define('DISALLOW_FILE_EDIT', true);{"id":632,"date":"2011-07-03T01:14:19","date_gmt":"2011-07-03T06:14:19","guid":{"rendered":"http:\/\/www.unfocus.com\/?p=632"},"modified":"2011-07-25T16:51:43","modified_gmt":"2011-07-25T21:51:43","slug":"performance-benchmarks-with-air-2-7-for-ios","status":"publish","type":"post","link":"http:\/\/www.unfocus.com\/2011\/07\/03\/performance-benchmarks-with-air-2-7-for-ios\/","title":{"rendered":"Performance Benchmarks with AIR 2.7 for iOS"},"content":{"rendered":"

I’ve been working on this Benchmark based on Iain Lobb’s BunnyMark<\/a>. Being a bit confused sometimes about what things speed things up or slow things down, I didn’t want to guess anymore, so I grabbed Iain’s code base (cause I’m lazy, and didn’t want to start from scratch), and added some tests for things I suspect are slowing things down (or speeding things up). I think this will also help shed some light on why some folks see a huge gain in AIR 2.7 CPU mode, while others do not.<\/p>\n

Some caveats – this only tests instances of flash.display.Bitmap on the display list, at the size they are, moving the way they move. It’s on my list to add Blitting (I have some initial work on that done, thanks to Iain, but I need to add the rotation, and alpha settings to it), and I’d like to add a vector test, and maybe some extra sized Bitmaps (I’ve heard that makes a difference).<\/p>\n

Enough! Here are some results – quality had no effect on GPU mode, so I included only one line:<\/p>\n

Note:<\/strong> some are reporting they see a difference in GPU mode, but I still don’t. Update:<\/strong> It appears some users are confusing “Mobile Performance Tester<\/a>” with BunnyMark, which explains the discrepancy. BunnyMark is not currently in any App Store, which is one key distinguishing feature. \ud83d\ude09<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
BunnyMark Results – 500 Bunnies<\/th>\n<\/tr>\n<\/thead>\n
Alpha<\/th>\n<\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/td>\n\u2713<\/strong><\/td>\n\u2713<\/strong><\/td>\n\u2713 <\/strong><\/td>\n<\/tr>\n
Rotation<\/th>\n<\/td>\n<\/td>\n<\/td>\n\u2713<\/strong><\/td>\n\u2713<\/strong><\/td>\n\u2713<\/strong><\/td>\n<\/td>\n<\/td>\n\u2713<\/strong><\/td>\n<\/tr>\n
CaB<\/th>\n<\/td>\n\u2713<\/strong><\/td>\n\u2713<\/strong><\/td>\n<\/td>\n\u2713<\/strong><\/td>\n\u2713<\/strong><\/td>\n<\/td>\n\u2713<\/strong><\/td>\n<\/td>\n<\/tr>\n
CaBM<\/th>\n<\/td>\n<\/td>\n\u2713<\/strong><\/td>\n<\/td>\n<\/td>\n\u2713<\/strong><\/td>\n<\/td>\n<\/td>\n<\/td>\n<\/tr>\n<\/tbody>\n
iPhone 3GS – GPU<\/strong><\/th>\n<\/tr>\n<\/thead>\n
FPS<\/th>\n24<\/td>\n18<\/td>\n17<\/td>\n22<\/td>\n13<\/td>\n13<\/td>\n19<\/td>\n1<\/td>\n19<\/td>\n<\/tr>\n<\/tbody>\n
iPhone 3GS – CPU<\/strong><\/th>\n<\/tr>\n<\/thead>\n
FPS-L<\/th>\n28<\/td>\n21<\/td>\n19<\/td>\n9<\/td>\n19<\/td>\n5<\/td>\n7<\/td>\n5<\/td>\n5<\/td>\n<\/tr>\n
FPS-M<\/th>\n28<\/td>\n21<\/td>\n19<\/td>\n4<\/td>\n18<\/td>\n3<\/td>\n7<\/td>\n5<\/td>\n3<\/td>\n<\/tr>\n
FPS-H<\/th>\n28<\/td>\n21<\/td>\n19<\/td>\n3<\/td>\n18<\/td>\n2<\/td>\n7<\/td>\n5<\/td>\n2<\/td>\n<\/tr>\n
FPS-B<\/th>\n28<\/td>\n21<\/td>\n19<\/td>\n3<\/td>\n18<\/td>\n2<\/td>\n7<\/td>\n5<\/td>\n2<\/td>\n<\/tr>\n<\/tbody>\n
iPhone 4 (Retina) – GPU<\/strong><\/th>\n<\/tr>\n<\/thead>\n
FPS<\/th>\n25<\/td>\n21<\/td>\n20<\/td>\n25<\/td>\n13<\/td>\n13<\/td>\n16<\/td>\n0.5<\/td>\n16<\/td>\n<\/tr>\n<\/tbody>\n
iPhone 4 (Retina) – CPU<\/strong><\/th>\n<\/tr>\n<\/thead>\n
FPS-L<\/th>\n32<\/td>\n23<\/td>\n20<\/td>\n10<\/td>\n21<\/td>\n6<\/td>\n8<\/td>\n6<\/td>\n7<\/td>\n<\/tr>\n
FPS-M<\/th>\n32<\/td>\n23<\/td>\n20<\/td>\n5<\/td>\n20<\/td>\n3<\/td>\n8<\/td>\n6<\/td>\n3<\/td>\n<\/tr>\n
FPS-H<\/th>\n32<\/td>\n23<\/td>\n19<\/td>\n4<\/td>\n20<\/td>\n2<\/td>\n8<\/td>\n6<\/td>\n2<\/td>\n<\/tr>\n
FPS-B<\/th>\n32<\/td>\n23<\/td>\n19<\/td>\n4<\/td>\n20<\/td>\n2<\/td>\n8<\/td>\n6<\/td>\n2<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Notes about the Benchmark:<\/p>\n