Odd Draw Behavior - Slow Ellipse2D Draw

Summary:
A game I have written uses bitmaps and lots of Ellipse2Ds. The Ellipses are in the background and occasionally need to all be redrawn (there could be many - let's say 100 at a time). This works fine on a Windows box, and Macs from G3 - current Intels. When I attempt to run the application full screen under X11 drawing the Ellipses takes too long, making the animation appear choppy. Running Java 5.0 under Debian.
Question
I am looking for suggestions on how to tweak this and/or a work around.
Details
I have tried -Xmx256M -Xsx256M switches (makes no diff). I am using a descent video card on the system. The Ellipse2Ds are drawn on top of a buffered image (as are the bitmaps). Garbage collection does not seem to be the issue. The processor is a little slower - its a VIA mini ITX - details here ...
http://www.hardwareinreview.com/cms/content/view/44/1/
I understand that it could be the hardware ... but just in case ... feedback appreciated.

I don't think it's very odd at all. On some of my computers windows runs fine, but anyn version of Linux is slow at drawing in comparison. I think this has to do with the state of the ATI drivers Linux has. Check out Phoronix for more information on the state of various graphics drivers in Linux.
Hopefully some nice Linux developer somewhere will drastically improve the performance of all video drivers. :)

Similar Messages

Random freezing and odd keyboard behaviors on 10.5.8

My random freezing started on 10.5.6. Updated to 10.5.7 and the problem persist. I have decided to upgrade to 10.5.8. Everything seemed fine for a day and a half. Then all of a sudden random freezing when starting, login, wake from sleep, etc.. and now odd keyboard behaviors. function keys randomly work when computer is not frozen.
I have tried every option listed in the discussions here and elsewhere. Does anyone have the same problems?

I have tried so far:
disk permissions repaired from dvd and startup,
disk first aid,cleaned caches, virus checked, zapped the pram, reset power management, ran apple hardware check, ran disk warrior, trashed network prefs, deleted sleep image, replaced directory with techtool pro, tried new user account, re-updated, removed 3rd party software, and nothing works?????

Odd display behavior

Happy New Year to All!
I just recently upgraded to Tiger. Since that time, I have noticed some rather odd display behavior on my screen. For example, when I am on web site that has a "fill-in" box, once I highlight that box to fill it in, a black rectangle stays in that position on my screen until I click on the desktop.
Likewise, if I highlight an icon on the desktop, a black rectangle outlines the icon and stays visible until I either highlight something else or click randomly on the screen.
I never noticed this behavior under Panther (10.3.9) and while it's not a huge deal, it does seem odd. It's as if the screen doesn't refresh after moving on. Any ideas?
Thanks,
Brad
P.S. After posting this message, a black rectangle corresponding to the message edit box stayed on my screen. Irritating.

Hi William,
Yes, I tried both and booted / used my Tech Tool Deluxe repair disk. All no help. Don't know if they are related, but also have other odd things going on.
Is it normal to have two Restart selections, two Shutdown selections and two Logout Selections under the Apple menu in Tiger? I do. Didn't in Panther, but can't remember if I've always had them in Tiger. Should I?
Also, can't get my scanner (All-in-one, printer works) to run in Tiger. All of these problems go away when I boot into my 10.3.9 clone on my Firewire drive. I'd really hate to return to Panther, but ...
Basically, I'm getting some very odd behavior, from a system that has always been relatively bullit-proof until now. I do all the req'd housekeeping stuff, have both Macaroni and Cocktail. Don't know what to try.
Brad

Odd usage behavior

iPhone v101: this odd usage behavior just started half a week ago. This is the second time it happened. During standby, the usage time continues to count forward. I leave my phone on my desk not using it, but the usage still goes up. For example, the usage was 1 hour and 20 minutes when I checked like three hours ago, now it says 4 hours and some minutes. WiFi off, not running any apps, great signal and I cannot find anything running. This behavior also appear to drain the battery a little bit. The phone was not being charged or on plugged in anywhere; just sitting still on my desk. Now, the usage is back to normal and not accumulating time when I am not using it. This is strange. Have anyone else seem this happening? I had the phone for over a month now...

I didn't notice it causing any battery drain, but it sounds like I am charging it differently than you are. Here's how mine worked:
I would plug the iPhone into the wall charger at night to charge while I slept. In the morning, it would show a full charge by showing a plug on the small battery icon, but the large green battery would still have a sliver that wasn't green. The usage and standby stats would show something in the 3-4 hour range, presumably the time since it hit full charge.
I would disconnect the phone, check my email, and then take it to my desktop computer and plug it in to sync and update all my podcasts, etc. I would leave it plugged into my desktop while I showered while I got ready for work. By the time I was ready to leave, it would again show a full charge on the small icon, but now my stats would be reset to 0 and would function normally throughout the day. But nothing I did ever stopped the stats from running up during the night after attaining full charge.
1.0.1 got rid of the stats problem (they always stay at 0/0 in the morning with a full charge until I unplug the phone) AND the weird behavior of the large green battery indicator went away. Now both indicators show full charge in the morning.

Odd rotation behavior

The following code is supposed to rotated a rectangle about a certain point. It rotates about that point but the rotation appears to speed up. The catch is that I need to keep track of five points as the rectangle rotates (4 points of the rectangle and the center). Any idea why it's speeding up?
public void paint(Graphics2D g) {
          //save g's transform
          AffineTransform savedForm = g.getTransform();
          pos_rotate = center;
          angle += Math.toRadians(5);
          //transform the shape
          AffineTransform newForm = AffineTransform.getRotateInstance(angle, pos_rotate.x, pos_rotate.y);
          GeneralPath gp = new GeneralPath();
          gp.moveTo((float)frontLeft.x, (float)frontLeft.y);
          gp.lineTo((float)frontRight.x, (float)frontRight.y);
          gp.lineTo((float)backRight.x, (float)backRight.y);
          gp.lineTo((float)backLeft.x, (float)backLeft.y);
          brick = newForm.createTransformedShape(gp);
          //save the new coordinates
          Point2D.Double[] oldPoints = { frontLeft, frontRight, backLeft, backRight, center };
          Point2D.Double[] newPoints = { frontLeft, frontRight, backLeft, backRight, center };
          newForm.transform(oldPoints, 0, newPoints, 0, newPoints.length);
          //paint the shape
          g.transform(newForm);
          g.setColor(Color.YELLOW);
          g.draw(brick);
          g.fill(brick);
          //set the old transform back
          g.setTransform(savedForm);
}

This method signature is incorrect and will/should not be recognized/called
public void paint(Graphics2D g) {Java will look for this signature
public void paint(Graphics g) {The acceleration is being caused by this
angle += Math.toRadians(5);I changed over to Swing/paintComponent for this. Although you can use the paint method
in swing components it is tricky and takes some care/practice.
import java.awt.*;
import java.awt.event.*;
import java.awt.geom.*;
import javax.swing.*;
public class SpinTest extends JPanel {
    Point2D.Double
        frontLeft = new Point2D.Double(100, 300),
        frontRight = new Point2D.Double(300, 300),
        backRight = new Point2D.Double(300, 100),
        backLeft   = new Point2D.Double(100, 100),
        center     = new Point2D.Double(200, 200),
        pos_rotate;
    double angle = angle = Math.toRadians(5);
    protected void paintComponent(Graphics g) {
        super.paintComponent(g);
        Graphics2D g2 = (Graphics2D)g;
        g2.setRenderingHint(RenderingHints.KEY_ANTIALIASING,
                            RenderingHints.VALUE_ANTIALIAS_ON);
        //save g's transform
        AffineTransform savedForm = g2.getTransform();
        pos_rotate = center;
        //transform the shape
        AffineTransform newForm =
            AffineTransform.getRotateInstance(angle, pos_rotate.x, pos_rotate.y);
        GeneralPath gp = new GeneralPath();
        gp.moveTo((float)frontLeft.x, (float)frontLeft.y);
        gp.lineTo((float)frontRight.x, (float)frontRight.y);
        gp.lineTo((float)backRight.x, (float)backRight.y);
        gp.lineTo((float)backLeft.x, (float)backLeft.y);
        Shape brick = newForm.createTransformedShape(gp);
        //save the new coordinates
        Point2D.Double[] oldPoints = {
            frontLeft, frontRight, backLeft, backRight, center
        Point2D.Double[] newPoints = {
            frontLeft, frontRight, backLeft, backRight, center
        newForm.transform(oldPoints, 0, newPoints, 0, newPoints.length);
        //paint the shape
        g2.transform(newForm);
        g2.setColor(Color.YELLOW);
        g2.fill(brick);
        g2.setPaint(Color.blue);
        g2.draw(brick);
        g2.setPaint(Color.red);
        g2.fill(new Ellipse2D.Double(frontLeft.x-2, frontLeft.y-2, 4, 4));
        //set the old transform back
        g2.setTransform(savedForm);
    public static void main(String[] args) {
        SpinTest test = new SpinTest();
        test.addMouseListener(new Spinner(test));
        JFrame f = new JFrame();
        f.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
        f.getContentPane().add(test);
        f.setSize(400,400);
        f.setLocation(200,200);
        f.setVisible(true);
class Spinner extends MouseAdapter {
    SpinTest spinTest;
    public Spinner(SpinTest st) {
        spinTest = st;
    public void mousePressed(MouseEvent e) {
        spinTest.repaint();
}Also, note what happens when your app is partially covered/uncovered by another app...
this is one reason we try to make changes to the animation variables in the
event code instead of by calling repaint.
It is usually not necessary to transform both the graphics context and a primitive
(GeneralPath).

Odd compiler behavior (re: generics)

Greetings, I am encountering a rather odd compiler generated error/warning with some code. The issue involves using an "independent" generic method of a raw-typed class. That is, the class has a variable T (call it Test<T> and has an instance method that returns ArrayList<String>. The problem: when a raw typed Test is used, the ArrayList returned is also raw. Here is some code to illustrate:
import java.util.ArrayList;
class Test<T>{
ArrayList<String> getChildren(){
    return null;
static void test1( Test t ){
    ArrayList<String> children = t.getChildren();
// WARNING:                      ^^^^^^^^^^^^^^^^
//Test.java:26: warning: [unchecked] unchecked conversion
//found   : java.util.ArrayList
//required: java.util.ArrayList<java.lang.String>
static void test2( Test<?> t ){
    // OK, since we have the type variable set.
    ArrayList<String> children = t.getChildren();
}If the type variable is specified as anything but raw, then it works fine.
Is there some explantation for this behavior? My argument: since the return type of getChildren doesn't use the wildcard, why does the type argument of the implicit instance matter? Is this a compiler bug or just a "sharp edge" in the specification (perhaps even reasonable or necessary).
I know the solution (or workaround). I am just seeking an explanation.
Thanks,
- Tim

I found a bug evaluation (6358084) that was closed as not a bug where
the evaluator wrote:
All members, regardless of their dependence of type parameters, are erased.
So the compiler is correct, the method call is unchecked.
Still, I am ignorant of the reasoning for independent or fully described types.

Odd RV042 behavior can someone help shed some light?

We have a group of 4 x RV042 routers linking the 3 branch offices back to the head office via VPN site to site tunnels. Also each of the sites has dual ISP connection, one from a cable provider and the other a DSL connection operating in smart link backup mode with tunnels configured for when each wan is active. In the past week one of the branch offices RV042 started failing over to the secondary DSL WAN link when the primary cable ISP WAN IP was expired and was being refreshed.
The odd behavior starts a few hours after the switch over, the system log reports that the router restarted but with no error reported and it shows the date being December 31 2002 beginning with a System Up message then the firmware message which is at 1.3.12.19 and establishes a connection with the primary WAN at this point I can ping the public IP but I cannot connect to the router nor can the users at the branch browse the web or establish VPN connectivity to the main office. This condition persists until the router is physically powered down and back on at which time it shows the correct date and time and all operations work well for the day. I must note the behavior seems to only occur between 5am - 7am so I want to believe that this behavior is tied to the ISP IP lease cycle, but it only affects one branch office with all offices using identical ISP's, setup and configuration.
Can someone help make sense of this situation as rebooting the router each day is becoming annoying?

Hi Garth,
Thank you for posting. I wish I had a brilliant explanation for you that would immediately solve the problem. Unfortunately I have not seen this before. There is a possibility that the configuration file is corrupt or has been changed inadvertently. I think I would set the router to factory defaults and manually reconfigure it. There is also the option of swapping the router for another. If the other router does the same thing it could point to an environmental issue. If this router does the same thing at the alternate location it could point to either a configuration or hardware issue.

Odd Gvim behavior in Fluxbox

I just started using gvim recently, and while it's a great editor, it has some weird resizing behavior. I'm not quite sure if it is gvim or fluxbox that is responsible, so I was wondering if anybody could help me stop it. Here's what happens:
I start Gvim and it looks like this. The dimensions of this window are set by fluxbox, previously I resized it to the size I wanted it and selected remember -> dimensions from the rightclick menue of its title bar. Notice where the bottom edge of the window is.
[URL=http://img236.imageshack.us/my.php?image=200607211843471280x1024scrottp6.png][/URL]
Then, when I open a newtab(:tabnew), this happens:
[URL=http://img233.imageshack.us/my.php?image=200607211844081280x1024scrotov2.png][/URL]
The bottom edge of the window jumps down, so it sits on top of the panel. Then, if I close the tab (:q) this happens:
[URL=http://img205.imageshack.us/my.php?image=200607211844251280x1024scrotch0.png][/URL]
It jumps way up! Anybody seen this before?

thlokey wrote:
> While dabbling in a site design, I ran into an odd
CSS-related
> behavior in IE6 that I've not seen before. An odd space
is produced
> between my two DIVs that disappears when I roll over the
links. To be
> clear..I do not want the space. Take a look here
>
>
http://www.todlokey.com/dbs
>
> And view the CSS
http://www.todlokey.com/dbs/css/cssDBS_030607.css.
>
> Any ideas to save me some time? Thanks.
Try this:
#header UL {padding-bottom:1px;}
Thierry
Articles and Tutorials:
http://www.TJKDesign.com/go/?0
The perfect FAQ page:
http://www.TJKDesign.com/go/?9
CSS-P Templates:
http://www.TJKDesign.com/go/?1
CSS Tab Menu:
http://www.TJKDesign.com/go/?3

Safari Odd Font Behavior

Hi, shortly before Christmas we replaced our old eMac with a new iMac running Leopard version 10.5.1 and Safari version 3.0.4. A couple of days ago, my wife installed some new software (Rosetta Stone, which is language learning software) and, since that installation, is reporting strange font behavior in Safari. Specifically, the spacing appears odd and letters appear to be overlapping where they shouldn't be.
Based on her description, I am suspecting there is a font conflict somewhere Rosetta Stone installed some fonts that are not playing nicely with Safari but, as I am still running Tiger and am 8,000 miles from home right now, I'm not in a position to tinker around in search of a solution. Any ideas/suggestions would be welcome.
Thanks.

Hi
I'm not familiar with Rosetta Stone, however, he two fonts that seem to adversely affect Safari are Helvetica Fractions and Times Phonetic.
Have your wife open Font Book in your Applications folder. Enter either of these names in the spotlight panel. If either appears, disable it via the Edit menu.
If the above referenced fonts applies to your situation, download a copy of Nuke Font. This will allow her to clean the Font cache.
When done, restart the computer/Safari.

Odd trackpad behavior, opens Command+ F in any program.

I have an odd behavior on my Mac that's starting to get pretty annoying.
I have a late 2009 model. If I am in say Safari, and I wiggle my finger near the bottom of the track pad, (wiggling up and down on the bottom center edge of the trackpad) it opens up the search bar for Safari.
Basically Command+F.
If I am on the desktop (Finder) and do the same action it opens Command+ F once again, this time just in Finder.
In iTunes, the same motion on the trackpad puts me in full screen, which is, you guess it. Command+ F.
I have tried to get rid of this behavior by terminal commands killall Finder, killall Dock, and also by moving the user/library/preferences folder contents to the desktop, so that programs rewrite their preferences, but everything still has the same behavior.
Any ideas?
- Pixel -

Never heard of anything like it, frankly. Run the Apple Hardware Test you'll find on one of the DVDs that came with your MBP, and make an exact note of any errors it reports. Then make a Genius Bar appointment at an Apple Store and demonstrate the problem to the Genius. Also show the Genius the text of the error report (if there was one).

Odd Sound Behavior After 10.6.2 Update

I've had a few odd sound issues since updating to 10.6.2:
1) Volume is set at 50% after some reboots (I normally have it set at 12.5%). This has happened about 1/2 the time after reboots (which have unfortunately been much more often in 10.6).
2) While iTunes is playing the sound stops and the sound icon in the menu bar greys out. iTunes keeps 'playing' but I don't get any sound. This lasts for 5 - 10 seconds and happens a few times an hour but without any predictability.
It may happen at times that iTunes isn't playing and I just don't notice it. It has happened while the system as announcing the time as well. This may be the cause of sound issue #3 but I haven't caught the sound icon in a 'grey state' when the notification sounds don't play yet.
When the sound icon is grey I get the beach ball and Safari doesn't respond (standard 'Application not responding' message). Other apps seem to work while this is happening (Finder, System Prefs - including sound panel, FireFox, etc).
3) Some expected sounds used for feedback in Finder & Mail don't always work. This is very intermittent and may be happening more often then I notice these are notification sounds and I don't realize that I've missed them. I don't know if it's associated to the sound failure mentioned in #2.
~
I've also had some other odd behavior:
* System won't wakeup from sleep (not always but often enough to require several reboots in the last week).
* Safari just eats up RAM.
* Safari has locked up for no apparent reason (no clue as to what sets it off).
* Safari has locked up while using a Java applet. This only happens on occasion but always happens when launching a GoToMyPC session (GTMPC works fine in FireFox, even while Safari is locked up). This is separate from the Safari issue noted above.
I'm using 10.6.2 w/ Safari 4.0.4 & iTunes 9.0.2 (Flash not installed)
James

You could try [these grey sceen, etc., troubleshooting steps|http://support.apple.com/kb/TS2570] ignoring any duplicaion of what you already have don. And, you could try [these general troubleshooting steps|http://support.apple.com/kb/TS1394] again ignoring duplication.
If all above fails, then I suggest you do a repair install of Sno by installing Sno over Sno as in an upgrade install. Such an install, while usually very safe, still is an install so be appropriated backed up if you proceed.

Odd issue with slow motion and exporting

Hello,
I have a project with footage taken at either 24p or 60i settings on a Canon HF200. I'm using 7.0.1, ProRes 422 selected. Frame size is set to 1920x1080. I'm exporting using the "Share" menu and the Apple TV preset.
The project is divided into sequences. One of these sequences (whose source material shot using the 60i setting) is set to play in slow motion using the "change speed" option set to 50%.
Here's the odd part, to me at least:
When I export just the slow motion sequence of the project, it plays perfectly (good quality, smooth not jerky, no shadows/trails). But when I export all sequences joined together, the slow motion sequence (and just the slow motion sequence) appears stuttery, with trails on every action by the actors.
Another odd bit: when I export just a subset of the project including the slow motion sequence and everything after it, again the slow motion sequence plays as intended, after export. So the problem only occurs if I export the slow motion sequence with anything before it included.
Thanks in advance for any help.

Problem solved, but I'm not sure if it was my re-importing the clips, or un-checking the "frame blending" option that did it. I did both for no particular reason, and one of them worked.

Odd dock behavior

Every so often out of the corner of my eye, I see some application launch and quit in the Dock — and it's fast too at probably under a second. It happened again this morning. Something is clearly launching and quitting. I've gone and looked at any logs I can find, and nothing appears to be out of the ordinary.
Is it anything I should be concerned about?
iMac G5 20" iSight; G3 Pismo Powerbook; LaCie 160G FW Mac OS X (10.4)

See, the thing is it happens fast. So fast I only notice the Dock expand and shrink in a split second.
It did it again this morning.
For now, I've disabled all widget exept the world clock and calculator. The only thing I can think of is that the weather widget was phoning home for current time/temp. But since this is a new behavior, it's doubtful.
Another odd thing that I noticed this morning: the Dock has acquired some icons that I have not put there.
iMac G5 20" iSight; G3 Pismo Powerbook; LaCie 160G FW Mac OS X (10.4)

Odd Stations Behavior

Ok
Aside from the problems everyone else is reporting I'm seeing really odd behviors. In trying to get a good all unplayed I tried creating my own.
I've now fiddled with a lot of settings getting differing results (none what I want)
So here is what I'm seeing, I've got 2 "Stations" which if you look at the settings have matching setup. here's the kicker. the little icon showing number of episodes is different for each. I've also made a third which has matching settings to the apple provided all unplayed... same deal.
in other words the code under the hood is compeltely messed up.
anyone else seen this?

I didn't notice it causing any battery drain, but it sounds like I am charging it differently than you are. Here's how mine worked:
I would plug the iPhone into the wall charger at night to charge while I slept. In the morning, it would show a full charge by showing a plug on the small battery icon, but the large green battery would still have a sliver that wasn't green. The usage and standby stats would show something in the 3-4 hour range, presumably the time since it hit full charge.
I would disconnect the phone, check my email, and then take it to my desktop computer and plug it in to sync and update all my podcasts, etc. I would leave it plugged into my desktop while I showered while I got ready for work. By the time I was ready to leave, it would again show a full charge on the small icon, but now my stats would be reset to 0 and would function normally throughout the day. But nothing I did ever stopped the stats from running up during the night after attaining full charge.
1.0.1 got rid of the stats problem (they always stay at 0/0 in the morning with a full charge until I unplug the phone) AND the weird behavior of the large green battery indicator went away. Now both indicators show full charge in the morning.

Odd Drummer behavior

Not saying drummers are odd... I'm using the Smash producer kit, and every now and then there's a snare hit somehow excluded from the Overheads, which makes the snare drop out of the mix a fair bit. When it happens, it happens at the same 3 points in the song. That said, it happens about 1 out of every 4 playbacks. It took me 1/2hr to figure out what was going on!
It's a pain because I never know until after a bounce whether I got a good take.
Anyone else seen this, fixed it, or reported it as a bug?

I didn't notice it causing any battery drain, but it sounds like I am charging it differently than you are. Here's how mine worked:
I would plug the iPhone into the wall charger at night to charge while I slept. In the morning, it would show a full charge by showing a plug on the small battery icon, but the large green battery would still have a sliver that wasn't green. The usage and standby stats would show something in the 3-4 hour range, presumably the time since it hit full charge.
I would disconnect the phone, check my email, and then take it to my desktop computer and plug it in to sync and update all my podcasts, etc. I would leave it plugged into my desktop while I showered while I got ready for work. By the time I was ready to leave, it would again show a full charge on the small icon, but now my stats would be reset to 0 and would function normally throughout the day. But nothing I did ever stopped the stats from running up during the night after attaining full charge.
1.0.1 got rid of the stats problem (they always stay at 0/0 in the morning with a full charge until I unplug the phone) AND the weird behavior of the large green battery indicator went away. Now both indicators show full charge in the morning.

Odd Draw Behavior - Slow Ellipse2D Draw

Similar Messages

Maybe you are looking for