Dilemma about using replay gain and EQ DSPs together

[R1sky]

Hello,

I'm new in using dbpoweramp, so I'm not much skilled in it. I wanted to ask about my problem:

I have few hundreads of CDs ripped in flac and I want to use all of them in my iPod with some EQ. However iPod's equalizer is very crappy - EQ is applied after amplifying and in most cases produces distortion (clipping). The only way to apply EQ is to lower the gain of track itself (using iTUNnorm tags (sound check) - Apple version of reaplygain - is useless, because just like EQ, sound check is applied after amplifying). I know it is a destructive way for audio files, but the only way to use EQ. But iPod's preinstalled EQ presets aren't good for me, so I'm thinking about using EQ DSP while converting flac into alac. All of songs I have are recorded at different volume levels (95-100 dB). So the question is:

What is the best way to adjust gain of all songs to the same level (95 dB), while also adding graphic EQ into them? The priority is to damage audio files as little as possible.


Note: None of my songs have replaygain tags in them, so probably it can't be done with "replay gain (apply)" DSP.

[Spoon]

Re: Dilemma about using replay gain and EQ DSPs together

Normally you do not boost frequencies, rather reduce the frequencies of all the other frequencies you do not want to be boosted.

[R1sky]

Re: Dilemma about using replay gain and EQ DSPs together

I was thinking about the same process. So lowering some frequencies should do the same equalization effect like boosting other, but the difference is, that when lowering frequencies, there shouldn't be more audible noise added compared to original file, just volume should be a little lower. Am I right?

Could you please give me a "step by step" guide how to add permanent replay gain - album gain to 95 dB (while not having RG tags in songs) and also adding EQ DSP? Which one should I put first in DSP effects list? Does it even matter?

Thank you.

[Spoon]

Re: Dilemma about using replay gain and EQ DSPs together

Correct

you would use these dsp effects:

bit depth - set to floating point
EQ
volume normalize (set to replaygain)
bit depth - to 16 bit (with triangular dither)

[R1sky]

Re: Dilemma about using replay gain and EQ DSPs together

Oh, great, I didn't know about using bit depth. If I understand it right, changing bit depth helps apply EQ more accurate. The last question I have is about using volume normalize. When I want to use it, the only replay gain option is track gain. But that's not exactly what I need. If I'll use it, songs that should be quiet will be at the same volume level as much "louder" songs. Is there any other way of doing this? (e.g. force dbpoweramp to add RG tags first and then apply bit depth, EQ, replay gain (apply) and another bit depth)

[Spoon]

Re: Dilemma about using replay gain and EQ DSPs together

Float allows it to clip without loosing information.

Replaygain will put all tracks the same level...

[R1sky]

Re: Dilemma about using replay gain and EQ DSPs together

Replaygain will put all tracks the same level...

I know, but that's the problem. When I'll want to play any gapless album, there won't be volume differences between "loud" songs and songs, that should be "quiet". Whole album dynamics will be gone. That's why I want to apply Album gain rather than Track gain.

[Spoon]

Re: Dilemma about using replay gain and EQ DSPs together

You have to: Add replaygain tags to all your files, then convert with the DSP Effect: ReplayGain (Apply) set to album gain.

[R1sky]

Re: Dilemma about using replay gain and EQ DSPs together

Ok, I thought the only way is to use a 2-step-method. Thank you anyway. By the way, when I select flac file and "convert" it to flac with RG DSP effect, does the program anything to file itself besides RG tag?

[Spoon]

Re: Dilemma about using replay gain and EQ DSPs together

Yes it reencodes, you should install the utility codec [Replay Gain] from codec central, then convert the FLAC file to this codec to tag the file.

[R1sky]

Re: Dilemma about using replay gain and EQ DSPs together

Please excuse me for still bothering you, I just wanted to make sure I don't mess up anything. So it's the same to convert FLAC -> FLAC with Replay Gain tag -> ALAC with applied Replay Gain and EQ, or to convert FLAC -> ALAC with RG tag -> ALAC with applied RG and EQ? As both are lossless formats, the result should be the same.

Anyway, I found something else I'm not sure about. I ripped a CD few months ago with dbpoweramp release 14.2 and now again with release 14.3. The first time I was just trying to rip CD with something different from EAC and I was satisfied with result (I only missed CUE sheet with pregaps detection). However I forgot to set offset for my BD drive. When I ripped the CD second time, I set offset, enabled C2 pointers and everything. I don't know much about CRC, but the problem is, that every song has different CRC and I can see some very minor differences in spectrogram. However when I verify both rips with CUE tools, they look ripped accurately. I'm adding rip info and CUE tools result below:


Rip 1:

Ripping with drive 'F: [Optiarc - BD RW BD-5300S ]', Drive offset: 0, Overread Lead-in/out: No
AccurateRip: Active, Using C2: No, Cache: 1024 KB, FUA Cache Invalidate: No
Pass 1 Drive Speed: 4, Pass 2 Drive Speed: 4
Ultra:: Vary Drive Speed: Yes, Min Passes: 3, Max Passes: 6, Finish After Clean Passes: 2
Bad Sector Re-rip:: Drive Speed: 4, Maximum Re-reads: 34

Encoder: FLAC -compression-level-5 -verify

.
.
.

17 Tracks Ripped Securely

(the CD was not found in the database, so there were 2 passes and then additional 3 ultra secure passes)



CUE tools:

[CUETools log; Date: 06/10/2012 22:51:34; Version: 2.1.4]
[CTDB TOCID: QVnZ6jlVHcpmRMQTDSOn1k4ZCu0-] found.
Track | CTDB Status
1 | (1/1) Accurately ripped
2 | (1/1) Accurately ripped
3 | (1/1) Accurately ripped
4 | (1/1) Accurately ripped
5 | (1/1) Accurately ripped
6 | (1/1) Accurately ripped
7 | (1/1) Accurately ripped
8 | (1/1) Accurately ripped
9 | (1/1) Accurately ripped
10 | (1/1) Accurately ripped
11 | (1/1) Accurately ripped
12 | (1/1) Accurately ripped
13 | (1/1) Accurately ripped
14 | (1/1) Accurately ripped
15 | (1/1) Accurately ripped
16 | (1/1) Accurately ripped
17 | (1/1) Accurately ripped
[AccurateRip ID: 002258c0-01b5606c-f90d6511] found.
Track [ CRC | V2 ] Status
01 [0a361b8b|38ac458a] (0+0/3) No match
02 [8cb44897|03afd169] (0+0/3) No match
03 [8783cfa1|d0398f06] (0+0/3) No match
04 [c6336511|9a300ab8] (0+0/3) No match
05 [65101416|b13e324b] (0+0/3) No match
06 [e2c1a6cf|00276965] (0+0/3) No match
07 [2dbd93cf|62667f98] (0+0/3) No match
08 [fef7dbdc|a8180ab2] (0+0/3) No match
09 [c603614f|2c22e04e] (0+0/3) No match
10 [4488c4a3|d255217b] (0+0/3) No match
11 [40e5166b|4a42a136] (0+0/3) No match
12 [6e327de1|5c6ba9c8] (0+0/3) No match
13 [8f005ee0|75c5fa58] (0+0/3) No match
14 [a31dcbfb|e964e08d] (0+0/3) No match
15 [12ba3568|2203e339] (0+0/3) No match
16 [d8eb3ed7|21882807] (0+0/3) No match
17 [723edc5e|dfcd0269] (0+0/3) No match
Offsetted by 6:
01 [1670e3b2] (3/3) Accurately ripped
02 [368c1667] (3/3) Accurately ripped
03 [204764bd] (3/3) Accurately ripped
04 [c7cb833b] (3/3) Accurately ripped
05 [83671400] (3/3) Accurately ripped
06 [689ca87b] (3/3) Accurately ripped
07 [0915b765] (3/3) Accurately ripped
08 [0154eed6] (3/3) Accurately ripped
09 [7a900b0d] (3/3) Accurately ripped
10 [41e59d95] (3/3) Accurately ripped
11 [bfbe1dc7] (3/3) Accurately ripped
12 [7a2feedf] (3/3) Accurately ripped
13 [02f6bbdc] (3/3) Accurately ripped
14 [5887ef3f] (3/3) Accurately ripped
15 [9b8a3220] (3/3) Accurately ripped
16 [29ea9a87] (3/3) Accurately ripped
17 [35a92b32] (3/3) Accurately ripped

Track Peak [ CRC32 ] [W/O NULL]
-- 100.0 [0D164B7E] [BE040041]
01 89.9 [79F558CC] [7A9FBFC0]
02 88.2 [56EDB16F] [E93F67EB]
03 88.2 [4E0034DB] [FAFFE555]
04 98.5 [186BE12A] [C877A02D]
05 87.2 [4873FC88] [7487D930]
06 97.1 [5A590E49] [5B745FA3]
07 92.6 [EF195FEE] [86D46DC3]
08 75.6 [BB9480F5] [3D000971]
09 96.0 [91D53110] [EECCE4DA]
10 73.7 [5D988707] [A7495742]
11 100.0 [CEF8ACC7] [46F2994B]
12 78.3 [8E676AAC] [023A15C9]
13 100.0 [7ACBC3FD] [B36CF69C]
14 82.3 [AAF71067] [75070511]
15 89.1 [41B38DA0] [94D78E03]
16 88.2 [18421D83] [523CB35C]
17 85.6 [42504C87] [45B445B1]




Rip 2:

Ripping with drive 'F: [Optiarc - BD RW BD-5300S ]', Drive offset: 6, Overread Lead-in/out: No
AccurateRip: Active, Using C2: Yes, Cache: 1024 KB, FUA Cache Invalidate: No
Pass 1 Drive Speed: 4, Pass 2 Drive Speed: 8
Ultra:: Vary Drive Speed: Yes, Min Passes: 3, Max Passes: 6, Finish After Clean Passes: 3
Bad Sector Re-rip:: Drive Speed: 4, Maximum Re-reads: 34

Encoder: FLAC -compression-level-5 -verify

.
.
.

17 Tracks Ripped Accurately



CUE tools:

[CUETools log; Date: 06/10/2012 22:54:56; Version: 2.1.4]
[CTDB TOCID: QVnZ6jlVHcpmRMQTDSOn1k4ZCu0-] found.
Track | CTDB Status
1 | (1/1) Accurately ripped
2 | (1/1) Accurately ripped
3 | (1/1) Accurately ripped
4 | (1/1) Accurately ripped
5 | (1/1) Accurately ripped
6 | (1/1) Accurately ripped
7 | (1/1) Accurately ripped
8 | (1/1) Accurately ripped
9 | (1/1) Accurately ripped
10 | (1/1) Accurately ripped
11 | (1/1) Accurately ripped
12 | (1/1) Accurately ripped
13 | (1/1) Accurately ripped
14 | (1/1) Accurately ripped
15 | (1/1) Accurately ripped
16 | (1/1) Accurately ripped
17 | (1/1) Accurately ripped
[AccurateRip ID: 002258c0-01b5606c-f90d6511] found.
Track [ CRC | V2 ] Status
01 [1670e3b2|43e4da4a] (3+0/3) Accurately ripped
02 [368c1667|abc53ba9] (3+0/3) Accurately ripped
03 [204764bd|67cfabc1] (3+0/3) Accurately ripped
04 [c7cb833b|98e3698c] (3+0/3) Accurately ripped
05 [83671400|cf1f5d40] (3+0/3) Accurately ripped
06 [689ca87b|83dcc00f] (3+0/3) Accurately ripped
07 [0915b765|3d4d82a9] (3+0/3) Accurately ripped
08 [0154eed6|a93e5fb7] (3+0/3) Accurately ripped
09 [7a900b0d|ddd86abc] (3+0/3) Accurately ripped
10 [41e59d95|cd64730a] (3+0/3) Accurately ripped
11 [bfbe1dc7|c7de392e] (3+0/3) Accurately ripped
12 [7a2feedf|6728bc96] (3+0/3) Accurately ripped
13 [02f6bbdc|e8cff85b] (3+0/3) Accurately ripped
14 [5887ef3f|9d0ae0ca] (3+0/3) Accurately ripped
15 [9b8a3220|a9fccb28] (3+0/3) Accurately ripped
16 [29ea9a87|706fc9a4] (3+0/3) Accurately ripped
17 [35a92b32|a0eac691] (3+0/3) Accurately ripped

Track Peak [ CRC32 ] [W/O NULL]
-- 100.0 [88FF0FC0] [BE040041]
01 89.9 [DF1F0AF0] [7A9FBFC0]
02 88.2 [0CDEBB3A] [E93F67EB]
03 88.2 [0CBF7DA7] [FAFFE555]
04 98.5 [64254466] [C877A02D]
05 87.2 [4FFF724A] [7487D930]
06 97.1 [71D61637] [5B745FA3]
07 92.6 [49495720] [86D46DC3]
08 75.6 [09DAAD21] [3D000971]
09 96.0 [95136552] [EECCE4DA]
10 73.7 [B9BD5B3D] [A7495742]
11 100.0 [3D5EAD55] [46F2994B]
12 78.3 [13959324] [023A15C9]
13 100.0 [15BD3449] [B36CF69C]
14 82.3 [91558E7A] [75070511]
15 89.1 [AC10AD3E] [94D78E03]
16 88.2 [699577A6] [523CB35C]
17 85.6 [1DE6DDE7] [45B445B1]


Could you please explain me, where is the problem and which rip should be accurate? And if you know about some better ripping settings, please letme know, thank you in advance.

[Spoon]

Re: Dilemma about using replay gain and EQ DSPs together

Ripping with a different offset would give a different CRC, but the rip is no less valid than the other...

If doing ReplayGain apply, then the file is changed at the audio level, it is no longer classed as lossless (as lossless means the same as the CD).

[R1sky]

Re: Dilemma about using replay gain and EQ DSPs together

If doing ReplayGain apply, then the file is changed at the audio level, it is no longer classed as lossless (as lossless means the same as the CD).

Yes, you're right. I just meant that as both FLAC and ALAC are lossless formats, the result at the end should be the same no matter if I convert FLAC -> FLAC RG tag -> ALAC RG apply or FLAC -> ALAC RG tag -> ALAC RG apply. Because when you use similar procedure with mp3 and aac, the final result is different for sure.