Input Gain Advice/Question in AT5

[Matthias Mueller]

Actually, ideally, existing presets would by default have "legacy mode" enabled, and then you can select "legacy" or "new way" on a per-preset basis.

Good idea! i can imagen this, that when you open an "old" preset, a small pop up window plops up and ask if you want to open this in "legacy mode" or "new mode". maybe with a checkbox to use your choice for all further presets you open from the "created prior this update catigory"(sorry i dont know how to say but i guess you understand what i mean). So that, when you work with a lot of "old" presets you dont get always this message and it starts to get annoying.

[Boogieshoes]

Some very interesting points in you have added, or perhaps more that I have missed in other conversation.

1. The legacy setting - I actually tested AT4 and thought I could use it as I could get some good BF fender sounds. Then when I went to buy it, it was AT5, stupidly I didn’t test that and the configs from AT4 didn’t translate to AT5. I asked about this and hoped I would be able to use AT4 under the AT5 paid license. That was a ‘no’. But nobody said anything about legacy setting…. I may go and look for it.

2. The suggestion about asking users for ‘Interface headroom’ info. That if used, needs some way of saying ‘plug you guitar in and adjust input gain to maximum at the loudest guitar volume/tone settings, that does NOT clip in the interface’. That depends on the interface and its capability to detect and report clipping (that isn’t universally delivered on devices).

Then AT software would ideally need some form of self calibration that is applied after physically indicating that you have configured the interface at ‘Max input no clipping’ setting.

Rhetorical Question - I suggest that would bin the need for the left hand input slider control in on the patch creation screen, possibly even the right hand output one too.

The problem is that there are many users who struggle to sort the basics of the interface. Added to this you would need to do the calibration when changing guitars as the interface will react differently to each guitar even if, in theory, the max output of the interface to AT should be the same level. I wonder how many forget this when changing guitars etc now as is.

Even with those thoughts this would also affect the modelling of the pedals, so how do they address the varying gains/levels and buffer impact of the pedals that are modelled, I am not sure that this would be so easy to address and some compromise would occur. This is way before considering speakers, cabs, rooms and mics. I went round in circles on this which is why I felt I’d hit analysis paralysis and why I think the stock answer is to tweak using your ears to get what you want. I have found it got too much and I just felt uninspired by the effort.

I sometimes use AT5 as a to capture some quick ideas, but I generally accept that it is just an idea to get the rhythm and melody before I forget it, and that is all. I have not kept anything in any recording after that.

[machinated]

2. The suggestion about asking users for ‘Interface headroom’ info. That if used, needs some way of saying ‘plug you guitar in and adjust input gain to maximum at the loudest guitar volume/tone settings, that does NOT clip in the interface’. That depends on the interface and its capability to detect and report clipping (that isn’t universally delivered on devices).

If there is an option to set the interface headroom info, then you wouldn’t need to measure the signal from the guitar. You just set the value and it’s done, the software would calibrate internally. This is what the headroom screw on UAD and some Waves plugins does.

There could just be an option in the plugin where you type 7dBu or 12dBu or 0.707V RMS=_____dBFS.

Once that data is entered, AT would internally adjust its input levels so it matches what the software is running on. The input slider would then be used to adjust for creative purposes.

The purpose of basing the input levels off the headroom of the interface is precisely so you don’t need to base the level off signal peaks (which vary guitar to guitar, player to player, pickup to pickup etc). It’s why some players using single coils have a better experience with Amplitube (because they are likely recording with less headroom)

[Matthias Mueller]

But nobody said anything about legacy setting…. I may go and look for it.

No need to look for it. There is no such a thing. Its an idea from me what ik multimedia could do, to prevent that presets, that were creatad before the update, gets unusable.

2. The suggestion about asking users for ‘Interface headroom’ info. That if used, needs some way of saying ‘plug you guitar in and adjust input gain to maximum at the loudest guitar volume/tone settings, that does NOT clip in the interface’. That depends on the interface and its capability to detect and report clipping (that isn’t universally delivered on devices).

Please forgett this thing with "set your interface just before clipping" in context with what we say. This is what IK tells, not I (or machinated).

Then AT software would ideally need some form of self calibration that is applied after physically indicating that you have configured the interface

yes

at ‘Max input no clipping’ setting.

No!

Rhetorical Question - I suggest that would bin the need for the left hand input slider control in on the patch creation screen, possibly even the right hand output one too.

kind of, it would be not that important anymore but can be used for creative settings. (Output is the same as before, nothing would change here

Added to this you would need to do the calibration when changing guitars as the interface will react differently to each guitar

)
No,the interface doesnt react differently, you just send more or less (louder/quiter) signal to it, depending on witch guitar/pickup/pickup height/pick/strings (also how hard you pick or strum) etc. you use. And we WANT all this changing parameters so we DONT compensate for that!

so how do they address the varying gains/levels and buffer impact of the pedals that are modelled, I am not sure that this would be so easy to address and some compromise would occur.

The same "rule" for amps applies to pedals (for all that are nonlinear at least)

[Boogieshoes]

Nearly with you on most points, however where you say ‘No’ to setting the interface for ‘Max input with No clipping’ I am of the view that the software should have asked to capture the raw guitar at this setting and from that level (which would most likely be different between interfaces) use that data to derive a ‘headroom’ pre scaling factor for the signal chain that the user puts into their ‘Preset’. That way the headroom needed should be consistent and the user is removed (except for playing the guitar at Max no clip) from determining the headroom.

In a minute somebody is going to come back and that just what we did… sort of even though it doesn’t feel that way. Sorry I took a timeout on replying today, had a friend over and we were playing the real amps and pedals…

I’ll take step back from this, though I have to say it has been a very interesting topic over most that I have scanned. I do look in, ever in the hope that the forum would have a bit more real world practical debate for using the package, but that simply doesn’t happen and it has meant trying everything on your own with a limited user guide. The YouTube stuff simply skims stuff, about the only review that hit on some the points we have discussed were in EytschPi42 review when he criticises the ‘levels’ …. As I say I have yet to find anything walks through this stuff to getting AT to work well. I haven’t looked at forums and debate on _ to see if there is anything to be gained from other Amp Sim users.

[Matthias Mueller]

Sorry i may not understand it correctly what you say but...

I am of the view that the software should have asked to capture the raw guitar at this setting

I think you overcomplicate things here. It could be as simple as that:
- Set your Interface to "0gain"
- Tell amplitube the value of headroom your interface has until 0dbFS
- Done. Amplitube has now all the info it needs.
Easy and accurate. No need to have some AI to capture any Raw guitar level at x level, or anything in this direction... its no rocket sience... . (+You made sure all "user variables", like pickup type etc. are preserved. )

-Because Amplitube now knows what level to expect from your interface, it can adjust by itself for its gain structure.
- This would be again nothing complicated, it just knows that when you type in eg. :
headroom = 12dbu @0dbFS it has to boost 12.79 for amp a, 16 (or what ever) for amp b etc. ...
In this scenario we users dont care anymore about whitch amp is xdbu = xdbFS. Amplitube itself will take care of it.

and from that level (which would most likely be different between interfaces) use that data to derive a ‘headroom’ pre scaling factor for the signal chain that the user puts into their ‘Preset’.

Yes the level you have to boost to get there is different for each interface, that (minimal) headroom thats left after boosting to "near clipping" is not. (at least i dont know any interface that doesent correspond to min headroom (clipping) = 0dbFS)
Amplitube can not smell what interface you used and how much you had to boost and also (as already said) most interfaces dont show you this amount in numbers.
+ Hot Input Signal brings you no advantage, but can be very dangerous for your recordings if you are not careful.

That way the headroom needed should be consistent and the user is removed (except for playing the guitar at Max no clip) from determining the headroom.

Exactly, congratulation. It doesnt matter now what guitar you plug in, how hard you play etc. Its all the same.
-----------------------------------------------------------------------------------------------------------------

There could be another possible solution to tell amplitube your interface headroom without even look into the specs. But you would have to create an feedback loop instead. (Output goes into input) But this would be way more complicate to programm and i dont even know if it would really work. So i dont explain it here further.

[Boogieshoes]

Mathias,

This thread went through various stages of debate, but as an owner of AT5 Max on V5.3 and not getting much if any real use out of the package for this and other reasons, I am wondering… Did you ever get an answer to you support request?

I gave up on my initial attempts long ago.

[Matthias Mueller]

Mathias,

This thread went through various stages of debate, but as an owner of AT5 Max on V5.3 and not getting much if any real use out of the package for this and other reasons, I am wondering… Did you ever get an answer to you support request?

I gave up on my initial attempts long ago.

Hi Boogieshoes,

No, sadly. Its a bit over a week or so now, so not too long. but i know that other people have submited their ticket way before and still have no answer. so... not much hope...

but i'm currently working on an "input gain list". Im maybe right there with some amps and wrong with others but i hope with feedback and suggestions from other users we can get this somewhat right.

Sad to hear that you can not achieve the results you want/expected with amplitube.
(but i dont work for them in any way, so i dont care either )
maybe you want to give it another try when i upload my list, shouldnt take too long anymore.

[Boogieshoes]

I am sorry to read that you haven’t had a response Mathias, that isn’t good.

I had a number of tech issues when I first installed AT5 in place of the demo version of AT4 as the package was not working well. In truth I was using a pc which complied with the basic system requirements for AT5 but it was really hard to get the package to work and it still fail. Tech support did reply but the info was pretty limited and I ended up using Resplendence latency analyser to help improve things, but got that recommendation in a DAW forum rather than here.

I am not quite clear how your input settings info will relate my interface (Octa-Capture) and the amps I have tried, Princeton, Twin Reverb, AC30 and Marshall of which I have the real items except for the AC30 here. With the amps I run Guitar into the following pedals Dunlop Wah, Boss FZ1 -Waza, King of Tone, Ibanez CP10, TS808, ADL FLL, SC10 and Retrovibe. Most of the time I don’t have the Wah in-line. I like to run the amps at the sweet-spot (extremely loud with the twin) so the Princeton is at about 4.5 Vol and cut the bass as the Princeton can flub out. Even with the Waza Fuzz set pretty strongly, winding the guitar volume control back to about 4 will clean things up and you can push things by simply upping the guitar volume and tweaking tone controls. It is this real dynamic capability that I find I simply cannot get to work with AT5. The modelled pedals have, as I said before, to be compromise choices over the real items I have, and putting the pedals in front of the interface I simply haven’t got to work well. What I get I’d describe as a cardboard cutout sound flat and lacking. I am wishing I had bought an iso cab for a speaker and used the amps.

[removed - please review the rules]

[Matthias Mueller]

I am not quite clear how your input settings info will relate my interface (Octa-Capture)

Im not an expert in this so maybe im wrong but i found this information.
Nominal Input Level
Input Jack 1—8 (1/4 inch TRS phone type)
-46 to +4 dBu

For me this would mean +4dbu @0dbFS. means you would need to boost +4.79 to get to 1vac=0dbfs

but i found also this in the specs of your interface

Head Room

16 dB

not entirly sure to what this information refers. but i dont think its the headroom that we are intersted in.

maybe someone will correct me if im wrong.

but you can also test for yourself. set input (on the interface) to 0, boost +4,79 in AT. when you play, is your level somewhere around 0dbfs or even above? then that should be correct.

[machinated]

I’d be careful with those values, but also because on the Octa-capture there are no DI inputs. The TRS input impedances are way too low for a guitar and will make the pickups sound bad. I’d recommend using a DI box and then measuring the headroom with a sine wave to be sure

[Boogieshoes]

The Octa-capture inputs 1 and 2 have Hi Z so they are are the only ones used for guitar with this device. If you want mega irony, I still have a POD XTLive that I can use as an interface too ( I thought modelling may have moved on a bit since 2008… err… well no not really is my view ).

This person https://www.youtube.com/watch?v=IRBLJobxNCY is playing about with things too, though he’s not particularly clear on what he’s using to boost prior to AT5.

If I go looking at his ideas then the specs for the Octa-capture aren’t that clearly defined but could be well below the figures he is showing for other devices as page 95 of the manual says input 1/2 -87dBu at 600 Ohms. The others, presumably when not Hi Z are shown as -56 to -6 dBu. I haven’t looked at the POD specs as it’s buried under other gear at the moment and doesn’t get used.

[Boogieshoes]

Interesting that your tests indicate 0 dB gain requirement for the Fender amps Mathias.

I have asked Roland for more detail, if possible, on the Octa-Capture input dBu figures as unlike other the spec does not quote this vs 0 dBf. Either way, I can’t see it fixing the issue that the Fender sims do not react to guitar volume control changes like the real amps do. That is without the pedal model buffer/gain issues that seem to also impact.

[Matthias Mueller]

Interesting that your tests indicate 0 dB gain requirement for the Fender amps Mathias.

0db in the list doesn't mean that you dont have to boost entirely. It could mean +12, +6, +?,... depending on your interface.

If you think a number on the list does not match to the real amp, please tell me what number/level would give you a more realistic behavour for this specific model.

In the end Amplitube is still an Simulation and not the real deal. So it will never sound and react 1:1 exactly the same. We can only try to get the best out of Amplitube but when the limit of this Software is reached we cant do anything about it.

As machinated already said:

but also because on the Octa-capture there are no DI inputs. The TRS input impedances are way too low for a guitar and will make the pickups sound bad. I’d recommend using a DI box and then measuring the headroom with a sine wave to be sure

Instead of this you can also try to use an entirely different audio Interface. (thats maybe suited better for guitar DI input)
+ You can also try to increase the Sample Rate, maybe this will also help a bit (if your computer can handle it without latency issues)

[dirtbag99]

I must be tone Def or just amplistupid max.i been messing with this
Thing for couple years.tryed every thing known to man and I think all I
get is 80% worthless frcking noise.it might sound faintly close but mine
has never sounded as good as others I heard...my dimebag sounds like
dirtbag.dont sound like that guy on you tube....the stoned guy who bought
Ola's it's...not even close.imat a loss.feels like i havent played since I started
messing with this crap..i gotta find something else before I go nuts...i guess I'm not
amplispunk compatable