yon%smltalk.cbm.dec.com@decwrl.dec.com (David A. Yon) (05/08/91)
Hey all, I've just purchased a 33Mhz 386 clone, which included a 124meg IDE drive (the Seagate ST1144A) with a generic IDE controller. In the course of researching this purchase (calling Gateway, Treasure Chest, etc) I was qouted data transfer rates of around 1.0-1.2mb/sec on IDE drives. Yet with this drive I'm only getting around 550Kb/sec! I purchased from a local dealer at a computer show, and when I called him about this problem, he said that he had never seen IDE drives do much more than 550Kb/sec. Now, I'm upgrading from a 12mhz 286 which had the ST296N/ST-02 SCSI combination, and I was getting 808Kb/sec out of that. (granted I've improved the access time from 28ms to 20ms, but I'm not sure which will have a larger effect on the performance). So... 1) Does anyone have this combination? What are your experiences? 2) Does anyone know the typical performance of IDE drives/controllers? Would the setup of the IDE card matter (there were precious few jumpers on the card) 3) Does anyone know what a jumper labeled "IRQ 14 buffered/unbuffered" on the IDE card would mean? The strange thing here is that the jumper is described as either shorting pins 1-2, or shorting pins 2-3. Yet there was no jumper installed at all. I remember that the ST-02 has the all-important "0 wait-state" jumper, which when enabled made a big difference in the transfer rate of my drives. I didn't see anything like that on my IDE card. Thanks in advance for any help... David Yon ---- yon%smltalk.cbm.dec.com@decwrl.dec.com yon@world.std.com 617-621-7427 (w) 508-443-7526 (h)
rreiner@yunexus.YorkU.CA (Richard Reiner) (05/09/91)
> I've just purchased a 33Mhz 386 clone, which included a 124meg IDE >drive (the Seagate ST1144A) with a generic IDE controller. >with this drive I'm only getting around 550Kb/sec! A friend and I each have the same drive in cached 386-33s, and we each get about 980 KB/sec. > I purchased from a local dealer at a computer show, and when I called >him about this problem, he said that he had never seen IDE drives do much >more than 550Kb/sec. The dealer is either inexperienced or a liar. I have almost never seen a non-tiny one (80 MB or more) do *less* then 700 KB/sec in a 386 machine, and I've seen plenty of them. //richard
bressler@iftccu.ca.boeing.com (Rick Bressler) (05/09/91)
I think the primary difference is in the benchmark used. For example, I see almost the exact same figures, (1.2MB/sec) with CORETEST and 5-600k/s with checkit and my AMI bios built in diagnostics. Probably, CORETEST is not getting very far outside of the built in cache, or is reading without moving the heads at all. It's hard to tell, since the 'real' configuration of the drive is hidden. On the other hand, CORETEST reports lower rates, down to 900k/s on the inner tracks, and this is as it should be, since the density of the data drops from 44 (I think) sectors / track on the outer tracks, to something like 32 on inner tracks. In fact, these numbers actually seem reasonable. Theoretically: 44 sectors / track * 3600 RPM at interleave of 1 should yield 1.28 mb / sec. 32 " " " " 937k/sec. These are actually about what CORETEST reports. Perhaps it is running efficiently enough that it is actually getting an advantage from the 1:1 interleave, where the other tests are real - life benchmarks where the 1:1 interleave is two fast. You will note the figures are almost exactly 1/2 those of CORETEST.
sigma@obee.ipl.rpi.edu (Kevin Martin) (05/09/91)
rreiner@yunexus.YorkU.CA (Richard Reiner) writes: >> I purchased from a local dealer at a computer show, and when I called >>him about this problem, he said that he had never seen IDE drives do much >>more than 550Kb/sec. >The dealer is either inexperienced or a liar. I have almost never >seen a non-tiny one (80 MB or more) do *less* then 700 KB/sec in a 386 >machine, and I've seen plenty of them. This depends enormously on how you measure transfer rate! CoreTest 2.92 or whatever the latest is reports my CP-3204F IDE drive at 1136 Kb/sec. This is in a noncache 386/25. But CheckIt! reports 546 Kb/sec. No other program tells me anything useful. All I need to know is that the drive is FAST, especially with a 3Mb cache. -- Kevin Martin sigma@ipl.rpi.edu "Can I kiss one of the bridesmaids instead?"
yon%smltalk.cbm.dec.com@decwrl.dec.com (David A. Yon) (05/09/91)
Hey all, Seems I've found the problem with the ST1144A. I called Seagate yesterday and they told me to try running Coretest 2.92, since that version handles IDE drives better. So I went home and ran Coretest again (that was the source of my original Data Transfer Rate quote, by the way), to find that it was version 2.91. "Aha!" I thought, "wrong version!" ...and then I looked at the result of the benchmark... 959Kb/sec! Huh...? I thought about it for second, and the disparity became clear. When I originally ran Coretest, the drive was partitioned the way my dealer had done, with a single 130meg C: drive. Call me wierd, but I *like* having several logical drives, so I went ahead and re- partitioned the drive despite the poor Coretest result. And then I never ran Coretest again until last night! My guess was that the repartitioning was enough to slightly change the logical-to-physical geometry mapping, causing an adjustment in how the logical cylinders mapped to the physical cylinders, or perhaps moved a logical set of clusters away from a locked-out sector. Either of these would slow down the DTR that Coretest reported. In the first case, if Coretest were reading a set of clusters that it assumed were on the same cylinder, but actually physically resided on two adjacent cylinders (do to the geometry translation that usually goes on in SCSI/IDE drives), the extra head movement would invalidate the DTR finding. In the second case, if the logical clusters resided among some locked-out bad blocks, the time spent skipping over those blocks would also cause a lower DTR to appear in the benchmark. At any rate, the drive is performing flawlessly, and blazingly fast! Thanks to all who responded, especially the email that I got from overseas this morning (Sweden, England, and Norway...gotta love the Internet!). David Yon
ericb@hplsla.HP.COM (Eric Backus) (05/10/91)
> I've just purchased a 33Mhz 386 clone, which included a 124meg IDE >drive (the Seagate ST1144A) with a generic IDE controller. In the course >of researching this purchase (calling Gateway, Treasure Chest, etc) I was >qouted data transfer rates of around 1.0-1.2mb/sec on IDE drives. Yet >with this drive I'm only getting around 550Kb/sec! I have a Gateway 33Mhz 386 Cache system with a Seagate ST1144A drive. If I completely strip the system down (NO config.sys or autoexec.bat files at all), Norton SI measures a transfer rate above 900 Kb/sec. With smartdrv.sys and a few other things installed, it measures more like 750 Kb/sec. Gateway originally told me the drive would do 1.25 Mb/sec, which may have been a little optimistic. I have no idea how accurate Norton SI is, but I do know I've been happy with my drive. -- Eric Backus ericb%hplsla@hplabs.hp.com (206) 335-2495
whitney@peewee.cs.unlv.edu (Lee Whitney) (05/10/91)
In article <1991May9.141234.4719@hollie.rdg.dec.com>, yon%smltalk.cbm.dec.com@decwrl.dec.com (David A. Yon) writes:
) At any rate, the drive is performing flawlessly, and blazingly fast!
Do you all think the 1144A is really that fast? After using both drives for a couple months it seems that it is slower than the conner drives (80 meg), although this is subjective.
Has anybody used both the 1144A and one of the Seagate Swift (15ms) drives? I am wondering how much of a practical performance difference there is between the two families.
Leemvolo@uncecs.edu (Michael R. Volow) (05/11/91)
Transfer rates in HD tests depend on block size. The version of Coretest that I have defaults to 32-64k block size. I believe that this is an unrealistic way to test hard disks, as programs often load in chunks of .5-2k. Why not re-run the test with Coretest, but specifying a small block size and then report the HD's transfer rate. -- Michael Volow, Psychiatry, Durham VA Med Center, Durham NC 27712 919 286 0411 Ext 6933 mvolo@ecsvax.edu