Image Map Image Map
Page 2 of 4 FirstFirst 1234 LastLast
Results 11 to 20 of 39

Thread: XT-IDE vs. SCSI on IBM XT/286

  1. #11

    Default

    Quote Originally Posted by 1ST1 View Post
    It's easy to get flash based IDE drive solution (CF with adapter), and that's much faster.
    No, it isn't

  2. #12
    Join Date
    Oct 2014
    Location
    near frankfurt/m, germany
    Posts
    981

    Default

    XTIDE_CORETEST.jpg
    This is XT-IDE with a CF card as drive 1 versus 20 MB MFM drive as drive 0.

  3. #13

    Default

    Quote Originally Posted by 1ST1 View Post
    XTIDE_CORETEST.jpg
    This is XT-IDE with a CF card as drive 1 versus 20 MB MFM drive as drive 0.
    Yes I'm sure it is faster than a MFM drive, but it's not faster than a good SCSI. Or probably even a modern IDE drive for that matter.

    http://www.vcfed.org/forum/showthrea...ine&highlight=

  4. #14

    Default

    It is highly illogical to demand speed from vintage PCs Unless you are member of a rabbit secta

  5. #15
    Join Date
    May 2011
    Location
    Outer Mongolia
    Posts
    1,775

    Default

    Quote Originally Posted by maxtherabbit View Post
    Yes I'm sure it is faster than a MFM drive, but it's not faster than a good SCSI. Or probably even a modern IDE drive for that matter.
    Unless you're running a highly unusual workload on your 286 access time is probably going to matter more than data transfer speed, and since flash solutions have nearly instantaneous random access times I would expect that a CF card would "feel" faster than even the most amazingly ridiculously fast SCSI drive, although realistically the difference probably wouldn't really be perceptable. Even a humble 8-bit XT-CF manages to benchmark at better than 500K per second in my 7.16mhz V-20 machine; 500Kps translates to being able to load essentially the largest possible DOS executable in one second; at that point would it matter much if it took half a second instead?
    My Retro-computing YouTube Channel (updates... eventually?): Paleozoic PCs

  6. #16
    Join Date
    Oct 2014
    Location
    near frankfurt/m, germany
    Posts
    981

    Default

    In that screenshot you can see 350 KB/s data transfer rate and a seek time of allmost ZERO. No cache, no mechanical SCSI drive can do that in an XT or 286 over the full capacity of the drive. I didn't expected the necessarity to explain that.
    Last edited by 1ST1; November 30th, 2019 at 11:45 AM.

  7. #17

    Default

    Quote Originally Posted by Eudimorphodon View Post
    Unless you're running a highly unusual workload on your 286 access time is probably going to matter more than data transfer speed, and since flash solutions have nearly instantaneous random access times I would expect that a CF card would "feel" faster than even the most amazingly ridiculously fast SCSI drive, although realistically the difference probably wouldn't really be perceptable. Even a humble 8-bit XT-CF manages to benchmark at better than 500K per second in my 7.16mhz V-20 machine; 500Kps translates to being able to load essentially the largest possible DOS executable in one second; at that point would it matter much if it took half a second instead?
    I know it's kind of buried in my thread, but the SCSI disk beat the CF card in both the "8K random, 70% read" and "sector random read" metrics of DISKTEST as well as sustained transfer metrics

  8. #18
    Join Date
    May 2011
    Location
    Outer Mongolia
    Posts
    1,775

    Default

    Quote Originally Posted by 1ST1 View Post
    In that screenshot you can see 1.4 MB/s data transfer rate and a seek time of ZERO.
    Maybe I'm misinterpreting the output, but it looks to me like it's 1.4MB/s over four seconds. (There's specifically a line that says "353.1KB/sec".) But the point is still clear from the "performance Index" output that this benchmark is *extremely* impressed by 1.5ms access time. As it should be.

    (Different benchmarks seem to differ a lot when it comes to data transfer rates. The 500+Kps I quoted was from Checkit. I don't think I've tried CORE on that machine.)
    My Retro-computing YouTube Channel (updates... eventually?): Paleozoic PCs

  9. #19

    Default

    Ya', nothing really compares to nitpicking over factors with fundamentally meaningless values.
    PM me if you're looking for 3" or 5" floppy disks. EMail For everything else, Take Another Step

  10. #20
    Join Date
    May 2011
    Location
    Outer Mongolia
    Posts
    1,775

    Default

    Quote Originally Posted by maxtherabbit View Post
    I know it's kind of buried in my thread, but the SCSI disk beat the CF card in both the "8K random, 70% read" and "sector random read" metrics of DISKTEST as well as sustained transfer metrics
    The Quantum Atlas has an 8MB cache buffer with a sophisticated prefetch algorithm. According to this page:

    https://www.lo-tech.co.uk/wiki/DOS_Disk_Tester

    That test, unless otherwise instructed, uses a 4MB test file. It's very possible the entire file fits into the cache. In which case, sure, it's going to be faster, it's running from actual RAM.
    My Retro-computing YouTube Channel (updates... eventually?): Paleozoic PCs

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •