-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
"Data: textarea" and "State" wrappers perform poorly on iPad. #432
Comments
It sounds OK for the text area and state wrappers to have poor performance on iPad. I don't recall if we requested testing those wrappers on iPad. |
@kathy-phet @arouinfar please decide if this issue is blocking for publication of GO 1.2. Also please provide guidance on whether QA should be testing these wrappers on iPad. |
@pixelzoom based on @samreid's comment, this is not a blocking issue.
Unless poor performance in a wrapper is indicative of poor performance of the student-facing simulation, I don't see any reason to test on iPad or Chromebook. However, this guidance should come from the PhET-iO team, so back to them. |
I agree, they should not be blocking. These wrappers test the top level of performance capabilities of PhET-iO, with printing tons to a text area and setting state many times back to back. In the vast majority of cases these are not what are exercised in student facing wrappers (though it is possible). These are cases where we leave it up to clients to test their wrappers on the devices their students will use. You can't expect an iPad to be able to set state on a complicated sim >1 time a second and still have good usability through that process. |
So... What is the guidance for QA? Should they be tested on iPad or not? |
It should not. In https://github.com/phetsims/QA/blob/master/documentation/qa-book.md#test-13-data-recording-wrapper-test I see that it says:
This is the same test, but it didn't get this documentation. @KatieWoe can you please add |
For the state wrapper, it is my recommendation to test on iPad. Here is what I would add as documentation for the test in the qa book:
|
Done in above commit |
As for #432 (comment), this is different from what we were previously instructed to do. Text area was specifically supported on iPad since colorized and json were not. Is this something the whole team wants changed? |
I worry that the issue is just because of the quantity of text output, and not because of internal data stream handlings. The metacog recording handles that case by piping the data to metacog (without printing it). @samreid should we test the text area wrapper on iPad? |
A spot check to make sure it runs on iPad sounds good, but text area wrapper performance does not need to be fast/smooth on iPad. |
This issue is still hanging around from a dev test that was done in April 2022. And this sim is scheduled for QA and publication as part of https://github.com/orgs/phetsims/projects/44, which is likely to happen while I'm on vacation. @zepumph @samreid is there anything to do here, or should this issue be closed? |
Yes this issue can be closed in my opinion. We do not support the PhET-iO wrapper suite in iPad. There is though reason to think that the state wrapper should have some performance rigor attached to it, but it isn't clear how much. Perhaps we need a larger discussion to know what is acceptable for setting state. i.e. 1 per second on powerful machines, and 1 per 3 seconds for iPad. |
Closing based on @zepumph's opinion. |
For phetsims/qa#798. @arouinfar @kathy-phet FYI.
In phetsims/qa#798 (comment) @KatieWoe reported:
"the textarea" is presumably the "Data: textarea" link in the Wrapper Index.
I'm surprised that these wrappers are being tested on iPad. @samreid and @zepumph can you please comment? Are these wrappers intended to be tested on iPad? Is their performance on iPad a concern?
The text was updated successfully, but these errors were encountered: