Thanks.
You also need to make sure to test clean versus wide on T+A settings - I preferred clean (piano upper register more natural to me).
Thanks.
You also need to make sure to test clean versus wide on T+A settings - I preferred clean (piano upper register more natural to me).
Ask Marc Coles at Sablon - think he has used it extensively with hifi man he 1000's @ home.
Sorry, don't know who Marc Coles is. I was just wondering if any of you guys tried this DAC with headphones.
He is a member of this forum: Sablon Audio (also happens to be dealer for SGM and Lampi) http://www.whatsbestforum.com/member.php?6908-Sablon-Audio
How good is the headphone amp in this DAC?
You also need to make sure to test clean versus wide on T+A settings - I preferred clean (piano upper register more natural to me).
I will be getting a DSD Dac 8 for a while as I am building a win 10 pro based system friend who will use this dac in a mega buck system to play all music at DSD512. I noted in the manual that there appears to be 2 windows drivers for this dac, so to play primarily, and this will probably be almost 100%, DSD512 what drivers does this Dac need, Just the ASIO or both? Also I have a full level 1 (Ric Schultz) modded Gustard X20u with the addition of a USB to I2S card that has the xmos U216 chip which handles DSD512 and I have also soldered in a OCXO 100 MHz clock much closer to the dual mono ESS9018 chips which significantly improves on the Gustards stock clock. I will obviously be comparing this modest, under $1500 dac with all mods and external PSU's, to the T&A. I was blown away by DSD512 when I finally got it to work smoothly. Like reported here the difference between DSD256 and 512 is not subtle the music becomes much more alive, the noise floor drops and the leading edges are much better defined giving you the you are there dynamics both micro and macro. Separation and that holographic 3D stage is the best I have heard, I am addicted.
So below are my comments on the T&A Dac 8 DSD vs. my very heavily modified Gustard X20u. I apologize for copying my comments from another forum but it is so long I hate to re-type.
I have (had) a T&A Dac 8 DSD in house. I was building a audio PC for a friend to incorporate, Roon, HQ player, Tidal and Audio Fidelizger so he could move into audio streaming after his very old Meridian Sooloos bit the dust. Well I had the Dac for approximately 20 days and ran it a little over 200 hours with music signal which 99.5% was at DSD512. Of course I compared it to my very heavily modified Gustard which has full Level 1 mod, a DIYinHK DSD512 USB to I2S card and a Abracon OCXO 100mHz clock soldered at the coax tower with a Acko regulator in front of it to provide clean power. Both the DSD card and Acko regulators are powered from external PSU's. My dac is fully DSD512 capable and is very stable at that resolution. As far as I know I am the only one who is getting DSD512 from this dac, at least in the forums I follow.
DSD512 is where the magic happens, going from DSD128-DSD256 is a subtle but nice jump, going from 256 to 512 is strangely a huge jump, one that I have no real technical explanation for. Common sense says you have a music file which is, most likely, 16/44.1 redbook resolution and you take this file and with the magic of HQ Player upsample it many, many times to DSD512. How can that improve it? If I have a photo at say 720P resolution and I suddenly increase (upsample) it to 4K resolution wouldn't the pixels now be more obvious, wouldn't it look worse? Strangely that does not happen when up sampling music and 512 is just more real, more being at the event than listening to a recording of the event. Background noise is lower, way lower (not measured) transparency is then obviously greater, leading edges better defined, bass becomes more focused and impactful, separation of the performers is greater and more 3D. Some very good recordings have made me gasp at how real the voice, guitar or violin can be. Almost eerie. I love it and I am never going back.
On my very first listen to the T&A at 512, it was not apples to apples (to the X20). The dac had a cheap power cord, and a stock USB cable, I did not have Roon or Audio Fidelizer pro on it. The dac also had just a few hours on it and it sounded thin and un-involving. As the hours on it increased and my work to make the comparison as close as possible continued I was finally ready to compare. Both dacs had the same power cable, the T&A had a slightly better USB cable from the same manufacture. Both Dacs had Roon, HQ Player and Audio Fidelizger Pro. Each dac was run from separate PC's but both PC's used Gigabyte MB's with Dac-UP USB out, seasonic PSU's and Win10 pro and the same power cords plugged into the same conditioner.. The Gustard was its usual very involving self, I just love the way it plays back music at DSD512. The T&A became a refinement of the Gustard they were close and I'll bet a lot of untrained folks would struggle to hear the difference. It had a bit quieter background, transparency was up a notch, leading edges had a bit more definition bass had more detail. where the pluck of a double bass string the resonance of it in wood body was more evident that with the Gustard, thou it may not of had the weight (close tho) that the X20 has. Also the separation and 3D placement of musicians was a bit more obvious. It also felt just a bit more real than the Gustard, less recorded more there. Now these are just subtle differences that I bet a lot of folks would not be aware of, but differences non-the-less that audiophiles pay huge $ for those type of gains. On another night I swapped the better USB cable on the T&A with the next one down the line in the Gustard, the gap closed slightly but I still give the nod to the T&A. I suspect if I decide to pop for the $400+ Pulsar OCXO clock the gap would almost vanish. For $4K the T&A may be the best sounding DSD512 Dac available and it is 44k and 48k compatible unlike the gustard which is only 44K. What that means is a 48K based files (24/48,96,192) is up sampled to 24.6 Mhz and a 44.1 file is up sampled to 22.6 mHz so in HQ Player you can check auto rate family which will lessen the CPU/GPU load somewhat. The Gustard will upsamaple 48K based files to 22.6, which increases CPU load as it essentially has to down sample from 24.6 to 22.6 mHz.
Tuesday I packed the T&A up to send to my buddy and next week I fly to install his new PC and T&A and it goes into a state of the art analog system with a modified struder reel to reel with very high quality tapes, and a SME30MK2 TT with top of the line dynavector cartridge. So we will see where the T&A at DSD512 fits into that picture. Last night I had to listen to just my Gustard and was not disappointed, it still is very involving, very transparent and very real like. Like I said they are close but ultimately if I had to choose just one cost not a consideration I pick the T&A. The consolation prize is not bad tho.
Not sure what forums you follow but I have been getting 512 without a hitch for months now.
Please tell us exactly how you are getting DSD512 from your Gustard. Thank you.
Please tell us exactly how you are getting DSD512 from your Gustard. Thank you.
@Priaptor yea, like cheating with T&A, just plug and play with HQP and magic DSD512. Gustard a bit more work.
Ric pretty sure you know what I did, paragraph 2 of my post goes over the item needed (DIYinHK DSD card, 768K they call it). There is a schematic to the USB I2S lines in Post 933 in the X20 thread on Head fi. Connect the Card to the I2S lines and add power. The Clock is just far superior to the coax cable as you know and the Abracon is superior to the Crystek.
Steve Williams Site Founder | Site Owner | Administrator | Ron Resnick Site Co-Owner | Administrator | Julian (The Fixer) Website Build | Marketing Managersing |