From eito at ubit.com Thu Nov 24 20:22:08 2005 From: eito at ubit.com (eito@ubit.com) Date: Fri, 25 Nov 2005 10:22:08 +0900 Subject: [Amrita2-users] template caching Message-ID: <20051125012208.GH2753@DEV-SI1.ubit.com> Dear amrita2 developers, I am wondering when template caching feature like amrita1 has will be implemented. I recently conducted some performance comparison research between amrita1 and amrita2. Amrita2 is faster by about 20% than amrita1. However, amrita1 with template caching option on is much faster, especially, when model data has array that contains many values. Then, I thought that if amrita2 has tempate caching feature, amrita2 would run faster than amrita1 with caching. Regards, :eito From tnakajima at brain-tokyo.jp Thu Nov 24 22:32:25 2005 From: tnakajima at brain-tokyo.jp (Taku Nakajima) Date: Fri, 25 Nov 2005 12:32:25 +0900 Subject: [Amrita2-users] template caching In-Reply-To: <20051125012208.GH2753@DEV-SI1.ubit.com> References: <20051125012208.GH2753@DEV-SI1.ubit.com> Message-ID: <438685C9.3040800@brain-tokyo.jp> Sorry for long interval. I have just re-started developing Amrita2 this week. I think Amrita2 has too many features for the first release. And I don't have enough time to make all features completed. So I'm now reducing some features of Amrita2 to release a stable version earlier. I think I can show it in one or two weeks. eito at ubit.com wrote: > Dear amrita2 developers, > > > I am wondering when template caching feature like amrita1 has will be > implemented. > > I recently conducted some performance comparison research between amrita1 > and amrita2. Amrita2 is faster by about 20% than amrita1. However, amrita1 > with template caching option on is much faster, especially, when model > data has array that contains many values. > > Then, I thought that if amrita2 has tempate caching feature, amrita2 would > run faster than amrita1 with caching. I agree with you. Amrita2 produces more concise code than amrita1. So if both have caching feature, Amrita2 will run faster because it will read compiled Ruby code faster than amrita1 and the code will run faster too. But I think it's important for me release 2.0. I will implement cache after release 2.0 .