Playmaker in production

[Rabagast]

Hi, everybody!

I found this page. Please, take your time to read this.

http://va.lent.in/should-you-use-playmaker-in-production/

What do you think of this man? Is he right?
Take an extra look at the picture with all this States in Playmaker. It made me laugh. I think he made this mess with the States on purpose.

[jeanfabre]

Hi,

 I have seen worse

He has some very good point, but I think it's more about a clash between a strong scripting background expert like him ( he's obviously very talented) where some rules can't be bent, more than anything else really.

He's basically saying: Hammers are dangerous, if you are going to use a hammer, chances are you are going to hammer your finger if you close your eyes. Should we stop using hammers? no, should we learn how to use a hammer, yep Would carpenters stop use hammers because of its potential for disaster, no... Do they hammer their fingers? definitely a few times already and counting . Basically, I guess Valentin was in a bad mood that day he wrote this post 

PlayMaker is a very open architecture, and global variables, open access to fsm content is something that gives nightmares to many developers include me, I admit. Yet, it does not invalidate the incredible power of PlayMaker, and therefore I am very cautious when working, enforcing strict policy when I work to never break the rules he's mentioning on big projects, and if in some cases, I deliberately use them to get to my result quicker, I know that what I am doing is dangerous and so I keep my eyes open basically... it's like doing extreme sports in a way

 I am using PlayMaker 24/7, and I must admit my general pressure level is far less then when I was doing conventional scripting. PlayMaker brings me more clarity over complex behaviours, conventional scripting is left in the starting block as far as I am concern ( like coroutines... I never liked them, I think this is far more dangerous  and breaching basics rules then anything else in Unity... same with components on gameobjects... it can get very ugly and dreadful pretty fast...) . If we would start listing all the things Unity doesn't do by the books... it would be quite tedious   It's simply a matter of knowing the tool you use.

What Valentin is not saying though, is that the opposite of PlayMaker can be a lot more destructive in many cases. Going for a true OOP approach with strong theoretical requirements requires careful planning, in some projects, pre production is huge, to get the oop schema and all the classes, interfaces etc that will form the application. It's a no go for more sensible budget. You can spend months with a team before you write the first line of code....

Who does that apart from google and other big companies?... so you go and start coding of course... d'oh... you will face heavy refactoring sooner or later because you created objects and structures that do not satisfy all your needs, and you can't break encapsulation, or bend oop rules... so what now? In these cases, a severe rewrite is required. Will your client be happy... no because he doesn't want to pay extra. Will your team be happy no, because in your team, you have some that will  screaming as they looks at your code as you tried to fix this. I have been there, I know how it goes... then deadlines gets pushed, tension rises, new members join the team and sometimes break the spirit and you start feeling more and more like dilbert...


The source of the issue with this is the customer ( read final expectations). Even with the best intentions, you define with the client a very accurate descriptions and start building what he said he wanted... only that down the line he changed his mind ( always for very good reasons)... ouch... your whole OOP structure is good for the bin, who's fault is it and who's going to pay for it? And why could you not consider flexibility when you coded? because it's impossible to predict these shifts... flexibility is only good when you define what that means...  You can not turn a giraffe into an elephant without careful planning (!), and the best you can do is to disguise it as an elephant for a saturday night party at best, not for a careful app store submission approval...

 With PlayMaker, you can bend rules, you can do duck tape programming and that's one big reason why the points Valentin raises are not really valid in the real world. This post from Valentin is a very much from a theoretical perspective, and should be definitely considered for future improvement of PlayMaker, that goes without saying.

I (We the community actually) have pretty much mentioned a lot of it to Alex, and I think he was clever ( dare I say genius) enough to keep Playmaker accessible, even if it meant breaking rules or not strictly enforce them. I think this is why PlayMaker is so successful and other people trying to do visual programming framework fails, because they don't let go on them rules.

 I'll happily carry on with PlayMaker, I can't get enough of it really. It's never been so pleasing to develop as far as I am concerned, and conventional scripting will have a hard time make me change my mind on this I am always on the look out for some clever frameworks and alternatives, but they never resonate in my so well in the end.

Bye,

 Jean

[MS80]

I read this article before I decided to purchase playmaker, on the other side tons of positive comments, feedbacks and articles how great playmaker is!

Using playmaker for 1,5 month now, I think it is one of best assets available, it is worth every little cent and more! I use it all the time, productivity and speed grows immediately...

Valentin Simonov is a talented and experienced programmer, maybe his point of view is deadlocked to the rules of "traditional" scripting... Michael Yin's and Richard Wilcoxson's comments hits the nail on the head!

[Marsh]

I think he is just not using the tool for what it was meant for. When you are a full fledged programmer and can do anything you want I dont see much reason to use PlayMaker, unless of course you are making a small game or a quick prototype.

On the other hand if you cannot program or your game is not overly complicated its perfect. We use PlayMaker in our live environment and it is quite handy for artists and game designers to get something looking exactly how they envisioned it.

Backend on the other hand is going to be handled by something a bit different.

I more see PlayMaker as a amazing tool for Indie developers, or big studios who want to give people more control of the product.

[jeanfabre]

Hi,

 Actually, I can do everything without playmaker, yet I am still using it. I am currently building a authoritative Server for a project, building complex modular vessels all connected physically with complex physics, 100% PlayMaker both on the client side AND server side! The custom actions and framework will be published on the wiki when I will reach a mature system ( using uLink). Definitely far from any beginners reach, but PlayMaker has a lot of reasons to be in "big" or complex productions as far as I a concerned AND I am experiencing this everyday.

The potential source of confusion here would be how far can you go without scripting knowledge, but that's a different deal, PlayMaker has still many advantages even if you could do it all by script. That's the catch. PlayMaker is perfect for describing processes, unbeatable by any conventional/non visual scripting IDE.

Bye,

 Jean

[Lane]

It lets lots of people do lots of cool stuff without learning lots of code.

If you're working at his level, you're going to have strong reservations about many things. That's fine, but not everyone can bench 480 lbs so many choose to use a brace to help with that, and they bench less and don't compete in competitions... Theres so many illustrations to clarify the picture, maybe the hammer was the best one.

Feedback is always useful though, theres some good food for thought in the article.

[Flying Robot]

I'd agree on one thing though. PlayMaker dearly needs a system to collapse nodes into compounds.

http://hutonggames.com/playmakerforum/index.php?topic=3855.msg17848#msg17848

This request has been generated for over an year. I don't know the progress on that. May be Alex have some good news for us.

[Rabagast]

In my opinion, Playmaker has made my dream to become a reality.
When I got Commodore 64 in the 80s, I loved to make levels in games.
But to make games was really difficult, I think.

The only thing I learned was this:

10 print "Hello!"
20 goto 10
run

Before I got Playmaker, I tried to learn Java Script, but it's so difficult to remember everything. So Playmaker is sooooo fantastic. I can at least make all the games I want to make. 

[Lane]

I'd agree on one thing though. PlayMaker dearly needs a system to collapse nodes into compounds.

http://hutonggames.com/playmakerforum/index.php?topic=3855.msg17848#msg17848

This request has been generated for over an year. I don't know the progress on that. May be Alex have some good news for us.

I think Alex has been working on some other stuff for 1.8 but perhaps the idea is on the backburner... Worth bumping, really. It's a great idea. Maybe the community can flesh out the suggestion and clarify the way forward.

[Marsh]

Actually, I can do everything without playmaker, yet I am still using it.

Fair enough, You have a lot more experience with the product then I. Though as you said by the sounds of it you are scripting a lot of your own stuff. I guess that is what i meant, without some knowledge of how to create actions yourself you may run into a bit of trouble.

[jeanfabre]

Hi,

 Yep, I do a lot of scripting, mainly ( purely )  custom actions and bridges for PlayMaker. And that's the catch, it's still 100% PlayMaker isn't it! that is I don't have a single behavior that works without PlayMaker being involved in the decision making process and flow of events. My scripting is really about building "lego" blocks, that on their own can not give life to the whole, instead they are simply part of the system handled by PlayMaker.
 
So, yes, in that regards, scripting is still important in big projects. And that's why Alex is aggressivly investing in people like me to produce and cover as much as possible and give this all back to the community. Everything that I build on various projects, end up on the wiki, and on this forum.

Bye,

 Jean

[redikann]

Playing off the Hammer comparison let's up it to a Post Hole Digger.
Option A - A classic Post hole digger is basically two wooden handled shovels joined where the handle meets the blade. It's a perfect tool for the job. It's safe. It will give perfect results in the right hands. But it's a lot of work. And the bigger the project that work is grossly amplified. There's absolutely no way to speed up the project except for brute force manning or man hours.
Option B - A small engine driven post hole digger. It's an engine with a 3-4 foot shaft with spiral blade that cuts into and removes ground very fast with no effort. Incredibly dangerous and very easy  to turn a project into a nightmare in the wrong hands.

Playmaker is option B. Anyone can purchase Playmaker and less than 30 minutes be walking around inside a 3d world with a camera and player controller with a 360 controller. That kind of power is extremely deceptive and possibly reckless. But it's awesome. Playmaker is easy to learn but hard to master. It's very easy to work into corners because you didn't visualize beyond that one task you were chipping away at.
It's a tool that will almost do all the work for you so it's your responsibility not to let it get out of your control. Sometimes that's the hardest part, not being seduced by the power of the tool. Forcing yourself to sit at the drawing board and think ahead instead of just going full throttle with Playmaker. I am definitely a non coder. I am a Playmaker addict. I am still getting burned by my love. I continue to get better at how I implement my control into this beast.

[jeanfabre]

Hi,

 It's so funny you make this comparision. I bought this year a post hole digger with an engine ( one man operation, Diameter 30 cm). This is the best diy tool I ever bought ( given my condition, I am a really weak person... being in front of the computer all day... d'oh ; I hate digging...). I can make 100 holes per day, as dip as I want, clean and perfect! but yeah... a friend wanted to try it out, cause I was showing it off of course man!!! he nearly broke it within the first 10 seconds of use, and really in his hands, it was such a deadly weapon of self destruction, I was amazed by the contrast between how I handled it even on my first use... I was shocked really, no trial, no analysis of what may happen, what could be going wrong, even not letting the engine warm up, straight in, full throttle, his whole body weight on it... brain dead basically...

You are making good points here. PlayMaker is really something that will suit people, while turn off others. But the bottom line is: You will get faster results with PlayMaker, and if you are careful and apply yourself, you'll be on top of the game and leaving everyone else in the starting block.

The other very important thing is that Playmaker IS NOT an alternative to NOT learning Unity, 3d, maths, vector operations, etc etc. You are still very much responsible for this. We try here with this forum to provide as much help and feedback we can, but ultimatly, it's really important that you sit down, and brainstorm what you want to do.

Typically, one way for me to work with PlayMaker is to create an Fsm with just the states, ( empty of any actions), just to "draw" the logic, and then I add the actions inside each state and further split a state into several, or a sub fsm or a template, or yet another Fsm. This is a very good way to become aware of what I want to achieve in a way. If what I read on the fsm and the flow of states and events is satisfying, then I proceed with creating all the actions I need inside each states.


 Bye,

 Jean


 

[Aaddiction]

Yes, the problem with this guy is that he's looking at Playmaker from the prism of his stiff university education and experience. He expects to see something that is doing the things better or equally compared to his own long-time programming routine.

For me Playmaker is not to make me better programer, as I'm not. For me and many others it's the tool that allows people with no coding experience (designers, artists) to start prototyping and creating their own games.

I don't want to be on his teaching courses and give him money, as I need years to reach what I currently do with Playmaker in two months. I don't want to expand my career as senior developer in JS or C#. All I want is to make my not-so-complex in terms of logic games a reality in Unity. I'm pretty specific with my needs. Playmaker is very close to give me all that with the help I get in the forum.

I want a complete product! I want my design ideas, models and textures and ultimately my game to get live now, immediately, on my own. How can he give me that from his article? If he dedicates on my projects for a year for free, sure, I'll let him write the code himself if he wants to.

Playmaker is like self-driving electric car which Google and other companies try to make. It's elegant, gives freedom and saves money. It helps people go to work while watching video, painting, chatting with friends or actually working. This guy on the other hand is the redneck in the heavy duty pick-up truck next to it, who says "Whoa, how can I transport my lumber and cows with this?".

Again, for a non-coder the visual scripting is an amazing thing for my needs and any other needs as well.

[jonathangersam]

When I started learning to code a decade ago, I messed up big time. Took a lot of reading and practice until I got good.

Why should we expect PlayMaker to be any different? Yeah we may get to results faster, but working PM at a high level also benefits from learning and practice.

Valentin suggesting to ditch PM is as nuts as PM-users asking him to ditch his Visual Studio IDE or whichever he's using. Anyone can blow up any tool, C# is no magic bullet if you don't do it in the context of learning best practices and OOP design patterns.

Now if only somebody could write Our Own big book of Design Patterns/Best Practices bible for PlayMaker, we can be total snobs as well.

If working with "pure" PM seems stifling to you, congratulations your skills have probably graduated to the level of need-to-learn-to-code PM Custom Actions.

The reason "code structure" matters is to accommodate change, so future development doesn't make your system crumble down. If you knew 100% what you were going to write, there's no need for safeguards via structuring/encapsulation/etc. But be careful not to corner yourself where Structure is so prioritized that implementing "the next big thing" is impossible.