<<
I'm puzzled. I'm running a MSI BxMaster that has built in IDE 66 Promise controller, and when I click on Internet Explorer, there is a delay of about 10 seconds, after which the window pops up and I see Explorer. The CPU is a PIII at 933, the memory is 512mb, the hard drive is a 30 gb Maxtor ATA 100 drive. I've got a second hard drive, a 40gb Quantum on the same IDE channel, and two CD-Roms on the other channel. The delay is driving me crazy, and I can't seem to figure it out. I didn't have it when I was running NT 4.0. The first drive is configured as one large NTFS drive, and the second is multiple logical drives. >>
Couple of things. Went to the MSI site and here's what it says about the onboard Intel controller for your board:
<<
An IDE controller on the Intel® 82371EB PCI Chipset provides IDE HDD/CD-ROM with PIO, Bus Master and Ultra DMA 33/66 operation modes. Can connect up to eight IDE devices ( 4 Ultra DMA33 + 4 Ultra DMA66 ). >>
Here's what it says about the onboard promise controller:
<<
Ultra DMA/ATA 66
Promise PDC20262:
-- Fully PCI Bus 2.0 Compliant.
-- Compatible with the latest PCI IDE, ATA-4, ATAPI and Enhanced IDE specifications.
-- Supporting up to four IDE devices.
-- IDE drive transfer rate is capable of up to 66MB/sec per. >>
So....it sounds like the onboard Intel controller can only deliver up to ATA 33, whereas the promise controller can deliver ATA 66. Right? I mean, that's what I make of the "Can connect up to eight IDE devices ( 4 Ultra DMA33 + 4 Ultra DMA66 )" statement. Man, it's wonderful how "concise" some companies can be in their product descriptions, isn't it?....
Alright, you identify your Maxtor drive as an ATA 100 drive, but you do not identify the ATA rating for the Quantum (which I assume you have setup in that order as master and slave.) Of course you know that since the Promise controller is only capable of ATA 66, that your Maxtor drive will only be able to deliver ATA 66, and I'm assuming it will automatically step itself down to IDE 66 on sensing the limits of the Promise controller. However, in order to get ATA 66, you have to be running an 80-pin cable between the Maxtor and the Quantum and the Promise controller (as a standard 40-pin IDE cable will automatically reduce the ATA mode for both drives to ATA 33.)
So here's one possibility: While your Maxtor is known to be ATA100 capable (and thus theoretically ATA 66 capable), if your Quantum drive is an ATA33 drive only, then this sets up the Master as an ATA 66 and the slave as an ATA 33, and both are running on the same channel. Some hard drive controllers do not LIKE the situation of running two different-timed ATA drives on the same channel, and have no mechanism to cope with it--thus, it can lead to lengthy access times such as the one you described, and even more worrisome, it can lead to lots of drive errors when accessing files between them. I had that precise problem with an Abit KT7 (KT133) mobo I had. That controller took extreme exception to my running two drives with different timings on the same channel--and I had lots of slowdowns such as you mention--and LOTS of errors. Actually the situation became untenable for me--I really couldn't use the system setup that way. So, I solved it at the time by installing a 40-pin IDE cable between both drives which leveled them out to ATA33 timings, and henceforward had no further problems either in access slowdowns or in write errors between the drives. (To be fair, I can't recall precisely which mobo I had with this problem--it may have been an Epox--I've had so many, it's hard to remember which was which at times.) Anyway, if your Quantum is by chance an ATA33, I'd recommend you install a standard 40-pin IDE between the two drives. (For that matter, if you discover that this is your problem and you have to run in ATA 33, what I'd do is disable the promise controller and just move everything to the Intel controller.)
Which brings me to the next issue--make sure that your onboard Intel controller is DISABLED in your bios entirely. The only reason to have it enabled at all along with the Promise controller is if you want to run more than 4 IDE devices, which apparently you don't. So if your Intel controller is still enabled , but with nothing connected to it, then your access delays could be the result of the system sorting out which IDE controller to use and taking a bit of time before it makes the right choice. So disable the Intel controller if you haven't already done that as currently you aren't using it.
Last, you mention that your Maxtor drive is "one big NTFS drive" and that your Quantum drive is "multiple logical drives." By your description then, it would seem you have multiple NTFS
partitions on the Quantum drive. It would be interesting to note whether you have one single
primary partition on the Quantum, ideally it should be the first partition, after which you'd create an
extended partition which would encompass the rest of the space on the Quantum, that you could then subdivide into as many logical drives as you need. Unless something is different with NTFS than with all other Microsoft file systems (FAT, FAT32, etc.),
every separate hard drive in your system
must have at least one primary partition--which is usually the first partition on the drive. Now, I may well be wrong about this and if I am I'd much appreciate it being pointed out to me....
But it's difficult to tell from your comments exactly how you have partitioned your second drive. If it's one giant extended partition subdivided into logical drives, with no primary partition present, it may well be that your system is pausing to figure out exactly what's what.
Here's the thing--is this the *only* problem you are having, that of IE taking way too long to load? You really don't say whether it is or isnt. But if it is, you can probably ingore all of my advice above as it probably doesn't apply. Tech support is wonderful, ain't it?
In that case, I'd simply recommend that you do a repair reinstall (as opposed to a "new" reinstall) of XP, because it may just be something relating to software and your initial installation--which you also don't talk about. Prior to the repair reinstall, though, you could run System File Checker in XP and if a system file has inadvertantly been overwritten SFC will replace it with the proper version, and that may fix your problem. If not, I'd say a repair reinstall, booting from the CD, is definitely in order.
Sorry about not being of more help, but the information you've given is scant--such as whether or not you did an initial clean install of XP Pro--or an upgrade--in which case doing a clean install is definitely in order. You don't mention a dual boot scenario, but I can't judge from your comments whether this applies, either.
But if you are intent on using the Promise controller then I would definitely check the Promise web site to see if new XP-compatible drivers have been released since XP shipped--and if so I'd definitely install them. Basically, though, something has made you conclude that this is related to your hard drive controller. Was this just a guess, or are you having other problems you haven't mentioned? If you are having those problems, which would explain why you'd narrow your focus to your hard drive controller as a culprit, then some of the advice above may apply. If your only problem is the delay in accessing IE, then your problem is not likely to be related simply to the hard drive controller--for instance, in some instances XP takes a strong dislike to over clocking--ESPECIALLY when aggressive memory timings are used. Do not assume that bios settings which worked properly for you under NT "should" also work the same under XP. They are two different OS's and in many instances they specifically work the hardware in different ways--so you should start with default bios settings and work up from there.
My latest mobo is an MSI K7T Turbo2 (v5), and runs Xp Pro and an Athlon XP 1600 +, and currently is rock-stable at a bus speed of 145MHz, which puts the overclocking of the cpu at the 1800 + level at 1.73v (according to the bios sensors), and the chip is so identified by XP running in this fashion. I chose this mobo specifically because it would allow me to use my existing 133 SDRAM while at the same time fully supporting the Athlon XP, Win XP, and even the new ATX 12-volt connector power supplies--which actually help even with Athlon XP systems because they isolate more cleanly the power delivered to the CPU from that delivered to the rest of the system via the standard ATX power connector. IE, these mobo connections, contrary to some popular opinion, are not for P4 systems only. Anyway, bottom line was that my prior board, an Abit KT133 (KT7) proved very problematic with XP when I invoked the asynchronous bus settings of 100MHz front side bus and 133MHz DRAM bus. In this state, every single 3D game I owned BSOD'd as soon as the game tried to switch to 3D mode--and the BSOD blamed the nVidia GF3 driver as the culprit. Turned out the solution was simple for that mobo--I just set the fsb and the DRAM bus to run synchronously @ 100MHz, and instantly every 3D game I had ran perfectly, and I never saw that "infinite loop" error again--not even once. With this MSI mobo, fsb and DRAM bus are running synchronously, although at 145MHz, and I haven't seen the "infinite loop" error even once. The point of all of this is just to demonstrate that XP works hardware subsystems differently than other M$ OS's, and so don't assume that bios settings which you set for NT should automatically be fine with XP. Not necessarily at all. BTW, the interesting thing about that Abit board with reference to the nVidia "infinite loop" errors under XP is that under Win9x the asynchronous bus setting with the same hardware worked fine with no problem--but under XP things work differently for different hardware--and so I was required to make some bios changes after installing XP that were tolerated without a problem under Win9x.
OK, time for me to sut up....
Hope you can glean something of interest or help out of this lengthy diatribe. Best to you!
Facebook
Twitter