[realmassy]

I must have missed that ^ gonna read it now, thanks!

 

[koso]

I must have missed that ^ gonna read it now, thanks!

If you have any specific questions, feel free to ask here (or on the other thread)

 

[sajunky]

I had Venus II DAC with no problems, very reliable, good quality DAC. If you are interestet, there is my Denafrips Venus II vs Holo Audio Spring 3 KTE comparison on "Holo Audio Spring R2R DAC" thread (post nr. 2720).

Your impression is matching many other users. Denafrips (all models) has an overal smooth presentation due to the linear interpolation (DSP-free oversampling method), maintaining real R2R properties. Holo Audio is not oversampling, but it is not NOS (as false advertised), as there is digital filtering in place and ultrasonic data scrambling that make sound more similar to Delta-Sigma DACs. A link to your post is here.

 

[richardloh]

Denafrips Terminator Plus First Impressions.

Just straight away, adding this to my setup has been the largest leap in sound quality yet. Immediately on first listen I noticed texture and details on tracks I hadn’t heard before. Usually it takes some time for a component to settle in before noticing these changes. Not this time.

Soundstage is monumental. It gives so much room for instruments to materialize out of thin air. This is aided by the darkest background I’ve personally heard from an audio component. Everything comes through with more confidence than on the Venus II and it’s not subtle either. Music sounds alive. The minimal amount of sibilance (if any) heard on the Venus II is practically erased. Pleasing, powerful, transparent, and resolving are some of the words that come to mind. It’s a serious piece of equipment that can transform your system, given one has the right level of gear. Height and depth seem limitless. Paired with the Susvara, decay is on another level with vocals especially creeping into haunting levels of transparency. Namely while listening to Lord of the Rings: Fellowship of the Ring Soundtrack in 24bit, there were moments where I took the headphones off and just sat there in disbelief. Just.. the amount of information there is mind boggling. As if this wasn’t addicting already, the Terminator Plus is a hypothetical audio crack machine with an unlimited supply.

I can see why so many review it with flying colors. It’s the best DAC I’ve heard to date & a remarkable piece of engineering. Happy to be part of the club & can’t wait for the many listening sessions this DAC is sure to elevate.

Nice ! Another happy Tplus owner not unlike me.

Just to share on your setup of Gaia and Tplus .. unless your usb source is not of certain level, pls do not hesitate to go directly to the Tplus as the usb implementation of the Tplus is excellent.

I had the allo usb signature and NuPrime S1 media players that will indeed improve going via Gaia but for my oladra upgraded Antipodes cx+ex stack, it deteriorates sound quality via gaia instead of directly into usb input of Tplus .. obviously as just tried again since foc for me.

Cheers and ymmv as usual .. cheers.

 

[richardloh]

I have a T2 and Gaia , can i use cheap HDMI and cheap BNC, or better use cheap hdmi and better Bnc or vice versa ?

Hi Mat,

For sure cheap proaudio BNC with tested performance specification and more expensive hdmi i2s cables as shared from some of the owners here ... without such specification.

A big ymmv since cables accentuate or hide strengths and weaknesses of each individual system i.e. final tuning.

Great that a well shielded cables works well in a system plagued by interference but it would do close to nothing if there is no need for such shielded cables in another system.

Spend our limited funds wisely .. 20% on all cablings, including powercords, dc cords, various interconnects, clock cables etc. w.r.t. equipment works for me.

Cheers.

Richard

 

[orsa99]

Sharp or slow filter? (Terminator not plus with DSP to 30.05.2019). What are your impressions? Which one do you prefer?

 

[cpurdy]

@PrTvI have some "bricked" gear that I had to throw away due to no support (LH Labs DAC as an example).

There's a name that I haven't heard in a long time!

I have a headphone DAC/amp from them. Fortunately it still works great. But yeah, their tech support was never amazing, to say the least.

 

[dcguy73]

Interesting video review on YouTube by @GoldenOne on the Terminator Plus in which we learn:

1.) It measures poorly — not due to its internal components, but due to its use of DSP processing (also observed by Stereophile).
2.) The NOS mode is not really non-oversampling at all.
3.) The Gaia DDC makes no impact on jitter measurement because the T+ internally reclocks everything anyway, and if you connect the Gaia to the T+ via clock cables as it says to in the manual, the left and right channels get flipped.
4.) The USB input is galvanically isolated (who knew? why don’t they publicize that?).
5.) Due to its high impedance, T+ works best with an active/non-passive preamp.
6.) The T+ sounds great with all genres and even with poorly-recorded music in a Technicolor, slightly oversaturated way vs. the Holo May DAC which gives less of a euphonic presentation but measures extremely well.

@GoldenOne does say he thoroughly enjoyed the sound of the T+ overall (his reference DAC is the Holo May), but this is the first review I’ve seen on the T+ with some real criticism in it. I wonder if Alvin is going to respond to these points and maybe fix some of the DSP issues in a firmware update.

 

[sajunky]

This is confusing or even incorrect on few items:

#1. It is commonly known that DSP processing improve measurements, not in reverse. Secondly, it impose that Holo May (used as for a comparison) do not do DSP, which is wrong. It has been discussed in this this thread, where the author was trying to promote Holo May using Denafrips Ares II as a shield. The same story is made for a different audience. To be clear, I don't claim that Denafrips do not use DSP processing, I am not sure. The whole story is an example of using smearing tactics to promote other product.

#2. This true that Denafrips NOS is not really NOS, but the same apply to the other brand used for comparison. See a discussion referenced above. According to the author's own definition Holo DACs technically are not NOS and do not sound NOS either. Both brands violate NOS principles, in a different way, a marketing is misleading in both cases.

#3. This aspect requires a separate thread not a utube platform. If is true that T+ reclock everything, it is really weird, as a clock originate from T+. A valid point is about swapping channels.

#4. I read on the Denafrips website that T2 and T+ use galvanic isolator on USB, I don't know why the author didn't learn.

 

[richardloh]

3.) The Gaia DDC makes no impact on jitter measurement because the T+ internally reclocks everything anyway, and if you connect the Gaia to the T+ via clock cables as it says to in the manual, the left and right channels get flipped.

Not aware channels get flipped in the manual and my soundcheck testing did not indicate so except via dual aes.

Only if you try to use dual aes will have either phase or/and channels get swapped and not other connection e.g. i2s ... still waiting for the fix from Denafrips.

Had you verified this ?

Cheers.

 

[GoldenSound]

#1. It is commonly known that DSP processing improve measurements, not in reverse. Secondly, it impose that Holo May (used as for a comparison) do not do DSP, which is wrong. It has been discussed in this this thread, where the author was trying to promote Holo May using Denafrips Ares II as a shield. The same story is made for a different audience. To be clear, I don't claim that Denafrips do not use DSP processing, I am not sure. The whole story is an example of using smearing tactics to promote other product.

DSP just means digital signal processing. It can mean anything from EQ to resampling to dithering to pitch shifting. It's a broad term and how it affects measurements will depend on what's being done.
I know you have some obsession with making accusations against me, but if you actually watch the video you'd have seen that I said I quite liked the effect the DSP was having in many situations. I didn't say it was bad, just different.
In regards to measurements, it's not that it measures 'poorly' per se, just that the DSP seems to behave strangely in certain situations causing things like a linearity test to look very odd.

And no, this isn't some 'smear tactic', if you think I'm lying then you can use Stereophile's measurement of the regular termi for example:

#2. This true that Denafrips NOS is not really NOS, but the same apply to the other brand used for comparison. See a discussion referenced above. According to the author's own definition Holo DACs technically are not NOS and do not sound NOS either. Both brands violate NOS principles, in a different way, a marketing is misleading in both cases.

We've argued that point to death in another thread and you've still yet to provide any evidence to counter anything I posted. I'm really not sure what your vendetta is here.

Not aware channels get flipped in the manual and my soundcheck testing did not indicate so except via dual aes.

I wasn't aware either, I only noticed it once a track that has stuff in the intro fully on the right came on and it was fully on the left now
I don't think it was related to the clock sync and just that the I2S pinouts were mislabeled. Easily fixed by changing I2S pinout or just using channel swap in roon, but figured it was worth mentioning.

I've no idea if it affects all units or in fact if it was the gaia or the termi. I just know that together with both of them set to the same pinout the channels were swapped

 

[chesebert]

DSP just means digital signal processing. It can mean anything from EQ to resampling to dithering to pitch shifting. It's a broad term and how it affects measurements will depend on what's being done.
I know you have some obsession with making accusations against me, but if you actually watch the video you'd have seen that I said I quite liked the effect the DSP was having in many situations. I didn't say it was bad, just different.
In regards to measurements, it's not that it measures 'poorly' per se, just that the DSP seems to behave strangely in certain situations causing things like a linearity test to look very odd.

And no, this isn't some 'smear tactic', if you think I'm lying then you can use Stereophile's measurement of the regular termi for example:

Correct me if I’m wrong, looks like gain and offset error potentially caused by mismatched resistors. What can I say, you get what you pay for. If you want R2R that measures well natively and not trying to error correct everything in DSP, you should just get MSB (or whomever uses their r2r models).

Given the significant difference in transfer function behavior between the 2 units I can say they rely almost exclusively on DSP to correct errors as opposed to trying to trimmed the resistors to a certain precision. This is like taking a badly exposed picture and trying to correct in post (not exactly but that’s the idea).

For all we know the company sent golden samples (properly trimmed resistors) to reviewers and influencers - although if they sent a unit with that performance to JA to test it’s not a good sign

 

[Riccioli30]

DSP significa semplicemente elaborazione del segnale digitale. Può significare qualsiasi cosa, dall'equalizzazione al ricampionamento, dal dithering al pitch shifting. È un termine ampio e il modo in cui influirà sulle misurazioni dipenderà da ciò che viene fatto.
So che hai una certa ossessione nel fare accuse contro di me, ma se davvero guardassi il video avresti visto che ho detto che mi piaceva abbastanza l'effetto che il DSP stava avendo in molte situazioni. Non ho detto che fosse male, solo diverso.
Per quanto riguarda le misurazioni, non è che misuri "scarsamente" di per sé, solo che il DSP sembra comportarsi in modo strano in determinate situazioni, facendo sembrare cose come un test di linearità molto strane.

E no, questa non è una "tattica diffamatoria", se pensi che stia mentendo, puoi usare la misurazione di Stereophile dei termi regolari, ad esempio:




Abbiamo sostenuto che questo punto alla morte in un altro thread e non hai ancora fornito alcuna prova per contrastare tutto ciò che ho pubblicato. Non sono davvero sicuro di quale sia la tua vendetta qui.

Neanch'io ne ero a conoscenza, l'ho notato solo una volta che si è accesa una traccia con roba nell'introduzione completamente a destra e ora era completamente a sinistra
Non penso che fosse correlato alla sincronizzazione dell'orologio e solo che i pinout I2S erano etichettati erroneamente. Risolto facilmente modificando la piedinatura I2S o semplicemente usando lo scambio di canale in roon, ma ho pensato che valesse la pena menzionarlo.

Non ho idea se influisca su tutte le unità o in effetti se fosse il gaia o il termi. So solo che insieme a entrambi impostati sulla stessa piedinatura i canali sono stati scambiati

DSP just means digital signal processing. It can mean anything from EQ to resampling to dithering to pitch shifting. It's a broad term and how it affects measurements will depend on what's being done.
I know you have some obsession with making accusations against me, but if you actually watch the video you'd have seen that I said I quite liked the effect the DSP was having in many situations. I didn't say it was bad, just different.
In regards to measurements, it's not that it measures 'poorly' per se, just that the DSP seems to behave strangely in certain situations causing things like a linearity test to look very odd.

And no, this isn't some 'smear tactic', if you think I'm lying then you can use Stereophile's measurement of the regular termi for example:




We've argued that point to death in another thread and you've still yet to provide any evidence to counter anything I posted. I'm really not sure what your vendetta is here.

I wasn't aware either, I only noticed it once a track that has stuff in the intro fully on the right came on and it was fully on the left now
I don't think it was related to the clock sync and just that the I2S pinouts were mislabeled. Easily fixed by changing I2S pinout or just using channel swap in roon, but figured it was worth mentioning.

I've no idea if it affects all units or in fact if it was the gaia or the termi. I just know that together with both of them set to the same pinout the channels were swapped

Very interesting and very well argued with objective technical findings. This (IMO) mark the difference from other reviews and makes reading enjoyable and informative. Bravo

 

[GoldenSound]

Correct me if I’m wrong, looks like gain and offset error potentially caused by mismatched resistors. What can I say, you get what you pay for. If you want R2R that measures well natively and not trying to error correct everything in DSP, you should just get MSB (or whomever uses their r2r models).

Given the significant difference in transfer function behavior between the 2 units I can say they rely almost exclusively on DSP to correct errors as opposed to trying to trimmed the resistors to a certain precision. This is like taking a badly exposed picture and trying to correct in post (not exactly but that’s the idea).

For all we know the company sent golden samples (properly trimmed resistors) to reviewers and influencers - although if they sent a unit with that performance to JA to test it’s not a good sign

It's unlikely to be this due to the consistency of the deviations.
Mismatched resistors would mean the linearity graph as a whole would just look poor and move above/below the 0dB reference point somewhat randomly. Whereas here we see steps above and below it of exactly the same value. And they also change depending on what I set the step value to.

This would indicate something in the digital domain/DSP rather than an inaccurate ladder.

 

[chesebert]

It's unlikely to be this due to the consistency of the deviations.
Mismatched resistors would mean the linearity graph as a whole would just look poor and move above/below the 0dB reference point somewhat randomly. Whereas here we see steps above and below it of exactly the same value. And they also change depending on what I set the step value to.

This would indicate something in the digital domain/DSP rather than an inaccurate ladder.

I was looking at the Stereophile’s graph. Perhaps you can also include the transfer function in addition to the linearity graph.

I think we are actually saying the same thing. The accuracy of the ladder relies almost exclusively on DSP algorithm rather than precision trimmed resistors.

In any case, these errors are terrible. Did you see this type of behavior with Holo dacs? Stereophile graphs look very clean and the dac they sent behaved exceptionally well. I just hope it’s not a golden sample or anything like that.