Wat voor mobo heb je ??
Misschien is dit iets voor je ? :
Dit komt van de abit faq, maar misschien ook wel van toepassing op jou systeem, give it a try.
It has been reported on the KT7A motherboard that with ACPI enabled under Windows 2000 the disk burst speed is considerably lower than when ACPI is disabled (20MB/s versus 80MB/s in the example quoted).
Slow disk burst speeds have been linked to problems with the network card sharing a hardware interrupt with another card. One user reported an increase in burst speed from 40MB/s to 50MB/s by resolving a conflict with his network card (see "How do I avoid IRQ conflicts?" in instabilities section). If your disk burst speed is unacceptably low, it may mean that you have a hardware IRQ conflict between two PCI devices that are not able to share hardware IRQs. Use the FAQ mentioned above to help troubleshoot this. Note, one user found that the IRQ sharing under ACPI caused his IBM DeskStar DJNA-371350 ATA-66 drive to have a burst speed of only 15MB/s under Windows 2000 , but this increased to 58MB/s when he disabled ACPI.
Significant performance increases can often be achieved by disabling the "PCI#2 Access #1 Retry" setting in the BIOS - this is enabled by default.
Note for some CDROM's, disabling DMA support can result in a significant speed increase. One user reports an increase from around 20x speed to nearly 70x speed on his Kenwood TrueX 72 speed CDROM player when DMA is disabled. One user found this helped also a problem with very slow disk performance (0.1MB/s) on IDE1 (with 7200rpm 40GB IBM and Western Digital disks and Windows 98). The speed increased to around 15MB/s - typical for PIO mode - but not as up to the 33MB/s+ that is achievable with DMA enabled.
Note that some users report a reduction in disk burst speed under Windows 2000 only with the WW and WZ BIOS versions.
One user reported that if his I/O voltage was raised above the default 3.3v then his disk's burst speed fell from 53MB/s to 28MB/s. Strange but true - apparently this problem is fixed in BIOS WZ and later.
One user found that with a CD-Writer and CDROM on their own channels as master, on IDE 1 and 2, the burst rate of the harddisk on the Highpoint Primary controller fell from 70 to 50 MB/s. This was solved by putting both the CD writer and CDROM and master/slave on IDE1 and leaving IDE 2 unused.
Try disabling unused COM ports or the LPT port to free up additional IRQs. This, particularly in a system with many PCI devices, may increase the burst speed
Heb je hier meer aan ?
:strip_icc():strip_exif()/u/1248/crop5734e2b56a28a.jpeg?f=community)
