TOPIC:

Speaking of running a pre-amp into another pre-amp. From the manual:

If audio playback is distorted, you may have too much gain internal to the DSP. Dirac Live can 
apply up to 10 dB of gain, so the output level should be kept lower than -10 dB to guarantee that there is no 
distortion with a full- scale input signal.

This issue typically occurs when volume control is being done downstream of the SHD Series 
processor. If so, set the SHD Series processor at -10 dB, and then use the downstream equipment to control volume.

Isn't -10 dB pretty loud? I think I usually run my system in the high -20's dB range.

-20db relates to a Voltage out from the SHD. The Vout will be effected by whether you're using RCA or XLR, and gains, filters, PEQs etc that have been applied. How loud that output then is depends predominantly on the gain of your amp(s) and the sensitivity of your speakers but in addition, how close your speakers are to room boundaries, how far away your listening position is and even whereabouts the sound engineer set the level of whatever you're listening to will all have influence.

Depending on my mood, when I'm listening and what I'm listening too, my range is between -36db and -16db. I believe typically below -48db you'll start losing resolution when listening to 16bit media.

Kind of curious about the best way to handle the volume. In a perfect world, I'd like to get a Mc tube pre-amp to play with, but they're seriously expensive - I'm thinking of trying something cheaper like a Schiit Freya+ to see if this is a Kool-Aid that I want to keep drinking. Also, the Freya+ has a passive mode so I can effectively resort back to my normal non-tube setup at the press of a button instead of fiddling with cables.

Assuming that I would need to have stream -> SHD -> Pre (via XLR) -> Amp (via XLR).

Would the best bet be to set the Pre volume to max (no attenuation?) and then just control volume completely from the SHD? Or does having the SHD before it in the chain mess up that plan?

The main problem with SHD before preamp is if you want to use subs, needing to use three or four SHD outputs.

Outputs 1&2 to the pre, 3&4 to the subs. In that scenario though the SHD MUST control volume so pre-amp would need to be fixed volume. I'd ask the OEM where the unity gain point was. But any accidental volume adjustment on the pre then messes the balance up. Begs the question if using subs doesn't it, why bother with a pre-amp?

entripy wrote:

The main problem with SHD before preamp is if you want to use subs, needing to use three or four SHD outputs.

Currently, I have my SHD sending 2 XLR to the amp and 2 XLR to the sub. I have high and low pass filters set on the corresponding XLRs. I was thinking that I could redirect the 2 amp ones to the pre then from the pre to the amp.

The only weirdness is this means some of the audio is going through tubes and some is staying full solid state. Although, ironically, it might mean I can then have the tube sound without sacrificing bass slam... ha ha

In a more perfect world, I'd just run the source to the pre to the SHD to the amp/subs... but to pull that off I need a new streamer (and maybe DAC) or I need to get a better tube pre. The ones I was toying with in my mind are 100% analog pre's.

asx77 wrote:

Outputs 1&2 to the pre, 3&4 to the subs. In that scenario though the SHD MUST control volume so pre-amp would need to be fixed volume. I'd ask the OEM where the unity gain point was. But any accidental volume adjustment on the pre then messes the balance up. Begs the question if using subs doesn't it, why bother with a pre-amp?

If I'm understanding volume control correctly, I *think* that the knob only adds attenuation - so it can only reduce a signal and not increase it. If that's the case, as long as one is set to max, I think the other would be setting the level.

If I'm doing PRE -> SHD then I think either can set the volume as long as the other is at max.
If I'm doing SHD -> PRE then the SHD needs to set the volume with the PRE at max since the pre doesn't control the sub.

Might be a little of an oversimplification because the pre's & SHD likely add gain (for sure the tube pre does... around 14 dB I think) so whichever device is set to max, I think it may need to really be max minus whatever that gain amount is.

And you're right - once that's done no one should ever touch the knob that's at "max".

Dirac recommend their software sits in the chain before any component that 'colours' the sound e.g. before a non-transparent DAC or a valve pre-amp. Otherwise they say you are risking loosing the colour due to the additional AD-DAC.

Don't know about the pre you have in mind but most pre amps will attenuated and boost. Irrelevant anyway. With SHD before the pre I'd set the SHD to -10db, have your sub(s) off and play a modern album that's mixed hot, turn the pre up to the max level you could ever wish to listen and then up it a bit more, thArts it's position,  next, reducing the SHD so you're at a typical listening level, turn the sub's on and adjust the gains until about level with the mains. Then set your delays, filters and/or crossovers etc and run dirac. 

Edited my previous post as I wrote the opposite of what I meant to! It's correct now. 

Interesting, do you know where they say that? I've been trying to find information on Dirac and colorization but haven't had much luck finding much other than a article that says in the next few years they'd like to have Dirac for tube setups (I wonder if that's my sign that what I'm thinking is a bad idea?).

Placing Dirac first in the chain does seem to best fit what I'm trying to do - and also allows me to continue to make use of the SHD as streamer and DAC. Also potentially save some money on a tube preamp since I could then get one without an onboard DAC.

Another option is that I can just run Dirac at  something like 300 Hz and under to tame the bass modes in the room. That's where I notice the most difference when I turn Dirac on and off.

I saw a video on YouTube a while ago where Dirac's Matthias Johansson was being interviewed.

The interviewer was saying how they didn't want to loose the sound of their DAC by using Dirac and Matthias explained that the colour could be kept by using Dirac while the music was still in the digital domain, so to place it before the DAC.

At some point in the same video he also mentions that Dirac and Valves can work fine together because of the non-linearity of the distortion.

If I get a chance I'll see if I can find and link the vid.

Much appreciated. It sounds like by letting Dirac do its work first, you can then try to put the goodness back in afterwards. 

Interesting that this is the opposite approach of the McIntosh and Lygdorf systems... but I think it was mentioned earlier in the thread that the RoomPerfect system tried to keep the colorization. I wonder if they needed to do that because they're stuck in the middle between the preamp and the amp. Another advantage to doing what you mentioned with the SHD being before the preamp, since the work can be done while things are still in the digital domain, it skips an extra round of A to D then D to A conversion which the processor-in-the-middle approach needs.

Hi, We've used the term colour frequently in this thread but I think it’s important to distinguish the types we're talking about. I kind of think of it as there being colour which is designed into a product and colour which is a consequence of design.

Colour that is designed in would be the particular response of a loudspeaker, by that I mean the shape of its curve in an anechoic chamber.Colour that is a consequence of design would be distortion which can be either pleasant or unpleasant.

When @entripy explained that RP keeps colour he was referring to the colour that is designed into the speaker i.e. its frequency response not any mechanical distortion that the speakers produce as a consequnce of their design.

I'm not sure what you mean by your first comment that "you can put the goodness back afterwards" as if we're talking about pleasant distortion, then as we've previously discussed Dirac doesnt remove it in the first place as its non-linear. If you mean Dirac flattens designed in colour and allows you to choose your own response through the use of a curve hence putting "goodness back" then we're all in agreement.

Found the video: 
Where to insert Dirac in the chain is discussed ~48mins in.

Thanks, that was a good video to watch.

Per the video, it sounds like the reason for putting Dirac ahead of the DAC was to avoid the second round of AD-DA conversion. The thought being if you liked your DAC, you would want it on the end of the digital processing chain to keep things in the digital domain and let the DAC do the 1 and only DA conversion at the end.

Thanks for the clarification, it does sound like I was getting my colors confused. I think the fixed frequency response (vs house speaker curve) is what I do want. I want to experiment with tubes - but it sounds like those are basically distortion machines (I'm assuming no affect on frequency response?) so it's sounding like Dirac shouldn't mess with that plan if I choose to take a sip of that Kool-aid.

eisenb11 wrote:

I want to experiment with tubes - but it sounds like those are basically distortion machines (I'm assuming no affect on frequency response?)
 

Well that all depends on the design. A well designed valve amp will keep distortion to a minimum and the distortion it does produce will be mainly 2nd order. The problem is there are a lot of brands out there that I suspect are producing too much distortion and the wrong kind to boot. Good designs are lightning fast and sound natural. Now part 2, "do they affect frequency response?" In actual fact yes because they have quite low damping factors which means their output is effected by the speakers impedance changes. I stumbled lucky with my speakers tbh as they have a very flat impedance curve with only a narrow peak around 80Hz so partner really well with valves.