True or false?!

[RainbowStorm]

For some time now I have been thinking about something: Signal/noise ratio in keyboards compared to signal/noise ratio in acoustic tracking.

Is it true that you are able to record with a much better dynamic range when you are using a really good mic + pre-amp and record an instrument acoustically compared to recording a synth/keyboard directly through its outputs? For instance, I have a Yamaha Tyros keyboard, which has both 1/4" stereo outputs and optional speakers. I have assumed that I get the best sound quality by recording it directly through its analog stereo outputs (it has no digital outputs) and the signal/noise ratio would then be something like 65 dB. I recently read an article where it was stated that keyboards often have a signal/noise ratio of no more than 50 - 60 dB from its outputs. Since I have an RME Fireface 800 I would lose nearly 30 dB of dynamic range by recording this way, since the Fireface can record at a signal/noise ratio of more than 90dB at 24-bit. Then I start thinking of my Mackie HR824 monitors that can produce sound with a signal-to-noise ratio of more than 100 dB and compare them with the TRS-MS01 Yamaha Tyros speaker set (2+sub woofer) which I have no signal-to-noise ratio figures on, those speakers could possibly have a better signal-noise ratio than 50-60 dB, so I could win some dB by micing the keyboard in a really quiet room instead?! Is this possible in practise?

But generally, are there any keyboards/synths with a much better signal-to-noise ratio through their outputs? Is this primarily why real instruments sound so much more present in a limited mix even though the keyboards can produce a very similar sound?

If you think about a really professional recording studio, what signal-to-noise ratio can some of the best studios acheive when tracking an instrument acoustically? I just want to know how far behind my recording specs are when I'm doing keyboard recordings in my home studio...!

For instance the 2005 tec winner mic AKG C 414 B-XL II (aimed mostly for the lead element) has a signal-to-noise ratio of 88dB, which is far away from something like 50 - 60 dB... How much of that 88dB is actually possible to capture when recording?