After more testing using MConnect Tidal via UPNP on the SA30, I have heard distortion on the following music file types:
24 bit 192KHz MQA
24 bit 96KHz MQA
24 bit 48KHz MQA
I have looked into the Lyngdorf streamers and they do not meet the technical specifications laid out by the SA30 (e.g. Lyngdorf TDAI-3400 has vastly inferior DAC and amplifier performance - ref Audio Science Review).
The only real alternative from a technical (and hence absolute) perspective is the SR250.
Also, I wonder if the SA20 has the same issues? - if not, the distortion is almost certainly related to either DIRAC or the streaming aspect.
Listen before you judge
First of all measurements do not tell the whole story, and second I have already seen some messed up measurements on ASR. Seems he is using same, not entirely correct, methodology to all devices (famous Denon AVR case, or… Arcam AV40).
In case of Lyngdorf it’s hard to say about DAC and amp, as… whole amp is a DAC
It’s one of the few fully digital amps.
As for SA30, it has annoying bugs. But to be honest there’s so few similar devices ticking all the boxes (all-in-one with room correction and HDMI), in the same price range, that for me it makes no sense to get rid of it. I can really live for some time with necessity to change source 1-2 times per day, or having to use remote to set correct volume (since it doesn’t work correctly in Roon, once max volume is set). I know those are software bugs, and will get fixed, hopefully sooner than later. Yes, it is annoying and I am pissed off at Arcam. I have a hard time understanding why it takes so long to fix bugs in what really is a simple device.
BUT I don’t see a real competition for SA30, especially if one is looking for class G or AB amp. SR250 is a striped down AVR, and for me definitely is not something I would consider after having SA30.
So for me I have to quote “a classic” - “patience my young padawan”
Haven’t seen any reports on distortion on SA20, but it’s much more simple device. Software is there only for control (inputs, volume, display etc). There’s no streaming and any kind of signal processing. DAC in those devices is quite simple implementation of ESS chips, with software controlling only filter selection (filters are built in the chip), so it’s definitely not DAC causing the issue, it’s software part of streamer or DSP.
As a Chartered Electronics Engineer, having worked with ultra high end FPGAs, DACs and DSPs, I can categorically say that it is 100% based on technical specification. There is no voodoo magic in audio or any other tech. based industry.
How something sounds is based on technical specification and this applies to all audio equipment including DACs, amplifiers, streamers and speakers alike.
Harmonic distortion, transient response, jitter, frequency response, time alignment, RMS power are all very real technical specifications.
The best sounding equipment are always the ones with the best technical specifications. Always.
The issue we have with the SA30 is that it has very high technical specifications (circa reference in class) but the implementation of (I suspect the incorrect sampling of the DAC, as it never seems to happen at Fs = 44.1KHz) causes terrible distortion (putting the SA30 to the bottom in class when this occurs).
I also wouldn’t knock the SR250 too much. I suspect the amplifier layout is very similar to the SA30 and can categorically say that my AVR-850 is a similar quality of sound to the SA30. However, I would give the edge to the SA30 and this is likely because of the far superior DACs (Sabre ESS9038) used.
The SA20 and SA10 don’t have Dirac and no streaming. Would be interesting to know of the new ST60 has the issues. You would assume architecture is comparable and same sabre DAC in play. Incidentally I have CDS50 too and the only trouble with that I’ve had so far is MusicLife, it’s not as technically capable as the SA30 for streaming and I didn’t buy it for that, only for CD and SACD.
Sorry but whilst I understand your comment about the best sounding amp being the one with the best technical specification, with all due respect that’s nonsense in reality In terms technically of anything measurable in a lab or whatever fine, but you’re forgetting about my source capabilities, my speakers, my room and my ears, and my own personal preferences which are totally subjective.
Ultimately if someone prefers A over B or vice versa only they can be the judge of what they hear, tech specs are irrelevant even if they use that a guide in making purchase decisions it shouldn’t be the only factor,
But rarely do we see decent set of measurements. Like in case of ASR it’s almost purely based on SINAD, which alone doesn’t actually tell you how something sounds. Have you seen there one measurement of transient response? Time alignment? There’s a reason people buy I.e. Arcam AVRs (which measured poorly at ASR) instead of Denon AVRs (which measured good). That reason is usually sound quality. So if for so many people Arcam sounds better, than certainly something is missing from ASR measurements. And that’s my point.
But SUBJECTIVELY not always devices which measure better, sound better. For example I love tube amps, especially single ended triode, which usually do not measure too good. I would even say usually they measure quite poorly. But what can I do? For me it sounds really good. Probably because of harmonic distortion. But actually (hope you can enlighten me here) for some reason SET amps tend to have really great imaging, holography effect. Do you by any chance know which parameters are responsible? Which measurements to look at? It’s definitely not only channel crosstalk.
I’m definitely not saying there’s any voodoo magic here. It’s just my experience is telling me, that average Smith doesn’t have full set of all relevant measurements, to tell what SOUNDS good only based on those.
If it would be DAC, that would be very strange, as is really very basic implementation and Arcam is actually good at HW part. Besides SA20 should also be affected. I still think it’s SW related and it’s somewhere in DSP. It was not there in 521 version. In 705 they were changing DSP to give Dirac more headroom (circa 10dB). Maybe it’s related. I think in case of digital sources signal is always going through DSP, even if Dirac is not really engaged.
So for now I just switch the input and restart playback. 5 seconds and distortion is gone. Annoying, but not deal breaker. Now if it’s not fixed soon, then it will be another story…
That’s right, they don’t. That’s why I think it’s not DAC related.
Now I thought the same about ST60. Would be good to know if it also has same issue.
As far as I know SA30 and ST60 share same platform and software (at least parts of it). So that would help us to understand where issue is.
If ST60 has same issue, it would be steamer issue. If it doesn’t, it would have to be DSP, as ST60 doesn’t have any AFAIK.
The overall technical specification is based on all of the components combined (source, amp, speakers).
Flat frequency response for all components is desirable, using DIRAC or other DSP with Harman curve (slightly elevated low frequency response) to provide proved response across thousands of people.
Tube amps are great at low volumes because they work in class A (lowest distortion category). They run very hot and are typically low power, so you are reliant on very sensitive speakers to provide SPL.
We slightly digress but there is one common theme… there are people who have a great deal of affection for Arcam on this forum, which is nice.
Now, if we can move back to 1.33 (Build 705), does anybody know what features will be lost?
Did I read that you need a developer subscription to move back?
Actually I think 521 is pre distortion, you can move back to it if you can put your hand on a copy, there was an interim 6xx I think before distortion arrive with the 705 release, For my use I am still finding it (521) preferable.
I’m assuming most of my issues are MusicLife app related (might not be of course but I don’t have the same troubles with the web client).I do have lip sync issues over HDMI ARC but suspect that is inflicted by my TV, LG is renowned for it and not acknowledging the problem. My viny and CD/SACD sound fabulous.
At least it seems Arcam don’t shy away from their software issues…and whilst I have no history with Arcam I feel invested in it given this for me was an expensive item, but also it does sound really great tbh … to my ears anyway
It’s not only about class A. There are good class A solid state amps, but they sound usually bit different, than tubes.
I don’t have affection for Arcam. Just can’t find anything else that would tick all the boxes (including WAF). I know issues will be fixed one day or another, and they are annoying, but not really make the amp useless, so worth to keep it and wait for fixes.
As for 521 FW. I have a copy, so you can PM me. No features are lost with older firmware, just a bit different screen looks in case of pre 705 ones. I also have some 6xx betas.
I’m not really using Music Life, so don’t know what bugs it has. Using Roon as a daily driver. ML very rarely as remote control.
As for HDMI, it might be TV, you’re right. Although before 705 I also had this problem. Now it’s gone. Using Panasonic CX700 series TV.
Ok, I have been testing 867 a bit the last few days. I fresh loaded 867 via USB, I did not load any Dirac filters. I have streamed all formats via Roon with no noticeable distortion for the most part. This morning I switched to an analog input and played some reel to reel via Direct. Sounded great . For grins I switched off the Direct, and immediately I heard bad distortion in the right channel. No Dirac filters being used. So the DAC by itself was the source of the distortion. I switched back to Direct, all was great, back out of Direct and the distortion was back immediately. I switched to NET and back to the analog input and the distortion was gone.
I think you are correct. I have never heard distortion in Direct mode via an analog input. If anyone else has heard distortion in direct analog mode then these SA 30’s will probably have to go to a repair center to have hardware replaced. So far I have only heard distortion when the SA 30 is using its DAC.
Ken
I can’t test out as I no longer have the SA30, but I’m sure I never had distortion when using the SA30 with my Chord Hugo 2 DAC. Definitely had distortion using the SA30 DAC though.
I had intermittent distortion on analogue inputs (cd) since I bought it back in february…(tried different units with same result…the only solution being the direct function) with v705 it got worse with distortion also on mqa tracks when streaming TIDAL,…(I therefore recently returned my SA30 to the dealer- almost a year after release and the problems were just getting worse.;(
I don’t think DAC is the source. As said before - i think, that in case of digital path (which is also the case for analogue sources if direct mode is not used) signal always goes via DSP and this is real source of issues.
I might be wrong though. It’s just “educated guess”.
You are probably correct.As I recall all signals are upsampled before passing to the DAC. But MQA signals would not be upsampled, correct? But your suggestion the trouble is in the ADC is probably correct, but I hear at times distortion with MQA and the ADC should be bypassed. But honestly this is over my head. I have even wondered if the SA30 is running out of memory to buffer the signal. But all I know is it is basically broken and I need to talk to my dealer ASAP. I am afraid if Arcam can’t fix it, they will just discontinue the model and we start trying to get warranty repairs etc…
Without doing any detailed testing, this is my perception as well - ie. that there are times when the processor (or its memory) can’t keep up with the demands of high res streaming, MQA, Dirac etc. This might explain why it’s not been fixed yet, because it can’t be fixed without improving the processor spec.
In terms technically of anything measurable in a lab or whatever fine, but you’re forgetting about my source capabilities, my speakers, my room and my ears, and my own personal preferences which are totally subjective.