Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 25, 2024 19:54:28 GMT
I am trying to decide which DAC to buy.
I have obsessed about ethernet noise already and my current delta sigma DAC is sounding much much better (more R2Risch sounding) then before my noise war started.
One obvious question when selecting the DAC is what type of DAC topology to get.
I have noticed a pattern where it seems like reviewers that don´t bother with addressing the digital noise seem to be more positive about R2R dac´s and have bigger issues with the delta sigma DAC´s.
Reviewers that have audiophile grade switches and clean power actually seem to think that the gap is not so big between delta sigma and R2R, and might even prefer chip based sometimes.
My theory now is that the resistor ladder topology might actually clean up some of the noise, which is why they come across as more analogue. But if you clean up that noise before the DAC then the gap (in terms of "analogue" sounding and better soundstage) vanishes, to some extent, and then delta sigma actually have some pros, like speed, for example.
I know that there are of course a lot of factors in play here! This is just a generalization.
What is your though on this? Do you think R2R is more immune to the digital domain noise and has a bigger advantage the more noise you feed your dac?
|
|
|
|
Post by John on Feb 25, 2024 21:00:51 GMT
I would say implementation plays the biggest part in this, in my experience both are affected by noise and benefit from working on reducing this as much as possible.
|
|
|
|
Post by MartinT on Feb 25, 2024 21:19:41 GMT
I agree that it's implementation, not chipset, that matters. I am tired of the number of times I have had the ESS Sabre discussion. No, it is not inherently bright any more than R2R inherently lacks resolution.
It's hard to be able to audition DACs but I would urge you to try to hear at least a good example of both. Do you have friends who can let you have a good listen to their DACs?
|
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 25, 2024 21:34:04 GMT
I agree that it's implementation, not chipset, that matters. I am tired of the number of times I have had the ESS Sabre discussion. No, it is not inherently bright any more than R2R inherently lacks resolution. It's hard to be able to audition DACs but I would urge you to try to hear at least a good example of both. Do you have friends who can let you have a good listen to their DACs? No, i can´t borrow but maybe i can try a few from a nearby store. I think i agree that i shouldn´t focus on that at all (even if i just did...). I just thought it was an interesting thought that maybe the ladder resistor is actually better at stopping that ethernet noise since the signal has to go thru all those steps. It makes sense that one single chip is more sensitive to noise, in my mind. But of course a better DAC is also handling that noise internally better since they have put more effort at it. Thanks for you replies! |
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 25, 2024 22:43:56 GMT
I agree that it's implementation, not chipset, that matters. I am tired of the number of times I have had the ESS Sabre discussion. No, it is not inherently bright any more than R2R inherently lacks resolution. It's hard to be able to audition DACs but I would urge you to try to hear at least a good example of both. Do you have friends who can let you have a good listen to their DACs? No, i can´t borrow but maybe i can try a few from a nearby store. I think i agree that i shouldn´t focus on that at all (even if i just did...). I just thought it was an interesting thought that maybe the ladder resistor is actually better at stopping that ethernet noise since the signal has to go thru all those steps. It makes sense that one single chip is more sensitive to noise, in my mind. But of course a better DAC is also handling that noise internally better since they have put more effort at it. Thanks for you replies! Let me just say that i want us to take away the implementation from the equation, in this topic. There clearly is a R2R vs Chip sound signature and i was trying to understand why that is. I know that the implementation can reverse those sound signatures but lets say the implementation is the same (even if that is impossible...). The typical R2R sound, if i may say so, is more "analogue", more full bodied, textured and has a bigger and deeper soundstage. At least that is an extremely common description that is often associated with R2R. It can´t be a coincidence that those exact aspects are getting improved when you lower the noise floor in front of a DAC, at least from what i have noticed and from what others are reporting. That "must" then mean that R2R is better at handling the noise natively, in my mind. I felt that it would be nice to understand why instead of only saying that it depends. They clearly sound different in their pure/stock form and understanding why would maybe help people understanding more, i was thinking. But i am not saying that i am right... |
|
|
|
Post by stellabagpuss on Feb 25, 2024 23:11:57 GMT
l have to echo John & MartinT view point. This last year l've been on that very journey,killing the noise.
Regardless of chipset, the more you dig, and in my case,modify. You realise that it's all down to design.
Of course this doesn't help if you can't listen to the DAC of choice. What l can say, l value the opinions of TAS members and there recommendations, that's what lead me to the R26. And as good as it is, l pushed the performance further, by modding the unit.
The downside of this rabbit hole of noise, its addictive, as it results in improvements,hell ...even not having LED results in an improvement..lt can almost be never ending.Not to mention voiding warranty,and possible resale.
What DAC are you interested in, perhaps the TAS chaps can advise and help.
|
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 25, 2024 23:34:14 GMT
l have to echo John & MartinT view point. This last year l've been on that very journey,killing the noise. Regardless of chipset, the more you dig, and in my case,modify. You realise that it's all down to design. Of course this doesn't help if you can't listen to the DAC of choice. What l can say, l value the opinions of TAS members and there recommendations, that's what lead me to the R26. And as good as it is, l pushed the performance further, by modding the unit. The downside of this rabbit hole of noise, its addictive, as it results in improvements,hell ...even not having LED results in an improvement..lt can almost be never ending.Not to mention voiding warranty,and possible resale. What DAC are you interested in, perhaps the TAS chaps can advise and help. Thanks! I want to have those things that are associated with the Gustard R26 for cheaper then the R26  . I know that it is really hard since the R26 is really good, and can be modded with the Leo Bodnar, for example. But i have started to realize, when looking at different reviews, that you can probably get similar traits also from something like the new SMSL VMV D2R or SU10 (chip based), especially when the noise is cleaned up in front of them. The question is if they get closer to the R26 (for example) or if the performance distance remain when exposed to the same low noise floor. I have a feeling that the chip based DACs can actually get closer to the R2R, in a low noise floor setting, as they react even more positively to the lowered noise floor. That is at least my theory that i would like input on. |
|
|
|
Post by stellabagpuss on Feb 26, 2024 0:11:15 GMT
All l can say from my experience, that l have had various ESS chips and AKM4499, they all have there sound signature, but nothing like the R2R in the Gustard R26, which offers a more analogue based sound.
l can't speak for every DAC, but don't know of any other cheaper DAC that would come close, and if there was, we would know about already. The ESS and AKM chips have a much lower noise performance than R2R, if you read ASR, it's one of the major points they mention on there reviews, the R2R is some roughly -20db a drift, but what many people forget, both R2R and Chips are already below the threshold of human hearing, showing that measurements can only go so far.
Personally l would hold out for used R26,if you looking for that R2R sound.
|
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 26, 2024 6:49:34 GMT
All l can say from my experience, that l have had various ESS chips and AKM4499, they all have there sound signature, but nothing like the R2R in the Gustard R26, which offers a more analogue based sound. l can't speak for every DAC, but don't know of any other cheaper DAC that would come close, and if there was, we would know about already. The ESS and AKM chips have a much lower noise performance than R2R, if you read ASR, it's one of the major points they mention on there reviews, the R2R is some roughly -20db a drift, but what many people forget, both R2R and Chips are already below the threshold of human hearing, showing that measurements can only go so far. Personally l would hold out for used R26,if you looking for that R2R sound. But did you try the cheaper ESS and AKM chip DAC´s after the noise reduction? I agree they will likely be "worse" sounding still! But I have a feeling they are much closer then one would think, in sound signature, since the noise is an extremely important factor and the cheaper DAC´s just need even more help lowering the noise floor since they don´t do it by themselves. The low noise floor in chip based DAC is very academical since you are probably no where near that noise floor in reality, the moment you connect something to it. People who wants that R2Risch sound tend to buy such an R2R DAC, or something that has those traits. But they haven´t usually explored what happened with their old ESS and AKM sound signature after going bananas with the noise reduction. They usually start the noise war after getting a really nice/expensive DAC. I agree that is seems like the R26 is hard to beat but i also haven´t read any review where an cheaper AKM, ESS or Rohm chip has been tested in a "perfect" noise setting. There is a gap there, i believe, since very few reviewers are testing the cheaper DAC´s in a very low noise floor environment. Having said this, i am very tempted to go for the R26 regardless. My biggest issue is that it is a bit bigger (deeper) then i want, but i guess i can live with that. |
|
|
|
Post by John on Feb 26, 2024 7:50:44 GMT
I had my fair share of Topping DACs and heard a few SMSL DACs. If you like a detailed DAC a good way to go Check out these DACs they are well reviewed and are R2R based. www.soekris.dk/products.htmlIn general R2R are known for their warmth and Sabre DACs for detail. But as we already said, implementation is the key to all of this. |
|
|
|
Post by MartinT on Feb 26, 2024 7:51:30 GMT
Example: I have an ES9038 chipset DAC, it's my second one so I know I prefer them. Many people suck in their breath and say "oh but the brightness" and other such things. It's nonsense and not inherent in the chipset. How do I know? Because I listen to mine every day and it faithfully reproduces what's in the recording.
HOWEVER, and this is a biggie: with any DAC you had better prepare the incoming signal well and do everything you can to remove electrical, radio and mechanical noise. Only then will you extract all that is possible - which in the case of the ES9038 is epic levels of fine detail, soundstage and dynamics.
If you are prepared to put the work in and want to hear everything you can possibly extract from the data, shop around for your preferred device and start your journey.
|
|
|
|
Post by John on Feb 26, 2024 7:54:37 GMT
|
|
|
|
Post by John on Feb 26, 2024 8:01:35 GMT
I can ensure you that even on a R2R based DAC decreasing electrical noise makes a big difference. As Martin says whatever you route to get the best out of them requires dealing with noise.
I know this is not helping you decide but listening to a few different DACs would be the best suggestion I can make as to which route
When you get either approach sorted the sense of dynamics and scale is so much fun to listen to
|
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 26, 2024 8:02:11 GMT
Example: I have an ES9038 chipset DAC, it's my second one so I know I prefer them. Many people suck in their breath and say "oh but the brightness" and other such things. It's nonsense and not inherent in the chipset. How do I know? Because I listen to mine every day and it faithfully reproduces what's in the recording. HOWEVER, and this is a biggie: with any DAC you had better prepare the incoming signal well and do everything you can to remove electrical, radio and mechanical noise. Only then will you extract all that is possible - which in the case of the ES9038 is epic levels of fine detail, soundstage and dynamics. If you are prepared to put the work in and want to hear everything you can possibly extract from the data, shop around for your preferred device and start your journey. Yes, that is basically what i am also trying to say, and what i also now start to see in reviews (in addition to you saying it), if you know what the reviewers infrastructure looks like, when they test something. Put a "sharp" sounding ESS implementation behind the Network Accoustic Muon Pro filter, and a really good streamer, and it will not sound sharp at all, i think. (ok, that is a very general statement, but i hope you understand what i mean) The gap is actually really small, much smaller then people think, in term of "analogue" sounding, between the different chipsets. It is more about the noise and maybe that the ESS and AKM chipsets are more vulnerable for that noise, even more so then the R2R? But again, there are a lot of other implementation factors, for sure! |
|
|
|
Post by MartinT on Feb 26, 2024 8:18:26 GMT
The gap is small between the chipsets, the gap in sound between implementations and optimisations is huge.
|
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 26, 2024 8:22:19 GMT
Wow, thanks for the link! Very interesting. |
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 26, 2024 8:47:17 GMT
Thinking out loud some more...
Could it rather be so that the ESS chipset is so capable (again, a generalization) that any imperfection (noise) gets amplified, hence coming across as sharp sounding. Feed that chipset with no noise, which is the very hard challenge(!), and it will reward.
|
|
|
|
Post by HD Music & Test on Feb 26, 2024 8:59:51 GMT
Yes the R26 cabn be made to really sing with the benefits of real tuxture and depth and decay as in three dimensionality. Generally (though not all) R2R present a more relaxed presentation that DS dacs do. Although score in the tonal recreation and trailing note body. DS tend to have more dynamics and a lively sound but that doesn't mean bright or upfront either.
As the chaps have mentioned you need to audition at home in your own time with your own music and not how many 'best mates'feel it sounds like, your music, your system.
It's no secret I'm known for not being a fan of ESS chips and find it hard to tune in to their superb sinad and hyper detail presentation, however currently I have been listening to a couple of dac's that use these chipset and for the first time its very listenable and enjoyable for longer periods. Given I have over one hundred dac's that a positive in my book.
All I can suggest is listen/listen/listen until the key fits that slot you wish for, after all its your own money and your music!
|
|
Tobias
Rank: Quartet
Posts: 320
|
Post by Tobias on Feb 26, 2024 9:07:56 GMT
Yes the R26 cabn be made to really sing with the benefits of real tuxture and depth and decay as in three dimensionality. Generally (though not all) R2R present a more relaxed presentation that DS dacs do. Although score in the tonal recreation and trailing note body. DS tend to have more dynamics and a lively sound but that doesn't mean bright or upfront either. As the chaps have mentioned you need to audition at home in your own time with your own music and not how many 'best mates'feel it sounds like, your music, your system. It's no secret I'm known for not being a fan of ESS chips and find it hard to tune in to their superb sinad and hyper detail presentation, however currently I have been listening to a couple of dac's that use these chipset and for the first time its very listenable and enjoyable for longer periods. Given I have over one hundred dac's that a positive in my book. All I can suggest is listen/listen/listen until the key fits that slot you wish for, after all its your own money and your music! Thank you so much, much appreciated input! I have to ask, to give some added context to your valuable answer. Could it be so that you have a better front-end (lower noise) then ever now? And that it is actually a big reason also to why even the ESS chipset is now also sounding listenable and enjoyable for longer periods? |
|
|
|
Post by MartinT on Feb 26, 2024 11:32:22 GMT
Could it rather be so that the ESS chipset is so capable (again, a generalization) that any imperfection (noise) gets amplified, hence coming across as sharp sounding. Feed that chipset with no noise, which is the very hard challenge(!), and it will reward. Yes, the ESS chipset reveals every tiny detail. The reward for getting it right is incredible fine detail, insight, dynamics and bass without losing soundstage and 3D. EDIT: to answer your final question, you do need a well sorted and balanced system as a revealing DAC can be too much for a poor system to handle. If you are going to pursue digital excellence, the whole system needs to be in line with that ethos. |
|
.
