Hi, I'm Alex (short for Alexander).
- Blog: kean.blog
- Built Nuke, Pulse, and other popular frameworks and tools
- Developed the initial CreateAPI version which is now maintained by @cookpad
Contacts
- Mastodon: @[email protected]
Objective-C libjpeg-turbo wrapper
License: Other
Hi, I'm Alex (short for Alexander).
Contacts
Does this wrapper decodes multiple images concurrently OR serially ? If it decodes serially then how can make it concurrent ?
Hi, I've tried libjpeg-turbo benchmark on iPhone 6 (with updated libjpeg-turbo to 1.4.0) and below are results, as you can see in some cases libjpeg-turbo is slower than AFNetworking, any ideas why?
2015-03-31 10:37:12.013 ---------------------------------------------------
2015-03-31 10:37:12.015 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (0.250000)
2015-03-31 10:37:12.114 SDWebImageDecoder:
2015-03-31 10:37:12.167 benchmark [46501786 ns, 46.50 ms, 0.0465 s]
2015-03-31 10:37:12.167 AFNetworking
2015-03-31 10:37:12.207 benchmark [40151786 ns, 40.15 ms, 0.0402 s]
2015-03-31 10:37:12.208 DFJPEGTurbo:
2015-03-31 10:37:12.251 benchmark [42878870 ns, 42.88 ms, 0.0429 s]
2015-03-31 10:37:12.251 ---------------------------------------------------
2015-03-31 10:37:12.251 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (0.500000)
2015-03-31 10:37:12.289 SDWebImageDecoder:
2015-03-31 10:37:12.330 benchmark [40412370 ns, 40.41 ms, 0.0404 s]
2015-03-31 10:37:12.330 AFNetworking
2015-03-31 10:37:12.366 benchmark [35781786 ns, 35.78 ms, 0.0358 s]
2015-03-31 10:37:12.366 DFJPEGTurbo:
2015-03-31 10:37:12.405 benchmark [39091786 ns, 39.09 ms, 0.0391 s]
2015-03-31 10:37:12.406 ---------------------------------------------------
2015-03-31 10:37:12.406 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (0.750000)
2015-03-31 10:37:12.452 SDWebImageDecoder:
2015-03-31 10:37:12.485 benchmark [33615286 ns, 33.62 ms, 0.0336 s]
2015-03-31 10:37:12.486 AFNetworking
2015-03-31 10:37:12.519 benchmark [32988328 ns, 32.99 ms, 0.0330 s]
2015-03-31 10:37:12.519 DFJPEGTurbo:
2015-03-31 10:37:12.566 benchmark [47049995 ns, 47.05 ms, 0.0470 s]
2015-03-31 10:37:12.567 ---------------------------------------------------
2015-03-31 10:37:12.567 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (1.000000)
2015-03-31 10:37:12.682 SDWebImageDecoder:
2015-03-31 10:37:12.736 benchmark [51647286 ns, 51.65 ms, 0.0516 s]
2015-03-31 10:37:12.737 AFNetworking
2015-03-31 10:37:12.777 benchmark [39616411 ns, 39.62 ms, 0.0396 s]
2015-03-31 10:37:12.777 DFJPEGTurbo:
2015-03-31 10:37:12.871 benchmark [93912786 ns, 93.91 ms, 0.0939 s]
2015-03-31 10:37:12.871 ---------------------------------------------------
2015-03-31 10:37:12.872 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (0.250000)
2015-03-31 10:37:12.924 SDWebImageDecoder:
2015-03-31 10:37:12.957 benchmark [32692745 ns, 32.69 ms, 0.0327 s]
2015-03-31 10:37:12.958 AFNetworking
2015-03-31 10:37:12.990 benchmark [31745370 ns, 31.75 ms, 0.0317 s]
2015-03-31 10:37:12.990 DFJPEGTurbo:
2015-03-31 10:37:13.028 benchmark [37825995 ns, 37.83 ms, 0.0378 s]
2015-03-31 10:37:13.028 ---------------------------------------------------
2015-03-31 10:37:13.028 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (0.500000)
2015-03-31 10:37:13.080 SDWebImageDecoder:
2015-03-31 10:37:13.116 benchmark [35420078 ns, 35.42 ms, 0.0354 s]
2015-03-31 10:37:13.116 AFNetworking
2015-03-31 10:37:13.153 benchmark [35493453 ns, 35.49 ms, 0.0355 s]
2015-03-31 10:37:13.153 DFJPEGTurbo:
2015-03-31 10:37:13.193 benchmark [40141620 ns, 40.14 ms, 0.0401 s]
2015-03-31 10:37:13.193 ---------------------------------------------------
2015-03-31 10:37:13.194 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (0.750000)
2015-03-31 10:37:13.257 SDWebImageDecoder:
2015-03-31 10:37:13.294 benchmark [37388245 ns, 37.39 ms, 0.0374 s]
2015-03-31 10:37:13.295 AFNetworking
2015-03-31 10:37:13.329 benchmark [34370203 ns, 34.37 ms, 0.0344 s]
2015-03-31 10:37:13.329 DFJPEGTurbo:
2015-03-31 10:37:13.376 benchmark [46458828 ns, 46.46 ms, 0.0465 s]
2015-03-31 10:37:13.376 ---------------------------------------------------
2015-03-31 10:37:13.376 Benchmarking JPEG decompression with image size: (2048.000000, 1536.000000), compression quality: (1.000000)
2015-03-31 10:37:13.510 SDWebImageDecoder:
2015-03-31 10:37:13.571 benchmark [58246203 ns, 58.25 ms, 0.0582 s]
2015-03-31 10:37:13.574 AFNetworking
2015-03-31 10:37:13.616 benchmark [40694995 ns, 40.69 ms, 0.0407 s]
2015-03-31 10:37:13.616 DFJPEGTurbo:
2015-03-31 10:37:13.711 benchmark [94622495 ns, 94.62 ms, 0.0946 s]
2015-03-31 10:37:13.711 ---------------------------------------------------
2015-03-31 10:37:13.711 Benchmarking JPEG decompression with image size: (8192.000000, 6144.000000), compression quality: (0.250000)
2015-03-31 10:37:14.314 SDWebImageDecoder:
2015-03-31 10:37:16.101 benchmark [1785519370 ns, 1785.52 ms, 1.7855 s]
2015-03-31 10:37:16.107 AFNetworking
2015-03-31 10:37:16.925 benchmark [818056370 ns, 818.06 ms, 0.8181 s]
2015-03-31 10:37:16.925 DFJPEGTurbo:
2015-03-31 10:37:17.500 benchmark [574367661 ns, 574.37 ms, 0.5744 s]
2015-03-31 10:37:17.500 ---------------------------------------------------
2015-03-31 10:37:17.500 Benchmarking JPEG decompression with image size: (8192.000000, 6144.000000), compression quality: (0.500000)
2015-03-31 10:37:18.523 SDWebImageDecoder:
2015-03-31 10:37:19.824 benchmark [1300961203 ns, 1300.96 ms, 1.3010 s]
2015-03-31 10:37:19.827 AFNetworking
2015-03-31 10:37:20.755 benchmark [927528620 ns, 927.53 ms, 0.9275 s]
2015-03-31 10:37:20.756 DFJPEGTurbo:
2015-03-31 10:37:21.413 benchmark [657550578 ns, 657.55 ms, 0.6576 s]
2015-03-31 10:37:21.413 ---------------------------------------------------
2015-03-31 10:37:21.413 Benchmarking JPEG decompression with image size: (8192.000000, 6144.000000), compression quality: (0.750000)
2015-03-31 10:37:22.005 SDWebImageDecoder:
2015-03-31 10:37:23.022 benchmark [1015919870 ns, 1015.92 ms, 1.0159 s]
2015-03-31 10:37:23.022 AFNetworking
2015-03-31 10:37:23.931 benchmark [909446995 ns, 909.45 ms, 0.9094 s]
2015-03-31 10:37:23.932 DFJPEGTurbo:
2015-03-31 10:37:24.657 benchmark [725024036 ns, 725.02 ms, 0.7250 s]
2015-03-31 10:37:24.657 ---------------------------------------------------
2015-03-31 10:37:24.657 Benchmarking JPEG decompression with image size: (8192.000000, 6144.000000), compression quality: (1.000000)
2015-03-31 10:37:26.502 SDWebImageDecoder:
2015-03-31 10:37:28.468 benchmark [1965668495 ns, 1965.67 ms, 1.9657 s]
2015-03-31 10:37:28.472 AFNetworking
2015-03-31 10:37:29.545 benchmark [1072792828 ns, 1072.79 ms, 1.0728 s]
2015-03-31 10:37:29.545 DFJPEGTurbo:
2015-03-31 10:37:30.762 benchmark [1216044953 ns, 1216.04 ms, 1.2160 s]
2015-03-31 10:37:30.860 Received memory warning.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.