[V4RB] Compiling PPC on an Intel machine -- more

[Bill Mounce]

Hmm. I copied all my programming over to a PPC, recompiled the app,  
and now everything works. I think there is a problem compiling PPC  
apps on an Intel machine. Can anyone confirm that they actually have  
done this?

Bill







On Oct 19, 2007, at 3:32 PM, Joseph Morgan wrote:

>
> On Oct 19, 2007, at 11:31 AM, Ruslan Zasukhin wrote:
>
>>> Are there are some specific speed advantages by going the  
>>> valentina way?
>>
>> A lots.
>>
>> For example
>>
>> 1)  Vlong.value as Ulong
>>     Vfield.value as Variant
>>
>>     RBDB have only Variant way.
>>    But when you know EAXCT type of value this speed up things in  
>> times.
>>     and problem here in fact that RB's type() function of VARIANT
>>     is about 100 times slower then it must be.
>>     I have report them this 2-3 years ago but not fix still.
> This hasn't been an issue for me in the past since I rarely use  
> VARIANT.
> I was told it was really slow so I always made sure to almost always
> know what type I am dealing with.
>
> In fact the only time I am using VARIANT, is when I am inserting  
> only one
> record at a time or reading in only one record at a time. And since  
> only
> one at a time is very quick, it seems VARIANT speed differences  
> would not
> make a difference.
>
>> 2) or ability in SqlSelect() specify array of binds.
>>     this can speed up thins up to 20 times
> I am not sure what a bind is, maybe I should look into this?
> Is this similar to binary link?
>
>
>
>
>
> _______________________________________________
> Valentina mailing list
> Valentina at lists.macserve.net
> http://lists.macserve.net/mailman/listinfo/valentina