Welcome to the Early Access programme for dCS v3.0 firmware covering:
Bartók DAC
Bartók Headphone DAC
Rossini DAC
Rossini Player
This build has passed internal verification. It’s functionally complete and close to being finalised. We’re inviting the dCS Community to validate in their own systems and share listening impressions.
What’s in v3.0
Mapper: new Map 4 & Map 5 (Map 1 - 3 unchanged)
PCM Peak Control: choose clipping or no clipping behaviour on PCM intersample overs
Volume Lock: new on Bartók & Rossini
DSDx4 Upsampling: new option in addition to DSD & DSDx2
Two new PCM filters at 48 / 88.2 / 96kS/s sample rates
Expanse 3 crossfeed: available on Rossini models and Bartók Headphone DAC; not available on Bartók DAC
How to participate
If you own a Bartók DAC, Bartók Headphone DAC, Rossini DAC or Rossini Player (either APEX or non-APEX variants) and would like to participate in this Early Access program:
Read through and complete the sign-up form found here.
Once completed, we’ll add you to the v3.0_early_access group and you will be able to see more information across several topics in this v3.0 Early Access category.
When you have been approved and can see the new topics, read through the Start Here topic to install and begin using the v3.0 firmware.
We will be running this program for 6 weeks. During the program, we will invite users to focus on testing and providing their feedback on specific features of the update.
Support note: With this early access, we run a best‑effort process and provide feedback where needed. We may not be able to respond to every post, and response times may be longer than usual.
Support time for standard (non-early access) firmware is unaffected.
I’m not sure whether replies to a news item are the right way to ask questions and comment, but here goes.
First, does dCS see the new mappers as genuine technical advances or just as different ways of balancing the various tradeoffs involved in the process, and which some owners might prefer to the existing mappers?
Second, I’m a little surprised by this:
PCM Peak Control: choose clipping or no clipping behaviour on PCM intersample overs
There are a couple of threads about this issue: “Inter-sample overs” and before that “Stupid question” (see especially the 14th post, by @ermos). My interest arose from reading the information article about the topic by John Siau of Benchmark. He pointed out that the problem is quite common and occurs on cds that, from the nature of the music and of the production, would certainly not have been intended to contain intersample overs. Tony Faulkner contributed to the “Inter-sample overs” thread and his view of them is very negative. However, @James has wondered whether producers actually intended the intersample overs (as a contributor to a dirty, compressed sound? See his August 2024 post). That seems to me to over-think the issue.
What is the reasoning behind offering this choice to owners?
On your car you have a lot of options, do you use them all ? No. From one driver to another expectation, experience, they will use or not those options. Same here, no ?
Take the opportunity to test, you never know what can happen.
Exactly this - they are new options that explore different trade-offs in operating the Ring DAC, that provide different sonic signatures. The two new options have each gone through internal listening, but we are keen to get user feedback on them as well.
With intersample overs, the point is not that an engineer is likely to have used intersample overs as a deliberate production technique - I think we can say with certainty that that isn’t going to be the case. But, if the equipment the mastering engineer used did clip on intersample overs (and just acting as the replay chain we have no way of knowing whether this was the case or not), then there is a strong argument that the DAC should in fact clip on intersample overs to match what the mastering engineer was hearing more closely, if matching hat the engineer in the studio heard is the goal.
I should note that the audio does not need to be compressed or intentionally made ‘dirty’ to contain intersample overs. As has been pointed out in the Benchmark article, they are a very common occurrence, and not just in compressed commercial recordings.
Given that, offering the choice of behaviour to the listener to allow them to adjust the behaviour allows for a greater degree of system control, and potentially better matching of replay system to capture / production system.
I should also highlight that this is Early Access, not general availability firmware - finding out how listeners use it and their thoughts on the feature one way or the other are precisely why it has been included here.
In a now shut down thread (where @Lee provided initial sound impressions) you wrote:
“Just a quick admin note to say that I am going to lock this post - there are dedicated threads within the V3.0 Early Access Category for those who have signed up to use for providing any feedback on the new features. Thanks all!”
My understanding is there are multiple threads available for viewing only to those who are participating in the program. While I understand threads about bugs, UI, etc. may only have relevance to those actively testing the firmware, impressions on sound may be of interest to the entire community. Vivaldi owners are not yet able to participate however I am sure remain interested in reading impressions of the new sound.
Perhaps you may contemplate making this category of threads available to the entire community - even if only in a view/no reply format (for those not participating in the program)
Gregg, you are asking for sight of what dCS may regard as commercially sensitive information.
Data of this nature would, if posted here , also then be available to dCS’ competitors. what the rest of us get once the beta is complete and the finished update released may have been changed /refined in the meantime as a response to test results in any case.
dCS’ business practices are up to them but I would expect test participants to be subject to an NDA and, if so, would not expect any details of the tests’ outcome to be published here. Of course once v.3 is released there may be loads of subjective comments made in the forum.
In any case it looks like the features are additions and selectable. You don’t have to use them if you don’t like them.
I would like to remind everyone who has accepted these terms to please confine their comments to the Early Access threads on this forum and not post on the open sections of this forum, on other forums, or on social media etc.
Huh. It’s not a huge deal either way, but this does seem a little heavy handed. I could understand an NDA if dCS had competitors who made dacs that functioned in exactly the same way and would take advantage of their ideas, but I don’t think that’s the case. Isn’t the dCS approach, in the sum of all its parts, unique?
Hello Andrew, no need to apologise. After this post I will be quiet on the topic and just await the results of the exercise with great interest. I suppose that the recruitment of testers in an open way on an open forum dedicated, amongst other things, to the discussion of dCS products does create certain expectations. However, although the interests of owners and dCS themselves are certainly closely aligned, they are not always going to be the same.
All the beta testing groups i have seen have all been a closed shop till it’s sorted. That’s exactly how I would expect them to run, any problems found put back to the manufacturer (dCS) rather than publicly on a forum, even if the forum is owned by dCS.
Anyway can’t wait to try and hear the new maps on the Vivaldi as I knew that new stuff was coming and is fantastic to see dCS still doing inpovements to its older range, especially the Vivaldi.
As Dunc says there should be a version of V3.0 for Lina ( subject to the outcome of the beta test).
I see that you are new having joined in June. This suggests that you may be unfamiliar with the update process. I thought that it may help you avoid any forthcoming disappointment if I explain that this type of update is not a single entity like a Mosaic one but may differ in release date for different ranges.
You speak of the 3.0 firmware. However you may read about Bartok or Rossini final versions being released without any mention of Lina ( or Vivaldi). These may come at different times later. As you observe:
