[The Matches A Band In Hope (Advance) 2008 FNT.rar
Hanne Rylaarsdam
I don't understand how your suggestions will help. With HQP I apply a fixed negative value to prevent EVER going over 0 dBFS and that is determined by my convolution filters boost independently of any given track.
It appears from my tests that PGGB changes the volume of each track to ensure optimal use of available dynamic range, at least during the eQing stage. In my tests, the gain was changed to bring the highest PCM sample value to 0 dBFS, that seems to be the peak normalization default in PGGB. I don't understand how your suggestions, would it be normalization, lowering highest value, will respect the original gain of each track.
Ok you claim offering a remastering tool and changing the volume of each track to ensure optimal use of available dynamic range is remastering. And I suspect it's considered a plus by PGGB aficionados. But I don't want to do that.
When doing EQ, it is by definition going to alter the peak levels of the track. For real-time processing, the safest route is to reduce the gain to where it never clips because it is hard to know in advance what the peak level will be. But this does not guarantee the peak levels remains the same across tracks either, as how the track is affected depends on the frequency content of the tracks. More than the peak levels, it is even harder to judge what the perceived loudness will be.
When left at the default (Auto gain with no peak normalization), PGGB matches the pre-eq peak to post-eq peak as a way to keep the peak gain the same (with the hope the post eq loudness remains the same), this is how I listen to music most of the time and I seldom find the need to adjust my volume any more than I would adjust the volume on the original tracks. The only time it normalizes to 0 is if you set the gain to 0. The only other time peak normalization is done when gain is set to auto is when there are inter-sample-overs that would result in clipping.
Neither applying a fixed gain nor ensuring pre eq levels and post eq levels remain the same are going to guarantee the loudness remains the same because that depends both on the eq you use and the frequency content of the track. The later approach tries to maximize the headroom and it comes very handy when there is not a whole lot of headroom to play with) for example I used to drive my high efficiency Omega CAMs direct from Chord DAVE) and having to reduce the gain by 6dB all the time would have a significant impact on how much I can get out of DAVE. Most importantly if headroom is not an issue, it still does no harm.
I do understand your point, which is it makes it hard for you to compare to a different method that does apply a fixed gain, but that has nothing to do with PGGB doing things differently or PGGB aficionados preferring it that way, but rather a limitation of real-time processing given both methods are bound to alter the relative loudness some way or the other. You are the first to even request a change like this (one I may be open to in the future for those who are in a similar situation as you).
What is your take about constant (meaning existing in the 4 compared tracks) difference of several dBs around 90 Hz compared to HQP with the WAter's DSOTM ? Same at 1600 with the Shirley Horn track ? With the Horn track I could confuse harmonica with an accordion with PGGB and it seems consistent with the shyness at 1600.
System: TT7 PGI 240v + Power Base > Paretoaudio Server [SR7T] > Adnaco Fiber [SR5T] >VR L2iSE [QSA Silver fuse, QSA Lanedri Gamma Infinity PC]> QSA Lanedri Gamma Revelation RCA> Omega CAMs, JL Sub, Vox Z-Bass/ /LCD-5/[QSA Silver fuse, QSA Lanedri Gamma Revelation PC] KGSSHV Carbon CC, Audeze CRBN
So let's imagine there's no HQP I'd still have an over loud track 1 and if I set volume based on it, I'd be frustrated with following tracks. Or say I set volume after listening to a favorite track (the rest of the album is more level consistent), track 1 would be awfully too loud.
From photography we know aperture vs speed priority ; there's been hypothesis about PGGB being frequency priority (vs time domain) ; my take at the moment is that PGGB's priority is optimal use of available dynamic range. Makes adagios to ravel in more "revealing" I guess
PGGB destroys my HQP settings. What I sometimes found pleasant before (a more intimate soundstage, slightly smoothed-over sound) is backfiring here. PGGB makes these tracks sound crisp-clean and huge (soundstage), turning a "modest"/intimate presentation into a big show. And because of the depth and layering I'm hearing things which I've neglected before (they were still there, just too congested).
One important thing - I will be testing with the gauss-half-band filter from now on. It is apparent that gauss-long, being shorter and not going all the way up to Nyquist, causes some (easily detectable) sacrifices in the size of the soundstage and the resultant presentation of sounds. I often prefer this, but this makes a comparison with PGGB quite pointless (when trying to reach semi-objective conclusions). Also, the half-band is non-apodising, like PGGB.
Now the size of the soundstage is similar (I think I'd still give the edge to PGGB), however to my ears PGGB suffers from the Sinc-L effect whereby although the soundstage is huge, the sounds within it are not as holographic as I'd like them to be and in comparison (note - only in comparison!) come across as a bit more 2D. The Gaussian filter (also in PCM) gives them more perceptible depth. Not the depth of the soundstage as such, here PGGB still rules - depth in terms of my perception of e.g. the drums being an event that has a very tactile 3D presence within the soundstage as opposed to being a slightly "flatter" event happening somewhere away from me (and, as a result - less engaging).
Maybe it's the headphones' limitation to present distant sounds with sufficient presence? Not sure, however that's what I'm hearing on my set-up. Listening to these particular tracks, PGGB'd, I kind of feel like a sound engineer where each sound is presented very clearly, nicely separated, on a huge canvass (soundstage) - making it easy for me to see how they are placed in the mix. There is engagement too, but it doesn't last long because my brain is just not convinced that this is "real enough" (could be highly subjective, of course). The gauss-half-band filter seems to fix that to a large degree, making each individual sound "pop out" more with the final result being that I'm simply more engaged and more interested in both each individual sound as well as the performance as a whole (with the gauss-long filter, I'd say the focus gets significantly shifted towards each individual sound as the size of the soundstage decreases).
IMHO it's such a fantastic way to demonstrate the power of PGGB 256 on the road, we could visit anyone we want since it's considered a relatively portable / transportable rig (e.g. with a pair of Abyss Diana) if you ask me
795a8134c1