OpenEuphoria

Re: speed of eu 2.5 + dynamic inclusion

• Posted by "Juergen Luethje" <j.lue at gmx.de> Nov 21, 2004
• 1033 views

Me wrote:

> Alexander Toresson wrote:
>
>> Now that I've tried the new interpreter, I find the parsing speed of it to be
>> unacceptable.
>> I do normally program everything on my laptop, to make myself able to work
>> anywhere.
>> It's not very fast. It's a p2 300. And because of that I get another
>> advantage:
>> I can easily see how a program of mine performs on a slower computer.
>> And I expect euphoria to work at an acceptable speed on older systems too.
>> Because I care about back-compatibility and I think it's one of the biggest
>> factors
>> one has to take into account when developing an application.
>>
>> The first application I developed was a DOS application, intended to run on a
>> 486 75.
>> I wouldn't dare of doing that now.
>>
>> Here are the times(p2 300):
>>
>> Simple win32lib application: 10s startup (eu2.4: 1.5s)
>> MEditor:                     1 min 10s startup (!) (eu2.4: 3.5s)
>
> Of course, I also see the disadvantage, but I think it mainly concerns
> to:
> o newbies
> o other people who don't have the binder/shrouder
> o the developement process of a program/library
>   ==> I hope we can inlude "shrouded" IL libraries!!

Now I know that this assumption was wrong. Rob reported that we can
*not* include pre-parsed code (= IL files) in a program. (
So I agree that for big libraries (such as Win32Lib) there is an
unbearable startup time!

<snip>

Regards,
   Juergen

-- 
Have you read a good program lately?