Geoff Canyon wrote:
On Tue, Nov 24, 2015 at 6:46 AM, Ali Lloyd wrote:
*Native string performance*
The performance of native string operations has been vastly improved, with
many achieving similar speeds to those in 6.7.
This is definitely improved. I had already been running tests using list
parsing, and I just ran that test in 8.0dp10, and while it is still slower
than 6.7.3, it is *much* faster than 7, or earlier releases of 8:
6.7.3: 1.77 seconds
7.0.3: 32.95 seconds
8.0dp1: 32.31 seconds
8.0dp10: 2.17 seconds
This is VERY encouraging.
Given the scope of what the team has been doing, from Cocoa to Unicode
and the hundreds of other enhancements and refinements along the way
(I've been surprised no one else is as excited about the addition of the
mobileGetLaunchData function), it would have been a bit much to expect
all that could come together and also be well optimized in the first pass.
After all, Python took a few builds moving from v2 to v3 to get
performance back, and even with fewer contributors it seems the LC team
has accomplished as much in about as much time.
Kudos to Malte for spearheading the benchmarking effort that helped
identify some of the specific string operations affected.
And of course kudos to the team for these optimizations, proving that
even if it takes a little time it's possible to add sweeping new
features and still maintain good performance.
--
Richard Gaskin
Fourth World Systems
Software Design and Development for the Desktop, Mobile, and the Web
____________________________________________________________________
ambassa...@fourthworld.com http://www.FourthWorld.com
_______________________________________________
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
