Performance differences between application vs. applet

Q

Qu0ll

Roedy Green said:
There are two reasons:
1. with an Applet, RAM is full of browser code and tables.
2. with an Applet, you have no control over the heap size.

With JSE 6 Update 10 you can control the heap size and I have found that it
makes no difference to the performance even if you specify a massive heap in
the JVM arguments.

There must be another reason.

--
And loving it,

-Qu0ll (Rare, not extinct)
_________________________________________________
(e-mail address removed)
[Replace the "SixFour" with numbers to email me]
 
J

John B. Matthews

"Qu0ll said:
With JSE 6 Update 10 you can control the heap size and I have found that it
makes no difference to the performance even if you specify a massive heap in
the JVM arguments.

There must be another reason.

More results: Mac OS X 10.4.11 (PPC), java version 1.5.0_13:

Command line:
Average = 250175.59494093293, Time = 26.057
Average = 250422.3576639226, Time = 26.064
Average = 250111.89527472467, Time = 26.248

Safari 3:
Average = 249768.90103053028, Time = 26.294
Average = 250040.5137461857, Time = 26.26
Average = 249876.6562412136, Time = 26.034

Firefox 2:
Average = 250055.24368220105, Time = 26.331
Average = 250055.44471629648, Time = 26.426
Average = 250061.03580421582, Time = 26.401

I'm not seeing much difference. I'm tempted to say the browser is as
fast as the application, but it may be that the application is as slow
as the browser. :)
 
Q

Qu0ll

[...]
More results: Mac OS X 10.4.11 (PPC), java version 1.5.0_13:

Command line:
Average = 250175.59494093293, Time = 26.057
Average = 250422.3576639226, Time = 26.064
Average = 250111.89527472467, Time = 26.248

Safari 3:
Average = 249768.90103053028, Time = 26.294
Average = 250040.5137461857, Time = 26.26
Average = 249876.6562412136, Time = 26.034

Firefox 2:
Average = 250055.24368220105, Time = 26.331
Average = 250055.44471629648, Time = 26.426
Average = 250061.03580421582, Time = 26.401

I'm not seeing much difference. I'm tempted to say the browser is as
fast as the application, but it may be that the application is as slow
as the browser. :)

Thanks for the report John. Perhaps it's actually a Java 6 problem?

--
And loving it,

-Qu0ll (Rare, not extinct)
_________________________________________________
(e-mail address removed)
[Replace the "SixFour" with numbers to email me]
 
A

Arne Vajhøj

Roedy said:
There are two reasons:
1. with an Applet, RAM is full of browser code and tables.

Possible but not not likely. The 3 big data structures only
need 240 KB.
2. with an Applet, you have no control over the heap size.

You can change it so you do have some control.

But heap size will not in itself effect performance. Either
the codes or it does not run.

Arne
 
Q

Qu0ll

Arne Vajhøj said:
Possible but not not likely. The 3 big data structures only
need 240 KB.


You can change it so you do have some control.

But heap size will not in itself effect performance. Either
the codes or it does not run.

I really want to get to the bottom of this. So far everyone I have asked
has been unable to explain the discrepancy including people from Sun who
were very surprised by my findings.

--
And loving it,

-Qu0ll (Rare, not extinct)
_________________________________________________
(e-mail address removed)
[Replace the "SixFour" with numbers to email me]
 
A

Andrew Thompson

I really want to get to the bottom of this.  

How did your test with killing the SecurityManager go?
..So far everyone I have asked
has been unable to explain the discrepancy including people from Sun who
were very surprised by my findings.

Another thing that occurred to me after J.B.M.'s
'null' result on Java 1.5 is that the applet and
JWS plug-ins were further integrated in Java 1.6.

Perhaps that is (somehow) responsible.
 
R

Roedy Green

There must be another reason.

I wonder if you have a very active display there is extra overhead
with the drawing happening in a browser's window. I would think it
would be arranged in such a way the browser had little to do with it
other that setting up the hole to draw in.
 
A

Andrew Thompson

I wonder if you ..

You? Not to put *too* fine a point on it, but the code
is available for *anybody* to test on their own machine.
As an aside, my machine is so damaged at the moment that
every time I boot it reports the monitor has no valid
driver.

The monitor works 'well enough' despite that, but I
doubt that any advanced graphics rendering is possible
(perhaps supported by the fact that DVDs play a little
less than entirely smoothly - combined with a recent
graphics oriented thread that showed frame rates of 320
FPS on other machines, while about 10 FPS on this one).
...have a very active display there is extra overhead
with the drawing happening in a browser's window.  I would think it
would be arranged in such a way the browser had little to do with it
other that setting up the hole to draw in.

I don't think that is true of *graphics intensive*
code, but that is *not* very relevant to Quoll's
original question that had no GUI until I inserted
one. And FTR - the GUI I put in is 'about as
basic as it can get' - it is just a button.

I have no specific complaint about you chiming in
on a thread in which you have not read all the
replies carefully (I know what it's like with these
long threads - and the wealth of carefully considered
information on your site makes any complaint look
like 'so much wasted bandwidth'), but I am not about
to let such a reply pass without comment.

(Harumph!)
 
R

Roedy Green

(Harumph!)

There is no need to harrumph because a brainstorm idea tossed out was
not the solution. It is wicked of you to try to discourage people
from tossing out half-baked ideas instead of finished solutions.

What you are doing is form of psychological entitlement, that you are
cosmic royalty and the universe is obligated to hand you only
finished perfect solutions on a titanium plate.

When people toss ideas out for me, a wrong idea can often lead to the
solution simply because it jars me to rethink my assumptions.
 
T

Tom Anderson

You? Not to put *too* fine a point on it, but the code
is available for *anybody* to test on their own machine.

I didn't read 'you' as meaning 'Andrew Thompson', but rather 'an arbitrary
person' - the use where you'd use 'one' in more formal language. Or, as
the OED puts it:

you, pers. pron., 2nd pers. obj. (nom.), pl. (sing.)

III. Special uses.

6. Denoting any hearer or reader; hence as an indef. pers. pron.: One,
any one.

1577 GOOGE Heresbach's Husb. II. (1586) 87 You shall sometime have one
branch more gallant than his fellowes. 1614 T. TOMKIS Albumazar I. iii,
With this [perspicill] Ile read a leafe of that small Iliade..as plainly
Twelue long miles off, as you see Pauls from Highgate. 1625 BACON Ess.,
Atheism (Arb.) 333 Nay more, you shall haue Atheists striue to get
Disciples, as it fareth with other Sects. 1707 Lond. Gaz. No. 4351/3 One
Red Buoy to the Eastward of you, as you pass this Chanel. 1726 SWIFT
Gulliver II. i, A child..began a squall that you might have heard from
London Bridge to Chelsea. 1865 RUSKIN Sesame i. 30 You can talk a mob into
anything. 1870 Good Words 133/2 The slope [is] so rapid that you can
scarcely find footing when once off the beaten road.
(Harumph!)

There, there. Have a nice cup of tea and a sit-down, and it'll all seem
better.

Hope your computer gets well soon.

tom
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,744
Messages
2,569,482
Members
44,901
Latest member
Noble71S45

Latest Threads

Top