CoffeeFiend

This page has all the Windows memory limits you may be interested into: http://msdn.microsoft.com/en-us/library/aa366778.aspx The 64 bit version of XP can use 128GB, as the previously linked page shows. For all practical desktop use (no specialized hardware, etc), 32 bit OS'es are limited to 4GB of address space, some of which is used by your video card(s), and some more by other hardware. So you'll always have less than 4GB available. See Mark's Russinovich's "Pushing the Limits of Windows: Physical Memory" blog entry for more details if you're interested.

I'd try deleting C:\Users\Your_User_Name_Here\AppData\Roaming\Microsoft\MMC\taskschd first (or just rename it). It could be a COM+ permission problem i.e. a registry (or file) ACL problem (seen that before), but that's kind of tricky to guide someone else into fixing Especially when you can't just guess what exactly is failing. Personally, I'd try running Process Monitor to find out what's going on when you load it (maybe you can zip up a capture of it, upload it on rapidshare or whatever, and post a link to that here?)

The issue you're having might also be due to a ODBC driver that isn't available on the x64 platform (e.g. Microsoft.Jet.OLEDB.4.0, for accessing older MS Access .mdb files) Not many options in that case: -not using Windows x64 for now (not a long term solution for sure) -moving to another database that has a driver (newer MS Access 2007 format, SQL Server Express, etc)

You have a bittorrent client running (and a port is open for it in your D-Link router obviously). There's lots of computers sending UDP packets to you, all coming in on your port 31848 (try rule "udp && !dns" to see it better), and their contents is definitely bittorrent DHT traffic (we can even see the SHA-1 hash of the torrent itself, one could even tell what you're downloading from that). The DNS requests made are made for everyone connecting on your port 31848 (probably the BT client itself, so it can show you the reverse DNS address instead of just IP)

What a completely nonsensical statement.It makes perfect sense. Back when Photoshop 2.5 was out (on 4 floppies -- back when we were running win 3.11), people had like single-digit memory sizes, and the OS used pretty much all of it. Opening any moderately sized image used all the resources you had pretty much. Back when Photoshop 4 was out (Win95 era), we still had very little RAM (most people having 16MB or less), and the OS used all of that too. Opening anything still used all your resources. Back when Photoshop 7 (CS) was out (early win XP era), we still had around 128MB of memory, and XP's core uses more than 128MB, yet again leaving you with hardly any resources. Opening anything even remotely big (high resolution scans, drum scans, etc) or doing anything more than than the very basics still quickly brought your machine to its knees. Nowadays, Photoshop CS3 opens on ~90MB of RAM which basically is nothing at all for any modern computer. You can also get 4GB of 800MHz DDR2 for as little as $70. I you have 4GB, and have Vista running + Photoshop CS3 open (nothing else, like in other scenarios), you still have over 3.5GB of RAM left to work with, without making use of any virtual memory (or 1.5GB on a box with 2GB) In the previous scenarios, you had a negative amount of RAM left after the OS loaded (anything you could use, was due to memory paging i.e. virtual memory), whereas now you have entire gigabytes left on commodity PCs! Even opening a 12 megapixel RAW file and adding a couple extra layers only uses an extra ~1% of my free RAM. Even an old box running XP with 512M of RAM could handle it. And processor wise, even a several years-old P4 will suffice for over 90% of tasks. In all those gigabytes of RAM free left, not only you can load a LOT of images at once, and do complex things with them, but you can actually multitask! Something I wouldn't even have attempted before (it would have been excruciatingly slow). I very often have Photoshop open at the same time than other apps that use hundreds of MBs of RAM and it works great. Long story short, save for a couple exceptional cases/uses, Photoshop isn't one of those monster processes that brings any computer to its knees anymore. It was a perfectly good point. I didn't think anyone wouldn't get it.

LOL. Good joke. I almost thought you were serious for a half second. I checked the date, nope. Not April 1st.

I have the old Universalis 10 (that's already 3 versions out of date) working fine under Vista SP1. They even have a page about it on their site here. For the camera, it's unfortunate, but it's a 2 megapixel point & shoot from 1999 (9 years old is a LONG time in the world of digital cameras), so it's understandable that at some point they cut the support... A $10 card reader works fine here indeed, or one could get a 6MP+ camera that gives much more detailed photos, has a better zoom range, works much faster, has a far better LCD and better/more modern menus/better processing features, more advanced functions (like histograms), uses more modern card types, can use FAT32 formatted cards too, has USB2, often records quicktime movies and such, and quite often nowadays even has an orientation sensor (no need to manually rotate your photos anymore), etc -- all for $100-ish. Might be time for an upgrade really. @geek: about ModifyPE, I'm not quite sure who to blame, as it checks arguments in a non-standard way (it doesn't even work if you type modifype, the program name itself is case sensitive too! it has to be ModifyPE)

Intel's naming has long been screwed up indeed. Celerons meaning anything from a P2-era chip to a number of modern core-based ones (some mobile, one now even dual core). It has essentially no meaning besides "budget". Same with the name Xeon -- anything from a P2 era CPU to an expensive, cutting-edge 45nm quad core CPU, it just means "for servers" basically. And even old familiar names get used in strange ways sometimes, like "Pentium Dual-Core", which is is actually Core-based (so not a Pentium at all). And naming processors by numbers that don't really reflect performance in any way... It doesn't make buying processors real simple (we're forced to look for benchmarks to really compare anything) AMD's naming scheme was far better (until the Phenom). It wasn't perfect, but you had a pretty good idea of what performance you were getting without looking at benchmarks all evening.

So you registered to point out something mostly irrelevant in a 6 month old reply to a question that's over a year old? ... Edit: If you had a problem, you could have actually mentioned it if you wanted help (e.g. "Hey guys, I have the same problem, and I tried x, y and z, any clues?" instead of just "it don't work") or some kind of useful answer (instead of being pointed out for digging up old threads for no apparent reason). But help you when you got that attitude? Yeah right. No point in repeating I_Broke_My_MHZ & cluberti's point either...

Sure, you can plug them speakers on it. The onboard sound card is overkill for that speaker set really. Don't expect great 5.1 sound from a $60 set of speakers... Like I said before, a decent, quality 5.1 sound system doesn't go for that much. Most of what you're paying is for retail sales profit, transport, the fancy box with full color printing, packaging, publicity/marketing, the factory's costs to assemble them, etc. That leaves around 6$ for the amp, 4$ for the sub, and and about 1$ per satellite speaker or so (the main part of that $ going towards the flimsy plastic enclosures). I've never seen even just a subwoofer that sounded good at anywhere near that price. 6 ghetto speakers don't really sound better than 3 better quality ones. The sub is an anemic 5" driver (doesn't move much air, doesn't go much below 50Hz), the overall response curve is likely nowhere near flat, and while it probably sounds plenty loud, the specs aren't exactly looking great (I mean, who rates their stuff at 10% THD?) I mean, they probably sound alright -- for budget speakers.

Microsoft SQL Server 2008 went RTM! Plenty of news around around the web. Google news has loads more as always. Product home page is here. The 180 day trial is already available. Seemingly the free Express Edition isn't online yet Edit: MS' site says: "SQL Server 2008 Express will be available for download at the end of August 2008. Until then, try the 180-day SQL Server 2008 Evaluation." Oh well.

And that's hardly a Windows-only thing (much less a Vista specific one). A quick peek a Ubuntu's bug tracker (just one distro) shows there's 46967 bugs currently open...

It's not exactly a fast server indeed (P4, single core, only 2GB -- a basic $350 Dell Vostro 200 desktop is faster), so I'd say 2003 too. It would be worthwhile to bump up the RAM for sure. But since there are so few clients, and that the box seemingly has to handle everything, it might be worth looking at Win 2003 Small Business Edition. You'll also get Exchange out of it, and even SQL Server if you go for the premium version although there's only so much that this box will run at once (and SQL Server Express Ed might suffice to your needs).

Yeah. That's pretty minor too. Even XP still has some glitches like that left after 7 years (I've seen some boxes where trying to create a new shortcut fails about 50% of the time). As long as they get around to fixing the more serious issues (like the status bar), I'm happy. And again, tested on many boxes or not, some of those I just can't reproduce. And talking about those, the Taskbar Bug" specifically (when you click where there's no icons), I noticed I get the "safely remove hardware" dialog or the volume control, depending on where in the empty space I double click.

No. your web browser might connect to port 80 on the destination box (web server) to get a web page, but that changes nothing at all. Windows (nor your web browser) itself wouldn't "open" (accept connections) on port 80 regardless. And if you're going thru NAT, port 80 should not be open either. In either case, the reply won't be on port 80 but rather a port number higher than 1024. In other words, when your computer send a SYN, it does it to destination port 80, but with a different source port (let's say 21075 -- it's as good as any other number really), and then you reuse those: the server sends it's SYN/ACK from port 80 to your port 21705, then your computer sends ACK (still src port 21075, dst 80). Now that the TCP handshake is done, your computer makes the HTTP GET or POST request itself (same ports yet again), and the answer from the web server (e.g. HTTP/1.0 200 OK) is just like its SYN/ACK packet (src 80, dst 21075). Your source port 80 isn't involved at any point in the process. There's no reason to have ports 21 ad 80 open, unless you run a ftp & web server.

An option so it will ignore quotes and output *everything* in hex? Or an option to use a different character to delimit what is to be converted? As for the zeros (everything being unicode), that could be optional if you wanted (I thought that's what you wanted in the first place). Simple to do anyways. Just let me know exactly what you want, and I'll make the changes.

Does that particular text file have a specific file name? Or does it change from one zip file from another?

Sure. But I'd need more infos first. I do PHP only as a last resort If you want someone to make an app or script something for you then we need more detail like I said. Things like: that text file inside the zip, that we're supposed to generate the zip filename from, what is it called? And things like what to do when the said text file is quite big, and that the generated filename would have to be like 5000 characters long? I suppose we could also drop all invalid characters... The idea is to unzip the said textfile, either using a command line util (for scripts) or a suitable library (e.g. SharpZipLib for a C# app), look in some text file (gotta know which one), extract the words (using a regular expression would work fine), generating the new name, renaming the zip file and deleting the temp file (unless it was unzipped to memory) It's not a 5 minute job, but it's still easy to do, as long as we know precisely what it has to do. The more details, the better.

I don't think you'll find ready-made apps to do something so unusual & specific. It could be done though (a library would do all the heavy lifting -- unzipping files)

Sorry for the late reply, had a nap, and it was a LOT longer than expected Try this app: http://www.zshare.net/download/1665788387ce49d0/ Syntax is: text2hex.exe sourcefile.txt destinationfile.txt a input text file with: "helloworld" this is a test "h l o r d" (&$^ " t esti ng" "helloworld" "h l o r d" in it, would create another text file with: 680065006C006C006F0077006F0072006C006400 this is a test 680020006C0020006F0020002000720020006400 (&$^ 200074002000650073007400690020006E006700 680065006C006C006F0077006F0072006C006400 680020006C0020006F0020002000720020006400 Which is exactly what you wanted if I understood what you wanted. Everything is proper unicode, and only the stuff between quotes is turned into hex. If you want something else than a command line app, it would also be trivial to make it into an app with a GUI where you can paste strings into and have it convert them (or even a web app)... Anything else just ask. It could be turned into a jscript too. I'll post the final source code when you're happy with it...

I hadn't chimed in yet, as I don't really care so much what brand I get (they mostly all work fine, and they're all dirt cheap), but samsung wise... I'd definitely skip on that. I just got rid of one. It didn't play nice at all (AHCI issues). I'd definitely get a Pioneer, NEC or LiteOn over one of those again... Just my 2 cents.

As Mr Snrub put it, no need to lose sleep over those. That's for ftp and web servers. If you're not running any, I'd look at the router's config (port forwarding specifically, as well as UPnP), and see what IP they're forwarded to (you very well could be in the DMZ too as Mr Snrub said, it would explain why so much stuff open/closed, and you definitely don't want to be in the DMZ) If you have a properly configured router that does NAT, you don't even need a firewall on your PCs. Everything port you didn't willingly open should be stealth then (the router wouldn't even know what to do with that traffic, so it just drops it)

That would be because it only does unicode searches then. So that's what I was thinking, it's not padding, it's just encoded this way. Arguably, one could say the real fix here is a hex editor that isn't limited in that way (perhaps I'm not getting the big picture, converting text to hex, to then do text searches with a hex editor?), but anyhow... As for the searching between quotes, the simplest thing would be to use a regular expression for that (e.g. "[^"]*"), and replacing in the original string all the matches by their hex representation (of the byte array returned when UTF-16 encoding it). Still simple to do. I'll try to find some time tonight to write you something simple in C# (should only take 5-10 mins really). You could still do it in JScript too. The biggest issue I can think of, is that .toString(16) doesn't return zero padded numbers (easily solved too).

Print screen, start > run > mspaint [enter], save as whatever.jpg, go to www.imageshack.us, browse, select that file, upload, and use the links it gives you (there's an insert image button). Word isn't an image format.

Yuck. I'd be willing to pay MORE so it DOESN'T come with Lotus Notes. Seriously though, don't see this making making much of a difference at all.