Hi Alessandro,

If you check Georg mail it says that the problem is only visible
when you run using the 'javaw' command, not when using the
'java' command as you seem to be doing.

I also thought it was some driver problem at first, and discovered
more by coincidence that the reason some machines worked and
some did not depended on which java command you used.

Best Regards,

Roger Berggren


Alessandro borges wrote:

>
>Hi Georg!
>
>I test your app I could not find any noticiable bug.
>I also could not reproduce the bugs you report.
>
>Runs fine with :
>
>Athlon 1.3GHz
>512 MB RAM
>JDK 1.3.1_03
>Java 3D 1.3 Beta 2 DirectX
>Win 2K Professional
>Savage4 16MB AGP video card
>
>
>C:\>java -version
>java version "1.3.1_03"
>Java(TM) 2 Runtime Environment, Standard Edition
>(build 1.3.1_03-b03)
>Java HotSpot(TM) Client VM (build 1.3.1_03-b03, mixed
>mode)
>
>j3d.debug reports (quick run):
>
>Java 3D 1.3 beta2, Windows version is 5.0 Build: 2195,
>Windows NT Service Pack 2
>
>[Display Driver] s3savg4.dll, S3 Inc. Savage4, Product
>5
>                 Version 12.1, Build 8009, VendorId
>389505648
>                 DeviceId 0x5333, SubSysId 0x8a22,
>Revision 0x69808
>  [Desktop Mode] 1024x768 75 MHz, D3DFMT_X8R8G8B8
>        [Device] Hardware Rasterizer
>        [Device] Reference Rasterizer (AA)
>
>==> start generating voxel
>==> initialCapacity: 1000
>==> number of generated spheres: 6
>==> number of generated points: 1410
>--> allBounds center: (-0.09676453471183777,
>-0.025539368391036987, -0.115838915
>1096344)
>Use S3 Inc. Savage4, Hardware Rasterizer
>
>Create device :
>Window 600x577 D3DFMT_X8R8G8B8, handle=e041e,
>D3DMULTISAMPLE_NONE, D3DSWAPEFFECT
>_COPY, D3DFMT_D24S8
>Max Texture Unit Stage support : 2
>Max Simultaneous Texture unit support : 2
>FPS:    33.738194
>
>
>Maybe you have a video card/OpenGL issue...not sure...
>That´s all ;)
>
>Alessandro Borges
>
>
>--- Georg Rehfeld <[EMAIL PROTECTED]> escreveu: >
>RE: [JAVA3D] Java 3D 1.3 Beta2Dear List,
>
>>over the weekend I installed JDK 1.4.0 and Java3D
>>1.3_beta_2 which went
>>fine, thanks J3D team.
>>
>>I'm experiencing a paint() problem similar to that
>>reported by Pasi
>>Paasiala ('covered part of Canvas3D goes gray').
>>
>>I verbosely report this, because I get part of the
>>error even with JDK
>>1.3.1 and the possibly related Bugs ...
>>
>> 4665482 - Missing first paint() event under JDK1.4
>>windows version
>> 4458337 - Merlin: repaints do not work by using
>>javaw on WinNT if
>>           double buffering is disabled
>> 4664490 - REGRESSION: AWT window background isn't
>>drawn on win32 using
>>           javaw
>>
>>... all name JDK 1.4.0 as the possible source of the
>>bug.
>>
>>The problem only shows up when 'javaw' is used, not
>>with 'java'. Similar
>>problems with J3D 1.2.1 did not show up. You'll find
>>a small application
>>attached for evaluation.
>>
>>My setup:
>>
>>- PIII 700MHz
>>- 320MB Mainmemory
>>- W2K, Servicepack 2
>>- nVidea Riva TNT with 16 MB Memory, AGP
>>- JDK 1.4.0 (build 1.4.0-b92) and
>>- JDK 1.3.1 (build 1.3.1-b24)
>>- J3D 1.3 beta 2, OpenGL
>>
>>Observations with JDK 1.4.0 (build 1.4.0-b92):
>>
>>1a. Click on 'Action' to pop up menu, release, then
>>click again --> the
>>    Menu disappears, then release ==> a gray area
>>appears where the menu
>>    was. Move the mouse over the Canvas3D --> gray
>>area disappears
>>
>>1b. Select 'Action/New Shape (grid 0.02)' to
>>generate many points and
>>    drop the framerate. Repeat step 1, but this time
>>do the second
>>    click/release really fast, so that both fall
>>into the same render
>>    frame --> the error does NOT show up.
>>
>>1c. Open the 'Action' menu again, then click/release
>>any menu item
>>    (including the noop one), don't move the mouse
>>yet --> the menu
>>    disappears OK, eventually new content is
>>displayed. Now move the
>>    mouse ==> a gray area appears where the menu was
>>and 1 (or few?)
>>    frames later disappears, causing flicker.
>>
>>2.  Overlap the 3D application partly with some
>>other window, then switch
>>    back to it, either by ALT-TAB or by moving the
>>mouse into the title
>>    bar and clicking --> the 3D content is repainted
>>OK. Move the mouse
>>    anywhere into the menu area of the 3D app ==> a
>>gray area appears
>>    where the other window overlapped. Move the
>>mouse over the Canvas3D
>>    --> the gray area disappears.
>>
>>3.  Resize the 3D app --> the Canvas3D correctly is
>>repainted. Move the
>>    mouse into the menu area without moving over the
>>Canvas3D ==> the
>>    whole Canvas3D goes gray. Move the mouse over
>>the Canvas3D --> canvas
>>    gets repainted.
>>
>>
>>Observations with JDK1.3.1 build 1.3.1-b24
>>
>>1a. Click on 'Action' to pop up menu, release, then
>>click again ==> the
>>    Menu disappears but a gray area is left behind;
>>release ==> gray
>>    area remains; move mouse to Canvas3D -->
>>repainted OK. This behavior
>>    does not show up always (~70% of the cases,
>>sometimes mouse movement
>>    in the menu area and/or opening the 'File' menu
>>forces the error).
>>    Could never reproduce bug with many points/low
>>framerate.
>>
>>1b. from above does not show up.
>>
>>1c. Open the 'Action' menu again, then click/release
>>menu item 'New
>>    Shape (grid 0.02)' forcing many
>>points/considerabe computation
>>    time/low framerate --> the menu disappears, a
>>gray area is left
>>    behind until J3D repaints to all black then ==>
>>the gray area
>>    flickers up for one frame again sometimes (20%
>>of the time).
>>    I consider this acceptable, but report it for
>>completeness.
>>
>>2.  Overlap the 3D application partly with some
>>other window, then switch
>>    back to it, by ALT-TAB --> the 3D app is
>>repainted OK just once.
>>    Again overlap and switch back, this time by
>>moving the mouse into the
>>    title bar and clicking ==> often (50% of the
>>cases) the Canvas3D is
>>    noticably repainted twice: grey area visible,
>>repainted, gray area
>>    AGAIN visible, then finally repainted. Most
>>visible with medium
>>    frame rate/medium number of points.
>>
>>3.  Resize the 3D app ==> the Canvas3D often (>50%
>>of the cases) gets
>>    repainted twice as with 2. above. Again most
>>visible with medium
>>    frame rate/medium number of points.
>>
>>
>>Conclusion
>>
>>Though with JDK 1.4.0 the bugs are more and more
>>obvious, it seems to
>>me, that the cause is not only in JDK 1.4.0 but
>>partly somewhere in J3D.
>>I hope my observations can help fix the bugs.
>>
>>Best regards
>>
>>Georg
>> ___   ___
>>| + | |__    Georg Rehfeld      Woltmanstr. 12
>>20097 Hamburg
>>|_|_\ |___   [EMAIL PROTECTED]           +49 (40)
>>23 53 27 10
>>
>>
>>----- Original Message -----
>>From: Pasi Paasiala
>>To: [EMAIL PROTECTED]
>>Sent: Thursday, May 02, 2002 1:26 PM
>>Subject: Re: [JAVA3D] Java 3D 1.3 Beta2
>>
>>
>>We seem to have a slight problem with Java3D 1.3
>>beta2.
>>What happens is the following:
>>When I put a window on top of the window containing
>>the Canvas3D and then bring the frame with the
>>canvas to top again by clicking
>>on on the window title bar, I get the scene full
>>rendered, but once I move the mouse to the canvas,
>>the part that was covered goes
>>gray.
>>If I then move the camera, the rendering is fine
>>again.
>>
>>[...]
>>
>>
>
>>ATTACHMENT part 2 application/x-zip-compressed
>>
>name=test.zip
>
>
>_______________________________________________________________________
>Yahoo! Encontros
>O lugar certo para você encontrar aquela pessoa que falta na sua vida.
>Cadastre-se hoje mesmo!
>http://br.encontros.yahoo.com/
>
>===========================================================================
>To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
>of the message "signoff JAVA3D-INTEREST".  For general help, send email to
>[EMAIL PROTECTED] and include in the body of the message "help".
>

===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff JAVA3D-INTEREST".  For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".

Reply via email to