nmX.Memnoch

No they aren't. NVIDIA is allowing the card makers to play around with clock speeds now. The base is 600MHz core / 1800MHz memory. If you do some research you'll see that they all have versions with both the base speed and versions with higher clock speeds. EVGA has the highest one right now with 700MHz core / 2000MHz memory. A 100MHz core clock and 200MHz (effective DDR) memory difference is a rather large gain. I'd say that probably pushes the card over GTS and pretty close to GTX speeds in realworld/benchmark usage. The problem is that they're back ordered at most of the popular places. http://www.newegg.com/Product/ProductList....=GeForce+8800GT

Yeah, the BACK-UPS line doesn't have near the manageability of the SmartUPS line. But at the same time the SmartUPS line is a lot more expensive. Try the PowerChute Business Edition and see if that works. It's free up to 5 nodes. Install all three components on the main machine (your server I assume) and then you can install the console on your workstation to manage it remotely. At any rate, you should still be able to get it to work using the built in Windows stuff. Also, don't plug your speakers into the UPS. You don't need sound for the PC to shutdown. Not to mention that if you have a 2.1 or 5.1 system, the sub could have adverse effects on the battery life (because the sub will suck more juice anytime there's a bass hit). This is the same reason that you should never plug a laser printer into a UPS.

No, the one labeled (System) under the Status column is the partition that contains the boot files. Are those all partitions on a single drive? If so, you're seriously hurting the performance of your system, not to mention making the drive work harder which will shorten it's lifespan. Backup your data and dump the partition structure. Just make it one large partition and use folders. It's beyond me why people think they can manage their data better using a separate drive letter than they can just using a folder.

There's your answer. It's pretty easy to setup, but you do have to install like three different components on the computer that has the actual (data) connection to the UPS. You can have it initiate a custom script (or utility) that would shutdown any other computer(s) plugged into the UPS. Of course, you would also have to have the switch plugged into the UPS (which you already plan on doing) so the shutdown command can be sent across the network. No sense in sending a shutdown command if the network ain't there. All you need is the following command in your .cmd file (this is for the XP version, the 2003 version requires a few more options): shutdown -m \\COMPUTERNAME -s -t 00 -f -m specifies the remote computer -s is for shutdown -t is the wait time -f forces applications to close BTW, you can also do this with the built in Windows XP and Windows Server 2003 UPS tools. In other words, the third party APC utility isn't strictly necessary. It does, however, monitor other settings about the UPS as well as providing for email notification. See above...APC's PowerChute Business Edition has this built in.Email notification doesn't do much good though if it's a neighborhood outage and the local ISP switch/router is down.

Still, from a performance stand point you still want a larger drive. Larger drives hold more data per platter (i.e. they have a greater aerial density). The higher the aerial density, the faster the drive performs and if you aren't putting a lot of files on the drive, then all of the files can be stored on the fastest part of the drive...giving you the best performance possible. Something doesn't sound right about that. For starters, there's no such thing as a "site license" for Windows...of any version. Second, you can't purchase an OEM license against a volume license. You can, however, purchase an OEM license when purchasing either a complete new pre-built computer (Dell, HP, etc) or if you're purchasing a qualifying part (motherboard). It is possible that you may have an employee purchase program where you work, which would allow you to purchase Microsoft software at discounted prices.Last, as to the SUS issue, you can fix that on your personal computer using GPEDIT.MSC. Just go in and set all of the Automatic Updates options to "Not Configured". Here are my hardware recommendations. I'm sure others will recommend some other choices as well. Note that I left off a case recommendation. Motherboard: (pick one depending on your budget) 1) Budget Choice: Asus P5K-E LGA775 P35 based $144.99 + $6.61 shipping 2) Enthusiast Choice: Asus P5E LGA775 X38 based Supports PCI Express 2.0 $234.99 + $7.14 shipping Processor: Intel Core 2 Quad Q6600 2.4GHz 2x4MB L2 Cache $284.99 RAM: Corsair XMS2 1066 PC8500 2GB Kit (2x1GB) $149.00 - $30 MIR = $119.00 Video Card: (get this single card in lieue of a low end CrossFire setup) XFX GeForce 8800GT based card This card is PCIe 2.0 capable so you'd get the full benefit on the X38 based motherboard. However, it will work on PCIe 1.1 motherboards. $269.99 + $6.33 shippping - BACKORDERED Hard Drive - Operating System: Seagate 7200.10 250GB SATA 3Gbps 16MB Cache $69.99 Hard Drive - Storage: Seagate 7200.10 500GB SATA 3Gbps 16MB Cache $119.99

Thanks for the info. I'm wondering though, if all else fails, would I be able to use a USB->Serial adaptor like this one. to create the interface? It looks like I might be able to plug that adaptor into the interface expander, and then just deal with USB cables from there. Heh...well, I've already tried that with the SmartUPS 700 I have on my machine. It didn't work but that could've been the specific USB->Serial adapter I was using. It wanted a driver and I'm not sure the driver ever installed right. I could get Windows built in UPS management to see the UPS for a split second and then it would lose connection. APC's management utility never even saw the UPS. I don't think so. I think it's "either/or, but not both".

If you get an APC SmartUPS based UPS (1000VA or higher...you'll want that much anyway for two systems to give them enough runtime to run any custom shutdown processes and then finish shutting down) you can also get an interface expander that will allow you to attach more than one machine. http://www.apc.com/resource/include/techsp...se_sku=AP9607CB If that doesn't come up (because of their country selection thing), the part number is AP9607CB. Of course, that does you no good if the system doesn't have serial ports (like the Asus P5B and P5K line. And some of the newer UPSes are coming with USB only connections. I'll look a bit more and see what I can find...one would think APC would have a USB interface expander. I have an APC SmartUPS 1400 that I use for my file server. It also has my cable/VOIP modem, router and switch plugged into it. It sits at about 15-20% load (if that) most of the time. That gives it PLENTY of runtime (especially if it's a short power outtage...I don't even have to worry about it shutting down).

Yeah...I didn't notice how old the original thread was until after I'd already replied. I could've edited the post but it was already too late anyway...

How would I go about restricting the services within group policy? Your right I am, this would prevent them from doing it again after I have the service started again. The best way to stop them from messing around with services? Don't make them admins! You can't restrict services with Group Policy until they're in an AD domain. Once you do that then you just create a Group Policy Object (GPO) that has the settings you want for the services. Then you apply that GPO to the OU that the computer accounts are in. Unfortunately, you can't Start a service if it's set to Disabled so the state has to be changed first. That script will do exactly what he wants as far as changing the state of the service. A mixture of that script and the one that Yzöwl posted could be used to both change the state to Automatic and then Start the service.

So what exactly are you needing in the way of parts? We know you need the motherboard and video card and I gather that you are also looking for a CPU (?), but what else do you need? RAM, SATA hard drive(s), power supply (probably so with the newer motherboard and video card), DVD burner(s), sound card, etc, etc. If you need RAM are you looking for something with DDR2 or DDR3? DDR3 is ramping up really fast, but it's also more expensive. Of course, the X38 boards are also going to be expensive. Speaking of, give us a rough estimate on how much you're willing to spend...that actually will go a long way towards dictating what we recommend. From what I gather, what you really care about is the Intel CPU and chipset. You can get some really good board from other manufacturers like Asus and Gigabyte. I don't typically recommend boards from MSI, DFI and abit, but some people have good luck with them.

Ok, you just made up my mind about something I've been thinking about for a while now. I do have a server with just such a configuration...as a matter of fact it has a 600GB array, a 1.7TB external array and two 1.8TB external arrays. The three external arrays are used as backup storage for all of my other servers...and routinely sit with about 1.5-1.6TB of data on each of them. And yes, the server is running 32-bit Server 2003 (won't support 2003 x64). I do have another server that's running Server 2003 R2 x64 that I've been thinking about moving the three external arrays to. It looks like my decision has been made...I'll be moving them next weekend.

I have to agree with puntoMX about the motherboard...don't get anything that's 975X based...or even 965 based (I know it's confusing but the 965 is actually newer than the 975X). Go with an X38 setup, or if you can wait a bit Intel has brought the release of the X48 forward by a couple of months. Rumor is that it will either be released later this month or sometime next month. As for the motherboard manufacturer, if you aren't going to overclock I don't see a problem going with an Intel manufactured motherboard. They're definitely stable and fast. You'll have a little bit of overclocking options with the newer Intel motherboards, but not near to the level of tweakability from something from Asus or Gigabyte. On to the video card...I'd have to recommend that you look at an 8800GT based card as well. NVIDIA just released that GPU and it's numbers are very promising. No, you won't be able to do SLI on an Intel chipset, but I really don't think that will matter much anyway once you see some of the numbers (both price and performance) for the 8800GT based cards.

Given the drives that you already have on hand, that's exactly the way I would set them up. Just keep in mind that if you lose one of the drives in the RAID0 array that you lose all the data. This shouldn't be that much of a problem since the main array is generally used just for the OS and apps. RAID0 will give you the performance you're looking for on your OS/apps drive. The only thing you'll need to worry about really backing up from your OS array is your browser favorites/bookmarks and any saved games (provided you play games). When you're done you should have a C: drive (the RAID0 array) and a D: drive (the RAID1 array) and nothing else. Don't partition the drives at all. Doing so will definitely hurt performance (as well as being harder to manage IMO). To be honest with you, my main workstation is configured like this: C: drive (OS/apps) -- 2 x 250GB RAID0 D: drive (data/downloads) -- 2 x 400GB RAID0 However, I have a file server with a RAID5 array (4x400GB) dedicated to storing my important data so I'm not really concerned with losing anything on my primary machine. Yeah, it would suck, but all of my important stuff is backed up or saved to another computer that has drive redundancy. Yeah, we all completely skipped over that question...not on purpose, we all just forgot to answer it. As you've already figured out, you have to have the RAID enabled in the BIOS and your arrays configured before you install Windows. Yes, you can do any mixture of array types that the controller and your drive configurations (size/type of drives) allows. There won't be any performance hit at all with mixing array types. RAID0 will give you both good read and write speeds which RAID1 will give you good read speeds with only a minimal (not noticeable) write speed hit. As I mentioned above, if you lose a drive in the RAID0 array you'll have to replace that drive, recreate* the array and reinstall. On the RAID1 array all you'd have to do is replace the failed drive, then the controller should automatically intiate a rebuild (if not, you can manually kick off a rebuild using either the controller BIOS or by using the Intel Matrix Storage Manager application in Windows).*Note the use of the term "recreate". In RAID array speak, the term "rebuild" is used for redundant arrays when you're simply replacing a failed drive. The redundancy is then "rebuilt" on that array. "Recreate" means you have to recreate the array from scratch and replace all of the files manually (because it's not a redundant array). Mount points are something done within Windows. They can be done with single drives or RAID arrays. Basically, instead of giving the drive a drive letter, you tell Windows to create a mount point on another drive letter. For instance, if I had a 500GB drive that was my D: drive and then I added another 300GB drive to the system but didn't want to have another drive letter, I could tell Windows (in Disk Management) to mount the drive at D:\Downloads. Anything saved to D:\Downloads would actually be saved to the new 300GB drive instead of on the 500GB drive.For your situation I really don't think it's anything you need to worry about.

Only if it's a multifunction device that uses domain authentication (i.e. LDAP lookup) http://www.microsoft.com/WindowsServer2003.../licensing.mspx

He can get the RAID/AHCI driver as well as the chipset driver from here. Note that I selected "Windows XP", but those drivers are also good for XP x64, Vista and Vista x64. The only exception is the driver floppy for x64 Windows. That can be found here. I don't like doing partitions either. I find that directories provide plenty of organization for my needs. Too many partitions can actually hurt performance anyway. A single partition can be defrag'ed putting the files on the fastest part of the disk. Partitions split the drive up so the read/write heads will have to travel more if you're writing/reading data on multiple partitions at the same time. I have to recommend the same. Leave the page file alone...just let Windows handle it. Windows XP does a great job of handling the page file on it's own without mucking around with it. I can make one, and only one, case for partitions. We used to partition drives at work into 40GB C: drives and D: was whatever was left (depending on drive size). We'd use the D: partition to store the Outlook PST's, My Docs and other data files. The reason we did this was if the OS got fragged for whatever reason we didn't have to worry about backing up too much data...just reinstall the OS, join the machine back to the domain and give it back to the user. We've gotten away from doing that though since XP is pretty stable...not to many worries about an OS install getting fragged. Plus we have a GigE network in the office now so backups don't take that long anymore. EDIT: Corrected "Leave the partition alone" to "Leave the page file alone"...dunno what I was thinking.

It's definitely worth it if you can afford it. And SATA drives being so cheap these days...it's hard to argue against it.

Yeah, but you'll need them on a CD/flash drive (or secondary hard drive) BEFORE you reinstall. Without the INF Update Utility none of your PCI/AGP devices will work to their fullest potential (sometimes the NIC won't work at all). You also won't get full USB 2.0 support without it. If you're doing an unattended disk you can have the drivers installed during setup.

The 169.02 betas were released on the 29th...but the 169.04's were released on the 30th. B) It also turns out that there was a 169.02 WHQL released...but only for the new 8800GT, and it was just released today.

I do know. It'll be horrible compared to a controller with a dedicated RAID processor. There are redundancy calculations that have to be done for RAID5 writes. Without a dedicated RAID processor, it'll use the system's CPU. If you're doing anything CPU intensive (playing a game, video editing, etc, etc) then the hard drive write performance will be terrible. This is why I recommended the 3ware controller for your file server. @kanataguy There's some pretty in-depth information in this thread. Most of the information there is probably more than you need know...but it's a good read anyway. For your situation I would recommend using a combination. Set two of the drives up in RAID1 and use the other drive as a stand-alone (and possibly as a backup alternative to the RAID1 set). Also, there shouldn't be any write slow downs with RAID1. A lot of people say this, but there aren't. The "slow down" most people mention is the difference between RAID0 writes and RAID1 writes (RAID1 offers no write speed benefits over a single drive, but with a good controller there's no slow down either). If you REALLY want the best mix of performance and redundancy, get yourself a fourth 500GB drive and setup RAID10. You can (potentially) lose two drives without losing any data. See the thread I mentioned above for more information on RAID10. Zxian is right about splitting up I/O as much as possible. Hard drives are the single slowest device in your computer so splitting up write and read I/O across as many drives as you can will help increase performance. Oh yeah...WELCOME!!

It has two SATA headers on board. They don't list a driver because you probably don't need it. With the early Intel SATA chipsets you didn't need the extra driver unless you were running in RAID mode. I recommend grabbing the latest INF Update Utility (aka Chipset Software Installation Utility). If you have onboard video you can grab the latest video driver from that same link.

hehe...not that I really meant to. I just thought it important to emphasize that he has a P4-M and not a P-M. You should be ok with one of those but we're not going to make any guarantees. It all depends on what support Acer put in the BIOS. Usually though, if it supports one C1 processor it'll support all of them (at the same bus speed). Honestly, I really don't see a reason why it wouldn't support the D1 revisions either. How do you know this? Are you referring to my BIOS. If you had the Pentium M like he originally thought, then he would've been correct. However, since you have a Pentium 4-M then you should have some upgrade options available. Again, we're not giving you any guarantees. To be completely honest, if you can even find a P4-M you'll probably spend more on the CPU than the laptop is worth.If you really want to give it a decent upgrade, give it some more RAM (assuming it has the default 256MB in it now, you really should get some more even if you get the CPU upgraded). You can get an additional boost by putting a 7200RPM hard drive in it. But, if you use it on battery a lot be warned that a 7200RPM drive will run the battery down faster. Unfortunately, I couldn't find any information on Acer's site that would provide a way to tell. However, I didn't look for very long though.

Personally, I tend to use /qb for any MSI install anyway so I can see if the process hung or not. The only difference between /qn and /qb is that the latter shows you a progress bar. @wigadore Adobe has a utility on their site called the InstallShield Tuner 7.0 for Adobe Acrobat. You can use it to customize the Adobe Reader 7.x install so that it's completely silent without any switches (you do have to use the decompressed files though). You can even use it to configure certain options (auto-accept EULA, don't display first run options, etc). If you move to Adobe Reader 8.x, the new utility is called the Adobe Customization Wizard 8. I'm actually curious why you're not using Adobe Reader 8.1.1. For that matter, I'm also curious why you're using an older version of JRE. Also, on your SAV install, I would install it unmanaged. You can migrate it to a managed state by copying the GRC.DAT file from \\SERVERNAME\VPHOME\ into %ALLUSERSPROFILE%\Application Data\Symantec\Symantec AntiVirus Corporate Edition\7.5\. SAV watches for this file every so often (or you can force it to read it by restarting the service or rebooting the workstation). If/when it finds it then it reads the settings from it and adjust the client as necessary. That will make it fully managed with whatever settings you specify (including the uninstall password) using the Symantec System Center console. You can also use the MSI to install it instead of setup.exe...it'll give you a bit more control over the initial settings. Here are the settings I use to install SAV 10.1.6.6000. I'm pretty sure they work for SAV 9.x as well: "<PATH>\Symantec AntiVirus.msi" /QB ENABLEAUTOPROTECT=1 RUNLIVEUPDATE=0 REBOOT=ReallySuppress

What he's saying is that the IUSR_* account is considered a system/service account, not the SYSTEM account. You never actually logon with that account so no profile is created.

Actually, his laptop does NOT have a Pentium M. It has a Pentium 4-M. They're compeltely different CPUs. He even linked (corrected link since he linked to the 272XVi but stated he has the 273XV) the information to his system. The Pentium 4-M's went to a higher clock, but his system won't support those higher clocked P4-M's. If you look at his CPU-Z data, he has this processor. He could only wish it was a Pentium M since at 1.7GHz it will run circles around a Pentium 4-M 1.7GHz.

They just put 169.04 on their site this morning. And yes...they're still beta (as are/were the 169.02's). http://www.nvidia.com/object/winxp_169.04.html http://www.nvidia.com/object/winxp64_169.04.html http://www.nvidia.com/object/winvista_x86_169.04.html http://www.nvidia.com/object/winvista_x64_169.04.html The latest WHQL drivers are still 163.69 for Vista and 163.71 for XP.