I think there is some confusion here. I create a new audio track, add Amplitube plugin as insert effect to the track, set record and monitor to ON. Then click transport record. I can then hear the guitar as it goes through the Amplitube plugin while I play/record.
I have Cubase 9.5 (latest update) Pro.
Exactly the same in Cubase but you don’t need to indicate specific input if you put Solo your tracks (guitar and audio which will record) you’ll be able to record only your guitar.
But seriously I don’t see the advantage to record in real time it’s so simple to render or mixdown and listen after the result … 
Hi dgrm44
I hate to bring this up, but when you play the track back, have you remembered to turn Input monitoring OFF, if you don’t you won’t hear a thing. I mention this because it seems you are doing everything right otherwise.
Best Regards, Dave
I don’t understand the first sentence? I have outlined two scenarios in this thread, first - recording a guitar track using a VST insert amp modeler plugin like Amplitube…and the second - using a second track to bounce the first track to. With the first scenario the input is my audio interface channel with my guitar coming in via a radial direct box. With the second scenario, the input to the original track would not matter, but the input to the new bounce track would be the original guitar track that I just recorded. Make sense now?
And for your second comment, its a matter of saving time. When I bounce a lot of tracks over the course of a project that extra time adds up.
Thanks. I figured that one out right away 
I have seen that behavior in other DAWS as well.
If you route your track to a group channel, you can use that group channel as a input for a second audio track.
You can also instead create a Output Bus that is not connected to any output, and use that for output of the first guitar track and input for the second track.
I would not bother, just do a “Render In Place” when you want to commit the changes to a new track.
I read about the group channel option in another post while researching, the Output Bus option is new but sounds like same amount of work. I hope Cubase will get updated to allow better routing…easier/quicker routing of tracks as inputs to other tracks (common feature in DAWS).
Btw - Render In Place is faster than real time and can cause dropout/pops/clicks/anomolies. I don’t trust faster than realtime anymore. Been burned too many times in the past. Thanks.
It may sound like a lot of work but it really doesn’t take long to set up (a minute?). If it’s something you’re going to use regularly you can have it set up in your templates, then it’s there waiting for you.
Unlikely this is very high on their list but more likely if you either add your +1 to an existing suggestion (it does come up occasionally from new users so may be there) or create one in the Suggestions bit of the forum.
https://www.steinberg.net/forums/viewforum.php?f=249
RIP shouldn’t introduce artifacts, there’s something wrong somewhere if it does. I use RIP on nearly all projects on many tracks without issue. One of my favourite features of the program.
Real time rendering or recording is more likely to introduce artifacts.
Non real-time rendering will render even even if your cpu can’t keep up, it will just take longer than real-time .
We could debate this back and forth. All I can do is go by my experience thus far and when I bounce with decent buffers on my high powered system in real time, I never have any issues in any DAW I use. Thanks.
Sure, there is no right or wrong.
Life is just to short for real time
So last night did a test with R.I.P. using EastWest Play VSTi. Cellos ended up having crackles after the render. So…I can spend my short life chasing down the perfect settings within the entire signal chain for this one VSTi track to fix this issue, or I can just bounce in real time and be done with it.
Note that I spent time writing the cello piece, listening to it in context with entire mix, then doing the R.I.P, THEN listening back to it so that I can verify it was a clean render…noticing the crackles that didn’t previously exist…undo the render, re-bounce in real time and finally listening again and all is good. My point being R.I.P adds more time as I have to listen to it and verify that it was indeed a good bounce before moving on in the mix.
Well, then there is something wrong somewhere.
I do use RIP daily without issues, including EW Play.
When I have had problems, it has been down to using buggy plugins, or using a bitbridge
+1 to this.
Seriously, if this were me I’d spend some time working out what is causing the problem as RIP is a great feature…makes what the OP is after really quick and easy.
Maybe I am dumb as f**k, but if you want to record wet, why don’t you just put the amp modeler vst on the input channel?
(I don’t really understand why you don’t just leave the plugin on the track, but I guess you manage huge projects and need to save processing power?)
Your not so dumb after all 
You answered your own question. When you have a large project with lots of vst guitar inserts and midi vsti’s, then it can help to bounce them down to audio while recording (audio interface buffers are low so that latency isn’t an issue while tracking the vst guitars) and later jack up your audio interface buffers and start mixing.
Something else I have noticed about RIP. It doesn’t disable the VST plugin or VST instrument. Note that there is a difference between activating/deactivating vs enable/disable plugins in a project. When you truly disable a plugin you gain all the cpu back it was using, this isn’t the case when using RIP and deactivating your plugins.
Actually, I am not sure if Cubase allows disabling of plugins…need to research that.
** UPDATE - Alt + Clicking bypass will disable the plugin.
So, while I’m not really sure if you want to use the tracks you have “bounced” when mixing, or if it’s just that you want guitars through amps while recording more stuff, I would suggest that the simplest (and by far least time consuming) way to accomplish a wet and a dry track at the same time is to simply record to two different tracks at the same time. Just create a new mono bus called wet guitar, set the input to the input you normally plug your guitar into, then put whatever amp plugin on the wet input and record both the wet and dry (from your normal bus) at the same time. Even if you take into account the fact that you chave to create two tracks, you’ll save a lot of time even if the song is just a minute long:)
That’s what I think I’d do.
Thanks for the suggestion, but I need to hear the vst plugin while I track the guitars ,then bounce it down to another audio track (which will include the vst guitar effect so its wet, not a dry track) and disable the vst plugin, thus regaining CPU as I track my guitars and midi virtual instruments. I hope that makes sense.
Yep makes sense, but that is exactly what happens if you do it the way I suggest (if you enable monitoring of the track recording the wet input). You will record a wet track and a dry track without any more work, and regardless of the amount of tracks you record there will only be one instance of the vst amp modeling running:) And you have the dry tracks to reamp or whatever if you later decide to change settings om the modeling.
