Visual Studio 2013 Released

jones_supa writes "Final releases of Visual Studio 2013, .NET 4.5.1, and Team Foundation Server 2013 are now available. As part of the new release, the C++ engine implements variadic templates, delegating constructors, non-static data member initializers, uniform initialization, and 'using' aliases. The editor has seen new features, C++ improvements and performance optimizations. Support for Windows 8.1 has been enhanced and the new XAML UI Responsiveness tool and Profile Guided Optimization help to analyze responsiveness in Windows Store apps. Graphics debugging has been furthered to have better C++ AMP tools and a new remote debugger (x86, x64, ARM). As before, MSDN and DreamSpark subscribers can obtain the releases from the respective channels, and the Express edition is available zero cost for all."

Comments

Who cares?

[faragon]

Visual Studio 2010 was already bloated and brain-dead. TFS sucks and the Git integration is poor. Not worth it, in my opinion.

Re:Who cares?

[Hsien-Ko]

So what new form of vendor lock-in does VS2013 do? Kick out XP SP3 users from executing?

Re: Who cares?

[Anonymous Coward]

Both VS and TFS 2012 were massive improvements over the 2010 editions for what its worth. 2013 seems more iterative and superfluous.

Re:Who cares?

[Anonymous Coward]

" that second part is a courtesy "

That's only what Apple users consider "courtesy," but since you explained it already it goes without saying.

-- Ethanol-fueled

Re: Who cares?

[Anonymous Coward]

I disagree

VS2012 was massive improvement in terms of features. Unfortunately, those features consumed A LOT of resources, to the point it was completely unusable on my computer (on start, after a few minutes, VS2012 would show a message saying "your computer is too slow for VS2012").

VS2013 is as feature rich (actually, more) than VS2012 *and* it consumes LESS resources than 2010. I have been using it since the Preview (with ReSharper and a few more plugins) and it's great.

Re: Who cares?

[Anonymous Coward]

I rather use 2010 as VS2012 IS EYESORE..

Re: Who cares?

[Anonymous Coward]

2012 was fine if you use the registry setting to fix the capital letters in the menu bar, and update 2 added a "Blue" theme. Then it looks like 2010, but with all the new functionality.

Re: Who cares?

[Barlo_Mung_42]

I find it easier on the eyes. There is a dark theme that makes my eyes feel less tired after hours of use.

Re: Who cares?

[LurkerXXX]

I've been using VS2012 on a 5 year old laptop, that was midrange at best when new. The requirements don't seem that steep.

Re: Who cares?

[ATMAvatar]

My experience was the opposite. VS2012 was night-and-day faster than VS2010 on my work machine, if only because it was much better at multi-threading. My peers had a similar experience. Perhaps my experience was different due to the fact that I don't run that many plug-ins.

VS2013 is an improvement as well, so I am curious to see how quickly I can get an upgrade approved.

Re:Who cares?

[ILongForDarkness]

VS2012 doesn't support XP as far as I know since .Net 4.5 doesn't run there and the main thing with VS2012 was support for Metro. So that ship has sailed.

I don't think it is vendor lock in to expect developers to be using a OS that is less than 10 years old.

Re:Who cares?

[Anonymous Coward]

Yes, VS2012 and VS2013 still support XP. I'm running some stuff on Server 2003 right now, that I compiled with VS2013RC.

Here's how it's done:
Windows XP Targeting with C++ in Visual Studio 2012

Works exactly the same in VS2013 also.

Re:Who cares?

[PmanAce]

How so? I'm using 2012 at work and feels just as good as any version so far. Debugging unit tests couldn't be easier, resharper helps a ton, etc. What bloat are you talking about that is preventing you from actually working?

Re: Who cares?

[inasity_rules]

Thank you. I had tried a while back to fix that in office, and failed. This will make Visual Studio 2012 usable for me.

Re:Who cares?

[Anonymous Coward]

VS2012 cannot be installed on XP. Original VS2012 also could not build for XP.
VS2012 SP1 added functionality to build for XP, but it's not enabled by default when creating a new project.

Re:Who cares?

[Anonymous Coward]

Ignore that, I clearly didn't think before I wrote!

-- Ethanol-fueled

Re: Who cares?

[Xest]

It was the UI that made me hate 2012. The largely black and white themed icons slowed me down in finding the file I wanted in solution explorer in larger projects which was fucking annoying. It took some used to having the menu bars shouting at you all the time too.

I also hate the fact that it's a step backwards feature wise in some ways also, no more automated generation of unit tests for a class when using MSTest for example. I've also found NuGet can be quite annoying with it breaking once or twice and me manually having to fix my project.

I've gotten used to it now, but all in all I preferred 2010. I never had any performance issues with it in the first place, so I've not noticed any kind of speed up in 2012 because there was seemingly no human perceivable speed up to be had on my hardware.

Still, I'll download 2013 tonight and have a look. It sounds like it's addressed some of the problems I had with 2012 at least and the OP's post was stupid.

VS2010 bloated? Where exactly. Braindead? What does that even mean? It's stupid? So why does it have the best Intellisense and automated refactoring tools on the market then? TFS sucks? Justification please? Git integration is poor? Go fix it then. That's the open source mantra isn't it?

Re: Who cares?

[nogginthenog]

I'm running it on a 4 year old Celeron laptop. Works great.

Re:Who cares?

[Anonymous Coward]

Ha ha, indeed. In my company we all are coding in notepad++ or xemacs and just using the VC++ compiler and libraries from a build script..

Re: Who cares?

[mrchaotica]

The real question is, why can't the Visual Studio programmers just use Windows Forms or XAML or whatever the Hell it is every other Windows application developer is "supposed" to use these days, so that VS looks and works like a "normal" application?

Re:Who cares?

[ILongForDarkness]

Sorry I should have been more specific VS 2012 doesn't run on XP as far as I know, you can target the platform but you can't run on it. You also give up features in .Net > 4 when targeting downwards which kind of sucks (async is your friend).

Re:Who cares?

[Anonymous Coward]

Anyone who needs git integration when there is a command line that is more powerful should turn in his programming card. The GUI is fine in VS, please get a modern computer.

Re:Who cares?

[TheyCallMeBruce]

Who writes to the Desktop anymore anyway. Since the Borg (M$) assimilated all the other Desktop authoring tools and forced developers into a bloated framework and everything moved away from this high handed environment and into the open source multi vendor choice driven free reining arena - who cares.

Re: Who cares?

[man_of_mr_e]

Nuget gets broken when using the standard mode. Part of the problem is that when you check in, it doesn't automatically select all files for checkin, and most people don't pay attention.

This is why the new(er) Package Restore mode works so much better (on top of not filling up your version control database with binaries).

The UI was largely addressed after a couple of months by a new version of the Theme Switcher and a hack to add in color icons. Many of the icons in 2013 are still monochrome, but a large number of important ones are color, and that helps.

Re: Who cares?

[man_of_mr_e]

Actually, the real question is... WTF are you talking about?

Re: Who cares?

[mrchaotica]

I guess I don't know... I was working off the premise that Windows applications ought to (and do) have a "standard," consistent look-and-feel, but then I just looked through the UIs of the 10 or so applications I have open right now and pretty much every single one of them is different.

Maybe the follow-up question should be "why can't Microsoft be less schizophrenic about UI standards?"

Re: Who cares?

[Anonymous Coward]

>Maybe the follow-up question should be "why can't Microsoft be less schizophrenic about UI standards?"

They were... back in the 90s. It was pretty good. Then they decided standards weren't important and their UI got ugly and confusing. I never understood it. The sad thing is that it's getting worse not better. Windows 8 is the ugliest version of Windows since Windows 2, and Windows 2 was ugly because it only supported 16 colors.

Re: Who cares?

[shutdown -p now]

If you're using C++, VS 2013 comes with a much better compiler in terms of standards compliance, with more C++11 features (notably, variadic templates), and even some small pieces of C++14.

Re:Who cares?

[Anonymous Coward]

Nothing wrong with having DVCS support in an IDE. It is nice that better IDE's can add/remove files from git automatically.

That VS doesn't have support for even a moderately good DVCS system is a black mark against it. That is doesn't support generic programming is the death knell.

WOW

[Anonymous Coward]

All this value free for the express edition! gotta thank GNU, if it weren't for them we'd be milked for way less stuff.

Re:WOW

[UnknownSoldier]

I know.

Used to use Dev Studio 6 heavy back in the day. Each Visual Studio just gets more and more bloated. :-(

Switched to Vim couple years back. One of the best text editors I have every used. (Note: They _all_ suck in some ways.)

At least, I'm freed from the Microsoft's VisualStudio & Apple's XCode lock-in now.

Re:WOW

[Anonymous Coward]

Switched to Emacs couple years back. One of the best text editors I have ever used.

Fixed that for you. Both were errors.

Re:WOW

[Anonymous Coward]

VSINATE (Visual Studio Is Not A Text Editor)

Re:WOW

[Gadget_Guy]

All this value free for the express edition! gotta thank GNU, if it weren't for them we'd be milked for way less stuff.

Actually, you can thank the Microsoft's own Platform SDK for all this free value. This included a free C++ compiler, and was released at the start of this century. It was originally for MSDN subscribers, but it was released to the public for anyone to download. If you want to thank anyone for this inital free release, I think it would be Watcom C++ which was released as open source in 2000 after commercial development stopped. At the time that was a much bigger competitor to Microsoft's dev kits than any GNU software.

Re: WOW

[Anonymous Coward]

emacs is a great os but only a mediocre editor

Re:WOW

[Anonymous Coward]

There's a commercial plugin which replaces Visual Studio's editor and which implements all of Vim's capabilities.

I haven't used Windows in over 10 years, and have never used Visual Studio. A friend who was in the same boat as me was forced to use VS for C# .Net development, and absolutely loved that plugin.

Re:WOW

[nojayuk]

I vaguely recollect someone years ago wrote an BASIC interpreter in Excel. It would even generate ASCII graphics. It wasn't fast but...

Re:WOW

[TranquilVoid]

There's a free plugin for VS2010 and on that replaces the editor with a vim-style one. It's not quite as nice as using gvim itself but really is fantastic. I don't know how developers can live with the standard editor's find tool.

The author orginally wrote it to teach himself F# - http://visualstudiogallery.msdn.microsoft.com/59ca71b3-a4a3-46ca-8fe1-0e90e3f79329

Re:WOW

[UnknownSoldier]

Emacs is OK; I would rank it #2. I prefer Vim over Emacs though. Vim feels more like an extension of my mind.

Programs!

[TechyImmigrant]

I look back with fondness for the times when a program was a set of instructions and declarations written in a programming language, rather than am odd derivative of C++ tied to a billion files in various XML schemas.

Re:Programs!

[SJHillman]

I look forward to the time when I can tell my computer, in plain English, what I need it to do and it just does it without having to program a specific application to do a specific function.

Re:Programs!

[TechyImmigrant]

Be careful what you ask for. Computers are vindictive. One that has free reign to misinterpret what you are asking for it going to be nothing but trouble.

Re:Programs!

[microbox]

If you see no value in fancy technologies, then don't use them.

Re:Programs!

[OzPeter]

I look back with fondness for the times when a program was a set of instructions and declarations written in a programming language, rather than am odd derivative of C++ tied to a billion files in various XML schemas.

Yeah and I remember hand crafting make files in order to build systems from all that carefully written C code.
 
I mean I really hate myself for clicking on the NuGet package manager that I installed in VS, browsing a huge number of open source solutions and downloading and installing libraries and libraries of useful code with almost a single click. Yeah .. progress sucks

Re:Programs!

[murdocj]

Using lots of libraries and components is great... when it all works. When your app won't build and you get an obscure error message from some package that you didn't even know you were using, it's not so much fun. I handcrafted make files as well. At least then, I knew what was going on, and what depended on what.

Re:Programs!

[Anonymous Coward]

I look back with fondness for the times when a program was a set of instructions and declarations written in a programming language, rather than am odd derivative of C++ tied to a billion files in various XML schemas.

Yeah...

cat configure.ac

420

Re:Programs!

[TechyImmigrant]

That's why I like my Apple 2e

Re:Programs!

[Horshu]

"In myyyy day, we wrote hand-tuned assembly, and we LIKED it!"

Re:Programs!

[Anonymous Coward]

I look forward to the time when I can tell my computer, in plain English, what I need it to do and it just does it without having to program a specific application to do a specific function.

You want that in real time too of course. I'd settle for a human that could do the same in real time, even if they charge $1,000 an hour. Business would pay 100x that "Write a program that cracks this 8192 bit encryption in real time, accessing all electronic databases in the world, popping up the target's photo id and other info, known associates, movement patterns, etc. You know, what every CSI and spy style show does on a regular basis.

Re:Programs!

[Anonymous Coward]

That's what NuGet is for, it's a package manager for libraries. Installing the latest version of X library for use in Y project won't cause compilation to stop on Z project etc, and you can easily see what libraries depend on what other libraries.

Re:Programs!

[Anonymous Coward]

Would you like a unicorn with your genie? Or just the regular fries?

Re:Programs!

[VortexCortex]

What better way to expand your attack surface.

Truly, in the Age of Information, the Hackers shall inherit the Earth.

Re:Programs!

[Wrath0fb0b]

The tension between KISS and DRY has always been there. Both are fundamental principles and yet at some level they are incompatible, since writing reusable code necessarily involves increasing its complexity. And the less you want to RY, the more complexity you have to build in.

The C++ STL is a shining example of this. Everyday developers shouldn't be writing their own lists and array and hashmaps. They definitely shouldn't write their own string utilities. And they shouldn't have to change those implementations whether they are working on regular strings or wide strings or with a HPC memory allocator. To deal with the genericity, STL is horrendously complex and Thor help me if I have to sit down with an error-page that's 5 pages long and 5 levels of template deep.

At the end of the day, you've just got to deal with that tension and decide what level of repetition (and the incumbent bugs and maintenance costs) you are willing to put up with to increase simplicity. If all you need is a simple array, don't use a library. If you are manipulating XML by using apos, on the other hand ...

Re:Programs!

[chrpai]

I'm a build and release engineer. I always know what depends on what.... especially when the developers don't.

Re:Programs!

[Anonymous Coward]

Maybe you should apply for a job in the janitorial field.

Re:Programs!

[chipschap]

Wish I still had my Epson 286. It was actually good for doing work.... I mean real work, not writing a letter with 100 fonts in it or making a spreadsheet so large and complex that no one could ever find all the errors in it.

Re:Programs!

[Xest]

I'd worry about a developer that doesn't even know what packages he is using.

It's not like NuGet provides a list of installed packages or anything. Oh wait.

You still know what's going on now. Scrap that. Competent developers still know what's going on now. The configs are still open and human readable, most people are aware of what dependencies they've added to their projects by simply not shutting down their brain whilst installing dependencies but I don't see how if you can't keep track of what you installed using say NuGet that you can magically keep track when manually adding dependencies to build files. Either you have some kind of issue of forgetfulness that makes you forget what dependencies you added to your project, or you don't. NuGet doesn't suddenly make it harder to remember what you added.

There's nothing magic, it's just that rather than type the config files yourself, there are tools to automate that now. They're still there, they're still accessible, you're still free to play with them as you wish.

Re:Programs!

[Anonymous Coward]

Mind you, some of that is down to C++ being a wordy and complex language for that sort of thing. Even the Java STL is lots cleaner to read (and read error messages from...), and Java is hardly a shining star in language design.

Re:Programs!

[ccanucs]

By' that wer't nowt. When I were't lad we toggled in th'instructions into't front panel. ;-)

Re:Programs!

[LordWabbit2]

Couldn't agree more. I miss those late nights paging through a thick ass stack of continuous paper shunted out by the mainframe's printer, looking for that single ascii character in the memory dump that said there was an error code after it.
I miss the magnetic tapes too, and how they would mysteriously disappear out of the tape vault, never to be seen my man or beast again.
I miss having to sleep at work at month end so you are instantly available to fix any issues instead of the fancy shmancy VPN tunnels they use now so you can sit on the couch at home.
I miss only having one colour on your monitor, so late at night in a dim office it looks like you have turned green and started decomposing.
And the IBM keyboards, they don't make them like that anymore. When you were typing the whole office knew about it.
I miss having to pour over the same code day after day trying to get it to fit into limited memory, tweak, tweak, tweak
Progress sure does suck



Actually I really do miss the keyboard, you could beat a cubicle neighbour to death with one and then carry on typing, and they were so heavy it would probably only take a couple swings.

Re:Programs!

[Anonymous Coward]

Riiiiigggghtttt. Because you could even make a spreadsheet that complicated back then. If you have a spreadsheet that large then you're doing it wrong. The data should be in a database and you should have an app doing whatever you need with that data.

Re:Programs!

[Anonymous Coward]

That's all well and good until the boss demands you waste days or weeks doing with the new shiny what used to take minutes or hours.

Re:Programs!

[Anonymous Coward]

... most people are aware of what dependencies they've added to their projects by simply not shutting down their brain whilst installing dependencies ...

How about being aware of the dependencies others have added to the project? There's the overhead of having some sort of guidelines to control that, or just rely on that people know what they are doing and problems are not hard to solve when they occur. The choice depends on the size of the company and the competency of developers.

Re:Programs!

[TechyImmigrant]

>"In myyyy day, we wrote hand-tuned assembly, and we LIKED it!"

I still do. Sometimes it's easier to write it in hex when there isn't an assembler because you just designed the microprocessor. I wouldn't say I like it, but it's certainly satisfying when it works.

Re:Programs!

[TechyImmigrant]

>The tension between KISS and DRY has always been there.
Yes, yes it has.

>The C++ STL is a shining example of this.
Yes, yes it is.

But C++ is in the middle. There's low level stuff (assembly or C running on little micros) and high level stuff (Python, Haskell etc). C++ is somewhere in the middle and the complexity is a symptom of that middly-ness. Generic code is easier in HLLs. KISS is easier with LLLs. I gravitate to the edges, language wise. DRY is nice, but thinking is hard.
 

Re:Programs!

[TechyImmigrant]

It's tempting.

Where is the RPM?

[Anonymous Coward]

I tried to do

yum localinstall visualstudio-2013.exe

but it wouldn't load on any of my Fedora or CentOS boxes. Tried the same with aptitude on my Debian boxes, same story.

Is someone gonna repackage this for our favorite distro? Really, these guys are worse than Canonical when it comes to supporting the community.

Re:Where is the RPM?

[Anonymous Coward]

*yawns*

Re:Where is the RPM?

[shutdown -p now]

I'm really sorry. We tried to build an RPM and a DEB, but for some reason no distro provides kernel32.dll in its repositories, and we need it as a dependency. I hope they fix that soon. ~

zero cost

[fermion]

If, as MS is so fond of saying, your time is worth nothing.

Or you could download full tool sets that are given away by every other developer. Apple, for instance, only charges $100 to develop on the iPad, giving the tools away. Eclipse is supported by many of the major vendors.

I realize that for a MS Shop the cost of Visual Studio is insignificant, but I can't even begin to comprehend why MS feels it needs to charge for the product. Everyone says that the Express version does everything anyone could want, but that is like saying the Home version of MS Windows does everything. We know it doesn't, and that it is that intentionally, so that people will pay for the instant upgrade.

Re:zero cost

[causality]

Apple, for instance, only charges $100 to develop on the iPad, giving the tools away.

Sure, and the dealership just GAVE ME the car I'm driving after charging me money for it! Wow that was nice of them.

Re:zero cost

[SJHillman]

"I can't even begin to comprehend why MS feels it needs to charge for the product"

I know, right? I don't know why the grocery store charges for hot dogs either. It's just a product.
More apps for the iPad means more app sales, which Apple takes a cut of, so that's a pretty bad example. Microsoft does give away the Express version, which is pretty decent for most non-commercial software.

Re: zero cost

[Anonymous Coward]

What can't you do in the express version that you need to do? The things they restrict you from are generally more team oriented and not geared towards independent developers. If you are part of a team like that your company probably has an msdn license and gets premium licenses with it for peanuts.

Not being an ass, actually curious what features are missing that makes you feel like its useful. I haven't used Eclyps very much but I personally find VS superior in every way (.NET development obviously).

Re:zero cost

[Anonymous Coward]

Apple, for instance, only charges $100 to develop on the iPad, giving the tools away.

Sure, and the dealership just GAVE ME the car I'm driving after charging me money for it! Wow that was nice of them.

Ignorance is bliss... Xcode is still free even if you don't want to pay $100 for a developer account.

Re:zero cost

[Anonymous Coward]

As is Visual Studio Express.

Re:zero cost

[causality]

Apple, for instance, only charges $100 to develop on the iPad, giving the tools away.

Sure, and the dealership just GAVE ME the car I'm driving after charging me money for it! Wow that was nice of them.

Ignorance is bliss... Xcode is still free even if you don't want to pay $100 for a developer account.

Actually you had to choose between two possible interpretations of what I said. 1) I am being facetious and am simply making a joke about the way he worded that, and 2) I was making a factual statement about developing software on (or for) the iPad. Because there was no additional context, you had to pick one. Naturally you chose the one that lets you make a smug comment while judging yourself smarter than me.

Is that bliss? Seems the product of a deep-seated (and horribly widespread) insecurity to me.

Re:zero cost

[Barlo_Mung_42]

I don't think you know what you're talking about. Developing for Windows / Win Phone is $19 and the express version does do everything most people will need.
Most people who pay for VS do so via MSDN which gets you a lot more than just VS.

Re: zero cost

[tangent]

The Express editions have a bunch of arbitrary limitations in them.

The two that bit me were:

1. You can't install plugins. I don't currently use any I can't live without, but several features in VS2013 -- e.g. NuGET, the thumbnail view replacing the scroll bar, better refactoring, visual indent level indication -- started out as plugins. Even if you take the view that eventually, all third-party plugin features eventually make it into the retail version, you're opting into being years behind the current state of the art.

2. The Express editions are artificially siloed into several versions, none of which has all of the features. If you need two features that are in different versions, at best you have to keep bouncing between the editions. If you need both features simultaneously, you're stuffed.

For me, the two features I needed simultaneously were the ability to create a mixed C# and F# program that ran on the desktop. To make a C# desktop app, you naturally need the desktop edition, but that edition doesn't include any F# support. For some demented reason, that's off in the Web edition, where it seems focused on ASP.NET development, not desktop development.

(And if you ask me why F#, well, this is Slashdot, isn't it? If I'd said Haskell instead, you'd just be nodding now. :) )

Re:zero cost

[Anonymous Coward]

Eclipse is crap compared to Visual Studio or Intellij IDEA. This isn't debatable.

Re: zero cost

[Anonymous Coward]

I just installed VS2012 the other day, so I can build example programs that were included with a DLL I was trying out. Won't work, it requires some .H file that turns out is only included in the paid version. That's great.. after installing 4GB worth of stuff, I'm stopped by lack of a little header file. Now I remember why I don't use their crap.

Re:zero cost

[betterprimate]

Huh? No, what you mean is that a Microsoft grocery store would charge for the hotdogs and buns just so I can buy the ketchup. From there, I can only sell their own brand hotdog in a square full of Microsoft employees.

Did I mention the hotdogs were five-years-old?

Re:zero cost

[causality]

Nah. It ain't "deep-seated". We just hate the same old bullshit by lousy "programmers". Apple likes developers and gives its tools away for free. Actually, the only company that doesn't respect its developers is Microsoft. But, oh wait, there are no real developers on the Microsoft platform. Apple and *nix has all the developers.

Ouch. Burn.

Prove me wrong. Show me any tool that is coded for or coded by Microsoft that is:

1) desirable 2) practical 3) intuitive

Waiting...........

If you're looking for a fan of Microsoft to defend the merits of their software, you're barkin' up the wrong tree, friend. I've been a Linux user since around 1996 or so and have no interest in Microsoft products.

I simply found it amusing the way that guy worded his sentence, saying that something was free ... after you pay. That's all. I have no idea how it can be so difficult to appreciate (or dislike) a simple jest.

Re:zero cost

[Anonymous Coward]

giving the tools away is a bit of an exaggeration but $100 isn't that bad for any programmer I know to pay except a college student. I'm not even sure why they charge that much for it.

Re:zero cost

[Anonymous Coward]

Did I mention the hotdogs were five-years-old?

.. and full of bugs?

Re:zero cost

[betterprimate]

I see now. It was late. I was drunk and trolling. A geek fight is better than a real one!

Still half-assed C++11 support

[mark-t]

(sigh)

Oh well... maybe next year they'll catch up. Oh wait, that's when C++14 is supposed to be standardized.

[double facepalm]

Re:Still half-assed C++11 support

[Anonymous Coward]

Still half-assed C++11 support

(sigh)

Oh well... maybe next year they'll catch up. Oh wait, that's when C++14 is supposed to be standardized.

[double facepalm]

I would give (almost) anything to work on that compiler exclusively : Think about us, poor developers, stuck with IBM XLC or (the worst of the worst) Solaris SunStudio and its poor support of (sigh) C++98 (98 as in 1998...).

C++ standard is evolving fast, so we can't expect all the compilers to offer an implementation in less than 6 months... But some compilers just remain in the last century... And Visual Studio is not one of them.

Re:Still half-assed C++11 support

[Anonymous Coward]

Why do you need that? Honest question.

Re:Still half-assed C++11 support

[Anonymous Coward]

Why do you need anything more than an assembler?

C++11 isn't needed, but many people (myself included) consider many of the language changes to be very beneficial to creating clean programs with better compile-time checking.

(In before "real programmers code by twiddling individual bits on disk.")

Re:Still half-assed C++11 support

[mark-t]

First thing that comes to mind? compile-time hashes used as case labels.

constexpr unsigned crc32_table(unsigned c,unsigned k=8)
{
return (k==0)?c:crc32_table((((c&1)?0xedb88320u:0)^(c>>1)),k-1);
}

constexpr unsigned crc32(const char *str, std::size_t len)
{
return (len==0)?0xffffffffu:((crc32(str,len-1)>>8) ^ crc32_table((crc32(str,len-1) ^ str[len-1]) & 0xFF));
}

constexpr unsigned operator "" _hash(const char *str, std::size_t len)
{
return crc32(str,len)^0xffffffffu;
}
...

which could then be used in code like this:
switch(tag)
{
case "show"_hash:
...//do stuff
break;
case "fill"_hash:
...// do stuff
break;
}

Re:Still half-assed C++11 support

[sideslash]

Why don't you just write a source code generation script to write that section of logic?

Re:Still half-assed C++11 support

[mark-t]

Because it's a pain in the ass, that's why.

Also, I don't like wasting my time writing tools to "fix" somebody else's partially complete implementation of something... in this case, C++11.

Yes, I'm lazy. I'm a computer programmer.

Re:Still half-assed C++11 support

[TheRealMindChild]

Because modern C++ developers want C++ to work like javascript

Re:Still half-assed C++11 support

[slack_justyb]

C++ standard is evolving fast

Does no one else find it funny that saying that about five years ago would have been met with "WTF?!!"

However, I do have to agree, VS still has half-baked C++ support period. It's neat that they have their own .NET stuff for C++, but I think they tend to think about that .NET stuff first and ISO C++ second. That's a shame really because I know quite a few (and maybe it's just the area I'm in) places wanting to hire those with C++11 skills.

so we can't expect all the compilers to offer an implementation in less than 6 months

Well the thing about it is that they've had longer than six months to prep for it. Especially for C++11. I get your point, but the other guys tend to build as the standard gets formalized, not wait until it is approved. Heck even the GNU guys are already baking C++1y support. That's what makes me think that my first point is more true than time frame reasons.

C++11 support isn't so much a need. The neat features brought with the new standard aren't a "MAKE OR BREAK" kind of thing. However, at the risk of making an oxymoron, C++11 new feature sets make C++ a great deal more readable and enjoyable to code in (I know I was like, shwhaaaaatt?!). Which kind of makes the .NET stuff a little less appealing (unless you're targeting the .NET run-time, which in itself is a whole another can of worms) since I've know quite a few folk to code C++.NET because it is easier than ISO C++. The run-time thing, to them, is just an added benefit. I think once a person starts to use C++11, it'll click what makes it great.

Think about us, poor developers, stuck with...

You are not forgotten, you have my deepest sympathies if you are still stuck with them.

Re:Still half-assed C++11 support

[Anonymous Coward]

...and have they ever bothered to upgrade the C compiler beyond ANSI 89??

Re:Still half-assed C++11 support

[mark-t]

Speaking for myself, there's some truth in that.

Only faster runtime.

Re:Still half-assed C++11 support

[Anonymous Coward]

http://cpprocks.com/c11-compiler-support-shootout-visual-studio-gcc-clang-intel/

Re:Still half-assed C++11 support

[AstrumPreliator]

You need to be careful with constexpr as it is not guaranteed to be evaluated at compile time. I don't have a link handy, but if I remember correctly the only time a constexpr function is guaranteed to be evaluated at compile time is if all of it's parameters are constant expressions and it is used in a constant expression. Compilers are of course free to evaluate constexpr functions in other situations, although to my knowledge neither clang nor gcc does this yet.

In your example "show"_hash and "fill"_hash should be evaluated at compile time. However if you had someFunc(int hash, int runTimeParameter); and you passed "show"_hash to someFunc, there is no guaranteed it will be evaluated at compile time.

Re:Still half-assed C++11 support

[Anonymous Coward]

You are full if sh*t. It is a compilation error to define a constexpr object that can't be evaluated at compile-time. constexpr functions can be evaluated at run-time, but NOT when used where a compile-time expression is required. The reason is to avoid having to write two identical functions.

Re:Still half-assed C++11 support

[AstrumPreliator]

I said nothing of the sort. I said that just because a function is constexpr doesn't mean it will be evaluated at compile time. Which apparently you agree with. So how am I full of shit again?

Re:Still half-assed C++11 support

[Anonymous Coward]

C++ has no concept of "at compile time", so therefore there is no way for the standard to state what must be evaluated at compile time. This fundamentally runs up against the "as-if" rule - if you can't tell the difference from inside the program between what the standard says should happen and what really happens, then that's all that is required. Here computation time is not considered a way of "telling". That said, it would be pretty asinine for a compiler to see a constexpr expression and then choose to go against the overwhelming expectation that it should be evaluated at compile time - if that's happening, I think you should offer evidence of that. The only way you could tell would be to look at the assembly output. It's true that you won't necessarily get compile-time evaluation from a constexpr function if you don't give it constexpr parameters, but that's obvious - you can't evaluate at compile-time what isn't specified at compile time.

Re:Still half-assed C++11 support

[mark-t]

If used on a context which requires a constant, a constexpr will *always* be evaluated at runtime.

If used in any other context, it's possible that it will instead output code to compute the value instead of evaluating it.

Re:Still half-assed C++11 support

[mark-t]

Looking at gcc 4.7.1's output with constexpr, I've found that putting it in a context which *requires* a constant, such as a case label in a switch statement, the compiler will faithfully output an evaluated value, as expected.

If it is used in more general contexts, however, it doesn't always work. My experience is that tail recursive constexpr's, in particular, did not always output the evaluated constant directly, but instead would often output the code to compute it per the algorithm described in the constexpr expression.

Re:Still half-assed C++11 support

[AstrumPreliator]

My last post was a bit sparse on details but I'll try and improve upon it here. The main point I was trying to make was that just because you define a function as constexpr doesn't mean it will run at compile time unless certain conditions are met. This obviously isn't a bad thing as you can re-use the function at runtime.
 
If you have a constexpr function and all of its arguments are constant expressions (such as literals) and it is used in a constant expression (switch case label, array size, non-type template parameter, etc...) or to initialize a constexpr variable then it is guaranteed to run at compile time according to the standard. Otherwise it is up to the compiler. I've noticed a lot of people think that merely passing constant expressions to a constexpr function is enough to guarantee compile time evaluation, but it's not. For instance:

constexpr int factorial(int n) { return n <= 1 ? 1 : (n*factorial(n-1)); }

int main() {
      char c[factorial(5)]; // Compile time
      switch(n) {
            case factorial(3): // Compile time
            break; }
      constexpr int i = factorial(4); // Compile time
      cout << factorial(8) << endl; // *Compile time or runtime*
      int i;
      cin >> i;
      int j = factorial(i); // Obviously runtime
}

Here is a page that talks about it (especially in the comments). I was also mistaken about clang and gcc. At higher optimization levels the factorial of 8 will be computed at compile time rather than runtime.

Re:Still half-assed C++11 support

[Anonymous Coward]

Hey at least they are releasing VS2013 in the 2013 calendar year, even if it is just in time for 2014...

Re:Still half-assed C++11 support

[shutdown -p now]

Actually, yes, that's one of the notable changes in VC++ 2013 - it now supports C99 _Bool, compound literals, designated initializers, and most of C99 new headers. Apparently, that's sufficient to compile ffmpeg, which was the point of the exercise. Still no VLAs, though.

Did they put Brief emulaton back in yet?

[Anonymous Coward]

If not, fuck 'em.

Learning this dross

[Anonymous Coward]

Just stepped into an organisation running TFS '08 & VS '10.

Coming from a background in open source, using Eclipse, SVN, Bugzilla & TRAC this MS stuff seems like absolute dross to me but I'm not in the position to change it yet.

Anyone have any advice regarding getting up to speed on this stuff. In particular the team I'm working with have NO concept of bug tracking which seems like madness. Is this side of TFS really so terrible?

Re:Learning this dross

[cbhacking]

That's mostly a problem of team, not tools. Lots of open source developers are shit at bug tracking too. TFS isn't my first choice of tool, but it works.

Also, to the extent that the tools are the problem, that's largely because you're using tools that are 3-5 years old. Updating to newer versions won't make them any more familiar to you (as if ability to adapt to tools isn't a vital skill for a professional coder...) but it will add a lot of functionality that you may be looking for.

It won't fix team stupidity, though. There's really no solution for that one.

I'd offer suggestions, but you completely neglected to explain what (in terms of the tools) you were having problems with, so rather than give a full course of
"MS Dev Tools 101: Tutorial for the Anonymous Coward Who Thinks it is "Dross"
I think I'll just suggest looking up online anything that you're trying to do and can't figure out; you're hardly the first person to use this software, even coming from a background like yours.

Re:Learning this dross

[murdocj]

Just stepped into an organisation running TFS '08 & VS '10.

Coming from a background in open source, using Eclipse, SVN, Bugzilla & TRAC this MS stuff seems like absolute dross to me but I'm not in the position to change it yet.

Anyone have any advice regarding getting up to speed on this stuff. In particular the team I'm working with have NO concept of bug tracking which seems like madness. Is this side of TFS really so terrible?

If you are referring to Bugzilla as your "bug tracker", god help you. What a nightmare of a user interface. It would be easier to track bugs by chiseling them into granite than to use Bugzilla.

Give Visual Studio a chance. I haven't used it for a few years but it's clean and works well. Don't pine for your old environment till you've tried the new.

Re:Learning this dross

[man_of_mr_e]

TFS does bug tracking. If they're not using it, that's their fault. It has change set integration (tying work items to changesets), and agile templates, although they're pretty out of date as agile has come a long way_

VS 2010 and TFS 2008 are dated, but they give you the tools you need. Bugzilla and Trac may have more features, but that comes at the cost of ridiculously complex interfaces which mere mortals can't figure out how to use (non-developers).

TFS has a web interface to allow end users to enter bugs, and there's a stand-alone client if they want to use that.

There are also tons of tools to integrate with more featureful tools like Jira and Trac, so you can map workitems and changesets, etc..

This isn't meant as a sales pitch, just that it *DOES* do what you need it to.. and there are ways to introduce better tools and still integrate.

TFS...

[dkegel]

Is it just me, or did other people read that as The F*cking Software?

Re:TFS...

[HalfFlat]

As someone who is obliged to use TFS, I would say that your reading is correct.

Re: TFS...

[RightSaidFred99]

Please. TFS 2010 Is very good. TFS 2012 is excellent.

Re: TFS...

[HalfFlat]

TFS2010 very good? Oh, my.

I've seen: check-ins transpose lines on check out; complete failures to update to actual latest versions of code; and random check-outs of code with no local changes.

Other fun aspects: can't unshelve to anything but the changeset that the shelf came from; industry worst? merge and diff tool; no non-connected way of getting changeset info for automatic version information; despite being a centralized model, local workspaces can't be moved (say, in the advent of hardware failure on a development machine). The only way I can be assured that the check-in state actually correlates with what I have locally is to manually run a compare over the project directory and check.

It's also terribly, astonishly, slow over a VPN. Start typing to make a change, only to have all but the first character thrown away as TFS laboriously attempts to check out the file first.

It is so crushingly painful to use now, that I honestly can't imagine they've fixed all their shit in two years to make TFS2012.

Re: TFS...

[bmajik]

check-ins transpose lines on check out; complete failures to update to actual latest versions of code; and random check-outs of code with no local changes.

If you can reproduce any of those things, please email me.

can't unshelve to anything but the changeset that the shelf came from

This isn't really legible, but I think you meant to say "branch" or "workspace". But in any case, it isn't true. You can use tfpt.exe (tf power tools) and force all kinds of "unsafe" things, like unshelving an add/edit shelveset across branch definitions.

local workspaces can't be moved (say, in the advent of hardware failure on a development machine).

This is also false. Moving an existing enlistment on another computer is something we do frequently

tfpt workspace /updatecomputername workspace

(ref: http://blogs.microsoft.co.il/blogs/shair/archive/2008/09/03/using-tfpt-command-line-tool.aspx)

I work on Visual Studio. All of Visual Studio is under TFS source control. We use it every day.

To dogfood scalability, we use a single instance for all of VS. It supports every single team within VS; thousands of simultaneous engineers with hundreds of branches across multiple sites (I'm in Fargo, and TFS is in an entirely different timezone)

Start typing to make a change, only to have all but the first character thrown away as TFS laboriously attempts to check out the file first.

Since I grew up using commandline SCC, I normally use tf.exe to open files for edit. You're referring to the feature in the code editor when you dirty a controlled file, we try to go create a pending edit. I don't know if that is still quasi-synchronous or not. In any case, there are certainly work- arounds -- like using the explicit checkout gesture in the solution explorer -- prior to typing in an editor window.

Re: TFS...

[HalfFlat]

I'm willing to believe things are going great in your environment — we have been plagued by problems. (Some of the gripes in my post may have been specific to TFS2008, though the mind-boggling line transposition was just two months ago.) We will almost certainly be upgrading TFS when we move to VS2013, though given some of the egregious compiler bugs present in the new release, we will probably wait until the first SP. In the meantime, we're migrating projects over to git, and ultimately we will probably not use TFS at all for source control.

It is good to know that some issues can be addressed with tfpt; it would have been very helpful to have that functionality accessible from within Visual Studio (hint, hint.)

I haven't any repeatable set-ups of brokenness, but things do seem to be way less reliable when we have files with mixed line-ending characters, or when TFS is operating in a non-constantly connected network environment (owing to the VPN link.)

Re: TFS...

[man_of_mr_e]

Ummm... How can you on one hand talk about your giddiness of moving to Git, and then complain about how things aren't accessible in VS? You have to drop to the git command line for a lot of things...

Tried it - was very underwhelmed

[Anonymous Coward]

I tried the new VS 2013 RTM build today on a Win 7 x64 SP1 system. I'm really unimpressed - yes, it's quicker than 2010/2012, but I only used it for 1 hour (it was a lunchbreak), and the IDE crashed on me twice in 1 hour, on a virgin system. I'd previously tried the preview and RC build of 2013 with no such problems. Apart from being slightly snappier, I really can't see what it's given us vs 2012. Most of the code I maintain/write is not pushing the boundaries of C++11 etc, so am much more interested in other tools - better debuggers, editors, more speed, better static analysis etc.

Yes you can still target XP with it, albeit using old compiler toolsets.
Other stuff to be aware of: MS are looking to drop MBCS support (at least in MFC). If you need MBCS support with MFC, you need to download an extra set of files.

But for me the killer is it's apparent instability - two IDE crashes in 1 hour is terrible. I didn't have those reliability problems before...

Just downgraded something to .NET 2.0

[jfdavis668]

At work, we just had to downgrade one of our products because the customer couldn't handle .NET 4.0. Will the world please catch up with Microsoft, please?

Re:Just downgraded something to .NET 2.0

[0123456]

I thought .NET was dead and the Microsoft future was HTML5 now?

Re:Just downgraded something to .NET 2.0

[xyourfacekillerx]

I also find myself having to use 2.0 a lot too.

Re:Just downgraded something to .NET 2.0

[Anonymous Coward]

what are the fuck are u smoking? Silverlight as a product is dead. Go troll somewhere else..

Re:Just downgraded something to .NET 2.0

[Anonymous Coward]

ahaha, losers. Tell your client there are reasons to move beyond the 286 architecture.

Re:Just downgraded something to .NET 2.0

[xyourfacekillerx]

Sometimes our clients are mom and pop shops. Right now I'm working with the county whose budget for software alone is $800 this fiscal year, $4,000 for hardware. That's the budget for the county department I'm with. Fortunately the consulting budget is any dollars, so I can offset the cost of upgrading their software with that, but still.... Not everyone you work with understands the benefit of running an up to date setup.

Re:Just downgraded something to .NET 2.0

[cbhacking]

What have *you* been huffing? .NET, in one form or another, is *the* main development framework Microsoft has been pushing the last few years, honestly.

Windows desktop pre-Win8: Native code or .NET.
Win8 / Windows RT apps: .NET (via the subset usable in WinRT), native code (same caveat), or HTML5/JS.
Windows Phone 7: .NET (via Silverlight) or .NET (via XNA).
Windows Phone 8: .NET (via WinRT subset for phone) or native code (WinRT).
Xbox 360 indie games: .NET (via XNA).

This goes back even further, actually, but those are just Microsoft's major ISV-facing platforms of this decade.

Re:Just downgraded something to .NET 2.0

[Anonymous Coward]

Apparently you missed the renewed interest in C++. .NET is still very popular, but the .NET team never sold the Windows development team on .NET, who went off in their own direction with Metro and additions to WinAPI. So, if we're talking the past two years, then .NET is definitely not *the* main development framework, it's C++ (i.e. native code). How have you missed this? There have been a ton of articles over the past couple of years analyzing Microsoft's schizophrenia.

Perhaps you were just working really hard on .NET code. There are Perl guys hammering away, still under the belief that Perl is still popular, too.

Re:Just downgraded something to .NET 2.0

[betterprimate]

Will the world please catch up with Microsoft, please?

LOL. We're waiting for Microsoft to catch up. It ain't 2008, bitch.

Re:Just downgraded something to .NET 2.0

[Xest]

The C++ frameworks have always had the cutting edge features first. MFC received ribbon support before the managed frameworks for example. This is because the managed frameworks usually just wrap around that anyway - i.e. WinForms wasn't much more than a wrapper around Win32 API.

But that doesn't mean they're the preferred, main, or recommended development framework. The Windows development team use C++ because they're doing OS development and it's the best tool for the job, coupled with the fact it's all built with that legacy-wise anyway.

Far and away the majority of Windows development is still done with .NET and that ain't going to change any time soon. It really doesn't matter how renewed the interest in C++, it's still not making any headway to become the primary development option on Windows - getting features first is irrelevant, it's always been that way, even before the false rumours of Microsoft killing .NET came about.

If anything, the failure to focus primarily on .NET by the OS team in Win8 is the reason why WinRT has been such a massive flop in the first place.

Re:Just downgraded something to .NET 2.0

[david_thornley]

Microsoft has been pushing the idea that native code development is back (not that I noticed it was gone, I just kept writing C++). This may not be the best idea (I find your suggestion about pushing .NET instead very plausible), but MS is at least publicly changing direction. Not as bad as Apple used to do, but it isn't pretty.

Re:Just downgraded something to .NET 2.0

[cbhacking]

I write equal amounts native and managed code. I'll grant that managed ha seen a resurgence, but the only way you could call it the "main" framework is to note that the last few tool versions have added more updates to native-oriented tools than managed-oriented ones... which sounds good until you realie that the native tools were left to languish for so long that these updates have been almost entirely a matter of catching up, while .NET has still gotten a bunch of cool new stuff like async.

Re:Just downgraded something to .NET 2.0

[shutdown -p now]

This is because the managed frameworks usually just wrap around that anyway - i.e. WinForms wasn't much more than a wrapper around Win32 API.

This actually hasn't been true for a while. You're right on WinForms, but that has been de facto deprecated from .NET 3 onwards, with WPF taking its place - and WPF doesn't wrap any OS API, it does everything down to rendering on its own (and that is directly on top of Direct3D). Similarly, WPF Ribbon does not wrap the OS ribbon, it reimplements it.

Also, MFC received ribbon support first simply because Microsoft has bought it from a third party company which implemented it already. I believe it is also a from-scratch implementation, not a wrapper over Win32.

VSS

[Anonymous Coward]

Oh my god if I have to use VSS one more time, I just might receive CANCER.

As a new user of Visual Studio

[Kwyj1b0]

I'd like to ask - what am I missing?

Until recently, I hadn't programmed in anything apart from Matlab in Linux (which has a crappy "IDE") in over ten years (the last version of VS I ever used in any way was VS6.0). Anyway, I started to work on Python and C++, and have so far found a lot of positives with the IDE (Ultimate VS2012 - free from my organization).

VsVim and PTVS let me use a vim like editing features, and Python Tools for VS has also performed well (interactive debugging, autcomplete and command help). On the C++ side, the debugger (for simple code at least) is straightforward. The Git integration could be a better, but I can quite easily drop into the command line and sync with GitHub.

Since I am still learning the tools (and I have used Linux a lot over the last five years, so I am OS agnostic) and the language, I'd like to know what I am likely to miss out by using VS over say Eclipse (or other tools). I tried Eclipse for about half a day, but I had a bit of problem getting the debugger to work for C++. Again, since I am just starting out, I like the convenience of an IDE, rather than using vim+gcc at the command line - I'm not even sure how I'd do a command line debug.

Re:As a new user of Visual Studio

[cbhacking]

Missing relative to other tools? Not terribly much, honestly; I wouldn't use VS for Java (by preference, I'd use NetBeans) or for POSIX native code, but both are possible. Some VS extensions are very handy; there's a tool for finding, installing and updating them called NuGet (should be built into current versions of VS, I think); you may want to check them out although it sounds like you've already found some plugins that you like. The git integration will probably improve over time; there has already been an update or two. Eclipse has slightly more refactoring power than is built into VS, but there are plugins for that and the Eclipse UI drives me nuts when I try to use it. The only major thing that comes to mind is that VS isn't going to run on anything except Windows (unless Wine support for it is a lot better than I remember) so, although there are Linux-compatible IDEs that can read its project files, it might not be the ideal tool for mixed environments.

Re:As a new user of Visual Studio

[Anonymous Coward]

Eclipse is still to this stay sluggish and the intellisense is slow enough that you tend to type faster than it, whilst in contrast Visual Studio's intellisense is snappy and lets you complete as you type churning out lines of code in just a handful of key presses.

It also has, as you've noticed, superior debugger support.

But really, Visual Studio shines the most if you're developing for MS technology, it's really designed for that and many features are explicitly for that, if you're doing C++ development then it's great, but Python? I'd imagine not so much, also it's pointless for Java.

Where VS stands out is if you have, say, the choice of developing a C# .NET application because your target is going to be Windows only and something else, say, Java and Eclipse application. In this case if there's no need for the portability of Java then it's just pointless not to go the .NET route, you'll get your application built way faster and have the tools and features you need to make sure it's much better quality.

So ultimately Visual Studios strengths are also it's weakness - it's the best IDE on the planet if you're building for the use cases it's optimised for, but if you have a different set of priorities, then you wont get any benefit from it over something else.

This is backed up by the fact I just don't really see any major complaints from C# .NET developers about Visual Studio who have also done extensive work with say, PHP and Zend's Tools or Java and Eclipse - they'll almost unanimously agree it's the best in business. The complaints I see about Visual Studio are those who want to use things with Visual Studio that were never designed to work with Visual Studio from the outset - why doesn't my Ruby on Rails with GIT project work well with VS! - perhaps because you're not using an IDE designed for that?

Compare and contrast to Eclipse, and it's a jack of all trades, it's designed to be so generic that it'll do anything, but that's also it's downfall to, it also results in a plugin architecture that just fails if you dare to mix too many with contrasting features such that you just end up having multiple installs of it instead, and it's extremely sluggish still.

Re:As a new user of Visual Studio

[shutdown -p now]

Out of curiosity, did you have a chance to look at the Python/C++ mixed mode debugging, and if so, how did you find it?

(I'm the PTVS developer who implemented it, and I'm always looking for feedback from users who use the feature on real-world applications, especially in terms of use cases, scenarios etc - i.e. what can be added or rearranged to improve the typical or not-so-typical workflow or make it more convenient. Bug reports are also always welcome, of course!)

Re:As a new user of Visual Studio

[shutdown -p now]

if you're doing C++ development then it's great, but Python? I'd imagine not so much

You'd be surprised. I dare say that we're neck to neck with PyCharm, and doing some things better than them - e.g. type inference for code completion (try some of the code snippets in this video in your favorite Python IDE, and see how it fares...). And no other Python IDE has anything like this, to the best of my knowledge.

I can fully understand where you're coming from - it's true that, historically, Microsoft developer tools have focused on supporting pretty much only Microsoft languages and frameworks. But times a-changing, and today the focus is being shifted towards making Windows a more attractive development platform for people with background in other technologies, including those that are traditionally closely associated with F/OSS.

Re:As a new user of Visual Studio

[Kwyj1b0]

Hi... I haven't used mixed mode debugging - I'm still learning the two languages independently.

Re:As a new user of Visual Studio

[shutdown -p now]

Thanks for getting back on this.

Generally speaking, any feedback that you can give to us on any area of PTVS is helpful. I don't just mean bugs, but generally what doesn't work well or fit right or breaks the workflow or is just generally counter-intuitive. We do prioritize user feedback when planning for next releases (e.g. for 2.0, we implemented 7 of our top 10 most voted features), so it's not just wasting your time pushing bits to /dev/null - and, of course, for every person who does report an inconvenience, there are ten who won't bother, but who will benefit once it's reported and fixed.

OTOH, if you really are just happy about everything you see, that sort of feedback helps, too - it promptly gets forwarded to higher management to help justify the existence of the project :)

"furthered"??? No such word...

[Anonymous Coward]

"Graphics debugging has been furthered"

I don't believe that 'further' is a verb.

Re:"furthered"??? No such word...

[PT_1]

"Graphics debugging has been furthered"

I don't believe that 'further' is a verb.

Not sure if I'm missing the joke, but further is a verb and furthered is its past participle.

Re:"furthered"??? No such word...

[EvanED]

further

verb
[with object]
help the progress or development of (something); promote:
he had depended on using them to further his own career

Origin:

Old English furthor (adverb), furthra (adjective), fyrthrian (verb)

oxforddictionaries.com

Unfortunately I don't have access to the OED any more so don't know places where I can get dated citations of use, but it's a good thing you don't have to believe in facts for them to be true.

Re:"furthered"??? No such word...

[Anonymous Coward]

Just goes to show that reality doesn't always conform to what you believe:

further
frTHr/

      [...]

verb
verb: further;3rd person present: furthers;past tense: furthered;past participle: furthered;gerund or present participle: furthering

        1.
        help the progress or development of (something); promote.
        "he had depended on using them to further his own career"
        synonyms: promote, advance, forward, develop, facilitate, aid, assist, help, help along, lend a hand to, abet; More

Re:"furthered"??? No such word...

[Anonymous Coward]

You believe wrong!

http://www.thefreedictionary.com/furthered

Re:"furthered"??? No such word...

[synaptik]

In English, almost any word can be verbed...

Re:"furthered"??? No such word...

[betterprimate]

You're not missing the joke. Everyone else is just missing English class.

Seems to be fine, but

[xyourfacekillerx]

1) Where the heck did they hide the option to set which version of .NET I want to target for my build? Argh.

2) what's a good free code editor or IDE for C# or F# that still does projects/solutions, source control (maybe), and let's me control the resource and assembly files on my own as well as regulate the build process too (i.e. command-line or integrated build tool)? cos I'm not liking 2013 or 2012 VS and MSbuild that MS offers right now is for 4.5.1x only. What if I want to compile a single source file and I don't want the stupid dev command line or MSbuild to do it for me?

Re:Seems to be fine, but

[BradleyUffner]

1) Where the heck did they hide the option to set which version of .NET I want to target for my build? Argh.

It's right there in the very first page of the project properties for me, where it's been ever since that was a configurable option.

Re:Seems to be fine, but

[shutdown -p now]

what's a good free code editor or IDE for C# or F# that still does projects/solutions

SharpDevelop?

MSbuild that MS offers right now is for 4.5.1x only

It requires 4.5.1 to run (but then why wouldn't you want to upgrade?), but it can build applications for any version of .NET from 2.0 up.

What if I want to compile a single source file and I don't want the stupid dev command line or MSbuild to do it for me?

You don't need any special command line, or MSBuild, to compile a single file. Just add csc.exe to your path and run it directly: csc foo.cs.

The GUI blows

[jez9999]

Maybe I'm more irritated by this than most, but I liked the VS2010 GUI; colorful icons, a relatively smart professional image. With VS2013 they appear to have tried to "geek it up" or something by making all the tool menus have CAPITAL headings which looks fucking retarded, and making most of the items monochrome (what is that, retro?) Apparently they're trying to 'draw my attention' to the code without distracting me with icons that are nice looking and, ya know, give you a clue what the fuck they do. It just looks like a trainwreck. If there's a VS2010 skin, that's the first thing to install.

Re:The GUI blows

[Anonymous Coward]

There are both of these, at least for VS2012:

http://stackoverflow.com/questions/12087949/is-it-possible-to-change-icons-in-visual-studio-2012
Turning the menus back to non-all upper caps is a registry hack.
The tool that gives colorful icons back to VS2012 needs to steal the icons from an installation of VS2010. The icons do not come with that app.

I can see having the monocromatic icons while the mouse is hovering over your code and you're presumably "not looking at them". But, could it have been simple enough to colorize them if the mouse moved over the toolbars?

Re:The GUI blows

[Anonymous Coward]

I encountered Windows Server 2012 the other day using RDP from AWS. Gawd that looked awful. I hit those utterly flag 2D-ish UI elements and thought "Windows 3.11."

What's wrong with these people? Just what is their major malfunction? It must be some manifestation of the Spielberg effect.

Re:The GUI blows

[jd2112]

Maybe I'm more irritated by this than most, but I liked the VS2010 GUI; colorful icons, a relatively smart professional image. With VS2013 they appear to have tried to "geek it up" or something by making all the tool menus have CAPITAL headings which looks fucking retarded, and making most of the items monochrome (what is that, retro?) Apparently they're trying to 'draw my attention' to the code without distracting me with icons that are nice looking and, ya know, give you a clue what the fuck they do. It just looks like a trainwreck. If there's a VS2010 skin, that's the first thing to install.

There is a registry hack to get rid of the dreaded ALL CAPS.
2012 Full: HKCU:\Software\Microsoft\VSWinExpress\11.0\General\\SuppressUppercaseConversion DWORD 1
2012 Express: HKCU:\Software\Microsoft\VWDExpress\11.0\Genera\\SuppressUppercaseConversion DWORD 1
For 2013 replace 11.0 with 12.0.

Re:The GUI blows

[norite]

Flat, minimalist 'design' (And I use that word loosely) is all the rage these days. Take a look at google+ ...it looks fucking hideous. There are plenty of other websites following this shitty trend, miles of brilliant whitespace everywhere, no borders around anything to give it some context, It gives me a headache and ensures I won't visit again. Office 2013 is just as awful; NOT ONLY DO THE RIBBON MENUS SHOUT AT YOU, it's a bland wasteland of empty ideas, with only three colour schemes - brilliant white, off white and slightly more off white.

It all looks bad now, but in 5 years time people will be shaking their heads thinking 'Just what the fuck were we doing?'

Re:The GUI blows

[betterprimate]

Just wait until tomorrow. What will happen with /. scheduled downtime?.. will they replace the current site with their Wordpress beta theme?

Re:The GUI blows

[betterprimate]

For those of you who have switched because of the dreaded Windows registry, say Amen!

I said Amen!!!!

Can I have an Amen??

C support stopped at ANSI 91

[Anonymous Coward]

As a C programmer, not useful; MS support ANSI C 1991 and that's it.

I'm moving to Linux on my next laptop, because of the NSA issues with MS, which will also let me move forward to the modern version of C.

Re:C support stopped at ANSI 91

[Anonymous Coward]

For those who want to stay on Windows and want good C support, another option would be to use Qt Creator, which ships with a modern C compiler (MinGW gcc). It's pretty good IDE for non-Qt projects too.

Re:C support stopped at ANSI 91

[shutdown -p now]

This version is a somewhat better at C99: it has compound literals, designated initializers, _Bool, and most C99 headers. Still not full support for the standard, but it should be much easier to compile a lot of code from the Linux land now.

Visual Studio? Released?

[Jeremiah Cornelius]

On bond, or recognizance?

Re:Visual Studio? Released?

[LifesABeach]

And in a unrelated news story today, 2 murders were released from prison in Florida. This O/S used to help the killers escape was, you guessed it; not Linux.

Re:Visual Studio? Released?

[Anonymous Coward]

Rumor has it that the Linux OS has been plotting to help Hans Reiser escape.

Re:Visual Studio? Released?

[twocows]

Don't worry, it's not free.

Too much hate

[Anonymous Coward]

I'm not a Microsoft fanboi, but for the sake of a bit of balance around here...

At least they've shown that their shift back to AOT compile development tools and C++ (even if adulterated with their extensions) is proving real. The more coders using generally portable languages the better. The alternative is they invest exclusively in their .NET stack.

from demos

[ILongForDarkness]

It seems that the editor changes are mainly a roll in of the powertools (I don't do client side web dev so javascript and ASP side of things don't matter to me). Makes me wonder: what will the next power tools be as it seems to be the only way I'll be getting new editor features?

I can't remember if VS2012 added it or not as my work developes mainly in 2010 but a big one I'd like to see is coding time checks on stored procedures for database projects. It annoys me that I have to migrate my database and run unit tests against my model to find out that one of my stored procs is trying to use a parameter in another proc called @Cust when it really is @Customer. This is something that is obvious if they just did a basic parse on the project contents. Probably "just" need to roll in the TSQL parser/lexer side of things from SQL.

Re:from demos

[Anonymous Coward]

Or, stop being a sheep and following the ORM herd?

I never encounter the problem you are encountering because *gasp* i write my own code instead of "saving time" and letting some tool create bland, non-optimized versions for me.

Dinosaur? Maybe. But I am more productive than most of my peers, and we never end up with sprint tasks like the one recently entitled "Change EF 4.0 out for EF 6.0" that takes a developer 2 weeks to perform.

Re:from demos

[ILongForDarkness]

What are you talking about? I'm talking hand written sql files not ORM. The test project in the solution targets a specific version of SQL Server so VS should know the semantics of the TSQL we are writing but it seems (at least VS2010) to only consider stuff in the current file and names from other files. Say you have a proc called dumb and another called dumber which takes @cust int as an argument.

If dumb tries to call dumper ... declaration/initialization code
exec dumber @customer = @bob

You can compile the database project just fine. Until you run a unit test and then you get the runtime errors from SQL saying that the proc expects parameter @cust or whatever.

zero cost?

[Anonymous Coward]

I tried to download it, then I was prompted to sign in.

Let me know when Express editions are available without the non-zero cost of the inconvenience of signing up for a Microsoft account. Previous versions of Visual Studio Express did not require a Microsoft account.

Re:zero cost?

[Anonymous Coward]

Oh you poor baby! It takes five minutes and none of the info you give them is verified.

Re:zero cost?

[andrew3]

It takes five minutes and none of the info you give them is verified.

Oh, so that makes it okay, does it? Almost all Microsoft services contain a termination clause which allows them to cancel a service for a user, or delete their account at any time.

That's right, the software on your computer is now being tied into Microsoft's services, so that the rights you once had disappear.

Re:zero cost?

[Xest]

What? one minute you're complaining about having to have an account to download, the next you're complaining that they might delete your account that you don't want in the first place.

Then you're jumping to some nonsense conclusion that by terminating your account they'll somehow hack into your computer and delete your software too?

This isn't Google apps. It's not a web based tool.

Re:zero cost?

[man_of_mr_e]

The prompt also had a link to skip logging in. You should pay more attention.

Link to.Net 4.5.1 ?

[Anonymous Coward]

How come the editors do not check the links before posting the submission ?

The link to the msdn blog doesn't give any pointer to .Net 4.5.1

And search engines only result in .Net 4.5.1 RC release.

Where is the real link to the final release of .Net 4.5.1 ???

Re:Link to.Net 4.5.1 ?

[RaceProUK]

Where is the real link to the final release of .Net 4.5.1 ???

Here. At a labour rate of $100/h, that would be a charge of $0.01.

Internet Explorer 10

[Anonymous Coward]

I'm still a bit miffed that VS2013 requires IE10. I'm unable to install it because it's currently banned at work.

Mandatory registration

[andrew3]

Writing a program in Visual Studio requires mandatory registration, or the program will refuse to start up. This also gives Microsoft to arbitrarily deny specific programmers the ability to publish a program.

Oh, and this, from the VS 2010 Privacy Policy, suggests that Microsoft can remotely target your computer after it does error reporting:

In rare cases, such as problems that are especially difficult to solve, Microsoft may request additional data, including sections of memory (which may include memory shared by any or all applications running at the time the problem occurred), some registry settings, and one or more files from your computer. Your current documents may also be included. When additional data is requested, you can review the data and choose whether or not to send it.

It's somewhat disappointing that Slashdot is used to advertise software like this. Fuck that, I'll stick with free (as in freedom) compilers like GCC, MinGW, LLVM etc. and free IDEs.

Re:Mandatory registration

[wonkey_monkey]

Microsoft can remotely target your computer

When additional data is requested, you can review the data and choose whether or not to send it.

Interesting use of "target."

Re:Mandatory registration

[Anonymous Coward]

Fuck that, I'll stick with free (as in freedom) compilers like GCC, MinGW, LLVM etc. and free IDEs.

I would love to, too... but VS is easy to setup and use. I tried to use Eclipse one - god what hell. Shit not compiling due to all manner of WTF path to lib errors out of the box. Seriously why, after _all_these_years_ and all the millions of users and contributors, can't the open source compiler community make something that you can just install and start coding with???? Hell there is even an example to work against - Visual Studio.

Re:Mandatory registration

[Xest]

"It's somewhat disappointing that Slashdot is used to advertise software like this. Fuck that, I'll stick with free (as in freedom) compilers like GCC, MinGW, LLVM etc. and free IDEs."

Yeah and free browsers like Firefox.

Oh wait, guess what Mozilla does when Firefox crashes? It does the following:

In rare cases, such as problems that are especially difficult to solve, Mozilla may request additional data, including sections of memory (which may include memory shared by any or all applications running at the time the problem occurred), some registry settings, and one or more files from your computer. Your current documents may also be included. When additional data is requested, you can review the data and choose whether or not to send it.

Sound familiar?

Re:Mandatory registration

[Anonymous Coward]

You're a fucking moron and an alarmist.

Same same snippet is true when any program crashes in Windows and prompts you send info back to the mother ship.

Now stfu you smelly hippie.

Re:Mandatory registration

[Anonymous Coward]

In all the years I have been farting around with MS stuff. I have had to do this a few times. Trust me on this. They do not remote into your computer. They *only* do it if you ask them to (and then you usually have to get 3-4th level support before they will consider it). That usually involves using one of your few free trouble tickets (meaning they will not do it), or paying for it (meaning they will do it if you twist their arm enough).

Typically it involves you setting up the DDK and them remoting in and them debugging it. If it turns out to be a bug in their stuff they gratis the case to you. If you just screwed up they charge you. If it is already fixed in a hotfix they will sometimes gratis it.

Microsoft is making it easy...

[betterprimate]

... to quit. It's because of Microsoft that I haven't coded in C++ for fifteen years. Really, is there a single developer on /. that prefers this environment?

Got to taunt: A C++ developer is only useful when he knows how to code in C.

Re:Microsoft is making it easy...

[Xest]

Given that most C++ projects I've received since forever from academic code on EdX through to proprietary game engine code contain a Visual Studio project I'd wager there's an awful lot of developers that prefer Visual Studio, and yes, I suspect a number of those are on Slashdot.

Though I find it a lot odd you say you haven't coded in C++ in 15 years.

Right, well aren't you just Mr Overqualified when it comes to judging then?

I haven't used a BSD distribution in about 15 years so it'd seem a little odd if I said "BSD is shit" given that I have no idea what the fuck it's even like nowadays.

Re:Microsoft is making it easy...

[Dutch Gun]

Apparently, about the time you stopped coding in C++ was the time I started professional C++ development (I'd been teaching myself for a few years before that).

You should take a look at what C++ 11 can do for C++ code. It's rather dramatic. Nowadays, I almost never use raw pointers or call new or delete, and follow RAII practices. Result? Memory management is nearly automatic (it feels almost like garbage collection), and leaks are pretty much forgotten. The class with an actual destructor in it is fairly rare in my codebase, since everything automatically cleans up after itself. It almost feels like coding in C#, just with nastier syntax.

Honestly, I'm enjoying working with C++ more than I ever have in my life. A lot of the tedium is gone, but the raw power and performance still remain. Also, with some real competition in the compiler space, Microsoft finally understands the importance of full compliance - they've announced that they're planning full C++11/14 compliance as an official goal. Here's hoping they'll stick to it, but they seems to be doing a reasonable job so far.

Re:Microsoft is making it easy...

[betterprimate]

I know I was as a troll, but this is actually a decent defense for C++. I'll give it another chance after I'm done learning Ruby.

Thanks for your post.

One major deficiency

[Z00L00K]

The installer that was removed at the introduction of VS2012 has not been re-introduced. That means that now the Nullsoft alternative is more attractive.

The hope that Microsoft would adopt ADA is of course futile.

Visual Studio

[Anonymous Coward]

Did the fix they GUI? Did they bring back setup and deployment projects? Did they listen to feedback? Did they act on feedback?

Questions over questions...

Windows SDK, VS Express, etc

[Zo0ok]

I basically just want C/C++ libraries, compilers and build tools. But not the GUI of Visual Studio.

It used to be possible to Download the Windows SDK/Platform SDK for no charge, and it contained all the command line tools and libraries need to build applications. Now: directly from the download page: "The Windows SDK no longer ships with a complete command-line build environment. You must install a compiler and build environment separately. If you require a complete development environment that includes compilers and a build environment, you can download Visual Studio 2012 Express, which includes the appropriate components of the Windows SDK."

Years ago, Visual Studio C++ Express was 32-bit only, and the Window SDK made more sense for me.

Anyone knows if Visual Studio Express 2013 for Windows Desktop now comes with the 64-bit compiler too? In that case I can forget about the Windows SDK for now. Otherwise I will need to rely on old Windows SDK that came with the compiler. Or install professional version of Visual Studio of course.

Re:Windows SDK, VS Express, etc

[shutdown -p now]

Yes, all Express editions that include C++ ship with a 64-bit compiler from VS 2012 onward.

(I still wish it was a separate download, though. A lot of people don't need to write code, just to compile downloaded stuff - e.g. when installing Python packages from PyPI)

Re:Windows SDK, VS Express, etc

[Zo0ok]

I agree about the separate download... I develop QT/C++ stuff sometimes. I develop in Linux, but I want to build executables for Mac and Win that I can distribute easily. There are different ways to do it, but last time I tried it, I got best result when I compiled QT myself, and then my Application against that QT build. (Perhaps at that time, it was the way to build pure 64-bit executable on Windows). So, yes, I just need the compiler, not the GUI.

But nevertheless it is good news that 64-bit tools come with Express.

SFTP/FTPS support?

[Anonymous Coward]

Until they support SFTP/FTPS within this application, I'm not interested.

Get with the times already Microsoft!

Microsoft platform is facing extinction

[Anonymous Coward]

It would be already dead if it was not for a couple of niche markets (finance for example), ignorant users, and ultra-cheap hardware ($399 shitty laptops that work for a year)
Who wants to use Windows? Once you try Mac, of Linux (if you are a bit more technical) - you wouldn't want to switch ...