You are not logged in. Please login or register.
SmoothVideo Project → Posts by SubJunk
You're a genius! MeGUI had a dummy version of opencl.dll in the x264 folder and now it works after replacing that with the one from my video card. I'm so happy to have this fixed, thanks a lot
By the way, GPU mode in SVP itself works, and so does the preview mode in MeGUI. The error only happens when trying to run the encoding, and other users are still reporting the issue (the latest report is at http://forum.doom9.org/showthread.php?p … st1710845)
*shrugs* guess I have to give up on GPU mode
I have a GT 640 using driver version 344.65.
This is the output from CLInfo:
Number of platforms: 1
Platform Profile: FULL_PROFILE
Platform Version: OpenCL 1.1 CUDA 6.5.29
Platform Name: NVIDIA CUDA
Platform Vendor: NVIDIA Corporation
Platform Extensions: cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll
Platform Name: NVIDIA CUDA
Number of devices: 1
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4318
Max compute units: 2
Max work items dimensions: 3
Max work items[0]: 1024
Max work items[1]: 1024
Max work items[2]: 64
Max work group size: 1024
Preferred vector width char: 1
Preferred vector width short: 1
Preferred vector width int: 1
Preferred vector width long: 1
Preferred vector width float: 1
Preferred vector width double: 1
Native vector width char: 1
Native vector width short: 1
Native vector width int: 1
Native vector width long: 1
Native vector width float: 1
Native vector width double: 1
Max clock frequency: 901Mhz
Address bits: 18397247599280160
Max memory allocation: 536870912
Image support: Yes
Max number of images read arguments: 256
Max number of images write arguments: 16
Max image 2D width: 32768
Max image 2D height: 32768
Max image 3D width: 4096
Max image 3D height: 4096
Max image 3D depth: 4096
Max samplers within kernel: 32
Max size of kernel argument: 4352
Alignment (bits) of base address: 4096
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: Yes
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: Read/Write
Cache line size: 128
Cache size: 32768
Global memory size: 2147483648
Constant buffer size: 65536
Max number of constant args: 9
Local memory type: Scratchpad
Local memory size: 49151
Kernel Preferred work group size multiple: 32
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1000
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue properties:
Out-of-Order: Yes
Profiling : Yes
Platform ID: 024C0578
Name: GeForce GT 640
Vendor: NVIDIA Corporation
Driver version: 344.65
Profile: FULL_PROFILE
Version: OpenCL 1.1 CUDA
Extensions: cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64
I haven't had a chance to install old drivers yet but it seems like other people are having the same issue recently, I just had a comment on my conversion guide "For those getting the “can’t read the video stream” error message : If you have added “GPU=true” in the script, remove it."
Chainik
Maybe some new video card drivers, I'll try old ones
Thanks for the replies 
MAG79
I used Driver Fusion to uninstall the drivers (in safe mode) and it didn't work.
Chainik
The list of missing DLLs for me is exactly the same as the ones in the article I linked to. I have tried with and without the Visual C++ Runtimes since that's what most Google articles seem to say is the cause, but it makes no difference.
Here's a screenshot of what Dependency Walker says for me: http://i.imgur.com/k7L3pYf.png
I have this issue now and I can't figure it out. I checked those folders for OpenDL.dll and it's in there.
I also used Dependency Walker but it seems to not be good for Windows 7, since I ran into the problem described at http://stackoverflow.com/questions/1702 … l-problems
I tried uninstalling Microsoft Visual Studio and installing the SDK recommended in that forum thread too but it still didn't work.
Hi guys, yeah like Chainik said, when using UMS it also has to do encoding which means it's slower than just using SVP.
Using AviSynth/FFmpeg should be faster than AviSynth/MEncoder, so if we can solve that runtime error it will be good. I'll try to help you solve that. Do you have the exact error?
The smooth tuning uses algo 23 I have been using it lately and I like it. try it and see
You can also try using Tuning="Smooth", it can have less artifacts with some content
TCmullet
I know what you mean about using the standard versions of things. In this case, though, the standard version of AviSynth has no multithreading, and for modern CPUs that means things are running at 1/8 or less of the possible speed.
The version I offer for download is one of SEt's builds. It is the most stable version I've tested (100% stability even with dozens of simultaneous plugins)
Oh and by the way, as a general note unrelated to InterFrame, the denominator 1001 is usually used for NTSC framerates. It makes it easier because it's just easy to remember, and accurate to more decimal places. So for 90fps it would be NewNum=90000,NewDen=1001. Not a big deal at all but thought you might be interested to know
TCmullet
Hi, thanks for reporting the error, and thanks MAG79 for letting me know.
I tried your script myself and it worked fine. It may be an error with the version of AviSynth you used. Please try the version in this zip file (AviSynth.exe) http://www.spirton.com/uploads/60FPS/Files-20120627.zip
Using SetMemoryMax 1024 is overkill, 512 is better since it is more likely to be stable.
This is the code template I would recommend for realtime processing with InterFrame:
Setmemorymax(512)
SetMTMode(3,4)
LoadPlugin("CORRECT_PATH/svpflow1.dll")
LoadPlugin("CORRECT_PATH/svpflow2.dll")
Import("CORRECT_PATH/InterFrame2.avsi")
ffdshow_source()
SetMTMode(2)
InterFrame(Cores=4)
SetMTMode(1)
GetMTMode(false) > 0 ? distributor() : lastChainik
That would be great for SVP 3.0.4 It is a minor change.
Hi guys
Just wondering if you plan to synchronise your version of mvtools2 with the official 2.5.11.2?
There is a change in it that will be used in an upcoming version of a filter I use.
Try increasing the searchparam values (not sure what the GUI option for that is, though)
MAG79
I edited the version information in the dll file so now it should work properly. It's here:
http://www.spirton.com/uploads/60FPS/Files-20110503.zip
Ah I see!
The file is SET's 2.6 build, which is the only 2.6 MT build AFAIK so it is probably relatively popular
That was silly of him to not use version information.
MAG79
Sure, here it is
http://www.spirton.com/uploads/60FPS/Files-20110430.zip
MAG79
Thanks for the reply
When I run SVP with my avisynth.dll, it doesn't let me run the program.
It just says "This computer has old Avisynth version (2.5). Version 2.5.8 is required. Install Avisynth right now?" then I click "No" because your version causes encoding crashes for me, then it says "The program completes its work, as not installed a new version of AviSynth v2.5.8", then the program closes.
I didn't know where to post this, but thanks a lot for the 3.0.2 update. The speed increase is impressive!
I checked 3.0.2 and it is still compulsory in the installer
In our tests on avsforum plevel=2 increases warping of lines compared to plevel=0 so I disagree with this
SmoothVideo Project → Posts by SubJunk
Powered by PunBB, supported by Informer Technologies, Inc.