[Matias]

bsckwan, I removed the player variable: my tests were done with an ASIO soundcard's SPDIF against DAC1 USB in WinXP. Same result though.

 

[03lab]

Quote:

Originally Posted by Crowbar /img/forum/go_quote.gif Do you drive or take the bus or another gasoline or diesel using vehicle? A liter of gasoline is 9,000 watt-hours, 11,000 for diesel. Using a vehicle to go a single kilometer pales in comparison to electronic equipment's standby power for a whole day.

Yes ... and at night, I even turn on the light sometimes.

I like to safe energy nonetheless though and on a global scale it does make a difference IMHO.

Anyway, I was more concerned about sending the DAC1 back for repair than power consumption in standby, as I always disconnect the power cord to my equipment when not in use.

 

[Wavelength]

Quote:

Originally Posted by poo /img/forum/go_quote.gif DAMN! I didn't realise that (or at least forgot). Is that a limitation of USB within the DAC1, or USB in general? Can I ask for a simple explanation as to why this is the case? Something to do with bandwith?

Poo,

Both Windows and MAC OSX have capabilities over 200Kbps with USB 1.1 protocols. The problem is with the TAS1020. It really does not have the buffer space to pull off 176.4/192k. For 24/96 operation the buffer needs to be at least 829.4 bytes (don't ask) for the TAS1020. No single ISO link in the TAS1020 can exceed 1024 bytes. It would require 1658 bytes for 192 which is more than the part has access to.

I am looking at some ARM controllers and other processors that handle both USB 2.0 speeds and have a lot larger buffer space.

I think ounce you get above 96k that you are pushing the host 1.1 ports as you doing about 5mbps. It would of course require 10mbps for 192 and considering how much is spent on the ports looking for a part that handles 480mbps might be the forward thinking way too go.

Thanks
Gordon

 

[EliasGwinn]

Quote:

Originally Posted by yourmando /img/forum/go_quote.gif the other day I noticed some loud hum from the listening position. I use the dac1 usb as my preamp, and it's the only thing connected to my speakers (no cable box or other sources). I finally narrowed down the problem--if I plug the dac1 usb into a different wall socket from the other components, the hum mostly goes away. (Still audible from near the speakers, but I don't mind that). The 2 different wall socket situation is not ideal because I have 2 power strips coming in across different sides of the room.

Hello Armando,

First of all, I would recommend avoiding the transformer at all cost. It will add significant distortion, which will absolutely be audible (assuming your system is high-resolution). I highly recomment the XLR connector with the unconnected pin-3.

The hum problem is strange, because usually ground-loops are worse when the components are plugged into different outlets. It sounds like there is an AC leak into the signal-ground in your amplifier. Are you good with electronics? It would be interesting to see what would happen if you connected the RCA shield to the chassis in your amplifier.

Let me know...
Thanks,
Elias

 

[yourmando]

Hi Elias,

Very interesting. No, I'm not handy with electronics. I'm a software engineer who has never soldered a thing. How would I connect the RCA shield to the chassis?

I'll definitely stay away from the transformer. I'd love to find the root cause anyway. Yup, the system Linkwitz definitely created a highly resolving system, combining a very interesting controlled directivity open baffle dipole design (including dipole woofers) and active crossover/eq...

Thanks again,
Armando

 

[EliasGwinn]

Quote:

Originally Posted by poo /img/forum/go_quote.gif DAMN! I didn't realise that (or at least forgot). Is that a limitation of USB within the DAC1, or USB in general? Can I ask for a simple explanation as to why this is the case? Something to do with bandwith?

96/24 is the upper limit of audio over USB 1.1, due to bandwidth limitations.

We could have went with USB 2.0, but there are no audio-specific USB 2.0 chips. Therefore, we would have to build custom drivers, and that would defeat our goal of making this device a completely 'native' solution.

Thanks,
Elias

 

[EliasGwinn]

Quote:

Originally Posted by erotisches /img/forum/go_quote.gif One question, if my PC is also a source for home theater, can dac1 be used simultaneously with the sound card? I also need digital out for the surround channel whereas ,if all output is fed through dac1, I will only have 2ch output. Digital output would be nice too ahhh I'm so greedy. Sorry if this is answered in FAQ, I admit that I haven't read it.

Can you go into more detail about what you're trying to do?

Thanks,
Elias

 

[EliasGwinn]

Quote:

Originally Posted by bsckwan /img/forum/go_quote.gif 1.) Mac OS X 10.5.3 with iTunes 7.6.2 playing AIFF files --> (via USB with 24 bit / 44.1 kHz setting in Audio Midi and standard USB cable) Benchmark DAC1 PRE Is there anyway to improve on 1 and whether this is usually the case with DAC1 - SPDIF better then USB?

If you are playing 44.1 kHz files, and the volume control is set to maximum, and you've followed the other 'Audio Wiki' recommended settings (no sound-check, no sound-enhancer, etc), then you've got bit-transparent playback.

If you've got bit-transparent playback on BOTH the CD transport and computer, they should sound identical. Don't assume that the CD transport is bit-transparent though...we've seen the opposite quite often.

Thanks,
Elias

 

[EliasGwinn]

Quote:

Originally Posted by 03lab /img/forum/go_quote.gif Anyway, I was more concerned about sending the DAC1 back for repair than power consumption in standby, as I always disconnect the power cord to my equipment when not in use.

Don't worry about this. You won't damage the DAC1 like this.

Thanks,
Elias

 

[EliasGwinn]

Quote:

Originally Posted by Wavelength /img/forum/go_quote.gif Both Windows and MAC OSX have capabilities over 200Kbps with USB 1.1 protocols.

This isn't true. I'm assuming you meant 200 Mbps, but even still, its not true. The highest data rate of USB 1.1 (Full Speed) is 12 Mbps.

Thanks,
Elias

 

[EliasGwinn]

Quote:

Originally Posted by yourmando /img/forum/go_quote.gif Very interesting. No, I'm not handy with electronics. I'm a software engineer who has never soldered a thing. How would I connect the RCA shield to the chassis?

Well, you could test to see if this eliminates the hum by using a scrap of wire and just touching the rca sheild to a bare part of the chassis. If so, then you could try something more permenant.

Thanks,
Elias

 

[G-U-E-S-T]

Quote:

Originally Posted by EliasGwinn /img/forum/go_quote.gif Hello Armando, First of all, I would recommend avoiding the transformer at all cost. It will add significant distortion, which will absolutely be audible (assuming your system is high-resolution)...Elias

Hi Elias,

Armando was asking about the Jensen transformer product, PC-2XR. These are not your ordinary transformers! Did you check-out its specifications? If not, could you please take just a moment to see them with graphs at this link: http://www.jensentransformers.com/datashts/pc2xr.pdf

These particular Jensen input-transformers are designed specifically for the use Armando is describing, and they are very well known for being transparent. They might be the best of their kind in the industry. I've also personally used them in hi-resolution systems and can vouch for their quality. So please forgive my double-questioning of your information, as I mean no disrespect - but are you absolutely sure about the accuracy of the advice you've given Armando here, about this particular Jensen transformer product?

 

[Crowbar]

Looking at the transformer specs, the phase distortion is not too bad, but the THD is unacceptable, especially at low frequencies. This is nothing close to transparent. Note also that they don't present a measurement of the hysteresis, which is a form of distortion specific to cored inductors. No surprise--why show even more of the distortion your equipment does in your marketing...

 

[G-U-E-S-T]

Quote:

Originally Posted by Crowbar /img/forum/go_quote.gif Looking at the transformer specs, the phase distortion is not too bad, but the THD is unacceptable, especially at low frequencies. This is nothing close to transparent. Note also that they don't present a measurement of the hysteresis, which is a form of distortion specific to cored inductors...

Typical THD is 0.015% at 20Hz and less than 0.001% at 1kHz (+4dBu levels). Can this possibly be audible? Plus the transformers actually decouple and remove existing ground-noise distortion otherwise riding on the audio signal through an unbalanced interconnect cable.

Also I've never heard of Jensen transformers suffering any problems whatsoever with saturation (hysteresis). Have you? They are used industry-wide in critical applications (recording, live, etc) and have a solidly established reputation for their quality.

Sorry if these are dumb questions/statements - I'm definitely not trying to be argumentative, just wanting to hash out this information for everybody's sake (mine included). What am I missing here? I'm ready to learn - thanks in advance for any replies...

 

[Wavelength]

Quote:

Originally Posted by EliasGwinn /img/forum/go_quote.gif This isn't true. I'm assuming you meant 200 Mbps, but even still, its not true. The highest data rate of USB 1.1 (Full Speed) is 12 Mbps. Thanks, Elias

Elias,

Sorry I meant >200K like above 192k the sampling rate. The USB 1.1 spec does not declare a maximum sampling speed. The only problem here is that the TAS1020 cannot do more than 24/96k as the buffer size for the endpoint cannot exceed 1024. Since the endpoint data in the TAS1020 is only 1304 of which much of that is used for the ROM code leaving only 1176 available for ISO data it does not make it capable of either 176.4/192k.

There are many capable Codec oriented ARM controllers capable of the higher rates. Centrance as did I decided that allot of the code was available with the Reference Code give by TI for the TAS1020. This gives you a boost up on getting your feet wet.

Many companies are now going to offer source code software to support the higher rates with much larger buffers.

Thanks
Gordon