|
|
| phaeron |
| Posted: Jun 12 2011, 11:36 PM |
 |
|

Virtualdub Developer
  
Group: Administrator
Posts: 7773
Member No.: 61
Joined: 30-July 02

|
| QUOTE (DarrellS @ Jun 10 2011, 04:29 PM) | | Got a nasty trojan last night that prevented my PC from booting. After trying to fix the problem using a boot disc failed, I installed a back-up image of my C drive and the Null Transform problem disappeared. | Uh... while it's good that you got your machine cleaned off, it's unlikely that this was the cause. There was a regression in test-8 that caused the null transform filter not to handle flipped bitmaps properly.
P.S. In the future, please indicate the test-* version you are using explicitly, as neither window you have in your image indicates this. The test release is only indicated in the initial window caption on startup. |
 |
| pureocean |
| Posted: Jun 13 2011, 02:16 AM |
 |
|
Advanced Member
  
Group: Members
Posts: 53
Member No.: 27197
Joined: 13-March 10

|
@phaeron: Thank you for new update. Also late thanks for you replied my messages.
Quote: | CODE | | I wasn't able to reproduce the WAV file issue either. If you have input plugins installed, try temporarily removing them to see if the problem goes away. There was a recent thread about ambiguities in file pattern selection that make me suspect this is a factor. |
You're right. Issue source: FccHandler's MPEG2 InputDriver (MPEG2.vdplugin). After this file deleted, *.WAV files can appear in "Audio from other file..." window.
Quote: | CODE | | The open file dialog crash looks like a problem with a video codec that Windows Explorer is trying to use to get preview information on a file. You might be able to reproduce this in Notepad if you set the file type in the dialog to All Files. The crash looks as it does because some codecs have the lame behavior of unloading their code when they detect a crash, resulting in the confusing crash dump. |
Actually, I suspect the mouse pointers. Because I don't use default pointers of windows. My mouse pointers: WinXPsetV3 Cursor 66.ani WinXPsetV3 Cursor 68.ani aero_busy.ani
Crashing issues is very rarely (in Save AVI window). Perhaps, a buffer overflow be caused by external pointers. I think, I need back to the original mouse pointers. |
 |
| DarrellS |
| Posted: Jun 13 2011, 06:12 AM |
 |
|
Advanced Member
  
Group: Members
Posts: 567
Member No.: 1061
Joined: 28-November 02

|
| QUOTE (phaeron @ Jun 12 2011, 04:36 PM) | | QUOTE (DarrellS @ Jun 10 2011, 04:29 PM) | | Got a nasty trojan last night that prevented my PC from booting. After trying to fix the problem using a boot disc failed, I installed a back-up image of my C drive and the Null Transform problem disappeared. |
Uh... while it's good that you got your machine cleaned off, it's unlikely that this was the cause. There was a regression in test-8 that caused the null transform filter not to handle flipped bitmaps properly.
P.S. In the future, please indicate the test-* version you are using explicitly, as neither window you have in your image indicates this. The test release is only indicated in the initial window caption on startup. |
In my first post, although I did not mention 1.10.1-test8 (I thought that I did), I did show a screenshot of the build 34592. I thought it was test8 since that is what my shortcut said but when I redownload the file a couple of days ago which didn't show the messed up frame, it said test7 (which is build 34093). I notice tonight that there is a test8 which was not on the front page a couple of days ago and it is the same build that I posted in the screenshot, test8.
I'll retry and see if test8 still shows the messed up screen like it did the other day.
EDIT: Yep! The problem is back in the version that I downloaded tonight. Test8 build 34592. |
 |
| Dstruct |
| Posted: Jun 14 2011, 02:23 PM |
 |
|
Advanced Member
  
Group: Members
Posts: 257
Member No.: 24005
Joined: 8-August 08

|
1.10.1-test10
"Save Default" isn't working properly in the "Resize" filter dialog. Doesn't remember "Lanczos3" and "Multiple of 2" settings!
FIXED (1.10.1-test10) |
 |
| pureocean |
| Posted: Jun 14 2011, 06:36 PM |
 |
|
Advanced Member
  
Group: Members
Posts: 53
Member No.: 27197
Joined: 13-March 10

|
About MPEG audio input module...
A sample audio (Format: MPEG 1.0 Layer 2). Cannot import. Error message: "Audio.mp2 is of an unknown or unsupported file type."
Is this normal? If so, it doesn't matter. 
Sample audio: http://rghost.net/10895751 |
 |
| DarrellS |
| Posted: Jun 14 2011, 10:33 PM |
 |
|
Advanced Member
  
Group: Members
Posts: 567
Member No.: 1061
Joined: 28-November 02

|
Null Transform problem fixed in 1.10.1-test10. Thanks! |
 |
| phaeron |
| Posted: Jun 18 2011, 09:58 PM |
 |
|

Virtualdub Developer
  
Group: Administrator
Posts: 7773
Member No.: 61
Joined: 30-July 02

|
Hmm, that Save As Default bug is interesting. I don't think the code for that's changed in a couple of years... surprised it didn't come up before. Anyway, it's fixed in test 11 (it wasn't syncing the state of the dialog before saving the config).
The MPEG audio code doesn't do signature detection and doesn't include the "mp2" extension. The reason is that signature detection is unreliable for MPEG audio -- Microsoft tried to do this in Windows Media Player, and the result was that it opened some AVI files as MP3 files -- and the .mp2 extension is more commonly used for video files. You can still open them as such if you override the file type. |
 |
| Mr.Bomber |
| Posted: Jun 19 2011, 07:21 PM |
 |
|
Newbie

Group: Members
Posts: 4
Member No.: 21190
Joined: 7-April 07

|
Thank you for corrected filter resize. There is else here is such small bug. The move of the performing the task is displayed in percent in window Job control and in headline main window Virtualdub. If in window status is switched to any tab, except the main. Image of the percent is frozen in headline main window. |
 |
| ale5000 |
| Posted: Jun 20 2011, 03:19 AM |
 |
|

Advanced Member
  
Group: Members
Posts: 1114
Member No.: 22180
Joined: 30-September 07

|
| QUOTE (phaeron @ Jun 13 2011, 12:35 AM) | | In any case, the WAV detection fix is in 1.10.1 test-10, so we'll see if that does the trick. | It works, thanks.
-------------------- New VirtualDub forum VirtualDub AIO (All-in-One installer for VirtualDub and plugins) Codec Toolbox RS (A tool to read/change merit of codecs and many other things) Input plugins for VirtualDub / ACM codecs / VFW codecs |
 |
| ale5000 |
| Posted: Jun 20 2011, 03:55 AM |
 |
|

Advanced Member
  
Group: Members
Posts: 1114
Member No.: 22180
Joined: 30-September 07

|
VirtualDub say "The video decompressor cannot decompress to the selected input format. Check for a Force YUY2 setting in the codec's properties or select a different input video format under Vido > Color Depth."
Test file (cut.demuxed.m2v): http://www.multiupload.com/SQRQRQKX3V You should have the MPEG2 input plugin. 1) Select Video => Color Depth 2) Then Decompression format => Other... => YCbCr (Rec.601) Limited 4:2:0 interlaced 3) Then Output format => 24 bit RGB (888) 4) Open the video 5) Now press the play output button
-------------------- New VirtualDub forum VirtualDub AIO (All-in-One installer for VirtualDub and plugins) Codec Toolbox RS (A tool to read/change merit of codecs and many other things) Input plugins for VirtualDub / ACM codecs / VFW codecs |
 |
| Dstruct |
| Posted: Jun 20 2011, 05:56 AM |
 |
|
Advanced Member
  
Group: Members
Posts: 257
Member No.: 24005
Joined: 8-August 08

|
| QUOTE (phaeron @ Jun 18 2011, 11:58 PM) | | Hmm, that Save As Default bug is interesting. I don't think the code for that's changed in a couple of years... surprised it didn't come up before. Anyway, it's fixed in test 11 (it wasn't syncing the state of the dialog before saving the config). | Thanks! Fixed in 1.10.1-test10. |
 |
| ale5000 |
| Posted: Jun 25 2011, 11:25 AM |
 |
|

Advanced Member
  
Group: Members
Posts: 1114
Member No.: 22180
Joined: 30-September 07

|
I have used File => Save segmented AVI (video: full processing mode uncompressed, audio: direct stream copy), then I stopped it and it is crashed:
| CODE | VirtualDub crash report -- build 34610 (release) --------------------------------------
Disassembly: 0041fc00: 8b09 mov ecx, [ecx] 0041fc02: 85c9 test ecx, ecx 0041fc04: 7409 jz 0041fc0f 0041fc06: 8b01 mov eax, [ecx] 0041fc08: 8b5004 mov edx, [eax+04h] 0041fc0b: 6a01 push 01h 0041fc0d: ffd2 call edx 0041fc0f: c3 ret 0041fc10: 51 push ecx 0041fc11: 56 push esi 0041fc12: 8bf1 mov esi, ecx 0041fc14: 8b0e mov ecx, [esi] 0041fc16: 85c9 test ecx, ecx 0041fc18: 7421 jz 0041fc3b 0041fc1a: 83c8ff or eax, 0ffh 0041fc1d: 8d5104 lea edx, [ecx+04h] 0041fc20: f0 lock 0041fc21: 0fc102 xadd [edx], eax 0041fc24: 83e801 sub eax, 01h 0041fc27: 750c jnz 0041fc35 0041fc29: 85c9 test ecx, ecx 0041fc2b: 7408 jz 0041fc35 0041fc2d: 8b01 mov eax, [ecx] 0041fc2f: 8b10 mov edx, [eax] 0041fc31: 6a01 push 01h 0041fc33: ffd2 call edx 0041fc35: c70600000000 mov dword ptr [esi], 00000000 0041fc3b: 8bc6 mov eax, esi 0041fc3d: 5e pop esi 0041fc3e: 59 pop ecx 0041fc3f: c3 ret 0041fc40: 8b4104 mov eax, [ecx+04h] 0041fc43: 3d38c86300 cmp eax, 0063c838 0041fc48: 7407 jz 0041fc51 0041fc4a: 50 push eax 0041fc4b: e831e11b00 call 005ddd81 0041fc50: 59 pop ecx 0041fc51: c3 ret 0041fc52: cc int 3 0041fc53: cc int 3 0041fc54: cc int 3 0041fc55: cc int 3 0041fc56: cc int 3 0041fc57: cc int 3 0041fc58: cc int 3 0041fc59: cc int 3 0041fc5a: cc int 3 0041fc5b: cc int 3 0041fc5c: cc int 3 0041fc5d: cc int 3 0041fc5e: cc int 3 0041fc5f: cc int 3 0041fc60: 64a100000000 mov eax, fs:[00000000] 0041fc66: 6aff push 0ffh 0041fc68: 6848d05f00 push 005fd048 (RtlUnwind@16+1582) 0041fc6d: 50 push eax 0041fc6e: 64892500000000 mov fs:[00000000], esp 0041fc75: 56 push esi 0041fc76: 8b742414 mov esi, [esp+14h] 0041fc7a: 89742414 mov [esp+14h], esi 0041fc7e: 83b9d800000000 cmp dword ptr [ecx+d8], 00h 0041fc85: c744240c000000 mov dword ptr [esp+0ch], 00000000 00 0041fc8d: 7412 jz 0041fca1 0041fc8f: 8b4c2418 mov ecx, [esp+18h] 0041fc93: 8b06 mov eax, [esi] 0041fc95: 8b503c mov edx, [eax+3ch] <-- FAULT 0041fc98: 6a01 push 01h 0041fc9a: 6a00 push 00h 0041fc9c: 51 push ecx 0041fc9d: 8bce mov ecx, esi 0041fc9f: ffd2 call edx 0041fca1: 8b06 mov eax, [esi] 0041fca3: 8b500c mov edx, [eax+0ch] 0041fca6: 8bce mov ecx, esi 0041fca8: ffd2 call edx 0041fcaa: 8b06 mov eax, [esi] 0041fcac: 8b5004 mov edx, [eax+04h] 0041fcaf: 6a01 push 01h 0041fcb1: 8bce mov ecx, esi 0041fcb3: c7442410ffffff mov dword ptr [esp+10h], ffffffff ff 0041fcbb: ffd2 call edx 0041fcbd: 8b4c2404 mov ecx, [esp+04h] 0041fcc1: 64890d00000000 mov fs:[00000000], ecx 0041fcc8: 5e pop esi 0041fcc9: 83c40c add esp, 0ch 0041fccc: c20c00 ret 000c 0041fccf: cc int 3 0041fcd0: 6aff push 0ffh 0041fcd2: 6873d05f00 push 005fd073 (RtlUnwind@16+15ad) 0041fcd7: 64a100000000 mov eax, fs:[00000000] 0041fcdd: 50 push eax 0041fcde: 64892500000000 mov fs:[00000000], esp 0041fce5: 51 push ecx 0041fce6: 56 push esi 0041fce7: 57 push edi 0041fce8: 6a28 push 28h 0041fcea: 8bf9 mov edi, ecx 0041fcec: e822ec1b00 call 005de913 0041fcf1: 83c404 add esp, 04h 0041fcf4: 89442408 mov [esp+08h], eax 0041fcf8: 33f6 xor esi, esi 0041fcfa: 3bc6 cmp eax, esi 0041fcfc: 89742414 mov [esp+14h], esi
Built on Althena on Sat Jun 18 14:40:53 2011 using compiler version 1400
Windows 5.1 (Windows XP x86 build 2600) [Service Pack 3] Memory status: virtual free 1896M/2048M, commit limit 2406M, physical total 2048M
EAX = 00000000 EBX = 00000000 ECX = 00000001 EDX = 0061242c EBP = 05e2ff34 ESI = 00cc71c8 EDI = 00000000 ESP = 05e2feb8 EIP = 0041fc95 EFLAGS = 00010202 FPUCW = ffff027f FPUTW = ffffffff
Crash reason: Access Violation
Crash context: An out-of-bounds memory access (access violation) occurred in module 'VirtualDub'...
...reading address 0000003C...
...while running thread "Processing" (thread.cpp:179).
Pointer dumps:
EDX 00612428: 00421440 00420990 0041fc60 004215f0 0041fab0 00421580 00421440 00421440 ESI 00cc71c8: 00000000 00cc74a0 002a52f4 00000000 00000000 00690056 00000000 00000000 ESP 05e2feb8: 00cd5130 05e2fee4 005fd048 00000000 004d7a7f 00cc71c8 00000001 00000001 05e2fed8: 00cd5130 004201cf 00cd5b20 05e2ff28 005fd048 00000000 004230d0 00cd5130 05e2fef8: 00000001 00000000 0013f104 00cd5b20 00ccace0 004023b7 05e2fcb0 00000001 05e2ff18: 00000000 00000000 00cd5b20 05e2ff00 05e2ff68 005fd370 00000002 05e2ff74 EBP 05e2ff30: 00000002 05e2ff74 00516ad6 00ccace0 0060dbe0 00000000 0063c870 000000b3 05e2ff50: 0063c858 0061255c 0063c870 000000b3 0063c858 0061255c 05e2ff9c 00605479 05e2ff70: 00000000 05e2ffac 005e35ae 00cd5b20 56e6877e 0013f104 00ccace0 00ccace0 05e2ff90: c0000005 05e2ff80 05e2fae0 05e2ffdc 005e6d30 5363102a 00000000 05e2ffec
Thread call stack: 0041fc95: VDAVIOutputFileSystem::CloseSegment() 004d7a7f: VDAVIOutputSegmented::finalize() 004201cf: VDAVIOutputWAVSystem::CloseSegment() 004230d0: _catch$?ThreadRun@VDDubProcessThread@@MAEXXZ$2() 004023b7: ?$VDProtectedAutoScope1::?$VDProtectedAutoScope1() 00516ad6: VDThread::StaticThreadStart() 005e35ae: _callthreadstartex() 005e3653: _threadstartex@4() 7c80b729: kernel32!GetModuleFileNameA [7c800000+b56f+1ba]
-- End of report |
-------------------- New VirtualDub forum VirtualDub AIO (All-in-One installer for VirtualDub and plugins) Codec Toolbox RS (A tool to read/change merit of codecs and many other things) Input plugins for VirtualDub / ACM codecs / VFW codecs |
 |
| Dstruct |
| Posted: Jul 5 2011, 04:29 PM |
 |
|
Advanced Member
  
Group: Members
Posts: 257
Member No.: 24005
Joined: 8-August 08

|
1.10.1-test11
-> View -> Log
Result: no checkmark in front of "Log" menu entry
Expected: checkmark in front of "Log" menu entry when Log window is opened (same as with Status window). |
 |
| Dstruct |
| Posted: Jul 13 2011, 01:54 PM |
 |
|
Advanced Member
  
Group: Members
Posts: 257
Member No.: 24005
Joined: 8-August 08

|
1.10.1-test11
-> Video -> "Filters" dialog
should remember the last clicked/selected Filter so that we immediately can click on "Configure" for example (without the need to select a Filter first) ... |
 |
| jpsdr |
| Posted: Jul 22 2011, 09:00 AM |
 |
|
Advanced Member
  
Group: Members
Posts: 335
Member No.: 20490
Joined: 23-December 06

|
Hello. In the VideoFilter.h of source code 1.10.1-test11 there is :
| CODE | prefetchProc = sizeof(VDXVideoFilterPrefetchOverloadTest(&T::Prefetch)) > 1 ? T::FilterPrefetch : NULL; prefetchProc2 = sizeof(VDXVideoFilterPrefetch2OverloadTest(&T::Prefetch2)) > 1 || sizeof(VDXVideoFilterPrefetchOverloadTest(&T::Prefetch)) > 1 ? T::FilterPrefetch2 : NULL;
accelRunProc = sizeof(VDXVideoFilterAccelRunOverloadTest(&T::Prefetch)) > 1 ? T::FilterAccelRun : NULL;
eventProc = T::FilterEvent;
accelRunProc = sizeof(VDXVideoFilterAccelRunOverloadTest(&T::Prefetch)) > 1 ? T::FilterAccelRun : NULL;
mSourceCountLowMinus1 = T::kMinInputCount - 1; mSourceCountHighMinus1 = T::kMaxInputCount - 1;
|
Is it normal to have twice accelRunProc, with the exact same line ? |
 |