I mainly use TransGainer on victims of the loudness war to give them a subtle boost. It works really well for this, results can be quite "invisible" sounding, resembling light handed EQing/expanding instead of transient design (this is a good thing, not a con).
I've been trying the new Precise mode for the same purpose though and can't get anywhere with it. I've only had a day with it, and I think I understand what the new controls do, but this new Precise mode is like Legacy/Sharp on steroids. It detects a lot more transients in general and things detected can register WAY higher on the meter. This is problematic since this erratic behavior is the "groove" for every other tweak after (if you increase trans gain, it will be jerky sounding quickly since it's starting with a jerky motion and also detected transients register higher so they are given the maximum trans gain value more consistently). My process in the legacy modes is pretty quick. Usually I know if TransGainer will work on it or I should something instead, and it's quite casual. Sometimes I'll push the trans gain super high to "pumping" levels and then increase sustain gain so that the "body" of increased transients also rises as well... it would only affect the most "1.0" transients though so it doesn't undo the trans gain. I think it helps to blend things in a better way than contour.
Anyways, I was just wondering.. is the Precise mode better for individual instruments and tracks? I get the feeling that is the point since it's so targeted and contains the additional controls... I don't really have access to high quality material, but I can imagine on that it wouldn't have such inconsistent behavior due to a better input and would work as intended then. I used to increase trans gain to 7, no sustain, low contour, low detect, but with this one, it gets crazy after only only 2! Very spiky.
I also have another idea. Inverse detection mode? What I mean by this is instead of high detection threshold only letting the most defined/loudest transients through and suppressing "dirty" transients, it would just do the opposite... or when it does detect strong/loud transients, those strong/loud ones don't make the meter go all the way up to 1.0. I'm suggesting this because in squashed songs, transients can be weakly defined and definitely not uniform - a limited drum hit may appear to really set the detection meter off one second, and a second later it's barely registering anything... this kind of action leads to random volume spikes and erratic sound. And it's mainly due to the fact that the "best" transients invoke trans gain (since only the transients that reach 1.0 use the full trans gain value). This also isn't good because those best transients could get increased way more than they should... I know, I could put a limiter on them after to get them under control or limit the source material, but that makes the problem worse because the material is already been limited in the first place! An inverse mode would place more priority on weak transients so they get most of the transient modification and the stray ones that really set the meter off don't get ridiculous volume spikes. I think you know what I'm getting at. A simpler alternative would be to have higher detected transients roll off or be pulled back, that way when the meter "surges", everything doesn't start sound "unstable". Loud transients wouldn't be treated as loud as they really are.
I guess that's all. I like picking your brain and brainstorming lol. Thanks for reading this again.
This topic was last updated 180 days ago, and thus it was archived. Replying is disabled for this topic.