One thing to mention regarding using a Pre-Amp instead of a Direct Box would be impedence, right? This is from the RME User Guide... http://www.rme-audio.de/download/fface_ucx_e.pdf The main difference between a line- and instrument input is the input's impedance. Via the op- tion Inst in the TotalMix channel settings, the input impedance changes from 10 kOhm to 470 kOhm (also called Hi-Z). At the same time the input sensitivity rises by 6 dB. The instrument input operates unbalanced.

.

.

Thanks, Mike. I think none of my pre-amps (Art Tube 40 dollar things, RME Quad Mic Pre etc) have the Hi-Z except 2 inputs on the front of the UCX. I only brought this up because I may not be the only user who has pre-amps that don't meet this requirement. .. Mike, I appreciate your time in answering my questions. These are things it takes a long time to learn and requires an understanding of the electronics I don't have. Thanks for sharing.

If you have this Art pre amp it has both a Hi Z and a XLR lowZ input. That's the Hi z input showing on the left. Change the tube in this puppy and you have a great little como Pre amp DI box. I also have to ground lift it when used with a snake, in the studio it's not bad. Depends on the type of music. But for the money it's a handy little box. 

On the right you see the Hi Z output that carry's on to the Amp and there is also the XLR out put that you use to record the dry signal. 

I just thought I'd point out that there isn't always a clearcut distinction between a DI and preamp. The Ultrasound DI Plus is a preamp. I use it with electrics, it's transparent and has plenty of headroom.
 
http://www.ultrasoundamps.com/image-viewer-diplus.html
 

I really enjoy these types of discussions as they provide a great opportunity for forum members to learn a bit more about the approach others are taking to track guitars.  Some interesting set-ups discussed here, some relatively straight forward as well as other more complex routing options.  While this is getting away from Sixfinger’s original question of the best way to mic an amp and route to an amp sim at the same time - for those inclined to experiment a bit, the 4 cable method (a variation on what Mike is doing) has been around for some time, although is geared a bit more toward effects processing than amp “blending”.  I tried the method myself a few years ago but found it a bit too noisy for my tastes.  Perhaps the equipment I was using was not as well suited to this method (given that some amps/ preamps are inherently more noisy than others).   Also have to be careful what type of signal is routed where - as it can lead to damage to one or more units.    
 
After reviewing a range of different products (switcher pedals, DIs, splitters – some of which had parallel outputs but were not actually “true bypass”) I ultimately chose the Mesa High Gain Amp Switch as the preferred option for combining two or more guitar preamps (although is way too expensive of a unit to recommend to anyone).  With the HGAS, the signal is split (as many as four ways) directly after the guitar without any noticeable signal loss or degradation.  This allows you to route to different guitar amps, processors and/or direct to the recorder all at the same time.  Made for some huge guitar tones with lots of characteristics, however after a few years of experimenting and getting older I came to the conclusion that less was, in fact, more and basically returned to tracking my main guitar preamp direct (i.e. not layered with other preamps).  While I tend to track heavy guitar tones, I usually dial back on the overdrive and effects going in as I found that even though it may sound a bit thin or “weak” during tracking it actually cuts better in the final mix with less dirt.  The preamp units I was originally layering with my main processor made for a really big sound but actually detracted from the overall punch of the tone.  The other thing to keep in mind is that when adding in two or more preamps at the source you may have to deal with grounding issues, impedance, latency, distortion (not the good kind but rather things like high-end sibilance), phasing and increased noise floor.  And with increased noise comes increased gating which can affect the dynamics, playability and color of the tone.  I've had much better luck "doubling" tracks (sometimes panned a bit different, with slightly different tone or a variation with a certain effect applied).
 
In my opinion - despite all the physics, math and supposed “rules” - it all comes down to the tone.  If it sounds good then it’s right…and doesn’t necessarily have to be complicated or cost a fortune (contrary to the Axe-FX approach to guitar tones – no offence Fractal Audio).  At the end of the day, 99% of the people who listen to the track will not be able to tell what equipment was used or how much it cost to track the guitars.  I’ve heard some great guitar tones recorded with relatively cheap gear.  Guitar tone is also subjective.  I have a friend who is a fabulous player but I really don't care for the tone he uses.  Again, comes down to personal choice and style.  Best advice I can give is to not be afraid to experiment.  Just because someone tells you it’s the “wrong” way to do it doesn’t actually mean it is.  What works for you may not work as well for someone else, and what may have sounded good on one project may not fit as well on the next project.  That’s what makes all of this so much fun……getting there.

Please play responsibly......

Mike getting emotionally involved ! How does it feel.   I think the preamp idea is OK for sure but the DI concept here is also getting an unfair beating. It is a very simple and effective option. Try both and see which the guitarist prefers.

The DI's high input impedance eliminates most input impedance loading issues. It does not add gain so it is not a pre amp but rather a buffer. Buffers can be well designed. (very low noise and distortion and high headroom) The gain is being added later by the preamp that is in your mixer or interface. Another reason leaving a pre amp out is also valid. Because two high gain stages in series are not usually necessary. That is where one could easily overload the other with silly settings. 

The fact that a guitarist may run the instrument through some sort of very transparent buffer before even feeding direct into an amp has never impacted on their performance in my experience anyway.

I have just finished tracking some guitars for a pop EP I am producing. He used a Tele a lot (I love that guitar!) but he tuned up with a beautiful Gibson hollow body jazz guitar on the second session. He had a Princeton amp as well as a custom designed Fender Twin. He ran his Gibson guitar at first straight into the Princeton and it sounded amazing. I inserted the DI into that path so I could get a total clean guitar sound and after careful checking it made no impact into his setup and sound. He still moved me to tears with his delicate playing. (BTW we did some hard rock and pop tracks with that Gibson guitar and it just screamed!)

The good thing about the DI is that as long it is a quality unit and a nice one that direct sound is going to be very nice and well recorded. It is so useful to have the guitar sound so untouched at this point too because the processing that follows it likes that sound as a starting point. I get better results from VST's and things when that sound has not already been pre amped in some way.

Mike I do appreciate the mention of phantom power supplies because I was only thinking today that it is a good point. A DC supply normally has a very low output impedance but a phantom DC rail has a higher output impedance. This and the fact the DI designer has no real knowledge of the quality of the phantom power that is coming in current capacity wise too. Makes me think maybe the battery might work better on a live gig for example. I have been lucky and never had a poor DI experience yet. I always take an active and a passive DI to any live mixes I do and always end up using them! I might setup the CRO and oscillator and do some tests such as determining max out level before clipping with the battery and the phantom power and try different sources of phantom power too. Mike you have got me interested in that with this thread and I think that is a positive.

The_7th-Samurai has brought up some good points especially in relation to multiple preamps in series and things. I am from the Hi Fi era of always heading toward the best sound with the least number of things in the chain. I have tried quite a few complex guitar signal recording techniques and after many years of doing it I seem to be going for the simpler options often. In this EP I am producing we got amazing results using a Tele running into a Timmy crunch type pedal and into the custom Fender Twin and the sound was just huge and recorded incredibly well with little fuss. The DI sent me this lovely clean sound that I have got so many options with later too. Incredibly simple setup with just two tracks recording and many options later.

Reamping requires some special considerations too. The box feeding the guitar amp for example. It is reducing the often pro line level output from a DAW down to guitar level. But also there is the quality of that input transformer. It cannot be too small a transformer otherwise it can saturate. Not all reamping devices are necessarily good and I think there would be more variances there compared to DI's. They should also include an equivalent circuit for a guitar too so the amp is really tricked into thinking a guitar has been plugged in. I have designed a nice reamp box but not built it yet. The reason is I am loving the VST's now and the need is getting less and less. 

Thanks, Cactus Music.

http://artproaudio.com/art_products/signal_processing/signal_channel_tube_preamps/product/tube_mp-original/

This says that the ART input impedence is...

Input Impedance 2k ohms (XLR), 840K ohms (1/4")

Above, RME says Hi-Z is 470.

Is 840 also hi-z?

High impedance is considered around 1Meg ohm or 1,000,000 ohms which is high. 840K ohm is also pretty high so yes 840K is high and so is 470K. (reasonably high)

It means that the total impedance the guitar sees is the input impedance of the amp and the DI in parallel. If the amp was 50K ohm and the DI was 840K then the guitar is looking into a roughly 47K ohm impedance (Impedances drop as they are connected in parallel) Not a big change there.

50K and 470K in parallel results in around 45K which is still not a great change either. Slightly lower but only by a small amount. 

The ART product above is a Mic Pre that can be used as a direct box. If you did not want the Pre amp part of it then a quality direct box alone would also serve that purpose. (just DI that is no gain) But if you are wanting a pre with gain this would be a great choice too. What is good about it is that it has its own power supply unlike a conventional DI so you don't have any battery issues or phantom power issues either. But you do need somewhere to plug it in!