[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [ide-dev] "Typing with pleasure"

On 24 April 2016 at 17:27, Doug Schaefer <cdtdoug@xxxxxxxxx> wrote:
On Sat, Apr 23, 2016 at 2:29 PM, Cole Markham <cole@xxxxxxxxxxxxxxxx> wrote:

Based on the post I read on the IDEA zero-latency implementation, it seems it only applies to the first keypress. Further, the testing tool used for these results is specifically designed to measure the optimization that the zero-latency mode uses. The tool does support rapid keypress testing, but I don't think it was used for this article. I think it was skewed to make the zero-latency mode look like the winner, but that's just a hunch. In either case, independent results with more real world conditions would answer some of these questions.

Interesting. It is certainly an area we should investigate. The use case I would think users really care about is when they're in the midst of writing a bunch of code and are touch typing, watching what they're typing on the screen. First keypress isn't really a driving issue.

To be clear, in case someone who hasn't read the article misunderstands, "first keypress" doesn't mean first keypress in an entire editor session, it means first keypress in a series of keypresses all occurring within a very small interval of each other. Quoting from the IDEA zero-latency article:

"This should be enough to significantly improve the editor latency. Keep in mind though, that if subsequentÂblocking or repainting takes more time than a period until the next key is pressed, a typing lag is still possible."


So yes, the benchmarks results are skewed to favor the first keystroke only (and thus favor IntelliJ), but the article also says they are working on a way to solve the latency problem for all keypresses (by avoiding how they currently do locks on their document model).