Listening Test using Spahm Clip

Notes by ff123

 

The Spahm clip can be found at the Lame Test Samples Page. This clip is interesting because Lame easily beats the FhG codecs. The artifacts in the Alternate codec and in mp3enc sound remarkably alike -- I suspect that the Alternate codec is a descendant of mp3enc.

At 128kbs, both of the "Alternate" codecs as well as mp3enc are astonishingly bad. Lame is an order of magnitude better on this clip at this bitrate. You'll notice that I still didn't bother to make any ABX runs on any codec at 128kbs -- they all produced easily recognizable artifacts.

At 160kbs, the "Alternate" codec and mp3enc are still faring worse than the others. The best codecs (Lame, FhG 1.263i) are still clearly different from the original.

Lame is still among the top at 192kbs. The other codecs are doing much better at this bitrate, so I actually did perform ABX tests.

At 256kbs, all codecs except FhG 1.263i and Xing are indistinguishable from the original.

 

Encoder bitrate settings ABX results Comment
Xing in AudioCatalyst 2.1 128 j-stereo, high freq didn't bother slightly worse than FhG 1.263i
Lame 3.87a (RH) 128 j-stereo, -h didn't bother best of the bunch; slightly better than FhG 1.263i
FhG 1.2 build 63i external codec (Radium) 128 j-stereo, hq didn't bother not too bad; but still noticeable
FhG "Alternate" codec (Sound Forge 4.5h) 128 j-stereo,
highest quality
didn't bother sounds like blowing bubbles in water; tied for worst along with the "Alternate-bad" and mp3enc
FhG "Alternate-bad" codec (MusicMatch 5) 128 j-stereo, very
high
didn't bother same as above
FhG mp3enc 3.1 128 j-stereo, qual 9 didn't bother same as above
FhG "FastEnc" codec
(MusicMatch 5)
128 j-stereo, fastest didn't bother not as bad as the "alternate" codecs or mp3enc, but worse than Xing, FhG 1.263i, and Lame
QDesign MVP 1.2 128 j-stereo, hq didn't bother hissing noise; not as annoying as the bubbling noises of the "alternate" codecs or mp3enc
Xing in AudioCatalyst 2.1 160 j-stereo, high freq didn't bother noisy
Lame 3.87a (RH) 160 stereo, -h didn't bother noisy
FhG 1.2 build 63i external codec (Radium) 160 j-stereo, hq didn't bother noisy; Xing, Lame, and FhG 1.263 are all similar in quality
FhG "Alternate" codec (Sound Forge 4.5h) 160 stereo,
highest quality
didn't bother noisy; worse than Xing or Lame or FhG 1.263
FhG "Alternate-bad" codec (MusicMatch 5) 160 stereo, very high didn't bother same as above
FhG mp3enc 3.1 160 stereo, qual 9 didn't bother same as above
FhG "FastEnc" codec
(MusicMatch 5)
160 j-stereo, fastest didn't bother noisy; on a par with "Alternate" or mp3enc
QDesign MVP 1.2 160 j-stereo, hq didn't bother still hear hiss
Xing in AudioCatalyst 2.1 192 j-stereo,
high freq
16 of 16 can hear pre-echo, which makes the opening sound mushy
Lame 3.87a (RH) 192 stereo, -h 13 of 16 harder to detect than others
FhG 1.2 build 63i external codec (Radium) 192 j-stereo, hq 15 of 16 "Tinkling" noise at very beginning of clip
FhG "Alternate" codec (Sound Forge 4.5h) 192 stereo,
highest quality
16 of 16 faint "tinkling" noise in right channel; also slightly noiser than original
FhG "Alternate-bad" codec (MusicMatch 5) 192 stereo, very
high
15 of 16 faint "tinkling" noise in right channel; also slightly noiser than original
FhG mp3enc 3.1 192 stereo, qual 9 16 of 16 "tinkling" noise is louder than with "Alternative"
FhG "FastEnc" codec
(MusicMatch 5)
192 j-stereo,
fastest
16 of 16 sounds "fat" right at beginning of clip
QDesign MVP 1.2 192 stereo, hq 13 of 16 harder to detect than others
Xing in AudioCatalyst 2.1 256 j-stereo, high freq 16 of 16 noisy (more noticeable in right channel)
Lame 3.87a (RH) 256 stereo, -h 9 of 16 indistinguishable
FhG 1.2 build 63i external codec (Radium) 256 stereo, hq 15 of 16 sounds slightly "fatter" near beginning
FhG "Alternate" codec (Sound Forge 4.5h) 256 stereo, highest quality 8 ov 16 indistinguishable
FhG "Alternate-bad" codec (MusicMatch 5) 256 stereo, very high 8 of 16 indistinguishable
FhG mp3enc 3.1 256 stereo, qual 9 9 of 16 indistinguishable
FhG "FastEnc" codec
(MusicMatch 5)
256 j-stereo, fastest 10 of 16 indistinguishable
QDesign MVP 1.2 256 stereo, hq 6 of 16 indistinguishable

 

Return to ff123's home page