Searched for "September 6, 2017" in this user's bookmarks. Found 42 results

« earlier   

David Owens II on Twitter: "The C++ debug numbers were: - std::list: 121ms - std::array: 1.97ms"
The C++ debug numbers were:
- std::list: 121ms
- std::array: 1.97ms

— David Owens II (@owensd) September 6, 2017
FavoriteTweet  owensd 
september 2017 by mjtsai
David Owens II on Twitter: "Final swift numbers (release build): - linked list: 375ms - array: 1.22ms (debug build): - list: 3339ms (3 seconds!) - array: 3.1ms"
Final swift numbers (release build):
- linked list: 375ms
- array: 1.22ms
(debug build):
- list: 3339ms (3 seconds!)
- array: 3.1ms

— David Owens II (@owensd) September 6, 2017
FavoriteTweet  owensd 
september 2017 by mjtsai
David Owens II on Twitter: "The release build performs just as well as the C++ version with the same sized data. The debug builds, well, not so much."
The release build performs just as well as the C++ version with the same sized data. The debug builds, well, not so much.

— David Owens II (@owensd) September 6, 2017
FavoriteTweet  owensd 
september 2017 by mjtsai
Panic Inc on Twitter: "🚚🆙 Transmit 5.0.3 is here! Enjoy a redesigned conflict resolution panel, macOS 10.13 support, tons more! Full notes: https://t.co/ciDx9zqu7g https://t.co/y6bmvlSUws"
🚚🆙 Transmit 5.0.3 is here! Enjoy a redesigned conflict resolution panel, macOS 10.13 support, tons more! Full notes: https://t.co/ciDx9zqu7g pic.twitter.com/y6bmvlSUws

— Panic Inc (@panic) September 6, 2017
FavoriteTweet  panic 
september 2017 by mjtsai
Joe Groff on Twitter: "@owensd Another linked list variant is a contiguous array of (value, next index) pairs, so you get locality and can efficiently insert by appending"
Another linked list variant is a contiguous array of (value, next index) pairs, so you get locality and can efficiently insert by appending

— Joe Groff (@jckarter) September 6, 2017
FavoriteTweet  jckarter 
september 2017 by mjtsai
David Owens II on Twitter: "@mjtsai Actually, they are even perf now. I was using Int instead of Int32, so the size was 64bits instead of 32bits like the C++ version. 🙄"
Actually, they are even perf now. I was using Int instead of Int32, so the size was 64bits instead of 32bits like the C++ version. 🙄

— David Owens II (@owensd) September 6, 2017
FavoriteTweet  owensd 
september 2017 by mjtsai
Michael Tsai on Twitter: "@owensd Thanks for updating."
Thanks for updating.

— Michael Tsai (@mjtsai) September 6, 2017
mjtsai  Twitter 
september 2017 by mjtsai
Michael Tsai on Twitter: "@daringfireball At this rate I’m expecting some Kubrick news by the end of the week."
@daringfireball At this rate I’m expecting some Kubrick news by the end of the week.

— Michael Tsai (@mjtsai) September 6, 2017
mjtsai  Twitter 
september 2017 by mjtsai
Joe Groff on Twitter: "@mjtsai I forget details, but exclusivity enforcement for passing pointers to ivars was relaxed to accommodate this kind of use. cc @pathofshrines"
I forget details, but exclusivity enforcement for passing pointers to ivars was relaxed to accommodate this kind of use. cc @pathofshrines

— Joe Groff (@jckarter) September 6, 2017
FavoriteTweet  jckarter 
september 2017 by mjtsai
Michael Tsai on Twitter: "@jckarter @pathofshrines Awesome, thanks."
Awesome, thanks.

— Michael Tsai (@mjtsai) September 6, 2017
mjtsai  Twitter 
september 2017 by mjtsai
Michael Tsai on Twitter: "@jckarter Because KVO checks are relaxed or because addresses of ivars are handled differently?"
Because KVO checks are relaxed or because addresses of ivars are handled differently?

— Michael Tsai (@mjtsai) September 6, 2017
mjtsai  Twitter 
september 2017 by mjtsai
Joe Groff on Twitter: "@mjtsai That use case should work with the latest Xcode 9 betas."
That use case should work with the latest Xcode 9 betas.

— Joe Groff (@jckarter) September 6, 2017
FavoriteTweet  jckarter 
september 2017 by mjtsai
Ryan Jones on Twitter: "Oh shit - BAD leak. - 21.3M Q1 OLED iPhone production - Touch ID *really was* on back - this shows *entire* ramp plans Oh. My. God. https://t.co/LTwjAZZ7Ov"
Oh shit - BAD leak.
- 21.3M Q1 OLED iPhone production
- Touch ID *really was* on back
- this shows *entire* ramp plans
Oh. My. God. https://t.co/LTwjAZZ7Ov

— Ryan Jones (@rjonesy) September 6, 2017
FavoriteTweet  rjonesy 
september 2017 by mjtsai
Benjamin Geskin on Twitter: "So Touch ID really was on the back at early development. Prototypes we saw. https://t.co/qGCvBQ2QVB"
So Touch ID really was on the back at early development. Prototypes we saw. pic.twitter.com/qGCvBQ2QVB

— Benjamin Geskin (@VenyaGeskin1) September 6, 2017
FavoriteTweet  VenyaGeskin1 
september 2017 by mjtsai
David Owens II on Twitter: "@mjtsai @jckarter It was a bug in the Swift code. The array should have been cleared out (or recreated). Only 2x slower than the std::vector now."
It was a bug in the Swift code. The array should have been cleared out (or recreated). Only 2x slower than the std::vector now.

— David Owens II (@owensd) September 6, 2017
FavoriteTweet  owensd 
september 2017 by mjtsai

« earlier   

Copy this bookmark:



description:


tags:



Copy this bookmark:



description:


tags: