100 dBa SPL peaks with live symphonic music. 90 dBa SPL with most good program material and 75 dBa SPL with over compressed stuff.
At 85 dB/W and 2M listening distance that would require 85 + 25 (dynamic range) +3 (headroom) + 6 (inverse square law) -15 (desired level 70 dBa SPL) = 104 dBa SPL out of the loudspeaker or 80 watts.
Last edited:
So just to be sure 70dBa SPL is -23dBFS give or take? (if taking the 0dFS definition of a sine wave that kisses FS one way. )
Your edit goes from 70 to 85? Also why 3dB headroom with a digital source that can't go above 0dBFS?
The 85 dB is the efficiency of the loudspeaker. The 3 dB is the margin to allow for DSP processing without clipping somewhere in the recording/reproducing chain.
ok so -31dBFS would be 70dBA for a good recording (0dBFS is -3dBRMS). Aligns to THD recommendations for the average residence. I think the loop is closed?
Schiit Ragnarok measurements now up for those who don't subscribe Schiit Audio Ragnarok integrated amplifier Measurements | Stereophile.com Not quite sure what to say to an amplifier that refuses to be measured other than I am not the target market!
Thanks for the link. It may suggest that a self-burned CD consisting of the 1812 Overture followed immediately by pianissimo solo female voice or solo violin, might offer some revelations to critical listeners.
Again, the real problem is not voltage swing, but current output. Perhaps I should make a 32W amp and call it the Dauphine. '-)
Thank you, SY. There are some fringes of what I did which I'm not so sure about, but that's also expected.
Oh, of course. Reality of theses today is that you really focus on papers and just glue them together, so the papers get the effort over the glued portions. 🙂
My advisor wants to graduate me, and I'm more worried about my committee falling asleep than lighting me up. No inter-committee rivalries to worry about, fortunately.
Okay, back to crest factors and sizing output capacity (I listen pretty quiet and near field, so attenuation is the name of my game). 🙂
I just fail to understand the need for a microprocessor-based bias control, save for novelty.
It's audiophile marketing genius. You can claim how great your amp is because of it, and then it becomes very hard to measure correctly. If it ever gets a bad review you can send them a firmware update to "fix" the problem.
You know, the usual sales blurb, cut and paste here. They say more dtail and better current transfer. To be fair, I did go for it after investigating several similar products in its time and found this one to be better than others, even if marginally on occasion.
The price was comfortable, €30/m.
Uh.. because it's so hard to make an amplifier that keeps its linearity over different power levels, I guess? I mean, heck, it might take as much as a $40 board to accomplish that very difficult feat. I have to wonder what exactly they were trying to fix that required all that research and uprocessor work.
My only (reasonable) thought is that yanking all the bias current out lets you get away with smaller heatsinks (as noted by the high idle temperatures). Use the microprocessor to hide that as much as possible.
except that JA review's fig 4 underbias seems inexcusable if it results in deadzone which many have pointed to as highly audible distortion - including the GedLee Metric
why wouldn't the fall back position be minimal (or is it Self's "optimal) Class B?
why wouldn't the fall back position be minimal (or is it Self's "optimal) Class B?
Last edited:
- Status
- Not open for further replies.