Re: VS2005 seems SLOOOOOW

that came out pretty rough now didn't it - a lot worse than I really
intended
actually I really like vs2005 - you just need a beast of a machine to run it
on.


"gerry" <germ@xxxxxxxxxx> wrote in message
news:eObksAeBGHA.984@xxxxxxxxxxxxxxxxxxxxxxx
> sorry I actually meant "managed code" rather than "managed c#"
>
> true, true & true
>
> but I fail to see how a slowdown being due to the jitter is any less of a
> .NET issue.
> if running a .NET managed app involves jitting then jitting has to be
> considered as part of the performance equation.
> vs2005 would seem to show that the insignifcant overhead imposed by the
> jitter is maybe not quite so insignificant after all.
>
> on the other hand, if design is the problem, was vs.net delegated to the
1st
> semester interns or is MS just generally incompetent at software design ?
> or maybe .net is just not realistically suited for large and complex
> applications ?
> not that I actually put much credence in my own words, but it could be
> looked at in this light.
>
> no matter how you cut, or how many excuses people come up with, vs2005 is
a
> dog - the how and the why are really irrelevant to the end user that doled
> out for it.
>
> a slightly off topic question - does code get jitted the 1st time it is
> encountered EVERY time an app is run or just the 1st time ?
>
>
> "Shawn B." <leabre@xxxxxxxx> wrote in message
> news:OAEbjScBGHA.1032@xxxxxxxxxxxxxxxxxxxxxxx
> > > I am assuming that this is due to vs2005 being written in .NET ( I'm
not
> > > certain tha it is - just a guess on my part )
> > > if it is then VS2005 really puts the lie to the 'managed c# is not
> > > significantly slower than unmanaged c++' line.
> >
> > I'm not gonna pull teethe here but, the "lie" that "managed c# is not
> > significantly slower than unmanaged c++", as you put it, doesn't account
> for
> > a few things. First, the JIT service needs to compile the managed code
> the
> > first time encountered and I'm willing to bet, with something like the
VS
> > 2005 IDE, there's a **lot** of code being encountered by the time you
are
> > using it. Second, it is likely programmed with a fair amount of
C++/CLI,
> > and C#, but not one exclusively, but that doesn't matter, IL is IL, C#
or
> > C++ or VB.NET (this is why it would be nice if the compilers did some
> > front-end optimizations before the JIT optimizer kicks in, to help it
out
> a
> > little regarding performance).
> >
> > Third, the internal design of the IDE and the services it provides may
be
> > terrible, but I wouldn't blame it on .NET. I've created infrastructure
> that
> > had issues and simply be changing the architecture I gave it orders of
> > magnitude performance boosts, and I wouldn't be surprized if the 2005
IDE
> is
> > in the same position. And fourth, the RTM release of the IDE, IMO (and
> that
> > of many others) is just beta 3 or RC2. But certainly not a final
release
> > I'm not dissing them, its a great tool, but in the meantime, I'm forced
to
> > use 2003 because 2005 has been unbarable on all my machines. Granted,
I'm
> > using Premium with Team Suite, and most people I know using the Express
> > editions aren't having any problems (wonder if there's a difference)...
> but
> > still. I just can't be productive in the 2005 edition (and I've never
> > installed a beta before the release). But, I really like the tool. I
> can't
> > wait until the scheduled service pack releases, but that is 6 months
away.
> > Actually, the Beta 2 install, running on Virtual PC with 768MB RAM (in
the
> > guest OS) out performed my 2003 install on my primary machine with 1GB
> RAM,
> > and this is a 3500+ 64-bit AMD we're talking about. But the final
release
> > really took a step backwards.
> >
> >
> > Thanks,
> > Shawn
> >
> >
>
>


.

Relevant Pages

  • Re: VS2005 seems SLOOOOOW
    ... jitter is maybe not quite so insignificant after all. ... semester interns or is MS just generally incompetent at software design? ... > front-end optimizations before the JIT optimizer kicks in, ... and I wouldn't be surprized if the 2005 IDE ...
    (microsoft.public.vsnet.general)
  • Re: Running both in the IDE and on a remote host
    ... I can't get this to work in the IDE. ... > Parser Error Message: Could not load type 'eNPTest02.WebForm1'. ... > The Design and HTML views of eNP02.aspx both display fine. ...
    (microsoft.public.dotnet.framework.aspnet)
  • Running both in the IDE and on a remote host
    ... Since debug mode opens IE and loads my page from localhost, ... I can't get this to work in the IDE. ... Parser Error Message: ... The Design and HTML views of eNP02.aspx both display fine. ...
    (microsoft.public.dotnet.framework.aspnet)
  • Re: VS IDE Blues
    ... Design View. ... > with the IDE that just shouldn't have made it through rigorous testing. ... > Microsoft even use the tool they are selling to the rest of us? ... > keyboard waiting for pages to render that I don't even want rendered! ...
    (microsoft.public.dotnet.general)
Quantcast