5 ms·

It plays really slow in Firefox, and there are also audio issues (they may be related?) But in Chrome, this plays really well! Great work!

tonyhart7 2y ago
ofc it is firefox
- ShinTakuya 2y ago
  Can I inquire as to why you feel so negatively about Firefox?
  - tonyhart7 2y ago
    burn a lot of fcking goodwill for measly 4% web user, turn out eating google money literally make you the next "google" chromium edge literally surpass firefox and with new policy it would eat more to the marketshare
    - ShinTakuya 2y ago
      Is it that surprising that Edge is surpassing Firefox given Firefox doesn't have the onboarding ramp of Microsoft and Google? Microsoft shoves Edge in your face every chance it gets, as does Google. What can Mozilla do to compete against that? Unfair to use market share as an argument.
      - tonyhart7 2y ago
        "What can Mozilla do to compete against that?" well if only Modzilla foundation spend money given to them to actual firefox development instead of ehm ehm DEI policy (lol) and pay executive millions of dollars at this rate even brave,opera browser would surpass firefox
        ShinTakuya 2y ago
        While I agree they shouldn't be putting their money towards political causes, the scrutiny of the budget is also political in nature. Of the $400 million in expenses, only around $15 million went to executives, and around $1 million went to the DEI stuff. The rest was split evenly between developing and marketing Firefox, which is basically what you were criticising them on not doing. If they got rid of all the executives and DEI, they could increase the other budgets by 2%. There is 0 chance that the 2% extra budget could help the market share and development in any significant way. And if you don't pay executives enough, you get worse executives who make worse decisions because the good ones will go to other companies. So you'll be complaining even more if the $15 million executive pay got dropped. So in short: I agree Mozilla shouldn't get involved in politics. Despite this, no reasonable person could look at the numbers and conclude it's having a material effect on the destiny of Firefox.
        tonyhart7 2y ago
        ok my past response is mostly for laugh, this is my serious response Firefox is literally default install in most if not all linux distro os, while I agree that situation is far from ideal but nothing is can be blamed other than themselves for the longest of time firefox always playing catch up with chromium to achieve parity, maybe just maybe we don't need more browser Microsoft literally give up their browser (IE), given how much resource MS has do you think firefox is doomed?? especially with DOJ force google to breakup chrome and cut off their spending on modzilla I mostly critize about firefox place in the market, Opera found an niche with Opera GX for gaming, Brave with adblock and privacy focused etc while firefox is what??? inferior of chrome???
        ShinTakuya 2y ago
        I can't reply further to this thread without this devolving into a flame war so I'll just observe that you've moved the goal posts claiming you were just joking, and refuse to engage further until you acknowledge you're either getting defensive for no reason or arguing in bad faith.
shakna 2y ago
WebRender recently got much slower in Firefox [0]. I haven't looked at the source yet, but if the bitmaps aren't all preallocated, then it'll run into this repeatedly. [0] https://treeherder.mozilla.org/perfherder/graphs?highlightAlerts=1&highlightChangelogData=1&highlightCommonAlerts=0&replicates=0&selected=87312,1964495950&series=mozilla-central,93432,1,13&series=mozilla-central,87324,1,13&series=mozilla-central,87312,1,13&series=mozilla-central,87336,1,13&timerange=5184000 https://treeherder.mozilla.org/perfherder/graphs?highlightAl...
adrian17 2y ago
For me it runs at full speed in both stable and developer versions of Firefox (but even then, audio is bad at times). But still, I looked at performance out of curiosity. After running it under FF's profiler for a moment, it looks like a sync executed after every opcode recalculates the framebuffer (`step_until_next_audio_buffer -> cpu::microops::read_byte_from_memory -> emulator::sync -> gpu::scanline::write_scanlines`) and uploads it to the canvas with `putImageData`, which takes ~35% of execution time. Sounds like that could be redundant work, since I'm pretty sure the canvas isn't actually redrawn until the wasm execution yields back to the browser. EDIT: also, the emulator can't really be ran in debug mode, since 1. it blows the wasm stack (presumably when creating the huge `Emulator` struct, a not uncommon issue in Rust; I had to hack in bigger stack size with linker args), and 2. is panics on int overflows `apu::step`.
- catapart 2y ago
  Yeah, still being on Windows 10, I almost didn't put the complaint in the comment. Lots of reasons it might just be an environment issue for me. But, in my development experience with firefox, the issues it seems to have always seem performance related (or compliance with web standards related), and I figured that even if others weren't seeing the issue, there was still probably something underlying that could be figured out. So thanks for going through the debug! It's nice to know that there may be an optimization available for the project, and it feels like it might even be an additional performance gain for chromium, if I'm understanding you right? Might be some reason they need to do that sync after every opcode, but I agree with your suspicion that the canvas is probably being over-rendered to.
  - adrian17 2y ago
    FWIW, the thing I've found as the slowest single factor on my machine, might not be the slowest on yours; I might have not noticed the real cause of slowdowns on your machine at all, and you may be the only one able to diagnose it. (and for the record, I'm on W10 too.) Also, my personal experience from optimizing https://ruffle.rs/ https://ruffle.rs/ is that in many cases Firefox's wasm runtime can behave slightly faster than on Chrome for some reason; and performance issues are often instead caused by other APIs like the canvas.

tonyhart7 2y ago

ofc it is firefox

ShinTakuya 2y ago
Can I inquire as to why you feel so negatively about Firefox?
- tonyhart7 2y ago
  burn a lot of fcking goodwill for measly 4% web user, turn out eating google money literally make you the next "google" chromium edge literally surpass firefox and with new policy it would eat more to the marketshare
  - ShinTakuya 2y ago
    Is it that surprising that Edge is surpassing Firefox given Firefox doesn't have the onboarding ramp of Microsoft and Google? Microsoft shoves Edge in your face every chance it gets, as does Google. What can Mozilla do to compete against that? Unfair to use market share as an argument.
    - tonyhart7 2y ago
      "What can Mozilla do to compete against that?" well if only Modzilla foundation spend money given to them to actual firefox development instead of ehm ehm DEI policy (lol) and pay executive millions of dollars at this rate even brave,opera browser would surpass firefox
      - ShinTakuya 2y ago
        While I agree they shouldn't be putting their money towards political causes, the scrutiny of the budget is also political in nature. Of the $400 million in expenses, only around $15 million went to executives, and around $1 million went to the DEI stuff. The rest was split evenly between developing and marketing Firefox, which is basically what you were criticising them on not doing. If they got rid of all the executives and DEI, they could increase the other budgets by 2%. There is 0 chance that the 2% extra budget could help the market share and development in any significant way. And if you don't pay executives enough, you get worse executives who make worse decisions because the good ones will go to other companies. So you'll be complaining even more if the $15 million executive pay got dropped. So in short: I agree Mozilla shouldn't get involved in politics. Despite this, no reasonable person could look at the numbers and conclude it's having a material effect on the destiny of Firefox.
        tonyhart7 2y ago
        ok my past response is mostly for laugh, this is my serious response Firefox is literally default install in most if not all linux distro os, while I agree that situation is far from ideal but nothing is can be blamed other than themselves for the longest of time firefox always playing catch up with chromium to achieve parity, maybe just maybe we don't need more browser Microsoft literally give up their browser (IE), given how much resource MS has do you think firefox is doomed?? especially with DOJ force google to breakup chrome and cut off their spending on modzilla I mostly critize about firefox place in the market, Opera found an niche with Opera GX for gaming, Brave with adblock and privacy focused etc while firefox is what??? inferior of chrome???
        ShinTakuya 2y ago
        I can't reply further to this thread without this devolving into a flame war so I'll just observe that you've moved the goal posts claiming you were just joking, and refuse to engage further until you acknowledge you're either getting defensive for no reason or arguing in bad faith.

shakna 2y ago

WebRender recently got much slower in Firefox [0]. I haven't looked at the source yet, but if the bitmaps aren't all preallocated, then it'll run into this repeatedly. [0] https://treeherder.mozilla.org/perfherder/graphs?highlightAlerts=1&highlightChangelogData=1&highlightCommonAlerts=0&replicates=0&selected=87312,1964495950&series=mozilla-central,93432,1,13&series=mozilla-central,87324,1,13&series=mozilla-central,87312,1,13&series=mozilla-central,87336,1,13&timerange=5184000 https://treeherder.mozilla.org/perfherder/graphs?highlightAl...

adrian17 2y ago

For me it runs at full speed in both stable and developer versions of Firefox (but even then, audio is bad at times). But still, I looked at performance out of curiosity. After running it under FF's profiler for a moment, it looks like a sync executed after every opcode recalculates the framebuffer (`step_until_next_audio_buffer -> cpu::microops::read_byte_from_memory -> emulator::sync -> gpu::scanline::write_scanlines`) and uploads it to the canvas with `putImageData`, which takes ~35% of execution time. Sounds like that could be redundant work, since I'm pretty sure the canvas isn't actually redrawn until the wasm execution yields back to the browser. EDIT: also, the emulator can't really be ran in debug mode, since 1. it blows the wasm stack (presumably when creating the huge `Emulator` struct, a not uncommon issue in Rust; I had to hack in bigger stack size with linker args), and 2. is panics on int overflows `apu::step`.

catapart 2y ago
Yeah, still being on Windows 10, I almost didn't put the complaint in the comment. Lots of reasons it might just be an environment issue for me. But, in my development experience with firefox, the issues it seems to have always seem performance related (or compliance with web standards related), and I figured that even if others weren't seeing the issue, there was still probably something underlying that could be figured out. So thanks for going through the debug! It's nice to know that there may be an optimization available for the project, and it feels like it might even be an additional performance gain for chromium, if I'm understanding you right? Might be some reason they need to do that sync after every opcode, but I agree with your suspicion that the canvas is probably being over-rendered to.
- adrian17 2y ago
  FWIW, the thing I've found as the slowest single factor on my machine, might not be the slowest on yours; I might have not noticed the real cause of slowdowns on your machine at all, and you may be the only one able to diagnose it. (and for the record, I'm on W10 too.) Also, my personal experience from optimizing https://ruffle.rs/ https://ruffle.rs/ is that in many cases Firefox's wasm runtime can behave slightly faster than on Chrome for some reason; and performance issues are often instead caused by other APIs like the canvas.