In my benchmark using the Dynamic library[1], my LCG stuff is much faster:
10 ReferenceType objects (Person):.
Activator:00:00:00.0843059
Activator: 00:00:00.3047689
Dynamic: 00:00:00.0179958
[1] http://www.codeplex.com/Dynamic
--
"I am Dyslexic of Borg. Resistors are fertil
Hi,
My issues aren't with the returned serialisable classes - it's with the
remote classes. Strangely, I can create an instance of my remote
factory, but when the factory tries to return an instance of another
MarshalByRef class, it fails with a "System.InvalidCastException: Return
argument has an
Sébastien Lorion <[EMAIL PROTECTED]> wrote:
> Ok, makes sense... But the question of Frans then remains unanswered.
> Without measuring, I would also have guessed that the generic version
> would be faster.
The cost differential for value types is even bigger. Same test program
as last time, only
Ok, makes sense... But the question of Frans then remains unanswered.
Without measuring, I would also have guessed that the generic version
would be faster.
Sébastien
www.sebastienlorion.com
On 2/28/07, Barry Kelly <[EMAIL PROTECTED]> wrote:
Barry Kelly <[EMAIL PROTECTED]> wrote:
> Your result
Barry Kelly <[EMAIL PROTECTED]> wrote:
> Your results appear to be anomalous because typeof(T) where T is a
> non-public type invokes more security checking. I get the following
> results:
Oops: I meant Activator.CreateInstance(typeof(T)) invokes more checking,
not typeof(T), where T is non-publi
Sébastien Lorion <[EMAIL PROTECTED]> wrote:
> Funny, I get exactly the inverse behavior both in debug and release ... ?!
> class MyReferenceType
Your results appear to be anomalous because typeof(T) where T is a
non-public type invokes more security checking. I get the following
results:
Ac
Hmm, my code:
using System;
using System.Collections.Generic;
using System.Text;
using System.ComponentModel;
using System.Data;
using System.Xml;
using System.Text.RegularExpressions;
using System.Diagnostics;
using System.IO;
using System.Collections;
using System.Configuration;
using System.Dat
Funny, I get exactly the inverse behavior both in debug and release ... ?!
using System;
using System.Collections.Generic;
using System.Text;
namespace ConsoleApplication1
{
class Program
{
static void Main(string[] args)
{
const int Count = 100;
DateTime now = DateTime.No
I was benchmarking some type instantiation code, and I stumbled upon this:
when I instantiate in a loop 1,000,000 times a class with the code:
MyType t = Activator.CreateInstance();
it takes 2800ms or thereabout.
When I use:
MyType t = (MyType)Activator.CreateInstance(typeof(MyType));
it takes
Okay, since this months-old thread has now turned towards AOP, I'll
chime in, since I've been working on a proxy-based AOP tool (and its
uses) for the last two years.
As I said the profiling API is not an optimal way of doing things (in
particulaer because there is no good way of updating symbol
Yes, sorry Public without Shared.
Thanks for pointing that out Ryan.
Ben
On 2/28/07, Ryan Heath <[EMAIL PROTECTED]> wrote:
> I've changed the variable declaration to "Public" and ran a test
> (making use of System.Threading.Thread.Sleep) and with the Shared
> declaration I was able to mimic th
I've changed the variable declaration to "Public" and ran a test
(making use of System.Threading.Thread.Sleep) and with the Shared
declaration I was able to mimic the behavior and with Public it seems
to be "thread safe".
With "Public" you mean, without the "Shared" keyword?
If not, then it is
Hi
I had a similar problem like this and managed to solve this by
implementing my own binder. How do you get hold of the object references
that you are unable to cast? In my case I couldn't do this in a certain
situation:
[Serializable]
class MyType {
...
public object Clone() {
Mem
I think I have found the problem.
In my base class I had some variabled declared as "Protected Shared ".
Multiple client sessions would create instances of classes derived
from this base class and assign values to the variables (via exposed
public properties). There was a situation yesterday wh
14 matches
Mail list logo
