OpenSesame 3.2.1 "Kafkaesque Koffka" released

sebastiaan

About this update

OpenSesame 3.2.1 Kafkaesque Koffka is the first maintenance release in the 3.2 series. It contains bug fixes and minor improvements, and should be a pleasant and safe upgrade for everyone who is using the 3.2 series.

The main reason for this quick maintenance release is an issue with variable properties of text elements in sketchpad items.

If you are upgrading from OpenSesame 3.1 or earlier, please see the list of important changes:

• http://osdoc.cogsci.nl/3.2/important-changes-3

As always, you can download OpenSesame from here:

• http://osdoc.cogsci.nl/3.2/download

Bug fixes and improvements

opensesame:

• Updated to 3.2.1
• Issue #580: Giving a variable color name to a text element in a sketchpad causes a crash (Bug)
• Issue #581: Variable z_index gives crash in sketchpad (Bug)
• Issue #583: Using 'response' as item name causes conflict with 'correct_response' variable (Bug)
• Minor translation updates

datamatrix:

• Updated to 0.8.1

andersony3k

Re Version 3.2.1. It will run experiments created in a previous version, but not experiments created in Version 3.2.1 itself. For example, if I simply launch Version 3.2.1 and simply run the default experiment without changing anything. An error message is generated:

"Stopped

The experiment did not finish normally for the following reason:

Python seems to have crashed. This should not happen. If Python crashes often, please report it on the OpenSesame forum.

Details

item-stack: ``"

And no matter how I change the experiment, I still get the error message.

I'm using Windows 7.

sebastiaan

Hi,

These Python crashes, which reflect a bug somewhere deep down in the libraries, are very annoying, and hard to debug because they aren't accompanied by a useful error message. Do your older experiments use a different backend? To me, that sounds like the most likely reason for they don't crash.

You could also try using the Python 3 version of OpenSesame. In principle, Python 3 is much nicer, but the downside is that some important packages (PsychoPy and PyGaze) still don't work with Python 3.

Cheers!
Sebastiaan

vatsalthebest

Hey!

Just a little heads up I am trying to update from Jazzy james version to Kafkaesque Koffka 3.2.1 But I cant seem to install the program using exe installer. Soon as the installer pops it seems to vanishes without any error or messages. I am using windows 10 on a 64 bit operating system.

What should I do?

vatsalthebest

Hey!

I am able to use the zip version. But still looking forward to use the installed version

Thanks
Vatsal

andersony3k

Thanks. It was not the backend (the backend was not different, and changing the backend in 3.2.1 doesn't avoid the Python 2.7 error).

For now, I'm simply reverted back to OpenSesame 3.1.9, and everything is working fine again.

[Re: Hi, These Python crashes, which reflect a bug somewhere deep down in the libraries, are very annoying, and hard to debug because they aren't accompanied by a useful error message. Do your older experiments use a different backend? To me, that sounds like the most likely reason for they don't crash.
You could also try using the Python 3 version of OpenSesame. In principle, Python 3 is much nicer, but the downside is that some important packages (PsychoPy and PyGaze) still don't work with Python 3.
Cheers! Sebastiaan]

sebastiaan

@vatsalthebest Have tried downloading the installer again? Perhaps the file got corrupted during downloading?

@andersony3k Thanks for reporting this. It seems that some other people are experiencing the same issue, so we're going to keep an eye on it. For now, what happens exactly if you try to run an experiment in 3.2.1. If you open the debug window, you'll see some messages, and these can help us track down when the crash occurs. Could you post that here?

andersony3k

Here's the entire contents of the debug window after I launch OpenSesame 3.2.1 and click the "Run Fullscreen" button. . .

Python 2.7.13

* Type "help()", "copyright()", "credits()" or "license()" for more information.
* Use the "print([msg])" function in inline_script items to print to this debug window.
* Inspect inline_script variables when an experiment is finished.


Starting experiment as ExperimentProcess-2
Expyriment 0.9.1b2-11-gc100ee8 (Python 2.7.13) 
Scripts/safelaunch-opensesame.py
openexp.sampler._legacy.init_sound(): sampling freq = 48000, buffer size = 1024
experiment.run(): experiment started at Sun Feb 18 08:11:07 2018
experiment.run(): disabling garbage collection

sebastiaan

@andersony3k Thanks, that's very useful. Two more questions:

• Do experiments without any text elements run properly?
• If you select the inprocess runner under Tools → Preferences, does the crash still occur?

andersony3k

When I launch OpenSesame and it opens the default, 'new experiment':

• If I change the runner to inprocess it runs without error (instead of resulting in a Python crash).
• With the runner still set to inprocess: if I delete the welcome message, or if I instead add a keyboard response object (keeping that object's default settings) OpenSesame crashes in a different way. Specifically, it just hangs for a long time. Eventually, Windows produces a notification that python.exe has stopped working. At that point OpenSesame is no longer running, so there's no debug window to look at.
• With the runner still set to inprocess: If I add to the default new experiment, a sketchpad or an empty sequence object, the experiment runs without error.

So it appears that there my be a subtle and/or complex incompatibility between Python 2.7 and OpenSesame 3.2.1.

andersony3k

Also note that even in OpenSesame 3.1.9, about a third of the time when I try to run an OpenSesame, experiment, there's still an error in launching the experiment. But in this situation, the experiment is fine IF it launches. But about a third of the time, it won't launch. It just hangs. And if I try to look at the windows task manager to see what's going on, I sometimes find that python.exe is not working.

sebastiaan

@andersony3k During a workshop today I noticed that quite a few people are affected by this, and I managed to find a workaround (I think). I opened a new topic for this:

• http://forum.cogsci.nl/index.php?p=/discussion/3829/

T-J-L

I have encountered another issue with the new version.
In an experiment that was running smoothly in the first version of Kafkaesque Koffka, the "show if" does not seem to work anymore.
I programmed an IAT with the possibility to correct answers, so the participants see a red X if the answer is wrong.
The setting, however, does not seem to alter the presentation, the X is always shown, no matter if the answer is right or wrong. I tried to program it with the script and with the user interface, it seems to save the settings but nothing changes in the behavior of the experiment...

Is there a way to fix this?
Best wishes and thank you for every hint!
Tim

sebastiaan

Hi Tim,

That was indeed a bug. It has been fixed already, and will be included in OpenSesame 3.2.2, which we will release soon. In the meanwhile, you could update the python-opensesame package to 3.2.2a1 (or later) in the Python package manager (under Menu → Tools).

Cheers!
Sebastiaan

T-J-L

Hi Sebastiaan

Perfect! I'll try that, thank you very much for the, as always, quick response.
Best wishes
Tim