Output Signal Path - Analog or Digital?

My current I/O box is a Behringer FCA1616. I feed the output into a Behringer Ultracurve EQ*. Then from the EQ to a power amp and my monitors. 

When I first put the current rack together, I ran TRS from the main outs of the 1616 to the XLR ins of the Ultracurve and XLR out of the Ultracurve to the TRS ins on my amp. The system is quiet and sounds great.
 
The other night I decided to experiment with using SPDIF from the 1616 to the Ultracurve. Plugged in a cable and it works great although I only experimented for an hour or so.
 
So which is better? Is there any advantage to doing the D/A conversion in the Ultracurve instead of the 1616? Might it improve latency?


*Yes I commit the cardinal sin of using EQ on the way into my reference monitors. Are there phase anomalies in my mixes? Not that anyone can tell. But I can mix without having to do any kind of mental gymnastics. The bass I hear in the mix is the bass that translates on most other systems. Same with the mids and highs. None of this 'I'll mix it a bit darker because these speakers are too bright' crap for me.

When I say it translates well, that's in regards to using an EQ verses those who believe one should never use EQ on monitor speakers. I use EQ and that aspect of my setup is not in question.

As for latency, I dunno, I was just speculating. Perhaps offloading some of the workload to a different box/processor etc I suppose. 

Spent more time working with it today. I've observed a few differences so far. First, the 1616 has a master output level knob and meters on the front panel. But both are only active when the signal is going to the analog outs. With the signal going to the digital outs, you no longer have a knob you can quickly grab to turn everything up or down.
 
The master out fader in Sonar still works to control output level. But you can't always grab that in a hurry if something loud and horrible suddenly happens. 

The other thing I'm noticing is the output seems louder overall. I normally kept the output knob on the 1616 all the way up or nearly so. But it seems I must have had the input attenuation set a little low on the EQ because the signal going through the the digital bus seems about 2-3 dB louder.
 
Actual sound quality is the same as far as I can tell. I'm going to continue to experiment for now and see if there are any other differences.
 
If nothing else, this exercise has shown me that I need to recalibrate the analog input levels on the EQ.