Post Mortem Note 01/07/2010
With the advent of 2.56A … 2.56 may be a nonuseful release.
2.56 and 2.56A are two different releases to emphasize the point.
This thread is now largely obsolete.
This thread was originally intended as a community inquiry but became a non community … failure.
I thank everyone for their useful comments.
I advise new readers to know this thread became a counter example to what community dialogue should be.
So feel free to skip any part of this thread or the entire thread.
If this were the days of old with paper …
I would send out my Thank You cards and throw this document in a burning barrel and roast marshmallows or 100% whole grain bread.
Just trying to save some people time.
Thanks.
Original Topic
Hello to all
I am using Blender 2.56 revision 33949 Windows 32 Bit … OS Windows 7
IMHO 2.56 seems to crash more on me then 2.55 … perhaps 3 times a day.
Recently I am using fluid simulation and particles more.
Could this be the cause of more crashes?
??Do others have the experience of
blender crashing when the user works with
water simulation and particle activity ??
…
EXTRA OPTIONAL
Such things are useful to know for software defect reporting.
For example … if I drag the frame marker during water simulation baking … blender seems to crash.
…
EXTRA
I have 6 months experience with blender total, 5 of which are 2.5x.
There is a crash bug in simulations/undo apparently that is fixed and will be in 2.56a to be released imminently… (and will already be in svn, but svn acces is out today as they’re upgrading the server…
If you find bugs in Blender, then please report them here. To be honest it’s a bit unfair to the developers to just say that “Blender 2.56 crashes more”, partly because they have spent a LOT of time fixing over 400 bugs since 2.55, and partly because you don’t explain how you’re getting Blender to crash (Other than that fluid baking problem, which I couldn’t repeat).
If nobody reports bugs, the developers won’t know about the bugs. Blender is huge (Almost a million source lines of code), and no one person can do it alone. You shouldn’t complain about bugs if you don’t report them.
I have a crash with particlesystem, type hair, hair dynamics enabled: setting amount to 0 and blender crashes.
I have another crash:
In uv/image editor, with image painting enabled and in texture paint mode, closing the properties panel with the mouse (not N) and blender crashes.
Could someone confirm and/or fill bugreports?
thanks
I can reproduce both of those (The first one is in fact already known I think) but maybe you should take this opportunity and learn how to report a bug. Go here and click “Submit new”, then follow the instructions there.
I am glad you are honest with your opinion.
However I feel your writing is misplaced and may deserve its own thread.
I just learned that the <<quick response>> 2.56A is released at blender.org.
I have installed it.
You wrote … <<unfair to the developers>>
What are the guidelines relating to and examples of fair and unfair statements <<to the developers>> ???
There is no contradiction where there is no specification.
Maybe that is a new thread.
If you would like to collaborate on a new thread please inform me.
Maybe I do not understand how you used the word <<unfair>>.
I asked a very objective question from the readers.
The operating system such as windows os and mac os can identify illegal operations by software and stop execution of that software. I call this for convenience sake a crash.
In that sense windows 7 os and mac os are objective measurements.
So you might redirect your statements of fairness to the OS manufacturers.
Getting the count of crashes in sometime range is objective and can be automated.
Comparing two numbers to see which is greater proceeds similarly.
Software Testing exists for a reason and beta means implies more testing by users who should voice their assessments and experiences.
That software infrastructure is not present on the desktop of all blender users at present … so I asked in this forum.
I have no problem reporting and discussing software defects.
I have contributed in some very small way already to defect reporting.
I have received defect reports myself on non blender projects.
Anyone can choose to improve software from a defect report or let software decay.
There are a number of articles about how many software writers process defect reports.
Software and defects are inseparable.
Great Software and open assessment are inseparable.
Blender can improve with customer feedback, automated assessment, praises and defect reports.
Microsoft compilers, and Mac compilers have experienced defects as well.
Every release with increasing version number of every piece software typically means that a defect was removed. That is software success.
The fact that we even must have antivirus software tells us that the OS and our knowledge about software is defective.
We are awash in defects and bacteria.
I do still use my OS.
I still use Blender.
Where was my statement of complaint???
Here is an example of complaint.
>>>> This green kool-sugar-water-aide taste bad and I do not like it and I am a dissatisfied customer.
Do you have some representative history with the developers? Are you one of them?
If you are, I want the inside scoop.
I have faith that the blender software writers can handle factual assessments of the software.
If you have superior knowledge to the contrary I would be thankful to receive it.
I would be a little surprised if they read this publicly readable forum. They are most welcome.
I am not one of the software developers. Though you have got me thinking.
You wrote
>>(Almost a million source lines of code)
How is that relevant to the question of crashing?
But since you opened the discussion …
If any group can rewrite blender in 0.5 million lines with less defects and more features … would that not be desirable??? Less can be more.
I think my post has some relevance for a few people, but not the whole blender community.
You might consider opening up a new thread … Praise for Blender 2.56A.
I try to tell people about Blender.
I include my likes and dislikes. I doubt I am alone in that activity.
If it your opinion that Blender should not be assessed please say so, in a new thread.
You dsavi wrote
>>You shouldn’t complain about bugs if you don’t report them.
?did you mean me (bezier_slinger) in particular or all users of blender artists?
ok thanks, bugs reported, we’ll see in how many days it will be fixed.
I think they are easy to fix because they are reproduceable, but I’m not a developper
Way over the top on that post. You could have gotten your point across in a more concise manner. I think the answer that you recieved was more than sufficient, bezier-slinger, and it seems as if you took offense. The point that was trying to be made here, was that this is not the place, in all reality, to report bugs or complain about the functionality of Blender when it does not work correctly. There is a place for that, and it is called the bug tracker. Taking a poll doesn’t really help, since not all users will use all of the tools, or some of the tools that you might or in the same way, and this poll was way too general in scope. However asking a question as to whether or not someone has experienced the same problems in order to determine a bug before posting in the bug tracker is welcome and encouraged, and you have done that.
The unfairness possibly lies in the fact that developers only stop by here once in a while, and to smear a development process without a proper defense could be taken wrong. I think it lies more in ridicule before the facts. I don’t, however, think that this was you intended, so perhaps unfair is an incorrect term, and not a proper defense posed by dsavi.
Be that as it may, perhaps a survey is not the proper route in this case. In development cycles that I have been a part of over the last few years, it is more productive to get a feel for the problem by asking a specific question about a specific problem and seeing if anyone else has had the same issue and collaborating about the issue before settling on a bug report. They will fix it. This, I feel, was your intention. Bottom line, report those bugs! They have a short list and yours will addressed quickly if the severity is high, and get rid of the general polls :).
There are no specific written guidelines, I am talking about being polite. Not fair as in “right and wrong”, but fair as in being overly critical.
Actually I think that your poll is quite subjective. You asked a vague question without asking for data to back it up.
(I underlined the part that I am replying to here)
It can be objective, but since you did not ask for data (As it is an “Informal survey”), it is not. On the other hand, I have the data that 449 bugs were fixed between 2.55 and 2.56 (More were fixed between 2.56 and 2.56a).
I don’t dispute anything here, but I would like to note that I was not talking about compilers or operating systems at all.
You complained by starting a thread that says that in your opinion, Blender 2.56 crashes more. That is fine and you are entitled to an opinion and the right to express it. But the way you are expressing it doesn’t seem very respectful to the developers (Just complaining about bugs instead of reporting them so that they can be fixed).
I am neither a developer nor a representative of any of the developers. There is no real “inside scoop”, as Blender has an open development process.
The Blender developers can handle factual assessments, I am not saying that they can’t. What I am saying is that this is not at all a factual assessment. I am also not saying that an assessment that says that Blender is unstable and crashes a lot can not be a factual assessment. You must provide evidence first, as I have with that link to the list of bugs fixed.
Some of them do read these forums occasionally.
Sure it can. My point is simply that nearly a million SLOC is a lot of code to debug.
I think this illustrates how badly you understood my post. I do not mean to offend, but English is not your first language
I mean everyone who uses Blender (We can assume that in 2011 nearly every Blender user has a working Internet connection, as they almost certainly downloaded Blender). Because the development process is open, they should use the opportunity to report bugs if they want a better working Blender. If they don’t, then it is partly their own fault.
DanaLou
Thank you for pointing out a defect in my poll.
Your observations are relevant and insightful.
May I ask you to advise me on any future poll?
(Shudder about poll creation)
I enjoy your icon. Is that you?
What is that all over you face?
We need more people like you in the World.
Carefree, Fun, Daring.
Thanks.
Please examine your m o o d in what you wrote above.
You seem angry and intent on pointing out fault.
There is so much to be happy about in Life. Please be happy.
My thread was only to inquire about the crash rate of 2.56 Blender to see if others had similar experiences.
I have no apologies for that.
dsavi consider creating your own thread about defects to be more effective.
You have motivated me with your … m o o d … to create my own new thread …
You are most welcome to write there
If you want people to report more bugs you should have your own thread because it would be more effective.
In many forums for many different products … not blender … introducing a secondary idea to a thread …not related strongly to the first idea … is deprecated because it creates non focused threads.
I believe it helps all users here to have focused threads.
Crash Rate inquiry and evangelizing defect reporting are not strongly related.
Please continue the evangelizing of Defect Report with a new and dedicated thread.
Defect reporting deserves its own thread.
Remember:
The version you are downloading from www.blender.org is NOT the latest. It’s the official release at a checkpoint.
if you want to download the latest version of blender, you should download it from www.graphicall.org . The builds you can download are updated daily. The higher the five digit number is, the newer the version is. Good luck.
).