I've now lived with Bryston passive and active speaker systems for the past several years, starting with the Model T Signature with its' very good external passive crossover modules. The system evolved through converting the Model Ts to full active using additional amps and the BAX-1. For quite a while now I've been living with the Model T10s. Anecdotally, this progression took me through at least four switches from active to passive, in part simply to implement change, but also on at least two occasions, to verify whether there was anything still preferable in the sound of the passive versions.
Please note that my opinions relate specifically to the Bryston components mentioned above, I have less experience with such comparisons with other products. (I do have some experience of larger mid-field active systems used in recording studios, but didn't own them.)
A year ago I finished a full review of the Active T10 system. It's a long review, below is an excerpt specifically discussing my observations when listening in both passive and active modes. (If anyone is interested in the full review, it's available from James Tanner, or me.)
"The Elephant In The Room
I hadn’t done a passive vs. active comparison between a T type speaker since upgrading my original Model T speakers quite a long time ago. The BAX-1 crossover uses DSP for very fine-grained control of crossover characteristics, much more so than can be achieved using passive components. Bryston claims that such levels of control even help to improve the sound power characteristic of the speaker, the relative evenness of the total envelope of sound all around the speaker cabinet, not just the frontal frequency response. The analog signal that leaves the preamp is converted to digital in the BAX, and back again to analog at the outputs to the amplifiers. I got to wondering: is all that A-D and D-A doing any harm? Somehow leaving something good on the cutting room floor, so to speak?
So I took the time to revert the system back to passive crossover operation, and listened to selections from a demo playlist, of both digital files locally stored, and vinyl from the SL-1210G turntable, using my old Shure V15 V with new Jico SAS/B stylus, eventually switching back to active mode, and re-listening to the same material. (Digitally speaking, I don’t stream digital files. I use music stored on local drives.) The comparison was completely subjective. I did no measurements, apart from checking volume levels.
The change to active mode was unquestionably an improvement, in every way that matters. Not only were dynamics improved overall, but micro-dynamics, that make for subtle shadings and details within the aural canvas were clearer, less homogenized. A specific instrument within an ensemble could suddenly be played or struck with more emphasis, without affecting the other sounds around it. The soundstage was at least as wide, high and deep as with passive, and with certain recordings, extends with a sense of three dimensionality well outside the speakers, and further back in depth. If you’ve ever heard a tall line-source speaker design, you already know that soundstage height is usually abundant. (Depth is a discussion unto itself – I’ve never sat in Carnegie or Massey hall and thought “I can hear the rear wall of the stage”. I tend to think of audiophile discussions of recorded depth to be relative to various recorded instruments and performers within the aural “stage”.)"
Brian