Voxengo Premium Membership - All Voxengo Plugins For a Fixed One-Time Fee
Forums     Plugins     Soniformer Input slope pt. 2

This topic was created before release of the latest product version, and it may contain details irrelevant to this version.  Replying is disabled for this topic.

Last

Next

Previous

First



The manual states that "modern music" has an input slope of 3.5 - 4.0.  However, most of my material flattens at about 5.0 - 5.5.  I record at 32/96.  Is this the reason for the higher input slope, or might it be my sound card records "darker"?  Also, I've been told to set the input slope to zero.  Do I do this prior to making initial adjustments to the input slope, i.e.  I call up Soniformer within my host program and the input slope is at 3.0.  Does setting it to zero, reset everything so I get an accurate input slope reading?  I have NOT been doing this if the above is correct, so maybe that's why the higher reading.  Help please, and thanks.

Right now I tend to use 4.5 dB/oct slope myself.  I think this is more universally-usable slope (and I do plan to update user's manual in that respect).

3.0 db/oct input slope will be too bright if you will be trying to balance it on the display. (by 'balance' I mean have spectral power at 100 Hz equal to that of 10kHz).  5.5 dB/oct is definitely darker sounding slope (from the point of view of 'balance' I've mentioned) - for example, dry acoustic guitar recordings may have this slope.  However, if you add high-hat to them, you'll easily achieve 4.5dB/oct slope.

When you set Slope to zero, you will be getting what you usually see on FFT analyzer.  Non-zero slope is what I think currently mostly welcome in any spectrum analyzer, because, for example, classic spectrum analyzers (what Soniformer resembles as well) had 3 dB/oct slope - this slope was created automatically due to band-pass filtering.

So, zero slope can only be given by FFT-based analysis.


Aleksey,

I turned the input slope down to zero, and got a pretty smooth decending graph that disappeared around 5.25kHz.  Two concerns.  One, when looking at the "bass" end of this spectrum (zero input slope) the amplitude (light blue lines) out is about 1/8 - 1/4 of an inch higher than the amplitued (dark blue lines) in, and there is NO amplitude above 5.25kHz.  Is this normal?  Also, I have to turn up the input slope (on my material) to 4.8 - 5.0 just to fill the bottom of the spectrum floor (get a flat spectrum, making sure I have lines across the entire bottom of the plugin screen).  The low bass section, (48.8Hz) in most cases is still higher then the top end of the high, (10.2kHz+) which leads me to think I need to tweak more so that both extreme ends of the spectrum are relatively equal in height.  I have always assumed that increasing the input slope would give me a brighter sound, due to more high being introduced, rather than a darker one.  Is this true, and am I correct in my assumption of what constitutes a flat spectrum?

Thanks


How darker and lighter bars look relative to each other depends on compression settings.  When no compression/expanding is taking place, they are equal in height.

Of course, when slope is zero, most of the higher frequency content will go out of range.

From my experience, the average picture you should see is a balanced 'inverted smiley' with bars around 100 Hz roughly equal to bars around 10.0k Hz.  AFAIK this picture is what almost every commercial recording strives to achieve.  Beside that, slope setting is what helps to get the desired brightness (4.5 dB/oct is what I usually see now).

Input slope increases higher frequency compontents on the spectrum analyzer hence making them look more powerful than lower frequencies so that you want to tame them even more.  That is why higher slope value leads to 'darker' sound.


Aleksey,

I tried your suggestions on a few of my songs.  With regard to 100Hz and 10kHz being roughly equal, should that be after the initial input slope setting or after tweaking the threshold, attack, etc. so the final specturm looks has this?  Also, when I'm monitoring the OUT GAIN, I see tips of blue lines from sub-bass to low midrange, then every now and then blue lines extend downward toward the -3dbl line.  What should I look for when "tuning" for gain adjustment?

Thanks again


daydad: I tried your suggestions on a few of my songs.  With regard to 100Hz and 10kHz being roughly equal, should that be after the initial input slope setting or after tweaking the threshold, attack, etc. so the final specturm looks has this?

This depends on your goal.  If you want 4.5 dB/oct and 100 & 10k are equal, you are close to 4.5 dB/oct.  But if you are adjusting slope without making any changes, this is the same as measuring the slope of the recording.

daydad: Also, when I'm monitoring the OUT GAIN, I see tips of blue lines from sub-bass to low midrange, then every now and then blue lines extend downward toward the -3dbl line.  What should I look for when ''tuning'' for gain adjustment?

First of all, I should know what do you want to achieve, to answer your question.  In the example you have posted Soniformer compresses lower frequencies.  If this is what you want, you do not have to do anything else.


Aleksey,

I guess I have been looking for an absolute "right" way to use these plug-ins, and maybe that in it self is the problem.  There is NO right way.  Since I'm working with relatively small near-field monitors I 'm trying to develop good translation from studio to car or portable cd player.  Sometimes things play better on one than on another.  I guess I'm looking for good if not perfect balance that translates universally.  If the spectrum lines on the highs are much shorter than the lows after using a preset like "solid", and turning the input slope to 4.5, is this simply because in that particular preset the highs are much more compressed (Threshold slope line)?  I've been trying to get the highs and lows (spectrum lines) equal height thinking this is what a balanced spectrum looks like.  That's why I've been going up to 6.0+ on th input slope.

With regard to the Out Gain, the low end stays fairly level however I see the spikes on the highs often (extending down toward -3dbl).  Otherwise, the lines just disappear midway.  Just wanted to know what the spikes meant, and what to do with them?  Again looking for balance and translation.

Thanks


daydad: If the spectrum lines on the highs are much shorter than the lows after using a preset like ''solid'', and turning the input slope to 4.5, is this simply because in that particular preset the highs are much more compressed (Threshold slope line)?  I've been trying to get the highs and lows (spectrum lines) equal height thinking this is what a balanced spectrum looks like.  That's why I've been going up to 6.0+ on th input slope.

After moving to 4.5 dB/oct slope you should have adjusted Soniformer settings as well - for example, use its Out Gain envelope to add some power to highs so that even at 4.5 dB/oct slope high and low heights are roughly equal (should be more of an inverted smiley picture, without too dip notches and peaks).

daydad: With regard to the Out Gain, the low end stays fairly level however I see the spikes on the highs often (extending down toward -3dbl).  Otherwise, the lines just disappear midway.  Just wanted to know what the spikes meant, and what to do with them?  Again looking for balance and translation.

These 'spikes' is nothing to worry about - they are the result of compressor starting to work (highs cross the threshold line).

By the way, as far as I understand, there is no such thing as 'perfect transferring' of audio track from one audio system to another.  All you can do is to achieve 'intermediate' sound that does not generate extremes on various sound systems (i.e. does not sound too boomy on sound systems that tend to sound boomy, and is not too sibilant on systems that tend to sound sibilant).


Aleksey,

These are my readings: At 4.5 dbl/oct the 10.4kHz spectrum line is at -48 and the 100Hz line is at -30.  I must create dark mixes or something.  When I move the input slope to 5.0, 10.4kHz moves up to above -41+ and 100Hz come down to -36.  I understand that I could squish the bass all the way down and use the OUT Gain to add 6 or 7 dbl to the highs, but won't I totally kill the song?  And..cause headaches for myself in Curve EQ?

Thanks once again,


I do not think you'll kill your song.  From my experience, adjusting slope of the recording (not adding any dips or notches) does not destroy sound - only changes its overall brightness.  This may not be an universal observation, of course, but it seems to work that way (our brains get accustomed to constant slope and may start perceiving 'dull' song as a brighter one, and brighter song may listen as a duller one - brains exhibit a kind of auto-correction).  Only when you compare the same song at different slopes you can tell which one is duller and which one is brighter.
This topic was created before release of the latest product version, and it may contain details irrelevant to this version.  Replying is disabled for this topic.

Last

Next

Previous

First