Very interesting comparison. I had never considered the SGM to be focused primarily on upsampling to DSD. But this model developed for bit perfect it. could be a great alternative.
A new star in 2019 : Innuos Statement Server
- Thread starter CKKeung
- Start date
Hi Wisnon,
It’s adjustable, 5/7/9/12v 2a but have been assured it will run at 12v 2.5a continuous for my router.
I ordered August for October delivery but delayed due to various problems.
Cheers
Blue58
Hey Blue 58,
The PH SR4 is a very nice LMPS.....I’d be surprised if it didn’t deliver a substantial uplift in SQ, especially if its replacing one of those nasty little SMPS that are typically supplied with routers.
I would be interested to hear how it goes and I’m especially interested to hear if the GA cable makes a difference as I have 4 unscreened DC cables on my network.
Do bear in mind that these power supply upgrades usually deliver a nice initial uptick in SQ followed by several weeks of deminished performance as the unit burns in. I’m guessing that the burn-in process is extended due to the low voltage and current draw....usually not enough to even warm up the heat sinks.
Hi,
My name is Nuno Vitorino and I'm the R&D Director at Innuos. Just to clarify a few comments with regards the ripping engine.
Our ripping engine is based on cdparanoia and adjusted to the offset of our specific drive. It will only rip bit-perfect - if it doesn't it will place the CD on Quarantine and report a damaged track.
We don't use the AccurateRip database because, unfortunately, it may be misleading. If it says it's an accurate rip, that's indeed the case. But there are multiple cases where it can report non-accurate rip when in fact it's perfectly ok namely:
1. Cd not on the AccurateRip database
2. A different pressing of the CD
3. A slightly different version of the CD
This would cause unnecessary user anxiety in that the system was reporting a non-accurate rip. So we prefer not to use it. A better alternative IMO would be CueDB as its algorithm would solve 2 and 3. However, it's a challenge enough to get metadata from most CDs nevermind being in the CueDB database. Our TEAC drive also features AccurateStream technology, which ensures a consistent ripping. AccurateRip was, IMO, more relevant in the past where drive offsets could actually vary during the rip, causing jitter. These do not and, as previously said, the offset has been adjusted to our drive.
As for ripping in AIFF, our WAV ripping already contains all necessary tags. We want the system to be as simple as possible and want to have only one fully uncompressed format. WAV would arguably enjoy better compatibility than AIFF. However, if user feedback would say to use AIFF instead of WAV, it's an easy change for us. We already support playing and editing AIFF anyway.
Hope this helps clarifying our options.
Nuno,
Good to see that WAV files can now be properly tagged, as that wasn't always the case, hence my early adoption of AIFF as preferred format, as its tag support was comprehensive from the start.
Thanks for the in-depth explanation about the ripping engine too.
Good to see that WAV files can now be properly tagged, as that wasn't always the case, hence my early adoption of AIFF as preferred format, as its tag support was comprehensive from the start.
Thanks for the in-depth explanation about the ripping engine too.
Nice to read from you - welcome to WBF!
It seems we have different views on this matter - I prefer the risk of the anxiety
caused by a few pseudo negatives due to absence than living permanently in the ignorance for the whole library. BTW, 2 and 3 are also very improbable, as AccurateRip keeps records of different pressings and versions and checks all of them.As I could see that no Linux ripper supports AccurateRip, probably the best way to surpass this problem would be checking a backup copy of the library using an Wisdows machine. Do you know of any software that checks an whole existing library with AccurateRip?
Nuno, thanks for chiming in. For some reason digital was my gravitation about 8 years ago when I got back into audio. I bought a Core Audio unit which was one of the first linear PS units. I was so impressed at the time I continued to follow one of the founders and use a full Mojo Audio suite today.
I have always admired the Innuos gear and consider it to be a leader in the pack. Good on you all for developing leading software in ripping and playback technology, as well as optimized computer boards and PS. I don't think there are many builders that use proprietary MB at this time. I'm sure its a significant production cost.
IMO, digital playback has leaped in performance at such a rate, the average user just does not understand and has a hard time accepting why things are as they are. I think these forum are great, but also lead to a lot of confusion. It's a lot of opinion based upon very limited experience. There is also no best practice at this time. Take for instance server technology. There is a one box and two box approach. There is native file and upsample topology. I dont see much literature on the Statement yet. Alex is indicating it is designed best for native, non upsampling pcm playback. I have always been sceptical of upsampling. Can you talk to how the statement is designed to surpass upsample performance with lower sample rate native pcm playback.
Thanks
I have always admired the Innuos gear and consider it to be a leader in the pack. Good on you all for developing leading software in ripping and playback technology, as well as optimized computer boards and PS. I don't think there are many builders that use proprietary MB at this time. I'm sure its a significant production cost.
IMO, digital playback has leaped in performance at such a rate, the average user just does not understand and has a hard time accepting why things are as they are. I think these forum are great, but also lead to a lot of confusion. It's a lot of opinion based upon very limited experience. There is also no best practice at this time. Take for instance server technology. There is a one box and two box approach. There is native file and upsample topology. I dont see much literature on the Statement yet. Alex is indicating it is designed best for native, non upsampling pcm playback. I have always been sceptical of upsampling. Can you talk to how the statement is designed to surpass upsample performance with lower sample rate native pcm playback.
Thanks
UPnP Server is indeed Asset. It's not possible to install MinimServer easily - this would require root access and would need extensive testing to ensure no issues would happen with our platform. We value staibility quite a lot with our platform. -We did discuss with Simon this possibility when we first developed innuOS but there were some requirements that we could not fulfill at the time. Asset does work very well with a wide variety of streamers (UPnP can be a real compatibility pain) and one thing we really like is the automatic scanning whenever new music is added. In the future we may talk again with Simon and discuss the possibility to integrate it.
Nuno
Nice to read from you - welcome to WBF!
It seems we have different views on this matter - I prefer the risk of the anxiety
As I could see that no Linux ripper supports AccurateRip, probably the best way to surpass this problem would be checking a backup copy of the library using an Wisdows machine. Do you know of any software that checks an whole existing library with AccurateRip?
Thanks for the warm welcome. Do bear with me as won't be following up forums thoroughly so I will only add my views every now and then.
We've long given up trying to please everyone - there are always some compromises one must take when doing a product for such a diverse audience - from technophobes that want little more than a "CD Player with storage" to advanced users which we could hire ourselves
. This is why there is choice and people should choose the solution that best fits their needs. As said before, we don't feel the need to check the AccurateRip database for the aforementioned reasons. But we can agree to disagree...
Nuno
Thanks for the kind words. The proprietary motherboard was indeed a significant project and, in our tests, resulted essentially on a "blacker" background. This is not too surprising as the main effort was removing EMI sources on the board.
There is a bit of confusion about the upsampling capabilities of the Statement or any of the MK3. PCM upsampling is quite light on processing and can be easily achieved on any of our servers. So upsampling to 768MHz is no issue at all if your DAC supports it. Upsampling to DSD takes a lot more processing particularly anything above DSD128. Our MK3 and Statement were able to upsample to DSD256 with the multi-core option in Roon. However, this may have been removed (or maybe the setting changed location, we are still looking) in Roon 1.6 which means only one core is use and in this case it can handle only upsampling to DSD128.
There are some design issues we came across when using faster processors on our systems - these are explained here: https://devialetchat.com/Thread-Innuos-Zenith-ripper-streamer?page=2
Regardless of the above, there are probably not a lot of DACs benefitting from DSD upsampling anyway. A good NOS DAC will benefit from PCM upsampling on the source, an FPGA based DAC will normally convert DSD to PCM internally anyway so upsampling to DSD will actually make it worse. Other DACs convert everything they receive to DSD so might as well let them do that. This is particularly true for higher end DACs. On a number of DACs we have tried, we have failed to get better sound quality from DSD upsampling than we would get from PCM upsampling. Feedback from our customers support this trend - they started initially with upsampling and as they went along, they reverted to simply using a bit-perfect output as clean as possible and let the DAC do its job.This is how we design our products. It's not the Statement that "sounds good", the Statement "simply" sends in the cleanest and best timed signal it can get so the DAC can do its best. Having a plain source connected to a great DAC is a bit like trying to use regular petrol on a Formula 1 car...
In the end, it's the same as any Hi-Fi component - try it out and check what sounds best to your ears. The upsampling capability is there and you can decide for yourselves.
Nuno
Thanks for your reply
Can you tell me if Asset correctly handles the metadata in waw files? Minimserver does it very well, Twonky (the server installed by default in my Melco) does it horribly.
We use ID3v2 tags on the WAV files and Asset handles that perfectly. This is similar to what dbPoweramp does.
Will your unit accommodate convolution files?
In one of my systems I currently use Roon into HQ Player with four files (two for each of two channels) to perform convolution. If I had to, I could use the convolution functionality built directly into Roon rather than HQ Player for that. In my system correction in the digital domain aids time alignment, volume alignment, and low-frequency room correction.
Thanks for the warm welcome. Do bear with me as won't be following up forums thoroughly so I will only add my views every now and then.
We've long given up trying to please everyone - there are always some compromises one must take when doing a product for such a diverse audience - from technophobes that want little more than a "CD Player with storage" to advanced users which we could hire ourselves
As said before, we don't feel the need to check the AccurateRip database for the aforementioned reasons. But we can agree to disagree...
Nuno
Thanks. Can you tell us how your software confirms the accuracy of the rip? Does it make multiple reads to validate them statistically?
I can attest that it does, much to my surprise. My previous experience with WAV files were probably like yours, not very good, that's why I asked for AIFF support with the Innuos folks. But once I used Asset, I noticed that metadata was recognized perfectly, same for album art. And the auto-scan is a great benefit as well.
The Statement sounds excellent! I have a demo unit which I don’t expect to leave my system.
I’ve been using it for the past 3 days, having changed from an Aurender N100h. Anybody who insists that all servers sound the same couldn’t miss the differences. My wife heard differences.
I am using the Statement as Roon Core. And streamer. Music is on a NAS and will eventually be imported to the Statement. Tidal is my on line source.
I’m no expert when it comes to high end digital. I can’t compare this piece with other high end digital pieces. Just the N100h. Bottom line is that relative to the N100h the Statement is more detailed, throws a larger soundstage and image, sounds more 3 dimensional (N100h flatter) and carries less of a digital sound signature. It had made my speakers sound larger and richer. It’s a way more open sound, top to bottom.
While the N100h made me feel I was sitting in the 6th or 7th row at a concert, the Statement makes me feel I’m in the front row.
I continue to listen to familiar tracks and am awed by what I’m hearing.
Plain and simple, it works. Really well.
Thanks to Dave from Audio Doctor for getting me this demo unit!!
popspin
I’ve been using it for the past 3 days, having changed from an Aurender N100h. Anybody who insists that all servers sound the same couldn’t miss the differences. My wife heard differences.
I am using the Statement as Roon Core. And streamer. Music is on a NAS and will eventually be imported to the Statement. Tidal is my on line source.
I’m no expert when it comes to high end digital. I can’t compare this piece with other high end digital pieces. Just the N100h. Bottom line is that relative to the N100h the Statement is more detailed, throws a larger soundstage and image, sounds more 3 dimensional (N100h flatter) and carries less of a digital sound signature. It had made my speakers sound larger and richer. It’s a way more open sound, top to bottom.
While the N100h made me feel I was sitting in the 6th or 7th row at a concert, the Statement makes me feel I’m in the front row.
I continue to listen to familiar tracks and am awed by what I’m hearing.
Plain and simple, it works. Really well.
Thanks to Dave from Audio Doctor for getting me this demo unit!!
popspin
Hello popspin,
Congratulations!
BTW pls transfer/copy your music files from your NAS to the internal storage disk of the Statement.
Playing the music files this way is significantly better sonically than playing them inside the NAS.
Files are being imported. Next step is to see if Roon will pick them up automatically or if I need to tell Roon that music is on the Statement now.Hello popspin,
Congratulations!
BTW pls transfer/copy your music files from your NAS to the internal storage disk of the Statement.
Playing the music files this way is significantly better sonically than playing them inside the NAS.
Curious; if there is a buffer why does music sound differently played off the local SSD and buffer vs a NAS and buffer?
Similar threads
