Discussion in 'Digital: DACs, USB converters, decrapifiers' started by purr1n, Nov 11, 2020.
SPL TwinTube! That brings back memories. I should reinstall my DSP plug-ins again, thanks for that
The filter coefficients is increased form 32 to 35 bits, one more bit was used for headroom as I needed to increase the filter gain a little, to match output level to 2V RMS at 0 dB. There is still around 0.4 dB margin so 0 dB digital in is -0.4 dB out to the resistor networks.... I just did a little testing to verify, when the dac2541 is running at +0.43 dB (in at -0.57 dB, volume at +1 dB), then the FFT looks like what purr1n get, tiny clipping.... So the dac2541 was definite feed more than 0 dB, although it's tiny margins....
Btw, the red clipping LED on the dac2541 is very precise, it do light up with tiny clipping....
Thank-you, I am interested in both auditioning and measuring the DAC2541.
@soekris Mod House Audio does not list dac2541, and shows all other Soekris products as "sold out." Will there be a US distributor?
Mod House is still my US reseller. the older DACS are all sold out, but he has been a little busy lately, should update his website soon....
I asked Mod House and they said it will be available to order “in the next week or so”.
Will the Soekris DAC sound different than Gungnir Multibit and closer to older PCM 63 R2R DAC? Surprisingly I find the older Gungnir Mulitibit to sound quite warm, maybe a bit too much for my taste.
The Soekris sounds closer to the Gungnir A2 multibit, not the A1 which was warmer. As far as PCM63 sound, that depends upon the DAC using the PCM63. I'd say that the Soekris is a more modern taken on the R2R sound. If you want organic, bloom, and dense tone, then probably best to look elsewhere. If you want a modern neutral sounding DAC with clear and textured lows, but with R2R sensibilities of fluidity and no hashy crap on the top end, then the Soekris is it.
Thank you for the tip, @purr1n. The PCM 63 DAC that I tried was a Parasound 1100HD but it has been heavily modified with a fully discrete analogue stage. Great tone and very balanced sounding, I think the Gungnir Multibit is more resolving and detailed but does not sound as natural.
Any idea how the Soekris holds up in layering? That's what I liked most with the holo spring/spring2 from auditioning. If the Soekris is no slouch in this regard that would definitely push me over the edge
Any opinion on the digital preamp section of DAC2541? I tend to stay away from using the digital volume control of certain DACs (by setting the volume at 0 dB, zero attenuation or bypass line level) and instead use the analog volume control of the headamp, preamp or integrated amp. But then again, I could be prejudiced here.
Does the DAC2541 perform wonderfully not only as a DAC alone, but also as a DAC/digital pre? If so, I could happily buy a power amp (maybe dual Aegir) instead of a preamp or integrated amp. However, if the digital preamp is as mediocre as the built-in headphone amp, then I guess I will have to look for a preamp or integrated amp after all.
Layering and separation didn't blow me away like with Wavelight/dream or Holo (at least in the width and NOS sense). While the Soekris managed a sense of openness while not unnaturally emphasizing the top octave, nothing else stood out. On the hand, I didn't sense any deficiencies either such as an overly compact headstage. One takeaway is no hole in the center effect. This has gone to @brencho and then @ChaChaRealSmooth for the loaner. I wish I more had more time, but I didn't want to hold on to this another week while I went on vacation.
DAC2541 is in, been warming up for about 12 hours, so I'll probably give it til Sunday before doing much critical listening.
Of course last night I took the opportunity to do casual listening and comparison. I can say this much: well done, @soekris. With the same audio running through both, improvements are evident even cold. The kind you can hear with quick attentive listening but more importantly the kind that make themselves evident without having to listen. As in noticing randomly "wait that doesn't sound as good as I was hearing, which DAC am I on? Oh, it's DAC1541" and switch, and "Aaaah, yes, there it is!". The same house sound, but like it's been working out and is clearly more athletic and better defined.
*** Important Note, these differences are dramatically reduced after warmup and level matching. See review below https://www.superbestaudiofriends.o...iew-and-measurements.10137/page-5#post-326918
Setup is Surface Pro/JRiver with linked Zones for DAC1541 and DAC2541 using WASAPI because the ASIO Driver only seems to let me address one dac at a time. Connection is USB out to both DACs from Surface Dock. From the DACS into a GoldPoint Selector, into a GoldPoint 47-step Level control and from there balanced out to the Genelecs. I have adapters from SE to BAL so I can compare BAL to BAL and SE to SE.
Heaphones are a convenience case for me, but I will compare using HD600 SE and BAL, but also taking the SE headphone outs to the Genelecs just because it's easy to do.
Review to follow after the weekend sometime.
Shitty phone pic proof:
Is the faceplate blacker? Or is it just the camera/lighting?
The faceplate is blacker, yes. A little misalignment of print as well that might irritate some OCD not worth making a big deal about IMO. I'll take better pics for the review.
I've used DAC1541 both with the GoldPoint and without using digital attenuation. I do adjust the input sensitivity on my monitors so I am not attenuating more than necessary in the digital realm, but I have not noticed a difference either way. I will do a more extreme test to see if/when I can hear an issue.
An option if you'd need to attenuate more than makes you comfortable would be to use something like the Naiant inline attenuators - $39 for a fixed value one (XLR) - in order to optimize your gain structure.
I must say it is nice to have such a small unit on my desk. I was relieved to put the GoldPoint in a drawer and declutter. FWIW I bought the GoldPoint setup just so I could have a relatively good quality way to switch gear and test stuff, and to see if I could hear a difference. I do like to play and see for myself but I'm no golden ear, and I find it hard to obsess on minutae at that level. If I have to wonder if I can hear it, I probably can't and that's good enough for me. I am using USB after all. I will get a Pi2AES at some point, but nervosa isn't driving that either.
Anecdotally Torq insisted he could hear a difference between 0dB and -1dB in his review of DAC1541 because the attenuation logic was active. I found it somewhat amusing that 0 is -1 on that unit and so he had the digital attenuation in place anyway.
The orignal frontplate for the dac1541 was manufactured in Denmark, but cost a fortune.... Now I have them made in China, actually with better quality, with laser cut text instead of screen printed. And they're way cheaper....
DAC2541 vs DAC1541 Review
This is a work in progress, so I will be adding to this as I go. But I have notable information/impressions so far.
*** TL;DR Vs my casual impressions ***
These two DACS become extremely close after warmup. There are subtleties between them, but in the grand scheme of things they are small.
Detail Vs Casual Impressions
In fact they are so close I thought I must have been a victim of expectation bias after my critical listening session, and/or my ears are shit. Then it occurred to me that the DAC2541 was cold and the other was warm when I first listened, and there is that 0.2dB bump as well. So I turned off DAC2541 overnight to test cold the following morning and eliminated the 0.2dB trim, and the difference heard in my initial impression was clearly evident. Interestingly this DAC sounds very good cold, it just sounds like it has a bit more 'relief' without this sounding sharpened or etched at all, and that +0.2dB makes it all the more evident. In fact I find this effect more pronounced in midbass through lower mids than in the treble region.
Warm DAC2541 up a couple of days, apply the trim and these differences all but disappear.
SE is where we do have significant differences. DAC1541 has a more focused central image SE vs BAL which a few of us have noted before (and some prefer over BAL). This difference is borne out testing DAC2541 vs DAC1541 - the same staging focus is evident on DAC1541 and is absent on DAC2541. The overall stage is a touch narrower on DAC1541, but this sense of center image focus seems greater than just some compression of stage width would account for.
So existing DAC1541 owners, this is not going to be a revolutionary upgrade, and if you are one who prefers SE then you won't find that advanatge in the newer model.
Actual audio impressions to come.
Setup is Surface Pro/JRiver with linked Zones for DAC1541 and DAC2541 using WASAPI because the ASIO Driver only seems to let me address one dac at a time. Connection is USB out to both DACs from Surface Dock. From the DACS into a GoldPoint Selector, into a GoldPoint 47-step Level control and from there balanced out to Genelec 8030B+7050B. I have adapters from SE to BAL so I can compare BAL to BAL and SE to SE. Position is nearfield, with reasonably flat measured FR using SonarWorks.
Headphones are 2019 HD600, switched beween the DACs SE using a TC-7240 switchbox and Monoprice RCA-->TRS adapters.
Both DACS had been on and playing music for 72 hours bvefore critical listening
The DACS are level matched using Decibel X on my LG V60. The matching was done using pink noise, averaged over one minute. Matching was done for both Line outs and Headphone outs and matched to a resolution of +/- 0.1dB.
Line outs (BAL) has a difference of +0.2dB for DAC2541
The Headphone outs (SE) have a difference of +2.4dB for DAC2541
Compensation was done using DMG TrackControl, applying appropriate trim to DAC2541.
All DSP was turned off for the purposes of this review, with the exception of DMG TrackControl for level matching. My GoldPoint attenuator has 1dB jumps which is not enough resolution to accurately compensate.
To be continued...
That's extremely good to know! I continue to enjoy my DAC1541 daily.
@GoodEnoughGear what happens if the 2541 is on for a long time and then you turn it off and on again? I wonder if the self calibration will matter.
Separate names with a comma.