Search Unity

  1. Good news ✨ We have more Unite Now videos available for you to watch on-demand! Come check them out and ask our experts any questions!
    Dismiss Notice
  2. Ever participated in one our Game Jams? Want pointers on your project? Our Evangelists will be available on Friday to give feedback. Come share your games with us!
    Dismiss Notice

UNITY 3 beta preview

Discussion in 'General Discussion' started by nikemerlino, Jun 23, 2010.

  1. Alex Mat

    Alex Mat

    Joined:
    May 14, 2010
    Posts:
    177
    Does the new PhysX update contains particles, fluids and ability to make destructible environments?
     
  2. Peter G

    Peter G

    Joined:
    Nov 21, 2009
    Posts:
    610
    No, sorry.
     
  3. Dreamora

    Dreamora

    Joined:
    Apr 5, 2008
    Posts:
    26,602
    particles always were physx to my understanding from the features offered.

    and fluids will never happen if we talk about the fluid simulation fluids as those require gpu acceleration that will neither ever happen due to its lack on osx and the requirement of the physx driver installation to even use a game.

    the forcefield fluids can be done through constant force blocks already in U2
     
  4. Yann

    Yann

    Joined:
    Oct 20, 2007
    Posts:
    431
    Talking about physics : does anybody know what happened to the DragRigidbody script in Unity 3 ? is it supposed to be used in a new way (different from simply attaching it to a camera) or does it need to be updated to work as it used to ?
     
  5. andeeeee

    andeeeee

    Joined:
    Jul 19, 2005
    Posts:
    8,768
    What is going wrong with DragRigidbody when you use it with Unity 3?
     
  6. Yann

    Yann

    Joined:
    Oct 20, 2007
    Posts:
    431
    @Andy : I create a new project with a cube on a plane, add a rigidbody to the cube and the DragRigidbody script to the camera. Press play, try to drag the cube : nothing happens. This is on Mac, btw.
     
  7. bryanleister

    bryanleister

    Joined:
    Apr 28, 2009
    Posts:
    130
    I thought you just applied the script to the cube, not the camera.:)

    Bryan
     
  8. Yann

    Yann

    Joined:
    Oct 20, 2007
    Posts:
    431
    @Bryan : this works too, but you don't need to apply it to every draggable object : as long at it is attached to one object at least (incl. a camera or an empty object) it will be sufficient to drag all the objects of the scene that have a collider and a Rigidbody. But whatever combination I try in Unity 3, I can't get it to work...
     
  9. Robert G

    Robert G

    Joined:
    Nov 17, 2009
    Posts:
    233
    For me this is still working in B6, but I'm on pc.
    Hope this helps.

    Regards, Robert
     
  10. andeeeee

    andeeeee

    Joined:
    Jul 19, 2005
    Posts:
    8,768
    DragRigidbody is working for me on a Mac using b7. If you're finding it doesn't work under b7 then please file a bug report (menu: Help > Report A Bug).
     
  11. Yann

    Yann

    Joined:
    Oct 20, 2007
    Posts:
    431
    Thanks, Andy and Robert. I will try a fresh OSX and Unity install on a second drive, and file a bug if it persists.
     
  12. Yann

    Yann

    Joined:
    Oct 20, 2007
    Posts:
    431
    OK, just filed a bug. On a clean OSX install with only Unity 3 beta 7, it now seems to work at first sight - but in fact it behaves strangely : you almost can't move objects vertically, as if they were too heavy - and sometimes they move in unpredictable directions. If you lower their mass by 1/2, you can have them leave the ground for a while - but after a few movements you cannot anymore. I still don't know if this is a real bug, or just me not understanding some new physics functionality in U3 :)
     
  13. taumel

    taumel

    Joined:
    Jun 9, 2005
    Posts:
    5,293
    I'm curious about how the process of reporting and fixing bugs in Unity3 works.

    I think it's wise to work with a limited testing/reporting audience in the beginning and to enlarge that base once things got more stable. I wonder when these points are meant to be set because i've already stumbled over quite some issues which never have been mentioned in the release notes so far and still exist...
     
  14. jashan

    jashan

    Joined:
    Mar 9, 2007
    Posts:
    3,166
    That's the way it's done. However, obviously not everything is perfect ;-) ... IIRC, "public testing" (for the people who have pre-ordered Unity 3) started around beta 4, so there already were quite a few issues fixed. But hey, Unity 3 is an awesome and huge update ;-)
     
  15. taumel

    taumel

    Joined:
    Jun 9, 2005
    Posts:
    5,293
    I also thought that they do it this way but i was just wondering because certain bugs aren't getting fixed since several updates which could mean that a) no one else finds them - some people have a talent for finding bugs or a good eye on quality - , b) isn't motivated enough to report them or c) they are more listed on a low fixing priority.
     
  16. jashan

    jashan

    Joined:
    Mar 9, 2007
    Posts:
    3,166
    I guess in the end, it's a matter of priorities. I'm pretty sure that every bug gets reviewed and assigned but it may sometimes take a while for even that to happen; and then it might take another while for the bug to actually get fixed. I remember one bug I filed which took 2 betas until it got fixed - but hey, meanwhile a whole lot of other bugs got fixed, so it didn't seem like nothing was done ;-)
     
  17. Don-Gray

    Don-Gray

    Joined:
    Mar 18, 2009
    Posts:
    2,278
    I've reported at least two (maybe more) bugs I found here that the Unity techs were not able to reproduce
    and since I wasn't able to send a project that had the bug it was considered closed.
     
  18. jashan

    jashan

    Joined:
    Mar 9, 2007
    Posts:
    3,166
    Basically, a bug that cannot be reproduced isn't really a bug. The time required to figure out what's going on usually isn't worth it. I don't know how UT handles this but usually, when you can't reproduce a bug, the best thing is to ask the customer to provide a way to reproduce it ... if that doesn't work, what should you do?
     
  19. Don-Gray

    Don-Gray

    Joined:
    Mar 18, 2009
    Posts:
    2,278
    Oh I understand it, hopefully they are just something in the mix here, otherwise others should report them as well.
    Some have been reported and some haven't but since I haven't been able to get past them (haven't tried in a while)
    I just stopped trying to use the features (mainly beast and umbra). I'll try again at least when the full release comes,
    hopefully things will have ironed out by then. Could just be me, I'm not the most experienced at everything in Unity.
     
  20. taumel

    taumel

    Joined:
    Jun 9, 2005
    Posts:
    5,293
    First of all that's technically wrong, as you most probably know on your own.

    Secondly if you care about the quality of your software, you also will try to hunt down such bugs. Hard to track down bugs are business as usual. Also it strongly depends on how serious the bug is. If you have written software for a longer period then you should have learnt that everything which does not work, will come back just like a bumerang.

    I've just read here on another thread, that HiggyB expects Unity3 to be released before Unite. Considering that Unity is still Beta and there still are quite some untouched issues, it might be wise preparing yourself for another typical, buggy, Unity release. You know, a modern x.0 release.
     
  21. jashan

    jashan

    Joined:
    Mar 9, 2007
    Posts:
    3,166
    Nope, I do professional software-development for a living, and a bug that cannot be reproduced simply won't be treated with particular priority in most cases. There are, of course, exceptions: I agree that it depends on the severity of the bug and that there are cases where you'd try to create a repro-case on your own (or have the Q/A team do that kind of research). But in many cases, that's just a waste of resources that no one has to waste.

    In many teams, the developers are not even bothered with bugs that the Q/A-team isn't able to reproduce. And that is for good reasons: Dev-time is costly, and no one wants to waste money ;-)

    If you have time for it. ;-) Think about it this way: You can spend a day to fix 10 bugs that can be reproduced and usually happen a lot (that's what makes them "reproducible"), or you can spend two days trying to set up a way of reproducing the bug ... and usually you have no idea whether or not you will have success in doing so.

    In purely test-driven development, you won't even think about fixing the bug unless you have written a test-case that reproduces the bug. So, in that kind of development-process, you don't only want to have a repro-case, you want to have a Unit Test that reproduces the issue. IMHO, that's usually taking it a bit too far - but there's beauty in it because once you have that test-case, that same bug will never ever leave the dev-team again (kind of like compilation problems ;-) ).

    I do make a difference between "can't be reproduced" and "hard to track down". Of course, there's a lot of issues that are hard to track down and that is in fact "business as usual". But I was talking about bugs that cannot be reproduced. That also happens frequently but I'd treat it quite differently (a bug can be hard to track down even if you have a perfect repro-case).

    And that is, by definition, the case for a bug that can be reproduced. When a bug "comes back just like a bumerang", there certainly is a way to reproduce it. So far, I've never got any bug I sent to UT confirmed that I couldn't reproduce myself. If I can't reproduce it - how should they reproduce it? And if they can't reproduce it, how should they fix it?

    And I'm very glad they seem to treat such bugs this way because it means they're not wasting their time and money (which in the end is - at least in part - the money I pay for the licenses) to hunt for ghosts.

    Of course, there are cases where you can do code-review to figure out and fix such issues - there's no rule without an exception. I guess that's the most tricky part for Q/A-teams: Deciding which bugs to pass on for "code-review" to the devs even when they can't reproduce the issue.

    And here "numbers" make sense (which is the reason why I *do* file bugs report even when I can't reproduce them): If it happens to a lot of people, it's worth the time. If it's just happening to one person, it's sad but most likely some sort of very esoteric configuration issue and as such a paid-support case (work with the customer to fix their problem and charge them for it ... business as usual ;-) ).
     
  22. TwiiK

    TwiiK

    Joined:
    Oct 23, 2007
    Posts:
    1,656
    Not sure why you have this attitude against Unity, but if those leftover issues are infact hard to reproduce bugs then it's perhaps wise to release version 3 and let a much larger user group try to reproduce them.

    I for one would not complain about a few bugs if I got the product much earlier. :) That goes for you as well, Blizzard. Diablo 3 tsk tsk...
     
  23. Dreamora

    Dreamora

    Joined:
    Apr 5, 2008
    Posts:
    26,602
    Its likely not against UT and Unity but bad experience from Unitys once main competition, which is in a state where nobody would recommend you to invest a single buck in it (Torque - Brett Seyler who joined UT months ago was before one of the leading forces over there and should know what happens if you put push out over quality especially if there are competitors around that aren't far enough behind you to be ignored ie UDK and Shiva)
     
  24. taumel

    taumel

    Joined:
    Jun 9, 2005
    Posts:
    5,293
    @Jashan

    >>>Nope, I do professional software-development...

    I know and that's the reason i wrote it this way but the way you treat bugs depends on quite some variables but it's wrong stating that a bug isn't a bug just because you don't have the ressources, philosophy or the motivation to track it down.

    You can say, in our project, at our company we treat it this way due to these reasons but it doesn't make the bug dissapear, actually it's still there and in more critical client driven projects or projects which have to be secure, you also treat it more seriously. Actually it's not about quantum physics. ;O)

    Anyway i hate typing longer posts on these iPhones. Although i don't have big fingers it kind of upsets me. This is not perfect, can you hear me Apple?!

    What i would like to see, after five buggy Unity years, is something which finally works, something were i don't have to be afraid when sending it to someone else or something that screws up when your hitting it a little bit below the surface. So far Unity has been unable to deliver such a quality and in my opinion they will loose ground in the future if they don't start seriously caring about this by now, because they will see more competition from above (traditional pro tools which come with more attractive licences) as well as from below (cheaper tools, established tools or open standards).
     
  25. jashan

    jashan

    Joined:
    Mar 9, 2007
    Posts:
    3,166
    I agree on that ;-) I guess saying "a bug that cannot be reproduced is not really a bug" was stretching it a bit, apologies for that ;-) basically, from the developer perspective, when a bug is closed with "unable to reproduce", it's gone - and we have many cases where they stay "gone" and never pop up again (even though some do).

    In fact, we do have quite a few bugs that can easily be "discussed away" and the reason for that is not that we're ignoring them but frequently because customers decide it's not a (real) problem for them after getting a proper explanation of why the system behaves that way. I guess the truth in those cases is that it wasn't a bug in first place ... but again: many many variables, and it only works when they have the correct values ;-)

    Argh, yeah ... some of the UI of the iPhone is a joke. Typing, for instance, just doesn't really work (especially the auto-correction wastes a LOT of my time ... but the small keys do as well, and I also don't have big fingers ... but I also don't have the elven fingers that keyboard was designed for. ;-)

    Sorry for being a little off-topic ... but ... I guess bugstuff is kind of on-topic in a beta preview thread ;-) ... and the iPhone stuff was fun ;-)
     
  26. jashan

    jashan

    Joined:
    Mar 9, 2007
    Posts:
    3,166
    What I see is that UT has obviously staffed up QA quite a bit. It's no longer a zero-person show (QA done by the devs), and it's also no longer a one-man show, either. ATM it's more people than I can easily remember the names of (I'm not trying hard, though) and my feeling is that they are treating the bugs in a very professional manner. Whether or not that will result in the quality you are looking for will have to show at release day and ... to be fair, let's say .1-release day (even with a large testing-group, with a system as complex as Unity is, issues can be expected to pop up only after release ... and yeah, you simply can't fix each and every bug if you want to release some day).

    UT certainly is an extremely ambitious team and sometimes that kind of ambition might result in more bugs than what everyone would like (but also a lot of pretty cool features that I guess not everyone desperately needs but those that do will appreciate them a lot). As far as I can see they're doing their best to improve QA while still keeping up significant pace with new features.

    A few bugs that kind of were showstoppers for me in 2.x and weren't fixed over a couple of minor releases "magically disappeared" in 3.0 beta X, without anybody even mentioning it. And I certainly liked that ;-)
     
  27. taumel

    taumel

    Joined:
    Jun 9, 2005
    Posts:
    5,293
    @Jashan
    Hey, no need for excuses, it's just a discussion. :O)

    I ended up turning the auto correction off which enhanced it significantly. Entering text also is more buggy in horizontal mode as soon as you come across the scroll-field-border of a textfield. I also wonder why you can't alter the split screen on a touch behaviour in order to better focus or enter stuff. Oh well and the selection can be a pain as well. ...

    I also miss those elven fingertips although i have more feine schmale Finger instead of Wurschtfinger or Pratzen but it's just not enough for any device smaller than an iPad. ;O)

    Regarding Unity:

    I will give my final verdict once it is out but seeing what has been fixed in the Betas so far and what has not, the time frame, what features they focus on, the way they are beeing implemented, what serious/crtitical stuff has been left out again, ... all such aspects make me tend to believe that they are still more kind of dunno, a teenager combo? and haven't really understood a significant part of what would be important.

    As i said in another thread already i was right about many aspects which they didn't want to do in the beginning, some of them made their way into the product through the years, some are still not in there. Maybe all will be fine in Unity 6.3.1 but why not doing it right from the start for a change?

    I would prefer beeing optimistic but i also know about the power of ignorance and the need and the importance of making your own mistakes. But sadly, if you've already been there, it's just lost time.

    Wie war das nochmal? Der Mensch wählt gerne erst dann die richtige Option, wenn keine Alternativen mehr existieren.
     
  28. Don-Gray

    Don-Gray

    Joined:
    Mar 18, 2009
    Posts:
    2,278
    @taumel
    I too refer to be optimistic and will remain.
    Though they may seem like a teenage combo in some respects,
    it is in retaining some of that that will keep them from dying a corporate mindset death, IMO.
    But, of course both sides (corporate and youthfulness) need a continual examination for things that are both good and bad.
     
  29. jrDev

    jrDev

    Joined:
    Nov 5, 2008
    Posts:
    1,701
    I have gotten a response from the Unity Team about the FBX jacked up problem which they say is caused by the FBX Importer which failed to import the t-pose/reference-pose correctly. I'm using FBX 2010 and Im not sure why this is exactly happening since Unity 2.X it worked perfectly.

    Oh well, just hope it gets fixed because its definitely leaving me peeved.
     
  30. taumel

    taumel

    Joined:
    Jun 9, 2005
    Posts:
    5,293
    @Don Gray
    Yes but at some point you should start feeling responsible for your work and regarding software this also involves a working product. If you want to make proper usage out of your water bucket, no one cares about the colours on it as long it leaks water through many tiny holes. The priorities should be rather obvious then.
     
  31. BearishSun

    BearishSun

    Joined:
    Dec 1, 2008
    Posts:
    175
    Are skyboxes bugged in Unity 3 b7? They seem to render in front of most geometry, instead of behind. I tried it in completely isolated scene with 1 camera, 1 skybox and 1 cube. Cube doesn't render if cameras ClearFlags=Skybox, renders fine if ClearFlags=SolidColor.
     
  32. Don-Gray

    Don-Gray

    Joined:
    Mar 18, 2009
    Posts:
    2,278
    Working in my upgraded (2.6 to 3 b3-b7) project it seems normal.
     
  33. BearishSun

    BearishSun

    Joined:
    Dec 1, 2008
    Posts:
    175
    Out of about 10 scenes in my project, the skyboxes worked as they should in only one of them.

    Even new simple scenes as I described don't show proper skybox behavior, only that single scene.

    Trying to figure out exactly why it happens before I file a bug report.
     
  34. Don-Gray

    Don-Gray

    Joined:
    Mar 18, 2009
    Posts:
    2,278
    All the scenes in my 5 scene project work normally here.
    Not trying to offend, just reporting.
     
  35. BearishSun

    BearishSun

    Joined:
    Dec 1, 2008
    Posts:
    175
    The only thing I changed between these two images is cameras Clear Flags from Skybox to Solid Color. You can see cube being rendered only in second case, although it's right in front of the camera.

    http://i54.tinypic.com/28jjxxl.png
     
  36. Yann

    Yann

    Joined:
    Oct 20, 2007
    Posts:
    431
    Now I can confirm that the Dragrigidbody problem I described earlier is caused by a bug : I have reproduced it on my second Mac, compiled two versions of the same basic test scene (with Unity 3 beta4 and beta7) : first build works as expected, second one doesn't work at all. Given the fact that I made my tests on two very different machines (an iMac 24 alu with Radeon videocard and a Mac Mini with NVDIA), I suppose other Mac owners should logically notice the same problem.

    [EDIT] Fixed in Unity 3.0.0f1 :D
     
  37. BearishSun

    BearishSun

    Joined:
    Dec 1, 2008
    Posts:
    175
    I'm also having an issue with the Pro water script/shader. In some scenes (haven't isolated which ones yet), if you lift the water plane high enough, as soon as its starts intersecting geometry Console goes wild with bunch of !IsNormalized(normal) errors that seem to be caused by camera from the Water.cs script (used for reflective or refractive cubemap rendering).
     
  38. MFUH

    MFUH

    Joined:
    Sep 9, 2010
    Posts:
    2
    Does anyone know how to fix this?

    transform.rotation assign attempt for 'SceneCamera' is not valid. Input rotation is { NaN, NaN, NaN, NaN }.
    UnityEditor.DockArea:OnGUI()

    The scene view is totally block! it doesn't let me work. :(

    Thanks
     
  39. MFUH

    MFUH

    Joined:
    Sep 9, 2010
    Posts:
    2
    For anyone having this issue,

    just revert the layout settings to factory settings!
     
  40. Bampf

    Bampf

    Joined:
    Oct 21, 2005
    Posts:
    369
    Unity 3 b7.

    I converted a iPhone 1.7 project over for the first time. Several things are giving me problems:

    1) Conditional logic like:
    #if !UNITY_IPHONE
    is firing now. Does this mean that the only way to test iPhone conditional logic is via Unity Remote, or a build to an actual device? I suppose I could define that code myself.

    2) iPhone iPad resolutions are not included in the Game View dropdown.

    Both of these are related to the merging of the iPhone and regular Unity IDEs, it's safe to say. Neither are showstoppers, thankfully.

    **EDIT**
    Ah, never mind, I didn't realize I needed to click the "switch platform" button! The project imported as a Mac/PC project, and confusingly, the iPhone project type was *selected* for me, but I needed to click the switch button to actually switch.

    I'm sure I'm not going to be the last person to be confused by that; ideally the import would do that extra step, but I figured it out.

    Now I've got some *real* bugs to deal with. Maybe more questions soon. :)
     
  41. Dreamora

    Dreamora

    Joined:
    Apr 5, 2008
    Posts:
    26,602
    1) conditional logic works on the device


    2) Yes they do. all 3 actually including the new 960x640.
    switch the project to iphone / ipad on your osx unity 3 install build settings
    with windows you can not do any iphone dev at all
     
  42. Bampf

    Bampf

    Joined:
    Oct 21, 2005
    Posts:
    369
    Thanks, dreamora. It's great to know I would have gotten an answer from the forums if I hadn't happened across it myself.
     
  43. Dreamora

    Dreamora

    Joined:
    Apr 5, 2008
    Posts:
    26,602
    hehe :)
    you happen to have it edited while I typed the answer
     
  44. larsbertram1

    larsbertram1

    Joined:
    Oct 7, 2008
    Posts:
    5,711
    working with umbra and beast i found it very unhandy to have just one checkbox ("static") to make the object being affected by occlusion culling and/or lightmapping. i might like to have objects added to occlusion culling but not to lightmapping (due to memory reasons e.g.), but there is no way to do so, is it?

    and what is about combined meshes (using the combine children script)? objects wqhich should be combined must not be check as "static" otherwise their meshes will disappear on runtime. any chance to lightmap those obejcts anyway?

    lars
     
  45. antenna-tree

    antenna-tree

    Joined:
    Oct 30, 2005
    Posts:
    5,325
    Since Unity 3 automatically batches static objects, using the combine children script isn't really necessary anymore AFAIK. But, if you feel it necessary, you can always mark them as static, lightmap, and then mark them as non-static again. Admittedly I haven't tested this though as the combine children scripts were abandoned for our new 3.0 demo.

    PS - I do find the single "Static" check for OC and lightmapping to be limiting as well. A few devs here agree, but it was just too late in the cycle to change this for 3.0. Perhaps this will be changed in a future release.
     
  46. Dreamora

    Dreamora

    Joined:
    Apr 5, 2008
    Posts:
    26,602
    actually you can change it antenna tree

    B6 or B7 introduced two new settings to the player settings, to enable static batching and dynamic batching.

    Its not as straight forward as being able to set it on a per object base but you can disable the static batching without changing static on the whole scene again.

    if you use the combine script you will likely want to do it everywhere or nowhere and thus this setting would likely be sufficient
     
  47. antenna-tree

    antenna-tree

    Joined:
    Oct 30, 2005
    Posts:
    5,325
    I was more referring to Occlusion Culling and Lightmapping both using the Static setting to be included in the lightmap calculation and and be marked as occluders in an OC bake. Sometimes this is not desired.

    But yeah, I believe if you want to manually combine your meshes at runtime instead of Unity doing it for you just turn batching off.
     
  48. Dreamora

    Dreamora

    Joined:
    Apr 5, 2008
    Posts:
    26,602
    Agreed.
    That way around you still need an editor scripts to disable all you don't want to be used in lightmapping.

    But that the opposite is already doable is surely some pretty good help for those who do their merging manually or have optimized assets and don't want to take the overhead from static batching
     
  49. larsbertram1

    larsbertram1

    Joined:
    Oct 7, 2008
    Posts:
    5,711
    hmm, i am a little bit confused right now. whenever i combine meshes the stats window gives me a lower number of draw calls. that is what i am after.
    but how about static or dynamic batching? does it only work on builts? but if so what is about the stats window? does it tell me the actual number of draw calls of the final built?
    right now i am using both: combine children on similar objects and batch static and dynamic objects in the player settings. if i get you right this is not a good idea…
    some words about optimization in the docs would be nice.
    thanks a lot.
     
  50. Dreamora

    Dreamora

    Joined:
    Apr 5, 2008
    Posts:
    26,602
    static and dynamic batching lower the amount of drawcalls too, if the material thats used is the same.

    the stats window gives you various stats, from the prebatched number to the emulated number that you can expect at runtime if you play it in editor
     
unityunity