VOGONS


socket 3 system on stereoids *Updated*

Topic actions

Reply 20 of 25, by Amigaz

User metadata
Rank Oldbie
Rank
Oldbie
5u3 wrote:

Oh, forgot about the NCR 810 SCSI card.

The board/chipset doesn't support PCI dividers, so other PCI harddisk controllers most likely would have problems as well.
If you want to run 40MHz on the PVI-486SP3, I'm afraid you'll have to use the crappy onboard CMD640 or insert a VLB/ISA based controller.

I'll stick at 33mhz FSB 😀 I won't be able to run any SVGA games at decent speed any way with 40mhz FSB
It would feel like going back to jurassic times if I went back to IDE again 😁

Reply 21 of 25, by Amigaz

User metadata
Rank Oldbie
Rank
Oldbie
Anonymous Coward wrote:

This is why I favour the VL/EISA boards. I can run the VL bus at any crazy speed I like but still run high performance disk controllers in spec.

Agree but still haven't found an Eisa mobo which has jumpers for FSB settings...

Reply 22 of 25, by MartinC

User metadata
Rank Newbie
Rank
Newbie

If your controller card is SCSI-2, how could your HDD have scored anything over 10MB's?

Win98 Gold: 1GHz PIII - GeForce2 - Voodoo2 - 768MB - SCSI 😀

Reply 24 of 25, by MartinC

User metadata
Rank Newbie
Rank
Newbie

http://en.wikipedia.org/wiki/SCSI

Confusing but regardless he scored 31,325 KB/s on one of the tests...

Win98 Gold: 1GHz PIII - GeForce2 - Voodoo2 - 768MB - SCSI 😀

Reply 25 of 25, by 5u3

User metadata
Rank Oldbie
Rank
Oldbie

I don't think the SST linear verify speed test actually transfers data over the bus. Even my Seagate 13GB IDE drive manages more than "20 MB/s" on the same board.
A SiS 496/497 board manages about 15 MB/s over the PCI bus (with a following wind 🤣). Actually it's more like 10 MB/s, because Amigaz' fast SCSI drive gets about the same results in the read speed tests as my crappy IDE drive.