-
Bug
-
Resolution: Duplicate
-
P4
-
None
-
1.2.0
-
x86
-
windows_nt
/*
======================================================================
Here is a new test program (based on the original) which will
measure the time spent in getToolkit() and getImage() separately.
The original program is at the bottom of the Description section.
tim.bell@Eng 1999-06-25
*/
/*
Running on a WinNT 400Mhz P2 system, I've noticed that
many of the Image operations are much (up to 50%)
slower while using JDK1.2rc1 as opposed to using
JDK1.2b4 for the same operations (i.e. getImage
and drawImage). The example program shows
loading speed for an image. A similar test using
drawImage will give similar results, RC1 is
significantly slower that beta4.
To run the following src, place a gif image in
the same directory as ImageSpeed.java (my image
was 490K in size). Name the gif test.gif. Set up
your environment to run jdk1.2beta4 and run the
test. Set up your environment to run jdk1.2rc1 and
run the test. You should see that jdk1.2rc1 is
considerably slower...
*/
import java.awt.*;
import java.util.Date;
import java.awt.MediaTracker;
public class ImageSpeed2 extends Component {
public static void outputMessage() {
System.out.println("java.version = " +
System.getProperty("java.version"));
System.out.println("Operating System Name = " +
System.getProperty("os.name"));
System.out.println("Operating system architecture = " +
System.getProperty("os.arch"));
System.out.println("Operating system version = " +
System.getProperty("os.version") + "\n");
}
public final static void main(String args[]) {
ImageSpeed2 imSpeed = new ImageSpeed2 ();
MediaTracker tracker = new MediaTracker(imSpeed);
Date toolkitStart;
Date toolkitEnd;
Date getImageStart;
Date getImageEnd;
long t1, t2, gt;
long t3, t4, lt;
outputMessage();
toolkitStart = new Date();
Toolkit tk = imSpeed.getToolkit();
toolkitEnd = new Date();
getImageStart = new Date();
// Image im = tk.getImage("test.gif");
Image im = tk.getImage("test.jpg");
tracker.addImage(im, 0);
try {
tracker.waitForID(0);
}
catch(InterruptedException e) {
System.exit(0);
}
getImageEnd = new Date();
t1 = toolkitStart.getTime();
t2 = toolkitEnd.getTime();
gt = t2 - t1;
System.out.println("getToolkit took: " + t2 + " - " + t1 +
" = " + gt + " milliseconds");
t3 = getImageStart.getTime();
t4 = getImageEnd.getTime();
lt = t4 - t3;
System.out.println("Load test took: " + t4 + " - " + t3 +
" = " + lt + " milliseconds");
System.out.println("Total time spent: " + (gt + lt) + " milliseconds");
System.out.println(System.getProperty("line.separator"));
System.exit(0);
}
}
/* End of ImageSpeed2.java */
======================================================================
Original test program appears below:
Name: tb29552 Date: 11/18/98
/*
Running on a WinNT 400Mhz P2 system, I've noticed that
many of the Image operations are much (up to 50%)
slower while using JDK1.2rc1 as opposed to using
JDK1.2b4 for the same operations (i.e. getImage
and drawImage). The example program shows
loading speed for an image. A similar test using
drawImage will give similar results, RC1 is
significantly slower that beta4.
To run the following src, place a gif image in
the same directory as ImageSpeed.java (my image
was 490K in size). Name the gif test.gif. Set up
your environment to run jdk1.2beta4 and run the
test. Set up your environment to run jdk1.2rc1 and
run the test. You should see that jdk1.2rc1 is
considerably slower...
*/
import java.awt.*;
import java.util.GregorianCalendar;
import java.util.Calendar;
import java.util.Date;
import java.awt.MediaTracker;
public class ImageSpeed extends Component {
public static void outputMessage(){
System.out.println("java.version = " +
System.getProperty("java.version"));
System.out.println("Operating System Name = " +
System.getProperty("os.name"));
System.out.println("Operating system architecture = " +
System.getProperty("os.arch"));
System.out.println("Operating system version = " +
System.getProperty("os.version") + "\n");
}
public final static void main(String args[]) {
ImageSpeed imSpeed = new ImageSpeed ();
MediaTracker tracker = new MediaTracker(imSpeed);
GregorianCalendar c = new GregorianCalendar();
Date start;
Date end;
long t1, t2, t;
outputMessage();
start = new Date();
//Image im = imSpeed.getToolkit().getImage("test.gif");
Image im = imSpeed.getToolkit().getImage("test.jpg");
tracker.addImage(im, 0);
try {
tracker.waitForID(0);
}
catch(InterruptedException e) {
System.exit(0);
}
end = new Date();
c.setTime(start);
t1 = c.get(Calendar.SECOND) * 1000;
t1 += c.get(Calendar.MILLISECOND);
c.setTime(end);
t2 = c.get(Calendar.SECOND) * 1000;
t2 += c.get(Calendar.MILLISECOND);
t = t2 - t1;
System.out.println("Load test took " + t2 + " - " + t1 + " = " + t + " milliseconds");
System.exit(0);
}
}
(Review ID: 42621)
======================================================================
======================================================================
Here is a new test program (based on the original) which will
measure the time spent in getToolkit() and getImage() separately.
The original program is at the bottom of the Description section.
tim.bell@Eng 1999-06-25
*/
/*
Running on a WinNT 400Mhz P2 system, I've noticed that
many of the Image operations are much (up to 50%)
slower while using JDK1.2rc1 as opposed to using
JDK1.2b4 for the same operations (i.e. getImage
and drawImage). The example program shows
loading speed for an image. A similar test using
drawImage will give similar results, RC1 is
significantly slower that beta4.
To run the following src, place a gif image in
the same directory as ImageSpeed.java (my image
was 490K in size). Name the gif test.gif. Set up
your environment to run jdk1.2beta4 and run the
test. Set up your environment to run jdk1.2rc1 and
run the test. You should see that jdk1.2rc1 is
considerably slower...
*/
import java.awt.*;
import java.util.Date;
import java.awt.MediaTracker;
public class ImageSpeed2 extends Component {
public static void outputMessage() {
System.out.println("java.version = " +
System.getProperty("java.version"));
System.out.println("Operating System Name = " +
System.getProperty("os.name"));
System.out.println("Operating system architecture = " +
System.getProperty("os.arch"));
System.out.println("Operating system version = " +
System.getProperty("os.version") + "\n");
}
public final static void main(String args[]) {
ImageSpeed2 imSpeed = new ImageSpeed2 ();
MediaTracker tracker = new MediaTracker(imSpeed);
Date toolkitStart;
Date toolkitEnd;
Date getImageStart;
Date getImageEnd;
long t1, t2, gt;
long t3, t4, lt;
outputMessage();
toolkitStart = new Date();
Toolkit tk = imSpeed.getToolkit();
toolkitEnd = new Date();
getImageStart = new Date();
// Image im = tk.getImage("test.gif");
Image im = tk.getImage("test.jpg");
tracker.addImage(im, 0);
try {
tracker.waitForID(0);
}
catch(InterruptedException e) {
System.exit(0);
}
getImageEnd = new Date();
t1 = toolkitStart.getTime();
t2 = toolkitEnd.getTime();
gt = t2 - t1;
System.out.println("getToolkit took: " + t2 + " - " + t1 +
" = " + gt + " milliseconds");
t3 = getImageStart.getTime();
t4 = getImageEnd.getTime();
lt = t4 - t3;
System.out.println("Load test took: " + t4 + " - " + t3 +
" = " + lt + " milliseconds");
System.out.println("Total time spent: " + (gt + lt) + " milliseconds");
System.out.println(System.getProperty("line.separator"));
System.exit(0);
}
}
/* End of ImageSpeed2.java */
======================================================================
Original test program appears below:
Name: tb29552 Date: 11/18/98
/*
Running on a WinNT 400Mhz P2 system, I've noticed that
many of the Image operations are much (up to 50%)
slower while using JDK1.2rc1 as opposed to using
JDK1.2b4 for the same operations (i.e. getImage
and drawImage). The example program shows
loading speed for an image. A similar test using
drawImage will give similar results, RC1 is
significantly slower that beta4.
To run the following src, place a gif image in
the same directory as ImageSpeed.java (my image
was 490K in size). Name the gif test.gif. Set up
your environment to run jdk1.2beta4 and run the
test. Set up your environment to run jdk1.2rc1 and
run the test. You should see that jdk1.2rc1 is
considerably slower...
*/
import java.awt.*;
import java.util.GregorianCalendar;
import java.util.Calendar;
import java.util.Date;
import java.awt.MediaTracker;
public class ImageSpeed extends Component {
public static void outputMessage(){
System.out.println("java.version = " +
System.getProperty("java.version"));
System.out.println("Operating System Name = " +
System.getProperty("os.name"));
System.out.println("Operating system architecture = " +
System.getProperty("os.arch"));
System.out.println("Operating system version = " +
System.getProperty("os.version") + "\n");
}
public final static void main(String args[]) {
ImageSpeed imSpeed = new ImageSpeed ();
MediaTracker tracker = new MediaTracker(imSpeed);
GregorianCalendar c = new GregorianCalendar();
Date start;
Date end;
long t1, t2, t;
outputMessage();
start = new Date();
//Image im = imSpeed.getToolkit().getImage("test.gif");
Image im = imSpeed.getToolkit().getImage("test.jpg");
tracker.addImage(im, 0);
try {
tracker.waitForID(0);
}
catch(InterruptedException e) {
System.exit(0);
}
end = new Date();
c.setTime(start);
t1 = c.get(Calendar.SECOND) * 1000;
t1 += c.get(Calendar.MILLISECOND);
c.setTime(end);
t2 = c.get(Calendar.SECOND) * 1000;
t2 += c.get(Calendar.MILLISECOND);
t = t2 - t1;
System.out.println("Load test took " + t2 + " - " + t1 + " = " + t + " milliseconds");
System.exit(0);
}
}
(Review ID: 42621)
======================================================================
- duplicates
-
JDK-4202846 Solaris Font code is inefficient
-
- Resolved
-