Thanks! Good to hear there don't seem to be any obvious flaws in FA's volume handling. This gain removal in the pre-amp is a smart trick. I need a lot of headroom for my woofers so I'd be more likely to resort to individual channel attenuation.
A tip from my days when I dealt with super efficient full range speakers - Duelund resistors can't be beaten for that. I can't justify the cost of their caps but those resistors are worth every penny.
I'm definitely not advocating for analog over digital but I think it is not so black and white. Digital, just like analog, can be done better or worse. Try comparing ESS chip in a good DAC against iTunes as the source in the same setup. Good example is Gustard DAC too, where they included both analog one step drop and digital volume control. Having said that, I'm leaning towards full digital path with the Fusions.
Ironically, that's the point that we were trying to make. It is ASR afficionados who often want to limit the quality to just SINAD reading. Believe it or not, I sometimes read their tests and I value the results for what they bring to the picture, which to me is value added and not the core assessment. I'm yet to find the test which translates to how realistic the instruments sound or how good imaging is. Let alone how well it will integrate with my gear. It is not ignorance of science on my part, it is the ignorance of reality amongst many of ASRers. It can sometimes be amusing too, e.g. total disregard of tubes. But generally I agree these tests bring more value than sponsored articles in most magazines.
I have found that with any gear that tests audibly transparent (and there is a lot of gear in that category), how realistic the instruments sound or how good imaging is totally depends on the recording, speakers, room, position of the listener in the room - and mood of the listener.
How many car magazines discuss steam engines?
My RME UFX is 12 years now. Plays like day 1 and get's a few software updates the year.
At least the USB/Firewire interfaces have some anti-jitter tech built in which works extremely well. Don't know the name, just have a look on their homepage and search if your card has this tech built in too. Not sure about the built in ones. Look here:
https://www.audiosciencereview.com/...yface-pro-fs-portable-interface-review.12313/
If you go with a digital source, you would want to send the signal with the highest output possible that still leaves headroom. This is to saturate all bits. I set the piCorePlayer to 90 percent. When I used the internal alsa mixer equalizer, this headroom was necessary to avoid clippling. I have however never experienced this since I use CamillaDSP instead to give me physical tone controls. The attenuation is done with Hypex IR remote, I never manipulate the source level. My startup volume is set to -40 dB in Hypex Filter Control, which is a good low level listening volume with my 94 dB efficient speaker.
That's EXACTLY the same you do here. Hypex plate amp does the volume control before D/A conversion. So it doesn't matter if you control the volume in the computer or in the DSP.
But don't worry - these amps have all their noises and artefacts so low you can not hear them any more.
With an analog signal you also have a noise floor and dynamic range - you can't hear details smaller as the noise floor! Similar is with digital - all your great 24bit information would be thrown away anyways when leveled down analog cause it's burried in the analog noise.
As long as your resolution is high enough and noise is low enough digital volume control works great!
(Actually you can hear "into" the noise floor. There is dithering. This Test CD has a level of -100dBfs recorded with dither which you can fully reproduce with 16Bit: https://www.discogs.com/de/release/8312637-Various-CD-1-Test-Disc
But that's not easy to understand ... )
But don't worry - these amps have all their noises and artefacts so low you can not hear them any more.
With an analog signal you also have a noise floor and dynamic range - you can't hear details smaller as the noise floor! Similar is with digital - all your great 24bit information would be thrown away anyways when leveled down analog cause it's burried in the analog noise.
As long as your resolution is high enough and noise is low enough digital volume control works great!
(Actually you can hear "into" the noise floor. There is dithering. This Test CD has a level of -100dBfs recorded with dither which you can fully reproduce with 16Bit: https://www.discogs.com/de/release/8312637-Various-CD-1-Test-Disc
But that's not easy to understand ... )
I think you are not using your Fusion optimally this way. Fine if it sounds good enough for you, but still...
First of all, 50 dB is anyhow quite a lot. It is a loss of about 8 bits. Do you ever use the -25..0 dB range during louder listening? If not, there is too much system gain.
Secondly, I would always recommend to attenuate at end of audio pipe, not at begin (presuming clipping is avoided). By doing this on PC side you will worse SNR or more rounding errors in ASR and DSP processing...
And input is 24 bit, so with a bit more than 8 bits loss you are getting below 16 bit resolution already at start of audio pipe!
The tweak can be different per channel, so you could do it only for mid and tweeter and keep woofer gain high. But I would first tune the speaker with software attenuation and see where you end. For me it works with same preamp gains as I do a lot boosting. But of course you need to watch out for clipping at high levels this way.
Yes, I was also considering the Duelands. Maybe ever I will try them, or make carbon resistors myself with pencil stifts. Or try the Mundorf foil resistors...
You can change the gain of the amp module in the plate amp if you never use full power. This will lower noise floor a few dB. In the 3-way modules the 100W amp is already better by 2dB when I remember right.
There are ome things you can do with a passive crossover you can't with active (dampen THD of a resonance, lower power amp noise). But I would not go as far as dampening the whole speaker, especially when system noise is already not audible.
There are ome things you can do with a passive crossover you can't with active (dampen THD of a resonance, lower power amp noise). But I would not go as far as dampening the whole speaker, especially when system noise is already not audible.
@JukkaM wrote that it is for when the kids are sleeping and he is still watching TV. Do those 8 bits really matter in that situation? It will still drown in the background noise.
Usually not an issue as the DSP works with greater temporary precision.
Commercial recordings, studios and domestic listening rooms don't have a noise floor exceeding a 16 bit range.
@Julf got me covered
I posted this scenario to demonstrate the benefits of high SNR and overall superiority of digital interconnects in an extreme use case. My most used range is in -30 dB ± 5 dB range, when audio oriented friends come by or when watching movies with high dynamic range it can rise to -20 and above. There was one occasion where I set it to 0 attenuation, but that was with separate mixer controlling the volume and it was a private disco night where I wanted to test the capability of my sound system, which is made from PA stuff. The FusionAmps were not the weakest link there.
You can be happy when a "loudness war recording" get's 10 bit of dynamic...
Btw - the ear only has 60dB of "natural" dynamics! It "switches gear" with the auditory ossicle to get at lower and higher range. That's the reason you can't hear silent stuff for a short while after loud signals. Psychoacoustics ... I love it! (You can read about masking effects, lot's of useful information and explains a lot)
Indeed. That is how perceptual encoders (MP3, AAC, Vorbis etc
work. I guess JJ Johnston (one of the fathers of MP3 and AAC) isn't on this site - he is an endless source of information on psychoacoustics. Some of his excellent presentations are online.I prefer transparency over cover ups... but ok...@Julf got me covered
If the noise is inaudible, the system is transparent.I prefer transparency over cover ups... but ok...
I was mainly talking about something else, but....
IMHO seeing everything as added noise is an oversimplification. The calculation errors in ASR chip can be data dependent. Same for PC side, but probably here at least some dithering is done to solve this, but at the cost of even more noise...
Anyhow, everybody is free to use their Fusion amps in whatever preferred way of course...
IMHO seeing everything as added noise is an oversimplification. The calculation errors in ASR chip can be data dependent. Same for PC side, but probably here at least some dithering is done to solve this, but at the cost of even more noise...
Anyhow, everybody is free to use their Fusion amps in whatever preferred way of course...