[FFmpeg-devel] AMR-NB decoder

[Robert Swain]

Hello,

2009/8/10 Colin McQuillan <m.niloc at googlemail.com>:
> Reynaldo, Robert, could you review your parts of this patch please? I
> checked the QCELP ref decoder before writing the comment.
>
> 2009/8/10 Michael Niedermayer <michaelni at gmx.at>:
>> On Mon, Aug 10, 2009 at 08:42:53AM +0100, Colin McQuillan wrote:
>>> 2009/8/9 M?ns Rullg?rd <mans at mansr.com>:
>>> > Colin McQuillan <m.niloc at googlemail.com> writes:
>>> >
>>> >> 2009/8/8 Michael Niedermayer <michaelni at gmx.at>:
>>> >>> On Sat, Aug 08, 2009 at 04:09:39PM +0100, Colin McQuillan wrote:
>>> >>>> 2009/8/8 Michael Niedermayer <michaelni at gmx.at>:
>>> >>>> > On Fri, Aug 07, 2009 at 08:23:53PM +0100, Colin McQuillan wrote:
>>> >>>> >> 2009/8/6 Michael Niedermayer <michaelni at gmx.at>:
>>> >>>> >> > On Wed, Aug 05, 2009 at 05:51:36PM +0100, Colin McQuillan wrote:
>>> >>>> >> >> Attached is a patch for an AMR-NB decoder.
>>> >>>>
>>> >>>> [...]
>>> >>>>
>>> >>>> >> > that should e a seperate patch
>>> >>>> >>
>>> >>>> >> I'll leave this one until I investigate a version for sparse vectors. Attached:
>>> >>>> >>
>>> >>>> >> 1. Helper functions for gain control in floating-point codecs
>>> >>>> >> I couldn't find a similar fixed point function to copy the function name.
>>> >>>> >>
>>> >>>> >> 2. Floating-point version of ff_acelp_high_pass_filter
>>> >>>> >
>>> >>>> >> ?acelp_vectors.c | ? 22 ++++++++++++++++++++++
>>> >>>> >> ?acelp_vectors.h | ? 27 +++++++++++++++++++++++++++
>>> >>>> >> ?2 files changed, 49 insertions(+)
>>> >>>> >> f1abbee9b62c1779fd5fb1c634d4ab4294d8611d ?get-set-energyf.patch
>>> >>>> >> Index: libavcodec/acelp_vectors.c
>>> >>>> >> ===================================================================
>>> >>>> >> --- libavcodec/acelp_vectors.c ? ? ? ?(revision 19606)
>>> >>>> >> +++ libavcodec/acelp_vectors.c ? ? ? ?(working copy)
>>> >>>> >> @@ -155,3 +155,25 @@
>>> >>>> >> ? ? ? ? ?out[i] = weight_coeff_a * in_a[i]
>>> >>>> >> ? ? ? ? ? ? ? ? + weight_coeff_b * in_b[i];
>>> >>>> >> ?}
>>> >>>> >> +
>>> >>>> >> +float ff_energyf(const float *v, int length)
>>> >>>> >> +{
>>> >>>> >> + ? ?float sum = 0;
>>> >>>> >> + ? ?int i;
>>> >>>> >> +
>>> >>>> >> + ? ?for (i = 0; i < length; i++)
>>> >>>> >> + ? ? ? ?sum += v[i] * v[i];
>>> >>>> >> +
>>> >>>> >> + ? ?return sum;
>>> >>>> >> +}
>>> >>>> >
>>> >>>> > ff_dot_productf)(
>>> >>>>
>>> >>>> Do you mean that ff_energyf is redundant? I've taken it out.
>>> >>>
>>> >>> hmm well, as you say it that way, ff_energyf() could be faster due to
>>> >>> fewer mem reads, if that is te case in practice it could be kept
>>> >>
>>> >> ff_energyf is reliably 4% faster in my test, so I'll add it back in.
>>> >
>>> > That function has high simdicity so it should be added to dsputil and
>>> > simdified.
>>>
>>> I'll try, but I didn't mean to imply that energy calculations are
>>> critical to performance. The slow parts of the AMR decoder are the IIR
>>> and FIR filters, which are already in celp_filters.c.
>>
>> if the SIMD energy is not reaching an overall (whole codec) 0.1% speedup over
>> using a more generic SIMD dot product then its probably not worth it and
>> could be ommited
>
> Great! Then I'll leave out ff_energyf, and avoid using "length" in the comments.
>
> [...]
>
>> alignment requirements are supposed to be written liks:
>> void ff_vp3_idct_put_c(uint8_t *dest/*align 8*/, int line_size, DCTELEM *block/*align 16*/);
>> also "/*" is not doxygen compatible and what the function does should be
>> more verbosely described, energy isnt a mathematically clear term, dot
>> product and sum or squares are.
>>
>
> Changed.

The only thing I don't like is the ff_scale_to() function name. I
wouldn't know what it did without reading the description. I know a
distaste for the use of 'energy' for a generic function has been
expressed but I'd understand something like ff_acelp_scale_energy().
Does anyone have any better suggestions? Reynaldo? Michael? Colin?

Regards,
Rob