VS.NET 2005 and SqlServer 2005 are both now available on MSDN for subscribers. I can't access the site right now, because it's totally burned down by the big amount of downloaders I guess, so I'll try tomorrow.

The .NET 2.0 builds of 1.0.2005.1's runtime libs are build with Beta 2 and will be updated with native builds against .NET 2.0 RTM at November 7th, as the version number for .NET 2.0 RTM is different.

Just got a license for LLBLGen too.

I cannot download the RTM's either. But Frans until the September 7 version is released it is possible to keep the .Net 2.0 beta 2 framework beside the RTM version?

Haven't installed the new LLBLGen yet, tried out the 2004 version some weeks ago. So you suggest to wait until the launch before installing a new build of LLBLGen together with the official VS 2005?

Kris wrote:

Just got a license for LLBLGen too.

I cannot download the RTM's either. But Frans until the September 7 version is released it is possible to keep the .Net 2.0 beta 2 framework beside the RTM version?

I don't think so, as they share folders.

Haven't installed the new LLBLGen yet, tried out the 2004 version some weeks ago. So you suggest to wait until the launch before installing a new build of LLBLGen together with the official VS 2005?

No, just go ahead The .NET 2.0 builds of the runtime libs are made with beta 2, my remark was targeted towards people who got the RTM of vs.net 2005 early and then want to use RTM builds of the .NET 2.0 runtime libs of llblgen pro: these are coming on nov 7. Till then it's beta 2 builds.

Frans,

The November 7 version DVD can be downloaded now on MSDN Subscriptions.

Kris wrote:

Frans,

The November 7 version DVD can be downloaded now on MSDN Subscriptions.

Yeah, I've 2 more minutes to go before it finishes . Yesterday I simply couldn't get on the site, it was flooded (looked familiar ) and this morning it was smooth as silk. Not even slowdowns. Remarkable, because normally msdn downloads were slowing down after a minute or so. (I haven't downloaded from that site in a while so it could be it was already fixed)

You are more lucky then I am; I'm behind a hard-to-breakdown firewall that caches all traffic first. So I'm condemned to download it tonight while all americans are awake and downloading too

Kris wrote:

You are more lucky then I am; I'm behind a hard-to-breakdown firewall that caches all traffic first. So I'm condemned to download it tonight while all americans are awake and downloading too

Kris, Just start the download and go home. It will be waiting for you in the AM. ;o)

Paul.Lewis wrote:

Kris wrote:

You are more lucky then I am; I'm behind a hard-to-breakdown firewall that caches all traffic first. So I'm condemned to download it tonight while all americans are awake and downloading too

Kris, Just start the download and go home. It will be waiting for you in the AM. ;o)

Paul,

I all got in last night already at my home office. Took me over 1 hour for all the images. Happy with my 20/1 broadband.

Just finished uninstalling all the other beta's. Installing will be done tonight.

Well finally downloaded everything from MSDN over the weekend, blew my laptop away and installed everything from stratch to ensure that nothing was going to spoil my VS2005 install.... well after 14 hours of installing absolutely everything I was up and running is VS2005... yeah baby :-).

I started to create some windows, console and website projects... awesome, everything works a dream and I knew it was definately worth it to go to the extreme of rebuilding the laptop in order to get the job done.

I think I started to run away with myself and decided to install some more Microsoft software, namely SQL 2005 with Business Intelligence (C00l) and Office Tools for VS2005... Well somewhere down the line I have managed to break my copy of VS2005; not so much completely broken but all the menu options to create/open websites has completely disappeared from the IDE

Well looks like I will spend another 24 hours trying to find out where they went but from what I am reading in the forums a lot of people are having a hard time with VS2005.

davisg wrote:

Well looks like I will spend another 24 hours trying to find out where they went but from what I am reading in the forums a lot of people are having a hard time with VS2005.

I've been using the beta for about two months now and I'm SO sick and freakin tired of all the crashes. However, one of my co-workers installed the RTM without formatting his computer and he's still getting the crashes.

After reading your experience I'm not sure if I'll upgrade from beta 2 just yet. Besides, I still need to figure out if TFS server was released and, if not, will the VS2005 RTM work with beta 3 of TFS server.

Bah...

alexdresko wrote:

davisg wrote:

Well looks like I will spend another 24 hours trying to find out where they went but from what I am reading in the forums a lot of people are having a hard time with VS2005.

I've been using the beta for about two months now and I'm SO sick and freakin tired of all the crashes. However, one of my co-workers installed the RTM without formatting his computer and he's still getting the crashes.

After reading your experience I'm not sure if I'll upgrade from beta 2 just yet. Besides, I still need to figure out if TFS server was released and, if not, will the VS2005 RTM work with beta 3 of TFS server.

Bah...

TFS Beta 3 Refresh went up on MSDN last night... and it works with the RTM.

We're running on WinXP x64 and are still having problems with VS.NET 2005 crashing on the RTM... especially on the "build in" cassini web server which simply "disappears" and doesn't catch any exceptions...

Marcus wrote:

We're running on WinXP x64 and are still having problems with VS.NET 2005 crashing on the RTM... especially on the "build in" cassini web server which simply "disappears" and doesn't catch any exceptions...

So what do you think about TFS thus far? It seems to be working well for my current project though I know I'm not using it to its fullest potential. Still, the company just bought a handfull of licenses and I think we're going to migrate that direction from a handfull of tools cobbled together to do the same thing. I don't think that handful of tools is quite as powerful as TFS out of the box, but it's definitely nice having everything integrated into one package.

I haven't had a crash yet. I did find a bug in the C# compiler yesterday though:

class Entity : IEntity
{
private string _foo;

public Entity(string foo)
{
_foo = foo;
}
}

class Entity2 : IEntity2
{
private string _foo;

public Entity2( string foo )
{
_foo = foo;
}
}

class Factory : IFactory<Entity> // A
{
public Entity Create()
{
return new Entity( "lalala" );
}
}

class Factory2 : IFactory<Entity2> // B
{
public Entity2 Create()
{
return new Entity2( "lalala" );
}
}

interface IFactory<T> where T : IEntityCore<T>
{
T Create();
}

interface IEntityCore<T>
{
}

interface IEntity : IEntityCore<IEntity>
{
}

interface IEntity2 : IEntityCore<IEntity2>
{
}

gives 2 CS0309 errors on the lines A and B:
Error 1 The type 'CS20Tests.Entity' must be convertible to 'CS20Tests.IEntityCore<CS20Tests.Entity>' in order to use it as parameter 'T' in the generic type or method 'CS20Tests.IFactory<T>' C:\Myprojects\VS.NET Projects\CS20Tests\CS20Tests\Program.cs 70 8 CS20Tests
Error 2 The type 'CS20Tests.Entity2' must be convertible to 'CS20Tests.IEntityCore<CS20Tests.Entity2>' in order to use it as parameter 'T' in the generic type or method 'CS20Tests.IFactory<T>' C:\Myprojects\VS.NET Projects\CS20Tests\CS20Tests\Program.cs 78 8 CS20Tests

Now, Entity implements IEntity, which inherits IEntityCore<IEntity> so this should work, because where T : IEntityCore<T> should be true: T is IEntity, and IEntity derives from IEntityCore<IEntity>. 

At least I think it's a bug.

Ladybug entry

I ran into the first intellisense 100% hang of hte IDE today. I have a very bad feeling about this... I really hate it when an IDE locks up or crashes.

Ladybug entry

Can someone try to reproduce it?

Frans, I could not repro.

As a side note, I've been using 2005 pretty extensively since beta 2. I think the thing has crashed maybe 2 or 3 times and locked up 2 or 3 times (about as much as 2003 does, or even Internet Explorer for that matter). I'm not sure if I'm just "lucky" or what.

Otis wrote:

I ran into the first intellisense 100% hang of hte IDE today. I have a very bad feeling about this... I really hate it when an IDE locks up or crashes.

Ladybug entry

Can someone try to reproduce it?

billb wrote:

Frans, I could not repro.

As a side note, I've been using 2005 pretty extensively since beta 2. I think the thing has crashed maybe 2 or 3 times and locked up 2 or 3 times (about as much as 2003 does, or even Internet Explorer for that matter). I'm not sure if I'm just "lucky" or what.

Otis wrote:

I ran into the first intellisense 100% hang of hte IDE today. I have a very bad feeling about this... I really hate it when an IDE locks up or crashes.

Ladybug entry

Can someone try to reproduce it?

hehe I think you're just lucky then I guess. MS confirmed it's a bug (it's actually already been reported on October 10th), but postponed it because they're already that far in the release cycle.

It happens because the intellisense dies in an infinite loop. To avoid it, you can create two files and it's gone. It's not taht likely you run into it, only in crappy code for testing like I wrote (and which made me run into this issue ).

The other issue I'm much more concerned about is the bug I pasted earlier in this thread, which is legit C# code but which doesn't compile.

To be honest, i have had extremely good luck with 2005 as well. But i also cant remember time when 2003 locked up or crashed though either. Now, infragistics databinding with llblgen is another matter

I did find a WSE 3.0 bug however, but thats still in beta.

Answer wrote:

To be honest, i have had extremely good luck with 2005 as well. But i also cant remember time when 2003 locked up or crashed though either. Now, infragistics databinding with llblgen is another matter

heh, let's rephrase that to "databinding in general". There's no control on the market which follows the set standard for databinding so you can write a class which binds to all those controls without problems. Not even the .NET native grids.

Admitted, some vendors make it a really tough task, and also MS could have done more in the documentation department.

MS has acknowledged the bug which hangs up the IDE though. Cyrus, who wrote the code which hangs, replied to the blogentry and appologized. What's great is that because all the detailed info is now there, you can avoid the bug altogether in your code.

Sorry, i shouldnt say strictly with LLBLGen as i made a custom entity class and it didnt work with mine either So it wasnt LLBLGens fault

I'm not sure if this has been covered in another thread yet since I cant find any related to .NET 2.0 but are you expecting any performance improvements for LLBLGen under 2.0. Just browsing around I came upon this article http://msdn.microsoft.com/data/default.aspx?pull=/msdnmag/issues/05/11/datapoints/default.aspx which says there should be huge performance gains with large datasets/datatables.

Is anyone else in general expecting/found any performance benefits with 2.0?

tmpreston wrote:

I'm not sure if this has been covered in another thread yet since I cant find any related to .NET 2.0 but are you expecting any performance improvements for LLBLGen under 2.0. Just browsing around I came upon this article http://msdn.microsoft.com/data/default.aspx?pull=/msdnmag/issues/05/11/datapoints/default.aspx which says there should be huge performance gains with large datasets/datatables.

I don't use datatables, but I expect a performance improvement with large collections as I will remove a bottleneck from the current code by architecting it differently. Though if you work with objects, instead of datatables, it will always be a little slower, due to the overhead in the entity object itself, but the overhead will be brought down to the absolute minimum.

Has anyone else experienced a dramatic slowdown on SQL Server 2005 vs 2000 when Refreshing the Catalog in LLBLGen designer?

Its taking me about 20 seconds to refresh in SQL Server 2005 whereas it was 2-3 seconds in SQL Server 2000.

Do you have the custom properties checkbox checked at the bottom of the dialog?

Otis wrote:

Do you have the custom properties checkbox checked at the bottom of the dialog?

Nope...

And the CPU on the server (dual proc Xeon) is 50% busy during the refresh...

I use different queries in some areas for sqlserver 2005, because of the fact that you can have a schema without a user in sqlserver 2005.

I think especially the PK retrieval for tables, can be a problem. On sqlserver 2000, the query is:

 SELECT db_name() as TABLE_CATALOG,
            user_name(t_obj.uid) as TABLE_SCHEMA,
            t_obj.name as TABLE_NAME,
            col.name as COLUMN_NAME
 FROM       sysobjects  c_obj,
            sysobjects  t_obj,
            syscolumns  col,
            master.dbo.spt_values  v,
            sysindexes  i
 WHERE
            permissions(t_obj.id) != 0
            and c_obj.xtype in ('PK')
            and t_obj.id = c_obj.parent_obj
            and t_obj.xtype  = 'U'
            and t_obj.id = col.id
            and col.name = index_col('[' + user_name(t_obj.uid) + '].[' + t_obj.name + ']', i.indid, v.number)
            and t_obj.id = i.id
            and c_obj.name  = i.name
            and v.number > 0
            and v.number <= i.keycnt
            and v.type = 'P'
            and user_name(t_obj.uid) = @sSchemaName
            and t_obj.name = @sTableName

on sqlserver 2005:

 SELECT  db_name() AS TABLE_CATALOG, s.name AS TABLE_SCHEMA, t.name AS TABLE_NAME, c.name as COLUMN_NAME
 FROM   sys.key_constraints k 
        inner join sys.tables t on 
        k.parent_object_id = t.object_id
        inner join sys.schemas s
        on t.schema_id = s.schema_id
        inner join sys.index_columns i
        on t.object_id = i.object_id
        and k.unique_index_id = i.index_id
        inner join sys.columns c
        on i.column_id = c.column_id
        AND t.object_id = c.object_id
 WHERE  k.type='PK'
        AND s.name = @sSchemaName
        AND t.name = @sTableName

Unique constraints are almost the same as that last query, for sqlserver 2005.