Successful re-amp without onboard USB audio (hybird analog/digital method)

shasha

Fractal Fanatic
EDIT: You can skip to the final results here rather than read this entire post of fail. :mrgreen

There's some decent info in here if you really like messing around (i.e., wasting time), but after a few days messing with this it's just not the best working method.


ORIGINAL POST:
OK, this one has been fighting me for a while now. Cliff has said that he would look into a future update in the firmware that would alleviate the small issues here and there with performing a re-amp without the onboard audio interface. I'll update this thread when and if he is able to do it (I would probably bet on that being likely knowing the big brain on that guy). Basically it will let us go 100% analog I/O for recording both wet and dry which of course means that we can use any sample rate in our projects and audio interface that we choose. For me that is an ideal situation for a lot of reasons, but mainly monitoring purposes. I'm really crossing my fingers for it to happen because it will just put the finishing touches on this thing.

But until that day here is what I've come up with for now. It is the most streamline method I could find to perform what I call a hybrid re-amp.

The stuff I'm using is an AxeFXII (duhh), a Focusrite Saffire Pro24DSP and StudioOne as my DAW.

Physical connections are as follows:
  • Guitar --> AxeFXII FRONT INPUT (sounds better)
  • AxeFXII OUTPUT 1/2 --> Saffire INPUT 3/4 (because they are the first non-mic preamp inputs and on the back)
  • AxeFXII SPDIF OUTPUT --> Saffire SPDIF INPUT
  • Saffire OUTPUT 5 (monitoring) --> AxeFXII REAR INPUT (L)
DAW configuration:
  • 48kHz 24-bit project (we're stuck with this for now)
  • Wet tracks are Saffire IN 3/4 (stereo) / feeding main mixer monitor
  • Dry tracks are Saffire SPDIF IN 1 (mono) / feeding a 'submix' to the Saffire 5 OUTPUT
  • Re-Amp tracks are Saffire IN 3/4 (stereo) / feeding main mixer monitor
Audio Interface configuration:
  • Dry tracks feed the main mixer output and are monitored via my software mixer for the Saffire. The only thing I have feeding my headphones/monitors is the DAW output which is the main mixer.
  • Wet tracks are routed to a sub mix or separate mix bus. In my case it's the Saffire OUTPUT 5. I can monitor this if I want, but I obviously choose not to except for verification purposes. It is how we feed the AxeFXII for the re-amp.



The only drawback here so far is that the AxeFXII HAS TO BE THE MASTER CLOCK SOURCE. I have made up so many damn cables this last week and wired this crap up so many different ways and tried everything....there are issues with every config I came up with. 90% of those issues came down to clock sync issues which will cause dropout and jitter. The real problem is that the AxeFXII will not run as slave and sync properly except for when the input is selected as AES input...then it locks right up and runs like a champ. Kind of hard to record a guitar into the AES input.


OK, so the basic operation is:

First take (initial wet and dry tracking)
  • AxeFXII I/O menu --> AUDIO tab --> SPDIF/AES SELECT -->SPDIF
  • AxeFXII I/O menu --> AUDIO tab --> USB/DIGI OUT SOURCE --> INPUT (this is what gives us our dry track)
  • AxeFXII I/O menu --> AUDIO tab --> INPUT 1 LEFT SELECT --> FRONT
  • Arm the first Wet and the Dry track for record and just wail away
Second take (re-amping)
  • AxeFXII I/O menu --> AUDIO tab --> INPUT 1 LEFT SELECT --> REAR
  • Arm the re-amp track, mute the first recorded wet track, go back in the timeline to the beginning of the dry take and hit record

The basic explanation of what's going on here is that we're recording ALL the wet tracks via analog outputs from the AxeFXII. We don't have to worry about clock issues anytime we're recording analog. The dry signal is only available via digital outputs right now so we are taking it off the SPDIF output. This is why we have to slave our audio interface to it. Then when we re-amp we are sending an analog output from the DAW to the rear inputs of the AxeFXII. Because it was recorded via the front input it has whatever special sauce Cliff put into that input; I just know that it sounds better and I am pretty sure that it's baked into the dry recording this way.

Advantages are that we don't have clock issues (and there are potential for a lot of them either in the recording or while monitoring which will confuse things because you don't know if its in the recording or not). The only thing that we have to change in the AxeFXII's menus are switching between front and rear input for wet and dry tracking. Monitoring works how it's supposed to from within the DAW; this was a major PITA going other routes. The process is smooth, no swapping cables and crap and once once you get it set up and working it's logical.

Disadvantage...well you have to remember to set your audio interface as the master clock after you are finished recording and editing or else it won't work and of course we are still tied to 48kHz projects for the time being.

Now you could go ahead and send the dry track back to the AxeFXII via SPDIF and it will work (just change the MAIN INPUT SOURCE to AES/SPDIF)....but you will also have to set the audio interface as the master clock at this point. Its an extra step and most DAW's will not release the driver (unless you set it in the app's advance settings). This way we don't have to deal with changing clock sources back and forth and the other thing is that if you set up your DAW to release the driver and you click anywhere outside the app's window it will stop recording or playing. Me no likey that way.

I was able to record for quite a while tonight (just horrible noodling) with zero dropout or glitches. Levels were really good throughout the entire process and the main thing is that everything just worked well.
 
Last edited:
The only way I could see a way to go all Analog for a reamp is if Cliff had an option for Output 1 to be the main "wet" signal and Output 2 to be the "dry" signal. Then set the Axe-Fx to use the Rear Inputs on the back as the input back from your audio interface of choice when doing the reamp. If the hardware/routing on the Axe-FX II is even connected in a way to allow this. I use the Output 1 to connect to my 16 channel firewire board. So currently I do not have the ability to reamp.
 
This might be exactly what I've been looking for... reamping with the Axe FX has been such a pain. most methods are either extremely limited or needs too much fumbling around to really be useful to me.
 
The only way I could see a way to go all Analog for a reamp is if Cliff had an option for Output 1 to be the main "wet" signal and Output 2 to be the "dry" signal. Then set the Axe-Fx to use the Rear Inputs on the back as the input back from your audio interface of choice when doing the reamp. If the hardware/routing on the Axe-FX II is even connected in a way to allow this. I use the Output 1 to connect to my 16 channel firewire board. So currently I do not have the ability to reamp.
That's exactly what Cliff is looking into. I asked for it in the Wish List forum the other day after M@ had brought up the idea in another thread. Basically having the ability to choose from a list of sources for output 2 rather than just have the option to mirror output 1.

If this happens it'll be awesome for more than a few of us IMHO.

The one other thing that I didn't bring up in here and does work accroding to others is to add an FX loop to each patch right after the input block. This will spit the input signal to output 2 and then you should be able to re-amp in a similar fashion. The downside is that every single patch has to have an FX loop in there. But if you were up to it you could probably do it.
 
The one other thing that I didn't bring up in here and does work accroding to others is to add an FX loop to each patch right after the input block. This will spit the input signal to output 2 and then you should be able to re-amp in a similar fashion. The downside is that every single patch has to have an FX loop in there. But if you were up to it you could probably do it.

That is how I re-amp using the Ultra, and it's easy, seems much easier than the method you're doing with your setup now. Either way, if cliff adds more options to the II in regards to re-amping that would be another huge bonus IMHO.
 
That is how I re-amp using the Ultra, and it's easy, seems much easier than the method you're doing with your setup now. Either way, if cliff adds more options to the II in regards to re-amping that would be another huge bonus IMHO.
Well I don't know if it's much easier....I know that my description is much wordier, but the basic worklow is going to be the same.

The big advantage of course in using an FX loop would be you can stay analog the whole way which means choosing your own sample rate and not having to set a master clock source.
 
UPDATE.

I've been messing with this all day long uninterupted (for a change). As of right now I cannot get this to work better than using SPDIF for everything. I tried using an FX loop going into another input on the audio interface and its just too noisey. Its not horribly bad, but its not a balanced output and while the humbuster output helps considerably it's got too much noise on it. Especially compared to the SPDIF which is just pristine (when you record it with the AxeFXII as the master).

But even worse (and I don't know how I missed this) is that the rear inputs on the AxeFXII are just noisey. I made 3 different cables, all with good quality shielded cable and no matter what there is an unacceptable amount of hum for something that is supposed to be a balanced input. I even tried making an unbalanced nstrument cable and using it and it was almost better, but there is a different kind of noise that it picks up. I tried making cables with grounds lifted and attached....not a bit of difference.

I'm thinking about snagging an SRC from somewhere for a few days just to mess with it.

This whole thing really drives home the importance of having a word clock. If it would somehow maintain a reference when it's not using the digital input as the audio source then that would solve all of this. Having the option of the output 2 source select isn't going to make any of this go away unfortunately.
 
I made the analog method work by doing fx block out and balanced analog back into rear input one for reamp. My interface is a FF800. I'm not having any noise issues that way.

I am getting noise when using the fx block in the signal chain and running the return back into the fx block. This method requires a slightly different switching technique but none the less, I'm getting lots of hiss(not hum) on the return to the fx block. It definitely sounds like bad signal to noise ratio. I'm not getting noise with the send.

Can't figure that one out, so for now I just use the fx block send for dry and input 1 front and rear in the i/o menu for play/reamp.


You've given this a Herculean effort for sure Shasha...8)
 
That's exactly what Cliff is looking into. I asked for it in the Wish List forum the other day after M@ had brought up the idea in another thread. Basically having the ability to choose from a list of sources for output 2 rather than just have the option to mirror output 1.

Having the option to set O2 as a dedicated dry out would be fantastic.
 
UPDATE.

I've been messing with this all day long uninterupted (for a change). As of right now I cannot get this to work better than using SPDIF for everything. I tried using an FX loop going into another input on the audio interface and its just too noisey. Its not horribly bad, but its not a balanced output and while the humbuster output helps considerably it's got too much noise on it. Especially compared to the SPDIF which is just pristine (when you record it with the AxeFXII as the master).

But even worse (and I don't know how I missed this) is that the rear inputs on the AxeFXII are just noisey. I made 3 different cables, all with good quality shielded cable and no matter what there is an unacceptable amount of hum for something that is supposed to be a balanced input. I even tried making an unbalanced nstrument cable and using it and it was almost better, but there is a different kind of noise that it picks up. I tried making cables with grounds lifted and attached....not a bit of difference.

I'm thinking about snagging an SRC from somewhere for a few days just to mess with it.

This whole thing really drives home the importance of having a word clock. If it would somehow maintain a reference when it's not using the digital input as the audio source then that would solve all of this. Having the option of the output 2 source select isn't going to make any of this go away unfortunately.

just a guess - may not be a good one:
I'm wondering if the hum on the rear input is due to an impedence mismatch..

when I eventually get my Axe2 I'll try this method - with the FX send / return blocks
and if I encounter noise I'll try it with my Redeye and see if that improves things..

I know that when I first started re-amping with my VG-99 that it was silent with a guitar jacked in, but nasty when the input was connected to an aux send from my mixer or AI.. when I threw a Redeye in between the mixer and the VG-99 things improved greatly..

when I eventually get to trying this out I'll let y'all know how it goes...
 
For the slave issue, you don't have to run an audio input into AES. Just clock.

For example, I've got an Apogee clock that outputs on AES. The AxeFx II will sync to that as a slave.
 
For the slave issue, you don't have to run an audio input into AES. Just clock.

For example, I've got an Apogee clock that outputs on AES. The AxeFx II will sync to that as a slave.
But will it remain locked while using the analog input on the AxeFXII?
 
Got some progress made. I mean part of progress is going through failure and getting to a point where you just realize that the best solution isn't as sleek or streamline as you would like, but nevertheless it is the best.

You can read post #7 to hear all about the issues with going the analog route FOR ME. Others may have better results, there is just too much hiss going analog even using humbuster cables and balanced audio wherever I could. I even borrowed a SRC today and messed with that for a bit. It does work and it actually worked very well, but unless I get two of them which I am not particularly interested in doing it's going to add more jumping through hoops than it's worth. The SRC in question is a cheap old Behringer SRC2496. I was able to go from 48kHz to 96kHz and back (with cable swapping and a bunch of button pushing because I only had one and didn't feel like making new AES cables right now) ...anyway the thing works surprisingly well. But the clock issues remain, you have to have the source as the master in every instance so if you are recording from the AxeFXII the audio interface is the slave and as soon as you go to re-amp you have to release the audio driver and then set the interface as the master.

Which brings me to the point that I am at right now which is something that works and works extremely well. SPDIF at 48kHz and just learn to like it. Now I'm just not a huge fan of SPDIF and I've never had this many issues with getting a good clean balanced audio signal, but I'm also used to messing with gear that is a lot higher quality (talking about my sub $500 audio interface here). Bottom line is that there may just not be a high enough SNR even though the AxeFXII's balanced outputs record wonderfully. But for the dry signal it is such a low level signal to begin with that it may just get masked along the cable.

OK, getting back to the point here, after making cables all weekend, routing signals all over the damn place, creating new presets and templates for my DAW and the audio interface's mixer/monitor software and then doing A/B/C/D/E/...../Z comparisons NOTHING can touch the quality of just using the damn SPDIF. The levels are always dead on and what cemented it firmly in the winner's circle was recording with one preset and then re-amping back into the same preset. If there is a difference at all I cannot for the life of me find it. I solo'd each track and toggled and other than an occasional glitch from the DAW while switching tracks there is absolutely no difference. Same amount of barely audible hiss (the good, real pickup/amp hiss, not crappy EMI induced hiss) and the levels are exactly the same.

Now if Cliff can somehow force the AxeFXII to always sync to the SPDIF input regardless of whether the input source is analog or digital then it'll be absolutely perfect, but even as it is right now just having to switch between slave and master (internal and external) for record and re-amp is a lot easier than all the other crap I tried and it's just about as perfect sounding as you can get. Once I got the routing set up in the DAW and audio interface it's a pretty elegant workflow.

Now there is one other possible solution to this which is a Digigram VX882 sound card. I remember messing with one a few years ago and its got 4 sets of stereo AES inputs, master clock in/out and 8 mono analog ins/outs (which share the digital path). I was talking about this thing a few weeks ago to someone and decided to do some reading up on it and it turns out that you can set each channel to be a slave or master independently. So in theory I could send the AxeFXII AES out to one AES input, set that input channel as slave and then route the playback audio from the DAW to a seperate channel, set that as a master clock source and feed the AES input of the AxeFXII. I'm just not sure if it will prevent dropouts completely. And of course the biggest issue with this plan is that I'd have to locate one of these suckers and last time I looked they're in the $2000 range. Oh but each channel has it's own SRC as well...which means any sample rate and bit depth for a project.
 
Last edited:
Its a bummer that your SNR is so bad.

I know I've said this before, but I'm using the FF800 and have no noise problems when going fx block send and rear input 1 for reamp return. I do get really bad SNR on the fx block return, go figure.


That being said, I'm making do with setting the levels manually which is a disadvantage as opposed to the the digital none the less.
This is why I've been in wonderment about the unit design. Very little flexibility, with the user struggling to make it work. You're a glaring example of this.

I'm not bitter or anything, so no need to defend the unit. I love my Axe. Its just plain odd to have all of the amazing engineering in this box coupled with the i/o nightmares...
I only have the consumer point of view, which is that every other digital audio capable device I know has more flexibility in routing/clock choices regardless of price range. So I wonder, "Why so complicated with this unit?"

I hope you can get comfy with your situation shasha, I feel for ya man...:(
 
Last edited:
I'll live. ;)

I do need to add one more thing here. I went into the SRC and took the analog out and fed it into my audio interface into both the straight balanced input and the mic/balanced input and got the same hiss. I know for a fact that the signal going into the SRC is extremely clean from the AxeFXII because that's what I'm comparing all of this too. So basically its not the AxeFXII that is creating the hiss.

I think that it is just the fact that the signal level is so low for the dry feed that its getting overwhelmed along the cable. Its the definition of SNR....small signal, too much noise.

I'm going to make up some AES cables tonight and try one more approach to this just for the sake of experimentation. I want to see if the SRC can convert to and from 96kHz/48kHz both ways without swapping cables. I should be able to do it with one SRC because it's got AES and SPDIF selectable sources so it's almost like a 2 way box (in theory). I know that the quality is good enough, just don't know if it's practical.

It won't change the situation with the master/slave clock thing, but it will get around the sample rate being fixed all the time.
 
Last edited:
I'm experiencing the exact same problem as described by Shasha, too much hiss when doing analog reamping.

My problem now is the dry tracks recorded (via effects send from output 2) seem to have a slight noise in it, however it is not visible in the DAW/Sappire Mixer meters. After adding some volume to the dry outs (by increasing preamp gain/ increase output levels in the axefx), the hiss is too loud whenever I'm trying to reamp a couple of high gain stuff. Using the noise gate can't really help much as I'm trying to do really tight palm muted stuff and the hiss will still be very evident in the recording.

The worst part is when I'm doing slow let-ring palm muted chugs, that's where the hiss really starts to become even louder than the original distorted guitar sound (as the distorted guitars starts to fade out)

I've tried varying output levels on my audio interface (Sapphire Pro40) by not adding any gain on the preamps but the hiss is still there, although much softer, but it is not really ideal as the dry signal would then be too soft.

That being said, the dry tracks recorded using the AXEFX2's USB works perfectly, no noise at all, but I don't intend to use the Axe2 to monitor my project's playback as it is rather unreliable.

This thread has been a great help for me, I might get my hands on some SPDIF cables if I have the chance, but if anyone has any other ways to record really clean dry tracks, I'm all ears :D
 
The SPDIF sounds as clean as the USB. I'm absolutely thrilled with the quality combined with being able to use my audio interface because of the ability to monitor how I like and the stability of the drivers.

I did find something really interesting called a Mutec MC-6. It can do bidirectional SRC and conversion and supposedly it will maintain clock stability regardless of source. In theory it would eliminate the need to switch the clock source and I'd be able to record at any sample rate. Unfortunately it's about $700.
 
Back
Top Bottom