Originally Posted by ss9001
due to limits on DSP power in consumer AVR & prepros, they have to downsample hi-rez audio formats to 48kHz.
this means all 88.2, 96, 192, 176.4 formats are downgraded to 48kHz/24 bit. even Anthem's receivers with ARC are limited the same way. this means all the upsampling those owners get from AL32 & similar is essentially wasted as soon as Audyssey is turned on - those companies' dirty little secret.
Audyssey says "not our fault" because they maintain the limitation is not in their software, as long as the receiver/prepro mfgs put in sufficient DSP's, which they don't to remain competitive. They maintain it's not their fault that no company wants to pay for the DSP chips it would take. what's remarkable is that the PC based Audyssey Pro software has the same limitation because it's still limited by the DSP's onboard the receiver/prepro. even though a PC is used to calculate the filters, they are still implemented & limited by the DSP's in the receivers. I find this troubling considering Audyssey charges $700+ for the software, per unit license & the improved mic. Even for $700 more, Audyssey's approach won't get you around the limitation. one has to ask, what does? to me, that's being disingenuous, since there is NO Audyssey prepro for any price that has the processing needed. so there is NO real world Audyssey implementation without downsampling. what's also remarkable is that there is not more uproar about this from owners. If Pioneer did something like this, I suspect Pioneer owners would be shouting all over this forum they've been cheated; I've seen posts threatening to sue Pioneer over i-phone app upgrades!
I'm hopeful Pioneer will come out with a new version in the near future to include sub EQ and improve the EQ part, maybe with parametric or some form of time domain filters. But I would NOT want to do ALL of this at the expense of downsampling SACD's, DVD-A's, 96KHz BD's and hi-rez audio files.
take Trinnov - it's a fantastic room correction, time domain system! but the true full resolution, full processing version cost about $15000 and even that was in an ext box, used a PC & required A-D-A conversions. if you want it built-in, done internally in the digital domain, it'll cost you $40000 from ADA. Dirac is another one, in a $25000 processor.
it's about compromises or what you're willing to pay not
to have them
Not sure that this response really belongs in the new Pioneer SC-7x thread, but I'll point out the following for those so inclined - from the POV of a former Pioneer user and a current Audyssey Pro user..
First of all, you're absolutely right that it's all about what you DON'T want to pay for. The truth is that AVR DSPs are well behind PCs in terms of potential processing power. But even if you've got 24/192 capable DACs, unless you're an analog purist, so what? As per Chris K. in a response quoted on the Audyssey thread this week (see the thread starting with http://www.avsforum.com/t/1430049/the-official-denon-avr-4520ci-thread/5190#post_23561433
), 48kHz sampling reproduces up to 22,000 Hz, well beyond the range of the most sensitive human hearing (20K), and thus having the extra processing power is pointless because the results aren't audible. So even if you're running MCACC and there's no down-conversion, how you tell if 24/96 or 24/192, let alone 32-bit processing matters vs. 24/48 in the audible range? Further, can the measurement tools that I'm aware for accessing actual impacts on room acoustics beyond the predicted effects from RC even measure 96 KHz? With the "gold standard" of REW, for example, the limitation on sampling frequency in the software is 48Khz.
Further, if you want full digital control and want an electronic solution beyond the existing room correction system in your current AVR, you have to spring for a Datasat RS20i or ADA system at the $15K or $40K level. Or if you're willing to live with A-D-A conversion, you can do the $15K Trinnov, wait for the alleged DL2/3 coming out as a standalone Dirac Live runtime box (as per Carl Huff, the developer, that's going to be in the $2K-$3K range, but it's been "45 to 60 days" away since 2011), or go the route of using a dedicated PC to run Dirac Live, and use something like a Lynx AES sound card as a bridge between your dedicated pre/pro and an amp. Which means separates, whether it's an actual pre/pro and amp, or two AVRs doing that duty. Either way, there's a lot of A-D-A conversion going on. How much it matters, once you've applying digital correction, is up to you.
Speaking of Dirac, it's only 8 channel processing, which has its own limitations if you're moving into front heights or wides, and even though the literature suggests it can do 8 channel/96 KHz processing on their website, you run into the same issues with accessment. BTW I see that Emotiva is releasing a XMC-1 with Dirac Live, but at an approximately $1500 price point, and their issues with TACT and implementing Trinnov, it's caveat emptor (buyer beware), and may take you back to the 48 KHz sampling limitation.
For all of this, I don't see anyone on AVS doing any A/B/X studies comparing MCACC to Trinnov, or Dirac to know if the time domain corrections really make a difference beyond theoretically. Dirac's gotten high praise for having a unique filter solution to work with both frequency and impulse response within the software, but those on the $20K+ thread that have Datasats or Trinnovs aren't measuring or telling, and the only other users of Dirac are folks with HTPCs that seem to be in their own world.
Ultimately it's a tradeoff issue - is spending $3K, at a bare minimum for the DL2/3 or building a HTPC for the sole purpose of processing Dirac, worth the hassle for something that "may" matter? I happen to have a spare AVR (SC-27) that I could use to power a Denon 4311-Dirac system, but at the end of the day, putting some of that money into room treatments is probably a better bet unless I'm feeling adventurous. Which being a tweaker I might...
To bring this remotely back onto Pioneer: what are the thoughts about the degree to which Class D power matters if you've got very efficient 8 ohm speakers? I've got Mythos STs, so for those of you that have had both Denon and Pioneer systems, and is there any real disadvantages to using, say, 140 watts @ 0.05% distortion with a traditional Class A/B amp vs. 140 watts @ 0.09% distortion for the Pio's Class Ds, all things equal (e.g. I suppose that means no room correction)?
I have no plans to upgrade to a SC-7x since I've still got the SC-27 if I want to have a cost-free "amp" if I look into Dirac, but what the thoughts about the amps in the older systems like your Susano (or an SC-27/37) vs. the newer SC-7x for more conventional speakers than, say, Magapans?
Parenthetically, if Pioneer eventually _does_ have a system like Dirac, I'd seriously considering it, as the only reason I moved to Denon in the first place two years ago was XT32+Audyssey Pro. Which as per the more advanced guys on the REW/Pro side, isn't nearly as "perfect" as some used to make it out to be, due to poor resolution Curve Editor control, a dual subwoofer distance bug, and what amounts to independent channel filter generation vs. more of a group/interaction approach. And most importantly, the "time domain" effects only work if the system's in minimum phase IIRC, which means in practice it's incidental to whatever XT32 is doing with frequency domain correction, So there's room for Pioneer to hit a "home run" if R&D eventually moves that way.