def-'s comments

On: Ask HN: Those making $0/month or less on side proj...

[def-]

I'm working on an open source online coop 2d platform game called DDraceNetwork: https://ddnet.org/

A lot of keeping infrastructure running, code reviews for the active developers we have, community management, some development: https://ddnet.org/news/ddnet-year-2021-in-review/

The community is the main reason for me to keep it running. We only cover server costs, but no one takes any payment to work on DDNet.

On: Time.gif (2017)

[def-]

Author here. The bandwidth usage was high with many open connections, I fixed that quickly by using LZW encoding instead of sending uncompressed data. I think you are referring to this thread: https://old.reddit.com/r/programming/comments/6t8gpq/timegif...

[Thaxll → def-]

Exactly, 5years ago already ... anyway I remember you fixed it in like 2 hours, great job.

On: Time.gif (2017)

[def-]

Try again, I think there was an unrelated issue. It load for me on Safari now, but runs super slowly in the beginning, so not as a real time clock.

[Jolter → def-]

Yep, it loads and ticks a bit slowly.

[dom96 → def-]

Safari truly becoming the new IE

[jagger27 → dom96]

Maybe in some ways. Back then devs seemed to care more about supporting IE a lot more than they do Safari now.

On: Time.gif (2017)

[def-]

Author here, my bad. There was a DoS attack against this a while ago and I set too strict limits after that. Didn't expect it to hit Hacker News again. Should be fine now.

[Kiro → def-]

What makes someone attack this?

[def- → Kiro]

I'm running an online game on the side and am thus commonly the target of DoS attacks. I guess it's related to this.

[Kiro → def-]

You mentioned in your blog post that you know who the attacker is. Do you know their motivation?

[def- → Kiro]

Destroying the fun for others and wasting time I figure. I received no demands or anything like that.

On: Permacomputing Wiki

[def-]

I was expecting this to be about Reversible Computing: https://en.wikipedia.org/wiki/Reversible_computing

On: How fast are Linux pipes anyway?

[def-]

Firefox is only Chrome-based on iOS.

[karamanolev → def-]

It's Safari-based, which is Webkit-based. Chrome is also Safari-based on iOS, because all the browsers must be. There's no actual Chrome (as in Blink, the browser engine) on iOS, at least in Play Store.

[Izkata → karamanolev]

> It's Safari-based, which is Webkit-based.

Firefox only uses Webkit on iOS, due to Apple requirements. It uses Gecko everywhere else. And I don't think it's ever been Safari-based anywhere.

[karamanolev → Izkata]

Replying to

> Firefox is only Chrome-based on iOS.

So I'm talking only about iOS. When I said it's Safari-based, I meant Webkit based, but I thought Firefox/Chrome actually pull parts of Safari on iOS. Quick research says that's wrong and they just use Webkit. Not an iOS dev, so someone can point out better sources for the 100% correct terminology.

[rwaksmunski → def-]

You mean WebKit.

On: Nim Version 1.6.6 Released

[def-]

  nim --styleCheck:error c x

This enforces the same capitalization style in your code:

  x.nim(1, 5) Error: 'my_var' should be: 'myVar'

On: macOS Setup after 15 Years of Linux

[def-]

Thanks, that's good to know, one less thing to need the mouse for!

On: macOS Setup after 15 Years of Linux

[def-]

You are right, vim is not required. SSH I misremembered, didn't install that. The rsync version on macOS is quite old (2006!) and didn't support all the options I wanted to use.

On: macOS Setup after 15 Years of Linux

[def-]

I forgot to mention that I have set hwdec=videotoolbox already, but vp9 and av01 from youtube don't get hw-decoded anyway.

> This is true for Homebrew formula as well, like the mpv formula, just not Homebrew casks. Most casks are proprietary closed source applications.

But even open source casks seem to be taken centrally instead of built again. My guess is that this is to keep the original developers' certificates.

[varenc → def-]

Thanks for the reply! Also wow I hadn't realize that Apple supports hardware decoding of Google's VP9 codec. Googling a bit, it looks like this is new in the M1 Macs. Though not sure if it supports AV01 yet? When I browse Youtube in Firefox I seem to get AV01 encoded videos. Are you sure Firefox on an M1 Mac is using hardware decoding? I wonder if they might just be using an optimized software decoder. Wish I had an M1 mac to test!

Frustratingly I couldn't find any description of videotoolbox codec support on Apple's documentation, though I did find that this VP9 constant exists: https://developer.apple.com/documentation/coremedia/kcmvideo... (but couldn't find an AV01 constant)

Re: Casks, I hadn't thought about it but you're totally right that this is probably mostly due to macOS developer signing issues.

[def- → varenc]

I get VP9 videos in Firefox apparently, probably depends on the video and your youtube settings:

Running MOZ_LOG="PlatformDecoderModule:5" firefox shows that VP9 works:

[RDD 15027: MediaSupervisor #1]: D/PlatformDecoderModule AppleVTDecoder[102edcfa0] Creating AppleVTDecoder for 3840x1920 VP9 video [Child 15023: MediaPDecoder #1]: D/PlatformDecoderModule AudioTrimmer[11fec72e0] ::PrepareTrimmers: sample[0,21000] no trimming information [RDD 15027: MediaPDecoder #2]: D/PlatformDecoderModule OpusDataDecoder[102ef4040] ::Decode: Opus decoder skipping 312 of 960 frames [Child 15023: MediaPDecoder #2]: D/PlatformDecoderModule AudioTrimmer[11fec72e0] ::HandleDecodedResult: sample[0,21000] (decoded[0,13500] no trimming needed [RDD 15027: MediaPDecoder #1]: D/PlatformDecoderModule AppleVTDecoder[102edcfa0] AppleVTDecoder: using hardware accelerated decoding

AV1 hw decoding on M1 Pro indeed is not supported even in Firefox.