NeXT TypedStream Data | 1993-05-23 | 7.6 KB | 129 lines
typedstream
IBObjectData
Object
CustomObject
Application
WindowTemplate
iiii***@s@
Panel
Responder
TextField
Control
TextFieldCell
ActionCell
A program to find your disk's performance.
Indices are given relative to the Maxtor P1-17S drive which is assumed to be a fast drive.
This program is free, you may copy and distribute it, but you're not allowed to sell it.
Helvetica-Oblique
Drive Performance
Helvetica
VERSION 1.1
Button
ButtonCell
NXImage
'COPYRIGHT 1991-93, UNTER ECKER SOFTWARE
by Oliver Unter Ecker
MenuCell
Matrix
@:@iiii
Info Panel...
Help...
List of drives...
ff@@#::s
[15@]
Times-Roman
Index
CustomView
PercentBar
kb/sec
START
)Consider your drive fast when index > 0.8
Write
ScrollView
ClipView
ciifffcfffs
[869c]{\rtf0\ansi{\fonttbl\f0\fswiss Helvetica;}
\margl40
\margr40
\pard\tx960\tx1920\tx2880\tx3840\tx4800\tx5760\tx6720\tx7680\tx8640\tx9600\f0\b0\i0\ulnone\fs36\fc0\cf0 Welcome to DrivePerformance\
\fs24 DP does not measure performance at a low (system) level, but tries instead to get values that should be obtainable from normal applications written in C.\
\i Bigger indices than 1.0 stand for a higher performance than the MAXTOR P1-17S drive (what we assume to be fast), smaller less.
\i0 \
DP uses the 'fread', 'fwrite' and 'fseek' C-calls and takes advantage of some handtuned parameters (e.g. buffersize). It creates two files of approx. 10mb in the /tmp directory.\
DP cannot disable disk caching. On systems with 32 mb RAM or more, the read value is true for moderately large files written immediately before they are read, but false in all other cases.
NXCursor
NXibeam
Scroller
_doScroller:
@@@ffs(
[2324c]{\rtf0\ansi{\fonttbl\f0\fswiss Helvetica;}
\margl40
\margr40
{\colortbl;\red0\green0\blue0;}
\pard\tx960\tx1920\tx2880\tx3840\tx4800\tx5760\tx6720\tx7680\tx8640\tx9600\f0\b0\i0\ulnone\fs36\fc0\cf0 List of drives\
\fs24 Here you find the performance of some tested drives. \
Future versions 1.x of DP will use the same algorithms but will show more tested drives. \
To add your drive please send non-NeXT mail to \
untereck@vipmzk.Physik.Uni-Mainz.de
\pard\fc0\cf0 \
\pard\tx520\tx1060\tx1600\tx2120\tx2660\tx3200\tx3720\tx4260\tx4800\tx5320\fc1\cf1 Capacity kb/sec write kb/sec read DP index Model\
100 MB 445 466 0.53 (Quantum LP105S) 3.5"\
200 MB 371 467 0.48 (Quantum LP210S) 3.5"\
400 MB 642 786 0.82 (Maxtor) 3.5"\
400 MB 669 692 0.79 (Seagate ST-1480) 3.5"\
660 MB 652 674 0.77 (Maxtor XT-8760S) 5.25"\
660 MB 688 726 0.79 (HP 97548S) 5.25"\
1.4 GB 797 891 0.98 (Maxtor P1-17S) 5.25"\
Non-NeXT drives:\
240 MB 413 439 0.50 (Quantum LP240S) 3.5"\
310 MB 471 502 0.57 (Seagate Wren 94171)\
349 MB 387 ? ? (Maxtor XT-8380S)\
340 MB 602 618 0.80 (Maxtor LXT-340) 3.5"\
426 MB 672 613 0.7x (Fujitsu M2623SA) 3.5"\
520 MB 808 692 0.9x (Fujitsu M2624SA) 3.5"\
633 MB 820 316 0.64 (Hitachi DK515C)\
955 MB 454 574 0.59 (HP 97549T) 5.25"\
1.0 GB 665 693 0.79 (Fujitsu M2266SA-512) 5.25"\
1.3 GB 559 528 0.64 (Micropolis 1908)\
Besides.\
The values shown above are making all drives looking worse than what they really are. Under NEXTSTEP they perform approx. two times worse compared to many other computers.\
The main reason for this is missing SCSI synchronous support in NEXTSTEP. This is a software item; the SCSI chip employed in NeXT computers, as well as nearly all SCSI adapters for Intel based machines support synchronous data transfer.\
Fujitsu drives seem to have the biggest variance, when measured with DP.\
Maxtor drives are the most silent, esp. 3.5 inch drives, followed by Quantum. Micropolis and Seagate drives are not silent anymore. HP and Fujitsu drives are really loud (but not Fujitsu's 3.5" drive).