You are not logged in. Please login or register.
RME User Forum → FireWire & USB series → Applying EQ on software playback (UFX+)
Hello guys,
please help a poor guy out.
For a long time now I've been trying to figure out how I could apply EQ, namely an HPF to software playback. In other words, I'm trying to apply an HPF to Youtube videos for example so I can better practice my bass along with the tracks.
How do I do that? Do I need to reroute the playback somewhere?
Thanks in advance.
Yes, reroute with Loopback from Totalmix into a DAW where you apply EQ and send to Totalmix again (to different channel number than before to avoid feedback!)
Hey, I'm back with one more question about this loopback thing.
I finally had some time to deal with this loopbak situation and I managed to route it so that now it works the way I want to, so thank you for that.
However, here's the thing. For context, I'm using Reaper DAW.
Prior to learning this loopback function, for the longest time ever, I was pretty much sure that the sound I'm hearing through the RME monitoring (real time) wasn't the same as the recorded signal in Reaper. I could only prove this by hearing it and knowing it sounds different but I wasn't too sure if it's just a different volume in signal or what.
However, sure enough, immediately after I routed the loopback back into my reaper I went and recorded two tracks - one was recording Mic12 channel directly and another, which was set to record Analog 3/4, channel that I set to loopback into the DAW. That channel only had Mic 12 enabled, meaning I was recording mono anyways.
And sure enough, after carefully matching gains of these two tracks, after flipping the phase on one of them, as expected, I didn't hear a null. There was a very obvious left over of specifically mid section - especially obvious in the high mids.
Meaning, there is in fact a difference between recording straight channels or loopback channels with exact same settings.
Is there an explanation for this? I'm not aware of any. Thanks!
this phase flipping is done in Totalmix ?
If its done in Reaper with a plugin, that plugin might induce a latency.
From what you wrote it seems one channel has a slightly bigger latency.
Loopback in Totalmix adds a tiny little bit of latency. Even one sample difference would cause an unsuccessful null test!
But I can ensure you, the loopback happens completely on the digital part of the interface and is absolutely lossless. No difference between the signals. Do your test again with a delay plugin on the track with the direct mic input. Increase the delay time in one sample steps. I'm sure you’ll get to a point where the signals completely cancel out each other.
The flip phase is done in Reaper but with a built in option. It is latency "free" because I can just copy paste the same track twice, flip phase on one and the sum is a null every time.
Hmm it's a weird one because in Reaper you do actually have a built in nudge tool and you're able to nudge the track left or right on a 1 sample precision and under no circumstance was I able to close in to the null. I also tried nudging and adjusting gain of the phased track to see if there's a volume difference.
This is Digicheck after just flipping the phase, no nudging:
Quite a lot of left-over material even though it's at -30db and I gain staged the tracks to -18db. I even tried helping myself with the Hornet VU meter.
I may be lacking a proper scientific approach here. But seemingly, if my ears aren't fooling me, recording through loopback instead of direct channel somehow seems to give a more pleasing timbre and I'm now not sure aymore whether I'm hallucinating or what. 
After nudging it left and right the best result I could get was nudging the loopback precisely 15 samples backwards, and the following is the result:
Particularly some noticeable leftovers in the 2k hi-mids which is actually what my ear hears when I just listen to the loopback and raw channel take. I always thought that the recording sounds more bite-y and real-time monitoring always sounded more "polite". Very interesting. I would be super curious to hear more about this if you guys have the will to comment on it.
Now regarding gain staging - to my understanding, if only gain staging was the issue in this case (from this point on), the flipped phase result should amount to basically the same EQ spectrum as the original, just lower volume. But this isn't the case here.
sure you bypassed the HPF plugin ? Or did you just switched off the HPF and left the plugin on ?
Here's my TotalMix screenshot:
Also all plugins and monitoring FX line in Reaper were disabled for this purpose. I only used 2 instances of VU meter for gain staging but have also tried doing the same without VU meter just in case if there was any difference but it turned out there was no latency present there either. The image was the same. In fact, dropping the Loopback track gain manually for -0.16 db gave me a bit better result but it only affected the volume, image stayed the same - I guess I at least managed to isolate the source of the difference in frequency graph to phase flip.
Edit: Nevermind the activated EQ under "Mic 12" channel, it wasn't on when I recorded the test sample.
Also, on the picture you will see "Monitor FX" being disabled in Reaper, but I also cleared all plugins in there so there was no bypass at all because no plugins were present.
I’ve done several TotalMix loopback / null tests. It’s 100% sample accurate, there is absolutely nothing to hear anymore as soon you find the perfect alignment. Maybe you have an activated EQ or such somewhere in the signal chain?
I’ve done several TotalMix loopback / null tests. It’s 100% sample accurate, there is absolutely nothing to hear anymore as soon you find the perfect alignment. Maybe you have an activated EQ or such somewhere in the signal chain?
There shouln't be because I'm recording straight into my UFX+, no EQs present, no plugins inside reaper, nothing in total mix, all FX faders are off. I'm really not sure what might be causing this.
But if all of you guys are getting perfect nulls then I must be doing something wrong but yet I don't know what that might be. I tested this on a very "barebone" setup, no preamps outside of RME interface. One bass guitar, one cable, that's it.
What are the buffer settings ?
Might be a different input buffer than output buffer (in Reaper) ?
That's possible in MacOS, in Windows this is probably not possible.
Buffer setting is set to 256 samples via MADIface settings and my instance of Reaper is set to copy paste that setting, so I'm not forcing internal buffering whatsoever.
Here's an interesting thing - through the same chain I created a loopback of a random song and imported that same song to a separate channel. And just like that, 15 samples of nudge backwards, there was indeed a perfect null. I have no idea why the same doesn't happen to recorded instruments in my session.
Could it be related to the mix of mono and stereo tracks maybe?
It crossed my mind too. I'll repeat the test with mono loopback.
But in any case, if that's the issue, it's actually a good one because I really prefer the sound I get through stereo loopback even though I have no idea what's going on.
Mono to stereo reduces signal level by 3 dB due to pan.
RME User Forum → FireWire & USB series → Applying EQ on software playback (UFX+)
Powered by PunBB, supported by Informer Technologies, Inc.