topbanner_forum
  *

avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Friday March 20, 2026, 9:28 pm
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Recent Posts

Pages: prev1 ... 334 335 336 337 338 [339] 340 341 342 343 344 ... 364next
8451
Carol, you're a win2k user I take it? :)

(XP only has regedit, regedt32 is just an alias for it. That regedit can set permissions and stuff, though).
8452
but perhaps that's what they've needed to convince them to cut out some of the fat and get back to focusing on quality.

Ho humm, cut fat and focus on quality? I just don't see that happening, with them focusing on eye-candy in Vista. Even WinXP came on a CD, Vista comes as a 3.19GB DVD... "but that's not a problem, modern harddrives are large!" - sure, but that size tells you something about system requirements as well. "But that's no problem, everybody have fast CPUs, lots of RAM, and powerful GPUs". No, not everybody has, and even for people that do, do you want your base operating system to chew up massive resources?

Their endless company acquisitions seems like "we've got to have that feature too", which either means MS gets bigger and suffer even lousier internal communication, or that there's less people to manage more code - resulting in ever-poorer software.

I'm not at all impressed. Too bad there's no viable alternative to Windows >_<
8453
Site/Forum Features / Re: Wanted: Cody Cartoons
« Last post by f0dder on July 20, 2006, 12:56 PM »
Cartoons!  :-*
8454
Living Room / Re: What's the best registry cleaner? Ask Leo says: none
« Last post by f0dder on July 20, 2006, 05:50 AM »
I am still using a old and safe cleaner program call RegCleaner which is a freeware but no longer supported,

If you're referring to the MS tool that has an icon of a cyan box with a red ribbon on it, that tool is not safe.

I haven't really noticed any cases where a registry cleaner caused any noticeable speed effect; the registry is stored in a binary and pretty efficient format. Looking up keys/values uses a binary search, which means that to find a key out of 4.294.967.296 takes a maximum of log(2**32)/log(2) compares - that's 32 compares. Of course most registries have a lot less entries than that, so it's even faster.

One thing that DOES help, though, is to keep your registry defragmented..
8455
Living Room / Re: Why is Microsoft software so unreliable?
« Last post by f0dder on July 20, 2006, 05:39 AM »
Personally I've never liked autosave - it's better to get a good habit of "one-on-control-anoter-on-s" regularly. I don't even think about that keyboard combo anymore, happens automatically. IMHO more reliable than autosave. (Oh, and VS saves on project builds btw.)

Use of temporary files for saving is generally a good idea though ("atomic commit" in a way). If it's not implemented in VS, I guess it's because "what could go wrong when saving a ~100kb file?" - *shrug*.

As for your laptop, the ACPI stuff needed for that to work well has horrible implementations on a lot of systems, so it's not necessarily Microsoft's fault. Problems can also be caused by third-party drivers not supporting the Windows Driver Model power-saving stuff properly.
8456
General Software Discussion / Re: Windows XP Myths
« Last post by f0dder on July 20, 2006, 05:18 AM »
And the only reason why there are more reported vulnerabilities for linux applications in the first place, is because they are easyer to spot, since it's all open source.
That's not necessarily true... first of all there's (private) tools for finding exploits, appearantly some of them are pretty efficient. But even without such a tool and without source, it's not necessarily hard to find an exploit. I was chatting with a grey-hat friend of mine while he looked for holes in either AIM or Yahoo chat (can't remember which one). It took him between 30-60 minutes to find a 0-day exploit.

On linux they usually get spotted, reported, and fixed quite fast, which adds to the security.
How long did the chunked-mode exploit exist in Apache before it was discovered? (Discovered by full-disclosure people, anyway ;) ).
8457
General Software Discussion / Re: Windows XP Myths
« Last post by f0dder on July 19, 2006, 03:07 PM »
just when i thought it was time to stop believing in the myths. oh well, back to being confused and unsure of what to believe - just as well i can't really be bothered with all this tinkering - my pc appears to working at the moment so i'll leave it well alone.
Well, those were the only two items I really disagree with :) - and leaving it alone if it works is probably a good idea. I've built up my collection of tweaks over several years, and merged them in my unattended XP setup CD. Otherwise I probably couldn't be bothered (but would swear at XP often ;)).
8458
General Software Discussion / Re: Windows XP Myths
« Last post by f0dder on July 19, 2006, 02:14 PM »
Humm, that article isn't entirely correct.

For instance...

Disable the Pagefile
They claim that there's no performance benefit from disabling the paging file... well, that's not true. Windows tends to page out to disk in a lot of situations where it doesn't really make sense (partially because of badly designed usermode programs, though). So if you have plenty (at least a gig, 2gig preferred) you can avoid some needless paging by turning off the paging file. It's not a big improvement for most situations, but it does help a little.

Of course if you run anything from Adobe, forget about turning off your paging file... Adobe apps are badly designed monsters that don't have decent memory management. And of course, some games have extreme memory requirements as well (funny enough a relatively simple game like PainKiller uses enough memory that a gig of ram + no paging file crashes on some levels).

Also note that this only works for XP, Windows 2000 and below require at least a minimal (~20meg) paging file, and will create one at boottime if you've disabled it.

The article also seems to confuse 80386 protected mode "virtual memory"/"paging" with the process of paging in/out from disk - just because 80386 paging is enabled doesn't mean you have to page (or swap) to disk.

Large System Cache Tweak
This tweak can be nice on Desktop machines, not just servers, depending on the way you use your system. For me this is a VERY nice tweak. It means that, for instance, when I've used nLite to create a slipstreamed install CD, once the files are prepared and I go to the ISO creation stage, almost all files will be in the filesystem cache, so almost all reads will go from memory instead of drive... lots faster.

but the changed pages occupy memory that might otherwise be used by applications is a moot point, since filesystem cache will always be dropped when applications request memory and there's currently not enough free memory. On workstations this increases paging and causes longer delays whenever you start a new app. is plainly wrong, there's actually a better chance of your .exe and .dlls being in memory (equals shorter loading time) if you have LargeSystemCache=1.

HOWEVER, never use LargeSystemCache=1 on machines with ATI graphics cards and drivers. ATI drivers are nasty... see this post.
8459
truth be told i'm wary of all of these executable-wrapper protection tools, and prefer using a full virtual machine tool like vwware or virtualpc.
Yeah, it's more secure. Anything based on API hooking shouldn't be too hard to circumvent. BufferZone does sound a bit interesting, though, in that it uses a kernel mode filter driver instead of simple ring3 API hooking.
8460
General Software Discussion / Re: Office or Vista: Pick one in 2007
« Last post by f0dder on July 19, 2006, 08:50 AM »
If you twisted my arm, I'd upgrade Office - at least I'd only suffer the extra bloat when writing letters, not when doing everything else.
8461
Living Room / Re: Want Vista NOW !!! Here's how ...
« Last post by f0dder on July 19, 2006, 08:42 AM »
Perhaps for the media portion they mean they can use QoS to allocate more bandwidth to streaming media. Increasing process priority on the player wont do this for you.
True enough, and there could be some advanced in kernel-streaming drivers as well (which has nothing to do with internet streaming btw), etc. There's nothing stopping QoS on XP though, and iirc win2k supports it as well - just a matter of making the media player utilize it. Oh, and the rest of your networking equipment.
8462
Living Room / Re: Want Vista NOW !!! Here's how ...
« Last post by f0dder on July 19, 2006, 05:25 AM »
Pretty boring and irrelevant article IMHO... also, what the heck do they mean by this?
However, users with dual-core PCs may notice a performance boost with Vista when they're multitasking, because Vista can take advantage of dual cores to run separate processes, something that XP can't do. So if you have a dual-core PC and you frequently run multiple programs simultaneously, you will lose the performance improvement that Vista can bring.
Even NT4 supports multi-CPU systems. And multitasking isn't done per-process anyway, it's done per-thread.

Microsoft also claims that multimedia will run better in Vista than XP because Vista can give streaming audio and video priority over other processes so that they won't be interrupted.
That sounds a bit silly as well - you can already manually (or with whatever tool) raise your media player process priority.
8463
Oh no :(

I hope they won't discontinue/mess up the free (sysinternals) tools - nor the (winternals) flagships like ERD commander. Or put a lid on Mark Russinovich.
8464
Developer's Corner / Re: its a myth, a MYTH I say!
« Last post by f0dder on July 18, 2006, 12:46 PM »
Idea is simple, .obj files no longer have machine code but rather the intermediate "p-code" the compiler is generating anyway, and optimizing is deferred until the link stage.

Which to me sounds as if it's not too far away from just-in-time compilation. I think in the future we will have more abstract machines which optimize themselves all the time; total virtualization so to say. I will have to pick up the proper technical terms so I can better explain what I mean.

The trend seems to be to generate the machine code as late as possible in the uhm "runtime/lifetime cycle" of The Code.

Perhaps - we're not there yet, though. Suns JVM has abysmal performance (MS JVM beat it by far), and while dotNET is pretty okay, it's still easy to beat with native code (code speed as well as memory use) - it looks like it's hard to beat dotNET for development time in a bunch of situations, though.

There's a lot of interesting techniques that can be applied with JIT optimization, some of which is hard to do with normal code optimization. For instance, taking note of branch prediction misses and rearranging the code to minimize this. However, the profiling overhead needed for doing this is not free - and it can be done with profile-guided "static" code generation as well (Iirc supported in vc6, vc2005, intel compilers).

Then there's more or less specialized pieces of code, where (on x86 anyway) a decent programmer can beat compilers by pretty large amounts. Compilers still don't have extremely good intrinsics support for new instruction sets (SSE/2/3, MMX) - besides those intrinsics makes code less portable, so it's safer and more portable to write an external assembly module. Even for code using the plain old instruction set, it's still possible to beat compilers - but of course this rarely makes sense (hence "specialized pieces of code").

As long as we stay on x86 (which will probably be pretty much forever, after AMD introduced x86-64 and since Apple moved to intel hardware, curse them both) I don't see JIT'ing replace normal code generation... unless there's enough SHEEP who blindly swallow VISTA.
8465
Developer's Corner / Re: its a myth, a MYTH I say!
« Last post by f0dder on July 18, 2006, 11:35 AM »
Not a bad little article, but nothing majorly new either. And not 100% correct...

Due to the way C works, it’s impossible for the compiler to inline a function defined in another source file. Both source files are compiled to binary object files independently, and these are linked.
Which is wrong, considering Link-Time Code Generation (visual C++ 2003 or later). Intel's C++ compiler has had it for a while too, called something else though. Works pretty well. Idea is simple, .obj files no longer have machine code but rather the intermediate "p-code" the compiler is generating anyway, and optimizing is deferred until the link stage.
8466
opendns... yay. As in, NO FRIGGING WAY!

Ping time to opendns.com - ~200ms
Ping to my ISP DNS server - ~30ms.

I bet my ISP has at least as big a cache as OpenDNS has, so no gains there, only slower response time. If OpenDNS does more agressive caching than the RFCs allow, they'll break stuff (at best just make DNS propagation slower than it already is).

And what's with the whole "fix typos" thing? Who decides whether a domain name is a typo or not? (hmm, they're ad financed... I wonder who decides...)

Also, some software depends on getting a "domain not found" instead of a bloody search page. And I'd be really unhappy about a search page myself, smells too much of adware.

 :down: :down: :down:
8467
BOCHS does full emulation of all instructions (S L O W!)

I have heard that about it. But I have also heard that it is of more use to someone writing their own OS, rather than someone looking to run an OS virtually in order to try one out, or for testing applications on one.
Indeed - since vmware and virtualpc can sometimes have problems with homebrewn kernels (mostly because of people doing things in somewhat "interesting" ways, though).

Bochs still has some bugs here and there though, and I bet QEMU does as well... you really need a couple of emulators as well as real hardware if you're doing OS development.
8468
And QEMU, which has a "kernel-mode acceleration module" for faster virtualization, whereas BOCHS does full emulation of all instructions (S L O W!)
8469
What does "elegant" really mean, in the context of programming?  Does it mean anything at all?
For me, it's code that you understand on the first read-through. Well-designed classes. And sometimes, just a "gut feeling" when something is done cleverly ("clever" doesn't mean "hackish" in this content, btw!).

mouser, et. al. - While I agree that your examples of bad code are indeed bad, I don't condem the language for it.  Just becuase you can use it to write bad code, doesn't mean you should.
-rover
You're right, of course... some langauges just lend themselves very well to writing hacky code. Heck, for some languages almost only hacky code is written :)
8470
Haha, what a fun toy  :-*
8471
Borland C++ Builder Contest / Re: Vortex Chat - Client Version
« Last post by f0dder on July 14, 2006, 03:51 AM »
I havent seen a chat program that messes with a web browser, installs spyware or viruses at all. What ones have you tried? I know gaim is perfectly safe, as is trillian, miranda, aim, msn, yahoo, icq and google talk. IRC (mIRC) is also a good chat protocol/series of networks.
Actually, AIM for one def. has spywhere included in the install. I am a computer technician during the summer and I know about this one because I, myself, have personally formatted an entire computer and then, reinstalled windows and put AIM on there... Wasn't long before I found things on the computer... Actually, about 5 min's. After installing and launching, I was able to view a few things that I didn't install. Hence, the fact that AIM and yahoo aren't really as safe as you think... Just because  everything looks fine doesn't necessarily mean that it is... I have found that out for sure! lol!

Was this a Windows XP or 2000 install without firewall/NAT enabled? If so, perhaps then that was the thing to blame, not AIM. I wouldn't be surprised, though, if there was annoying stuff bundled with AIM.
8472
Living Room / Re: Firefox 2.0 beta available
« Last post by f0dder on July 13, 2006, 02:16 PM »
Spell checking in a web browser? Wtf? >_<
8473
Living Room / Re: Zidane's head butt mashup videos
« Last post by f0dder on July 13, 2006, 11:44 AM »
Heh, just shows what a bunch of morons soccer players are  :-\
8474
I tend to agree with the stuff you've just posted here, mouser. The forms of multi-assign shown is mostly for crap-coders and leetkids.

It can, however, be useful for languages that support returning multiple values from a function - alleviates the need for endless pointer-to-return-value function arguments :) (but of course, "use a struct" would be a viable option as well.)
8475
Living Room / Re: Microsoft: End of support for Windows 98 and Windows Me
« Last post by f0dder on July 13, 2006, 11:38 AM »
Just for the record, 9x includes ME. It's still the same shoddy old "dos extender gone wild". So you shouldn't say "upgraded from 9x to ME" but "down^H^H^H^Hupgraded from 98 to ME".

Too bad about lack of driver support, otherwise I'd much rather be running NT4 than any 9x, if the hardware was that limited. But that's probably because I'm a programmer and can't live with the leaks and instability of 9x (ùser-mode errors being able to cause BSODs? No way, man.)
Pages: prev1 ... 334 335 336 337 338 [339] 340 341 342 343 344 ... 364next