kjbii

Thanks again, Keith, for your instant response. Considering the complexity of Acourate, I will start with the simple 2-channel system described above, using speakers that do not have a time alignment problem. I'm at the beginning of a long learning curve, and am happy to benefit from your experience. If you started with a Marchand, you have been there before me.

Hi Michael, I'm still digging through your old posts. Did you decide to abandon this alternative? Is Dirac an option, for delay and crossover alignment? thanks, Ken

Mark, thanks for your comments, no worries about my thread. I'm looking for the same solution that you describe, ethernet input and the ability to use different DACs in a biamped system with the advantages of Acourate correction and the possibility of adding two or four channels at some later time. The choice of DAC for the bass drivers is not important, it is the requirement of keeping my mid/upper range DAC that is my real concern. Please post your experience with the Hapi, and any further refinements in your plan based on reflection or hard empirical experience. I'm a complete neophyte, and greatly appreciate Keith's and your comments and corrections to rule out dead ends at the planning stage. It's helpful to know that systems 2 and 3 are not useful options, and I will read more about the Hapi and RME HDSPe card system. With luck, you both will have explored these alternatives by the time that I get there. thanks and regards, Ken

My chain of understanding of Acourate is clearly missing a few links, and it will be helpful to reduce the basic calibration and playback setup options to the simplest possible 2-channel system, with passive crossovers in the speakers. In this system the Klipschorn bass driver has a delay of about 13 ms compared with the mid-range driver. This is my primary reason for considering Acourate. As I understand from the discussion above, the Tascam USB-66 has a microphone preamp, and connects by USB to a PC, and to the DAC by SPDIF. Acourate generates test tones/sweeps that are sent to the speakers, and the speaker output is sent back to Acourate through the microphone. Acourate then generates FIR filters that are used by JRMedia for convolution that corrects the output. Once the Acourate filters are installed in JRM, the Tascam can be removed from the system - it is used in the calibration step only. Will this calibration time-align the output that is sent through the DAC to the speakers? This may be the same question that you asked earlier, "Would I be able to set a partial (low frequency) target curve along with the crossover/delay in acourate?" (6/22/13). I didn't find an answer to your question, but understand that Acourate can provide "time domain" correction. Uli's 2006 "Thoughts about crossovers" seems to contemplate this correction. files.computeraudiophile.com/2013/1202/XOWhitePaper.pdf Any comment or advice you have would be most helpful, since you seem to have started with the same questions and goals. Ken

Hi Keith, this provides extremely useful and detailed answers to most of my questions, and leads back to the clocking/synchronization question. I can’t wait for your report on the RME HDSP card driving 4 DACs. At this point, I would settle for two. I am wed to the Directstream DSD DAC as my core component. It can be configured to accept various digital inputs, including USB and Ethernet. I prefer the Ethernet connection to avoid myriad USB issues, as well as to provide server and control capability, but can use any of the other inputs including XLR (AES/EBU). The DAC upsamples data from all sources to 10x DSD rate, buffers the incoming data, and after processing it is downsampled to 2x DSD rate and output through a passively filtered analog section rather than an active analog output stage. This does not mean much to me, but it may be significant for integration/synchronization with other components such as a Hilo/Hapi/RME HDSPe AES32. The DAC “handles the PCM conversion from AES/EBU, S/PDIF, TOSLINK, I2S and USB without recovering a clock, by simply watching for the edges and making decisions about what they mean in context. The output clock’s rate only depends on the long term average rate of the inputs not on any edge or other local feature.” Maybe this explains the lack of a clock input. I would also appreciate any insights that Dallas has on this question. My goal is to biamp/triamp two (or three) horns, crossing over from the bass bin to a midrange driver at 440 Hz. I have been using an ESS AMT as the high range driver in a triamped configuration, crossing over at 2200 Hz, but think that it is superfluous, in view of the very low tweeter volume at my typical (70 dB) listening levels. Crossing over at 1700/1800 does not improve speaker performance, despite (or due to?) the colorless and instantaneous response of the AMT. Instead, I will try the BMS4592nd passive crossover, using the integrated second high frequency driver. Its coaxial orientation should provide time alignment with the midrange driver, as well as vertical and horizontal linearity. Apart from other advantages of DSP upstream or downstream (DEQX), time alignment of the bass bin and midrange horns is critical. A delay of about 13 ms in the mid-range is about right. One possibility (system 3) is to use an analog active crossover, and a 2-channel AD/DA delay filter, rather than a digital crossover, if such a device exists. Biamping and time alignment provide an image that is almost holographic, even using an inexpensive and noisy breadboard Behringer crossover. System 1 and System 2 describe my limited understanding of a software-implemented upstream DSP crossover, and a downstream DEQX crossover, converting the “perfect” DAC two-channel output to a digital signal fed to the DEQX. It’s a lot to pay for a solution that may significantly degrade the quality of the signal. I don’t know what to expect. Thanks for your suggestion of the RME HSDPe card, which is a relatively inexpensive alternative that would surely drive my essential DAC. It will take some time to go over and digest your extensive comments, which have saved me endless frustration and at least $16000 to date. I'll be back when I start setting up Acourate. many thanks, Ken

but thanks, and OUCH! That's quite a bite, the amps must make up the difference in $100K. What horns are you driving?

Keith, thanks for your comments - I didn't know that Acourate could generate crossovers, and have months to spend tinkering with tuning a modular system. I've gone back through your other posts, and am at the same spot you started - what modules do I need, and how do the BYO DACS interface with the PC/Acourate/HQP/JRM? What is necessary for making the measurements for Acourate? Did you find a way to use the high-end DACs you initially had in mind, in a parallel configuration? If the only solution is a multichannel DAC or A/D D/A (Hilo, Hapi) this is a significant restriction. Can the mid-range from the crossover can be passed through a Hilo/Hapi to a high-end downstream DAC? Dallas had comments on clocking and drift that I could not understand, but they sounded like serious reservations. I'm on the verge of a major DAC upgrade, but would reconsider if it can't be used in a software-implemented active crossover system. That would be disastrously wrong. thanks, Ken

Keith, thanks for your informative posts, and clear comments on the competing advantages of the DEQX and higher quality DACS. I need an active crossover to time-align the bass horns in a biamped (or triamped) system, and the optimum delay is large, about 13 ms. This time alignment alone does make a massive difference in imaging, even when using a basic digital crossover and primitive DSP adjustments. Most of my listening is from web sources, using a Berkeley Alpha interface to reduce jitter and a PS Audio DAC, with the advantages you have described. It seems evident that inserting the DEQX downstream in this chain would merely sacrifice these advantages, and that inserting an ADC wouldn't change the limitations imposed by the DEQX DACs. The result could be better just by eliminating the intermediate DAC and feeding the signal from the interface to the DEQX digital inputs. Is there a configuration that could win more and lose less? Since time and phase adjustment are my primary goals, it might be feasible to feed one delayed digital output from the DEQX to my external DAC and the mid/high amp, and to feed an analog output from the DEQX to the bass amp, where I'm less concerned with nuances of clarity and tone quality. Latency in the external DAC should not be an issue. Or there might be a simpler way to time-correct one channel from an analog crossover. I would appreciate your comments on any of these approaches, and of course hope to benefit from the experience of other members of the forum who have considered similar issues. Biamping and time coherence are critical, and there must be better solutions. thanks to all, Ken