Page 79 of 82 FirstFirst ... 29697778798081 ... LastLast
Results 1,561 to 1,580 of 1624

Thread: Blender 2.8 development thread (Widgets and Eevee)

  1. #1561
    Quote form code blog.
    The “everything nodes” concept will have to wait a while. It’s not something for 1 person (or part timer) to do in a short time, especially not now we’re doing so many other projects already. This project also depends on successful completion of “depsgraph”, which will take many months – if not half a year.
    In dev meeting, Ton is talking about a first 2.8 as a beta. Like it was the case for 2.5.
    It means that 2.81 , 2.82, 2.83 ..? would be dedicated to user feedback, design issues (Workspaces, topbar, key-config, ...) and fixing and optimizing of basics of 2.8 (Layer system, EEVEE, Clay Engine, asset-engine, etc...).

    When this awful period would be passed, everybody would want a useable Blender and a period of calm with small problems to solve.
    So, I seriously doubt that everything nodes would come before the last 3 releases of the series.

    I disagree with Ton's method. IMO, it makes no sense to launch a beta to calm people or reassure them about blender progress. We have buildbots and dev blog and artists making videos and demos before any official release.
    Blender progress is obvious. We don't need to keep this bad habits to schedule unusable betas.
    Launching a beta with obvious missing features will just result in a noisy user feedback.
    People will continuously report about these missing features and devs will waste their time to close them.
    A beta is not needed as long as there are obvious lacks from module owners and Blender Institute artists point of view.
    He probably does it to convince some supporters. But the result is a lost of efficiency that result in a waste of their money.



  2. #1562
    Member JustinBarrett's Avatar
    Join Date
    Jul 2009
    Location
    Trier(near), Germany
    Posts
    2,021
    While I do agree the main point of releasing for re-assurance is weak ....at some point it is better to release to the community for that exact reason though....many eyes see more bugs. I would just limit the bug reporting in phases 2.81 layer system bugs, 2.82 shading bugs etc etc until 2.84 all bugs being accepted. I also think the bug tracker should be blacklisting people based on number of bugs submitted versus the amount that were actual bugs.

    but this is just my opinion.
    "The crows seem to be calling my name." Thought Kaw.
    Myrlea, "The Shepherd's Quest" formerly "Valiant" [project]



  3. #1563
    Member Ace Dragon's Avatar
    Join Date
    Feb 2006
    Location
    Wichita Kansas (USA)
    Posts
    28,163
    Originally Posted by zeauro View Post
    I disagree with Ton's method. IMO, it makes no sense to launch a beta to calm people or reassure them about blender progress. We have buildbots and dev blog and artists making videos and demos before any official release.
    Blender progress is obvious. We don't need to keep this bad habits to schedule unusable betas.
    Launching a beta with obvious missing features will just result in a noisy user feedback.
    People will continuously report about these missing features and devs will waste their time to close them.
    A beta is not needed as long as there are obvious lacks from module owners and Blender Institute artists point of view.
    He probably does it to convince some supporters. But the result is a lost of efficiency that result in a waste of their money.
    The issue is that holding the official 2.8 builds until they have full usability will cause a lot of impatience among users who are used to the frequent releases of 2.6x and 2.7x. In any case, early 2.8 builds done by the BF will need to carry a clear message that they are betas and that you may notice bugs, crashes, and missing features. Users will also need a message stating that scenes and created in the beta and addons designed for the beta are likely to break in later builds.

    That said, even the betas may not start to come out until the end of this year or even next year, there's still a lot of time to get to (at least) a minimum state of usability.
    Sweet Dragon dreams, lovely Dragon kisses, gorgeous Dragon hugs. How sweet would life be to romp with Dragons, teasing you with their fire and you being in their games, perhaps they can even turn you into one as well.
    Adventures in Cycles; My official sketchbook



  4. #1564
    Originally Posted by Ace Dragon View Post
    That said, even the betas may not start to come out until the end of this year or even next year, there's still a lot of time to get to (at least) a minimum state of usability.
    You are right on this point. And it is what was eventually kept on dev meeting report.
    Today, sergey get started the massive tracker cleanup.



  5. #1565
    I'm agree with the thing that show betas to people create noises. But in this case is more a problem of communication. The principal doubts of people are about layers, draw modes,... things that a simple public message will solve.



  6. #1566
    Member Indy_logic's Avatar
    Join Date
    Nov 2008
    Location
    Portland, Oregon
    Posts
    998
    The daily builds create noise but they also create a lot of good hype too. Lots of Eevee videos on YouTube increases interest in Blender and hopefully that leads to more funding.

    IMHO, I think it's a mostly positive thing. Even if the result is users crying about the lack of functionality or broken features, this still helps the devs in one way or another. You can kind of get a feeling for how people will react when the features are worked out. Also, feedback while still in the design phase is something that most developers don't actually do. And it's that level of transparency that allows Blender development to thrive. As far as I can tell, development secrecy causes more issues than it solves.
    Last edited by Indy_logic; 15-Sep-17 at 02:08.



  7. #1567
    This is a long thread, but does the newest build of 2.8 fix the Udim problem that I get with 2.7? e.g. Udims are being skipped over when I'm using Renderman. Haven't tested it with cycles 2.79.



  8. #1568
    Hey, I am only occasionally checking the progress of Eevee, so I am not much in the loop. Could someone please tell me what happened to antialiasing? I remember seeing on quite early spot on the roadmap, yet all the Eevee tests I see so far seem to miss it.

    Thanks



  9. #1569
    Member dfelinto's Avatar
    Join Date
    Jun 2005
    Location
    Rio de Janeiro, Brasil
    Posts
    667
    Originally Posted by rawalanche View Post
    Hey, I am only occasionally checking the progress of Eevee, so I am not much in the loop. Could someone please tell me what happened to antialiasing? I remember seeing on quite early spot on the roadmap, yet all the Eevee tests I see so far seem to miss it.

    Thanks
    Good timing. AA is expected to be tackled in the end of September, starting next week.
    www.dalaifelinto.com - @dfelinto
    Game Development with Blender [released June 2013]
    Latest BGE Project: Dome+Cosmic Rays+Blender = Cosmic Sensation (2010):BN - files - presentation



  10. #1570
    Originally Posted by zeauro View Post
    Quote form code blog.

    I disagree with Ton's method. IMO, it makes no sense to launch a beta to calm people or reassure them about blender progress. We have buildbots and dev blog and artists making videos and demos before any official release.
    Blender progress is obvious. We don't need to keep this bad habits to schedule unusable betas.
    Launching a beta with obvious missing features will just result in a noisy user feedback.
    People will continuously report about these missing features and devs will waste their time to close them.
    A beta is not needed as long as there are obvious lacks from module owners and Blender Institute artists point of view.
    He probably does it to convince some supporters. But the result is a lost of efficiency that result in a waste of their money.
    There was already noise with the 2.8 developer builds on the tracker. That will always exist.
    No matter the format, it'll need to clearly define areas what Blender bug reports will be accepted depending on the completeness.



  11. #1571
    Member JustinBarrett's Avatar
    Join Date
    Jul 2009
    Location
    Trier(near), Germany
    Posts
    2,021
    Thanks for the update Dalai.

    I've not really been messing with 2.8 at all tbh...I did look at one of the first builds before the shaders were updating properly...and I do not see me using it until my current game is complete...it probably will not even be out yet but I do like to keep up with blender development because at some point I WILL be using it and it helps to not just jump into it without any prior knowledge..

    I do have one hope...later down the road when the interactive mode is being worked on...I would love to have more functions exposed...there are a lot of bpy ops that would be really beneficial for that....creating geometry alone would be great...anyway...said my peace for today.
    "The crows seem to be calling my name." Thought Kaw.
    Myrlea, "The Shepherd's Quest" formerly "Valiant" [project]



  12. #1572
    Originally Posted by zeauro View Post
    Quote form code blog.


    In dev meeting, Ton is talking about a first 2.8 as a beta. Like it was the case for 2.5.
    It means that 2.81 , 2.82, 2.83 ..? would be dedicated to user feedback, design issues (Workspaces, topbar, key-config, ...) and fixing and optimizing of basics of 2.8 (Layer system, EEVEE, Clay Engine, asset-engine, etc...).

    When this awful period would be passed, everybody would want a useable Blender and a period of calm with small problems to solve.
    So, I seriously doubt that everything nodes would come before the last 3 releases of the series.

    I disagree with Ton's method. IMO, it makes no sense to launch a beta to calm people or reassure them about blender progress. We have buildbots and dev blog and artists making videos and demos before any official release.
    Blender progress is obvious. We don't need to keep this bad habits to schedule unusable betas.
    Launching a beta with obvious missing features will just result in a noisy user feedback.
    People will continuously report about these missing features and devs will waste their time to close them.
    A beta is not needed as long as there are obvious lacks from module owners and Blender Institute artists point of view.
    He probably does it to convince some supporters. But the result is a lost of efficiency that result in a waste of their money.
    I disagree with you, and this has been tested.

    There are daily builds, but a lot of people are not willing to test it until it's called RC (release candidate). And there are people not willing to test it until it's release build - hence the need for a, b and c version/releases. This happen for every releases. No matter how many time Ton begged people to test the RC, there are people who waits for release version, test it, and report bugs as they have issues with their production.

    I guess to launch a beta is good. It's marketing. And besides, after previous meeting, the bug tracker has been labelled to not receive 2.8 bugs as its still in heavy development.

    My 2 cent.



  13. #1573
    I regret to have mentioned beta. It was not scheduled or kept in final dev report.
    I don't deny the fact that no matter what testing system is in place, there are people that would only test an official release.
    I just wanted to say that is a waste of time to launch a beta when obvious problems and limitations are known.
    Because the same way some people are not testing testing builds ; there are some people who are not reading warnings, articles, using dbo search and will complain about the lack of completeness of a beta. It was proven by 2.5 development.

    I know that you are probably used to test betas of games that are clearly incomplete.We are not talking about a game, here.
    We are talking about a tool that should demonstrate that is capable and efficient.
    There was a time and many people for who beta was meaning Release Candidate.
    If something is entitled just 2.80, people does not expect it to be a RC or a beta.
    If something is entitled 2.80 beta, people expects that next release is 2.80 or 2.80 beta 2, not 2.81 beta.
    There are new blender users, every month. And when 2.80 will be released, you can be sure that some of them will thought like that.
    Here, everybody is always complaining about how standards could increase accessibility of Blender.
    Why nobody cares about standard in naming ?
    Do you really think that you have to fool people to have them testing efficiently 2.8 ?
    It is not a newbie who just don't know how to use blender and don't know that 2.80 is a beta that will post a pertinent bugreport.

    Is bugreporting so low that we need that ? Currently, there are 278 open bugreports. There are a hundred of open To Dos. 76 open design tasks.
    2.8 bugtracker is limited to dev.
    Do you really think that when it would be open to users after months of waiting ; it would not quickly jump to several hundreds of reports just with daily builds ? Nowadays, with actual community ? I sincerely don't think it would make sense to add an hundred of parasite reports with a beta.

    It is not just a matter of collecting the biggest number of issues, we have to think about people that would have to treat them.
    It is not the same thing to have to triage a dozen of bugreports per day and several dozens.
    What is important is the quality of the feedback.
    If you receive a clear report that help you to fix a problem that occurs for thousands of users, you will be more efficient than if you triage a hundred of reports about the same problem before solving it.
    IMO, it would be an error to increase dev workload before having a clue that it will start to decrease.
    Especially when your team is partially made of volunteers.

    I don't know why I posted about beta. Maybe I had the feeling that things will be as chaotic for 2.8 as it was for 2.5.
    But it was not the conclusion of dev meeting and instead they started to clean the tracker (300 bugs at that moment).
    So, it was not a feeling that corresponds to current reality. No dev really cares about a 2.8 release at this moment.
    It is a distant problem for more of them.

    So, please, don't add to my mistake and don't reply to my rhetorical questions for the moment.
    Think about it and share your points when question will be more imminent.
    For the moment, try to pursue talk on 2.8 development. Eevee has soft shadows, now. What's next ?



  14. #1574
    Member tyrant monkey's Avatar
    Join Date
    Oct 2007
    Location
    Windhoek, Namibia
    Posts
    5,189
    Originally Posted by zeauro View Post
    that things will be as chaotic for 2.8 as it was for 2.5.
    Maybe I remember things a bit differently but I don't recall things been chaotic for 2.5 People were doing pretty much what is happening now they would take the builds play around, post a video there was a general buzz and excitement but I think most people knew it was a work in progress that you didn't use for serious work until it was out of Beta.

    If anyone suffered the chaos of 2.5 it was the guys making Sintel but not really general Blender users.

    I don't know if their(Blender developers) development model is good or not but I certainly don't want to back sit drive them because these guys have overall always managed things well.
    I have 500 bad drawings in me before the good ones. And I have 319 to go, here they are.



  15. #1575
    Originally Posted by tyrant monkey View Post
    Maybe I remember things a bit differently but I don't recall things been chaotic for 2.5 People were doing pretty much what is happening now they would take the builds play around, post a video there was a general buzz and excitement but I think most people knew it was a work in progress that you didn't use for serious work until it was out of Beta.

    If anyone suffered the chaos of 2.5 it was the guys making Sintel but not really general Blender users.

    I don't know if their(Blender developers) development model is good or not but I certainly don't want to back sit drive them because these guys have overall always managed things well.
    I totally agree, IIRC there was downloads for 2.49 and 2.5 next to each other, so it was obvious that 2.5 series was WIP. That was great because even if 2.5 was not complete at some point it was usable so you can choose when it's ok for you to switch. Maybe Ton wants to release intermediate version to help things getting organized and made step by step.



  16. #1576
    Does someone know if the new Eevee roadmap planning is already online?



  17. #1577
    Nice overview about the new Layers and Collections in Blender 2.8:
    https://code.blender.org/2017/09/vie...d-collections/
    I look forward to see the interface for changing the View Layers in the viewport
    My personal portfolio at marekmoravec.cz



  18. #1578
    Member BrilliantApe's Avatar
    Join Date
    Dec 2003
    Location
    Barcelona, Spain
    Posts
    2,682
    Originally Posted by maraCZ View Post
    Nice overview about the new Layers and Collections in Blender 2.8:
    https://code.blender.org/2017/09/vie...d-collections/
    I look forward to see the interface for changing the View Layers in the viewport
    The collections idea sounds interesting, but it would be nice to see how this improves on regular layers. Hard to tell how this will work without a good UI example..
    Working on an epic creative project with superheroes, aliens and lots of T&A - stay tuned!
    My hardware: Core i5 4690 3.5ghz - Geforce 960 2GB - RAM 8GB, Wacom Bamboo Fun Pen&Touch
    My software: Blender, Krita | My BA sketchbook



  19. #1579
    Will we be able to hide individual objects or will it be collections only? So far it's just collections, that will take some time getting use too.



  20. #1580
    Originally Posted by BrilliantApe View Post
    The collections idea sounds interesting, but it would be nice to see how this improves on regular layers. Hard to tell how this will work without a good UI example..
    You can see a basic implementation in the 2.8 builds already. The new layer system can already nest layers so you can make a nice hierarchy of layers. The naming of layers works as well.
    More advanced tools to handle all the objects and layers will surely come in the future. I guess that hiding objects will be implemented also and if not it should not be so difficult to make a python script to do just that (move selected objects into new collection, assign the new collection to a new View layer and finally assign the view layer to active viewport).



Page 79 of 82 FirstFirst ... 29697778798081 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •