Reamping adds major noise - what's wrong? [Solved]

Yes, I read your post and am not questioning it. I just would like to understand why the impedance change tames the top and bottom end but this forum might not be the right place to discuss physics.
 
Yes, I read your post and am not questioning it. I just would like to understand why the impedance change tames the top and bottom end but this forum might not be the right place to discuss physics.
Look at the speaker tab of the amp block and the impedance peaks. What happens in mismatches they flatten...
 
I finished testing this morning

This file includes six sound bytes - DI track and then re-amp track for each of my three "DI's".

1) 18i8 Instrument in DI track
2) Previous DI track reamped through AX8
3) ISA one DI track
4) Previous DI track reamped through AX8
5) AX8 DI track
6) Previous DI track reamped through AX8


There are large chunks of silence between the sound bytes, but that is not real silence, you are hearing the actual noise level - Tape was still rolling, so to speak.

Except for level adjustments (LUFS level around -20 dBLUFS), absolutely no processing has been done in Reaper.

My Focusrite 18i8 only has main outs - so I had to use one of the mains to send to AX8 - That is actually not that bad, since my main out has an analog volume control, so I could adjust the volume going into AX8 in the analog realm. I had no problem getting sensible levels into AX8.

I apologize for the slightly bad (very bright) re-amp tone, but other than that, I have no reservations re-amping with AX8 like this. I will not be buying a re-amp box.
 
I find it weird/unlikely that AX8 should add more noise than your average DI box / audio interface.

I understand that a re-amp box might be needed, but the initial DI should not be necessary. Maybe @FractalAudio or @Danny Danzi can share their experience in this regard?

EDIT: I will get my hands on a Focusrite ISA One tonight, so I can compare all kinds of DI (Guitar->Interface; Guitar-> DI->Interface; Guitar->DI (thru)->AX8->interface;Guitar->AX8->Interface)

I've never actually tried to literally re-amp through the AX8. My issue was attempting to get the DI part right. The quickest and best remedy in my opinion is to run SPDIF if you have an interface that supports it. Your DI will be pure, clean and reamping will be the same as using an MKII/XL+.

The cleaner the signal, the better the re-amp. That said, I have been able to create a pretty awesome DI from my AX8 just using it as is. BUT...that won't help with the issue at hand here, which is actual re-amping. I've not tried that, but I should probably give it a shot at some point just to see how it reacts.

Now, I'm still a total n00b on this stuff other than what works for me using my XL+ via SPIDIF or using the XLR's on the XL+ for reamping and DI. But for the AX8, I'm not sure how you would feed your signal back in. My first choice (again, I'm clueless here...but it would be the first thing I try) would be to run an output from the interface I have the AX8 out 2 going into, into the input of the AX8.

To describe my setup for DI on the AX8 a little further using the options that matter for me in a DI situation....

For DI creation

Levels page
Instrument pad: 6.0
Out 1: +4
Out 2 FX send pad: 0.0

Audio page
Out 1: Stereo
Out2 Fx Send Echo: Input 1

Physical out 2 jack sent to input of soundcard interface.
Physical output 2 knob on unit set for 1 o'clock.

This gets me around the -12dB range for my DI on the AX8. Now, I COULD run SPDIF from the AX8 to my interface, but I have my XL+ already running that way. So, because I don't really use the AX8 for DI or for reamping having the XL+, I just wanted to see how good I could get the DI. The above works really well.

If I were to try and reamp, I would just take the output of the interface and run it into the input of the AX8. I may have to mess with the interface +4/-10 settings there...but I believe this would give me decent results. I'll try to give this theory a shot and get back to you. I'm nearly done for the day and it wouldn't take me long to try.

The interface I'm using in this particular example is an old Echo Layla 24/96. Old but powerful....and with good converters and an acceptable noise floor.

OK, paused my writing and tried this. It's acceptable with low and extremely acceptable outcome. I'm sure the power users will chime in and correct me even further. As I said, this is one area I am not schooled in and am just trying things at this point.

For re-amping:

After you create the DI as I mentioned above, you would make the following changes IF you were going to just take the interface out and run it back into your AX8 input.

Output of interface (I used out 3 since 1&2 are being used for something else.) to the input of the AX8.

The only change I made for the re-amp was to change the output 1 nominal level on the "levels" page to -10 so that the noise would be reduced. There is an ever so slight little crackle, but I've re-amped in other situations with noise WAY worse than this, and we dealt with it. The noise here can be removed via the gate to where it is 0. My AX8 gate is at -80. Bringing it to -74 totally removed the light noise/crackle. There are no audible artifacts or excessive distortion present on dirty tones.

Clean tones are perfect and though this may not be the correct way to do this, I'd have no problems using this method for a client. Worst case scenario, I might have to slip edit in a few places or use one of my noise plugins. But I'm not hearing any noise with the gate at -74 for dirty patches, and even at -80, it was tolerable and could only be heard in empty spaces in the audio. But when I say the noise/crackle is light, I'm dead serious. It's not enough for me to worry about if the AX8 was all I had.

Not sure any of this will help, but it's definitely acceptable on my end until someone more knowledgeable than me can point you in the right direction. I'd call this method "re-amping made possible with a little "r". :)
 
P.S. One other thing, I used no FX loop blocks or other stuff within my patches. All I did was run the DI right into my patch as it was...no other bells and whistles. I'm actually quite pleased at the results.
 
Hey Danny, what a nice post - so much information to take in, and hence the late answer :)

But I guess my conclusion is perfectly in line with yours; with a little care, DI recording and re-amping through AX8 with no extra gear (other than an audio interface) works perfectly and yields a high quality result.
 
Hey Danny, what a nice post - so much information to take in, and hence the late answer :)

But I guess my conclusion is perfectly in line with yours; with a little care, DI recording and re-amping through AX8 with no extra gear (other than an audio interface) works perfectly and yields a high quality result.

Thanks! Yeah, we definitely share the same view on this. The biggest issue with the AX8 is creating a DI with it. If you use a direct box or spdif etc, it's perfect for the reamping process.

If you use the AX8 itself like I did in my example, you pick up a few little artifacts but nothing that is a show stopper. My recommendation is a good direct box if the interface being used doesn't support spdif.

That other thread I was involved in that you referenced before....I learned a lot from that as well as what Cliff offered about DI signals. It's amazing how even a fairly clean signal will STILL have some noise when fed through the axe or even a VST.

As soon as you do it correctly via digital or with a good direct box, you improve reamping tenfold. I sat there attempting to create clean, dry signal feeds at -12dB for about an hour.

Every one of them looked good and sounds good on their own. I even would compare them to a SPDIF signal, and listening to them, I could not hear an audible difference. But when you reamped with my home made DI vs a real one....this is where the huge differences came in for me.

In a nutshell, as soon as you do a DI the right way, it's so great you don't even know you're using it. It's literally like being plugged in and playing in real time with no noise or other artifacts.

But again, my last test the other day worked out quite well considering it was another home made DI that was not done correctly. If someone doesn't have a decent interface, ya gotta go direct box....which if we look at how cheap they are, it's a very small price to pay to be able to reamp more successfully with an AX8. :)
 
Figured I'd add a final comment to this thread to summarize for others.

Some in this thread are able to use their interface line out directly to AX8 instrument input for re-amping. That did not work for me, with my Steinberg UR44 interface. I was experiencing excessive noise. In the end, the solution for me was a re-amp box. Those can be quite expensive, however I found a highly recommended one for around 60 USD that did the trick (Orchid Electronics Amp Interface). It came with a nice volume-knob that, together with the input volume indicator on the AX8, made it very easy to match the level of my guitar directly into the AX8.

Tracking direct signal: Guitar > Interface Hi-Z input > DAW
Re-amping : DAW > Interface Line Out > Re-amp box > AX8 > Interface Line input > DAW

Hope that's useful to someone sometime :)

Cheers, and thanks to everyone who contributed in the thread.
 
Thanks for the feedback. I wonder why you get such excessive noise without the reamp box? maybe, it is just the fact that you lack the analog volume control, so you have to heavily reduce the level in the digital domain.

Great you got it sorted. To be honest, I was almost disappointed that my reamping was flawless without a box, because, I really wanted to get the Orchid one :p
 
Line level out to line level in is as close as you can get to a unity gain transfer, no color added, minimum distortion in the analog world. You will not get as good a result by sending line level back to the AX8 instrument level input. A conversion box (line to instrument) is going to sound somewhat different than the impedance loading of a guitar pickup circuit. The AX8 is not an amp, so you don't need to reamp through the guitar input to get to the preamp as you would with a real amp. Go in the FX return (In 2), set to left only in I/O Audio (its even designed for it).

Record the initial DI from the FXLoop Out 2 (echo Input) and into Line In on the Interface to record into the DAW. (Meanwhile, the full AX8 amp & effects is being recorded via Outs 1 or SPDIF).

For re-amping: Return the DI track from the DAW out to the Interface's Line level balanced out and into AX8 Line level balanced In 2 FX return. By adjusting (attenuating in this case) the return level from the DAW send until the gain matches (hitting the amp model input with the same level).

Sending the DI through the same effects chain in the AX8 and re-recording it via SPDIF sounds identical to the original AX8 amp and effects recorded at the same time as the original DI (I can't tell them apart. Also: wave forms look the same). No discernible added noise (the RF hum of the ungated silent guitar would mask what there might be). I've also recorded DI via the UAD Apollo, a Focusrite ISA1, and an ART DMPA Tube pre: they all sound different from each other and the DI from the AX8, in part because the pickup loading and therefore frequency response is different into other units, or when guitar is split to two inputs. So, you won't be able to exactly replicate the original tracking through the AX8 when you reamp unless you used the AX8 for the DI (if you just wanted to change up one phrase for example). I've attached a couple of screen captures from AX8edit showing the routing for DI and for the reamp.
Twinv2-Reamp.jpg
 
Last edited:
Line level out to line level in is as close as you can get to a unity gain transfer, no color added, minimum distortion in the analog world. You will not get as good a result by sending line level back to the AX8 instrument level input. A conversion box (line to instrument) is going to sound somewhat different than the impedance loading of a guitar pickup circuit. The AX8 is not an amp, so you don't need to reamp through the guitar input to get to the preamp as you would with a real amp. Go in the FX return (In 2), set to left only in I/O Audio (its even designed for it).

Record the initial DI from the FXLoop Out 2 (echo Input) and into Line In on the Interface to record into the DAW. (Meanwhile, the full AX8 amp & effects is being recorded via Outs 1 or SPDIF).

For re-amping: Return the DI track from the DAW out to the Interface's Line level balanced out and into AX8 Line level balanced In 2 FX return. By adjusting (attenuating in this case) the return level from the DAW send until the gain matches (hitting the amp model input with the same level).

Sending the DI through the same effects chain in the AX8 and re-recording it via SPDIF sounds identical to the original AX8 amp and effects recorded at the same time as the original DI (I can't tell them apart. Also: wave forms look the same). No discernible added noise (the RF hum of the ungated silent guitar would mask what there might be). I've also recorded DI via the UAD Apollo, a Focusrite ISA1, and an ART DMPA Tube pre: they all sound different from each other and the DI from the AX8, in part because the pickup loading and therefore frequency response is different into other units, or when guitar is split to two inputs. So, you won't be able to exactly replicate the original tracking through the AX8 when you reamp unless you used the AX8 for the DI (if you just wanted to change up one phrase for example). I've attached a couple of screen captures from AX8edit showing the routing for DI and for the reamp. Note: the FXLoop position (in the reamping image) doesn't have to be in the leftmost slot, nor need to have a shunt to the left side. That signal is coming from the FX return, not guitar in.

View attachment 36445


Maybe you should read post #5 ;)
 
Good question. Seems line out > re amp box > regular input would be the way to go then? Anyone tried that? Don't have re amp box.

Yes, get a reamp-box. I have found this to be the solution as well. Going through the IN2-Fxloop created a whole lot of noise.
 
Maybe you should read post #5 ;)
I did. Also the one about poor noise floor. Maybe folks are cranking the gain on the DI - I don't know. The recorded DI signal through the AX8 into the DAW is under -24db. As far as "shouldn't be using the FX loop for reamping." : well, I tell myself it must be a misunderstanding and happily continue to successfully reamp.

All I can say is I've done it dozens of times now, (as have clients): those of you that actually try it will see it works as I described. True, I have a great interface in the Apollo: but as this is line level with minimal gain adjustment: I would think any decent pro or prosumer device should be able to do this reasonably well.
 
Last edited:
I did. Also the one about poor noise floor. Maybe folks are cranking the gain on the DI - I don't know. The recorded DI signal through the AX8 into the DAW is under -24db.

All I can say is I've done it dozens of times now, (as have clients): those of you that actually try it: you'll see it works fine. its easy. True, I have a great interface in the Apollo: but as this is all line level transfer, and audio 101 gain staging: any decent prosumer device should be able to do at least reasonably well.

I'm sure it works but it wasn't "what the ax8 was designed for". I've reamped literally dozens of times aswell (for my clients) out of line 3/4 of my Apollo into input one and back into line 1 (via reamp box of course) flawlessly with zero issues just as I would any one of my real amps and this is my preferred method. Cheers
 
Last edited:
Okay: so here's a quick, but more or less empirical test. A DI was recorded from the AX8 output 2 (no FXLoop, echo input). I set input signal level to around -24db in Logic (Out 2 on the AX8 set about 2/3 up). The SPDIF original recording was done at the same time (Full output on Out 1). The signal was re-amped with a 0db send from Logic (Apollo Send +4db / AX8 return +4db - return mode left only). A second re-amp was then made at -1.6db (metered difference between original SPDIF and 0db reamp SPDIF).

When playback levels are matched: The three SPDIF amp & effects tracks all sound the same to me: But: as you can see from the wave forms: on the 0 db re-amp the dynamics are somewhat different - and are slightly different on the -1.6db version from the original. The signal is hitting the front end of the amp model slightly harder: so the amp sound is subtly different - even if I can't hear a difference (via JBL LSR308). Maybe a nudge of -0.1db (for a -1.7db send) would get even closer: but this is one unit at the finest level of resolution in Logic. The display of amp modeling shows its dynamically responding even to .1db changes in the input. (limit of resolution and repeatablity) (Also, I question how accurately Logic's wave form display can represent this information?). I'm not going to run a null test. With accurate gain level matching the AX8 does an impressive job of re-amping to closely match the original: not that one may need this accuracy: but its evidence of extraordinarily high audio fidelity.

I can't audibly detect additional noise (by turning up the monitors) and it doesn't show up on Logic's meters which go to -60db.(I rolled guitar volume knob to zero at end -so any noise would be circuit only) The guitar noise itself is -54db when I'm lucky, more usually -48db. I've also reamped DI tracks from other sources and they sound better with the FAS modeling (well, duh) than whatever plugin I used in the past.

So, make it easy on yourselves. Re-amp with the AX8 and some quality cables. It works great. Really.

DI-&-Orig.jpg
 
Last edited:
Just curious, what happens when you run the DI you created from the AX8 output 2 into a high gain sound?

You should definitely pick up a little more noise than the SPDIF signals. Nothing that the gate can't clear up used lightly, but that's usually where you pick up more noise.

Light to mildly dirty won't show as much noise as a full blown, high gain sound. Especially in the beginnings and tail ends. They usually need to be slip edited tighter than usual or gated as I'm sure you know. Or are you using high gain sounds already and I'm totally misunderstanding?
 
Did a quick test: Cranked the Euro Uber model for a focused, saturated tone with inherent pre and power amp distortion and compression for a hot signal -6db at the DAW output. Yes, noise is evident in quieter passages, original SPDIF or re-amped. I take it as normal. What I think is going on: The guitar noise level is also being amplified 30 ish db, but is not compressed nearly as much by the amp, since its mostly below the threshold of saturation and compression the rest of the signal is being crushed with. The noise of the guitar volume on 10, not being played is now at -18db = Loud (gate off) It was about -45db in the Fender Twin example. The noise of the circuit (guitar volume = 0) originally below -60db is now boosted to -35 db, which is masked by the noise contribution of the guitar when its on. DI reamped or original SPDIF sound the same. So, as I see it, the degree to which a cranked amp (or model) acts as a compressor bringing up the inherent guitar signal noise floor is the significant factor, the process of send and return for the DI is below that threshold.

Note: the UAD Apollo is an outstanding interface and may be a significant factor in my results. Its also a tribute to the quality of the AX8 AD/DA converters.
 
Last edited:
Did a quick test: Cranked the Euro Uber model for a focused, saturated tone with inherent pre and power amp distortion and compression for a hot signal -6db at the DAW output. Yes, noise is evident in quieter passages, original SPDIF or re-amped. I take it as normal. What I think is going on: The guitar noise level is also being amplified 30 ish db, but is not compressed nearly as much by the amp, since its mostly below the threshold of saturation and compression the rest of the signal is being crushed with. The noise of the guitar volume on 10, not being played is now at -18db = Loud (gate off) It was about -45db in the Fender Twin example. The noise of the circuit (guitar volume = 0) originally below -60db is now boosted to -35 db, which is masked by the noise contribution of the guitar when its on. DI reamped or original SPDIF sound the same. So, as I see it, the degree to which a cranked amp (or model) acts as a compressor bringing up the inherent guitar signal noise floor is the significant factor, the process of send and return for the DI is below that threshold.

Hmm...thanks for doing that. I seem to be getting different results on my end. If I can get a few minutes, maybe I'll create a video just to show you what *I'm* getting over here. Not that it will matter anyway...lol...but at least we can see if one of us may be doing something wrong or learn from the other. :D

On my end, using my XL+ for the DI internally (XL+ interface) is the same as my SPDIF sound. I can't even recall hearing any noise with these. Super light if any. But the DI from the AX8 definitely picks up a little noise. Again though, it's nothing super bad but a considerable amount more compared to the other two methods. I'll try to post something up with a high gain sound just to see if you and I are getting similar results.

That said, I still have no issues reamping into the AX8. We can adjust the pad and remove even more noise because we're recording digitally and have plenty of room. Also, with a lower pad, though the sound on the AX8 is lower, if we were to normalize (ugh, that's a bad word for me in the recording world lol) or better yet, input trim for gain staging, when the sound gets louder, you don't pick up much more noise. When the pad is hotter, you pick up more noise and it's destructively there. So we're still in really decent shape.

Not that this is important....but....I'm a control freak anyway and always remove dead spots in all my audio. So even if there were little sections of noise, I'd just slip edit them. But the gate does a nice job too the way I have mine set. It takes the noise out perfectly without choking off the sound. A lot of guys like to leave their volume knobs open and gate so there is 0 noise. I come from the days of "you better learn how to shut down that knob, tap dance on 3 drive pedals and kick on the gate all in one movement!" LOL! So I'm a volume knob whore. If I'm not playing, it's not open. I'm so fast turning it up and turning it down that it's become part of my playing. This allows me to use less gate as when I play, the sound of my guitar will mask any noise and my gate never cuts off my tones. :)

I'll see if I can find some time and do something with this. Thanks again for looking into it.
 
Back
Top Bottom