Essentially, yes. So what it means is that it's not simply the preamp signal with cab emulation. You get the power amp interaction as well.
Yes, the cab emulation is added post this signal.
This is NOT what Line6 support says. Below is a conversation I had with Line6Hugo (hmerida) about cab modelling, and LVM. Hope it provides some information on Line6's stance. Bear in mind that I asked basically these questions in a different support ticket, and some of the answers I got there do NOT agree with this info. So I think Line6 themselves have different answers depending on who you ask... Professional, eh?
Line6Hugo: When connected via L6 LINK, the cab model in the DT25 is simply being controlled by the HD500 User Interface. Using the default cab models (the ones which automatically pair with the preamp) should provide the same results when using the same preamp and default cab model in the DT25 stand alone.
Me: "In your explanation, the cab modeling is placed before the power amp, not after. Is this really the case, and if so, why is that? It makes much more sense to have it afterwards. Also, why is there cab modeling active in the dt25 standalone when it has a cab and speaker anyway?"
Line6Hugo: - Cab modeling exists in the DT25 to create a complete amp tone. If you prefer to turn it off you may do just that with a connected HD500 or in standalone mode when connected to a MIDI controller accessing the DT25 2.0 firmware features. Having the cab modeling before the power amp in the DT amps ensures a consistent experience when using with or without a POD HD device.
Me: "How does the use of LVM mode and the direct out affect the placement and operation of cab modeling?"
Line6Hugo: - LVM has no impact on the cab modeling. Direct out uses the same cabinet model through a model of the mic’ed DT25 cabinet. This provides a tone which matches the cab model that comes through the DT25 speaker through a mic, as you would here in a recording or through the PA… essentially the same experience you would get with a microphone but direct via XLR.
Me: "Finally, I notice a large difference on the B channel when I activate LVM. My test case is as follows: dt25 a and b channels set with all knobs to 12oclock, topology 2, P-75 Pre (no cab) I have a footswitch set to toggle the amp block on and off and they sound similar (but far from identical). However, if I engage LVM and toggle the amp block on and off, the B channel now sounds very different (much more distorted). Why is this, nothing in your explanation suggests why the power amp/LVM would behave differently."
Line6Hugo: - More information is required to properly answer this question – normally toggling an amp model bypass (on a connected HD500) also toggles the amp between channels A and B. If channel B is not set identically to the amp model when active, the sounds will be different. Also, each channel (if switching) must have its topology settings set identically. Engaging LVM does two things: uses HD Technology to deliver the preamp and power amp modeling (whereas non-LVM provides just the preamp model) as well as scales the level of the Master Volume knob on the DT25.
Some more conversations with support. might be of interest:
With the DT in LVM and using the direct out, disabling the cab modeling on the HD500 will remove the cab modeling from the output of the DT.
Removing the cab modeling causes the mic modeling to be removed as well.
We tested this scenario here and our ears told us there was no perceived change in tone when changing mic models while the cab modeling is disabled. Thus, the mic modeling does nothing when no cab is being modeled.