Does this make sense? GNX -> amp

ponq

Well, warning first. Untested material ahead. Input: 1 beer, output: this msg.

Background: I've always been struggling with getting the right sound at different locations. My own KX1200 keyboard amp servers my modeller (gnx4) pretty well. Recently at a jam it sounded awesome through a PA-system. Regulary at bandrehearsel through PA-system sounds mediocre...sometimes pretty good, often weak. Hard to cut through with crisptones when both the keyboard and the vocals are coming out of the same box. Running through a guitaramp is rubbish. I get sounds aimed at killing cats. Or birds. But not nice 'tones'.


I've read all the stuff about cabinet modelling and such. It's a bit hard for me to do, since I don't have the practice-amp at home. The other guitarist in our band plays through stompboxes to guitaramp. Same type of amp I could use, but he gets nice tones. Whatever I did thus far on the gnx into the amp (instrument in, line in, eq-settings changed, tried all kinds of amp models and such, it always sounded crap...uhm yeah...spkr cmp ON).

So here's the idea. What I need is to get a sound out of the gnx which is equal to the sound out of the stompbox-setup. That way the *same* sound will go into the guitaramp and (hopefully) the same sound will get out. Therefor what I need is an amp/cabinet model which doesn't model a real amp/cabinet, but morphs the singal in such a way that it turns out to be a stomp-box signal. As if we bypass ampmodelling.

I don't know if this is possible (and it's not likely possible for me to test), but can you sample the signal coming out of the stompbox-chain into a amp/cabinet model for use with the gnx-series? The same way all those great sounding amps/cabs have been sampled? That way we're not treating the amp/cab-model as a modelling-thing, but as a last effect applied to the chain.

Breakdown:
goal: disable amp/cab modelling so I can use the gnx as an mfx-unit with a guitaramp (for occasions when PA isn't available/sounds crap).

What to do:
-'model' the signal coming out of the stompbox-chain (\"real\" stompboxes, not the one on the gnx) with various settings (eg OD on/off, chorus on/off) so you get a general idea/shape/tone of the signal.
Eg, like if you plug in the stompbox-chain into a PA system (without mic'ing the amp, directly into the PA) it sounds thing, weird etc. That's for a reason (I guess). I hope we can model that reason.


Chances are I don't get my idea across :P, but plz ask for explanation. In my head this idea makes sense

iliace

Are you talking about getting a stompbox-level output from the GNX? So, not quite line level, more like instrument level. That going into the amp would drive the amp at appropriate levels, thus making it respond as it ought.

I think the first appropriate adjustment would be in the compressor module. First, a slower attack... low threshold... low ratio... adjust the makeup gain as needed to drive the amp. This will ensure both sufficient level (hopefully not too high) as well as good dynamic response.

The \"Direct\" model, in my experience, produces noticeably less signal than an amp model. You have to have gain almost at 99 and volume close to there as well, to get a matching signal level to an amp-model-based patch. But the problem is the coloration that occurs when you have the gain up so high. So, how about lowering the input gain on the amp model, and adjusting the model post-volume to bring down/up the signal level to where it needs to be.

The one part of your explanation that I'm not understanding is the PA hookup. Are you talking about GNX->KX->PA? Or GNX->KX->mic->PA? What amp is your other guitarist using?

guitar3456

What kind of guitar amp do you have?

Here's the thing.. amp models model the cabinet (already voiced and colored) for that amp model. Most guitarists with simple amp and stompboxes have basically one voicing. A modeler tries to achieve several. Problem is the amplifier will color the already colored (voiced) amp model. So transparency is req'd of your amplification system.

One way around this is to use various cab models first before you start tweaking EQ's and stuff. What may not work in headphones or KB amp may sound much better on the guitar amp.

Without knowing the guitar amp and cabinet you use with your GNX it's hard to determine what needs to be done to \"counter\" the colored model.

I do it with cab model changes. Try various cab models and see what you find that closely uncolors the sounds. Now use the cab tuning function to fine tune what you hear.

For example, if you had an amp model that used a Vintage 30 cab model, try an american 2x12 model and see what happens. Use the cab tuning +/- to set \"X\" semitones. + will brighten, - will darken. Also, try combinations in this manner.. say you have a Brit Stak model- put the same model in both red and green. Use different combinations of cab models from the drop down menu or other cab customs you may have saved on your PC. With all this connected to the guitar amp and GNX use the warp control to find a precise area that works at \"Stage Volume\"; not headphones. It will be different. Save the amp model after you get where you want it. In either the User Custom Area or on your PC. From Xedit do this: File> Save custom Genetx model> Pick a location. Then, simply reload that saved model for usage. See if that works... also, keep gains lower when using a guitar amp GNX combination. It gives the amp room to work.

ponq

Iliace: There's 3 setups available to me:
-at home (or places I can take my own kb-amp)gnx->KX1200
-at rehearselplace, there's the mediocre PA (sound is a bit shrill, even on vocals/keys). And there's the 'normal' solidstate guitaramp. I'm trying to get the gnx to sound acceptable (not awesome/great, just, uhm acceptable.

-g3456: I know about the amp/cab modelling thing, I've tried some. But... it doesn't really matter what amp/cab model I use. I'll tell you what I did:

I made 6 patches with Engl amps (from SS). Standard cab used. Varied cabtuning from -1.5 to +1.5. There was *no* difference in sound at all. They *all* sounded very very shrill, especially on OD/HG settings. In fact, every amp with OD/HG or stompbox dist from gnx sounded very very harsh. As if all the mids and bass were gone.

-I've got two sets of patches, one with Vox ac30's and one with Fenders. ac30's sound warm and dark, fenders sound brighter. They sound the same through the guitar amp. Both shrill, unusable.

-I've loaded the gnx with like a dozen amps. Clean they aren't bad, but once I use the gnx-OD/HG channel or enable a stompbox on the gnx the sound is gone, it's shrill harsh etc.


-the amp I use overthere is a 65watt Hughes and Kettner Club Reverb (attax-series). (Celestion Rockdriver jr 12\", 8 ohms speaker)

I found a manual which describes the pre-amp section of the amp:
(instrument) input: -10dBV/ 1 M ohm
line in (fx in): 0 dBV / 47k ohms

I think Iliace could be right, the signallevel into the amp coming out of the GNX maybe to high. Would giving too much imput normally result in a shrill sound? I use to have 'standard patch settings' (patch level at 50) and the 1/4\" out (mono) level about 25% turned up.
If I'm right, you'd normally connect the gnx to the line-input, since gnx gives line-level output. However, there didn't seem to be much difference using the line-in or the instrument-in. In fact, when using the line in I had to control volume at the gnx, since I'd bypass the pre-amp on the amp. When using the instrument-in I could at least use some EQ to kill some highs (the sound was still unusable, far from warm/fat etc).

Our other guitarist uses the same type of amp (there's two overthere) with 'normal' stompboxes. It's not the amp, I've used both amps with likewise results.

-just to add, it's not my guitar either. I've switched 'rigs' with the other guitarplayer and my guitar sounded awesome through his stompboxes (nice 'twang' on OD, warm broad sound, etc), his (america fender strat) guitar sounded like mine through the gnx.

Hope you understand what I'm saying. If not, it's probably my competence with the english language I'll gladly try explain again if I'm unclear, plz say so.

ponq

any more ideas or tips?

guitar3456

For guitar input usage, the GNX is very high as far as output level and that will in fact give your amp probs.

Try this, reduce the amp levels and amp gains by 20%. Finally, adjust the master preset volume on the patch to 11:00 position. See what happens from there. Obviously your cab is coloring the sound a great deal. Try other cab models to counteract the effect. Let me know what you find. If you prefer, send me an email.

mtprosound@wowway.com

iliace

It sounds like the line-in would be a good candidate for GNX running to the amp. 0dbv is close to the standard +4dbv line level. -10 is a \"sub-line\" level which is common on consumer-grade equipment. It should also work with the 1/4\" outputs on the GNX (instrument level signal, by comparison, is usually like -50, so -10 is a lot of headroom).

The shrillness is probably due to overloading the amp. Let me think about this some more...

dj482

Why not try running the output of the GNX into the effects return on the H & K amp? That input should be set up to take the line-level output of the GNX, and should use only the power amp section of the H & K, avoiding most of your tone-coloring worries.

Good luck!

guitar3456

Coloring is the voicing of any amp. Influenced by design and speaker selection. Modeling captures that coloring. This is why they sound different in EQ etc. If using a guitar amp, reduce gains.

Stompboxes are pretty much transparent. This is why they don't have as much impact on overall sound, except for the pedal settings themselves. GNX units are preamps. Patches are models of amps including cabinets, and their associated voicings.

Think of it as taking a triaxis, ADA mp1 etc (preamps) through your guitar amp input. Those units are not so transparent either, however they do not include cabinet modeling which accounts for additional coloring.

The best would be to have a Full range cabinet model for Live usage. Direct cab selection for Live amp models is another option. This is one we are doing extensive research on. It's not as easy as it sounds.

We are working with some cab designers to get a better way to implement modeling applications through guitar amps. The idea is to neutralize the effect of coloring for as many given speaker typed voicings.. ie Vintage 30's, G12T75's, Eminence variations, Jensen, etc etc.

You cannot necessarily reverse the freq response which would be easier. Instead it's a method of direct modeling that is as nonlinear as the original speaker in any guitar cabinet. If this is accomplished, you can make much better use out of any GNX processor/guitar amp combination.

iliace

G, you have a few of these right? http://www.behringer.com/DEQ2496

Couldn't you use the AEQ module with a reference mic to do frequency correction like you mentioned, in real time? In other words, running post CIT -> DEQ -> power amp ->etc

guitar3456

yes I do. We've looked at a few options and have found a SW based rta solution that is nuts and bolts.. The waterfall pattern, and 3D views show mid and HF variances you cannot detect with a simple rta.

The Ap units are proving response of speakers inconsistently. None have come close to mfg graph specs. So, the reference starting point was nil. We've tested a few dozen, many out of the same cab..old, broken in, and some right out of the box. It makes perfect sense as to coloring, phasing, and other problems users have had.

the cab guys have an Sa3050 something or another rta/spectrograph that they have measured the exact units. So, we have decided to pick a happy medium and try it out. The problem is the nonlinearity at various levels of volume.

Seems the most commonality in low cost amps are using Rockdrivers, Eminence (Tonker spec'd), Jensen, and Cel made customs. Mid to High price amps- V30's, G12T75, G12H, etc. Higher items are using EV, JBL, Tone Tubby and a few others. So, this narrows down the mix a bit. I think we can narrow the tolerance of the neutralized cab to about 10%.. which is better than 50-60% users have now. This can open the possibility of limited tweaks for Live modeling.

OMT- the cab on a guitar amp is half the battle.. the preamp coloring can (and we've verified it) add a bit more, BUT most eq's a flat at center on almost any guitar amp. That is.. flat as to the voicing of that mfg's amp design. Again.. that varies.