topbanner_forum
  *

avatar image

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

Login with username, password and session length
  • Saturday December 20, 2025, 1:14 am
  • 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 ... 96 97 98 99 100 [101] 102 103 104 105 106 ... 364next
2501
General Review Discussion / Re: Daemon Tools
« Last post by f0dder on March 09, 2010, 03:48 AM »
Dunno about spyware in later versions - there's been some toolbar stuff which you probably don't want installed, but afaik you can choose not to install it. But really, if you don't need the piracy related features of daemon-tools, go for magicdisc or virtual clonedrive. I did so a couple of years ago, and haven't missed d-t :)
2502
General Software Discussion / Re: Why the aversion to .NET Frameworks?
« Last post by f0dder on March 09, 2010, 01:08 AM »
SchoolDaGeek: sounds to me more like HP has a modified windows setup that has b0rked stuff, and possibly also that you've b0rked stuff additionally with nLite (which is easy to do, been there & done that).

I just don't understand why we need to download and install a 200mb library of 'possibilities' to use a program that is 4mb in size....
-SchoolDaGeek (March 08, 2010, 10:43 PM)
Because it gives developers at crapload of functionality that they don't have to re-invent, and by having the users download the runtimes you don't have to distribute multi-megabyte executables.

Yeah, the dotNET runtimes are huge, and sometimes I wonder if they have to be that huge. But there's lots and lots and lots of functionality in there.
2503
General Software Discussion / Re: Versioning Systems, for Small Enterprise ...
« Last post by f0dder on March 09, 2010, 01:04 AM »
I use CS-RCS for my product development.  Since I am a single developer, I can use the free version.  Check it out.

http://www.component...re.com/products/rcs/
People still using RCS in 2010? Unbelievable :)
2504
General Software Discussion / Re: Linus Up Close and Personal [via Glyn Moody]
« Last post by f0dder on March 09, 2010, 01:03 AM »
ZP rocks :)
2505
Then Phillipe Kahn came along with Turbo Pascal and I'm pretty sure I paid $69 for it!!
-cranioscopical (March 08, 2010, 03:31 PM)
That sounds almost as if Philippe wrote it - whereas Turbo Pascal was actually written by the (Danish :)) [urlhttp://en.wikipedia.org/wiki/Anders_Hejlsberg]Anders Hejlsberg[/url] (yes, the guy who went on to create C#).

SPO actually optimized the intermediate code in TPU files
@f0dder - Amazing. Can you imagine somebody doing something like that today? They'd get their butts sued off before they could hit the save key on their compiler. How times have changed...
Oh, I don't really think so - you're working at the .obj level, not modifying the compiler. A friend of mine has a software protection product that works on the .obj level :)
2506
Yep, it is :)

It was a pretty interesting program, especially for it's time - there wasn't nearly as much stuff (freely) available on compiler optimization, and Borland's TPU library format wasn't officially documented... SPO actually optimized the intermediate code in TPU files (which would then be linked to .exes), it wasn't just a "unit full of optimized functions".
2507
I started programming using Turbo Pascal 6 - Borland's IDEs were lightyears ahead of the competition back then, and the compiler was pretty fast (faster than Borland C++). Also, a really cool feature was "build-to-memory", which was great during run-debug-modify development cycles, since harddrives were PIO and amazingly slow back then. Integrated help was also very nice. The generated code was, as usual with Pascal/Delphi, relatively sucky, though... so you used "sally's peephole optimizer", customized runtime, and often had to resort to writing assembly snippets when you needed speed.

I also used Borland Pascal 7 (note the turbo->borland rename after v6), which unfortunately dropped the compile-to-memory functionality. Did a bit of 16bit DPMI programming, but found it too limiting - this is why I ended up switching to C++, and frankly I never looked back.
2508
Living Room / Re: Help me build my new Home Theater PC
« Last post by f0dder on March 08, 2010, 02:03 PM »
You can go with Windows built-in support for RAID-1 on dynamic volumes - should work pretty well. Caveat is that it's only available in server editions unless you patch system files, desktop windows only offers striping >_<. Basically any drive you can get today is going to be pretty crazy fast compared to what was available a few years ago... I'd probably go for one of the WesterDigital 'green' series, and probably 5400rpm... going to be PLENTY fast for media use, uses less power, and afaik is also pretty silent.

Fast OS boots? Sure, a velociraptor or a SSD buys you a bit there, but imho the BIOS screens etc. take just as much time as the actual windows loading, so meh. Where a SSD really wins is on application startup time, but if you're mostly going to run media player, don't bother. Really, don't bother. You could even go for a non-mirrored partition on the data drives.

If you go for non-windows RAID, I've got good experience with Intel's RAID Matrix onboard stuff - it even does striping of reads on mirror volumes, giving very fast read speeds. Remember to always (re)enable RAID after flashing bios, though, or you can get some fscked up system state... especially if your system partition is also mirrored :)

Dunno about CPUs, but AMDs are cheaper - haven't looked at charts for a while, though, so I dunno if they're cheaper-per-performance as well, and you have to factor in the cost of motherboards as well. If you're going to do transcoding, you probably do want a quadcore chip.
2509
Developer's Corner / Re: Resources for learning git?
« Last post by f0dder on March 08, 2010, 01:55 PM »
Could I copy the repository from my linux server and do git-svn locally with a file:// reference?

Why not make the git clone of the svn repo on the server? Though git needs to be installed there of course. It's only a once off process though anyway, it'd probably finish in the time you'd spend looking for a faster solution ;)
I do have git there (as well as gitosis - seems pretty nice), but apparently not the git svn extensions... they don't seem to be in gentoo's portage, and I'm too lazy to figure out where to find them at the moment - other than that, yeah, it'd probably work ;). If it can do file:// urls, copying the repos to my workstation and doing it from here would work too... guess I'll end up testing that before too long :)

Anyway, I'm still interesting in hearing what experiences other people have, especially if using nonstandard folder structure (fSekrit initially did, version history became rather messy after I fixed it - this might be an opportunity to use git's history-editing capabilities). Also, "git tag -l" doesn't show any of my tags, but I can see tag references in TortoiseGit's interface... they seem to reference the "remote" location, though. So the conversion process indeed doesn't seem to be 100% automagic.

By the way, a completely unrelated question: would it be possible to "slice off" parts of a repository? Let's say, not so hypothetically, that once I reach version 1.5 I'd like to opensource fSekrit, but not provide source or history of anything older than 1.5... but I'd like to be able to keep the entire history myself, resulting in a setup somewhat similar to:
local repo: full version history
local server: full version history
public server: history from 1.5 and onwards
2510
Developer's Corner / Re: Resources for learning git?
« Last post by f0dder on March 08, 2010, 01:20 PM »
So, lemme re-use this thread, seems appropriate enough...

I've been playing around with GIT for a bit now, and feel relatively comfortable with it - the basics (enough for using it basically in a "mostly centralized" mindset) are definitely drop-dead simple, topic-branch workflow doesn't seem too complicated and definitely has value, and the more fancy stuff can be ignored until you find yourself needing it.

In fact, I'm comfortable enough with git that I'd like to conver my existing subversion repositories over. I've found a bunch of guides, and one using git-svn seems to be able to do the trick... however, I want to make sure I get my entire version history over, I want it "in a sane state" (ie, svn /tags should be git tags and not branches, there should be no references to the old svn repo when done, et cetera). I think the guide I found will ensure this (even if I might have to do some manual metadata pruning and tag pruning), but I'd like to hear if any other members have experience with this process?

Furthermore, the git-svn conversion process is slow, even across my gigabit LAN. fSekrit is a pretty moderately sized repository (91 commits, some 900(?) changesets, 1.4meg server-side repository). Could I copy the repository from my linux server and do git-svn locally with a file:// reference?

Also, are there any post-conversion tips? I realized that after the initial test conversion of fSekrit, the .git folder was pretty darn large - the actual filesize was a bit more than the server-side subversion repository (iirc), but the actual on-disk usage was much larger because of filesystem cluster size. After "git repack -a -d", I went down to a 423kb repository, 728kb on-disk --- smaller than the original subversion repository. If it really did grab the entire version history and relevant metadata, that's pretty nice (and consistant with what I've read elsewhere).

And gosh darn, the git network protocol is SO much faster than subversion's. In the time it takes me to commit ~10 file changes for fSekrit, I could push ~3.5meg/~500files of Notepad++ CommunityRelease via git... not to mention that regular non-pushing commits under git are of course instantaneous because they're done locally.
2511
General Software Discussion / Re: Linus Up Close and Personal [via Glyn Moody]
« Last post by f0dder on March 08, 2010, 12:45 PM »
Much as I admire (and owe) Linus Torvalds, I'm slowly coming to the realization that the more I get to "know" him, the less and less I seem to like him.
Amen to that.

He might get some software stuff right (the git core datastructures (though not the frontend, or how things were initially unix-style duct-taped together) is a good example - basing a kernel on minix in the 90'es definitely wasn't), and listening to him being a royal asshole can be fun for a limited amount of time... and a royal asshole does seem to be a pretty fitting description of him, adding in "overinflated ego" as well.

Perhaps he's nicer in IRL, especially if you don't write code or talk software politics.
2512
I hope anon will keep on banging the DRM servers :)
2513
Living Room / Re: Pirate vs. Paying Customer illustrated
« Last post by f0dder on March 08, 2010, 12:06 PM »
UbiSoft lied about DRM Servers - http://bit.ly/aKEbCg
Yup - and before anybody else posts anything about that event in this thread, take it here instead :)
2514
General Software Discussion / Re: alternative to filehamster?
« Last post by f0dder on March 08, 2010, 07:01 AM »
Something like that could very well be useful, but I see that as a relatively specialized application, rather than automatically monitoring for changes and automatically committing?

For document authoring, a system with GIT integration could actually be pretty cool - but I'd still want the commits (and pushes) to be controlled rather than automated.
2515
Living Room / Re: New Generation Game Demos
« Last post by f0dder on March 08, 2010, 02:47 AM »
If the demo is already nearly full version, then the fullversion will be released next day or something?
There's a lot of ways to go about this. If the game isn't entirely done, then "all the available content" could be preloaded, requiring perhaps just a few hundred megabytes being downloaded once the game is entirely done.

Or it could be that the demo is released at the same time as the full retail, just as a way to evaluate whether you want to play the game.
2516
Nuts & Bolts had the best defragger I have ever used. It was lightning fast and quite thorough, unless you chose the option to 0 the free space, which then it was just thorough. I really wish they would have continued the development of that. I really miss it.
Amen to that! - it did some pretty hefty pre-planning, which enabled it to be very smart about how it moved data around... it defragged properly, and it did it fast. Dunno if it's possible to do it this well today; will probably require an offline filesystem defrag at any rate.

Faked multitasking from back in the days of DOS. "When men were real men; women were real women; and those little fuzzy creatures from Sirius were real little fuzzy creatures from Sirius."
...and 16-year-old girls weren't FBI agents.

A couple of my favorite old pieces of hardware have to be the Commodore C=64:
c64.jpg
This is where I whetted my teth on Maniac Mansion, Zak McKracken, Commando, Last Ninja and whatnot. From the times where I didn't even know you could buy games, since none of the stores around where I lived had any.

And the Commodore Amiga 600:
commodore_amiga600_front_1.jpg
Which was a real gaming beast, with great graphics and sound. Cannon Fodder, Gods, superfrog, prince of persia, persian gulf inferno :-*, Another World (aka. Out Of This World), flashback, Hunter and a LOT more. Lots of fond memories!
2517
Living Room / Re: New Generation Game Demos
« Last post by f0dder on March 08, 2010, 01:49 AM »
Sounds pretty lame to me - it will severely distort the gameplay, you won't get a proper feeling for how it really works. And the method would only work for relatively open-ended games, not for linear progression. And why have people download a full game (some are up to 16 gigabytes now!) if you can get away with a couple gigabytes of demo? (yeah, OK, grabbing the full thing means "unlocking" will be faster, but meh).

Isn't going to mean much for pirates, the already release retail game versions before it hit the stores.
2518
General Software Discussion / Re: alternative to filehamster?
« Last post by f0dder on March 08, 2010, 01:43 AM »
Hm, automated git committing?

Doesn't sound like a super good idea to me. Either it has to simply watch for changes, and commit when that happens - which could check in files that are only partially updated. Otherwise, you need integration with the specific tools used, which you likely won't be able to get everywhere you want. And even that probably won't work out too well if you have the habit of saving often (I hit ctrl+s pretty much after every sentence when editing text).
2519
Living Room / Re: Pirate vs. Paying Customer illustrated
« Last post by f0dder on March 08, 2010, 01:40 AM »
Lawl, just think about checking your ping when playing Assassin`s Creed - so you wont lag out and quit without saving your game to the server :D
Saves are still stored locally - you can put them "in the cloud" as well, but saves being cloud-only i FUD.

That still doesn't change my stance on protections like that being inherently evil & only hurting legitimate customers.
2520
The only problem is that neither of them will run on 64-bit Win7 no matter what you do. Might just be some sloppy coding since they work just fine under 32-bit Win7.
In which way do the apps fail on 64bit Windows?

Mayhaps it's a registry key issue - if you export keys from 32bit Windows and import directly on 64bit with a .reg file, the values will go in normal HKLM, instead of the Wow6432Node subkey where you want them...
2521
Living Room / Re: Antivirus companies support virus writers?
« Last post by f0dder on March 04, 2010, 06:07 PM »
How do you propose to secure things with CA?
I'm not proposing to "secure things with CA" - but SSL certs (and code signing certs) need the CA system unless you want to rely on self-signed certs (and how do you verify the validity of those, then?).

In case of SSL certificates, you know, there's no bigass warning for real spoof site. The only warning you have for real spoof is lack of tiny yellow lock icon.
Which is enough for power users (the ones that be keeping their software up todate, unlike regular users). Authentication isn't the only thing SSL does, though, confidentality and tamper-resistance are just as important.

The bigass warning is mostly shown to customers of legitimate businesses whom forgot to pay racket money (forgot to renew certificate).
And I do believe this is a problem. SSL certs and code signing certs are a bit on the expensive side. Code signing certs are slightly difficult to obtain, but that's mostly a positive thing, though.
2522
First, I'd try simply "extracting" the installed critical software. you say it's dotNET, it's probably going to be pretty easy, without too much registry/filesystem entangling. If that works, simply do a clean Win7 install on her hardware, problem solved :)

I'd probably still do a P2V of the existing XP install first, just to be on the safe side... sometimes it takes a couple of months before you realize that you forgot backing up or extracting whatever single item.
2523
Living Room / Re: Antivirus companies support virus writers?
« Last post by f0dder on March 04, 2010, 05:34 AM »
Use firefox, keep it up to date, its usually fixed for exploits sooner than any use of exploit appears in the wild (which is also sooner than antivirus responds). Geez.
The browser is only one part of the exploit vector equation - you're forgetting flash and java, which aren't always fixed in a timely fashion.

Fixing security holes with a third party code blacklist for known uses of that security hole in the wild, that's just wrong. It's like you have a  digital lock on your door, with password code, and instead of updating lock's firmware you also install second lock that has camera that blocks entry for people whom look like known criminals.
A decent anti-malware product wouldn't just be blacklisting static code sequences, though, so this comparison doesn't really work. A better one would be a cop stopping a guy pulling a gun before he pulls the trigger.

Ditto by the way for digital certificates and 'certificate authorities'. Extortion scheme, pure and simple, not very effective for protection because it is possible to steal certificate, but extremely effective for having various people make billions by doing very little. Everyone who doesn't pay up is subject to plain libel delivered when user tries to run the application* The libel also devalues genuinely useful warnings.
Unfortunately there's too many CAs and some have been way too lax on security... but how do you propose to secure things without a CA?
2524
T-Clock / Re: T-Clock 2010 (Maybe...)
« Last post by f0dder on March 03, 2010, 09:39 PM »
As I'm sure you know, it's quite common in C to store pointers in int variables.  While this might not be a good practice in general there are times when it makes sense, such as for SetWindowLongPtr() where all the API is interested in is storing a small blob of data - it doesn't care what that data is (the user of the API gets to decide what the data means).
A void* is a better choice - it's opaque ("don't go around manipulating this like you woudl an int"), you don't run into trouble on platforms where sizeof(int) != sizeof(T*) (16- and 64-bit x86 platforms), et cetera. 8 of the 10 documented indices for Get/SetWindowLongPtr are pointer types.

On the other hand, Microsoft should have made it so that passing a LONG_PTR type (whether by casting or not) to SetWindowLongPtr() should not have caused a warning whether the build was for 32 or 64 bit. That's really where something fell down here.  The whole point of a type like LONG_PTR is so that the conditional code that Stoic Joker ended up using wouldn't be necessary.
And, at least on VC2008, there is no warning when doing the single cast. Can't see why there would be with other versions of the compiler, either, really - but mistakingly using SetWindowLong (instead of SetWindowLongPtr) at 4am because you haven't had enough coffee... then yeah :) (or perhaps some flaky old PlatformSDK version - but I kinda doubt that as well).

The PSDK is a big effing mess with some of the worst C coding styles, it's a shame MS has never cleaned it up. Stuff like using enums instead of #defines, inline functions for the ApiNameA/W distinguishing, etc.
2525
Heh wtf, boxer hasn't had wordwrap until now? O_o
Pages: prev1 ... 96 97 98 99 100 [101] 102 103 104 105 106 ... 364next