USB Direct vs Audio Interface

[Bodde]

with the TC I have cable 4 in's and 4 out's and 4 firewire in's and out's

guitar -> TC in1 -> TC firewire out -> DAW: this is the dry
TC has a mixer inside it which 'sends' in1 to jack out3: cabled [via a Redeye] to the Axe INSTR in
Axe output1 cabled to the TC in's 3 and 4: so I can hear the Axe as per normal operation with amp / cab / fx
TC out 1 and 2 to monitors
DAW master outs to TC firewire 1 and 2 so I can hear the mix
all that records the dry guitar..

I have another config in the TC that takes the dry guitar from the DAW firewire 3, routes it to TC out3 which is cabled to the Axe INSTR in
the Axe plays back via cables into the TC ins 3 and 4 just as before, which are then routed via TC firewire to the DAW to record the wet
so that is the reamping config..

it works fine.. I just don't do this anymore.. in fact, I don't use the TC at all now..

ok, in my question I was asking about the simplest setup. No reamping just straight simple recording.
The question is: Can you hear any difference in sound (or noise ratio) when recording through you soundcard (Axe fx stereo output cables to soundcard and then to DAW) or recording through your Axe fx with USB (so no external soundcard)? No reamping.

 

[Boerboel666]

what is your setup?

At the mo my set up is axe fx via usb -> Mac book pro retina (Logic x) -> apogee ensemble -> Monitors.
Any tips would be appreciated.

 

[Boerboel666]

I did a video on this myself
You need an additional sound card, i.e. the one that comes with the mac book or whatever.

Thanks I'll check out the video.

 

[clarky]

At the mo my set up is axe fx via usb -> Mac book pro retina (Logic x) -> apogee ensemble -> Monitors.
Any tips would be appreciated.

try making an aggregate interface with both the Axe and the Apogee
use the Axe to get to / from the DAW [to record and reamp]
use the apogee to monitor the mix and monitor the Axe [via the Axe's output jacks]

 

[clarky]

ok, in my question I was asking about the simplest setup. No reamping just straight simple recording.
The question is: Can you hear any difference in sound (or noise ratio) when recording through you soundcard (Axe fx stereo output cables to soundcard and then to DAW) or recording through your Axe fx with USB (so no external soundcard)? No reamping.

ok.. in my own setup [just recording] I didn't really notice any noise that stood out as being bad / unusually high
but then it was Axe -cabled-> Mackie -firewire-> DAW
the Mackie is pretty good noise wise
so.. recordings via USB vs cables didn't seem to show any noticeable differences with respect to audio quality or signal to noise ratio..

but then.. if the signal path from the Axe to the DAW is good in both scenarios you really should expect them to be pretty much the same..
if there were differences my ear didn't detect anything..
Note: my ears have been thrashed half to death as a result of decades stood within the Marshall's blast radius..
so maybe I'm not the best judge.. but everything seemed fine to me..

it's reamping the dry where I noticed the improvement..
and even with my gig thrashed ears.. the difference was undeniable..

 

[plyall]

Clarky/all -

Doing a little research and testing I found that I am able to use an aggregate device that includes both of my RME units and the AXE via USB (actually, knew this somewhat earlier), but ALSO that I can use the Audio/MIDI setup tool in the Mac to resample the AXE-FX to 44.1k on the fly (see attached picture):



In this case, RME #2 is providing clock reference and the Mac is resampling the 48K -> 44.1K using the DRIFT check box. I don't know if this is also successful with larger differences such as 88.2K or 96K, but it does appear to remove the hard restriction that the Axe be recorded at 48K, at least on the Mac.

While doing this test, I did re-verify my level testing regarding USB vs. Analog. The analog was a hair hotter, but that's probably attributable to where I have the gain pots set on the RME. I will say that the USB processed output looked pretty impressive as it applies to being full scale output w/o clipping. The DI signal looked a bit anemic compared top the analog DI, but then again the gain pots on the RME are likely the difference. I guess that particular signal could always be normalized to get some of its juice back.

All in all, I'm pleasantly surprised with my revisit of the USB interface, along with the fact I don't have to be locked at 48K (on the Mac anyway). I will test the higher sample rates later to see if they can be handled as well.

If only the DI (DO really) wasn't so low...

 

[plyall]

I thought this might also be worth seeing - it's a picture of the analog DI (top) compared to the USB DI (bottom, in stereo) after it had been normalized by my DAW (Logic):

 

[BillyZeppa]

Interesting. Does ASIO4ALL allow you to connect the AXEII to a PC DAW via USB, while connecting another ASIO PC interface at the same time? Or is this strictly for apple... or?

 

[exoslime]

i use it the analog way on my 16channel mackie onyx 1620 (the old one) firewire interface, as i also need the remaining input channels for recording other stuff like vocals, etc.. and i dont like switching the audiointerfaces in my cubase projects as it would always totally change the predifined track input settings.

Also the mackie onyx is the perfect central station for me, where i can flexible put my outputs anywhere in my musicroom where i want.. headphones, Full Range PA (ideal when just jamming or fiddling around) or studiomoniros
but i was worried that maybe the additional D/A A/D Conversion results in a sound loss, so i recently did a comparision and made a test, recorded digitally via usb, kept the DI and then reamped through USB and recorded back analog and then flipped the phase.. perfect cancelation of the sound..
then i did it the opposite round, recorded everything analog, made a reamp via usb and recorded via usb and did the same again, perfect phase cancelation. so nothing to worry about me that one or another has a "better" sound.

so in my setup, both ways are good to go, but i guess if you going to have a cheap audiointerface with cheap converters and preamps, there might be a noticable differnce.

also i prefer analog outputs as i can use my complex routing setup
for each guitar track i record 3 tracks in Cubase
1x Stereo and 2 mono Tracks
(the guitar goes straight into Instrument Input, no DI Box, etc, as my opion is that the AXEFXII Instrument is TOP Notch and i defintly want to use this and not hook up any di-box or signal speaker in front of my axe to get a for example a nice DI Signal, as you would do with an analoge tube amp setup for example)

Axe FX Output 1 is the Stereo Signal, it includes the Guitar Amp inlcuding Reverb, Delays etc.

i use this signal always when i play guitar, as i like a wet sound with lots of reverb
but for recording this is often not ideal, so what i do is in my presets, that i split the signal and send it out of output 2

that gives me 1x Mono Track that i run out from the CAB Block, with no Reverb and Delays,
and another Mono Track that i use as DI

so Main Output 2 Left is my DI Signal, and Main Output 2 Right is my Mono Guitar Sound that i use in my cubase sessions, for like rhythmn guitars.
you can easily do this with an FX Block in your presets

and looks like this:

on Lead or Solo Takes i may use the Stereo Track, or mix it with the Mono track. also for cleans i consider to use the wet stereo track..

i would love to make an aggreate interface, using my axe FX with USB simultany with my Mackie Onyx.. but somehow that doesnt work on PC..

 

[dosilegecko]

No empirical evidence of noise (and I'm pretty picky). Also, more signal also means more dynamic range, and more 'bits'. Conversely this means less signal = less bits. For me I'd rather have the range.

The dynamic range comment would be correct as the hotter the signal, the more of the maximum range of the digital environment you are using. If you're in 24bit, you have 2^24 possible voltage levels. However with the axe fxs noise floor, a relatively small signal would be fine. I typically record at -12db with good results.

The number if bits comment is not correct. You always have 24 bits (or 8,16,32,etc) regardless of signal level. Many of them may be zeros, but there are still the same number of bits total to make up the word length of 24.

Sent from my SCH-I545 using Tapatalk

 

[Thomas Larsson]

I agree with Plyall.
You get less digital resolution (numbers to represent the "voltage-picture-of" the signal if you record 20dB lower .
Especially the low levels (ring out´s etc..) will benifit if you use all bit´s. It´s like pixels in the voltage domain. I don´t think we went to 24 bit´s of noise issues rather for sound quality reasons. If we hear it , I don´t know. But theoretically, fuller level recording -better copy.

the noise thing yes.. you'll get more noise using the jacks..

but more dynamic range?? more bits?? I don't think this is true at all..
if I recall correctly, whilst on the surface this appears to be a reasonable thought, in real terms I don't think this is actually true..
I think that [for clever reasons that I don't fully understand], the signal with the lower level should contain just as much dynamic detail as a larger one when it's in the digital domain..

try thinking about it this way.. [maybe this is not a great justification, but it works for me]..
we all agree that on the digital audio engineering smart-ass scale, Cliff is a hard 10..
so.. we'd also all agree that the decisions he makes regarding the Axe and the signal path are going to be right up there on the top floor of extreme cleverness..

now take a look in the Axe-II manual at the Input / Output block diagram [a few pages from the last]
you'll see that the USB output marked USB 2/3 is just after the Main Input Source switch
you'll also see that the USB input feeds the same switch..
also, the INSTR input also feeds the same switch..

this tells me the following:
- the signal from the guitar [after whatever smart stuff happens at the INSTR INPUT] has all the digital information that is needed for the Axe to do it's thing..
- the dry signal recorded from the Axe USB 2/3 has to be exactly the same as the original dry guitar signal because it's sent to the DAW from the same switch and therefore at the same point in the signal path..
- the recorded dry signal sent from the DAW to the Axe arrives at the same switch and enters the signal processing path in the same manner as if you jacked your guitar straight into the front panel INSTR IN socket
and if this USB input signal was different, playing the guitar straight into the Axe and playing back the recorded dry from the DAW would yield different results..
so surely if this was true it'd render the whole USB IN for recording the dry pointless…
could you really see someone like Cliff / FAS settling for that?? the whole FAS attention to detail thing is kinda nuts [luckily for us]

this all tells me that the USB recorded dry signal contains all of the frequency and dynamic information as if you jacked the guitar into the front panel…
if jacking your guitar into something else, boosting the signal [for greater dynamic range] and then pumping that into the Axe via USB was the way to go, I think we'd all be doing it.. imagine.. "make your Axe sound better and have better dynamic range.. jack into this gizmo first.. then USB to the Axe rather than use the INSTR IN.."
and clearly this makes no sense..

so.. should you boost your recorded dry in the DAW, the only thing you actually add is level..
and the dynamic relationships between the samples should therefore remain the same.. but louder..
so it seems to me like you're telling the Axe to read the same book [in this case, a sequence of digital samples]… but with a bigger font..
because you've not captured anything different.. you've just made it louder..

 

[Thomas Larsson]

Yes of course the bit-depth is allways 24 . But if we have 8 zeros (caused by low level) all the time it will behave as 16 bits.

And if I put it this way. If you record extremly low level so only the last bit moves (23 zeros) you will get a lousy sound quality even if the noise figures are messured low.

Agree ?

The dynamic range comment would be correct as the hotter the signal, the more of the maximum range of the digital environment you are using. If you're in 24bit, you have 2^24 possible voltage levels. However with the axe fxs noise floor, a relatively small signal would be fine. I typically record at -12db with good results.

The number if bits comment is not correct. You always have 24 bits (or 8,16,32,etc) regardless of signal level. Many of them may be zeros, but there are still the same number of bits total to make up the word length of 24.

Sent from my SCH-I545 using Tapatalk

 

[Thomas Larsson]

Sorry if I am stupid but what can you do with this agreggate-stuff that you can´t do with normal use ?

I use a RME fireface and the axe on USB. Then I choose (in logic) where to pickup or send signal.
Is there anything better than that we can do ?

I have found that my axe-2 works fine in my 44,1kHz songs by itself, really.

Clarky/all -

Doing a little research and testing I found that I am able to use an aggregate device that includes both of my RME units and the AXE via USB (actually, knew this somewhat earlier), but ALSO that I can use the Audio/MIDI setup tool in the Mac to resample the AXE-FX to 44.1k on the fly (see attached picture):

View attachment 20355

In this case, RME #2 is providing clock reference and the Mac is resampling the 48K -> 44.1K using the DRIFT check box. I don't know if this is also successful with larger differences such as 88.2K or 96K, but it does appear to remove the hard restriction that the Axe be recorded at 48K, at least on the Mac.

While doing this test, I did re-verify my level testing regarding USB vs. Analog. The analog was a hair hotter, but that's probably attributable to where I have the gain pots set on the RME. I will say that the USB processed output looked pretty impressive as it applies to being full scale output w/o clipping. The DI signal looked a bit anemic compared top the analog DI, but then again the gain pots on the RME are likely the difference. I guess that particular signal could always be normalized to get some of its juice back.

All in all, I'm pleasantly surprised with my revisit of the USB interface, along with the fact I don't have to be locked at 48K (on the Mac anyway). I will test the higher sample rates later to see if they can be handled as well.

If only the DI (DO really) wasn't so low...

 

[dosilegecko]

Yes of course the bit-depth is allways 24 . But if we have 8 zeros (caused by low level) all the time it will behave as 16 bits.

And if I put it this way. If you record extremly low level so only the last bit moves (23 zeros) you will get a lousy sound quality even if the noise figures are messured low.

Agree ?

Yes this makes sense, I wasnt trying to correct anyone, more just clear things up because when I read it at first it wasnt clear to me what he meant, thats all.

Sent from my SCH-I545 using Tapatalk

 

[Loquenau]

@Thomas Larsson: I'm a novice here.....I'm just using Audacity, and then regardless of interface (input source), I use the 'amplify' and 'limiting' algorithms to increase the recorded track(s) output. It seems to work just fine. From what I've read, others do similar with other DAWs and various plug-ins.

the beauty of 48k and above is that it stands up better to the processing that happens during mixing / producing
cos when you change eq, add fx, compress etc in the digital realm, the higher the sample rate the better..
it lessens / better hides the artefacts created by the act of manipulating the digital audio..

so the rule of thumb is to start at a high sample rate and remain there until the final act of creating the master that's bounced to CD

I think Cliff has talked about and said some of this. But.....who the fuck uses CDs anymore?? And why???

 

[bigswifty]

Glad to see this thread took off! It is humbling to know that I'm not the only human who is bogged down at times by frustrating audio troubleshooting..

Interesting. Does ASIO4ALL allow you to connect the AXEII to a PC DAW via USB, while connecting another ASIO PC interface at the same time? Or is this strictly for apple... or?

ASIO4ALL indeed allows you to do this! I have come to appreciate it 10 fold over the last couple of weeks. When I run ASIO4ALL as my main device, I can record in from both my Axe-FX (USB), MIDI Controller (USB direct to CPU) and UR22!
EDIT: And it seems to adjust the sample rates accordingly.. Though I have now set it to default at 48kbs.

 

[Warrior]

I'm running Reaper on a Mac-Mini:

Axe-FXII>Mac-Mini via USB>Reaper (48k)>Focusrite Saffire Pro 24 DSP>Monitors.

My Audio Devise is set to Axe-FX II as my "Input" and my Saffire Pro is set as my "Output" via aggregate device.

When I need to use a mic I change my input to Saffire Pro. I wouldn't be able to record my Axe-II at the same time via USB running like this but it hasn't been an issue for me in my workflow.

 

[BillyZeppa]

Glad to see this thread took off! It is humbling to know that I'm not the only human who is bogged down at times by frustrating audio troubleshooting..



ASIO4ALL indeed allows you to do this! I have come to appreciate it 10 fold over the last couple of weeks. When I run ASIO4ALL as my main device, I can record in from both my Axe-FX (USB), MIDI Controller (USB direct to CPU) and UR22!
EDIT: And it seems to adjust the sample rates accordingly.. Though I have now set it to default at 48kbs.

Sounds fantastic, dbrozz. I have been to the web site of what looks like the creator of AISO4ALL and downloaded the file. Is there a user guide available, or 'installation & use' readme? Is it fairly dummy proof? :emmersed:

 

[plyall]

Sorry if I am stupid but what can you do with this agreggate-stuff that you can´t do with normal use ?

I use a RME fireface and the axe on USB. Then I choose (in logic) where to pickup or send signal.
Is there anything better than that we can do ?

I have found that my axe-2 works fine in my 44,1kHz songs by itself, really.

Sure - not have to switch interfaces all of the time. I use 2 RME Fireface 800's and the Axe FX II as my audio devices, and the aggregate let's me treat them all as one atomic device. Most of my I/O is on the RME's, so being able to route audio to/from the RME's while using the Axe is exactly the use case for using an aggregate device.

Regarding use at 44.1 over USB, you really need rot have the DRIFT CORRECTION box checked to be assured that you're getting a good resample (unless Mavericks somehow does this under the covers now).

 

[bigswifty]

Sounds fantastic, dbrozz. I have been to the web site of what looks like the creator of AISO4ALL and downloaded the file. Is there a user guide available, or 'installation & use' readme? Is it fairly dummy proof? :emmersed:

Oh yea, BillyZeppa. ASIO4ALL is a pretty straightforward install. So much so I can't even recall the install process on about 3+ PC's.
Just boot it up, let it do its thing, let Reaper know it's in command and you'll be good to go! One of the best freeware installs available along side AdBlock for Google Chrome :eagerness: