[esimms86]

I'm considering buying one, is the customer service usually bad? How reliable are the units?

In the 3 1/2 years that I’ve had my A16 it has never had a problem that a firmware upgrade didn’t fix. So yes, it’s generally very reliable outside of the specific anecdotal and real experiences of a few HeadFiers. In at least a couple of well described cases(e.g., see above) it has taken exorbitant amounts of time for owners to get their units back after sending them in for repairs.

Beyond that, I’ll say no more about their customer service.

 

[zdjh22]

Take everything James says which a gigantic tub of salt. My unit has just been sitting there since May 19th, which was a Friday. I actually spoke with him on the phone on the 19th, and his exact words were "these issues don't take long to diagnose, so we'll likely have it figured out by Monday and get back with you".

Never heard back from him, and then I had to find out from John (Litlgi74) like two and a half weeks later that they were inquiring about the specific problems I said my unit was having. James couldn't be bothered to actually contact me about it.

Two more weeks go by and I send him an email asking for a realistic ETA on when I could actually expect to get my unit back, and to be honest about it even if they don't think it would be until the end of the summer. He then tells me they just started looking at it that day. Then more back and forth happens, he finally says they fixed everything and asked me if I was ready to have it shipped back. I tell him yes, then more crickets and the following week he goes on a two week vacation.

Thanks. Yes, it looks like “ready to ship this week” (which was last week) got interrupted by the two weeks the Smyths aren’t in the office.

 

[zdjh22]

I'm considering buying one, is the customer service usually bad? How reliable are the units?

I heard back quickly from James Smyth on an e-mail inquiry from late May about purchasing - just 3 or 4 days later. He estimated a 21-day lead time, and with shipping to the US my A16 was in my hands on the 22nd day.

After one week of very satisfying use, my A16 stopped working (“error code 003”). James answered my query in about 2 days, and said unfortunately that meant the DSP soldering job needed repair and the unit would have to go back. Smyth Research is paying the return shipping (when that will happen, I don’t know). It took forever for UPS to get my shipment to the UK through customs, and after I knew it had been delivered and waited an additional week I finally emailed James to ask for confirmation of receipt and the status of the repair. A week later he said the repair was done and the A16 was undergoing testing, anticipated to ship “this week” (that email was one week ago, no shipping notification was sent). And then another person on the forum advised that the Smyths were out for two weeks so don’t expect to hear anything for a while.

So, communications can be fairly rapid, or very slow, depending on the calendar. Judging from my experience and another forum member’s just upthread, the information received from Smyth Research tends not to be accurate. It looks like all customer communications go through one person (James). Two weeks to do the solder repair is pretty slow, considering the unit was built in a few weeks (I assume, since a lead time was given). It’s clear that an additional 7-day burn-in at Smyth Research prior to the original shipment would have had a good chance of catching the problem. On a roughly $5K item I would say a soldering issue is not a testament to good quality control (there are other instances of “error code 002” in this thread that also required DSP soldering repair). It is certainly possible that they are outsourcing some of the assembly to a 3rd party (the surface mount soldering work, for example) and perhaps the QC issues aren’t all in house.

I am a bit worried that, since my A16 has to run the 2.15 firmware (for the new version of the ARM motherboard), the PRIR issues with sound shifted to the right discussed upthread will bite me and I’ll have to wait for the next firmware. I may have seen a similar issue on my unit doing HPEQs and that functionality is more important to me than doing my own PRIRs.

 

[Sanctuary]

I heard back quickly from James Smyth on an e-mail inquiry from late May about purchasing - just 3 or 4 days later. He estimated a 21-day lead time, and with shipping to the US my A16 was in my hands on the 22nd day.

After one week of very satisfying use, my A16 stopped working (“error code 003”). James answered my query in about 2 days, and said unfortunately that meant the DSP soldering job needed repair and the unit would have to go back. Smyth Research is paying the return shipping (when that will happen, I don’t know). It took forever for UPS to get my shipment to the UK through customs, and after I knew it had been delivered and waited an additional week I finally emailed James to ask for confirmation of receipt and the status of the repair. A week later he said the repair was done and the A16 was undergoing testing, anticipated to ship “this week” (that email was one week ago, no shipping notification was sent). And then another person on the forum advised that the Smyths were out for two weeks so don’t expect to hear anything for a while.

So, communications can be fairly rapid, or very slow, depending on the calendar. Judging from my experience and another forum member’s just upthread, the information received from Smyth Research tends not to be accurate. It looks like all customer communications go through one person (James). Two weeks to do the solder repair is pretty slow, considering the unit was built in a few weeks (I assume, since a lead time was given). It’s clear that an additional 7-day burn-in at Smyth Research prior to the original shipment would have had a good chance of catching the problem. On a roughly $5K item I would say a soldering issue is not a testament to good quality control (there are other instances of “error code 002” in this thread that also required DSP soldering repair). It is certainly possible that they are outsourcing some of the assembly to a 3rd party (the surface mount soldering work, for example) and perhaps the QC issues aren’t all in house.

I am a bit worried that, since my A16 has to run the 2.15 firmware (for the new version of the ARM motherboard), the PRIR issues with sound shifted to the right discussed upthread will bite me and I’ll have to wait for the next firmware. I may have seen a similar issue on my unit doing HPEQs and that functionality is more important to me than doing my own PRIRs.

Just a heads up regarding "return shipping". They don't ship it back fully insured. They only cover up to $1,500. What this means if it a warranty return gets lost is anyone's guess. When they ship out new units what it means is if it gets lost, it covers their cost to build you another and ship you that one. For a unit under warranty repairs though? James told me to make sure my unit was fully insured when I shipped it, which was $4,275. For the return trip he was acting like it was going to be my responsibility to handle the shipping back to me if I wanted it to be fully covered. Haven't heard anything more on the subject since May 19th though, so I guess I'll find out more in just over a week...

Never have I dealt with such customer service like this in my nearly fifty years of existence.

Regarding the firmware though: unless the new motherboards are simply incompatible with prior firmwares, you might be able to just use a previous one. I hope to god I don't get sent back a buggy unit that I can't use for who knows how long until they do yet another new firmware. But since they said they had to replace hardware, that's likely what my unit will be saddled with as well. So I'll functionally still not have a working unit.

edit: Or possibly not. I thought it was a general speaker shift when listening to PRIRs, but it seems to be only an issue if you make new PRIRs with the new firmware. PRIRs made with previous firmware should still work fine on 2.15?

 

[sander99]

my A16 has to run the 2.15 firmware (for the new version of the ARM motherboard)

Regarding the firmware though: unless the new motherboards are simply incompatible with prior firmwares, you might be able to just use a previous one.

From 2.15 release notes:

APM110 C1 Hardware Backward Incompatibly
Realiser A16s manufactured since Q3 2022 use a revised APM110 decoder sub‐assembly, known
as APM110‐C1. This hardware is incompatible with A16 firmware revisions 2.12 and earlier.
Unfortunately, there is no retrospective solution to this problem other than to avoid downgrading
with older firmware. Should APM110‐C1 decoders come in contact with old firmware the
decoding capability may become inoperative.

 

[zdjh22]

Just a heads up regarding "return shipping". They don't ship it back fully insured. They only cover up to $1,500. What this means if it a warranty return gets lost is anyone's guess. When they ship out new units what it means is if it gets lost, it covers their cost to build you another and ship you that one. For a unit under warranty repairs though? James told me to make sure my unit was fully insured when I shipped it, which was $4,275. For the return trip he was acting like it was going to be my responsibility to handle the shipping back to me if I wanted it to be fully covered. Haven't heard anything more on the subject since May 19th though, so I guess I'll find out more in just over a week...

Never have I dealt with such customer service like this in my nearly fifty years of existence.

Regarding the firmware though: unless the new motherboards are simply incompatible with prior firmwares, you might be able to just use a previous one. I hope to god I don't get sent back a buggy unit that I can't use for who knows how long until they do yet another new firmware. But since they said they had to replace hardware, that's likely what my unit will be saddled with as well. So I'll functionally still not have a working unit.

Buried in the 2.15 release notes, on the last page, is this discussion that indicates that new units cannot employ firmware before 2.15.

 

[sander99]

Ah, I only now realise it concerns both the different APM decoder bord and the different ARM processor...

 

[esimms86]

Buried in the 2.15 release notes, on the last page, is this discussion that indicates that new units cannot employ firmware before 2.15.

Does this have any effect on compatibility of future firmware versions with the vast majority of existing A16’s using the original APM 110 and ARM?

 

[zdjh22]

Does this have any effect on compatibility of future firmware versions with the vast majority of existing A16’s using the original APM 110 and ARM?

I would doubt it in the short to medium term. Ensuring backwards compatibility is straightforward, as long as (1) there is sufficient flash memory space in the device for any extra code involved, or there is a process during the firmware update to selectively install packages tailored to the hardware present, and (2) there is intent by Smyth Research to do so, given that there are software development costs associated with supporting multiple flavors of the embedded hardware.

I bet that if Smyth Research survives (and that’s a big “if”), in the long term older hardware will get locked out of new firmware, much as you see with other devices like smartphones and computers. But I think this will be many years out.

IRL I was involved with software in data acquisition for physics and astrophysics experiments that had very long lifetimes (10 to 15 years). We managed to keep development environments viable well past the end of the experiments, although sometimes heroic measures were involved (virtual machines running very old instances of the operating system and development software, and/or maintaining very old computer hardware).

 

[VandyMan]

The firmware and drivers for the APM board are provided by Momentum Data Systems; so the continued support will largely be up to them. However, unless there is a new protocol (HDMI 3?) that needs to be supported, I don't think there will be an issue for a long time.

 

[GeorgeA]

The streamer Eversolo DMP-A6 used in conjunction with Realiser A16 can bitstream multi-channel DSD and PCM audio signals (up to 5.1) via HDMI cable. As multi-channel DSD signals aren’t compatible with the Realiser A16 (“support is not imminent”), my attached pictures show D2P (DSD to PCM) multi-channel output from Eversolo DMP-A6 as rendered by my Realiser A16 unit.

 

[zdjh22]

The streamer Eversolo DMP-A6 used in conjunction with Realiser A16 can bitstream multi-channel DSD and PCM audio signals (up to 5.1) via HDMI cable. As multi-channel DSD signals aren’t compatible with the Realiser A16 (“support is not imminent”), my attached pictures show D2P (DSD to PCM) multi-channel output from Eversolo DMP-A6 as rendered by my Realiser A16 unit.

From the APM-110 data sheet, DSD decoding is supported by that particular board (which is used in the A16). Interesting that it is not enabled. Perhaps a licensing issue, and the cost would be too high? Certainly similar and higher bitrate signaling over HDMI for DTS, DD, and Atmos is already working in the A16, so it’s not likely a technical issue related to the frequency of the signals.

 

[Litlgi74]

The streamer Eversolo DMP-A6 used in conjunction with Realiser A16 can bitstream multi-channel DSD and PCM audio signals (up to 5.1) via HDMI cable. As multi-channel DSD signals aren’t compatible with the Realiser A16 (“support is not imminent”), my attached pictures show D2P (DSD to PCM) multi-channel output from Eversolo DMP-A6 as rendered by my Realiser A16 unit.

Just a reminder... If you want to hear 5.1 as it's intended... You really should use a listening room with Ls and Rs virtual speakers. ±100–120°

 

[GeorgeA]

Just a reminder... If you want to hear 5.1 as it's intended... You really should use a listening room with Ls and Rs virtual speakers. ±100–120°

Got it. I was so curios to see how those two devices can work together.

 

[GeorgeA]

From the APM-110 data sheet, DSD decoding is supported by that particular board (which is used in the A16). Interesting that it is not enabled. Perhaps a licensing issue, and the cost would be too high? Certainly similar and higher bitrate signaling over HDMI for DTS, DD, and Atmos is already working in the A16, so it’s not likely a technical issue related to the frequency of the signals.

Excerpt from FAQ page at https://smyth-research.com/frequentlyaskedquestions/

Q. Will the A16 support SACD (DSD)?

Yes in conjunction with the higher sampling rate PCM modes we plan to support 2-ch DSD input via HDMI and will decode internally to 32-bit floating point PCM @176.4kHz before processing by the SVS algorithm. Headphone output will be analog and SPDIF 24-bit @176.4kHz. DST is not currently supported by our HDMI interface so multi-channel DSD support is not imminent.