valentina 2

[Olivier]

New tests :

Stopwatches indicated below are valid ONLY as comparison between 
Valentina 1 and 2.
They does not need to take them theoretically. They do not calculate 
the RAW speed of Valentina 1 or 2.
Because there are several factors which SLOW DOWN Adds (display of a 
progressbar, several methods of processing of strings before 
"addrecord", strings got back by a listBox, etc.).

BUT they are the SAME factors for Valentina 1 and Valentina 2.

Valentina 1 cache : 20 Mo
Valentina 2 cache : 50 Mo
DebugLevel disabled.

3 string fields. 1 boolean field. 2 strings are indexed. 1 string is 
UNIQUE.

Add x lines		With cursor		With API		Valentina 1

100 lines			2 s				2 s			2 s
600				2				2			2
3000			9				8			3
5000			17				16			3
8000			35				34			4
17000							109			6
25000							180			8
36000			329				322			10

There the cursor is pratically also fast as the API.

-------------------

NO unique or indexed field :

Add x lines		With cursor		With API		Valentina 1

3000 lines		3 s				3 s		
8000			6				5
17000			13				10			4
36000			32				22			7

The difference of speed between Valentina 1 and 2 is sharply less 
important here.
Besides, the speed of Valentina seems proportional among adds.
There is thus a problem in the processing of the indexs / unique by 
Valentina 2.
Here, API is a little faster than the cursor.

olivier




Le 10 janv. 05, à 05:52, Olivier a écrit :

>>
>> Oliver, another point.
>>
>> When you do this bench you must make sure that
>>
>>      V4RB_Log File is disabled !
>>
>> So is it ?
>>
> Arrghh....no.
>
> I do new bench now.
>
> olivier
>
> _______________________________________________
> Valentina-beta mailing list
> Valentina-beta at lists.macserve.net
> http://lists.macserve.net/mailman/listinfo/valentina-beta
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 1843 bytes
Desc: not available
Url : http://lists.macserve.net/pipermail/valentina-beta/attachments/20050110/cd9ff58a/attachment.bin