. It will also shoot more, esp. as you get near the final buzzer; and of course pull the goalie for an extra attacker. These are the only three things that could count as comeback logic that I am aware of and I think they are all legitimate, in concept at least (I'll come back to this in a sec).

There is a fair amount of code in gameplay that is OLDER than 3 years and I suppose it is possible that there is something in there that is treating things differently based on the score and/or time remaining for the CPU. I would be surprised if that was the case, and very surprised if it there was anything like that which has a significant effect. I can say with near certainty that there is nothing in the goalies, shooting or skating code (to list a few core areas) that is cheating because all of these areas are new or have had major overhauls in the last 3 yrs.

All that said, I can't ignore the repeated claims that people are experiencing drastic CPU comebacks that feel completely contrived. A few things that could be causing it:
-Unbalanced strategies -- maybe on `all out attack` the CPU AI is too good
-The increased tendency for the CPU to take any shot means he is taking shots he normally wouldn't and for some reason these are actually more effective
-Confirmation bias -- when it comes back you remember it and put a mental tick in the proof of comeback logic box but do not track the negative evidence the same way. Unless you are recording stats it is pretty much a given that this is at least part of the issue -- that is just human psychology

This is something I will investigate further for NHL10. I can do things to test it, like take away the (legit) comeback logic that is there now and see how it plays, or play CPU vs CPU, same team, but lock the game at 1-0 for one of the teams with 45 seconds left and let them play and see if one team dominates. And of course root through the older legacy code and get a handle on it so that if there are weird quirks in there, causing this sort of problem or any other, we know about it and can start cleaning things up.

It is not a part of our philosophy to orchestrate the game results in any way. The CPU team should be trying to achieve its goals, but it should be doing so under the same rules and physical constraints as you. On All-Star this is what the game is supposed to be doing and if we learn that it isn't we'll fix it.