3 ms·

I've recently tested 20 different webgl libraries. I was looking for anything with hierarchical scenes that performance-wise would be as fast as hand tuned vani

by dvh 6y ago

I've recently tested 20 different webgl libraries. I was looking for anything with hierarchical scenes that performance-wise would be as fast as hand tuned vanilla webgl. Only play canvas came close but it has one annoying feature. There is IDE with the same name (from the same company) and every time I searched how to do X in playcanvas, first 10 items in search were always for ide (e.g. go to menu x, choose y, move slider to z), not library (use function fooBar(x,y,z)). This made development extremely annoying so I abandoned it and went vanilla webgl. They should rename library to something else (libcanvas or something). Imagine you are deep in the "zone", solving difficult problem, you write search query while still desperately trying to stay in the zone and every time you press enter Google stabs you in the back. It was hell.

ovenchips 6y ago
Thanks for the feedback! Good to hear that you experienced good performance from PlayCanvas in your tests. The majority of PlayCanvas users to opt for the Editor rather than using the Engine standalone, so yeah, the developer resources are biased a bit in that direction. But we are working hard to rebalance things towards the engine. For example, we're about to release a new engine-only examples browser app that's really cool. Key an eye out for it. :)
- adenozine 6y ago
  This seems really tone-deaf to me, to sorta market a mostly unrelated feature that's upcoming when a developer just decided against using your product for practical reasons. I suppose it's nice to know that you've at least acknowledged that there's an issue, but it seems weird to take so lightly someone using your product, finding it so difficult to deal with that they dropped it, and then even dropping an emoticon at the end. Just my two cents.
  - ovenchips 6y ago
    No problem! Thanks for your feedback too. The examples browser is really designed to be a vehicle to learn and experiment with the engine run-time. It should be a great resource for engine-only developers - so hopefully the dvh will find it interesting. We've done a ton of work on the API reference recently too (https://developer.playcanvas.com/en/api/ https://developer.playcanvas.com/en/api/). It's a huge task, but yeah, we're working night and day to make the runtime easier to work with.
    - shkkmo 6y ago
      It seems like none of that work solves the underlying issue that was raised which was specifically search ambiguity between the library and the editor. Any response should discuss the possibility to changing the library name or creating search friendly alternative name (such the use of "golang" to alow easy searching for content related to the highly ambiguously searchable Go programming language.
      - ovenchips 6y ago
        I've kicked off a new forum thread on this topic: https://forum.playcanvas.com/t/improving-search-results-for-engine-only-developers/19981 https://forum.playcanvas.com/t/improving-search-results-for-... Excited to get peoples' thoughts on it.
  - greshario 6y ago
    It's a marketing thing - the engine is free and open source, the editor is not.

by dvh 6y ago

ovenchips 6y ago

Thanks for the feedback! Good to hear that you experienced good performance from PlayCanvas in your tests. The majority of PlayCanvas users to opt for the Editor rather than using the Engine standalone, so yeah, the developer resources are biased a bit in that direction. But we are working hard to rebalance things towards the engine. For example, we're about to release a new engine-only examples browser app that's really cool. Key an eye out for it. :)

adenozine 6y ago
This seems really tone-deaf to me, to sorta market a mostly unrelated feature that's upcoming when a developer just decided against using your product for practical reasons. I suppose it's nice to know that you've at least acknowledged that there's an issue, but it seems weird to take so lightly someone using your product, finding it so difficult to deal with that they dropped it, and then even dropping an emoticon at the end. Just my two cents.
- ovenchips 6y ago
  No problem! Thanks for your feedback too. The examples browser is really designed to be a vehicle to learn and experiment with the engine run-time. It should be a great resource for engine-only developers - so hopefully the dvh will find it interesting. We've done a ton of work on the API reference recently too (https://developer.playcanvas.com/en/api/ https://developer.playcanvas.com/en/api/). It's a huge task, but yeah, we're working night and day to make the runtime easier to work with.
  - shkkmo 6y ago
    It seems like none of that work solves the underlying issue that was raised which was specifically search ambiguity between the library and the editor. Any response should discuss the possibility to changing the library name or creating search friendly alternative name (such the use of "golang" to alow easy searching for content related to the highly ambiguously searchable Go programming language.
    - ovenchips 6y ago
      I've kicked off a new forum thread on this topic: https://forum.playcanvas.com/t/improving-search-results-for-engine-only-developers/19981 https://forum.playcanvas.com/t/improving-search-results-for-... Excited to get peoples' thoughts on it.
- greshario 6y ago
  It's a marketing thing - the engine is free and open source, the editor is not.