A 4-cable-method comparison

[Orvillain]

So I did a comparison of the Axe FX III, Helix, and GigRig G3. I wanted to know which would provide the cleanest signal path when connected up in 4-cable-method. Now the GigRig G3 is an analog switcher, so there isn't any AD/DA conversion; not until you stick a bunch of digital pedals in the loops anyway.

So here's what I found in terms of noise floor.

Helix:

Axe FX III:

GigRig G3:

Ignore the peak max figures. We're looking at RMS momentary for our purposes.

Ranking is:
Axe FX 3 (not by much though!)
Gig Rig G3
Helix

But they're all WELL WITHIN tolerance.

-----------------

Now here are some sweeps of the entire setup. The amp is my Satriani JVM set to OD Red, because I want as many harmonics as possible in order to excite and potentially overdrive the I/O stages of each bit of kit.

Red is Helix.
Green is Axe3.
Orange is GigRig G3.

So I think we can more or less conclude that in the grand scheme of things, if you're running 4-cable-method with any of these devices.... you can't really go wrong.
The differences across the frequency range are infinitesimal.

The thing I do notice is that both the Axe3 and Helix have more oscillations in the frequency response from 3kHz upwards. Whereas the G3 seems to be flatter. But again, the difference is no bigger than 0.4 SPL in my test. I don't really know why this should be actually. If anyone has any ideas, I'm all ears.

Ultimately though, you're probably not going to be able to hear that much of difference.

 

[Orvillain]

Just an addendum regarding latency:
From previous measurements, we know that the Axe FX III has lower latency than the Helix, but not by huge amounts.

Helix would add 1.792ms with a blank preset. But this figure is for 2-cable-method based on previous tests. You'd probably need to double it to 3.584ms. So I think Helix would add this much latency in 4-cable-method to your rig, all of the time.

AxeFX III would add 1.25ms with a blank preset. But this figure is for 2-cable-method based on previous tests. You'd probably need to double it to 2.5ms. So I think Helix would add this much latency in 4-cable-method to your rig, all of the time.

The GigRig - assuming no pedals in the signal path at all - would not add any additional latency, because it is an analog switcher, and therefore no conversion stages are required. Once you start bringing pedals in and out, then the digital ones will add latency in some cases (unless they're analog dry thru) and your analog ones probably won't shift it perceptibly.

Now let's say that your Helix and Axe FX III are delivering the goods when it comes to 4-cable-method performance. Good noise floor, good latency performance.... but gosh darnit, you just really need another digital pedal in your rig. You need to remember that this additional digital pedal can also potentially add additional latency.

Also the effects you choose for your patches, they will also change the latency somewhat; particularly pitch based effects.

 

[SillyOctpuss]

I was a real wet blanket on using the Helix in 4cm for a while but this was based on me being able to hear a difference in my spare room when switching helix in and out whilst looking for the difference.

When I got the levels sorted to balance the effects loop and started using it in 4cm at rehearsals and gigs at volume it just sounds like my amp and I don't even notice there's anything there.

I definitely don't notice any extra noise in day to day use vs going straight into my amp.

 

[Cirrus]

Re; those oscillations, could be shaped dither noise? Just an on the toilet thought, could be bollocks.

 

[EOengineer]

I was a real wet blanket on using the Helix in 4cm for a while but this was based on me being able to hear a difference in my spare room when switching helix in and out whilst looking for the difference.

Ive battled with Helix Floor and HXFX for years and eventually gave up on 4cm. I want 4cm to work for me, and maybe I’ve just never nailed the levels, but even at rehearsal I feel like I’m losing dynamics or responsiveness in the top end.

I need to take another swing at it again. It is mighty convenient.

 

[FractalAudio]

The oscillations are caused by leakage. You're getting a mix of wet and dry which is causing slight comb filtering. This is almost certainly occurring outside of the units somewhere. Either in the interface or the cabling.

 

[Orvillain]

The oscillations are caused by leakage. You're getting a mix of wet and dry which is causing slight comb filtering. This is almost certainly occurring outside of the units somewhere. Either in the interface or the cabling.

Hmmmmm. Maybe the amps effects loop??

Wouldn't explain the difference with the G3 though. How odd.

Preset on both the Helix and Axe3 were in line with the 4-cable instructions from the manuals. Humbuster cables in the case of the Axe3, but not Helix.

I had an Orchid Electronics 1:1 isolation transformer in the chain going into the front of the amp too.

 

[FractalAudio]

Hmmmmm. Maybe the amps effects loop??

Wouldn't explain the difference with the G3 though. How odd.

Preset on both the Helix and Axe3 were in line with the 4-cable instructions from the manuals. Humbuster cables in the case of the Axe3, but not Helix.

I had an Orchid Electronics 1:1 isolation transformer in the chain going into the front of the amp too.

Probably the effects loop. The G3 is analog so it doesn't have any latency therefore comb filtering doesn't occur.

Effects loops are notorious for leakage (looking at you Mesa/Boogie). Leakage almost always gets worse as you go up in frequency.

 

[Orvillain]

Probably the effects loop. The G3 is analog so it doesn't have any latency therefore comb filtering doesn't occur.

Effects loops are notorious for leakage (looking at you Mesa/Boogie). Leakage almost always gets worse as you go up in frequency.

Interesting!!

If I drag myself out of bed from this bloody flu I've got, I might try another amp to check. Maybe the Rockerverb.

The FX loop on the Dual Recto I have is pretty pants; I disable that one always!

 

[Alex Kenivel]

I love grafs.

Get well Drew

 

[SillyOctpuss]

Ive battled with Helix Floor and HXFX for years and eventually gave up on 4cm. I want 4cm to work for me, and maybe I’ve just never nailed the levels, but even at rehearsal I feel like I’m losing dynamics or responsiveness in the top end.

I need to take another swing at it again. It is mighty convenient.

I found that comparing my Helix in front of the amp vs 4cm I had to bump the volume on the effects loop by 2.8db or I was losing signal vs just going in the front. The jvm loop is footswitchable so I could A/B really quickly. Once I'd balanced those levels I've not had any problems since.

 

[Orvillain]

The Orange Rockerverb MKIII does seem less prone to the leakage issue that Cliff talked about. Kinda cool!

 

[Orvillain]

For anyone curious about the Dual Rectifier loop, here is what happens to the frequency response when you engage it:

This is sweeping the input of the clean channel. But no knobs moved during the different sweeps. Just the red curve is the loop bypassed, and the green curve is the loop enabled.

 

[Sascha Franck]

Effects loops are notorious for leakage (looking at you Mesa/Boogie).

Ah, so I haven't been fooling myself when I noticed that back in the days.

 

[texhex]

Wow, thanks for this post @Orvillain. Doing some 4CM research and this is fantastic.

 

[laxu]

This reminded me of several years ago when I ran the Helix Floor in a 7CM setup through a Bogner and Victory. That was still no issue with latency or anything and sounded pretty glorious.

Later on I had a pedalboard with all the pedals connected to the Morningstar ML5 loop switcher, for lowest latency/capacitance. But honestly that was just overly complicated when I had several digital pedals in there.

I turned that into a MIDI controllable setup where the digital pedals are all just chained together until the amp input, then a few more in the fx loop. That's something like 10 A/D/A conversions when all pedals are on at the same time? Still zero issues and since these are Strymons the latency is about 3ms when you exclude the analog dry-through pedals.

Aside from poorly designed fx loops, the things to actually worry about are IMO input buffering, isolated pedal power and ground loops.