René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"[MNT]: TimedAnimation event loop interval","articleBody":"### Summary\n\n`TimedAnimation` resets the event timer at the end of its `_step()` method: \r\n\r\nhttps://github.com/matplotlib/matplotlib/blob/f7b3def52b39a070064e0e713042cd948c985643/lib/matplotlib/animation.py#L1446\r\n\r\nAs far as I can tell, this is by design. The original commit included this comment:\r\n\r\nhttps://github.com/matplotlib/matplotlib/blob/70e401fe681b6e32ed5fdbf7eea182828887d75b/lib/matplotlib/animation.py#L314-L318\r\n\r\nTo me, this was suprising and it took me a while to get to the bottom of it. From the documentation I took that the event loop calls the animation target at exactly the `interval` specified. However, the current implementation means that it actually calls it at `interval + time_in_callback`. In case the callback does some heavy lifting, like processing data or communicating with an instrument, this can add significant overhead to the real interval between frames.\r\n\r\n\u003cdetails\u003e\r\n\u003csummary\u003e\r\nA MWE highlighting the effects of resetting the interval:\r\n\u003c/summary\u003e\r\n\r\n```python\r\nimport matplotlib.pyplot as plt\r\nimport numpy as np\r\nfrom matplotlib import animation\r\n\r\ndef update(frame):\r\n    if len(frame) \u003c 2:\r\n        return []\r\n    diff = np.diff(frame)\r\n    avg.set_text(avg_txt.format(diff.mean()*1e3))\r\n    cur.set_text(cur_txt.format(diff[-1]*1e3))\r\n    return [avg, cur]\r\n\r\n\r\ndef frame():\r\n    times = deque(maxlen=100)\r\n    while True:\r\n        time.sleep(0.1)\r\n        times.append(time.perf_counter())\r\n        yield times\r\n        \r\nfig, ax = plt.subplots(figsize=(1,1))\r\nax.axis('off')\r\navg_txt = 'avg: {:.3g} ms'\r\ncur_txt = 'cur: {:.3g} ms'\r\navg = ax.text(0.5, 0.75, avg_txt.format(0), va='center', ha='center')\r\ncur = ax.text(0.5, 0.25, cur_txt.format(0), va='center', ha='center')\r\n\r\nanim = animation.FuncAnimation(fig, update, frame, repeat=False,\r\n                               cache_frame_data=False, interval=200)\r\n```\r\n\r\n\u003c/details\u003e\r\n\r\nWith the current behavior, this results in an effective interval of 300 ms. Commenting out [the line resetting the interval](https://github.com/matplotlib/matplotlib/blob/f7b3def52b39a070064e0e713042cd948c985643/lib/matplotlib/animation.py#L1446) results in the effective interval I would expect from `interval=200`, 200 ms.\n\n### Proposed fix\n\nThere might be reasonable usecases for resetting the interval, but I cannot see any right now. Hence, I would propose to leave the timer running uninterrupted, i.e., dropping\r\n\r\nhttps://github.com/matplotlib/matplotlib/blob/f7b3def52b39a070064e0e713042cd948c985643/lib/matplotlib/animation.py#L1446\r\n","author":{"url":"https://github.com/thangleiter","@type":"Person","name":"thangleiter"},"datePublished":"2024-08-02T09:13:46.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":3},"url":"https://github.com/28647/matplotlib/issues/28647"}