CHORD ELECTRONICS DAVE
Jan 10, 2017 at 6:06 AM Post #6,736 of 25,850

Dave, have you noticed and improvement in the 1200E since you got the DAVE?
 
I thought it was as good as it was going to get with the BLU and Indigo(QBD76) direct to the amps. I have a small room and when the volume went up, quite high mind, the system could get a bit shouty. This I thought was associated with the amps and their class A/B operation. However, since I got the DAVE this shoutiness has gone for the same volume levels. I now have to be careful as the house foundations are very old…
 
The Chord amps are better than I thought they were.
 
Jan 10, 2017 at 6:34 AM Post #6,737 of 25,850
Dave, have you noticed and improvement in the 1200E since you got the DAVE?

I thought it was as good as it was going to get with the BLU and Indigo(QBD76) direct to the amps. I have a small room and when the volume went up, quite high mind, the system could get a bit shouty. This I thought was associated with the amps and their class A/B operation. However, since I got the DAVE this shoutiness has gone for the same volume levels. I now have to be careful as the house foundations are very old…

The Chord amps are better than I thought they were.


When one considers the price of the SPM 1200 mk2 compared to some of its larger brothers I think this amp is very good value. Certainly it has no trouble showing off the attributes of Dave. The only limit I have found is in the A/B nature if I wind it up crazy loud. (Something I have only done once and only noticed once.) It can as you say lose some of its smoothness. I am looking forward to having the digital signal travel from Dave to the new amps but I have to say, after a year with Dave I am still rediscovering music I Loved but had stopped listening to that I now know still has something to offer. Dave still delivers the magic for me despite the passage of time and the new craving for Blu2. :sweat_smile:Lately I have even been buying old 70's compilation CD's of songs I loved back then but were not greatly recorded and yet even these have something to offer. I am hearing timbre I had never heard in 40 year old recordings. I also listened to Doris Day's 'When I fall in Love' from 1952 and the quality of the vocal recording is breathtaking. Without getting too deep, my day is enhanced by these new revelations and I feel privileged to be honest.
 
Jan 10, 2017 at 7:32 AM Post #6,738 of 25,850
  Actually I have had the half M taps since September on my home theatre. With Blu there is a dither switch for CD - but when you are using the SPDIF this switch controls the delay, and there is a video mode which reduces the latency from 0.66 seconds to 100 mS which is fine for projectors and audio delay set to 0.
 
But over Christmas I discovered a cool feature with JRiver. I play Blu ray discs with JRiver, and you can set the delay up to 2.5 seconds, so the audio is delayed by 2.5 seconds. Then I thought what if it accepted a negative delay? That is delaying the video not the audio? And it works - setting it to -0.56 gave me perfect lip sync! So I can play a blu ray disc with JRiver and the full 1M taps with the M scaler engaged.
 
Rob

and the beauty of j river, it allows the asio selection for audio output with movies which no other media software allows to my knowledge. 
 
Jan 10, 2017 at 7:33 AM Post #6,739 of 25,850
  I am still getting to grips with how good the M scaler is -

 
Hi Rob, when you've fully worked it out :xf_eek:), can I suggest 2 things in the (hopefully not too distant) future:
 
1. A standalone M-Scaler in a smaller, simpler box and a smaller price - for those who don't need the CD transport
 
2. Even better, an upgrade to DAVE to include all the SQ benefits of the M-Scaler, but fitted within the existing DAVE box. I'm guessing this will need h/w as well as s/w changes. I'm sure many existing DAVE owners would be prepared to pay a fair price for such an upgrade.  
 
Jan 10, 2017 at 7:50 AM Post #6,740 of 25,850
Hi Rob, when you've fully worked it out :xf_eek:), can I suggest 2 things in the (hopefully not too distant) future:

1. A standalone M-Scaler in a smaller, simpler box and a smaller price - for those who don't need the CD transport

2. Even better, an upgrade to DAVE to include all the SQ benefits of the M-Scaler, but fitted within the existing DAVE box. I'm guessing this will need h/w as well as s/w changes. I'm sure many existing DAVE owners would be prepared to pay a fair price for such an upgrade.  


I agree with you but I'm concerned about the power requirements of 2. This is the most power intensive, noisy part of the device and to increase the requirements so much... you're probably better off with a new device imho.
 
Jan 10, 2017 at 8:22 AM Post #6,741 of 25,850
 
The USB port is bi-directional, so it can input any sample rate/32 bits and at the same time output any sample rate/32 bits. USB is galvanically isolated. I suppose that via an appropriate app one could then transmit to Dave via USB. But the BNC outputs are galvanically isolated too so that won't represent a SQ degradation as its the galvanic isolation that makes by far the biggest difference.

Rob


Rob are you able to reveal if Hugo 2 USB is galvanically isolated? Thanks.

No it's not as Hugo 2's primary use is on trains, planes and automobiles and currently our USB decoding chip consumes too much power - which with galvanic isolation would mean the USB chip would flatten the source battery too rapidly. It's less of a problem with mobile sources; and I have much improved the RF filtering and isolation on the USB power and ground too, so RF noise can't get into Hugo's ground plane.
 
Rob
 
Jan 10, 2017 at 1:06 PM Post #6,743 of 25,850
Romaz
One thing that I am surprised you didn't focus on when describing the Blu2 improvements is timbre of instruments. Perhaps because there was just so much of everything else! The reason I say this is that I can tell quite a noticeable difference in this area through the Dave when I compare the Red Reference upscaling to 88.2 from 44.1. Particularly noticeable on electric guitar. Instruments that I know well like Fender Strat or Gibson LP. I would expect the same of any instruments you are particularly well acquainted with in an analogue environment. Would be interested to hear your thoughts on this aspect of the sound?

Edit: 'Acoustic Twelve String' is also very noticeable because of its complexity.

The DAVE has always excelled when it comes to timbre accuracy and so the M-scaler did nothing to change that opinion.  One thing that perhaps was more prominent during my brief listen was timbre variation.  The timbre contrasts among similar instruments was much more evident.  This is what I meant by the violins sounding homogenous in the MSB room, this performance difference was very apparent.
 
Jan 10, 2017 at 6:52 PM Post #6,746 of 25,850
What I can't get my head around is why just a sixfold increase in tap length can transform the sound.

Now that you've proven that there is a huge new vista of performance at 1 million taps and that DAVE reveals it so well, I thought I'd pose some questions:

Is DAVE fully revealing what these 1 million taps are doing? It seems unlikely, since you were still hearing large improvements when you got to 350dB noise shaping resolution.

Since DAVE can reveal the benefits of > 164,000 tap filtering, isn't is possible to write software for an ordinary computer that, for example reads a WAV encoded 44.1kHz 16-bit file and produces a 768kHz 24-bit WAV file that's used the WTA algorithm at an arbitrary tap length? Sure, this wouldn't be in real-time, but going to 2 million or 10 million taps to hear what happens when you play this file into DAVE has got to be tempting...

The WTA algorithm isn't using a pure sinc function (since sinc requires infinite taps) and additionally it seems to be "tuned for transients". I'm wondering if there comes a point at which a finite tap-length sinc function becomes identical in its result to a WTA function for the same number of taps? Does the difference between the two functions reduce with increasing tap-length? If so, is there a chance that would coincide with the tap-length that produces no additional gain in quality?

Now playing: Tindersticks - She's Gone
 
Jan 10, 2017 at 7:28 PM Post #6,747 of 25,850
  Actually I have had the half M taps since September on my home theatre. With Blu there is a dither switch for CD - but when you are using the SPDIF this switch controls the delay, and there is a video mode which reduces the latency from 0.66 seconds to 100 mS which is fine for projectors and audio delay set to 0.
 
But over Christmas I discovered a cool feature with JRiver. I play Blu ray discs with JRiver, and you can set the delay up to 2.5 seconds, so the audio is delayed by 2.5 seconds. Then I thought what if it accepted a negative delay? That is delaying the video not the audio? And it works - setting it to -0.56 gave me perfect lip sync! So I can play a blu ray disc with JRiver and the full 1M taps with the M scaler engaged.
 
Rob

 

 
Jan 10, 2017 at 8:27 PM Post #6,749 of 25,850
Now that you've proven that there is a huge new vista of performance at 1 million taps and that DAVE reveals it so well, I thought I'd pose some questions:

Is DAVE fully revealing what these 1 million taps are doing? It seems unlikely, since you were still hearing large improvements when you got to 350dB noise shaping resolution.

Since DAVE can reveal the benefits of > 164,000 tap filtering, isn't is possible to write software for an ordinary computer that, for example reads a WAV encoded 44.1kHz 16-bit file and produces a 768kHz 24-bit WAV file that's used the WTA algorithm at an arbitrary tap length? Sure, this wouldn't be in real-time, but going to 2 million or 10 million taps to hear what happens when you play this file into DAVE has got to be tempting...

The WTA algorithm isn't using a pure sinc function (since sinc requires infinite taps) and additionally it seems to be "tuned for transients". I'm wondering if there comes a point at which a finite tap-length sinc function becomes identical in its result to a WTA function for the same number of taps? Does the difference between the two functions reduce with increasing tap-length? If so, is there a chance that would coincide with the tap-length that produces no additional gain in quality?

Now playing: Tindersticks - She's Gone


Great point / question!


PS

Does anyone now if Jude have abandon Head-fi and moved to Tokyo for good?? :wink:

No Video coverage of all new products on CES ??
 
Jan 11, 2017 at 12:06 AM Post #6,750 of 25,850
That was my thought too, as for me personally I would rather have a slower and less expensive FPGA, it does not need to be real time at all, or this processing done in software that runs on a windows even it if takes one hour per album. 
 
Anyway necessity is the mother of invention. Hope this can be done off line without such an expensive FPGA.
 

Users who are viewing this thread

Back
Top