This is the Read Me file for MacBench Version 1.1. It contains the MacBench license agreement, installation instructions, and important notes about MacBench 1.1.
Ziff-Davis is releasing this version of MacBench to let you obtain realistic, reliable performance measurements on both existing Macintosh® systems and the new line of Power Macintosh™ systems. MacBench 1.1 is also useful for purchase evaluations during the period of transition from 680x0-based Macintoshes to Power Macintoshes. With MacBench 1.1, you can assess the effects of such system changes as running extensions and drivers in native or emulation modes, using video accelerator cards, and so on.
This Read Me file describes the differences between versions 1.0 and 1.1 of MacBench, but it does not restate information in the manual, "Understanding and Using MacBench™ 1.0," that comes with MacBench.
This Read Me file also contains information on enhancements in MacBench 1.1 that affect 680x0-based Macintoshes.
This Read Me file is the only documentation for the MacBench 1.1 release; as a result, it is LONG. Read this file before running MacBench 1.1. We also encourage you to refer to the MacBench 1.0 manual.
Before you run MacBench, you must first read and agree to the following license agreement.
LICENSE AGREEMENT FOR ZIFF-DAVIS' MACBENCH™ 1.1
READ THIS AGREEMENT CAREFULLY BEFORE USING THE SOFTWARE EMBODIED IN THE MACBENCH™ 1.1 DISKETTE (OR, IF DOWNLOADED, IN THE DOWNLOADED FILE(S)). Embodied in the MacBench 1.1 diskette ("diskette") (or, if downloaded, in the downloaded file(s)) is the MacBench 1.1 computer program and related documentation (the "Software"). Ziff-Davis Publishing Company, L.P., having a place of business at One Park Avenue, New York, New York 10016 ("Ziff") is the licensor under this Agreement and you are the licensee. By using the Software, in whole or in part, you agree to be bound by the terms of this Agreement. If you do not agree to the terms of this Agreement, promptly return the Software (or, if downloaded, delete the Software) to the Ziff-Davis Benchmark Operation at One Copley Parkway, Suite 510, Morrisville, North Carolina 27560. Title to the Software and all copyrights, trade secrets and other proprietary rights therein are owned by Ziff. All rights therein, except those expressly granted to you in this Agreement, are reserved by Ziff.
1. Limited License
This Agreement grants you only limited rights to use the Software. Ziff grants you a non-exclusive, non-transferable license to use the Software on a single dedicated Macintosh® computer or on a file server networked with multiple Macintosh computers for the sole purpose of conducting benchmark tests to measure the performance of Macintosh computer hardware and operating system configurations. You have the right to make a single copy of the Software for archival purposes and the right to transfer a copy of the Software across a network only to the computers attached to the network. Ziff hereby grants you the right to publish benchmark test results obtained by you from your use of the Software provided that with the publication of each such result you:
A. Identify Ziff-Davis, the name and version number of the benchmark Software used (i.e., Ziff-Davis' MacBench™ Ver. 1.1);
B. Identify the exact system unit name, amount of secondary RAM cache, if any, CPU accelerator processor speed and type, if any, amount of RAM, hard disk size, and size of hardware hard disk cache, if any, of the computer used for the test (e.g., Apple® Macintosh IIci with 32 KB external RAM cache, WXY Corp. 33-MHz 68040 XYZ CPU accelerator, 8 MB of RAM, and 80 MB hard disk, or Apple Power Macintosh 6100 with 60-MHz PowerPC™ 601 CPU, 8 MB of RAM, and 250 MB hard disk);
C. Identify the exact display name, video card name (or on-board video, if applicable), amount and type of video RAM, display resolution, and color depth, (e.g., Apple Macintosh 21-inch Color Display using on-board video with 1 MB of VRAM and running at 1152 by 870 pixels with 256 colors );
D. Identify the operating system version (e.g., System 7.1), size and type of software disk cache, if any (e.g., 32 KB System 7.1 disk cache), and any other special conditions used to achieve the result (e.g., disk compression utility ABC version 1.1 enabled);
E. State that all products used in the test were shipping versions available to the general public;
F. State that the test was performed without independent verification by Ziff; and
G. Follow proper trademark usage and acknowledge Ziff's trademark rights (e.g., "[ ] achieved a Processor score of 39.89 on the MacBench™ 1.1 Processor test. MacBench is a trademark of Ziff-Davis Publishing Company, L.P.").
This Agreement and your rights hereunder shall automatically terminate if you fail to comply with any provision of this Agreement. Upon such termination, you agree to cease all use of the Software, to delete the Software and to destroy all copies of the diskette and other materials contained in this package in your possession or under your control, or, if downloaded, to destroy any and all copies of the Software in your possession or under your control.
2. Additional Restrictions
A. You shall not (and shall not permit other persons or entities to) rent, lease, sell, sublicense, assign, or otherwise transfer the Software or this Agreement. Any attempt to do so shall be void and of no effect.
B. You shall not (and shall not permit other persons or entities to) reverse engineer, decompile, disassemble, merge, modify, include in other software or translate the Software, or use the Software for any commercial purposes, except for the publication of test results, as provided above.
C. You shall not (and shall not permit other persons or entities to) remove or obscure Ziff's copyright, trademark or other proprietary notices or legends from any of the materials contained in this package or downloaded.
3. Limited Warranty and Limited Liability
THE SOFTWARE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTY OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE RESULTS AND PERFORMANCE OF THE SOFTWARE IS ASSUMED BY YOU, AND ZIFF ASSUMES NO RESPONSIBILITY FOR THE ACCURACY OR APPLICATION OF OR ERRORS OR OMISSIONS IN THE SOFTWARE. IN NO EVENT SHALL ZIFF BE LIABLE FOR ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE SOFTWARE, EVEN IF ZIFF HAS BEEN ADVISED OF THE LIKELIHOOD OF SUCH DAMAGES OCCURRING. ZIFF SHALL NOT BE LIABLE FOR ANY LOSS, DAMAGES OR COSTS, ARISING OUT OF, BUT NOT LIMITED TO, LOST PROFITS OR REVENUE, LOSS OF USE OF THE SOFTWARE, LOSS OF DATA OR EQUIPMENT, THE COSTS OF RECOVERING SOFTWARE, DATA OR EQUIPMENT, THE COST OF SUBSTITUTE SOFTWARE OR DATA, CLAIMS BY THIRD PARTIES, OR OTHER SIMILAR COSTS.
THE ONLY WARRANTY MADE BY ZIFF IS THAT THE ORIGINAL PHYSICAL MEDIA IN WHICH THE SOFTWARE IS EMBODIED AND WHICH IS DISTRIBUTED BY ZIFF SHALL BE FREE OF DEFECTS IN MATERIALS AND WORKMANSHIP. ZIFF'S ENTIRE LIABILITY AND THE USER'S EXCLUSIVE REMEDY SHALL BE LIMITED TO THE REPLACEMENT OF THE ORIGINAL PHYSICAL MEDIA IF DEFECTIVE. THE WARRANTIES AND REMEDIES SET FORTH HEREIN ARE EXCLUSIVE AND IN LIEU OF ALL OTHERS, ORAL OR WRITTEN, EXPRESS OR IMPLIED. NO ZIFF AGENT OR EMPLOYEE, OR THIRD PARTY, IS AUTHORIZED TO MAKE ANY MODIFICATION OR ADDITION TO THIS WARRANTY.
SOME STATES DO NOT ALLOW EXCLUSION OR LIMITATION OF IMPLIED WARRANTIES OR LIMITATION OF LIABILITY FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES; SO THE ABOVE LIMITATIONS OR EXCLUSIONS MAY NOT APPLY TO YOU.
4. U.S. Government Restricted Rights.
The Software is licensed subject to RESTRICTED RIGHTS. Use, duplication or disclosure by the Government or any person or entity acting on its behalf is subject to restrictions as set forth in subdivision (c)(1)(ii) of the Rights in Technical Data and Computer Software Clause at DFARS (48 CFR 252.227-7013) for DoD contracts, in paragraphs (c)(1) and (2) of the Commercial Computer Software-Restricted Rights clause in the FAR (48 CFR 52.227-19) for civilian agencies, on in other comparable agency clauses. The contractor/manufacturer is the Ziff-Davis Benchmark Operation, One Copley Parkway, Suite 510, Morrisville, North Carolina 27560.
5. General Provisions
Nothing in this Agreement constitutes a waiver of Ziff's rights under U.S copyright laws or any other Federal, state, local or foreign law. You are responsible for installation, management, and operation of the Software. This Agreement shall be construed, interpreted and governed under New York law. If any provision of this Agreement shall be held by a court of competent jurisdiction to be illegal, invalid or unenforceable, the remaining provisions shall remain in full force and effect.
INSTALLING MACBENCH™ VERSION 1.1
You can install MacBench 1.1 from a diskette or a file you downloaded from ZiffNet, a Ziff-Davis on-line service. MacBench 1.1 comes as a self-extracting archive.
NOTE: If you get MacBench 1.1 on a diskette, the name of the file is MacBench 1.1.sea. If you download MacBench 1.1 from ZiffNet, the name of the file is MACBEN.SEA. The application package — which includes the MacBench 1.1 application, this Read Me file, manual, test resources, sample test results, and test suites — requires approximately 2.6 MB of hard disk space. Each version of the manual (Microsoft® Word 5.1 and MacWrite® II) takes an additional approximately 350 KB of disk space. If you are short on disk space, you can delete the version(s) of the manual you do not need.
To install MacBench 1.1:
1. If you have MacBench on a diskette, insert the diskette into the diskette drive and double-click on the diskette's icon. If you downloaded MacBench from ZiffNet, make sure the MACBEN.SEA file is on your hard drive.
2. Double-click on the .sea icon for MacBench (MacBench 1.1.sea or MACBEN.SEA).
3. When the Aladdin Systems copyright window appears, click on Continue.
4. At the dialog box that appears, enter the name you want the MacBench folder to have. (The system will warn you if you give the name of an existing folder.) You can also specify in the box a location for the folder. You must always install MacBench on your hard drive.
NOTE: You can name this folder anything you want; we recommend, however, that you name the folder MacBench 1.1.
5. Click on Save.
6. When the dialog box appears saying the installation was successful, click on Quit.
When the installation finishes, the MacBench folder appears on your hard drive. To save space on your hard disk, you can now delete the MACBEN.SEA file.
IMPORTANT MACBENCH 1.1 NOTES
Read these notes before running MacBench 1.1.
1) You must read and agree to the license information at the front of this Read Me file before running MacBench. The same licensing information appears on the screen of your Macintosh® or Power Macintosh™ when you first run MacBench and in the on-line documentation. If you do not agree to the licensing information, delete all copies of MacBench. If the Ziff-Davis Benchmark Operation (ZDBOp) sent you the MacBench diskette, return it and all accompanying materials (including any documentation) to ZDBOp at the following address:
Ziff-Davis Benchmark Operation
One Copley Parkway, Suite 510
Morrisville, NC 27560
2) To run, MacBench requires:
• A Macintosh Plus or later running System 7 or later.
• A minimum of 4 MB of RAM.
• At least 8 MB of disk space to run the disk tests.
The disk tests give best results with 32 MB of free disk space; you can, however, run the disk tests with only 8 MB of free disk space.
If your 680x0-based Macintosh has a floating-point unit (FPU), MacBench 1.1 includes a test to measure its performance. For more information, see note 11 below.
3) MacBench will not run on System 6. If you try to launch MacBench while running System 6, MacBench will immediately quit.
4) In general, MacBench 1.0 and MacBench 1.1 test results are comparable. If you run MacBench 1.0 and 1.1 on a 680x0-based Macintosh, you should get similar results and scores.
The method MacBench 1.0 used to time its tests, however, unfairly penalized Power Macintosh systems. The penalty was the result of the effects of the Power Macintosh's 68040 LC emulation on the timing method MacBench 1.0 used. MacBench 1.1 corrects this problem. Due to this correction, MacBench 1.1 emulation test results on a Power Macintosh will be slightly higher than the results of MacBench 1.0 on the same Power Macintosh. For more information on the Power Macintosh's native and emulation modes, refer to note 8 in this section.
IMPORTANT! To get accurate emulation results on a Power Macintosh, always use the MacBench 1.1 emulation tests.
5) Power Macintosh systems use the new PowerPC reduced instruction set computing (RISC) CPUs. MacBench 1.1 lets you obtain realistic, reliable performance measurements on the Power Macintosh.
If you are using a 680x0-based Macintosh, this update will benefit you through minor enhancements. Scores you get using MacBench 1.1 on a 680x0 Macintosh are comparable to the scores MacBench 1.0 produced on the same systems.
In addition, you can use the benchmark to compare MacBench 1.1 results from both Power Macintoshes and 680x0-based Macintoshes.
6) If you plan to test systems with 24-bit video adapters, you may have to increase MacBench's memory partition size by about 100 KB. Otherwise, the system could freeze when running MacBench, and you would have to reboot.
To increase MacBench's memory partition size, quit MacBench if it is open, and then select the MacBench icon. Choose Get Info from the File menu to display the MacBench Info window. In the Memory requirements section of the window, highlight the number displayed in either the Minimum size box or the Preferred size box. Then, type the number of kilobytes of memory you want MacBench to set aside for itself and close the MacBench Info window.
MacBench 1.1's default minimum and preferred memory size is 2048 KB. After adding 100 KB to either the Minimum or Preferred sizes, the memory size should be 2148 KB.
7) MacBench's Video and Processor tests run a fixed length of time on all Macintoshes, regardless of system speed. For example, each individual Video test lasts 15 seconds, while the CPU test lasts 30 seconds and the FPU test lasts 50 seconds.
The time the Disk tests take to run, however, depends on your system's speed. Specifically, the Sequential Read and Write 1 KB and 8 KB tests may take a while to run. The MacBench base machine, a Macintosh LC III, runs the Sequential Read 1 KB test in less than four minutes, the Sequential Read 8 KB test in less than 2 minutes, the Sequential Write 1 KB test in less than 11 minutes, and the Sequential Write 8 KB test in less than 3 minutes. The LC III ran all the Disk tests in 45 minutes.
Because the Disk tests move a great deal of data to and from the disk, setting the system disk cache size to a high value causes the Disk tests to run for a longer period of time. (Generally, the higher the disk cache setting, the lower your Disk scores. On the MacBench base machine, an LC III, the system disk cache is set at 32 KB.) Use the Memory control panel to set the disk cache size. For best performance results, we recommend you set your Macintosh's disk cache size to 32 KB before running the Disk tests.
The slowest Macintosh in the ZDBOp Research Center, a Macintosh SE, runs all the MacBench tests in about two hours.
8) To understand how MacBench works on Power Macintoshes, you need to understand the difference between the two main types of Macintosh software a Power Macintosh can run: native and emulation.
"Native" refers to software compiled in PowerPC machine language to run directly on the IBM/Motorola PowerPC CPU. Native applications for a Power Macintosh typically will run faster than comparable applications on a 680x0-based Macintosh.
"Emulation" refers to software compiled in 680x0 machine language to run on Motorola 680x0- based Macintoshes. While PowerPC and 680x0 machine languages are different, Power Macintoshes can still run Macintosh programs written for the 680x0 CPU by using an emulator program to translate the 680x0 instructions into PowerPC instructions. So, while a 680x0-based application is running on a Power Macintosh, the system is translating the application into PowerPC instructions.
This translation process is invisible, but it does affect the speed of 680x0-based Macintosh programs. You can run MacBench 1.1’s tests in emulation mode on a Power Macintosh to determine how quickly your Power Macintosh will run existing 680x0-based applications.
9) When you start MacBench 1.1, it will detect whether it is running on a 680x0-based Macintosh or a Power Macintosh and then run the appropriate code. Power Macintosh users can select whether they want MacBench to run its tests in native or emulation mode.
If you’re running MacBench 1.1 on a 680x0-based Macintosh, you will see no significant changes in the interface between MacBench 1.0 and MacBench 1.1. Note 12 of this section describes the minor changes to the System Information display. Refer to the MacBench 1.0 manual for more information on how to use the benchmark.
On Power Macintoshes, MacBench 1.1 by default launches its tests in native mode. Only the MacBench 1.1 tests, however, run native; the MacBench 1.1 user interface — which includes the menus and the buttons — runs in emulation.
If you want to run MacBench 1.1 tests in emulation mode on a Power Macintosh, double-click on the Power Macintosh Emulation icon in the MacBench 1.1 folder or drag the Power Macintosh Emulation icon onto the MacBench 1.1 icon. A dialog box then asks you to confirm that you want to run MacBench's tests in emulation mode. If you do, select the Emulation button. If you want to run MacBench's tests in native mode, select the Native button. The dialog box disappears, and MacBench opens in the mode you selected. You will see this dialog box every time you launch MacBench in emulation mode.
The Power Macintosh Emulation icon exists for your convenience in running the MacBench tests in emulation mode. You can move the Power Macintosh Emulation icon to any folder on your hard disk. It does not have to be in the MacBench 1.1 folder. You can use the Power Macintosh Emulation icon no matter where it is on your hard disk. When you double-click on it or drag it onto the MacBench 1.1 icon, MacBench displays the dialog box described above.
Native and emulation modes only have meaning for Power Macintoshes; 680x0-based Macintoshes do not have these different modes of operation. As a result, MacBench 1.1 does not offer tests in native or emulation modes for 680x0-based Macintoshes. If you are running MacBench 1.1 on a 680x0-based Macintosh, double-clicking on the Power Macintosh Emulation icon launches MacBench as if you had double-clicked on the MacBench 1.1 icon.
10) MacBench 1.1 on a Power Macintosh does not display whether it is running its tests in emulation or native mode. If you want to see the mode in which MacBench's tests are running, select System Info from the Tests menu or click on the System Information button. The next-to-last category in the General Info display is "Power Macintosh mode." For Power Macintoshes, this entry appears as either "Native" or "Emulation." For 680x0 Macintoshes, this entry appears as "N/A."
11) MacBench 1.1, like MacBench 1.0, detects whether your 680x0-based Macintosh has a hardware floating point unit (FPU). If your system doesn't contain an FPU, MacBench offers a software-only version of its FPU test. If your system does have an FPU, MacBench offers you the choice of running the software FPU test or the hardware FPU test.
For 680x0 Macintoshes, MacBench 1.1's FPU tests return the same scores and results as the MacBench 1.0 tests.
The situation is a little different for Power Macintoshes. Because all generations of the PowerPC CPU include a built-in FPU, a native-mode software FPU test has no meaning. As a result, if you are running MacBench 1.1 tests on a Power Macintosh in native mode, you can run only the hardware version of the test.
If you are running MacBench 1.1 tests on a Power Macintosh in emulation mode, you can run only the software version of the FPU test. This is because the Power Macintosh's emulator program does not translate hardware FPU instructions in existing 680x0 applications into instructions the PowerPC's built-in FPU can use.
12) MacBench 1.1’s System Information display now reports the following additional information pertinent to the Power Macintosh.
Processor Type: MacBench 1.1 now recognizes Power Macintosh CPUs in the Processor Type category of the General Info display. MacBench 1.1 recognizes the following types of processors:
• all 680x0-series CPUs
• all PowerPC-series CPUs
MMU Type: MacBench 1.1 recognizes the following types of Power Macintosh memory management units (MMU) in the General Info display:
• 601 built-in MMU
• 603 built-in MMU
• 604 built-in MMU
• 620 built-in MMU
Power Macintosh Test Mode: In the General Info display, the next-to-last line in the leftmost column reads "Power Macintosh Mode." For a 680x0-based Macintosh, the entry for this category will always be "N/A." For a Power Macintosh, the entry will be either "Native" or "Emulation." In MacBench 1.0 results files, these fields are blank.
Modern Memory Manager: In the General Info display, the last line in the leftmost column reads "Modern Memory Manager." For a 680x0-based Macintosh, the entry for this category is "N/A." For a Power Macintosh, the entry will be either "Active" or "Inactive." In MacBench 1.0 results files, these fields are blank.
13) MacBench 1.1’s interface — the buttons, results window, menu choices, dialog boxes, and so on — has not changed significantly from that of MacBench 1.0. On Power Macintoshes, the interface runs in emulation mode; the native tests, however, run in native mode.
MacBench's tests, such as Disk, Video, and Processor are resources in a Test Module folder. (The MacBench installation process placed a Test Module folder inside your MacBench 1.1 folder.)
These test resources are "fat" code resources, meaning they contain code that lets them run in either native or emulation mode. Each test resource has two parts: one with the test routines compiled for 680x0 CPUs, and one with the test routines compiled for PowerPC CPUs.
14) MacBench 1.0 did not let you select optical or CD-ROM drives for the Disk tests. In MacBench 1.1, you can select any SCSI drive connected to your Macintosh, including optical drives.
Note, however, that while you can select any drive, you cannot run MacBench 1.1's Disk tests on all drives. The reason is that MacBench's Disk tests write a file on the disk device they are testing, something that is not possible on read-only drives. In MacBench 1.1, therefore, you can select a CD-ROM drive, but you will not be able to run the Disk tests on it.
15) All the test results files in this release of MacBench are MacBench 1.1 results files and are in the MacBench Results folder. MacBench 1.1 includes the following additional files that were not in the MacBench 1.0 release:
• Power 6100/60
• Power 6100/60 w/L2
• Power 6100/60 w/AV
• Power 7100/66
• Power 7100/66 w/L2
• Power 7100/66 w/AV
• Power 8100/80
• Power 8100/80 w/AV
• Quadra 610
• Quadra 650
• Quadra 660AV — config 1
• Quadra 660AV — config 2
• Quadra 800
• Quadra 840AV — config 1
• Quadra 840AV — config 2
• Quadra 950 — config 1
• Quadra 950 — config 2
("Config 1" refers to a ZDBOp Research Center configuration of this system. "Config 2" refers to a MacUser Labs configuration of this system. The "w/L2" refers to a system configuration using Apple's unified 256 KB level-two cache card. The "w/AV" refers to a system configuration using an Apple AV-based video card.)
16) MacBench 1.0's online help contained some scrambled text, misspellings, and other errors. MacBench 1.1 corrects these errors in the online help.
17) MacBench 1.1 requires that you reread and agree to the License Agreement when you first start MacBench 1.1.
18) MacBench 1.1 looks for the MacBench Preferences file in the System folder's Preferences folder. If the MacBench preferences file is not in that folder, or if a Preferences folder does not exist, MacBench looks for the file in the MacBench 1.1 folder.
KNOWN PROBLEMS IN MACBENCH 1.1
1) The MacBench 1.0 Read Me file stated, "The current set of test results cannot be removed from the Results window." It is more accurate to say you cannot remove any Current System test results after you have run a test or tests.
2) When running the Disk tests, occasionally one test in a test suite may return a number about 10-20% lower than expected. This problem seems to happen most on 68040-based machines, such as the Centris 610 and Quadra 900. You can tell if an unusual test result exists by running the Disk tests several times and then comparing the scores.
3) MacBench may crash on a Macintosh LC III or LC 520 after completing the Video Scroll test, leaving the system frozen and the test results unrecorded. If this happens, you will need to reboot the Macintosh.
4) If you select Run Demo and run MacBench for an extended period of time, the program may eventually freeze. To work around this problem, either increase MacBench's memory partition size by about 100 KB, or restart MacBench every two hours or so. To stop the demo, press Command-period. MacBench should quit after it finishes running the current test.
5) Pressing Command-period does not always cancel a sequence of tests after the current test has finished, but even then MacBench will stop after completing all its tests.
6) When your Macintosh is not connected to a printer, MacBench does not correspondingly gray out the File menu's Page Setup and Print Graph options.
7) Under certain unusual conditions, the result graph bars in MacBench's main window may disappear. This can occur for certain types of result files (such as those for a Macintosh LC III), when many results (from 8 to 10) are open, when a slow machine (such as the Macintosh SE) is the 100% System, and when a range of 300% or less is active.
8) The MacBench application window, which holds the function buttons, contains a System Information button, an icon representing your Macintosh model. Beside the button is the model and model number of your Macintosh, the Macintosh's RAM size, hard disk size, and the System version it is running. If, however, your Macintosh has no SCSI 0 device (the SCSI number for an internal hard disk), this display will report a 0 MB hard disk size.
9) If you have a small amount of free disk space on the hard disk you're testing, and you choose Run All Tests from the Tests menu, MacBench displays the following message: "An error 8001 occurred while executing the last test" for each Disk test. The test fails, and you return to MacBench.
10) The following sequence of actions will cause MacBench to close with an error type of 1 on the Macintosh IIsi, the Performa 200 and 400, and the PowerBook 170:
• choose System Information to display the System Information window
• close the window
• choose Settings
• close the Settings window
• choose System Information again
• select SCSI Info from the pull-down menu
This behavior is most likely to occur after a clean start of MacBench, but not after adding or manipulating results.
11) On some Macintoshes, the CPU test may occasionally cause the machine to hang. Should this problem occur, you can work around it by increasing MacBench's memory allocation size by a megabyte and then running the CPU test. You can then return the memory allocation size to its original value.
12) MacBench 1.1 displays the menu bar when running any Video test in native mode on a Power Macintosh. This behavior does not affect the results of any test.
13) For Macintosh SE users, MacBench 1.1 corrects a MacBench 1.0 reporting error that affected the CopyBits Large, CopyBits Small, and Video Mix tests. The error affected only the reporting of the test results, not the results themselves. SE users will notice that MacBench 1.1 will report higher results for these tests than did MacBench 1.0. The Macintosh SE results file in this release of MacBench is a MacBench 1.1 results file that has correct values.
14) Under the following conditions, MacBench does not test the disk you selected:
• Select a SCSI device that is not the startup disk.
• From the pop-up menu beside the Run function button, select All Disk.
• Click the Run function button.
Under these conditions, MacBench executes all the Disk tests on the startup device, but not on the device you selected.
You can work around this behavior by selecting the non-startup device and then choosing from the Tests menu either All Tests or any Disk test.
ON-LINE DOCUMENTATION
MacBench comes with the following on-line documentation:
Read Me
This file contains information about licensing MacBench, which you must do before you run MacBench. This file also tells you how to install MacBench and provides any last-minute notes about MacBench that do not appear in the manual. (You are reading this file now.)
MacBench Doc (MS Word)
This is a Microsoft® Word (version 5.1) copy of the manual. If you have Word on your Macintosh, you can print a formatted copy of the manual. If you have a word-processing application that accepts Word documents, you may be able to print a copy of the manual that retains much of its formatting.
MacBench Doc (MacWrite)
This is a version of the Microsoft Word manual that ZDBOp saved in MacWrite® II format. This version of the manual contains the same technical information as the original manual and much of the original document format. Because the conversion to MacWrite II affected the manual's page formats and page numbers, however, this version does not include page numbers in the table of contents. If you have access to MacWrite and not Word, you should be able to print this version of the manual.
TECHNICAL SUPPORT
If you have problems installing or running MacBench 1.1, please contact ZDBOp. You can contact ZDBOp in three ways:
1. If you have a modem and communications software, you can reach ZDBOp via ZiffNet, a Ziff-Davis on-line service. One way to access ZiffNet is by using CompuServe®. ZiffNet's ZDBOp forum (GO ZDBENCH) lets you communicate your technical questions and comments to Ziff-Davis. (Access to CompuServe is available for a fee.)
2. You can fax your questions and comments to MacBench Technical Support at the ZDBOp benchmark request fax number:
919-380-2879
3. You can mail your questions and comments to ZDBOp at the following address:
Ziff-Davis Benchmark Operation
One Copley Parkway, Suite 510
Morrisville, NC 27560
Attn.: MacBench Technical Support
TRADEMARKS
Apple® and Macintosh® are U.S. registered trademarks and Power Macintosh™ is a trademark of Apple Computer, Inc.
Compuserve® is a registered trademark of CompuServe Incorporated.
MacBench™ is a trademark of Ziff-Davis Publishing Company, L.P.
MacWrite® is a U.S. registered trademark of Claris Corporation.
Microsoft® is a U.S. registered trademark of Microsoft Corporation.
PowerPC™ is a registered trademark of International Business Machines Corporation.
StuffIt Deluxe™ is a trademark of Aladdin Systems Inc. and Raymond Lau.