IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Question about --dm-presets
Chu
post Oct 28 2001, 16:00
Post #1





Group: Members
Posts: 15
Joined: 28-October 01
Member No.: 367



Well, with hard drive prices finally firmly under $2 a gig, I think it's about time to rip my CD collection, because i'm getting sick of porting my binder around. I've been doing a lot of research, and I think I'm gonna end up using standard or xtreme presets, but have some questions.

#1: Can you give me an idea of exactly what the differences between standard and xtreme? I've done some sample rips, and the only tracks i've noticed a real difference on are "noisy" songs, especially some industrial stuff, but are there particular types of songs where the difference is much more acute?

#2: A lot of people have told me to investigate standard -Z if I don't mind the filesizes xtreme produces. What exactly does -Z do? I've looked at the docs and it's only listed as "expiremental."

#3: metal . . . hrmm . . . What exactly about this makes it better for metal then standard? Also, and I guess this ties in with that question, but how exactly are we defining metal? I imagine my Pantera cd's would firmly fall into this category, but would some more heavy guitar stuff like Siamese Dream fall under here too, or some of this "numetal" stuff like Staind?

#4: I've been looking around here and r3mix, and people have been mentioning that dibrom is about to update his standard presets. I am a patient soul, but what type of improvements are we talking about exactly?

In any case, thanks for the input.

-Chu

P.S: Excuse my poor spelling.
Go to the top of the page
+Quote Post
RD
post Oct 28 2001, 17:50
Post #2





Group: Members
Posts: 151
Joined: 29-September 01
Member No.: 31



Well Dibrom is really the expert about these things....but I'll try to save him some work :-)

#1: Can you give me an idea of exactly what the differences between standard and xtreme? I've done some sample rips, and the only tracks i've noticed a real difference on are "noisy" songs, especially some industrial stuff, but are there particular types of songs where the difference is much more acute?

xtreme is identical to standard except for a lower ath curve.
While this should make mp3s sound a little "cleaner" there is a sample where it sounds worse than standard (high hat sound i believe) As such Dibrom generally recommends standard over xtreme as xtreme is more experimental... his new compile will be better than all of these... but is not out yet...

#2: A lot of people have told me to investigate standard -Z if I don't mind the filesizes xtreme produces. What exactly does -Z do? I've looked at the docs and it's only listed as "expiremental."

-Z is a toggle switch now, and in this case turns noise shaping off...it seems that for some reason turning it off will help standard to encode transients better...

#3: metal . . . hrmm . . . What exactly about this makes it better for metal then standard? Also, and I guess this ties in with that question, but how exactly are we defining metal? I imagine my Pantera cd's would firmly fall into this category, but would some more heavy guitar stuff like Siamese Dream fall under here too, or some of this "numetal" stuff like Staind?

Nothing about it makes it better; and it is not better. It is a compromise line. Some people don't want metal mp3s to rise above 200 kbps in filesize and this line is supposed to prevent such a rise....it also tries to fix some problems that plague the now passe r3mix line...for example, dropouts, ringing etc.

#4: I've been looking around here and r3mix, and people have been mentioning that dibrom is about to update his standard presets. I am a patient soul, but what type of improvements are we talking about exactly?

Yes, no date given yet for release but his new compile will have bettter noise shaping/measuring and lower filesizes than --dm-preset standard -X3 which currently gives the best quality...
problem clips like fatboy.wav should also be improved...

Take care,
RD
Go to the top of the page
+Quote Post
JohnV
post Oct 28 2001, 20:18
Post #3





Group: Developer
Posts: 2797
Joined: 22-September 01
Member No.: 6



QUOTE
Originally posted by RD
-Z is a toggle switch now, and in this case turns noise shaping off...it seems that for some reason turning it off will help standard to encode transients better...

-Z doesn't turn off/on noise shaping. With -Z you toggle between noise shaping 1 and 2. Noise shaping 2 is the default with nspsytune. Noise shaping 2 is supposed to be more "intelligent" and save bits, while noise shaping 1 uses larger "margins" and that leads to higher bitrate. Noise Shaping 2 ain't intelligent enough with certain types of signals (eg. Bass pulses ala Fatboy).
QUOTE
Yes, no date given yet for release but his new compile will have bettter noise shaping/measuring and lower filesizes than --dm-preset standard -X3 which currently gives the best quality...
problem clips like fatboy.wav should also be improved...
Take care,
RD

I don't think there will be new noise shaping models. Still the same 1 and 2. But there will be new noise measurement setting. What I have understood is that Dibrom combines several existing noise measuring methods to create better working combination.


--------------------
Juha Laaksonheimo
Go to the top of the page
+Quote Post
RD
post Oct 29 2001, 00:02
Post #4





Group: Members
Posts: 151
Joined: 29-September 01
Member No.: 31



I wrote in haste (had to take the wife-to-be out to shopping but still wanted to give Chu some answers...I had a feeling I'd make some mistakes...

JohnV is quite right on both points.

However, I wonder if Dibrom's new noise measuring stuff is designed to work with noise shaping 1 or 2? or both? But maybe only optimally with one of the two models?

RD
Go to the top of the page
+Quote Post
JohnV
post Oct 29 2001, 00:40
Post #5





Group: Developer
Posts: 2797
Joined: 22-September 01
Member No.: 6



QUOTE
Originally posted by RD
However, I wonder if Dibrom's new noise measuring stuff is designed to work with noise shaping 1 or 2? or both? But maybe only optimally with one of the two models?


The whole point of the tweakings is that noise shaping 2 is used to keep the bitrate down with "normal music", but still achieve higher than noise shaping 1 quality with clips like fatboy. Also transient handling in general should be much better.

Dibrom's goal is to implement the tweakings also for vbr-mtrh (--r3mix uses vbr-mtrh), which means high encoding speed... One problem there at least was, is that vbr-mtrh doesn't have all the noise measurement methods vbr-old has. Also vbr-mtrh's noise measurement methods are technically implemented a bit differently than vbr-old's, so separate tweakings (or at least testings) must be done for both vbr-modes.


--------------------
Juha Laaksonheimo
Go to the top of the page
+Quote Post
Chu
post Oct 29 2001, 06:25
Post #6





Group: Members
Posts: 15
Joined: 28-October 01
Member No.: 367



Thanks for the feedback. I think i'll be waitig for the new compile to start ripping. Oh, and by the way, will there/are there any good presets for ripping to ~128K for portables?

-Chu
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 26th December 2014 - 04:14