[ On Fri, January 22, 1999 at 09:25:38 (+0800), Chan Yiu Wah wrote: ]
> Subject: 10G IBM ide hard disk
>
> I am going to buy the 10G IBM hard disk and I would like to konw if anyone
> have encounter any problem with it.  If so,  can you share your experience 
> with me.  Thanks.

I've just recently installed a number of IBM DGHS09U (8748MB) Ultra-Wide
SCSI drives.  I don't know if these would be similar to the one's you're
referring to, or not.

Most modern IBM SCSI drives incorporate their intelligent error handling
and predicting technology, which seems pretty nifty, but I've yet to see
it actually do anything.

We also had a number of IBM DDRS-34560W Ultra-Wide 4GB drives on the
same i686 machines, which have onboard Adaptec AIC-7880 controllers.

Oddly enough even though the 9GB drives have a 512KB cache and the 4GB
drives only have a 6x64KB/3x128KB cache, the 4GB drives attain about 1.7
times the write speed as the 9GB drives.  The former will write (using
dd from one raw drive to another, with seemingly the optimum size 64KB
buffers) at about 9.1MB/s, and the latter will only write at about
5.2MB/s.

(One interesting aspect of these higher-speed drives is that if you copy
the entire drive sequentially you can clearly see the speed differences
between outside cylinders and inside cylinders!)

Unfortunately with the NetBSD Adaptec driver (1.3.3 and -current) the
drives never negotiate any more than 10MHz (with 16-bit) transfer rates,
even though they claim to be able to go to 80MHz.  Perhaps with a higher
transfer rate negotiation the 9GB drives would perform properly....

FYI here's a sample of the boot messages:

Jan 20 14:40:15 proven /netbsd: ahc1 at pci1 dev 4 function 0
Jan 20 14:40:15 proven /netbsd: ahc1: interrupting at irq 15
Jan 20 14:40:15 proven /netbsd: ahc1: Using left over BIOS settings
Jan 20 14:40:15 proven /netbsd: ahc1: aic7880 Wide Channel, SCSI Id=7, 16 SCBs
Jan 20 14:40:15 proven /netbsd: scsibus0 at ahc1 channel 0: 16 targets
Jan 20 14:40:15 proven /netbsd: ahc1: target 0 using 16Bit transfers
Jan 20 14:40:16 proven /netbsd: ahc1: target 0 synchronous at 10.0MHz, offset = 0x8
Jan 20 14:40:16 proven /netbsd: sd0 at scsibus0 targ 0 lun 0: <QUANTUM, VIKING 4.5 SCA, 8600> SCSI2 0/direct fixed
Jan 20 14:40:16 proven /netbsd: sd0: 4345MB, 6144 cyl, 8 head, 181 sec, 512 bytes/sect x 8899737 sectors
Jan 20 14:40:16 proven /netbsd: ahc1: target 1 using 16Bit transfers
Jan 20 14:40:16 proven /netbsd: ahc1: target 1 synchronous at 10.0MHz, offset = 0x8
Jan 20 14:40:16 proven /netbsd: sd1 at scsibus0 targ 1 lun 0: <QUANTUM, VIKING 4.5 SCA, 8600> SCSI2 0/direct fixed
Jan 20 14:40:16 proven /netbsd: sd1: 4345MB, 6144 cyl, 8 head, 181 sec, 512 bytes/sect x 8899737 sectors
Jan 20 14:40:16 proven /netbsd: ahc1: target 2 using 16Bit transfers
Jan 20 14:40:16 proven /netbsd: ahc1: target 2 synchronous at 10.0MHz, offset = 0x8
Jan 20 14:40:16 proven /netbsd: sd2 at scsibus0 targ 2 lun 0: <QUANTUM, VIKING 4.5 WSE, 880P> SCSI2 0/direct fixed
Jan 20 14:40:17 proven /netbsd: sd2: 4345MB, 6144 cyl, 8 head, 181 sec, 512 bytes/sect x 8899737 sectors
Jan 20 14:40:17 proven /netbsd: ahc1: target 3 using 16Bit transfers
Jan 20 14:40:17 proven /netbsd: ahc1: target 3 synchronous at 10.0MHz, offset = 0x8
Jan 20 14:40:17 proven /netbsd: sd3 at scsibus0 targ 3 lun 0: <IBM, DDRS-34560W, S71D> SCSI2 0/direct fixed
Jan 20 14:40:17 proven /netbsd: sd3: 4357MB, 8387 cyl, 5 head, 212 sec, 512 bytes/sect x 8925000 sectors
Jan 20 14:40:17 proven /netbsd: ahc1: target 4 synchronous at 10.0MHz, offset = 0xf
Jan 20 14:40:17 proven /netbsd: sd4 at scsibus0 targ 4 lun 0: <IBM, DCAS-34330, S65A> SCSI2 0/direct fixed
Jan 20 14:40:17 proven /netbsd: sd4: 4134MB, 8205 cyl, 6 head, 171 sec, 512 bytes/sect x 8467200 sectors

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>