Title

Roy Longbottom at Linkedin  DiskGraf Disk, LAN and USB Benchmark Results On PCs

See     Example Log File     and     Results and Examples

Description

DiskGraf Benchmark measures disk writing and reading speeds at different block sizes, producing a graph of megabytes per second and CPU utilisation. It also does the same for bus/DMA speeds and measures random reading time of various sized files. Results are saved in a text file. This document is a summary of maximum speeds shown for a range of PC systems. Pre-compiled versions of the benchmarks can be found in DiskGraf.zip which also contains the source code and more detailed explanations. The benchmark can also be used for testing disks on a Local Area Network and for CD/DVD drives with RW discs (see Results). See also CDDVDSpd Results.htm for disk and other device speeds at different file sizes and My Main Page for other PC benchmarks and results. Latest benchmark is IOSpeed that combines test functions from DiskGraf and CDDVDSpd.

The benchmark writes a number of files (default 5, maximum 16) then reads them, each at block sizes 1 KB to 1024 KB. Default file size is 8 MB, but this can be increased to up to 10 GB. For compatibility with Windows 9X, CPU utilisation is measured via a second thread that soaks up free CPU cycles. This leads to the utilisation results only being valid on single CPU systems.

DMA speed is measured by repetitively reading the same block. Normally, this will be from the disk’s buffer, directly over the bus, to the CPU. These tests again use block sizes 1 KB to 1024 KB. Note that measured disk transfer rates will be less than those on the DMA test.

The random test measures random reading time in milliseconds of 1KB blocks out of 1, 2, 4 and 8 MB or larger (up to 1 GB, depending on maximum file size used for the other tests). As shown in the results, this helps in identifying the disk’s buffer size and RPM. On the latest disks, with buffer sizes of 8 MB or greater, the default file size has to be increased (menu option) to produce appropriate measurements.

The benchmark program uses FILE_FLAG_NO_BUFFERING in the CreateFile function to tell Windows not to keep the data in main memory based File Cache. Without this setting, data would generally be read from memory instead of from the disk. The program has an option to enable caching and show memory to memory data transfer rates.

A 64 bit version is now available in More64Bit.zip with minor changes described in 64 Bit Disk Tests.htm. These include a different method of measuring CPU utilisation, applicable also for dual CPU systems. Results are included below (see q64AMax1 - XP x64, jC2DSea2 - 64 Bit Vista, also yDiskUSB2) and, as one would expect, are usually similar to 32 bit version speeds. Measurements include results for one of the fastest 2009 disks in an eSATA enclosure (system jC2DSea3). For further details See Report.

Results for year 2009 include a WD low power Green Disk (qPheWD1 using 64-Bit Windows 7). WD do not publish the usual performance specifications such as RPM and latency for these disks . Instead, they just indicate IntelliPower, with repudiated marketing suggestions that the disk speed varies between 5400 and 7200 RPM. The consensus is that the speed is (was?) fixed at 5400 RPM. This benchmark indicates maximum writing speed of 94 MB/second and 104 MB/second reading, better than some recent 7200 RPM disks. Extended Random Access Times suggest that the RPM might be somewhat higher than 5400.

Further 2009 disk results are for a Solid State Disk (SSD), where maximum reading speed of more than 200 MB/second could be limited by SATA 2 bus speed. Although this is much faster than current hard disk drives, writing speed is not as good, but constant random access time is superior at 0.1 milliseconds. A 2011/12 drive (jI5SSD) demonstrates SATA II speeds of greater than 200 MB/second on both writing and reading but random access times are slower at 0.3 milliseconds.

2014 additions are for a fast 7200 RPM disk drive (ji7WD1) with reading speed up to 170 MB/second and 500 MB/second DMA. Then a USB 3.0 disk drive (yDiskUSB3a), three times faster than via a USB 2 port and a USB 3 Flash Drive (zUSB3CF), writing at up to 116 MB/second and reading to 255 MB/second.


Example Results Log File


AMD XP 2.34 GHz, Hitachi 7K250 123 GB Disk, ATA100, 7200 RPM, ATA133 mainboard, Win2K


 ###########################################################
     Disk Test DiskGraf Version 1.3A Tue Apr 20 20:59:56 2004
 
              Copyright Roy Longbottom 2001

  Windows NT Version 5.0, build 2195, Service Pack 2
  CPU AuthenticAMD, Features Code 0383FBFF, Model Code 00000681, 2338 MHz
  Memory 523760 KB, Free 419596 KB
  F:Disk/partition 2051 MB, Free 2039 MB
  Files 5 Size 8 MB Block Sizes 11 Minimum RAM Data
  Notes: nVidia driver 4.12 

 Block   File ->
   KB    1     2     3     4     5 
        Write MB/Second
    1   5.9   6.1   6.2   6.2   6.2
    2  12.1  12.1  12.1  12.1  12.0
    4  22.0  22.1  22.2  22.3  22.2
    8  38.9  39.5  39.6  40.0  39.7
   16  61.9  52.5  58.2  61.8  58.4
   32  59.4  64.0  52.3  58.1  58.7
   64  62.4  64.0  62.6  62.6  67.0
  128  58.3  64.0  59.4  52.6  63.0
  256  54.0  65.8  39.0  55.2  64.0
  512  61.6  56.2  66.3  58.0  66.5
 1024  59.1  50.1  62.1  58.2  52.1
        Read  MB/Second
    1  10.7  10.8  11.0  10.7  11.0
    2  19.6  20.4  20.0  20.1  20.4
    4  29.8  30.1  29.4  29.6  29.5
    8  47.5  48.9  49.6  49.0  49.1
   16  51.0  52.0  52.1  51.6  55.2
   32  53.6  52.2  52.3  54.6  55.2
   64  52.4  52.3  52.3  52.3  52.4
  128  51.2  52.7  52.6  55.7  52.7
  256  51.4  53.4  53.3  53.3  53.4
  512  54.2  55.3  57.8  55.0  55.1
 1024  52.0  55.0  57.8  55.0  55.1
        Write CPU Utilisation
    1  18.6  16.9  17.1  16.7  16.9
    2  16.9  16.1  16.2  16.4  16.6
    4  15.4  14.8  15.5  14.9  15.3
    8  15.8  13.5  14.4  13.9  13.6
   16  10.8   8.9   9.9  10.6  12.0
   32   7.6   7.5   6.3   8.7   7.0
   64   2.5   2.5   2.4   2.8   2.6
  128   2.1   2.2   2.2   1.6   4.4
  256   2.1   2.1   2.0   1.6   2.0
  512   2.0   3.6   2.1   1.5   3.9
 1024   3.1   1.3   1.9   3.8   1.4
        Read  CPU Utilisation
    1  26.5  26.8  27.9  26.7  27.5
    2  24.4  24.8  25.1  26.2  24.9
    4  17.9  18.1  17.7  18.7  18.2
    8  15.4  14.9  15.1  16.8  15.1
   16   7.3   7.5   9.9   9.5  10.2
   32   5.6   6.4   5.5   3.7   3.6
   64   2.6   1.2   1.2   2.5   1.3
  128   1.1   1.5   2.4   1.3   1.1
  256   3.3   1.0   1.0   1.1   1.1
  512   1.2   1.2   1.3   1.6   1.2
 1024   1.1   1.1   1.3   1.1   2.8

 DMA Reread block up to 100 times at 1 to 1024 Kbytes

 Read  MB/Second
  10.8  19.1  46.3  64.9  80.3  86.9  90.8  91.0  91.2  91.1  91.1
 Read  CPU Utilisation
  27.0  52.5  30.2  40.8  12.4   6.5   3.0   2.8   3.3   3.1   2.9

 Random Read 1KB out of 1, 2, 4, to file size Mbytes

 Read  Milliseconds
 0.511  1.88  3.94  4.39 0.000 0.000 0.000 0.000 0.000 0.000 0.000
 Read  CPU Utilisation
   4.0   2.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0

               End Disk Test Tue Apr 20 21:00:57 2004

To Start


Examples and Results


DMA Operation Random Reading File Cache Operation
PCI and SCSI Speed Over Whole Disk Write Caching
USB Disks USB and Firewire Disks Local Area Network
USB and Compact Flash 4 to 16 GB USB Flash Drives DMA Problems
Driver Problems Disk Writing Speed Results Disk Reading Speed Results
Disk DMA Speed Results Disk Random Speed Results Extended Random Speed Results
CD and DVD RW Drives Systems Under Test Key

Results tables cover a range of disk drives providing data transfer rates between less than 1 MByte Per Second to greater than 200 MBytes Per Second, the highest being for RAID configurations. Speed is generally dependent on disk rotation speed (RPM) and amount of data on a track. However, it can be limited by DMA/bus speed with other variations due to bus driver used, mainboard, BIOS settings, Operating System version, disk space used, data fragmentation and, to a small extent, CPU and memory speed (these affect CPU utilisation). Comparison with the best results can help to identify problems, some of which are shown below.

CPU utilisation reduces with increased block sizes to close to zero percent on the latest processors, where measurement is not very accurate. The tables show CPU utilisation for block sizes up to 64 KB.

The original version had FILE_FLAG_WRITE_THROUGH in the CreateFile function. This made no difference in performance using Windows 9X but produced slower writing speed via Windows NT/2K/XP, with more significant differences on the later Operating Systems. The benchmark program was revised to remove this flag, the original being indicated by # in the writing speed table. Examples of differences are:



                  CPU  Disk            Writing MBytes/Second at KB Block Size              
  Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

  q64ASeaR  Old  1991 15000 0.33 0.62  1.2  2.4  4.3  7.3   13   21   29   32   52 
  q64ASeaR  New  1991 15000  7.4   13   20   39   60   82  106  108  112  133  133

  q64AMax1  Old  2210  7200    7   10   19   21   23   26   27   31   31   33   34
  q64AMax1  New  2210  7200   17   29   45   58   54   61   60   54   59   61   62

  oThoHi1   Old  2080  7200  0.6  1.3  4.2   14   26   30   26   35   34   36   38
  oThoHi1   New  2080  7200  6.5   12   22   36   45   52   47   46   47   47   46

  iP4Sea2   Old  1900  5400  1.1  1.8  2.7  4.3  7.0   10   11   13   13   15   15
  iP4Sea2   New  1900  5400  3.6  6.4   10   18   20   21   21   21   21   21   21

To Start


DMA Operation

Following are example of results that might be expected from the DMA test, which is measuring data transfer speeds from the disk’s buffer. Some examples show that speed can reduce at larger block sizes. It is not clear why this happens as the buffers are large enough to hold all the data and, sometimes, disk reading speeds can be faster than the lower figures. It must have something to do with repetitively reading the same data and, maybe, the disk reading ahead on the next track.

Two examples are given for q64AMax1. This has a 16 MB buffer and maximum data transfer rate of 66 MB/second. The latter, at 8.33 milliseconds rotation time, suggests a maximum track capacity of 550 KBytes. Yet it can not read the same 256 KB continuously.



                  CPU  Disk            MBytes/Second at KB Block Size
  Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

  gp3SeaA        1400  7200    7   11   15   20   23   25   26   27   26   20   16  UDMA  33  
  nTbdIBM8       1400  7200    8   17   33   43   50   56   59   60   60   60   60  UDMA  66
  nTbdQua        1200  7200    6   11   21   35   53   65   74   77   81   84   83  UDMA 100
  oThoHi1        2080  7200   12   25   48   71   78   87   91   93   93   93   93  UDMA 133
  q64AMax1       2210  7200   22   39   60   85  104  118  125  131   41   29   38  SATA
  q64AMax1       2210  7200   23   38   60   85  104  118  126  131   34   53   59  SATA
  jC2DSea1       2400  7200   15   18   35   68  109  145  173  208  229  237  243  SATA2
  ip4WD2         2411  7200   15   27   40   57   70   79   84   87   88   30   40
  q64ASea1       2150  7200    9   18   32   49   65   77   86   91   89   30   40

To Start     Results/Examples Index



Random Reading

As indicated above, the random test reads 1 KB blocks out of 1, 2, 4 MB or higher, the upper limit depending on the file size chosen for other tests. When this fits within the disk’s buffer, the reading speed is reasonably fast. When the size is greater than the buffer (or some smaller default), the average reading time can be expected to be slightly higher than half the disk revolution time. Here, the milliseconds for the various disk RPM specifications are:

4200 - 7.1 ms  5400 - 5.6 ms  7200 - 4.2 ms  10000 - 3.0 ms  15000 - 2.0 ms

Following are examples from the later detailed table to show what can be expected. With the 8 MB buffer, a lot of accesses are from the disk, but a test run with 16 MB files is really required.



           Buffer  CPU  Disk    Milliseconds at MB  file size
  Key          MB  MHz   RPM   1    2    4    8   16   32   64

  nTbdMax1      2 1533  5400  2.0  4.4  6.3  6.8  7.5
  mDurMax2      2  950  7200  0.8  2.6  4.3  4.9
  q64ASea1      8 2150  7200  0.4  0.8  2.0  3.9
  q64AMax1     16 2210  7200 0.04 0.04  0.2  0.9  2.9  4.1  4.9

To Start     Results/Examples Index



File Cache Operation

The following results show example reading speeds when the benchmark option is chosen to allow Windows to cache the data in memory. Writing speeds tend to be slightly slower than normal, when using this option. The table shows that there can be variations on the same PC, depending on the version of Windows, and speed can reduce with larger block sizes, as on the DMA test.



                  CPU  Disk            MBytes/Second at KB Block Size
  Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

  mDurMax1  W98   950  7200  132  168  137  175  198  210  153  140  141  140  144
  mDurMax2  W2K   950  7200  216  244  260  266  262  250  159  156  156  156  156
  iP4Sea2        1900  5400  340  455  554  633  682  706  703  388  263  266  262
  iP4Max         2400  5400  293  427  406  632  794  907 1004 1004  467  390  391
  q64AMax1 DCDR  2210  7200  690 1016 1292 1519 1630 1619 1601 1591 1352  840  838
  jC2DSea2 DDR2  2400  7200  420  717 1089 1512 1816 2000 2233 2331 2358 1995 2064
  

To Start     Results/Examples Index



PCI Bus and SCSI Operation

The standard 33 MHz PCI bus has a maximum speed of 133 MB/Sec and 93 MB/sec has been recorded with UDMA 133 on the DMA tests. As can be seen in the reading speed table, CPU utilisation using SCSI is much lower than for IDE disks. So, it comes as rather a surprise that PCI/SCSI disk reading speeds (and via the DMA test) have been at 60 MB/Sec or less. Except for a RAID0 set up, this did not matter much until recent times, when a single disk could achieve this data transfer speed.

The following show reading speeds on a 15K RPM disk (R0) and two in RAID configurations, all via an Ultra 160 controller. The tests were run at the different PCI bus speeds shown and results were similar to those measured on the DMA test. The disks have maximum rated speeds of 68 MB/Sec. Calculations indicate that the PCI/SCSI overheads for each block transferred are higher than the data transfer time. See later results for fast speeds via 133 MHz PCI-X.



             Bus  CPU  Disk            MBytes/Second at KB Block Size
  Key        MHz  MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

  oPalSeaR0   50 1700 15000  2.5  5.3   10   19   29   44   44   56   57   57   57
  oPalSeaR2   34 1700 15000  1.7  3.6  7.0   13   21   30   40   47   51   56   58
  oPalSeaR2 42.5 1700 15000  2.0  4.4  8.4   15   26   38   48   59   64   68   70
  oPalSeaRd   50 1700 15000  2.2  5.0   10   18   29   43   57   64   73   79   82
  

To Start     Results/Examples Index



Speed Over Whole Disk

Modern disks do not have the same amount of data on each track, but the surface is divided into zones which do have equal track capacities. The inner zone might have half the amount of data than in the outer zone and transfer rate will at half speed. Following are two examples of measurements in different partitions, where the last results are at the 92% and 95% point.



                  CPU  Disk            MBytes/Second at KB Block Size
  Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

  iP4Sea2    50% 1900  5400  6.0   11   15   21   28   28   28   28   28   27   28
  iP4Sea2    42% 1900  5400  6.2   11   14   24   24   24   24   24   24   24   24
  iP4Sea2     8% 1900  5400  6.2   11   19   19   19   19   19   19   19   19   19

  oPalIBMRd  22% 1396  7200   11   18   26   29   46   65   68   63   61   58   62
  oPalIBMRd  71% 1396  7200   12   20   30   39   55   56   63   63   62   58   62
  oPalIBMRd   5% 1396  7200   11   19   35   34   36   35   34   35   35   33   31

Speed can be slower than expected if a disk or partition has lots of data. The data might also be fragmented and a benchmark file could be scattered over a wide area, producing slow results. This is usually indicated by variations in speeds on the five (or whatever) files used for testing. At one time, tests showed that Windows wrote files in the first unused space. You could nearly fill the disk then delete almost all of the files but the next files saved would be on the slower part of the disk. This might still apply.

On a modern fast disk, maximum speed might not be shown with 8 MB files. The following shows the difference between reading 8 MB and 32 MB files, where the maximum transfer rate is greater than 100 MB/sec on a RAID system.



                 CPU  Disk            MBytes/Second at KB Block Size
 Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

 iP4WDSRd  8MB  3633 10000   14   23   37   53   66   53   88   83  100  104  102
 iP4WDSRd 32MB  3633 10000   14   24   37   54   68   65   89  103  108  109  108
  

To Start     Results/Examples Index



Write Caching

Depending at least on the bus driver and disk buffering strategy, sometimes writing speed can appear to be faster than reading and producing an apparent higher data transfer speed than the disk’s specification. This is due to Windows informing the benchmark program that transfer has been completed when data has been sent to the disk’s buffer.

The following shows measured writing and reading speeds on a SATA RAID0 system. This has two disk drives with 8 MB buffers and the maximum drive speed rating is around 60 MB/Sec. Default file size of 8 MB is used.



                CPU  Disk            MBytes/Second at KB Block Size
Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

 iP4HiSRd Write 3000  7200  5.0   10   19   35   56   69   82  110  155  166  168
 iP4HiSRd Read  3000  7200  6.4   18   29   47   69   55   67   68   85  105  101
  

To Start     Results/Examples Index



USB Disk Drives

The following show writing, reading and DMA speeds of disk drives via high speed USB 2. Three of the examples are for the same Hitachi 7200 RPM disk drive, FAT32 formatted, which is probably capable of achieving 60 MB/Second, same as the HS USB 2 maximum speed. In all cases, reading speed can be assumed to be the same as that for repetitive reading from the disk’s buffer for the DMA test.

Using WinXP, default operation Properties Policy is Optimise For Quick Removal (OQ below). Unlike with Flash Memory USB devices, the alternative Optimise For Performance (OP) is available for USB disks. Using OP, writing and reading speeds are very similar but writing speed is faster than OQ, due to overheads. Assuming a constant data transfer speed, overheads per block transferred can be calculated, as shown by the minimum values in the table. The larger program block sizes have much higher overhead, as though transferred in multiple blocks of 32 KB to 64 KB. CPU utilisation figures have not been shown as, in most cases, they were very small. The year of purchase of the PCs is shown. Comparing these and reducing overheads, indicates that the USB controller speed might determine the penalty of older technology. Results are now included for 64-Bit Vista, showing that the 64 bit version of the program produces the same performance as the 32 bit variety and writing speeds are slower than via XP x64.

The last two results are for a 2.5 inch 5400 RPM disk in a tiny enclosure that also obtains power from the USB. The second one is for NTFS formatting where writing small block sizes is faster and similar to results using Optimise For Performance. With FAT, writing speed is similar to the 7200 RPM disk but reading can be somewhat slower.



                   CPU                  MBytes/Second at KB Block Size
  Key              MHz  Mode     1    2    4    8   16   32   64  128  256  512 1024   Max

  Win2K - 2001 card Acomdata box, WD disk
  yDiskUSB2d      1533 Write   0.3  0.6  1.3  2.1  3.7    6    8    9    9    9    9     9
  yDiskUSB2d      1533  Read   0.6  1.2  2.4  4.1  6.2   10   13   13   13   13   13    13
  yDiskUSB2d      1533   DMA   0.7  1.3  2.6  3.6  7.8   10   13   13    8   13   13    13

  WinXP - 2001 card - Freecom disk - Write O/H 3.1 ms
  yDiskUSB2a  OQ  1900 Write   0.3  0.6  1.3  2.5  4.3    7   10   11   12   13   13    13
  yDiskUSB2a  OQ  1900  Read   0.7  1.3  2.6  5.2  7.7   10   14   14   14   14   14    14
  yDiskUSB2a  OQ  1900   DMA   0.7  1.3  2.6  5.1  7.8   10   14   14   14   14   14    14

  WinXP - 2001 card - Freecom disk - Write O/H 1.5 ms
  yDiskUSB2a  OP  1900 Write   0.7  1.3  2.6  5.1  7.7   10   14   14   14   14   14    14
  yDiskUSB2a  OP  1900  Read   0.7  1.3  2.6  5.2  7.8   10   14   14   14   14   14    14
  yDiskUSB2a  OP  1900   DMA   0.6  1.2  2.6  5.2  7.8   10   14   14   14   14   14    14

  Win2K - 2004 on-board - Freecom disk - Write O/H 0.5 ms
  yDiskUSB2c  OP  2080 Write   2.0  4.4  7.7   12   18   23   26   26   26   26   26    26
  yDiskUSB2c  OP  2080  Read   2.0  4.0  6.9   12   18   23   28   28   28   28   28    28
  yDiskUSB2c  OP  2080   DMA   2.0  3.9  7.5   12   18   23   28   28   28   28   28    28

  WinXP x64 - 2005 on-board - Freecom disk - Write O/H 1.0 ms
  yDiskUSB2b  OQ  2210 Write   1.0  1.9  3.8    7   12   19   25   29   30   32   30    32
  yDiskUSB2b  OQ  2210  Read   2.0  3.9  7.7   13   18   27   34   34   34   34   33    34
  yDiskUSB2b  OQ  2210   DMA   2.0  3.2  7.8   13   16   25   31   32   33   33   33    33

  WinXP x64 - 2005 on-board - Freecom disk - Write O/H 0.5 ms
  yDiskUSB2b  OP  2210 Write   1.9  3.9  7.7   12   20   28   34   33   33   33   34    34
  yDiskUSB2b  OP  2210  Read   2.0  3.9  7.9   13   18   27   33   33   35   35   35    35
  yDiskUSB2b  OP  2210   DMA   1.6  3.9  6.4   13   18   27   32   32   33   33   33    33

  64-Bit Vista - 2007 on-board - Freecom disk - Write O/H 1.0 ms
  yDiskUSB2f  OQ  2400 Write   0.9  1.8  3.7    7   11   16   21   22   24   25   25    25
  yDiskUSB2f  OQ  2400  Read   2.0  3.9  7.8   12   18   25   33   33   33   33   33    33
  yDiskUSB2f  OQ  2400   DMA   2.0  3.9  7.7   13   18   26   30   33   33   33   33    33

  64-Bit Vista - 2007 on-board - Freecom disk - Write O/H 0.5 ms
  yDiskUSB2f  OP  2400 Write   1.9  3.8  7.6   12   18   22   26   26   26   26   26    26
  yDiskUSB2f  OP  2400  Read   2.0  3.9  7.8   13   18   25   33   33   33   33   33    33
  yDiskUSB2f  OP  2400   DMA   2.0  3.9  7.7   13   18   26   33   33   33   33   33    33

  64-Bit Vista - 2007 on-board - Freecom disk - 64 Bit Version
  yDiskUSB2f  OQ  2400 Write   1.0  1.9  3.7    7   10   16   21   23   24   25   25    25
  yDiskUSB2f  OQ  2400  Read   2.0  3.9  7.7   12   18   25   33   33   33   33   33    33
  yDiskUSB2f  OQ  2400   DMA   1.9  3.9  7.8   13   18   26   33   33   33   33   33    33

  64-Bit Vista - 2007 on-board - Freecom disk - 64 Bit Version
  yDiskUSB2f  OP  2400 Write   1.9  3.8  7.6   12   17   22   26   26   26   26   26    26
  yDiskUSB2f  OP  2400  Read   2.0  3.9  7.8   13   18   25   33   33   33   33   33    33
  yDiskUSB2f  OP  2400   DMA   1.9  3.9  7.8   13   18   26   33   33   33   33   33    33

  64-Bit Vista SP1 - 2008 on-board - Freecom disk - 64 Bit Version
  yDiskUSB2f  OP  2400 Write   1.9  3.8  6.2   10   15   21   25   25   26   26   26    26
  yDiskUSB2f  OP  2400  Read   1.9  3.9  6.2   12   18   24   33   33   33   33   33    33
  yDiskUSB2f  OP  2400   DMA   1.9  3.9  6.3   12   18   25   33   33   33   33   33    33

  64-Bit Vista SP1 - on-board - FAT - WD Passport Disk 2009
  yDiskUSB2g  OQ  2400 Write  0.89  1.8  2.8  5.3  9.3   14   19   22   24   20   17    24
  yDiskUSB2g  OQ  2400  Read   2.5  4.6  6.9   11   17   21   26   26   25   26   25    26
  yDiskUSB2g  OQ  2400   DMA   2.5  5.0  7.8   14   21   28   33   33   33   33   33    33

  64-Bit Vista SP1 - on-board - NTFS - WD Passport Disk 2009
  yDiskUSB2g  OQ  2400 Write   2.5  3.8  7.5   12   17   21   25   23   24   24   24    25
  yDiskUSB2g  OQ  2400  Read   2.5  4.7  7.7   14   20   27   33   31   31   33   31    33
  yDiskUSB2g  OQ  2400   DMA   2.5  5.0  7.8   13   21   28   33   33   33   33   33    33

  64-Bit Windows 8.1 - on-board USB 3 - NTFS -  Seagate Expansion  SRD00F1 Disk 2014
  yDiskUSB3a  OQ  3700 Write    11   21   28   53   84   99   98   99  100   90  101   101
  yDiskUSB3a  OQ  3700  Read    12   24   31   57   87  111  112  111  112  108  110   112
  yDiskUSB3a  OQ  3700  DMA      7   14   26   48   80  123  161  191  217  229  232   232

  64-Bit Windows 8.1 - via USB 2 - NTFS -  Seagate Expansion  SRD00F1 Disk 2014
  yDiskUSB3a  OQ  3700 Write     4    7    9   14   19   23   25   27   27   29   29    29
  yDiskUSB3a  OQ  3700  Read     4    8   10   16   25   26   31   33   33   34   34    34
  yDiskUSB3a  OQ  3700   DMA     3    5    8   13   21   25   29   32   33   34   34    34

  

To Start     Results/Examples Index



USB and Firewire Disk Drives

The disk used for these tests is a 120 GB WD 1200JB, IDE, 7200 RPM capable or running at 748 Mbits/sec and mainly in a Hiyatec enclosure with USB and Firewire connections. Other results are via a Sohouse IDE/USB adapter and some for an earlier system (different disk).

Earlier Systems - Limited results for a PC from 2003 showed 16/24 MB/second writing/reading speed for USB 2, 19/33 MB/second for Firewire and DMA reading speed results shown below, where Firewire is again faster. Next details are for a 2003 Laptop where USB 1 is shown to be significantly slower than Firewire and the latter as fast as some modern systems.

Same Disk, Same PC, Different Interfaces - XP x64 2210 MHz has two Firewire sockets. FW1 is on a Creative sound card and FW2 from the Motherboard, the latter being up to 15% faster on writing and mainly 28% faster on reading. Then there are the Hiyatek and Sohouse USB interfaces on the 1830 MHz laptop and 2400 MHz desktop. Here, Hiyatek is up to 50% faster. Best gains tend to be at the largest block sizes.

Same Disk, Same Adapter, Different PCs - There are significant differences in speeds on the three PCs tested (like up to 35% on writing) and it is not all down to CPU MHz.

Same Disk, Same PC, Same Adapter, USB and Firewire Connections - The best Firewire speed advantages are on the 1830 MHz laptop, more than twice as fast as USB at small block sizes, 10% faster on reading with large block sizes and similar speed writing large blocks. The 2210 MHz desktop shows the same gains on reading but lower gains on writing small blocks with USB 20% faster writing with large block sizes. Other observations to note are faster maximum DMA reading speeds of 39 to 40 MB/second with Firewire and lower CPU utilisation for both writing and reading.



                    All Results FAT Format, Optimise For Quick Removal

                   CPU                  MBytes/Second at KB Block Size
                   MHz  Mode     1    2    4    8   16   32   64  128  256  512 1024   Max


  USB v Firewire - WinXP 2003 
  yDiskUSB2e      3000   DMA   1.4  2.7  5.3    9   14   19   24   24   25   23   25    25
  yDiskFwire      3000   DMA   2.3  4.4  8.0   14   20   27   33   37   38   39   30    39

  Laptop - WinXP 2003
  Hiyatek         1800 Write  0.93  4.0  6.4   10   15   20   24   25   26   27   27    27
  XP FW           1800  Read   5.8   10   15   17   24   32   33   33   33   33   34    34
  AMD CPU         1800   DMA   5.6   11   17   24   29   35   37   38   39   20   30    39

  Hiyatek         1800 Write   0.1  0.2  0.3  0.5  0.6  0.8  0.9  0.9  0.9  0.9  0.9   0.9
  XP USB 1        1800  Read   0.3  0.5  0.7  0.8  0.9  1.0  1.0  1.0  1.0  1.0  1.0   1.0
  AMD CPU         1800   DMA   0.3  0.5  0.7  0.8  0.9  1.0  1.0  1.0  1.0  1.0  1.0   1.0

  Same Disk, Various Systems
  Hiyatek         2210 Write  0.96  3.3  6.0  9.4   14   18   21   22   23   23   23    23
  XP x64 FW1      2210  Read   5.8   10   13   17   17   25   26   27   27   27   27    27
  AMD dualcore    2210   DMA   6.4   11   16   20   24   26   27   28   28   20   24    28

  Hiyatek         2210 Write  0.97  3.8  6.7   11   15   20   23   24   25   25   25    25
  XP x64 FW2      2210  Read   6.4   10   17   18   25   33   33   35   34   34   34    35
  AMD dualcore    2210   DMA   7.3   12   19   26   32   36   39   40   40   20   30    40

  Hiyatek         2210 Write  0.94  2.2  3.7  7.3   12   18   26   31   32   33   33    33
  XP x64 USB      2210  Read   2.5  4.8  7.3   13   17   27   31   32   32   32   32    32
  AMD dualcore    2210   DMA   2.6  5.2  7.8   17   21   28   34   35   15   30   28    35

  Hiyatek         1830 Write  0.94  3.8  5.9  9.7   14   18   21   23   24   24   24    24
  Vista 32 FW     1830  Read   5.7  8.7   13   17   24   31   33   31   34   35   35    35
  Core 2 Duo Mob  1830   DMA   4.6  8.7   16  4.8   30   33   35   39   39   20   30    39

  Hiyatek         1830 Write  0.78  1.4  2.8  5.2  9.0   14   19   22   24   24   24    24
  Vista 32 USB    1830  Read   2.1  3.2  6.3   10   17   25   30   31   31   31   32    32
  Core 2 Duo Mob  1830   DMA   1.9  3.3  6.8   10  9.1   26   32   32   32   20   29    32

  Hiyatek         2400 Write  0.95  1.9  3.5  6.3   11   16   21   23   24   25   24    25
  Vista 64 USB    2400  Read   2.4  4.7  7.2   13   17   27   31   31   32   31   32    32
  Core 2 Duo      2400   DMA   2.6  4.1  7.8  4.7   16   28   33   33   33   20   29    33

  Sohouse         1830 Write  0.69  1.3  2.4  4.2  7.0   10   14   15   16   16   16    16
  Vista 32 USB    1830  Read   1.6  3.1  5.8  8.5   13   18   21   21   22   22   22    22
  Core 2 Duo      1830   DMA   1.5  3.2  6.1  8.9   14   18   22   22   22   20   20    22

  Sohouse         2400 Write  0.95  1.6  3.0  5.3  8.6   12   15   16   17   17   17    17
  Vista 64 USB    2400  Read   2.5  4.7  7.3   10   15   20   23   23   23   23   23    23
  Core 2 Duo      2400   DMA   2.6  5.1  7.8   13   14   21   24   24   24   20   20    24

  

To Start     Results/Examples Index



Local Area Network Speed

Following are examples of performance of 10 and 100 Mbits/second Local Area Networks. These were via an older version of the benchmark which, as indicated elsewhere, was slow on writing. So, an example of writing speed via the latest version is shown. DMA speeds were generally the same as reading speeds, except for the example (Win98) shown, where data is cached in the local system’s memory. Measured CPU utilisation on LAN tests can be greater than 30% on the 1900 MHz CPU but it might indicate polling time that could be used by other processes, if needed.



                  CPU Speed           MBytes/Second at KB Block Size
  Key             MHz  mbps     1    2    4    8   16   32   64  128  256  512 1024  Max

  Write
  zLAN10mbs LAN   450    10  0.72 0.72 0.72 0.73 0.85 0.92 0.96 0.97 0.96 0.96 0.94  0.97
  zLAN100m1 LAN  1900   100  0.90  1.5  2.5  2.9  4.7  6.0  6.5  6.7  6.8  6.8  6.2   6.8
  zLAN100m2 LAN   750   100  0.66  1.1  1.7  1.9  3.0  4.8  6.2  6.5  6.4  6.4  6.5   6.5
  zLAN100m3 LAN  1900   100  0.68  1.1  1.7  1.9  3.1  4.6  6.2  6.4  6.3  6.2  6.5   6.5
  Later          1900   100   1.1  2.1  2.9  5.6  6.9  7.5  8.5  9.2 10.2 10.4 10.5  10.5
  zWiFi56        1830    56   0.6  0.9  1.4  1.8  2.1  2.3  2.5  2.6  2.7  2.8  2.8   2.8

  Read
  zLAN10mbs LAN   450    10  0.70 0.68 0.71  0.8  0.9  1.0  1.0  1.0  1.0  1.0  1.0   1.0
  zLAN100m1 LAN  1900   100   3.2  4.7  6.7  7.7  8.7  9.3  9.4  9.6  9.7  9.8  9.8   9.8
  zLAN100m2 LAN   750   100   3.4  4.6  7.0  8.0  9.2  9.9 10.0 10.2 10.4 10.5 10.5  10.5
  zLAN100m3 LAN  1900   100   3.4  4.6  7.0  8.0  9.2  9.9 10.0 10.2 10.4 10.5 10.5  10.5
  zWiFi56        1830    56   0.5  0.8  1.4  1.6  2.2  2.5  2.7  3.0  3.0  3.0  2.9   3.0

  DMA
  zLAN10mbs LAN   450    10    54  104  196  245  309  363   87  136  131  135  126   363
  zWiFi56        1830    56   0.6  0.9  1.6  1.9  2.4  2.5  2.8  2.9  2.9  3.0  3.0   3.0
 


To Start     Results/Examples Index



USB, Compact Flash and Secure Digital

The results for this benchmark are for old Compact Flash Memory cards which restrict the maximum reading speed. The following show speeds via a reader on an LP Port, PCMCIA on laptops and different memory cards via USB 1 and High Speed USB 2. See CDDVDSpd Results.htm for a more appropriate benchmark with results on faster cards.

The last results shown are for a USB1 to USB1 LAN type communications box.

In order to demonstrate similar characteristics, new results have been added for SD cards used with a camera. The speeds can be compared with those for 4 GB Flash Drives below.



              CPU   Device          MBytes/Second at KB Block Size
  Key         MHz               1    2    4    8   16   32   64  128  256  512 1024  Max

  Write
  zCFlashLP   950  LP Port                  0.15 0.18 0.21 0.23 0.23 0.24 0.26 0.27  0.27
  zCFPcmcia   400   Pcmcia   0.28 0.35 0.40 0.42 0.44 0.45 0.45 0.45 0.45 0.45 0.46  0.46
  zCFPcmci2  1800   Pcmcia   0.07 0.13 0.20 0.28 0.36 0.41 0.45 0.47 0.48 0.49 0.50  0.50
  zUSB1CF    1900     USB1   0.05 0.08 0.13 0.21 0.29 0.36 0.40 0.43 0.44 0.45 0.45  0.45
  zUSB1CF    1900     USB1   0.03 0.05 0.09 0.16 0.24 0.32 0.39 0.44 0.44 0.46 0.50  0.50
  zUSB2CF    1900     USB2   0.05 0.09 0.14 0.22 0.30 0.37 0.42 0.44 0.46 0.47 0.47  0.47
  zUSB2CF    1900     USB2   0.03 0.06 0.11 0.18 0.27 0.39 0.45 0.50 0.54 0.55 0.62  0.62
  zUSBComms  1900     USB1        0.12 0.15 0.25 0.39 0.42 0.50 0.54 0.57 0.57 0.59  0.59

  zSD32MB    2400     USB2   0.03 0.06 0.12 0.25 0.52 0.73 0.89 1.00 1.10 1.10 1.20  1.20
  zSD2GB     2400     USB2   0.13 0.27 0.52 0.98 1.90 3.40 5.40 7.70 9.70 9.30 8.90  9.70


  Read
  zCFlashLP   950  LP Port                   0.1  0.1  0.1  0.1  0.1  0.1  0.1  0.1   0.1
  zCFPcmcia   400   Pcmcia    0.7  0.9  1.1  1.2  1.2  1.3  1.3  1.3  1.3  1.3  1.3   1.3
  zCFPcmci2  1800   Pcmcia    0.8  1.1  1.3  1.5  1.6  1.6  1.6  1.6  1.7  1.7  1.6   1.7
  zUSB1CF    1900     USB1    0.1  0.2  0.4  0.6  0.7  0.8  0.9  0.9  0.9  0.9  0.9   0.9
  zUSB1CF    1900     USB1    0.2  0.3  0.4  0.5  0.7  0.8  0.8  0.8  0.8  0.8  0.8   0.8
  zUSB2CF    1900     USB2    0.3  0.5  0.8  1.1  1.3  1.4  1.5  1.5  1.5  1.5  1.5   1.5
  zUSB2CF    1900     USB2    0.6  1.0  1.4  1.6  1.8  1.9  1.9  1.9  1.9  1.9  1.9   1.9
  zUSBComms  1900     USB1         0.1  0.2  0.3  0.4  0.4  0.4  0.5  0.5  0.5  0.5   0.5

  zSD32MB    2400     USB2    1.1  1.7  2.3  2.8  3.2  3.4  3.6  3.6  3.6  3.5  3.6   3.6
  zSD2GB     2400     USB2    1.1  2.0  3.6  6.1  8.9 11.8 16.9 16.9 17.0 16.9 17.0  17.0


To Start     Results/Examples Index



4 to 16 GB USB Flash Drives and More

The benchmark was run on two 4 GB Flash Drives with FAT and NTFS formatting, using both 32 bit and 64 bit benchmarks via Windows XP Pro x64 and 64-Bit Vista, plus the 32 bit version using Windows 2000 (Systems see Key jC2Dsea2, q64AMax1 and oThoHi1).

Results via Vista and x64 were essentially the same as were those for the 32 bit and 64 bit benchmarks. So, the following are example results and some of these show the same performance. The Windows 2000 system is the only one where the optimise for performance setting works. This leads to faster writing speeds with FAT formatting and small block sizes. NTFS has the same impact. The second flash drive is clearly much faster, except reading with small block sizes. The benchmark was also run using 1 GB file size to measure random access times (see below). These were quite consistent on the second drive.

The first drive was found unsuitable for Vista ReadyBoost, reported as insufficient write performance at 1723 KB/sec. The minimum requirements for ReadyBoost are 2.5 MB/sec for 4 KB random reads and 1.75 MB/sec for 512K random writes. The second drive, with NTFS format, was reported as reading at 5486 KB/sec and writing at 1814 KB/sec.

The reading speed makes sense, compared with these benchmark results. Random reading of 1 KB at 0.52 milliseconds is at a speed of 1.92 MB/sec, virtually the same as that for the benchmark reading 1 KB blocks, with 5.4 MB/sec at 4 KB. On the other hand, it is difficult to see where the 1814 KB/sec comes from (particularly if it is really for 512 KB) with this benchmark indicating greater than 3 MB/sec. My other benchmark (see CDDVDSpd Results.htm) shows that individual small files can be written faster than the ReadyBoost measurements. I suppose that one way to achieve the low speeds would be to produce a fragmented data space where writing 512 KB fills in the spaces, effectively as small block sizes.

Results added for zUSB3CF, a 16 GB USB 3 Stick.



                                          MBytes/Second at KB Block Size
  Key                Win      1     2     4     8    16    32    64   128   256   512  1024

  Write
  zFD4096A     FAT    2K   0.21  0.37  0.46  0.52  0.59  2.60  2.60  2.60  2.60  2.60  2.50
  zFD4096A     FAT    64   0.07  0.10  0.14  0.24  0.30  1.80  2.40  2.60  2.60  2.50  2.40
  zFD4096A    NTFS    64   0.21  0.37  0.47  0.54  0.59  2.50  2.50  2.70  2.70  2.60  2.60

  zFD4096C     FAT    2K   0.62  0.97  1.40  2.20  2.90  3.80  4.40  4.50  4.50  4.30  4.40
  zFD4096C     FAT    64   0.17  0.32  0.56  1.00  1.80  2.70  3.60  4.10  4.40  4.50  4.50
  zFD4096C    NTFS    64   0.61  0.95  1.40  2.30  3.40  4.20  4.80  4.00  3.90  3.60  3.30

  zUSB3CF     NTFS     8      6    14    21    58    80   108   108    69    95    78    47
  zUSB3CF      FAT     8      7    15    31    59    80   105   112   114   114   111   116

  Read
  zFD4096A     FAT    2K   2.20  3.80  5.20  6.80  7.70  8.60  9.10  9.20  9.20  9.20  9.20
  zFD4096A     FAT    64   1.90  3.90  5.20  6.90  7.80  8.60  9.10  9.10  9.20  9.20  9.20
  zFD4096A    NTFS    64   1.90  3.90  5.20  6.80  7.80  8.70  9.10  9.20  9.10  9.20  9.20

  zFD4096C     FAT    2K   1.90  3.10  5.40  8.80 11.40 14.60 16.60 16.60 16.60 16.60 16.60
  zFD4096C     FAT    64   1.90  3.10  5.40  8.90 12.40 14.60 16.50 16.30 16.30 16.80 16.60
  zFD4096C    NTFS    64   1.90  3.10  5.40  8.90 12.30 14.60 16.50 16.60 16.50 16.60 16.50

  zUSB3CF     NTFS     8      8    15    28    51    86   131   176   213   237   253   254
  zUSB3CF      FAT     8      8    15    29    51    87   130   177   213   238   254   255

  Random reading milliseconds 1 KB from 1 MB to 1024 MB data

  zFD4096A     FAT    64   0.51  0.50  0.50  0.51  0.51  0.50  0.50  0.50  0.89  1.61  2.39
  zFD4096A    NTFS    64   0.65  0.54  1.31  1.70  1.27  1.17  0.84  0.71  0.59  2.18  2.22

  zFD4096C     FAT    64   0.51  0.52  0.52  0.51  0.52  0.53  0.54  0.57  0.54  0.61  0.61
  zFD4096C    NTFS    64   0.52  0.52  0.56  0.55  0.52  0.53  0.54  0.58  0.60  0.62  0.66

  zUSB3CF     NTFS     8   0.24  0.26  0.31  0.32  0.32  0.33  0.34
  zUSB3CF      FAT     8   0.29  0.29  0.32  0.33  0.34  0.34  0.35

To Start     Results/Examples Index



DMA Problems

Following are three examples of DMA found not to be working on running the benchmark and results after enabling DMA. The slow PIO mode is identified by CPU utilisation approaching 100%. The results suggest that the relative performance loss without DMA has become progressively worse. The fourth example shows variations on the same PC when booting to Windows 98, which uses UDMA33, and Windows XP at UDMA100. Disk reading speed is shown but there were similar differences on writing.



                  CPU  Disk            MBytes/Second at KB Block Size                          % CPU Utilisation
  Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024     1   2   4   8  16  32  64

  K63WD           400  5400  1.4  1.4  1.6  2.4  2.7  3.9  4.3  3.3  3.7  3.8  4.0    87  87  87  79  75  75  66
  K63WD     DMA   400  5400  4.4  5.8  5.9  6.0  6.1  6.3  6.9  8.0  8.0  8.1  8.0    65  72  66  64  63  65  69

  eCelMax1a       950        3.3  3.9  4.4  4.6  4.8  4.8  5.0  5.0  5.0  5.0  5.0    80  92  93  95  97  93  93
  eCelMax1a DMA   950        7.5   12   18   23   26   28   29   30   30   30   30    54  41  26  17  10   5   2

  gP3IBM1        1000  7200  2.1  2.3  2.3  2.3  2.4  2.3  2.3  2.3  2.4  2.4  2.4    96  96  94  95  97  94  95
  gP3IBM1  DMA   1000  7200   11   23   32   31   32   32   32   32   32   31   32    64  71  49  25  12   6   3

  Win98 UDMA33 WinXP UDMA 100 
  oPalWD2  W98   1466  7200  9.3   13   16   18   21   20   20   19   20   24   26    45  29  17   9   5   2   1
  oPalWD2  W2K   1466  7200   11   16   24   23   24   43   44   43   44   44   43    23  19  20   8   5   4   2

Some excessive DMA speeds can be produced when Norton Anti-Virus is in use. It seems that this software reads the first part of a file using the Windows file caching option. Then the DMA test is repetitively reading the data from memory. The following examples show this effect up to 16 KB block size.



                   CPU  Disk            MBytes/Second at KB Block Size                    % CPU Utilisation
  Key              MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024    1   2   4   8  16  32  64

  eCelNA2     AV   900         50   93  167  234   51   56   59   58   22   23   23  100 100 100 100  76  42  29
  eCelMax2    AV   700         59  105  172  286  382   50   40   34   32   20   24  100 100 100 100 100  14   7

To Start     Results/Examples Index



Driver Problems

The following shows results of an experiment in comparing Microsoft and Intel bus drivers. Both produce a strange reduction in performance at large block sizes but in different places.

Next is an example of how poor performance was corrected by re-installing the bus driver.



                  CPU  Disk            MBytes/Second at KB Block Size
  Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

  Write
  ip4WD2  Intel  2411  7200  1.0    6   10   18   28   40   51   30   63   64   65
  ip4WD2    MS   2411  7200  1.0    6    9   18   28   40   36   35   63   64   65
  Read
  ip4WD2  Intel  2411  7200  9.5   19   37   41   42   42   42   19   25   22   28
  ip4WD2    MS   2411  7200  9.5   19   37   41   42   42   43   43   43   28   35

  Driver Reloaded
  nTbdMax1  Wr1  1533  5400  1.4  2.6  5.6  6.7  9.3   10   10   11   11   11   11
  nTbdMax1  Wr2  1533  5400  1.5  3.0  5.4   11   19   24   26   26   27   27   21

This example demonstrates different speeds on a board where the disk can be plugged into a RAID or IDE connector.



                  CPU  Disk            MBytes/Second at KB Block Size
  Key             MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

  oPalIBM3 RAID  1539  7200  6.6   13   16   19   21   26   23   26   26   27   24
  oPalIBM3  IDE  1539  7200  6.6   12   22   36   37   37   37   38   37   40   40
  

To Start     Results/Examples Index




CD and DVD Writers

The benchmark can also be used for testing CD/DVD drives with RW discs when packet writing software is available, such as Nero InCD. Following are example results on a DVD drive. Note the variations in speed and the high CPU utilisation.


 Pentium 4 1.9 GHz, Liteon LH-20A1H DVD +RW 8x, DVD Disc Rating ??, 
 using Nero InCD, WinXP

 ###########################################################
     Disk Test DiskGraf Version 1.3A Sun Jun 24 16:33:49 2007
 
              Copyright Roy Longbottom 2001

  Windows NT Version 5.1, build 2600, Service Pack 1
  CPU GenuineIntel, Features Code 3FEBFBFF, Model Code 00000F12, 1900 MHz
  Memory 523760 KB, Free 250432 KB
  F:Disk/partition 4479 MB, Free 4478 MB
  Files 5 Size 8 MB Block Sizes 10 Minimum RAM Data

 Block   File ->
   KB    1     2     3     4     5 
        Write MB/Second
    2   2.1   3.4   1.7   3.4   3.4
    4   2.1   1.7   3.4   3.4   3.4
    8   1.3   3.4   3.4   3.4   1.8
   16   2.2   3.4   3.4   1.8   3.4
   32   2.2   3.4   1.6   3.4   3.4
   64   2.0   1.8   3.4   3.4   3.4
  128   1.3   3.4   3.4   3.4   1.7
  256   2.0   3.4   3.4   1.7   3.4
  512   2.1   3.4   1.7   3.4   3.4
 1024   2.0   1.7   3.3   3.3   3.3
        Read  MB/Second
    2   2.6   3.2   3.2   2.6   5.1
    4   2.6   3.2   3.2   2.6   5.1
    8   2.6   3.2   3.2   2.6   5.1
   16   2.6   3.2   3.2   2.6   5.1
   32   2.5   3.2   3.2   2.7   5.1
   64   2.6   3.2   3.2   2.7   5.1
  128   2.6   3.2   3.3   2.6   5.2
  256   2.6   3.2   3.2   2.7   5.2
  512   2.6   3.2   3.3   2.7   5.2
 1024   2.6   3.2   3.2   2.7   5.1
        Write CPU Utilisation
    2  55.1  85.2  45.1  85.0  84.6
    4  32.5  28.0  47.7  48.1  47.9
    8  12.0  28.7  28.7  28.4  18.4
   16  13.2  20.0  20.5  12.1  20.2
   32  10.4  15.9   8.6  15.6  15.4
   64   7.6   9.3  13.3  12.7  13.3
  128   4.8  11.4  11.4  11.7   8.0
  256   5.5  10.3  10.5   5.9  10.9
  512   5.7   9.4   5.4   9.8   9.5
 1024   5.9   5.0   7.8   8.2   8.5
        Read  CPU Utilisation
    2  41.0  47.6  47.4  38.2  74.1
    4  35.1  40.8  39.9  33.3  63.7
    8  29.9  34.3  34.0  27.4  53.8
   16  26.0  30.6  30.9  26.3  48.6
   32  24.1  28.7  28.5  23.6  46.4
   64  23.9  27.5  27.5  22.7  44.5
  128  23.4  26.5  26.4  21.7  43.8
  256  23.3  29.4  27.1  22.3  43.7
  512  21.5  26.5  26.1  21.5  42.8
 1024  21.7  26.1  26.6  21.2  42.4

 DMA Reread block up to 100 times at 1 to 1024 Kbytes

 Read  MB/Second
   3.1  10.5  24.7  39.9  56.7  72.7  75.4  72.1  75.1  74.6  76.4
 Read  CPU Utilisation
 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

 Random Read 1KB out of 1, 2, 4, to file size Mbytes

 Read  Milliseconds
  5.86  19.6  40.3  14.9 0.000 0.000 0.000 0.000 0.000 0.000 0.000
 Read  CPU Utilisation
   6.0   2.3   2.5   2.2   0.0   0.0   0.0   0.0   0.0   0.0   0.0

               End Disk Test Sun Jun 24 16:39:13 2007
  

To Start     Results/Examples Index



Disk Results

The following results include some for the 64 bit version as well at those at 32 bits - q64AMax1 via Windows XP Pro x64 and jC2DSea2 via 64 Bit Widows Vista. Results for the latter are also provided for the two 200 GB partitions C: and D:. A surprise is that speed on the smaller block sizes is much slower on the first partition (C:). The same was apparent on a PC with Windows XP. Perhaps there is a Windows overhead checking data transfers on the System partition (there was no Anti-Virus software in use).


Writing Speed


               CPU  Disk            MBytes/Second at KB Block Size                          % CPU Utilisation
Key            MHz   RPM     1    2    4    8   16   32   64  128  256  512 1024  Max     1   2   4   8  16  32  64

a486DX2          66  3600           0.35 0.38 0.54 0.71 0.94  1.1 0.94 0.97  1.1    1            22  19  26  33  44
bPenSea         100  5400 0.95  1.2  1.5  1.5  1.6  1.7  1.7  1.8  1.7  1.8  1.5    2    84  82  87  87  90  94  95
dPProQua  #     200  4500                0.17 0.33 0.59  1.0  1.5  2.1  2.7  3.6    4                 4   6  10  16
fPIILap         300  5400  2.5  4.8  5.4  5.4  5.4  5.2  5.4  5.4  5.2  5.4  5.4    5    23  22  13   8   4   3   2
dPProQua        200  4500  1.2  2.3  4.1  4.8  5.4  5.8  5.8  5.8  5.8  5.3  5.3    6    36  52  77  83  88  90  88
eCelLap         400  4200  2.5  3.6  4.1  4.8  5.1  5.1  5.2  5.4  5.4  6.1  6.1    6    23  14   9   5   3   2   2
K62Sea2         380  5400  2.4  2.9  3.5  3.4  3.4  3.5  3.5  3.5  3.5  5.6  6.3    6    29  18  12   8   7   6   5
cPMMXQua        200  5400  1.3  2.5  4.8  7.4  7.3  7.3  7.4  7.4  7.1  7.1  7.1    7    29  42  50  81  74  71  72
gP3Max    #    1400        1.2  2.4  4.1  6.3  7.3  9.4  9.6  9.8   11   11   11   11    18  16  13  11   7   5   2
nTbdQuaRd IR#  1400  7200 0.15 0.27 0.47 0.90  1.6  3.1  5.0  9.0   11   11   12   12    12  12  12  13  13  13  13
pBarLap   #    1800  4200 0.68  1.2  2.2  3.6  5.8  8.1  8.7  3.1  5.3  8.4   12   12     6   5   5   4   3   2   0
eCelMax         450  7200  2.4  4.8  7.8   11  9.2   12   12    9   12   12   12   12    21  22  15  12   5   4   2
fPIIFuj         450  5400  2.5  4.7  6.7   11   12   12   12   12   12   12   12   12    22  21  15  12   7   4   2
oPalMaxRd $IR# 1667  7200            0.2  0.3  0.6  1.2  2.0  3.1  6.7  8.4   12   12     D
gPIIILap        650        7.6   10   12   12   13   13   12   13   12   12   12   13    41  32  19  11   6   3   2
iP4Sea2   #    1900  5400  1.1  1.8  2.7  4.3  7.0   10   11   13   13   15   15   15    16  14  14  11   7   4   1
K62Sea    #     500  7200  1.1  2.1  3.1  6.1  5.7  6.6  6.5   10   14   15   17   17    63  58  46  33  20  11   7
gp3SeaA   #    1400  7200  1.3  2.4  4.2  5.5  8.5   12   13   16   16   16   17   17    15  14  10   9   7   4   3
fPIISea1  SC    450 10000  7.9   14   17   17   18   17   17   17   17   18   17   18    74  58  39  22  14   8   6
iP4NLap   #    1700  5400 0.75  1.4  2.8  4.7  9.0   15   18    5    8   12   16   18     8   8   8   7   6   5   1
iP4Sam    #    3066  5400 0.52  1.1  1.9  3.9  8.6   12   14   14   18   16   14   18     3   2   2   3   2   1   1
eCelMax2  #     700        1.7  3.0  5.0  7.2  8.9   10   11   12   17   18   17   18    30  27  25  16  10   6   3
K63WD           400  5400 0.18 0.36   15   16   16   16   16   16   16   16   18   18     1   1  20  13   8   6   6
oPalSea3  SC#  1667 15000           0.66  1.4  2.5  4.9   10  8.9   14   16   18   18     D
oPalIBM2  #    1466  7200 0.67  1.6  3.0  5.8  9.4   13   17   16   18   18   16   18     6   7   7   7   5   3   2
eCelMax1        950  5400  6.6   11   14   16   18   18   19   18   19   19   19   19    36  31  20  13   8   5   4
oPalMax1  #    1667  7200  1.1  2.3  3.2  5.1   10   18   18   17   16   17   19   19     D
eCelMax2        700        7.4   11   15   17   18   19   19   19   19   19   19   19    40  35  23  16  10   7   6
iP4Sea2        1900  5400  3.6  6.4   10   18   20   21   21   21   21   21   21   21    24  20  17  15   7   4   1
eCelNA2         900        5.3  9.9   16   20   22   22   22   22   22   22   22   22    58  55  46  31  19  13   9
oPalQuaS  SC#  1533 10000           0.26 0.49  1.1  1.8  4.1  6.2   10   16   22   22     D
mDurMax2        950  7200  4.2    8   11   18   22   24   23   24   24   24   24   24    36  37  24  20  13   8   9
iP4Sea3   #    2533  7200  2.1  3.9  6.2  8.9   12   18   23   23   22   23   27   27    16  17  15  10   6   3   1
nTbdMax1  #    1533  5400  1.5  3.0  5.4   11   19   24   26   26   27   27   21   27    10  10   9   8   8   3   1
iP4Sea1   #    2016  7200  1.9  3.1  4.4  7.6   13   19   20   22   21   22   27   27    30  36  26  27  22  15   7
oPalWD1   #    1410  7200  1.4  5.5   11   18   26   25   23   27   28   27   28   28    19  29  45  39  27  14   6
oPalIBM3       1539  7200  6.8   13   18   22   25   27   28   28   28   28   28   28    41  37  27  15   9   5   2
nTbdWDI1  #    1300  7200  1.7  6.0   13   20   28   26   25   29   29   28   28   29    17  39  43  30  22  10   5
eCelWD1         466  7200 0.96   10   17   21   24   27   28   28   29   30   29   30    11  52  42  28  21  15  11
oPalSea2 %SC#  1700  7200  2.0  4.0    7   11   12   18   20   19   23   23   30   30    19  23  20  14   6   5   1
iP4Xeon   #    2175        1.8  3.6  6.3   10   14   22   24   24   25   24   30   30     D
iP4Sea4   #    2533  7200  1.7  2.1  3.9  6.7   14   21   25   28   31   33   31   33    14   9   6   5   5   4   2

jAtoLap   SA   1600  5400  4.2  8.2   15   25   32   32   34   34   34   34   34   34    11  10   9   6   6   3   1

q64ASea1  #    2150  7200  1.8  2.5  3.7  8.2   14   16   21   27   30   33   34   34     8   6   4   5   4   2   1
nTbdQua        1200  7200  3.8  7.5   15   29   36   36   35   35   35   35   35   36    14  14  14  14   6   4   2
gP3Sea15  %SC#  667 15000 0.23 0.45 0.88  1.7  3.6  5.7   11   12   18   27   36   36     D
gP3WD1    #    1000  7200  1.7  5.9   14   24   35   34   32   36   34   37   36   37    12  30  36  25  27  11   6
iP4EWD1   SA   3000  7200  8.0   13   16   19   22   24   24   26   30   33   37   37    11  10   5   3   1   1   0
iP4WD3    #    2533  7200 0.75  4.1   12   21   31   34   34   34   34   37   38   38     6  16  17  27  19   8   3
nTbdIBM8  #    1400  7200  1.0  2.3  5.0   10   17   19   22   24   33   38   30   38    15  17  16  19  14   8   5
q64TLap  D:SA  1900  5400  8.6   15   25   32   34   38   36   37   37   37   33   38     D
gP3RAID   IR#  1000       0.95  1.8  3.5  6.2   11   18   24   29   35   39   39   39     D
oThoWD3   #    2300  7200 0.78  5.6   12   20   33   32   35   38   39   39   38   39     5  13  18  10  14   7   3
oPalWD2   #    1466  7200 0.86  6.8   13   24   35   34   36   40   39   40   38   40     8  23  24  17  20  10   5
gP3IBM1        1000  7200  7.6   15   27   41   41   41   41   41   41   41   41   41    49  48  46  26   8   5   3
jC2CLap    SA  2000  5400    3    4    7   10   20   24   34   34   44   37   34   44    18  17  18  16   8   5   3
oPalSeaR0 %SC# 1700 15000  1.4  2.4  3.5  6.1   10   15   25   35   42   45   45   45    15  14  14  11   8   5   4
oPalSeaRd %SR# 1700 15000  1.6  2.9  5.4  9.0   15   22   30   35   41   44   46   46    16  15  14  11   9   6   4
iP4Hit1        1900  7200  6.2   12   21   36   43   44   42   47   45   44   44   47    44  42  40  32  14   8   3

jC2DLap    SA  1830  5400    4    8   15   26   44   42   47   45   44   29   44   47     D
jC2DSea2 C:SA  2400  7200    6    8   14   17   25   35   38   42   46   46   52   52     D
jC2DSea2 C:SA  2400  7200    5    8   10   16   24   30   19   30   42   49   35   49  D&14   9   6   4   1   1   0
jC2DSea2 D:SA  2400  7200   10   20   40   47   47   47   52   51   51   50   49   52     D
jC2DSea2 D:SA  2400  7200    9   16   31   36   38   45   45   44   45   44   43   45  D&21  19  13  10   0   2   0

q64ASeaR  SR#  1991 15000 0.33 0.62  1.2  2.4  4.3  7.3   13   21   29   32   52   52     D
oThoHi1        2080  7200  6.5   12   22   36   45   52   47   46   47   47   46   52    16  15  14  10   5   2   1
iP4Max         2400  5400   10   21   38   51   52   52   52   52   52   49   53   53    45  47  37  27  10   4   1
K62IBM          267  7200  5.8   11   19   28   38   46   50   46   52   53   54   54    85  91  91  84  66  50  40
nTbdIBMRd IR#  1400  7200  1.4  2.5  3.7  7.3   12   24   34   44   46   54   52   54    15  17  13  13  11  10   8
jC2DSea1   SA  2400  7200  5.4  8.2   15   17   27   35   38   43   48   54   57   57     D
q64AMax1  &SA  2210  7200   17   29   44   46   51   55   59   55   61   58   59   61  D&23  18  11   4   3   1   1
q64AMax1  SA   2210  7200   17   29   45   58   54   61   60   54   59   61   62   62     D
oPalIBMRd $#   1396  7200  1.3  2.7  5.2  8.7   19   24   40   54   48   62   62   62    21  20  21  19  16  21  18
iP4WD1    #    2411  7200 0.88  6.2   13   25   37   47   55   60   62   63   64   64     5  23  28  27  20  13   7
ip4WD2    #    2411  7200  1.0  5.6   10   18   28   40   51   30   63   64   65   65     8  32  30  31  26  17   4
gP3IBMRd  SR#  1000 10000  1.0  1.9  3.4  6.2  9.6   18   29   46   57   69   52   69     D
iP4SeaRd  SR#  3633 15000  1.2  2.3  4.1  7.1   12   22   31   43   53   61   71   71     8   7   7   5   4   3   2
iP4WD4    #    3000  7200 0.94    5   15   29   50   61   75   82   85   86   84   86     4   6   9  18  16  10   7

jC2DSea3  &ES  2400  7200  9.3   18   29   46   66   82   94   99   98  100   99  100  D&32  31  23  21  17  10   9
at 145 GB &ES  2400  7200  8.4   16   28   46   67   87   99  100   99  100   98  100  D&27  25  24  20  15  10   8
at 290 GB &ES  2400  7200  9.1   16   28   45   65   67   67   67   67   68   67   68  D&30  29  24  20  14   8   5

qPheWD1   &SA  3000  5400   11   22   39   63   64   89   75   94   94   86   92   94  Q&14  14  14  15  13  10  10
at 224 GB &SA  3000  5400   12   26   45   55   85   84   89   90   88   88   87   90  Q&15  16  14  13  11   8   9
at 448 GB &SA  3000  5400   13   26   42   59   73   72   74   71   75   72   76   76  Q&16  16  15  14  10   9   7

q64ASeaR  *SR  1991 15000  7.4   13   20   39   60   82  106  108  112  133  133  133     D
iP4WDSRd  AR#  3633 10000  3.4  7.8   15   30   46   62   88  109  154  160  159  160    27  30  26  25  18  12  13
iP4HiSRd  AR   3000  7200  5.0   10   19   35   56   69   82  110  155  166  168  168    11  12  12  11   9   4   3

jI5SSD     SA  1600   SSD   12   23   42   72  109  155  187  216  227  232  218  232     D
jI7ISSD    SA  2800   SSD   12   24   32   30   29   38   37   46   55   83   81   83     Q
ji7Hi1     SA  2800  7200  3.5  6.8   13   25   46   79   95   82   90  104  103  104     Q

ji7WD1     SA  3700  7200    5   11   17   26   37   46   54   78  110  136  126  136  &Q 2   2   2   1   2   2   1
at 200 GB  SA  3700  7200   19   36   67  114  140  138  137  138  138  136  140  140  &Q 6   5   5   4   2   2   2
at 400 GB  SA  3700  7200   19   37   67  116  114  118  119  112  116  119  112  119  &Q 5   5   4   5   2   2   1

q64ASeaR2 *SR  1991 15000  6.5   13   23   41   69  101  147  190  200  218  225  225     D

# = Old Version  % = PCI 50MHz  SA = SATA  SC = SCSI  AR = SATA Raid  IR = IDE RAID  SR = SCSI Raid  D = Dual CPU
Q = Quad CPU ES = eSATA $ = High Speed PCI    * = PCI-X/64bit 133 MHz   & = 64 bit version  C:/D: 200 GB partitions

To Start     Results/Examples Index


Reading Speed


               CPU  Disk            MBytes/Second at KB Block Size                          % CPU Utilisation
Key            MHz   RPM     1    2    4    8   16   32   64  128  256  512 1024  Max     1   2   4   8  16  32  64

a486DX2          66  3600           0.79  1.1  1.0  1.1  1.3  1.3  1.2  1.2  1.3    1            66  59  46  48  54
bPenSea         100  5400  1.2  1.3  1.3  1.6  1.6  1.7  1.7  1.7  1.7  1.7  1.4    2    97  97  98  97  95  98  99
dPProQua        200  4500                 3.4  3.5  3.5  3.3  3.1  3.2  3.1  4.3    4                71  76  77  27
dPProQua        200  4500  2.6  2.7  2.9  3.3  3.3  3.8  4.4  4.4  4.3  4.2  4.2    4    70  72  70  74  73  80  93
fPIILap         300  5400  3.4  5.3  5.3  5.3  5.3  5.5  5.5  5.3  5.5  5.5  5.3    6    29  11  13   7   4   3   2
K62Sea2         380  5400  2.1  3.4  4.6  5.7  5.3  5.1  4.9  4.9  4.8  6.9  6.1    7    25  20  12   6   4   3   2
eCelLap         400  4200  3.7  5.5  6.7  6.8  6.7  6.8  6.7  6.7  6.7  7.2  7.2    7    32  23  14   7   4   2   1
cPMMXQua        200  5400  5.4  6.2  7.0  7.7  7.5  7.5  7.4  7.5  7.2  7.2  7.2    8    86  85  93  76  70  66  67
eCelMax         450  7200  6.5  9.9   12   12   12   12   12   12   12   12   12   12    47  38  23  11   6   3   2
fPIIFuj         450  5400  5.4  7.9   11   12   12   12   12   12   12   12   12   12    42  28  21  12   5   3   2
gPIIILap        650         11   15   15   15   15   15   15   15   15   15   15   15    61  41  21  10   5   3   1
K63WD           400  5400  5.8  8.3  8.5  8.8  8.9   16   16   16   16   17   16   17    33  22  12   7   4   2   2
fPIISea1  SC    450 10000  8.0   14   17   17   18   18   18   18   18   18   18   18    73  65  42  22  13   8   5
gp3SeaA        1400  7200  4.6  6.8  9.0  9.7   14   15   18   20   20   20   20   20    15  10   5   4   4   2   1
K62Sea          500  7200  4.0  6.7  9.4   13   14   16   18   20   20   19   19   20    34  40  14  19  12   6   5
pBarLap        1800  4200   17   18   19   20   21   20   21   21   21   21   21   21    42  31  17   8   3   1   0
gP3Max         1400        5.2  8.7   12   20   22   22   22   22   22   22   22   22    19  13   7   8   6   3   2
iP4Sea2        1900  5400  6.1   11   14   25   26   26   25   25   25   25   25   26    29  27  16  12   6   2   0
iP4NLap        1700  5400   12   20   26   26   26   26   26   26   26   26   26   26    77  65  39  18   8   3   0
nTbdMax1       1533  5400   11   19   19   20   21   26   27   27   27   27   18   27    10  16   1   3   1   0   0
mDurMax2        950  7200  7.9   10   14   23   27   27   27   26   26   27   26   27    60  36  27  23  13   6   8
eCelNA2         900        5.0  9.5   18   27   27   27   27   28   28   28   28   28    31  25  23  26  14   8   5
eCelMax2        700        8.9   14   19   24   27   29   30   30   30   30   30   30    49  39  26  17  10   6   4
eCelMax1a       950        7.5   12   18   23   26   28   29   30   30   30   30   30    54  41  26  17  10   5   2
oPalWD1        1410  7200  7.4   13   13   14   14   31   32   26   25   25   25   32    31  35  20  10   5   6   3
oPalMax1       1667  7200  6.9   12   10   13   17   23   29   30   32   32   31   32     D
iP4Sam         3066  5400  5.9   10   18   32   29   30   30   26   31   32   30   32    14  14  11   9   4   2   1
gP3IBM1        1000  7200   11   23   32   31   32   32   32   32   32   31   32   32    64  71  49  25  12   6   3
nTbdWDI1       1300  7200  8.3   13   13   14   14   32   33   27   16   27   25   33    36  34  17   8   4   5   2
gP3WD1         1000  7200  8.1   13   14   15   14   33   33   26   27   16   25   33    30  27  13   5   3   4   2
nTbdQua        1200  7200   13   34   34   33   35   35   34   35   34   35   34   35    47  57  30  15   7   4   2
iP4EWD1    SA  3000  7200  9.1   14   19   22   22   22   28   27   29   36   37   37     6   5   4   3   1   1   0
oPalIBM2       1466  7200   15   27   32   32   32   32   32   32   32   32   38   38    43  39  25  13   6   2   1
jC2CLap    SA  2000  5400  5.4  8.9   16   18   29   33   33   36   36   38   37   38    24  19  14   9  10   6   6
iP4Xeon        2175        6.5   11   18   20   22   16   16   38   40   40   39   40     D
iP4Sea1        2016  7200  5.3  9.0   12   16   22   39   34   40   39   39   38   40    24  28   8  12   9   8   2
iP4Sea4        2533  7200  4.2  7.3   10   16   33   30   30   34   39   40   40   40    14  10   4   2   2   0   0
oPalQuaS  SC   1533 10000           0.65  1.2  2.4  4.5  9.3   20   25   35   40   40     D
q64ASea1       2150  7200  6.4   11   11   22   28   26   28   32   35   38   41   41     8  10   2   5   3   0   0
iP4Sea3        2533  7200  7.3   10   12   18   33   38   40   41   40   40   40   41    27  20  11   6   6   2   0
oPalSea2 %SC   1700  7200  5.5   10   12   14   16   28   35   38   40   40   41   41    20  22  14   7   3   2   1
oPalIBM3       1539  7200   11   21   33   41   41   41   40   41   41   40   40   41    64  61  41  27  14   6   3
eCelWD1         466  7200  7.3   11   15   16   16   41   42   37   42   40   39   42    65  48  34  20  12  11   9
ip4WD2         2411  7200  9.5   19   37   41   42   42   42   19   25   22   28   42    18  25  16   0  11   8   0
iP4WD3         2533  7200  5.2   14   22   22   23   37   40   36   38   43   42   43    21  13   5  10   4   2   0
jC2DLap    SA  1830  5400  5.3   10   20   29   39   41   44   40   35   33   32   44     D
oThoWD3        2300  7200  6.7   16   23   23   23   36   41   44   44   44   43   44    15  19  15   4   3   2   1
oPalWD2        1466  7200   11   16   24   23   24   43   44   43   44   44   43   44    23  19  20   8   5   4   2
gP3RAID   IR   1000        8.8   16   23   30   33   39   40   41   41   45   39   45     D
iP4Max         2400  5400   15   26   41   46   46   46   46   46   46   46   46   46    62  52  42  21   9   3   0
nTbdQuaRd IR   1400  7200  1.3  1.3  1.2  1.5  3.1  5.3   11   18   28   42   46   46    16   9   6   4   5   5   6

jAtoLap   SA   1600  5400  4.4  8.6   16   27   46   47   42   43   40   40   40   47    10   8   9   6   6   3   1

K62IBM          267  7200  7.2   13   22   31   41   45   46   44   47   47   47   47    99  98  96  80  57  40  27
iP4WD1         2411  7200  9.9   17   38   41   43   47   46   25   25   22   30   47    25  27  24  17  13   9   7
iP4WD4         3000  7200   11   20   25   25   25   46   47   47   47   45   45   47    22  19  12   5   1   1   0
q64TLap  D:SA  1900  5400   10   20   33   49   48   48   49   50   46   50   40   50     D
iP4Hit1        1900  7200   11   21   28   37   50   51   50   49   50   49   49   51    53  45  32  19   9   5   1
nTbdIBM8       1400  7200  7.1   13   22   36   46   49   51   51   51   49   49   51    25  32  10  24  15   9   5
oPalSea3  SC   1667 15000             25   34   42   49   53   54   54   54   54   54     D
gP3Sea15  $SC   667 15000   10   19   34   54   56   56   56   56   56   57   56   57     D
oThoHi1        2080  7200   12   21   32   51   57   57   57   57   57   56   57   57    27  25  18  14   8   4   1
oPalSeaR0 %SC  1700 15000  2.5  5.3   10   19   29   44   44   56   57   57   57   57     6   7   3   6   4   0   0
oThoHi1        2080  7200   11   21   32   46   53   56   60   60   60   60   59   60    26  24  18  13   7   3   1
q64AMax1  &SA  2210  7200   20   34   59   51   56   52   61   56   62   61   60   62  D&12   5   5   1   2   1   0
q64AMax1  SA   2210  7200   21   38   53   59   54   66   64   59   64   63   62   66     D
oPalIBMRd $IR  1396  7200   11   18   26   29   46   65   68   63   61   58   62   68    52  56  34  17  10  16   9

jC2DSea2 C:SA  2400  7200    7   10   14   16   30   38   42   49   55   55   59   59    D
jC2DSea2 C:SA  2400  7200    7   11   10   20   27   32   41   47   56   56   57   57  &D10   7   5   3   2   0   0
jC2DSea2 D:SA  2400  7200   14   27   47   57   58   58   61   56   57   57   57   61    D
jC2DSea2 D:SA  2400  7200   13   23   46   54   57   62   69   62   55   54   53   69  &D20  23  11   8   2   0   0

jC2DSea1  SA   2400  7200  5.9   11   18   25   35   49   55   59   62   72   71   72     D
iP4SeaRd  SR   3633 15000  1.7  3.2  6.3  8.9   19   29   28   36   51   65   75   75     3   3   3   1   1   0   0
nTbdIBMRd IR   1400  7200   11   19   25   36   49   71   71   67   73   74   76   76    69  59  45  31  17  16   9
oPalSeaRd %SR  1700 15000  2.2  5.0   10   18   29   43   57   64   73   79   82   82     7   9   8   7   7   5   3
gP3IBMRd  SR   1000 10000  1.3  2.4  4.1  6.0  8.9   15   30   65   81   91   44   91     D
iP4HiSRd  AR   3000  7200  6.4   18   29   47   69   55   67   68   85  105  101  105     6  12  11  10   7   1   2
iP4WDSRd  AR   3633 10000   14   24   37   54   68   65   89  103  108  109  108  109    52  43  34  23  15   6   5

jC2DSea3  &ES  2400  7200   11   20   39   60   81   98  115  115  112  112  115  115  &D30  30  28  27  19  13   8
at 145 GB &ES  2400  7200   11   20   36   58   79   97   99   98   97   98   96   99  &D28  28  24  21  15  11   5
at 290 GB &ES  2400  7200   11   20   36   58   66   67   64   65   63   65   63   67  &D32  27  24  21  14   7   5

qPheWD1   &SA  3000  5400   12   24   39   77   78   97   91  103   98  101  104  104  &Q13  12  14  14  11   9   7
at 224 GB &SA  3000  5400   16   28   49   84   92   95   94   94   93   93   91   95  &Q15  14  13  13  12   9   7
at 448 GB &SA  3000  5400   14   27   51   79   77   79   82   81   79   78   80   82  &Q12  13  12  10   8   5   5

q64ASeaR  *SR  1991 15000  8.6   16   28   46   68   87  106  120  125  128  129  129     D
oPalMaxRd $IR  1667  7200             21   45   67   90  113  129   55   74   72  129     D

jI5SSD     SA  1600   SSD  138  158  139  152  181  189  195  200  207  210  213  215     D  
jI7ISSD    SA  2800   SSD   25   45   74   98  120  131  158  192  198  206  217  217     Q
ji7Hi1     SA  2800  7200  6.3   13   25   41   68  111  104  115  125  116  125  125     Q
 
ji7WD1     SA  3700  7200    8   14   19   29   38   57  105  134  138  153  152  153  &Q 2   1   1   1   1   1   1
at 200 GB  SA  3700  7200   23   41   79  135  153  155  163  158  170  160  159  170  &Q 4   4   4   4   5   4   2
at 400 GB  SA  3700  7200   22   45   77  121  125  128  135  134  129  139  123  139  &Q 4   5   4   4   3   3   2

q64ASeaR2 *SR  1991 15000  6.2   12   23   41   60   95   97  173  193  209  221  221     D

 % = PCI 50MHz  SA = SATA  SC = SCSI  AR = SATA Raid  IR = IDE RAID  SR = SCSI Raid  D = Dual CPU  Q = Quad CPU
 ES = eSATA  $ = High Speed PCI   * = PCI-X/64bit 133  MHz    & = 64 bit version  C:/D: 200 GB partitions

To Start     Results/Examples Index


DMA Speed


                 CPU  Disk            MBytes/Second at KB Block Size                        % CPU Utilisation
Key              MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024  Max   1   2   4   8  16  32  64

a486DX2           66  3600  1.2  2.1  3.3  0.5  1.0  0.7  1.0  1.1  1.2  1.3  1.3    3 100 100 100  18  42  27  43
dPProQua         200  4500  1.6  2.4  3.2  3.7  4.0  4.3  4.3  1.9  2.1  2.7  3.0    4  53  68  81  88  92  96  95
dPProQua         200  4500  1.6  2.4  3.3  3.6  4.1  4.3  4.4  2.3  2.7  3.1  3.6    4  54  67  80  85  93  96  98
eCelLap          400  4200  2.6  4.0  6.4  5.9  8.1  9.3   11   11  3.5  4.4  5.4   11  28  22  20  14  10   7   5
bPenSea          100  5400  4.2  7.3   11  2.7  2.1  1.9  1.8  1.6  1.6  1.7  1.8   11 100 100 100  98  99 100  99
K62Sea2          380  5400  1.7  3.9  5.3  8.5   11   14  5.6  5.6  5.6  6.4  6.9   14  29  38  23  20  18  12   3
fPIISea1    SC   450 10000 0.20 0.30 0.70  1.3  2.6  5.3   10   11   14   14   17   17   5  81  26   4   6   5   4
fPIIFuj          450  5400  2.3  4.2  7.7   12   17   17   20  3.8  5.6  7.5  7.8   20  24  22  21  19  11   9   5
K62Sea           500  7200  5.6  8.6   14   18   21   23   25   26   26   26   21   26  69  53  41  30  19  11   7
eCelMax          450  7200  3.0  8.4   11   17   22   25   26   23   10   12   12   26  28  39  25  19  12   6   5
gp3SeaA         1400  7200  6.8   11   15   20   23   25   26   27   26   20   16   27  32  26  17  13   6   3   2
gP3Max          1400        4.7  9.2   15   21   25   27   29   30   15   17   20   30  22  22  19  12   8   4   3
eCelMax2         700        5.1  9.0   14   19   24   27   29   30   30   19   20   30  43  36  26  18   9   5   3
eCelMax1a        950        7.4   12   17   23   26   28   29   30   30   22   30   30  53  43  31  18   9   7   2
cPMMXQua    AV   200  5400   11   20   31   14   12   12   12  5.6  5.6  6.4  6.9   31 100 100 100  97 100 100 100
mDurMax2   W2K   950  7200  6.0   12   18   24   29   32   32   31   15   20   24   32  48  46  35  24  15   9  11
mDurMax1   W98   950  7200  3.3  6.7   13   20   27   32   36   34   15   20   24   36  24  24  24  20  12   7   4
oPalQuaS    SC  1533 10000 0.20 0.30 0.60  1.3  2.6  5.2   10   21   20   28   39   39   D
eCelWD1          466  7200  8.4   14   24   34   43   50   28   47   15   30   40   50  89  81  64  46  33  21   9
iP4Sea2         1900  5400  6.0   13   21   31   39   45   51   50   11   15   18   51  29  30  24  18  10   5   1
oPalSea3    SC  1667 15000  5.7   10   18   27   37   45   51   52   52   41   41   52   D
iP4Sea2      N  1900  5400  5.7   13   20   28   37   45   49   52   11   52   30   52  34  32  25  20  10   5   3
oPalSea2   %SC  1700  7200  5.7   12   20   29   37   45   50   52   53   30   30   53  27  28  25  17  11   6   3
K62IBM           267  7200  5.9   11   20   29   39   48   53   51   51   57   57   57 100 100 100  90  68  48  36
oPalIBM3        1539  7200   11   22   34   43   50   55   58   57   58   58   58   58  73  74  56  38  20  11   5
iP4Hit1         1900  7200   11   21   34   44   51   56   59   61   61   61   61   61  58  64  42  27  14   7   2
nTbdIBM8        1400  7200  7.5   17   33   43   50   56   59   60   60   60   60   60  76  50  50  33  19  11   6
iP4Sea3         2533  7200  7.6   15   22   37   46   53   59   61   61   30   28   61  32  33  25  19  11   5   2
oPalMax1        1667  7200  4.6  8.6   16   27   41   53   62   68   71   70   28   71   D
nTbdMax1        1533  5400  6.4   13   22   34   48   60   69   72   11   23   22   72  16  15  13  10   7   4   2
iP4Sea1         2016  7200  7.5   15   27   40   52   63   70   74   70   30   34   74  55  57  51  37  24  14   8
iP4SeaRd    SR  3633 15000  1.7  3.3  6.4   12   21   35   51   55   66   77   68   77   4   3   5   3   3   2   1
iP4Max          2400  5400  5.3   10   19   31   46   62   74   78   74   45   43   78  23  23  20  16  11   6   3
gP3IBM1         1000  7200   11   24   39   53   65   73   78   73   78   79   78   79  77  88  73  50  30  17   9
nTbdQuaRd   IR  1400  7200  3.9  7.2   13   24   38   54   63   73   78   80   79   80  65  62  58  52  42  30  44
oPalSeaR0%  SC  1700 15000  2.3  4.6  8.7   16   27   42   57   71   77   80   52   80   7   7   7   6   5   4   2
iP4NLap         1700  5400  4.9   10   17   30   42   57   70   80   83   82   75   83  34  35  30  26  17  11   7
nTbdQua         1200  7200  6.2   11   21   35   53   65   74   77   81   84   83   84  26  23  22  19  14  10   5
iP4Sam          3066  5400  6.7   13   23   36   53   67   78   85   83   60   26   85  18  17  15  13   8   6   3
pBarLap         1800  4200  6.7   14   19   34   52   66   77   83   85   86   80   86  24  25  17  15  11   7   4
iP4Sea4         2533  7200  8.5   24   32   49   65   78   83   88   87   23   22   88  38  53  35  26  16   9   3
iP4WD1          2411  7200   13   27   40   57   70   79   32   87   88   30   60   88  48  50  37  26  16   9   0
ip4WD2          2411  7200   15   27   40   57   70   79   84   87   88   30   40   88  57  50  37  26  16   8   4
iP4WD3          2533  7200   14   19   39   55   69   79   85   89   89   39   40   89  66  82  45  31  18  17   4
oPalWD1         1410  7200   12   21   32   50   65   77   83   89   89   89   40   89  71  66 100  69  28  15  15
q64ASea1        2150  7200  8.8   18   32   49   65   77   86   91   89   30   40   91  23  23  20  15  10   5   2
oPalWD2         1466  7200   15   25   37   56   70   80   87   91   91   91   48   91  59  50  50  28  18  11   5
oThoWD3         2300  7200   15   28   44   59   73   30   89   92   92   76   40   92  51  49  38  27  16   3   5
iP4WD4          3000  7200   13   32   49   65   77   85   90   55   92   92   40   92  31  37  33  18  10   4   1
oPalIBM2        1466  7200   16   30   49   65   77   85   90   92   92   92   91   92  56  53  42  28  16   9   5
oThoHi1         2080  7200   12   25   48   71   78   87   91   93   93   93   93   93  28  31  29  22  12   7   3
gP3Sea15   $SC   667 15000  5.6   11   21   30   57   74   99  109  111   45   49  111   D
jC2DLap     SA  1830  5400  5.3   15   16   43   64   81   97  107  110  113  112  113   D
oPalIBMRd  $IR  1396  7200   13   26   42   58   72   89  111  112  112  113  113  113  82  92  75  52  32  35  17

jAtoLap     SA  1600  5400  6.0   11   21   36   58   80  101   80  117  117  118  118   6   6   5  13   3   8   3

jC2DSea3 D& ES  2400  7200   11   20   39   61   82   93   99  118  109  107  108  118  35  31  31  24  21  11  13

iP4EWD1     SA  3000  7200   14   25   42   57   80   99  112  117  119  119  117  119   7   2   2  13   1   2   2
jC2CLap     SA  2000  5400  5.6   12   27   36   66   89  108  120  121  118  118  120  32  33  36  23  27  16   8
q64TLap     SA  1900  5400  5.7   17   31   50   73   95  112  123  124  125  115  125   D
q64AMax1   &SA  2210  7200   23   39   62   86  105  119  127  131   38   59   60  131 D16  20  18  19   8   7   3
q64AMax1    SA  2210  7200   22   39   60   85  104  118  125  131   41   29   38  131   D
q64ASeaR   *SR  1991 15000  5.3   11   19   33   54   79  103  119  115  130  131  131   D
oPalMaxRd  $IR  1667  7200  7.9   15   27   44   65   93  114  114  145  153  158  158   D

jI5SSD      SA  1600  SSD   4.1  6.9   13   21   34   57   88  131  118  160  189  189   D
jI7ISSD     SA  2800  SSD   8.0   11   16   36   54   89  128  169  172  173  164  173   Q

iP4WDSRd    AR  3633 10000   16   28   45   62   32   88   95  135  188   42   83  188  60  55  41  28   5   9   4
qPheWD1  & SA2  3000  5400  9.3   19   33   60   88  130  166  186  197  203  204  204 Q 9  10   7  10  11  10   7
iP4HiSRd    AR  3000  7200  9.8   20   42   71   91  105  126  183  204  207  208  208  19  20  20  17  11   6   7
jC2DSea2 & SA2  2400  7200   13   26   49   80  120  165  205  220  220  219  206  220 D31  29  31  28  17  10  12
jC2DSea2   SA2  2400  7200   12   25   50   82  133  165  206  235  236  227  209  236   D
q64ASeaR2  *SR  1991 15000    6   12   23   43   74  117  158  196  209  145  233  233   D
jC2DSea1   SA2  2400  7200   15   18   35   68  109  145  173  208  229  237  243  243   D
ji7Hi1      SA  2800  7200  7.1   14   28   55  107  208  229  247  249  251  251  251   Q
ji7WD1      SA  3700  7200   12   23   45   84  143  229  316  401  458  486  508  508   Q

% = PCI   SA = SATA  SA2 = SATA2  SC = SCSI  AR = SATA Raid  IR = IDE RAID  SR = SCSI Raid  D/Q = Dual/Quad CPU
 ES = eSATA  * = PCI-X/64bit 133 MHz     $ = High Speed PCI    AV = Norton AV in use   & = 64 bit version

To Start     Results/Examples Index



Random Access Time Small Files


         Buffer  CPU  Disk    Milliseconds at MB  file size
Key          MB  MHz   RPM   1    2    4    8   16   32   64  

a486DX2           66  3600 12.7 19.1 21.4
dPProQua      S  200  4500  9.9 11.0 10.6 16.0
dPProQua      S  200  4500 10.0 12.9 12.5 12.8
eCelLap          400  4200  9.9  9.7  9.6 11.9
K62Sea2       E  380  5400  8.5  7.6  9.3
cPMMXQua      S  200  5400  4.9  7.4  8.2  8.6
fPIIFuj       Q  450  5400  7.7  7.9  8.1  8.5
pBarLap       2 1800  4200  1.3  4.5  6.3  8.3
oPalWD1         1410  7200  0.5  1.9  6.7  7.1
eCelNA2          900        5.5  5.7  5.9  7.1
iP4Sam          3066  5400  4.9  6.6  6.1  7.0
iP4Sea2       2 1900  5400  4.2  5.8  6.0  7.0
eCelMax1a        950        1.2  3.4  5.4  6.9
nTbdMax1      2 1533  5400  2.0  4.4  6.3  6.8  7.5
eCelMax       Q  450  7200  6.5  6.9  6.2  6.5
iP4Sea2       2 1900  5400  1.9  5.9  6.5  6.4
gP3Max          1400        1.9  3.9  5.0  6.0
jC2CLap     SA  2000  5400  0.6  0.3  2.0  5.9  8.5
iP4Max          2400  5400  2.2  3.9  5.8  5.8
iP4Sea1       2 2016  7200  0.9  3.1  4.2  5.7
gp3SeaA         1400  7200  1.3  2.7  5.1  5.6
nTbdQuaRd   2R? 1400  7200  0.6  1.7  4.4  5.6
iP4Sea3       2 2533  7200  1.0  3.0  4.2  5.5
nTbdIBM8      2 1400  7200  0.7  2.2  4.1  5.1
K62Sea        2  500  7200  0.6  3.3  4.3  5.0
oPalSea2      2 1700  7200  0.7  3.0  3.5  4.9
iP4Sea4       8 2533  7200  0.4  1.0  2.7  4.9
mDurMax2      2  950  7200  0.8  2.6  4.3  4.9
mDurMax1      2  950  7200  0.8  2.3  4.2  4.8
nTbdQua       2 1200  7200  0.9  3.0  3.8  4.8
eCelWD1          466  7200  0.4  2.0  4.2  4.7
gP3IBM1       2 1000  7200  0.7  2.7  3.5  4.7  5.4
iP4EWD1       8 3000  7200  0.2  0.9  2.4  4.5
oPalMax1        1667  7200  2.1  3.5  4.4  4.4
K62IBM        2  267  7200  0.6  1.7  4.2  4.4
iP4Hit1       2 1900  7200  0.7  2.6  3.8  4.4
oThoHi1       2 2080  7200  0.3  2.3  3.5  4.4
oPalIBM3      2 1539  7200  0.5  2.7  3.8  4.3
fPIISea1      1  450 10000  6.0  5.0  4.8  4.3  4.7
oPalIBM2      2 1466  7200  0.4  2.7  3.7  4.1
q64ASea1      8 2150  7200  0.4  0.8  2.0  3.9
oPalIBMRd   2R2 1396  7200  0.1  0.6  1.9  3.8
eCelMax2         700        1.9  2.8  1.8  3.8
iP4NLap      16 1700  5400  0.2  1.0  2.8  3.3
q64TLap       8 1900  5400  0.2  0.2  1.1  3.2
jC2DSea2     16 2400  7200  0.1  0.5  1.2  3.0  6.1  5.2  5.4
oPalSeaR0     8 1700 15000  0.4  1.1  2.2  2.9
oThoWD1       8 1737  7200  0.1  0.2  0.7  2.8
iP4WD4        8 3000  7200  0.1  0.2  0.8  2.8
jC2DSea1     16 2400  7200  0.2  0.1  0.5  2.8
iP4WD3        8 2533  7200  0.1  0.2  0.6  2.6
oPalSea3        1667 15000  0.2  0.7  2.1  2.4
oPalWD2       8 1466  7200  0.1  0.1  0.5  2.3
jAtoLap         1600  5400  0.2  0.3  0.8  2.2
iP4SeaRd      8 3633 15000  0.8  0.7  0.9  2.1
gP3Sea15      8  667 15000  0.3  0.6  1.3  2.0
q64ASeaR    3S8 1991 15000  0.2  0.4  0.6  1.5
ip4WD2        8 2411  7200  0.1  0.2  0.8  1.2
iP4WD1        8 2411  7200  0.1  0.1  0.6  1.1
iP4HiSRd    2R8 3000  7200  0.1  0.3  0.7  0.9
q64AMax1     16 2210  7200 0.04 0.04  0.2  0.9  2.9  4.1  4.9
q64ASeaR2   4S8 1991 15000  0.2  0.3  0.5  0.9  1.3  1.8
iP4WDSRd    2R8 3633 10000  0.1  0.1  0.3  0.9  1.8  3.4
jI5SSD          1600  SSD   0.3  0.3  0.3  0.3  0.3 
jI7ISSD         2800  SSD   0.1  0.1  0.1  0.1  0.1  0.1
ji7Hi1       16 2800  7200  0.1  0.1  0.4  0.6  3.0  6.3
jC2DSea3 &   16 2400  7200  0.3  0.4  0.4  0.4  2.8  4.8  6.1
ji7WD1   &   64 3700  7200 0.08 0.08 0.08 0.08 0.10 0.17  3.7
q64AMax1 &   16 2210  7200 0.04 0.04 0.04 0.05 0.70  5.0  6.4
 
2R = 2 Disk SATA Raid     3S/4S = 3/4 Disk SCSI Raid   SA SATA
S  = Small  Q = 256 KB   E = 128 KB  & = 64 bit version

To Start     Results/Examples Index



Extended Random Access Times

Following are random reading times of files up to 1 GB. The laptop disk runs at 5400 RPM and, except for the WD Green drive, the others run at 7200 RPM. The results are influenced by buffer size, where speeds are very fast when data is cached there, but with reducing impact as the file size increases. Then latency (half disk revolution time) come into play at 4.2 milliseconds for 7200 RPM and 5.6 milliseconds at 5400 RPM. Finally, head switching and movement affects larger files, influenced by the number of heads/platters and amount of data per track. Fragmentation will also have some impact. Results for a SSD are also included, showing a constant 0.1 milliseconds random reading time.

The WD green disk (qPheWD1) performs well on these tests, initially casting some doubt that the rotation speed was 5400 RPM. Then tests were run on the tiny WD Passport USB drive (yDiskUSB2g) where performance is similar on the larger files. Note access times on ji7WD1 with 64 MB buffer.


            Buffer  CPU  Disk    Milliseconds at MB  file size
   Key          MB  MHz   RPM    1    2    4    8   16   32   64  128  256  512 1024

F: yDiskUSB2g    8 2400  5400  0.4  0.4  0.4  2.8  5.2  6.4  7.5  7.5  8.1  7.9  8.2

C: jC2DLap       8 1830  5400  0.2  1.5  2.4  5.4  9.1  9.5  9.5  9.8 10.8  9.6 10.5
D: jC2DLap       8 1830  5400  0.3  1.5  2.6  6.0  6.4  7.2  8.4  8.2  8.6  9.6 10.6

C: qPheWD1  &   16 3000  5400  0.1  0.1  0.1  0.1  0.3  3.7  6.3  6.9  7.2  7.0  8.4
D: qPheWD1  &   16 3000  5400  0.1  0.1  0.1  0.1  0.1  2.8  5.4  7.4  7.6  7.7  7.6 
E: qPheWD1  &   16 3000  5400  0.1  0.1  0.1  0.9  2.4  4.5  6.0  8.2  8.1  8.1  8.0

C: q64AMax1     16 2210  7200 0.04 0.04 0.04 0.05  1.3  5.2  7.4  8.7  9.1  9.2  9.6

C: jC2DSea2 &   16 2400  7200  0.1  0.1  0.3  2.5  6.3  8.0  8.1  7.3  6.1  6.3  6.7
D: jC2DSea2 &   16 2400  7200  0.1  0.2  0.2  3.6  5.1  5.7  5.9  6.0  5.7  6.0  6.7

G: jC2DSea3 &   16 2400  7200  0.3  0.4  0.4  0.5  2.7  5.1  5.8  6.3  6.9  7.9  8.1
H: jC2DSea3 &   16 2400  7200  0.3  0.4  0.4  0.5  2.9  5.4  6.2  7.0  6.9  7.0  7.2
I: jC2DSea3 &   16 2400  7200  0.3  0.4  0.4  0.8  3.4  5.3  6.4  6.6  7.3  7.4  7.5

C: ji7WD1       64 3700  7200 0.26 0.27 0.16  0.4  2.0  2.7  4.9  6.7  7.7  8.3  8.4
D: ji7WD1       64 3700  7200 0.09 0.16 0.17  0.4  0.6  1.5  2.9  4.2  4.5  5.0  7.2
J: ji7WD1       64 3700  7200 0.09 0.09 0.09  0.6  0.9  1.7  3.7  4.7  5.0  4.9  7.3

C: jI7ISSD         2800  SSD   0.1  0.1  0.1  0.1  0.1  0.1  0.1  0.1  0.1  0.1  0.1
   
    Partitions GB jC2DLap C: 70,  D: 70,   qPheWD1 C: 224, D: 224, E: 146, 
                  jC2DSea2 C: 195, D: 177, jC2DSea3 G: 145, H: 145  I: 30 eSATA
                  yDiskUSB2g C: 250,       ji7WD1  C: 200, D: 400, J: 400

To Start     Results/Examples Index


Configuration Key

Code       Configuration

a486DX2    80486DX2 66.7 MHz, 3600 RPM, Win95 (1994)
bPenSea    Pentium 100 MHz,  Seagate ST5850A, 850MB, 3600 RPM Disk, Win95, ISA connection (1995)
cPMMXQua   Pentium MMX 200 MHz, Quantum Fireball ST 2.1AT, 2.1GB, 5400  RPM disk, Win95 (1997)
dPProQua   Pentium Pro 200 MHz, Quantum Fireball TM2100A, 2.1GB, 4500 RPM disk, NT 4.0 (1995)
dPProQua   Second result version 1.3A
eCelLap    Celeron 400 MHz Laptop, Fujitsu MFH 2043, 4.3 GB 4200 RPM disk, Win98 (1999)
eCelMax    Celeron 450/100MHz, Maxtor DiamondMax Plus 2500 7200 RPM disk, Win98 (1999)
eCelMax1   Celeron 950/100 MHz, Win98, Maxtor DMA66 disk, 5400 RPM (2001)
eCelMax1a  As CelMax1, DMA not working WinXP, different Maxtor disk (2003)
eCelMax1a  Second result DMA working via reinstalling primary IDE
eCelMax2   Celeron 700 MHz, 80 GB Maxtor disk, UDMA 33. Results via Win98 and WinXP (2003)
eCelNA2    Celeron 900/100 MHz, WinME (2002)
eCelWD1    Celeron 466 MHz, WD 60 GB 7200 RPM disk, win98 (2002)
fPIILap    Laptop, Mobile Pentium II/PE 300 MHz, 5400 RPM disk, Win98 (1999)
fPIIFuj    Pentium II 450 MHz, 512K cache, Win98, Fujitsu MPC 3102AT 10.2GB 5400 RPM Disk (1999)
fPIISea1   PII 450, Win98, LVD 10Krpm Seagate ST39102LW SCSI, 9GB (1998)
gP3IBM1    PIII 1000 MHz, IBM Deskstar 60GXP, 40GB, 2MB cache, 7200 rpm, ATA100: DMA not enabled (2001)
gP3IBM1    DMA enabled
gP3IBMRd   Dual PIII 1 GHz, RAID0, 4 x IBM 10K RPM SCSI160 18.3 GB, Adaptec 3200 64 bit RAID card 80 MB
           cache, Win2K (2001)
gP3Max     Powerleap CPU upgrade to 1400 MHz, 20GB Maxtor disk, Win2K (2003)
gP3RAID    Dual Pentium III 1 GHz, IDE RAID0, Win2K (2002)
gP3Sea15   Dual Xeon 667 MHz, Seagate Cheetah Ultra 160 SCSI 15K, ST318452LW, Adaptec AIC-7899, 
           Win2K (2002)
gp3SeaA    Powerleap CPU upgrade  1400 MHz, 40GB Seagate Barracuda, 7200 RPM, ATA/100 drive mobo 
           ATA/33 Win2K (2003)
gP3WD1     PIII 1000, Intel drivers, WD400BB, 7200 RPM, partitions C, D, E (2001)
gPIIILap   Toshiba Satellite 2800 Laptop, PIII 650 MHz, WinME (2000)
iP4EWD1    P4E Prescott 3.0 GHz, WD2000JD 7200 RMP, 8 MB buffer, 200 GB SATA disk (3/4 full), WinXP (2004)
iP4Hit1    P4 1900 MHz, Hitachi 7200 RPM 7K80, WinXP
iP4HiSRd   P4 3 GHz, two 80GB Hitachi 7K250 Deskstars, 8MB cache, SATA RAID 0, 128K stripe on 
           Intel ICH5R  WinXP (2004)
iP4Max     P4 2405 MHz, Maxtor 5400 RPM, 80 GB drive (D:) Normal results + same file cache enabled (2003)
iP4NLap    Laptop, 1700 MHz P4-M Northwood CPU,  WinXP, Toshiba MK4019GAX 20 GB disk 5400 RPM (2003)
iP4Sam     P4 3.06 GHz, Samsung SV08134 disk, 80 GB, 5400 RPM, Win XP (2003)
iP4Sea1    P4 2016 MHz, Seagate Barracuda IV 7200 RPM disk, WinXP (2001)
iP4Sea2    P4 1900 MHz, 3 partitions, Seagate ST340810A 40GB 5400 RPM disk, WinXP (2001)
iP4Sea2    2 other partitions and using Win File Cache
iP4Sea3    P4 2.53 GHz, Seagate ST380021A Barracuda 4 80 GB 7200 RPM, WinXP (2003)
iP4Sea4    P4 2.53 GHz, Seagate Barracuda V, ST3120024A, 8MB cache, WinXP (2003)
iP4SeaRd   P4 3633 MHz, 2 Seagate Cheetah 15K RPM disks, Adaptec cotroller, RAID0, WinXP (2003)
iP4WD1     P4 2411 MHz, Western Digital 120GB disk 8MB cache, 7200 RPM, WinXP (2002)
ip4WD2     P4 2411 MHz, WD 100GB 8MB Intel and MS drivers, 7200 RPM, WinXP (2002)
iP4WD3     P4 2.53 GHz,  Western Digital 80GB HD 8MB cache, WinXP (2002)
iP4WD4     P4 3.0 GHz, WDC WD800JB 7200 RPM 8 MB buffer, Win2K (2003)
iP4WDSRd   P4 3363 MHz Abit 1C7G Canterwood,SATA RAID0, 2 WD Raptor disks, 10,000 RPM/8MB cache (2003)
iP4Xeon    Dual Xeon 2.2 GHz, IDE disk, Win2K (2002)
jC2CLap    Compaq 6720s, Celeron 550 2.0 GHz, Fujitsu 5400 RMP Disk, Vista Home Basic (2008)
jAtoLap    Samsung N140 Netbook, 1.6 GHz Atom, 160 GB 5400 RPM disk, WinXP (2010)
jC2DSea1   Core 2 Duo 2.4 GHz, Seagate ST3320620AS SATA Disk 320 GB, 16 MB buffer, WinXP (2006)
jC2DSea2   Core 2 Duo 2.4 GHz, Seagate ST3400633AS SATA Disk 400 GB, 16 MB, Vista 64 Bit (2007)
           4 GB DDR2 RAM, 32 bit and 64 bit results
jC2DSea3   As jC2DSea2 but Seagate ST3320613AS 320 GB, in Akasa 1.5 Gb/sec eSATA enclosure (2009)
           Results early tracks, at 145 GB and 300 GB
jC2DLap    Acer 7720G, Core 2 Duo 1.83 Ghz, Hitachi HTS5425 5400 RPM disk, 8 MB cache, Vista 32 Bit (2008)
jI5SSD     Samsung Series 9 13.3-Inch Laptop, 1.6 GHz Core i5-2467M CPU, 128 GB SSD, Win 7 (2012)
jI7ISSD    Intel Corei7 860 2.8Ghz, Intel X25-M 80GB SSD, 64-Bit Windows 7 (2009)
jI7Hi1     Core i7 930, 2.8 GHz, Hitachi HDS 721010CLA332 1 TB 7200 RPM disk, 64-Bit Windows 7 (2010)
ji7WD1     Core i7-4820K 3.7 GHz, Asus P9X79 LE, 4 Channel 32 GB 800 MHz, DDR3 RAM, 1 TB WD CAVIAR 
           BLACK WD1003FZEX SATA 6 Gb/s, 64MB cache, Windows 8.1, 2014
K62IBM     AMD K6-2/266MHz, IBM Deskstar 80GB 7200rpm 2MB cache, Win98 (2002)
K62Sea     AMD K62 500 MHz,  Seagate Barracuda 20GB drive, Win2K (2001)
K62Sea2    AMD K6-2 380 MHz,  Seagate ST36531A 6.5GB, Win98
K63WD      AMD K6-3 400 MHz, WD136AA 5400 RPM disk 13 GB, Win 98SE (2000), slow with DMA on
K63WD      Second results DMA off and third DMA on with new IDE driver (2000)
mDurMax1   Duron 950 MHz, Win98SE, Maxtor DiamondMax Plus 40 53078H6, 7200 RPM, IDE UDMA66  (2000)
mDurMax1   Second results using Win File Cache (2000)
mDurMax2   As DurMax1, but Win2K (2000)
mDurMax2   Second results using Win File Cache (2000)
nTbdIBM8   Tbird 1.4 GHz, IBM 180GXP 60GB 2 MB buffer 7200 RPM, Win2K (2003)
nTbdIBMRd  Tbird 1.4 GHz, Promise FastTrak 100-TX2 EIDE RAID, 2 x IBM 60GXP, 7200 RPM, 2MB, Win2K, (2001)
nTbdMax1   Tbird 1544 MHz,  Maxtor 60GB, ATA100, 2MB, 8ns, 5400RPM, NT4 (2001)
nTbdMax1   Second results, drivers re-installed - faster writing (2001)
nTbdQua    AMD Tbird 1200 MHz, Quantum Fireball AS 20 GB 7200 RPM ATA100, Win98 (2001)
nTbdQuaRd  Athlon 1.4 GHz RAID0 with the KK266R, 2-40Gb Quantum Fireball 7200 rpm, WinXP (2002)
nTbdWDI1   Tbird 1296, VIA driver, WD400BB 7200 RPM, partitions C, D, E (2001)
oPalIBM2   AMD XP 1700 1.47 MHz, IBM disk 60GB 7200 rpm 2mb buffer,. Win2K (2001)
oPalIBM3   AMD XP1800 1539 MHz, IBM Deskstar 120GXP 41.2GB UDMA 100, Win98 (2003)
oPalIBM3   Second result NAV enabled, third result  IDE socket instead of RAID
oPalIBMRd  Athlon 4 1397 MHz, partitions C, D, E, 2-IBM 60GXP 40GB, RAID 0, HPT 370A, Win2K (2001)
oPalMax1   Dual athlon 1.67 Ghz, Maxtor ATA133 80GB 7200, onboard primary IDE, Win2K (2002)
oPalMaxRd  Dual athlon 1.67 Ghz, 3ware: Escalade 7810 w/ 8 Maxtor ATA133 160GB 5400, raid 0, Win2K (2002)
oPalQuaS   Dual AMD MP 1533 MHz, Quantum 10K RPM disk, Adaptec 29160 U160,  WinXP, (2002)
oPalSea2   Athlon XP 1700 MHz, Barracuda IV, 7200 RPM, Adaptec 2110S, SCSI Ultra 160, WinXP (2002)
oPalSea3   Dual Athlon 1667 MHz, Adaptec 19160 SCSI, 15K RPM Seagate Cheetah disks, Win2K (2002)
oPalSeaR0  As oPalSeaRd PCI but single Seagate 15K RPM Cheetah ST318452LC, PCI 50 MHz (2002)
oPalSeaR2  As oPalSeaRd PCI but bus at 34 and 42.5 MHz  (2002)
oPalSeaRd  Athlon XP 1700 MHz, RAID0 Adaptec 2110S, SCSI Ultra 160, 2 x Seagate 15K RPM  ST318452LC,
           18.4 GB, 8 MB buffer, PCI 50 MHz, WinXP (2002)
oPalWD1    Athlon XP 1410 MHz, ASUS A7V266-E,  WD40G-7200 RPM (2002)
oPalWD2    AMD XP 1.47 GHz, SLOW, Western Digital WD1200JB-00CRA1, Win98 (2003)
oPalWD2    Fast results WinXP
oThoHi1    AMD Tbred 2.08 GHz, Hitachi 7K250 123 GB, 7200 RPM, 2 MB, Win2K (2004)
oThoHi1    More results using nVidia IDE driver instead of MS
oThoHi2    AMD Tbred 2.08 GHz, Hitachi 7K250 123 GB, 7200 RPM, 2 MB, Win98 (2004)
oThoHi3    AMD Tbred 2.38 GHz, Hitachi 7K250 123 GB, 7200 RPM, 2 MB, Win2K (2004)
oThoHi3    More results using nVidia IDE driver instead of MS
oThoWD1    AMD Tbred 1.73 GHz, WD 120 GB disk, 7200 RPM, 8 MB cache, WinXP (2003)
oThoWD3    AMD Tbred 2.30 GHz, WD 120 GB disk, 7200 RPM, 8 MB cache, WinXP (2003)
pBarLap    Laptop, AMD XP 1800 MHz, WinXP, Toshiba MK4018GAS 40 GB disk, 4200 RPM, 2MB  (2003)
q64AMax1   Athlon 64 X2 2.21 GHz, 300 GB Maxtor disk 7200 RPM SATA 16 MB Buffer, WinXP Pro X64 (2005)
q64ASea1   Athlon 64 2.15 GHz, 120GB Seagate Barracuda V, 8MB cache, Win XP (2003)
q64ASea2   As q64ASea1 with MS driver plus new IDE 2.0 nVidia driver 3.66
q64ASeaR   Dual Opteron 246 CPUs 2 GHz, 3 x Seagate 15K RPM Cheetah ST318452LC, 18.4 GB, 8 MB buffer,
           Adaptec 39320D-R PCI-x/64bit 133 Mhz, WinXP (2003)
q64ASeaR2  Dual AMD Opteron 246 CPUs 2 GHz, 4 x Seagate 15K RPM Cheetah ST318453LC ultra 320 disks, 
           18.4 GB, 8 MB buffer, Adaptec 39320D-R PCI-x/64bit 133 Mhz, WinXP (2003)
q64TLap    Acer 5520G Laptop, Turion 64 X2 1900 MHz, WD2500BEVS disk 5400 RPM, Vista (2007)
qPheWD1    Phenom II X4 3.0 GHz, 640 GB WD 5400 RPM WD6400AACS disk, 16 MB buffer, 64-Bit Windows 7 (2009) 

yDiskUSB2a P4 1900 MHz, Freecom Disk 7200 RPM, 8 MB buffer, (Hitachi HDT72252 5DLAT80), WinXP
yDiskUSB2b Athlon 64X2 2.21 GHz, Freecom Disk 7200 RPM, 8 MB buffer, (as above) WinXP x64
yDiskUSB2c AMD Tbred 2.08 GHz, Freecom Disk 7200 RPM, 8 MB buffer, (as above), Win2K
yDiskUSB2f Core 2 Duo 2.4 GHz, Freecom Disk 7200 RPM, 8 MB buffer, (as above), 64-Bit Vista
yDiskUSB2g Core 2 Duo 2.4 GHz, WD 2500MER Passport 250 GB, 5400 RPM, 64-Bit Vista
yDiskUSB2d External USB 2.0 Acomdata box with WD disk, AMD CPU, Win2K.
yDiskUSB2e Same disk USB 2.0 and Firewire, P4 3 GHz
yDiskFwire Same disk USB 2.0 and Firewire, P4 3 GHz
yDiskUSB3a Core i7-4820K 3.7 GHz, Seagate Expansion  SRD00F1, 1TB USB 3.0, 5400 RPM, 2014

zCDRW      Duron 950 MHz, Iomega ZipCD USB Drive (6 x 4 x 4) via benchmark on CDRW disc, Win98
zCFlashLP  Duron 950 MHz, San Disk Compact Flash Card Reader on LP port. Benchmark on 20 MB CF, Win2K
zCFPcmcia  Celeron 400 MHz Laptop, Compact Flash in PCMCIA adapter, Win98
zCFPcmci2  AMD XP 1.8 GHz Laptop, Compact Flash in PCMCIA adapter, WinXP
zDVDRW     Pentium 4 1.9 GHz, Liteon LH-20A1H DVD +RW 8x, disc 4x, Nero InCD 5, WinXP
zFD4096A   Staples USB Flash Drive 4 GB
zFD4096C   SanDisk Cruzer Micro 4 GB, Enhanced for ReadyBoost
zLAN10mbs  Disk on remote Duron 950 PC. Test on Celeron 450 MHz via 10 Mb/sec Lan, Win98.
zLAN100m1  100mbs LAN disk on remote Duron/Win2K server, program run on P4 1900 WinXP
zLAN100m2  100mbs LAN disk on remote P4/WinXP server, program run on Duron/Win2K
zLAN100m3  100mbs LAN disk on remote Fujitsu Laptop WinXP, program run on P4 1900 WinXP
zSD2GB     System jC2DSea2, 2 GB SanDisk Ultra II 110x speed rating SD Card, SanDisk Reader (2009)
zSD32MB    System jC2DSea2, 32 MB SD Card supplied with Canon camera, SanDisk Reader (2009)
zUSB1CF    P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, old card, USB 1.1
zUSB1CF    P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, new card, USB 1.1, 2003
zUSB2CF    P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, old card, High Speed USB 2
zUSB3CF    Core i7-4820K 3.7 GHz, 32 GB SanDisk Extreme USB 3.0 Flash Drive, 2013
zUSB2CF    P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, new card, High Speed USB 2, 2003
zUSBComms  Disk on remote Duron 950 PC. Test run on P4 1.9 GHz WinXP. Link Belkin USB Direct Connect
zWiFi56    jC2DLap above wireless connection 56 kbps


To Start      Results/Examples Index



Roy Longbottom at Linkedin  Roy Longbottom July 2014

The new Internet Home for my PC Benchmarks is via the link
Roy Longbottom's PC Benchmark Collection