Possible problem case with --alt-preset normal rev6

2001-11-21 23:49:18

Hi to all,
First of all congratulations to Dibrom and everyone else involved for a very high quality preset. While testing I came across this case where the above preset  produced a somewhat problematic result.
http://www.geocities.com/sad_baffoon/temple.zip
(Rainbow -  Temple Of the King, LPAC file, 336 KB)

Around the 0.5 sec mark, I notice some chirping on the right channel (accoustic guitar). I've confirmed this through ABX. Please check this out and tell me what you think

Zbutsam

Possible problem case with --alt-preset normal rev6

Reply #1 – 2001-11-22 00:10:21

Thanks, I'll take a look.

Can you test with --alt-preset high, and --r3mix, to get a point of reference on how well they do on this clip vs normal?

I probably won't be able to listen to this myself till later tonight.

Possible problem case with --alt-preset normal rev6

Reply #2 – 2001-11-22 01:36:24

To give you a description of the problem, there seems to be some defect on the original sample as well, a small chirp around the 0.5 sec on the right channel. When encoded to mp3 with --alt-preset normal, this artifact becomes louder, more "watery" sounding and seems longer in duration. You have to turn the volume a bit up to listen to it and, perhaps, move the balance slider fully to the right. --alt-preset high shows the same behaviour. --r3mix sounds somewhat better. I hope you will be able to reproduce this problem

With regards
Zbutsam

Possible problem case with --alt-preset normal rev6

Reply #3 – 2001-11-22 12:59:47

Hmm, i believe you´re talking about the chirp which happens about 0.85s (lame decoded) at the same time with guitar sound and loud drum.

Yeah, it's easily audible in the original as well (right channel).
What is your abx score if you just concentrate on the chirp?

Personally i hear more clearer differencies for both r3mix and alt-normal by just listening the loud drums and guitar notes (although you have to listen very carefully).

Don't get me wrong, but I believe you found this chirp also from the original after you were trying to pick up artifacts from the encoded ones? Just in my opinion there are some more obvious issues than this chirp...

Anyway, I'm not sure for how many people any of these are audible...

Even bigger problem in my opinion is that alt-normal is about 60kbps bigger than r3mix. And since I hear differencies with both alt-normal and r3mix, I'm not sure this bitrate difference can be justified...