search for: loopfilt

Hi people, As I said before: I did the IDCT to run on the FPGA. My friends from university did the Reconstruction routines running on the FPGA. I'm helping with the LoopFilter, and it is almost there. (all VHDL) I did a small profiling of the libTheora running on a Altera Stratix II device: The processor used was the NIOS II with 8Kb of data and instruction cache, branch prediction and hardware divider. (this is the more roubust NIOS II version). I decoded some f...

Hi folks, FYI: I've finally made some benchmarks for inline-assembler versus intrinsic based mmx code. I've just applied the changes to the fragment reconstruction functions as writing the IDCT and loopfilter have not been ported yet. Nevertheless here are some numbers: As a baseline I'll take the current version from the trunk with all inline assembler functions enabled. Lower values mean lower performance. All functions with inline-asm: 100% inter_mmx replaced by C-fun...

...ity have an interest in a multi-thread decoder implementation? I'm starting to work with multi-thread and I thought that Theora Decoder is a good choice for me, because I had been working with it in a FPGA implementation and I have experience with the library. I'm thinking in working with LoopFilter at first. Do you think I could start with it or there is a better choice? Thanks -- Leonardo de Paula Rosa Piga Undergraduate Computer Engineering Student LSC - IC - UNICAMP http://www.students.ic.unicamp.br/~ra033956

Great news! Theora is running on FPGA. After almost a year of a great effort we have Theora validated on FPGA. Now I will try to integrated the hardware with a video controller to see the video! I completely implemented the ExpandBlock, CopyRecon, LoopFilter and UpdateUMVBorder functions. The ReconRefFrames function was partially implemented and the part before will run on a software compiled for Altera Nios2 Processor. The implementation is based on version alpha6 and due FPGA SRAM memory constraints the maximum resolution is about 280x210. Increa...

I'm trying to implement the dct_decoder's LoopFilter function in a FPGA. I need to know what is the maximum memory allocation size which display_fragment and recon_pixel_index_table points. These are fields of PB_INSTANCE structure that is declared in the codec_internal.h Thank you =========================== Leonardo de Paula Rosa Piga Compute...

Hello, First of all, I would like to say that my work that I wrote in the other email would be to do in hardware the functions: CopyRecon, LoopFilter and UpdateUMVBorder. These are modules that Leonardo had made, but it wasn't ok in FPGA. When I had a chat with Leonardo we were thinking in rewrite these module for to do this running in FPGA (to debug in a Hardware level is much more difficult, sometime is faster rewrite the code). But, Leo...

As it turns out, with the proprietary firmware, the VP3 and VP4 interfaces are identical. Furthermore, this is all already implemented for nvc0. So these patches (a) move the easily sharable bits of the nvc0 implementation into the nouveau directory, and then (b) implement the other parts in nv50. The non-shared parts are still largely copies, but there are some differences, not the least of which

...oc_dequant_tables_init 2 3.5e-04 2 5.2e-04 ld-2.3.2.so (no symbols) 1 1.8e-04 0 0 dump oc_huff_tree_mindepth 1 1.8e-04 0 0 dump oc_huff_tree_occupancy As you can see loopfilter needs some work. After loopfilter opts I will take a look again on oc_state_frag_recon_mmx and oc_state_frag_inter2_mmx and maybe I can add quantization in MMX too. Derf have you any objections againts the patch? Thanks regards Rudolf -------------- next part -------------- diff -Naur a/lib/id...

Hi Stepan, Sorry for the delay. It's great that you are working on MergeFunctions as well and I agree, we should definitely try to combine our efforts to improve MergeFunctions. Just to give you some context, the pass (with the similar function merging patch) is already being used in a production setting. From my point of view, it would be better if we focus on improving its capability

quite some speed improvement indeed. attached the updated patch to apply to svn/trunk. j -------------- next part -------------- A non-text attachment was scrubbed... Name: theora-mmx.patch.gz Type: application/x-gzip Size: 8648 bytes Desc: not available Url : http://lists.xiph.org/pipermail/theora-dev/attachments/20040824/5a5f2731/theora-mmx.patch-0001.bin

...0 0.01 68843 0 0.01 68843 load.ll 5 38589 0 0.01 38569 0 0.01 38569 loadpg.ll 2 73982 0 0.01 73949 0 0.01 73949 lobject.ll 12 62363 0 0.01 62348 0 0.01 62348 LockedStream.ll 7 16832 0 0.01 16800 0 0.01 16800 long_term.ll 4 66031 0 0.01 66001 0 0.01 66001 lookup.ll 2 68844 0 0.01 68813 0 0.01 68813 loopFilter.ll 8 157592 0 0.02 157571 0 0.02 157571 loop_unroll.ll 602 1276589 8 0.29 1276255 * * * lopcodes.ll 0 5630 0 0.01 5615 0 0.01 5615 loslib.ll 17 46260 0 0.01 46245 0 0.02 46245 lowercase.ll 4 10513 0 0.01 10498 0 0.01 10498 lparser.ll 69 259824 0 0.03 259809 0 0.03 259809 lpbench.ll 8 51184 0 0.02...