on 1/5/02 3:21, REALbasic Network Users Group at
realbasic-nug at lists dot realsoftware dot com wrote:
>>> Ruslan Zasukhin is the developer of Valentina, a high-performance
>>> database engine available to RB coders as a plugin. His tests suggest
>>> that, in its current state, RB is about 10% slower than C++. As RB is
>>> worked over by its developers, that percentage will likely shrink.
>>
>> That isn't always true. Sometimes yes, but sometimes, no.
>
> Right. I assume that Ruslan meant 10% on average, whatever that means.
I have test only loop in 100,000 which calls plugin function.
And I have compare speed of this loop to C++ code.
So I have test seep of function call located in plugin.
This was the most interesting thing to me.
Other IDEs do calls in 5-20 times slower.
Although in this loop was many other REALbasic code: calculation of
variables and strings...And this also was in avg on 10% slower of C++...
I can believe that there is NOT optimal operations yet in RB.
But in general I was very glad to see such speed.
And this was for RB 2. May be now even better.
--
Best regards,
Ruslan Zasukhin
-------------------------
Paradigma.
e-mail: ruslan at paradigmasoft dot com
web : http://www.paradigmasoft.com
To subscribe to the Valentina mail list
send a letter to valentina-on at lists dot macserve dot net
|