Sorry for banging a (very) old drum, but we really need it to be faster than 2.3 which was significantly faster than 3.3...
Printable View
4.0.2 does not fix my performance problems.
Chrome is wicked fast, but my grid in a container in a border layout, under IE9, takes about 2-3 seconds from the time the grid data arrives until the grid and browser are usable. In the interim, the browser is hung. Under IE8 it's 5-7 seconds until it's usable.
The problem also occurs when I resize the browser window - it takes the same about of time I indicated above to re-layout the page.
stevil
P.S. Allow me to say that it IS an improvement over 4.0.1 - it's just not really usable in IE.
For those who still have doubts about the fact that 3.4 is faster than 4.0 try this and see how the theme switching in 3.4 is fast while in 4.0 is very slow:
http://dev.sencha.com/deploy/ext-3.4...mes/index.html
http://dev.sencha.com/deploy/ext-4.0...mes/index.html
Ok so here's the current state of play (Themes benchmark wise) with 4.0.2 release build
See attached Benchmarks
Compared to 4.0.1
IE 8 : 1.6 seconds slower
Chrome 11 : 0.17 faster
Firefox 3.6 : 1.4 seconds slower
Opera 11 : 0.1 seconds slower
Compared to 3.3.1
IE 8 : 11.8 seconds slower / (444% slower)
Chrome 11 : 0.6 slower / (162% slower)
Firefox 3.6 : 2.2 seconds slower / (155% slower)
Opera 11 : 1.6 seconds slower / (202% slower)
I think the big and serious question to the Dev Team is: How confident are you that you can at least get 4.x to the equivalent speed of 3.3.1 in the very short term and moving forward do you expect 4.x will have superior performance?
A lot of us using 4.x need to make big decisions (with financial implications) on current projects. Do we stick with 4.x or take the hit and go back to 3.x. 4.x gives me everything I need, except performance, the key factor being IE is totally unusable under 4.x, Themes is bad but when you bolt in a large app, it grinds to a halt.
Hats off the Dev team, EXTJS in general and the accomplishments in 4.x to date are amazing you are all obviously very skilled at your craft. However I feel that I've broken a golden rule of IT, "don't believe the hype" I've not broken that rule for many many years but had a very good feeling that that you would prove me wrong and 4.x will live up to its billing. I'm still hopeful that you can nail this performance issue, manly because I really don't want to move back to 3.x and loose the goodies in 4.x that will help make my app a success.
Best
MrSparks
I can only add to what has been said above. We are making significant, long-term financial decisions based on the expectation (hope) that ExtJS will be a serious competitive framework in the near future. But we need to understand that Sencha acknowledges and is serious about determining the cause (and solutions) of these problems, especially IE. You cannot dismiss IE - for example in our public sector market, IE is the 90% dominant web browser so it simply has to work.
We are extremely serious about this and are working round the clock to identify and correct the performance problems with IE8. Ext JS 4 has an amazing architecture and is an exceptional foundation for modern web apps but all that power has visibly regressed performance in some situations. We're totally confident that we'll get these resolved, and I hope we're doing a better job of communicating here and giving you feedback on what we're doing to address these issues.
I'm more worried that Sencha seems surprised everytime by regressions. It seems they don't know how to create automated tests.
Use cases are easy to find...
Try selecting couple of rows in the table one after another using IE9:
http://dev.sencha.com/deploy/ext-4.0...Dashboard.html
http://dev.sencha.com/deploy/ext-4.0...Dashboard.html