Advanced search

On other websites (9858)

• Anatomy of an optimization: H.264 deblocking

  26 May 2010, by Dark Shikari — H.264, assembly, development, speed, x264

  As mentioned in the previous post, H.264 has an adaptive deblocking filter. But what exactly does that mean — and more importantly, what does it mean for performance? And how can we make it as fast as possible? In this post I’ll try to answer these questions, particularly in relation to my recent deblocking optimizations in x264.

  H.264′s deblocking filter has two steps: strength calculation and the actual filter. The first step calculates the parameters for the second step. The filter runs on all the edges in each macroblock. That’s 4 vertical edges of length 16 pixels and 4 horizontal edges of length 16 pixels. The vertical edges are filtered first, from left to right, then the horizontal edges, from top to bottom (order matters!). The leftmost edge is the one between the current macroblock and the left macroblock, while the topmost edge is the one between the current macroblock and the top macroblock.

  Here’s the formula for the strength calculation in progressive mode. The highest strength that applies is always selected.

  If we’re on the edge between an intra macroblock and any other macroblock: Strength 4
  If we’re on an internal edge of an intra macroblock: Strength 3
  If either side of a 4-pixel-long edge has residual data: Strength 2
  If the motion vectors on opposite sides of a 4-pixel-long edge are at least a pixel apart (in either x or y direction) or the reference frames aren’t the same: Strength 1
  Otherwise: Strength 0 (no deblocking)

  These values are then thrown into a lookup table depending on the quantizer: higher quantizers have stronger deblocking. Then the actual filter is run with the appropriate parameters. Note that Strength 4 is actually a special deblocking mode that performs a much stronger filter and affects more pixels.

  One can see somewhat intuitively why these strengths are chosen. The deblocker exists to get rid of sharp edges caused by the block-based nature of H.264, and so the strength depends on what exists that might cause such sharp edges. The strength calculation is a way to use existing data from the video stream to make better decisions during the deblocking process, improving compression and quality.

  Both the strength calculation and the actual filter (not described here) are very complex if naively implemented. The latter can be SIMD’d with not too much difficulty; no H.264 decoder can get away with reasonable performance without such a thing. But what about optimizing the strength calculation? A quick analysis shows that this can be beneficial as well.

  Since we have to check both horizontal and vertical edges, we have to check up to 32 pairs of coefficient counts (for residual), 16 pairs of reference frame indices, and 128 motion vector values (counting x and y as separate values). This is a lot of calculation; a naive implementation can take 500-1000 clock cycles on a modern CPU. Of course, there’s a lot of shortcuts we can take. Here’s some examples:

  • If the macroblock uses the 8×8 transform, we only need to check 2 edges in each direction instead of 4, because we don’t deblock inside of the 8×8 blocks.
  • If the macroblock is a P-skip, we only have to check the first edge in each direction, since there’s guaranteed to be no motion vector differences, reference frame differences, or residual inside of the macroblock.
  • If the macroblock has no residual at all, we can skip that check.
  • If we know the partition type of the macroblock, we can do motion vector checks only along the edges of the partitions.
  • If the effective quantizer is so low that no deblocking would be performed no matter what, don’t bother calculating the strength.

  But even all of this doesn’t save us from ourselves. We still have to iterate over a ton of edges, checking each one. Stuff like the partition-checking logic greatly complicates the code and adds overhead even as it reduces the number of checks. And in many cases decoupling the checks to add such logic will make it slower: if the checks are coupled, we can avoid doing a motion vector check if there’s residual, since Strength 2 overrides Strength 1.

  But wait. What if we could do this in SIMD, just like the actual loopfilter itself? Sure, it seems more of a problem for C code than assembly, but there aren’t any obvious things in the way. Many years ago, Loren Merritt (pengvado) wrote the first SIMD implementation that I know of (for ffmpeg’s decoder); it is quite fast, so I decided to work on porting the idea to x264 to see if we could eke out a bit more speed here as well.

  Before I go over what I had to do to make this change, let me first describe how deblocking is implemented in x264. Since the filter is a loopfilter, it acts “in loop” and must be done in both the encoder and decoder — hence why x264 has it too, not just decoders. At the end of encoding one row of macroblocks, x264 goes back and deblocks the row, then performs half-pixel interpolation for use in encoding the next frame.

  We do it per-row for reasons of cache coherency: deblocking accesses a lot of pixels and a lot of code that wouldn’t otherwise be used, so it’s more efficient to do it in a single pass as opposed to deblocking each macroblock immediately after encoding. Then half-pixel interpolation can immediately re-use the resulting data.

  Now to the change. First, I modified deblocking to implement a subset of the macroblock_cache_load function: spend an extra bit of effort loading the necessary data into a data structure which is much simpler to address — as an assembly implementation would need (x264_macroblock_cache_load_deblock). Then I massively cleaned up deblocking to move all of the core strength-calculation logic into a single, small function that could be converted to assembly (deblock_strength_c). Finally, I wrote the assembly functions and worked with Loren to optimize them. Here’s the result.

  And the timings for the resulting assembly function on my Core i7, in cycles:

  deblock_strength_c: 309
  deblock_strength_mmx: 79
  deblock_strength_sse2: 37
  deblock_strength_ssse3: 33

  Now that is a seriously nice improvement. 33 cycles on average to perform that many comparisons–that’s absurdly low, especially considering the SIMD takes no branchy shortcuts: it always checks every single edge! I walked over to my performance chart and happily crossed off a box.

  But I had a hunch that I could do better. Remember, as mentioned earlier, we’re reloading all that data back into our data structures in order to address it. This isn’t that slow, but takes enough time to significantly cut down on the gain of the assembly code. And worse, less than a row ago, all this data was in the correct place to be used (when we just finished encoding the macroblock)! But if we did the deblocking right after encoding each macroblock, the cache issues would make it too slow to be worth it (yes, I tested this). So I went back to other things, a bit annoyed that I couldn’t get the full benefit of the changes.

  Then, yesterday, I was talking with Pascal, a former Xvid dev and current video hacker over at Google, about various possible x264 optimizations. He had seen my deblocking changes and we discussed that a bit as well. Then two lines hit me like a pile of bricks:

  <_skal_> tried computing the strength at least?
  <_skal_> while it’s fresh

  Why hadn’t I thought of that? Do the strength calculation immediately after encoding each macroblock, save the result, and then go pick it up later for the main deblocking filter. Then we can use the data right there and then for strength calculation, but we don’t have to do the whole deblock process until later.

  I went and implemented it and, after working my way through a horde of bugs, eventually got a working implementation. A big catch was that of slices: deblocking normally acts between slices even though normal encoding does not, so I had to perform extra munging to get that to work. By midday today I was able to go cross yet another box off on the performance chart. And now it’s committed.

  Sometimes chatting for 10 minutes with another developer is enough to spot the idea that your brain somehow managed to miss for nearly a straight week.

  NB: the performance chart is on a specific test clip at a specific set of settings (super fast settings) relevant to the company I work at, so it isn’t accurate nor complete for, say, default settings.

  Update: Here’s a higher resolution version of the current chart, as requested in the comments.
  

  http://git.videolan.org/?p=x264.git;a=commit;h=b9ec3ea27812288485329ac7771143021cd4917b

• SNES Hardware Compression

  16 June 2011, by Multimedia Mike — Game Hacking

  I was browsing the source code for some Super Nintendo Entertainment System (SNES) emulators recently. I learned some interesting things about compression hardware. I had previously uncovered one compression algorithm used in an SNES title but that was implemented in software.

  SNES game cartridges — being all hardware — were at liberty to expand the hardware capabilities of the base system by adding new processors. The most well-known of these processors was the Super FX which allows for basic polygon graphical rendering, powering such games as Star Fox. It was by no means the only such add-on processor, though. Here is a Wikipedia page of all the enhancement chips used in assorted SNES games. A number of them mention compression and so I delved into the emulators to find the details:

  • The Super FX is listed in Wikipedia vaguely as being able to decompress graphics. I see no reference to decompression in emulator source code.
  • DSP-3 emulation source code makes reference to LZ-type compression as well as tree/symbol decoding. I’m not sure if the latter is a component of the former. Wikipedia lists the chip as supporting "Shannon-Fano bitstream decompression."
  • Similar to Super FX, the SA-1 chip is listed in Wikipedia as having some compression capabilities. Again, either that’s not true or none of the games that use the chip (notably Super Mario RPG) make use of the feature.
  • The S-DD1 chip uses arithmetic and Golomb encoding for compressing graphics. Wikipedia refers to this as the ABS Lossless Entropy Algorithm. Googling for further details on that algorithm name yields no results, but I suspect it’s unrelated to anti-lock brakes. The algorithm is alleged to allow Star Ocean to smash 13 MB of graphics into a 4 MB cartridge ROM (largest size of an SNES cartridge).
  • The SPC7110 can decompress data using a combination of arithmetic coding and Z-curve/Morton curve reordering.

  No, I don’t plan to implement codecs for these schemes. But it’s always comforting to know that I could.

  Not directly a compression scheme, but still a curious item is the MSU1 concept put forth by the bsnes emulator. This is a hypothetical coprocessor implemented by bsnes that gives an emulated cartridge access to a 4 GB address space. What to do with all this space? Allow for the playback of uncompressed PCM audio as well as uncompressed video at 240x144x256 colors @ 30 fps. According to the docs and the source code, the latter feature doesn’t appear to be implemented, though; only the raw PCM playback.

• Anomalie #2132 (Fermé): Vider les referers

  18 June 2011, by cedric -

  Super, je ferme le ticket !