Thread: Re: [GD-Windows] VS.net rants, was Re: VC++ lag (Page 2)
Brought to you by:
vexxed72
From: Neil S. <ne...@r0...> - 2003-06-13 14:12:54
|
> [2] Keyboard shortcuts: I was so used to doing a "Find in Files" in VC6 and > then stepping through the reuslts with F4. Why did they have to take that > away? Now I *have* to use the mouse to step through search results, because > F4 is exclusive to the compiler output list. You can map F4 to Edit.GotoNextLocation and Shift-F4 to Edit.GotoPreviousLocation to achieve this. The only difference is that it doesn't (usually) automatically switch to stepping through the find list if you have the task list open and then do a find in files. Once you click in the find in files window, it remembers to step through that from then on. Strangely, it seems to auto-switch to the task list after you do a build, so it would be good if they fixed it so it would work both ways. Here's one from me: For some reason, they have changed the functionality of vector::clear() to actually deallocate memory, rather than simply doing erase(begin(), end()), which leaves the memory allocated. If, like a lot of people, you are relying on the performance characteristics of non-deallocation, you could be suffering a serious performance loss from this "small" change. I was more than a little bit disgusted when I discovered this. Needless to say, we don't use that particular implementation any more. - Neil. |
From: Enno R. <en...@de...> - 2003-06-13 14:51:04
|
> For some reason, they have changed the functionality of vector::clear() to > actually deallocate memory, rather than simply doing erase(begin(), end()), > which leaves the memory allocated. Oh, that's good to know, thanks. I never realized what a hassle the STL can be when switching compilers (or more precisely, STL versions). Especially if you buy in libraries that use it - we've got one. We now need to ask the vendor to compile us a new version that fits with the VC7.1 version of STL. Other libraries (we use ACE) need to be recompiled for every new STL version. There's probably no really good solution for this... Enno. |
From: Ben C. <be...@gu...> - 2003-06-13 18:11:05
|
On Friday, June 13, 2003, 1:44:24 PM, someone wrote: > I can add a few cents here, too... ...and a "few" more... 1) Various source-control operations cause the Solution Explorer to spend several minutes jumping randomly between files in the project (apparently opening/closing/saving them or whatever) if you have many windows open. 2) In fact, can the Solution Explorer jump-to-current-file behaviour be turned off? It's occasionally useful, but far more often just annoying. 3) Editing multiple project properties at once is buggy, as is editing multiple configurations - frequently the properties window mysteriously looses changes, vanishes altogether, displays "there are no properties for the selected item", or, in extreme cases, simply crashes VS.net entirely. I've had to edit the .vcproj files by hand a couple of times to make changes thanks to this... 4) If you do a "get latest" from within Sourcesafe recursively, and several .vcproj files are updates, then VS.net only notices the first one that's changed, and blindly goes on using the others without reloading them. 5) "Get latest" from within the IDE is useless because it seems to operate on the *current* .vcproj file, not the one in Sourcesafe (or a merge of the two), and hence misses newly-added files. 6) Custom build step dependencies are crippled - it seems impossible to set up a configuration that allows (for example) compiling a vertex shader and then "incbin-ing" the resulting bytecode into a header file for inclusion in the project in such a way that everything rebuilds correctly without resorting to multiple rebuilds and/or abuse of "touch" on things. 7) Can we have an option to turn off the task list? It's just a pointless recap of the information in the build window (usually with the important bits missing), and more often than not you spot the error in the build window, go to read it and... blam, you're looking at the task list instead because the build just aborted. 8) An option to suppress the comments dialog when checking out files would be good. We never use them, but I've lost track of how many checkouts now have comments along the lines of "nt blah=0;" because we're all used to just hitting ENTER to clear the "check out now?" prompt... 9) I haven't tried too hard to make this work so it could just be a misplaced setting, but Japanese SJIS text support in the editor is broken (it worked in VC6). 10) And yes, yet another plea for a one-step "show settings for current project and build" button. And, if possible, a convenient way to change the "Debug->Command line parameters" setting for the current project (in fact, the option to have that as a edit box on the toolbar would be best). ...er.. I think I'll stop now ^-^; -- Ben Carter - Neko Technologies - be...@gu... http://www.neko-tech.com/ - http://www.absoluteterror.com/ ---------------------------------PGP Key available on request--- "Broken mirror, a million shades of light, the old echo fades away. But just you and I can find the answer, and then we can run to the end of the world." - Small of two pieces, Xenogears |
From: Enno R. <en...@de...> - 2003-06-14 07:51:14
|
Ben Carter wrote: Lots of stuff about SourceSafe that I wouldn't know about, happy CVS user that I am. But I've got one answer: > 2) In fact, can the Solution Explorer jump-to-current-file behaviour > be turned off? It's occasionally useful, but far more often just > annoying. Yes, it can. Options/Environment/Projects_and_Solutions has a checkbox for "Track Active Item in Solutions Explorer". Not sure if it had that in the first one, but it's in 2003. > 7) Can we have an option to turn off the task list? It's just a > pointless recap of the information in the build window (usually with > the important bits missing), and more often than not you spot the > error in the build window, go to read it and... blam, you're looking > at the task list instead because the build just aborted. In the same Option dialog, "Show Task List window if build finishes with errors", seems to be new with 2003. Enno. |
From: Enno R. <en...@de...> - 2003-06-13 12:02:05
|
Carsten Orthbandt wrote: > I'd like to mention that I reported several bugs of VS.Net > without any response. And hey, in VS.Net 2003, none of these > bugs was fixed. I've reported only one bug so far in VS.Net - through the Online Assisted Support system. This may sound like I'm hired to do propaganda for MS, but I had prompt and competent support. Their incident tracking system is really nice, too. Not oly did I get the bug confirmed by them, they also tried the problem in VS.net 2003, made sure that it had been fixed by the STL changes in there, and sent us free upgrades to VS.net 2003 for all our licenses. Enno. |
From: Rich <leg...@xm...> - 2003-06-13 15:15:02
|
In article <009401c3318e$72581c80$0301a8c0@pc350>, "Colin Fahey" <cp...@ea...> writes: > [1] I guess this will only ever happen to any person > once, but it sure is shocking: > > Deleting files from the project deletes the source > files themselves! This doesn't happen with all projects all the time. It depends on how the files are included in the project. It is documented under Visual Studio .NET / Developing with Visual Studio .NET / Managing Solutions, Projects, and Files / Solutions as Containers / Projects as Containers / Item Management in Projects. You do read the documentation on a new product before you start using it, don't you? ;-) -- "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ: <http://www.xmission.com/~legalize/book/> izfree: Open source tools for Windows Installer <http://izfree.sourceforge.net> |
From: Colin F. <cp...@ea...> - 2003-06-13 23:25:07
|
> > [1] I guess this will only ever happen to any person > > once, but it sure is shocking: > > > > Deleting files from the project deletes the source > > files themselves! > > This doesn't happen with all projects all the time. It depends on how > the files are included in the project. It is documented under Visual > Studio .NET / Developing with Visual Studio .NET / Managing Solutions, > Projects, and Files / Solutions as Containers / Projects as Containers > / Item Management in Projects. > > You do read the documentation on a new product before you start using > it, don't you? ;-) Let's just say that Microsoft gave notice about their plans to destroy my files just as the Vogons gave notice of the destruction of the Earth in "Hitchhiker's Guide to the Galaxy" (Douglas Adams): Vogon: There's no point in acting all surprised about it. All the planning charts and demolition orders have been on display in your local planning department on Alpha Centauri for fifty of your Earth years, so you've had plenty of time to lodge any formal complaint and it's far too late to start making a fuss about it now. [...] Vogon: What do you mean you've never been to Alpha Centauri? For heaven's sake mankind, it's only four light years away you know. I'm sorry, but if you can't be bothered to take an interest in local affairs that's your own lookout. Energise the demolition beams. This parallels Arthur Dent's experience of the notice of the plan to demolish his house: - ...You hadn't exactly gone out of your way to call attention to them had you? I mean like actually telling anyone or anything.' - But the plans were on display...' - On display? I eventually had to go down to the cellar to find them.' - That's the display department.' - With a torch.' - Ah, well the lights had probably gone.' - So had the stairs.' - But look you found the notice didn't you?' - Yes,' said Arthur, `yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying "Beware of The Leopard". --- Colin cp...@ea... |
From: David N. <dno...@mi...> - 2003-06-13 16:41:33
|
I would say the default setting is wrong. I was pretty annoyed by that one too (specially when you use a source control that doesn't integrate with vs) -----Original Message----- From: gam...@li... [mailto:gam...@li...] On Behalf Of Rich Sent: Friday, June 13, 2003 8:15 AM To: gam...@li... Subject: Re: [GD-Windows] VC++ lag=20 In article <009401c3318e$72581c80$0301a8c0@pc350>, "Colin Fahey" <cp...@ea...> writes: > [1] I guess this will only ever happen to any person > once, but it sure is shocking: =20 >=20 > Deleting files from the project deletes the source > files themselves! This doesn't happen with all projects all the time. It depends on how the files are included in the project. It is documented under Visual Studio .NET / Developing with Visual Studio .NET / Managing Solutions, Projects, and Files / Solutions as Containers / Projects as Containers / Item Management in Projects. You do read the documentation on a new product before you start using it, don't you? ;-) --=20 "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ: <http://www.xmission.com/~legalize/book/> izfree: Open source tools for Windows Installer <http://izfree.sourceforge.net> ------------------------------------------------------- This SF.NET email is sponsored by: eBay Great deals on office technology -- on eBay now! Click here: http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5 _______________________________________________ Gamedevlists-windows mailing list Gam...@li... https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows Archives: http://sourceforge.net/mailarchive/forum.php?forum_id=3D555 |
From: Rich <leg...@xm...> - 2003-06-13 19:12:04
|
In article <6C3...@RE...>, "David Notario" <dno...@mi...> writes: > I would say the default setting is wrong. AFAIK, its not a setting, its a property of the project IDE. C#/VB.Net projects apparently don't use links, i.e. the project item -is- the file, while C++ projects do use links, i.e. deleting the project item doesn't touch the file. -- "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ: <http://www.xmission.com/~legalize/book/> izfree: Open source tools for Windows Installer <http://izfree.sourceforge.net> |
From: Carsten O. <car...@se...> - 2003-06-14 09:04:43
|
>=20 > -----Original Message----- > From: Jon Watte [mailto:hp...@mi...]=20 > Sent: Friday, June 13, 2003 7:55 PM > To: gam...@li... >=20 >=20 > * Is there a key for go-to-next-Find-In-All-Files-match ? >=20 > I haven't found it. I don't like reaching for the mouse just to=20 > navigate my Find results. "Go to next task" does not do it; it only=20 > seems to work for compile errors etc. Yes, there is. The default is (IIRC) Ctrl-F6 and Ctrl-Shift-F6. But I always remap it to Alt-Down and Alt-Up. The GUI commands are Edit.GotoNextLocation and Edit.GotoPreviousLocation. These commands step through the build log and find in files results depending on which of both is currently visible (wonder what happens if both are). Carsten Orthbandt Founder + Development Director SEK SpieleEntwicklungsKombinat GmbH http://www.sek-ost.de Wenn ich Visionen habe, gehe ich zum Arzt. - Helmut Schmidt =20 |
From: Gareth L. <GL...@cl...> - 2003-06-14 22:34:40
|
I thought vc 7.1 (vs.net 2003) supported partial template specialization ? > -----Original Message----- > From: Parveen Kaler [mailto:pk...@al...] > Sent: 14 June 2003 21:00 > To: gam...@li... > Subject: RE: [GD-Windows] VC++ lag > > > On Thu, 2003-06-12 at 13:05, David Notario wrote: > > If you guys can provide a list things where vs can improve (from its > > latest version), with arguments such as (X sucks because > this, this and > > that. Blah doesn't work in this situation), I'll be happy to forward > > your list of petitions to the VS team. > > > Since no one else has mentioned it. How about partial template > specialization sometime soon? > > I can live with most of the quirks of the IDE. > |
From: Parveen K. <pk...@al...> - 2003-06-17 04:07:39
|
On Sat, 2003-06-14 at 15:34, Gareth Lewin wrote: > I thought vc 7.1 (vs.net 2003) supported partial template specialization = ?=20 Hmm, looks like it does. Not to mention Koenig Lookup as well. http://msdn.microsoft.com/visualc/productinfo/overview/whatsnew.aspx Not sure how they came up with the 98% conformance number. I suspect it involved someone's nether regions. It's to bad we're using 7.0 at work. Sheesh, time to upgrade again. |
From: Dan T. <da...@cs...> - 2003-06-17 05:06:00
|
I'd have to comment on a couple of bugs. 1) Unix newlines. Dear god this caused trouble at my school... We had to write a game in 10 weeks that had a bunch of stuff.. we used CVS, which is fine, but one guy checked in stuff from a unix prompt, and merging filled the code with unix newlines. The guy did something to try and strip them out(millions of tools that are supposed to work...dunno what he did), but what we were left with was basically double spaced code. Works, Compiles. Until you try to edit it. Then invisible newline problems show up, and the compiler freaks on strange invisible tokens. The compiler should ignore anything non-alphanumeric/Windows Newline in the source. (read as ignore anything not displayed in the editor window!). This was an issue with .Net, havent tried 2k3 yet. (class is over, no point since I don't use *nix) 2) There doesn't seem to be a way to get from the build window to the last window you were working on. One might *think* Ctrl-Tab would do it, but that moves you to the next source window based on the last source you were working on(does that make sense?), so in essence you have to hit ctrl-tab twice. Kinda irritating. -Dan ----- Original Message ----- From: "Parveen Kaler" <pk...@al...> To: <gam...@li...> Sent: Monday, June 16, 2003 9:07 PM Subject: RE: [GD-Windows] VC++ lag |
From: Enno R. <en...@de...> - 2003-06-17 07:53:25
|
Dan Thompson wrote: > I'd have to comment on a couple of bugs. > > 1) Unix newlines. Dear god this caused trouble at my school... We had to > write a game in 10 weeks that had a bunch of stuff.. we used CVS, which is > fine, but one guy checked in stuff from a unix prompt, and merging filled > the code with unix newlines. The guy did something to try and strip them > out(millions of tools that are supposed to work...dunno what he did), but > what we were left with was basically double spaced code. Works, Compiles. I would say your problem is mostly with wrong use of CVS, not with Visual Studio. You should never mix Unix/Windows newlines in your code, and as long as you don't mix unix and windows sandboxes, you're fine. Fix the CVS contents, and Visual Studio will be happy again. Worst thing you can do to CVS, by the way, is to use ssh from cygwin. Totally messes up the repository, (you get \r\r\n newlines when checking out on windows with non-cygwin cvs). Enno. -- <xterm> The problem with America is stupidity. I'm not saying there should be a capital punishment for stupidity, but why don't we just take the safety labels off of everything and let the problem solve itself? |
From: Carsten O. <car...@se...> - 2003-06-17 07:39:57
|
No idea on the first one, but getting from the build output back into the text editor is as easy as pressing ESC... Carsten Orthbandt Founder + Development Director SEK SpieleEntwicklungsKombinat GmbH http://www.sek-ost.de Wenn ich Visionen habe, gehe ich zum Arzt. - Helmut Schmidt =20 >=20 > -----Original Message----- > From: Dan Thompson [mailto:da...@cs...]=20 > Sent: Tuesday, June 17, 2003 7:20 AM > To: gam...@li... >=20 > I'd have to comment on a couple of bugs. >=20 > 1) Unix newlines. Dear god this caused trouble at my=20 > school... We had to > write a game in 10 weeks that had a bunch of stuff.. we used=20 > CVS, which is > fine, but one guy checked in stuff from a unix prompt, and=20 > merging filled > the code with unix newlines. The guy did something to try and=20 > strip them > out(millions of tools that are supposed to work...dunno what=20 > he did), but > what we were left with was basically double spaced code.=20 > Works, Compiles. > Until you try to edit it. Then invisible newline problems=20 > show up, and the > compiler freaks on strange invisible tokens. The compiler=20 > should ignore > anything non-alphanumeric/Windows Newline in the source.=20 > (read as ignore > anything not displayed in the editor window!). This was an=20 > issue with .Net, > havent tried 2k3 yet. (class is over, no point since I don't use *nix) >=20 > 2) There doesn't seem to be a way to get from the build=20 > window to the last > window you were working on. One might *think* Ctrl-Tab would=20 > do it, but that > moves you to the next source window based on the last source you were > working on(does that make sense?), so in essence you have to=20 > hit ctrl-tab > twice. Kinda irritating. >=20 > -Dan >=20 > ----- Original Message -----=20 > From: "Parveen Kaler" <pk...@al...> > To: <gam...@li...> > Sent: Monday, June 16, 2003 9:07 PM > Subject: RE: [GD-Windows] VC++ lag >=20 >=20 >=20 >=20 > ------------------------------------------------------- > This SF.Net email is sponsored by: INetU > Attention Web Developers & Consultants: Become An INetU=20 > Hosting Partner. > Refer Dedicated Servers. We Manage Them. You Get 10% Monthly=20 > Commission! > INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php > _______________________________________________ > Gamedevlists-windows mailing list > Gam...@li... > https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows > Archives: > http://sourceforge.net/mailarchive/forum.php?forum_id=3D555 >=20 >=20 |
From: Gareth L. <GL...@cl...> - 2003-06-17 08:25:42
|
> Hmm, looks like it does. Not to mention Koenig Lookup as well. > http://msdn.microsoft.com/visualc/productinfo/overview/whatsnew.aspx > Not sure how they came up with the 98% conformance number. I > suspect it > involved someone's nether regions. Isn't that a bit unfair ? I know it's Microsoft and all[1] but you haven't used vc7.1 so why automatically assume they cheated ? > > It's to bad we're using 7.0 at work. Sheesh, time to upgrade again. > ^^ My point. ________________________ Regards, Gareth Lewin Programmer, Sudeki http://www.xbox.com/Sudeki [1] I use the term "I know it's Microsoft" because other people automatically distruct and/or hate them. Personally I reserve feelings for people. |
From: Rich <leg...@xm...> - 2003-06-17 16:57:57
|
In article <015e01c33490$092ffa20$6500a8c0@quiksilver>, "Dan Thompson" <da...@cs...> writes: > 1) Unix newlines. Dear god this caused trouble at my school... We had to > write a game in 10 weeks that had a bunch of stuff.. we used CVS, which is > fine, but one guy checked in stuff from a unix prompt, and merging filled > the code with unix newlines. Then you used CVS wrong. I've done this plenty of times -- shared code between PCs and unix boxes -- and if you get unix newlines on a PC or PC newlines on a unix box then you have used the tool incorrectly. -- "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ: <http://www.xmission.com/~legalize/book/> izfree: Open source tools for Windows Installer <http://izfree.sourceforge.net> |
From: <cas...@ya...> - 2003-06-17 20:12:42
|
Rich wrote: > Then you used CVS wrong. I've done this plenty of times -- shared > code between PCs and unix boxes -- and if you get unix newlines on a > PC or PC newlines on a unix box then you have used the tool > incorrectly. Yes, I also had that problem some time ago, and it was because of cygwin. Cygwin allows you to change the newspace mode, and one of them produces that problem. If I remember correctly, I was using the unix mode, don't know why, but switching to the other mode solved the problem. Ignacio Castaño cas...@ya... |
From: Dan T. <da...@cs...> - 2003-06-17 21:06:49
|
I personally wasn't having any problems - just the one guy who wouldn't listen to us telling him to check in through windows and not SSH. However, I think this is all beside the point - I think the compiler should just toss out characters that aren't doing anything... Since code is just text, that basically limits you to letter, numbers, spaces, tabs, and windows newlines, right? (Not excluding the shift-something characters). I wouldn't think it would be difficult to just ignore those... unless I'm missing something. -Dan ----- Original Message ----- From: "Ignacio Castaño" <cas...@ya...> To: <gam...@li...> Sent: Tuesday, June 17, 2003 1:18 PM Subject: Re: [GD-Windows] VC++ lag > Rich wrote: > > Then you used CVS wrong. I've done this plenty of times -- shared > > code between PCs and unix boxes -- and if you get unix newlines on a > > PC or PC newlines on a unix box then you have used the tool > > incorrectly. > > Yes, I also had that problem some time ago, and it was because of cygwin. > Cygwin allows you to change the newspace mode, and one of them produces that > problem. If I remember correctly, I was using the unix mode, don't know why, > but switching to the other mode solved the problem. > > > Ignacio Castaño > cas...@ya... > > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: INetU > Attention Web Developers & Consultants: Become An INetU Hosting Partner. > Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission! > INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php > _______________________________________________ > Gamedevlists-windows mailing list > Gam...@li... > https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows > Archives: > http://sourceforge.net/mailarchive/forum.php?forum_id=555 > |
From: Pierre T. <p.t...@wa...> - 2003-06-12 16:00:34
|
> Delete your .opt file, normally solves that. It's checking for dependancies > and something go screwed up ( I guess ) Sigh.... Unfortunately I'd like to keep that one alive, as it's where currently open files are stored (I think). I like to keep some files open from day to day (the files I'm currently working on). Anyway I guess I'll give it a try. >The .ncb file is another pest. No problem with those as I delete them frequently (each time I make a backup, to save some space) >Do you use VisualAssist, or anyother kind of VC++ pluggin ? No. >Do you have workspaces that contains a lot of subprojects, with the >mainproject having all other projects considered as dependancies of the main >? Yes. Pierre |
From: Mickael P. <mpo...@ed...> - 2003-06-12 16:19:58
|
>> Do you have workspaces that contains a lot of subprojects, with the >> mainproject having all other projects considered as dependancies of >> the main ? > > Yes. Ok. Each time I have a workspace with this kind of situation the first build of the complex project after loading visual is always slow. It seems that visual check each file date of each subtarget to see if it must be rebuilt, and it takes an insane amount of time for some strange reason I didn't understand :) Try to add a new "hello world" project in your workspace. Save the workspace and quit visual. Reload visual, select the "hello world" as active project, build. It should start instantaneously. Quit and reload, select your complex main target, and build. It should take forever before starting building. Mickael Pointier |
From: Jon W. <hp...@mi...> - 2003-06-12 19:38:27
|
Hi again, I noticed you released 1.3 of Opcode. Gotta check that out sometime=20 soon! (That "time" thing ...) Anyway, my question for you is where you found LSS-triangle collision=20 code, which you talk about in the notes about Opcode. I looked at=20 Magic Software, and they actually don't provide that. I want to get=20 some readable code to show a particular stick-in-the-mud around the=20 office that it truly is easier to use LSS-tri intersection that=20 returns time of earliest intersection, than first moving the sphere=20 and then detecting what happened and trying to patch it up from=20 there (complete with tunneling through walls and everything). If you know of a link that could help me out, or could donate a page=20 of code, I'd be much obliged. Cheers, / h+ |
From: Jon W. <hp...@mi...> - 2003-06-12 19:44:04
|
> Hi again, >=20 > I noticed you released 1.3 of Opcode. Gotta check that out sometime=20 > soon! (That "time" thing ...) And this was of course aimed at Pierre and shouldn't have gone to the=20 list. Here's hoping you got amusement out of my personal confessions :-) Cheers, / h+ |
From: Pierre T. <p.t...@wa...> - 2003-06-12 21:54:05
|
>Anyway, my question for you is where you found LSS-triangle collision >code, which you talk about in the notes about Opcode. I looked at >Magic Software, and they actually don't provide that. I want to get >some readable code to show a particular stick-in-the-mud around the >office that it truly is easier to use LSS-tri intersection that >returns time of earliest intersection, than first moving the sphere >and then detecting what happened and trying to patch it up from >there (complete with tunneling through walls and everything). Let me reply to the list as well, just to make it clear : - The LSS-triangle collision code really comes from Magic Software. It's actually a segment-triangle distance query. - Before you tell me this doesn't give the earliest time of intersection, let me point out some code that does : http://www.three14.demon.nl/ . There was a bug in his code but I reported it to the author some weeks ago so it should be fixed now. Pierre |