|
Replies:
8
-
Last Post:
Jul 21, 2008 3:19 AM
by: demonduck
|
Threads:
[
Previous
|
Next
]
|
|
|
|
|
|
Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 12, 2008 5:25 PM
|
|
|
What are you people doing? 0.7 draws half as fast as 0.5 and 1.6.0.5 was already slower than 1.5.0.13
http://pancyl.com/VSync.htm
draws half as fast as in 0.5 and 0.5 is about half as fast as 1.5.0.13.
What are you doing??? Are you trying to put us all out of business?
And the LiveConnect bug is still disabling JavaScript <->applet communication.
http://pancyl.com/aerialpan.htm
Click on a thumbnail and get a memory error in FF2.
Plus, I have 4 JRE's installed but have no way to select which JRE is run by the plugin.
I used to be able to do that.
WHAT ARE YOU PEOPLE DOING?????
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 16, 2008 4:38 PM
in response to: demonduck
|
|
|
The liveconnect issue seems to have been fixed in our upcoming JRE 6u10.
You can download it from:
https://jdk6.dev.java.net/6u10ea.html
It works for me with FF2 or FF3.
|
|
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 16, 2008 5:13 PM
in response to: ccheung
|
|
|
Humm...I downloaded the Kernel Installer as a test. I'm still using FF2. I had 1.5.0.16 installed.
Summary: I think the plain JRE install is actually simpler and cleaner than the kernel install. I still get a lot of LiveConnect spew but no memory error.
The Kernal Installer seems to work. But it's a stop and go process that I think is not as clean and simple as a straight JRE online/offline install. In other words, it seems to involve the user in the installation process even more than a simple JRE install.
And I had to restart my browser two or three times to get my applet to run. It would wedge. And it is solid and well tested and never wedges.
But 0.10 does at least let my applet use LiveConnect with out a memory error which is a big step in the right direction. But I still get a lot of spew and a delay. I understand that FF3 doesn't have this problem. But I like to hang back and stay with what I think the majority of users still have which I think is FF2.
Here is the spew I see with trace=5:
--------
Java Plug-in 1.6.0_10-rc
Using JRE version 1.6.0_10-rc Java HotSpot(TM) Client VM
User home directory = C:\Documents and Settings\Administrator
----------------------------------------------------
c: clear console window
f: finalize objects on finalization queue
g: garbage collect
h: display this help message
l: dump classloader list
m: print memory usage
o: trigger logging
p: reload proxy configuration
q: hide console
r: reload policy configuration
s: dump system and deployment properties
t: dump thread list
v: dump thread stack
x: clear classloader cache
0-5: set trace level to <n>
----------------------------------------------------
PanCyl v0.2.1.h MsgScreen
Trace level set to 5: all ... completed.
basic: Referencing classloader: sun.plugin.ClassLoaderInfo@9300cc, refcount=1
network: Cache entry not found [url: http://pancyl.com/black.jpg, version: null]
network: Connecting http://pancyl.com/black.jpg with proxy=DIRECT
network: Connecting socket://pancyl.com:80 with proxy=DIRECT
basic: Loading applet ...
basic: Initializing applet ...
basic: Starting applet ...
basic: completed perf rollup
PanCyl v0.2.1.h MsgScreen
liveconnect: JavaScript: calling Java system code
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: calling Java system code
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: calling Java system code
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: calling Java system code
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: calling Java system code
[ about a thousand more lines that I didn't copy ]
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: calling Java system code
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: calling Java system code
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: UniversalBrowserRead enabled
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: UniversalBrowserRead enabled
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: UniversalBrowserRead enabled
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: UniversalBrowserRead enabled
liveconnect: JavaScript: default security policy = http://pancyl.com
liveconnect: JavaScript: UniversalBrowserRead enabled
liveconnect: JavaScript: default security policy = http://pancyl.com
network: Cache entry not found [url: http://pancyl.com/images/aerialpan/pano_waikiki_sunset_large.jpg, version: null]
network: Connecting http://pancyl.com/images/aerialpan/pano_waikiki_sunset_large.jpg with proxy=DIRECT
network: CleanupThread used 12 us
network: Downloading resource: http://pancyl.com/images/aerialpan/pano_waikiki_sunset_large.jpg
Content-Length: 1,118,794
Content-Encoding: null
network: Wrote URL http://pancyl.com/images/aerialpan/pano_waikiki_sunset_large.jpg to File C:\Documents and Settings\Administrator\Application Data\Sun\Java\Deployment\cache\6.0\4\4fe15104-14121c6f-temp
network: CleanupThread used 12 us
|
|
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 20, 2008 11:03 AM
in response to: demonduck
|
|
|
What does it mean for an application to "wedge" or for LiveConnect to "spew"?
|
|
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 16, 2008 5:59 PM
in response to: ccheung
|
|
|
I uninstalled the kernel installer and did the offline JRE install. Much neater, cleaner and faster than the kernel installer -- IMHO.
I noticed that in IE6, the Tools menu no longer has the Java Console menu item. 0.10 is installed and runs in IE6 but no way to open the Java Console from the Tools menu.
I can open the console from the tray icon.
FF2 still has the Java Console menu item in the Tools menu.
|
|
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 20, 2008 6:25 AM
in response to: demonduck
|
|
|
I'm fairly sure both have been discussed elsewhere on this forum.
* The tools option is gone because plugin2 supports multiple JVMs per browser instance
* The LiveConnect "spew" is due to the architecture of FF2 and the old plugin, one of the reasons plugin2 requires FF3 is they changed a bunch of FF internals which won't be backported to FF2.
The extra messages are 100% harmless, aside from making the console harder to read (and if you're unlucky, a small browser freeze while they appear)
|
|
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 20, 2008 10:30 AM
in response to: andrewherron
|
|
|
> I'm fairly sure both have been discussed elsewhere on
> this forum.
>
> * The tools option is gone because plugin2 supports
> multiple JVMs per browser instance
> * The LiveConnect "spew" is due to the architecture
> of FF2 and the old plugin, one of the reasons plugin2
> requires FF3 is they changed a bunch of FF internals
> which won't be backported to FF2.
>
> The extra messages are 100% harmless, aside from
> making the console harder to read (and if you're
> unlucky, a small browser freeze while they appear)
Right... No sense whipping a dead JVM but there are things called sub-menus. And all I know is what's in front of me and that's FF2.
They appear to be harmless in 0.10rc but in 0.7 there is a memory error associated with the spew.
Not everyone is going to upgrade to Plugin{NG,2} and FF3 -- and I'll be the one getting the "...you suck..." emails.
|
|
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 20, 2008 9:59 PM
in response to: demonduck
|
|
|
> They appear to be harmless in 0.10rc but in 0.7 there
> is a memory error associated with the spew.
In my experience, FireFox memory issues come not from the messages but rather from FF2 holding on to the applet instance after destroy() is called. We had to systematically null out all of our instance variables to keep the memory usage as low as possible, and eventually you'll still run out.
> Not everyone is going to upgrade to Plugin{NG,2} and
> FF3 -- and I'll be the one getting the "...you
> suck..." emails.
The application I work on was first released in the days of JRE 1.3 with IE6 or IE on Mac, so I've seen more than my fair share of those... With applets you just have to find workarounds for the annoying platform and browser issues because half the time customers refuse to upgrade.
I can still remember the time we wrote a C app to handle copy & paste because of a clipboard bug on a very specific JRE 1.4.1 version that one of our customers had deployed company-wide 
|
|
|
|
|
|
|
|
|
Re: Just updated to 1.6.0.7 -- it sucks worse than 0.5
Posted:
Jul 21, 2008 3:19 AM
in response to: andrewherron
|
|
|
Thanks for the input...
The memory error I see in FF2 comes when I try to use LiveConnect to make my applet fetch an image from the server. There was spew in 1.5 but it eventually connected with the image and ran ok. In 1.6 up to 1.6.0.7, there's spew followed by a memory error. This happens for me the way I run the applet at startup time.
If my applet happens to run, I am moderately strict about nulling variables and calling gc() at various stages.
Applet deployment was supposed to be easy but is one of the hardest problems with Java programming.
|
|
|
|
|
java.net RSS
