Tips for getting your questions answered?

So I've posted a few questions on here over the past month and haven't had any luck getting any answers. These aren't questions that are that out there as far as ease of concept. One was a Mecanim question that could pretty much be answered by anyone who is well experienced with Mecanim and the other was a MipMap question that seemed like it should have been pretty easy to answer by someone experienced in that area as well.

I could be wrong and these are both actually very difficult questions to answer, but since they both involved a "What is the best way to approach this type of situation" approach I figured I'd at least get a bite or two.

I've tried wording my questions in such a way that they aren't too long, but provide adequate information as to what I'm trying to do and what my problem is. Here is the overall structure I try to adhere to:

1. Explain my situation, what I'm trying to achieve.
2. Point out the issue I'm having that is preventing me from reaching that goal.
3. Provide additional information as to what steps I've taken in attempting to solve the problem.

So am I missing something here? I'm really hoping that this is just something that got buried in the hustle and bustle of forum activity, as tends to happen in forums, and not something more nefarious. It would really suck if I'd been "blacklisted" in some way because I had the gall to suggest that maybe the recent release of Unity wasn't pitch perfect in every way.

I also feel that one of the major issues with "community based support" is the issue of diffusion of responsibility. If no one answers a question because each person thinks someone else more experienced and helpful is going to answer, then that question is just going to get buried no matter what.

But of course, I can't rule out that I'm doing something inherently wrong here in my question asking process. If so, please point that out to me and I'll take the appropriate steps to correct it.

 

Wouldn't it be funny if nobody answered this question?

 

I was actually going to add that to the end of my tirade, but elected not to since the post was already far longer than it needed to be.

 

learrn tips from how this question gets answered

 

Generally they won't get answered. If you've a legitimate coding problem that shows you make a concerted effort to solve yourself you may have better luck, especially on the Unity Answers.

 

Can you elaborate on what constitutes a "legitimate" coding problem, versus a bogus one? Does my three step process fail this test in some way?

 

Effort.

Experienced coders are pretty good at detecting effort in the questions asked and in the original problem being asked about.

And so, as you've noted, they've basically stopped doing so.

It should be a big red flag to you that Unity Answers is moderated. And that's the place you should be asking technically intermediate questions really. Advanced questions you can't solve are ones you'll solve yourself or pay for a solution.

But if you have a one liner question about what this is or that is in the Unity editor or a C# class declaration problem that shows you are trying to learn the basics, really, then experienced users will often answer because it doesn't require they reproduce a coding problem that they don't have the questioner's resources to reproduce and then expect them to solve that problem for the questioner. They are not going to spend a lot of time coding for someone that's copy & pasting and that questioner's having made no prior effort to learn how to program.

 

So you're saying intermediate->expert questions aren't meant to be asked on the forum? The forum is a place for beginners only?

Sound's perfectly reasonable.

Neither of my questions required any kind of reproduction. These aren't "Unity crashed, I can't fix it" or "Help make this thing do this for me". They are "How would you approach this situation" types of questions.

1. This is a question about the best approach to mipmap handling for a given setup.
2. This is a question about the limits of the Mecanim system.

I don't think my problem is a lack of effort. I'm not saying that just to defend myself, I honestly do not believe I haven't given this enough though or looked into it enough. I usually am able to solve issue on my own through a rigorous searching process, and when that fails I come here for help.

It feels like you think this is the standard "Guy just downloaded Unity and wants to have the forum community build a game for him" (which does happen a lot so it's a perfectly understandable misconsception to have) when it's actually a "Guy has been using Unity for 5 years, needs help approaching a new situation and is looking for experienced users who have faced this issue before to provide suggestions."

 

Or when someone is basically trying to get their script written for them. It doesn't happen that often but there was a user putting up threads "asking" for assistance but had put literally no effort into it themselves both prior to the thread and after receiving some guidance.

 

1. isn't a common question. As fast as things get bumped around here someone that actually knew about that sort of thing just wasn't around during the hour that you posted it and no one bother to dig into old posts looking to help some random soul out.

2. has a lot of words, its high level stuff, kind of hard to picture in your head and no pictures or references. I would assume it just didn't seem like an interesting post for people that knew about it to look into.

A lot of times you'll get a better response if your post is simple, to the point, targets a very specific thing or asks a direct question, and maybe has any pics/references related to the topic.

But, there's not really a secret formula. A lot comes down to timing and user interest. C# questions about accessing some thing in Mecanim may get answered because there is a large pool of C# users around here versus a high level topic about how to approach something.

The best I've found with dealing with Mecanim is just trying things out yourself and seeing if it works. Getting some other packs with existing controllers and dissecting them is helpful as well. Theres a lot of deep features no one even really uses or talks about that you just have to pioneer into using.

 

Sounds reasonable and I can definitely see what you're saying.

This has pretty much been my standard approach to just about everything in the coding world. The only issue is that trying things out on your own might get you something that works but that doesn't mean it's a good or optimized approach. I've been able to go back over the code I wrote years ago and bring it up to higher standards because I've worked in that particular area for a very long time. With things like Mecanim, I haven't really done any experimenting and I am more prone to making those kinds of mistakes.

It seems like my issue is just a combination of bad luck and prevalence of the issue being reported. I do feel, however, that the point of a forum like this is to house information about best practices and approaches to any given situation. That the forum acts as a place for those with personal experience in any given area to provide help to those new in that given area. To help them avoid common pitfalls and to help guide their approach in a manner that will ultimately result in their goals being achieved in the most efficient way possible.

It seems from what I've heard so far that it ultimately comes down to user interest and commonality.

 

I would say that's a fairly accurate statement. Its not that you shouldn't ask them, its that there are very few who will be able to answer them, and youll get tumble weeds on your posts (as you found).

There are some other things to consider though. Choice of sub-forum. You choose animation, which is probably the most suitable place, however with all the sub-forum dilution, you might find animation is a low traffic area. I NEVER look there personally.

As mentioned above by Lanefox. Its about time and place. There is a lot of noise on the forums and it would be easy for your question to go to page 2 in under a day. Most people wont look past page 1. You might pick the time when the person you need to answer your question isn't around.

I looked at your animation question. How I would probably do it is setup multiple states then jump straight to the state I wanted to play by name (animator.play / animator.crossfade) . Im not sure if sub-states can be created at run time.

 

Its frustrating sometimes, yeah. There really just isn't that big support group for some features though. Less common things will always be subject to users pioneering into them and finding better ways years later, like you mentioned. But we're all kind of in the same boat in that case so if your first topic fails, try digging into it and making a more specific topic later about trouble you're having with it. Even follow up with solutions - those are great threads to find.

 

you might be able to poke the unity guy in this thread for some clarity...

http://forum.unity3d.com/threads/go...de-and-not-to-a-specific-state-within.311676/

 

I've heard your complaint from others and myself too so it's not like I have learned what you are experiencing personally. And I've learned that venues such as this are really only practical for beginners that get their quick answers from professionals at work when they are not busy and feel like chatting. Nothing against anyone that didn't bother to or couldn't answer my questions.

Truthfully, personally I need to pick a Unity sub-forum and a sticky for my monitor as a reminder that each time I visit the forums I should try to answer the 1st 5 technical questions I understand and can offer a solution for. So I now have a sticky on my monitor. Thanks.

1. Question one is a case of wanting to discuss design that requires intricate knowledge of the details of a fast changing animation system that most people, as in 99% of them, in the forums don't need to involve themselves with learning. So really, relative to the forums that is an advanced design question that should be posed to a peer that you work personally with a similar and more of the needed experience if you have that type of luck. And it comes down in the question you are asking the reader to design for you. I think you're personally quite capable of the design yourself and most readers will detect that too and recognize you are wanting to chat and maybe team up with a similarly experienced developer and if they lack that or don't have interest in that they will ignore the question. Nothing wrong with your question but it's not a surprise that the forums failed to find a match. You should try a local Unity group.

2. Question 2 is you state the problem and lack of solution to it and that's it's been discussed multiple times in the past in the forums. The problem is a characteristic for the solution chosen. To get rid of the problem you need a different solution. And so a new design. Again, it's a similar concept to the 1st question. The forums won't find matches to such relatively advanced questions most times and so you should try and join a local Unity group if you are lucky enough to live near one or a local programming group and introduce the members to Unity if there is not a Unity group nearby.

It comes down to we all need to develop own skills to better help each other and if you're an advanced user relative to the forums well you are fortunate.

 

I very much appreciate the help James, that's actually something I was considering as well. The truth is that Mecanim (at least from what I can gather from my own experience) is not tailored for dynamic game development, but rather is more statically-oriented. Rather, that Mecanim is pretty much meant to be handled directly from the editor on a specific this-for-that basis. The legacy animation system is probably my best bet when looking for a modular workflow. It just means I'm going to have to create something special to handle it, which I'm not worried about really.

That makes sense. Is there some kind of information repository (aside from the Wiki which hasn't really been touched since 2013) where people can post their own experiences working with Unity to achieve specific results that otherwise aren't outlined in the documentation?

I've done a lot of work getting Unity to do things it wasn't specifically designed to do and I'd be more than happy to post what I've learned and how I learned it. It just seems like there are a lot of different avenues and sub-sectors that aren't necessarily as intuitive as they may seem to others who have used them for a long time.

It would be nice to be able to archive information and solved issues in a master thread list somewhere, rather than having people use the search function only. Especially considering the fact that thread titles rarely depict with accuracy the issue being handled.

 

I have specific relationships with a couple of other experts in the scripting forum. We all hang around to discuss advanced topics. If something comes up that's beyond me, I can ask one of those guys specifically. And generally they will answer.

It's also worth pointing out that to get these relationships you have to answer far more questions then you ask.

 

I really wanted to do something dynamic with hardkour, but every time I looked at mecanim my brain imploded. I was very resistant to wanting to use it. Took me a while to even figure out I could jump straight to an animation via code (without using flags).

I suspect you could build your own dynamic system though if you had references to all the animations, and rather than using substates, you managed the transitions yourself through plays/crossfades and merged that with co-routines/events to set the timing. Then you would only need to know the sequences (animation name + time)... generally that's what a lot of substates are.

 

and the docs are a bit meaty :/ i really should do an unofficial tutorial.

to OP: pics help a ton. if you have a .gif of your issue, even better. If its something that you can find in the docs, even if you have to read a little, I might just throw a link at you and say "read this, please"

 

A picture is worth a thousand words. I think, any question about any graphical issue must contain at least a pair of screenshots.

I've read your post about mipmaps and can't say for sure that I've understood it completely. Maybe it's just because English is not my native language, maybe it's because I have no idea of what "isometric distance" is, maybe it's something else. With the lack of visual reference I can't visualize the problem using the textual description only. Usually I skip questions that I'm not sure I fully understand, even if I probably could answer them.

If the question was "My textures are too blurry at the distance but I don't want to switch the mipmaps completely off. What could I do?" then I would suggest four possible directions to look into:
1) Enable anisotropic filtering if it's not enabled yet.
2) Texture.mipMapBias
3) Custom mipmaps
4) Custom shader that can control mipmap levels.

 

The docs could definitely do with some love.

 

I may have been lucky but so far people on here have been great at pointing me towards possible solutions. I try, normally not to successfully, to keep it as short as I can while providing the info needed. I also try to include what I've tried or already looked at as well as any images I can find that might help so anyone reading it will know I've tried to resolve it myself before seeking help. I've also seen people just saying that they've got something working a certain way but asking is there a better way to do it, which again shows that they did make the effort.

All of that is pretty much what you are doing so the only other factor could be that I'm still looking at what I think many people find as the easy stuff so the questions haven't been that specialised (yet) whereas yours might be towards that intermediate level as others mentioned.

 

I think it's awesome you have that, I imagine you've found some invaluable information through those relationships. Honestly, the only thing I have of that type (at least at a local level) are fellows from Blizzard who have dabbled in hobby projects and just overall messing around with Unity, but no one who is actually experienced enough with it to create that particular kind of group dynamic.

I think it's worth noting, however, that those relationships, while very helpful, are only helpful to yourselves versus the whole community. This is why I'm hoping there is an information depot (again, not the Wiki) where you could post your findings so that other people who come across the same situation would benefit. If that doesn't exist then perhaps it would be beneficial to us all to look into creating something like that? I'm game.

I think that is possible, yes. However, at that point isn't the main benefit of Mecanim lost? From what I understand, Mecanim's main feature is that it has the visual component that handles the blending/state changes in a more visually pleasing flow. If you bypass all of that then what are the benefits over just using the legacy animation system? I'm also pretty sure that Mecanim has a stricter access restriction on member functions for accessing/editing variables such as clip weight/speed/layer/blend transitions. It may be worth looking into deeper on my end though, thanks for the tip!

That's what I keep hearing. Trouble is conceptual questions aren't very easy to take screenshots of. You'd most likely have to draw some kind of diagram. Given how varied the human visual thought process is, that can be more of a hindrance than a benefit at times.

I appreciate you taking time to try and answer my question. Here is what I discovered in my own research/experience:

1. Anisotrpoic filtering, whether set to 0 or 16 is not going to solve mipmapping issues where sharp paneling lines are concerned. Aniso is mainly geared toward maintaining texture detail at steep viewing angles.
2. This is a possible fix, but only on a local level. Isometric point of view means a camera position that is high above the player. Games that use this are: Diablo, Fallout, Arcanum, Neverwinter Nights, etc.
3. Another good idea, however it is still just a local fix to a global problem. I need all mipmaps on all textures to behave a certain way.
4. To my understanding that isn't possible. At least from the several threads I've read that talk about controlling how mipmap level usage, it isn't something you are able to control yourself.

For that particular problem, I still don't have a good answer. I'm hoping to avoid changing every single texture in the game to DDS to fix this. The optimum way to handle it (and I don't know if this is possible now verusus Unity 3 which is the last relevant thread I could find on the subject) would be to set a minimum mipmap level for textures. Essentially, be able to tell unity "Only go to miplevel 1/2/3 not 4/5/6 which are at resolution (128, 64, 32) which completely destroy fine detail.

Whatever method I find out, I'll make sure it's publicly accessible to the community.

I do appreciate all the help. I was beginning to think you guys didn't like me

 

Sorry, I think I mislead you. My network of relationships is with other users I've met on the forum. We have experts in pretty much everything here. There is the guy who wrote an entire frame work to go with unity. There is the C# expert from building windows aps. There is the guy that works for Disney and has a decent feel for the industry. There is the guy that does Unity based simulations unrelated to games, who knows the other side of the industry. There are open world graphics artists who can tell you the exact limits of the engine. There are pixel artists. There are a ton of generalists. There are the community managers for all things licensing. There are various Unity devs for bugs.

I guess my point was that sometimes knowing who can answer, and tagging them in your question, goes a long way. Its as simple as @McMayhem

On an interesting side note, I've only met one individual in person that I'd previously known on the forums. Was at a party, totally unrelated to Unity, and someone walked up to me and introduced themselves and asked me if I was the same guy who did the YouTube tutorials on the UI system. That was kind of creepy and cool at the same time.

All of my technical posts are available on this forum or answers. I make a habit of directing anyone that comes to me for personal help to the appropriate forum section as well. The only conversations I keep private are the ones relating to specific business work. If someone is paying me to work on their project, or vice versa, the details stay out of the public forum.

 

Check this. However, I'm not sure it works on all the platforms. I made a little change to the standard surface shader template and it works:

Code (csharp):

1.         void surf (Input IN, inout SurfaceOutputStandard o)
2.         {
3.             float mipmapLevel = 5;
4.             float4 s = float4(IN.uv_MainTex, 0, mipmapLevel);
5.             fixed4 color = tex2Dlod(_MainTex, s) * _Color;
6.             o.Albedo = color.rgb;
7.             o.Alpha = color.a;
8.         }

The mipmap level is hardcoded, but you can try to calculate it on the fly from the distance to the camera.

From watching one of the videos from one of the recent Unites this year I remember some guy asking one of the developers to implement this feature, and the answer was that they could do it pretty easily with close-to-zero time and effort. I'm still waiting for it.

PS

with images please

 

What's the bug report? If you included your views and concerns then my doc team will know it's work that users want.

 

There's no specific answer to that type of question. Even though they're not technically difficult they're potentially very time consuming to write about because it's not an "answer", it's a "discussion".

There's another way to view it, which is that the answer is almost always "do some experiments to see what works best in your scenario". Chances are that nobody who happens along will have had an exactly equivalent scenario in the past to give you a specific and detailed answer. (And even if they did... would you take that random claim from someone arbitrary on the Internet as gospel without verifying it for yourself anyway?)

Also, while you seem to be addressing most of it anyway, see the link in my sig. It was a huge help to me in my early days on the 'net.

 

Me too. I periodically get PMs asking for help, and my response is always to start a thread and I'll answer it where others may also benefit from it.

 

You could always bait with an easier problem. List 2 issues. The first, 90% of the community can solve. The second, your actual problem. More views & replies for #1 will get you more views for #2. You keep a small and very strictly on topic discussion about #2 and your discussion will either bring you closer to solving the problem yourself or keep the thread alive and visible long enough for some expert to notice it.

The docs are America. Some are richly detailed and have more than they need, some have almost nothing.

I'm surprised that doesn't get threads closed. Maybe there should be a sticky of questions asked that could be answered by the docs. But that'd probably be too big of a post... people should just read the docs.

 

But this is a forum. Isn't discussion the primary function? I feel like UnityAnswers would be the place to get "answers" to specific issues, whereas this would be the place to "discuss" things such as best practices and approaches.

Maybe it's in the way I'm presenting this but I'm not looking for an "end-all be-all" type of answer. And while someone may not have the exact same experience with the situation, they may indeed have been down a similar enough road to know how to help. That's kind of the key to human innovation.

You're right, I wouldn't take a claim at face value, but I don't think that's what this particular conversation is about. If someone where to say matter-of-factly "This is how you do it." versus saying something like "You might want to try this." the reaction to both statements is the same, you try what is suggested.

I still feel like a lot of these issues can be solved by some kind of better archival representation of issues presented both here and on UnityAnswers. Even something as simple as a sticky thread organized into categories for each sub-forum that have relevant threads linked. I feel like that would be much more helpful than relying on the search function.

Either way, I feel like I have a much better understanding of how things work around here and across the various support portals designed to help Unity users. I very much appreciate the added information and suggestions from the community and will certainly take it to heart.

You have my thanks.

 

A lot of questions require a "deeper than surface level" knowledge of the personal projects people are working on. And, unfortunately, this takes some time (and work) to learn and understand how a person has their project set up, and what their goals are -- in order to answer what seems like even simple questions. And that's hard to do over forums.

Most people are not willing to spend that extra amount of time, since a lot of the users on this forum are busy with their own projects etc.... It's not really that these people are bad (I'm included in this), it's just kind of the nature of the work involved. Of course, there are exceptions, and it's really cool when they are able to help.

Personally, I have trouble conceptualizing and understanding what people are asking. I really want to help people more, but simply cannot (in most cases). I've spent some time perusing the "Scripting/Coding" subforums looking for questions I might be able to answer, and 95% of them I simply don't have the level of expertise needed to answer their question properly or, in some cases, I simply can't conceptualize their question.

So, generally, I only have the amount of time and expertise needed to answer some of the most basic questions. All this being said, I wish I could help more. And I enjoy being able to help beginners, when I can. But it's not all so simple.

Edit: Also, I forgot to add that I really think breaking up what used to be a small number of sub-forums, into a whole bunch of different ones, actually did dilute the number of people who would view your question. And so, in a way, the re-structuring of the website actually did kind of hurt beginners in a way.

 

For what it's worth I haven't really managed to get much or any help from these forums since I began using them. I gave up asking questions in the end. I've recently asked a couple and these were largely unanswered. Generally, as a moderator, I do see most people asking the most basic of questions over and over and over again, that are easily answered with a trivial google search, while intermediate and advanced questions are ignored.

Do still have quite a few smart shader guys lurking around though so perhaps it's just the manner questions are asked!

 

I thought you knew everything already Hippo. I am surprised by this lol.

 

Literally what do you mean by not getting your questions answered? I post a question and it gets answered within an hour! A strong community is the strength of unity! Are you getting irrelavant answers or something?

 

Im not fussed enough to write a bug report, but following in the original context of this thread, here's an example with zero information other than the function header...

http://docs.unity3d.com/ScriptReference/Animator.Play.html

Sure its generally pretty obvious what the function does, but its not hard to dig through other topics and find similar stuff.

I started writing an editor script recently. That was a brain imploding task with the current docs. Managed to make good progress eventually, but that was more about stumbling on snippets around the net. It sure wasnt from the docs.

 

Did you bother reading any of the other replies? Doesnt seem like it.

 

You received straightforward answers because your question was very simple and asked in a very active section. His questions are far more difficult to answer and the appropriate forum sections are likely less active.

 

Wouldn't that just be... a copy... of the docs?

 

Absolutely, and ideally an answer will ultimately become a discussion. It's just that often those discussions are started by something quite specific, mostly because if something is really specific then it's easy to get started on. For a broader question you have to think about where to start with your answer, and that extra layer of effort can be offputting.

 

As awful as it sounds, in my "off work" hours, I get lazy too

 

Lazy? Or pragmatic? We can't all help everyone.

 

Yes.

I'm so lazy that I do things right the first time, because it's too much work to fix things later. Be efficient. Be lazy.

 

I understand why you think this, but I can factually say there are people within this exact thread that I have asked offline questions - and they have given some great advice - that have lead to better understanding of Unity processes - which have in turn benefited other people here on the forum.
I enjoy helping people out when I know a solution to an issue they are having. I also like directing people to the learn section which seems like an invisible section of the forum to A LOT of people who post questions, especially in the animation forum, which I frequent quite often for selfish personal knowledge gaining reasons.

I also looked at your mecanim question and gave it some thought before passing on it, because the concept you detailed is really a unique setup - I think. And pardon if this is direct, but the concept you detailed is beyond intermediate imo and you noted in your post you are just beginning with mecanim, so it really is a question that the answer to is - gaining experience with the system will provide those answers.

I actually started watching your mecanim thread though (selfish reasons) because I was interested in any answers that were posted and also interested in seeing if you would post up the solution once you had worked through it.

As Lane kinda said - I also find threads, where the OP comes back to and answers his own questions with a proper solution and reasons other options didn't work some of the most useful threads on here.

All questions I've read of yours are very well structured and thoughtful.

 

Just left out the complicated ones