SPL Volume8 Review

[wisechoice]

Wow, very interesting. Can you tell me what possible negative effects that might have, if any?

Our wiring here is very old, so old that it crumbles in the fixtures if you touch it. No ground. We live on the third floor of a multi-unit building so there's also no hope of changing that. But all audio connections are balanced.

Also, is there any point in doing this if I can't hear the power supply noise when nothing is playing?

 

[EnceinteAcoustix]

I did this to see where the noise problem was coming from. This modification means that this device no longer has electrical safety. Do not reproduce this if the electrical installation of your house is not good.
Regards

 

[wisechoice]

OK, thanks. Did you actually hear the noise?

 

[gandalfandula]

The DDRC-88D, Okto and Ultralite Mk5 all have volume control knobs although the Okto is by far the nicest. The Mk5 knob is rather small and the DDRC-88D doesn’t have a display (although one can be DIYed rather easily with minidsp-rs). I’ve been meaning to DIY a better knob for my Ultralite Mk5 + CamillaDSP setup but just haven’t gotten around to it. I have some nice rotary encoders on hand but just need to write some code.

I am just not sure when the SPL makes sense. If you are doing multichannel you are a probably sticking with your AVR for volume control.

For stereo DIY active it is not hard to beat the performance of the SPL with a decent DAC. Both the MOTU Ultralite Mk5 and the Okto dac8 pro are lower noise than the SPL. Heck a RME Fireface 800 from the mid 2000s (available for ~$400 used) can actually beat the Okto in terms of residual noise if used on the lowest output setting. If you are adverse to DIY or using the DAC volume control I think it makes more sense to add a digital output miniDSP Flex on the front end of your system for volume control / indication. Will definitely give you better system SNR with any decently low noise DAC than the SPL.

Michael

I too was interested in this for its application to active multi-way stereo listening. Unfortunately the performance doesn't seem to be worth the cost in this case. The advantage to a high quality analog domain volume attenuation, as I understand it, is that you could benefit from some additional gain from an analog preamp(s), e.g. the benchmark LA4 rather than just going straight out of a good DAC like the okto that you mentioned. This might be important if you've had to reduce your headroom due to DSP adjustment in the digital domain.

 

[Andysu]

at £800 for 8 channel and what appears be D25 connector input/output and has no other function . my Sony SDDS D3000 can act as 8ch fader with same common D25 input/output and other inputs and only cost one for £65 and one for £80 . go figure . £800 for common D25 computer connectors what is the hi-fi world coming too ?

not fully bench tested to see how far off the input/output voltage levels are i do suspect some divination and that can easily be corrected so all levels shows equal frequency voltage levels .

so it's like a not good design for home cinema remote fader that all cinemas use . and often remote cinema faders can be cheap some even think they are worth £400 for CP650 remote fader ?

remote cinema fader 6ch for Dolby CP500 connected to back of processor for few simple pin wires . CP500 has it's own actual remote fader but as long as correct vairable resistor used , it worked for the moon landings so shouldn't give me a problem .

 

[mdsimon2]

I too was interested in this for its application to active multi-way stereo listening. Unfortunately the performance doesn't seem to be worth the cost in this case. The advantage to a high quality analog domain volume attenuation, as I understand it, is that you could benefit from some additional gain from an analog preamp(s), e.g. the benchmark LA4 rather than just going straight out of a good DAC like the okto that you mentioned. This might be important if you've had to reduce your headroom due to DSP adjustment in the digital domain.

Ehh, having a DAC with higher output voltage does the exact same thing and pretty much any multichannel DAC with balanced outputs has plenty of output for amplifiers of typical gain.

I am using Hypex NC252MPs with 1.66 V input sensitivity with my Okto which has 4.2 V output. Because I am using the Okto volume control I attenuate any boosts in DSP so I never clip in the DSP. Due to the output voltage if I go above -8 dB on the volume control it is possible to drive the amplifiers in to clipping. Important to note that there is no free lunch here, you can either clip the DSP or clip the amplifier and although I think clipping the amplifier is a better idea I have not really gained any headroom.

Michael

 

[wisechoice]

Regarding the poor measured channel separation on the Volume8, it is utterly inaudible when I mute any given channel in software, turn off the other speakers and put my ear directly to the tweeter of the software-muted channel. This is with the volume dial at 100.

The same is true when passing the signal through an old parallel printer AB box (this one: https://www.amazon.ca/Way-Switch-Box-DB25-Female/dp/B000I9B2K0). Which I've read is the main concern with not using a specially made AB switch for audio.

To be honest, I thought that crosstalk this bad would be audible, but it certainly doesn't seem to be. If someone can propose another listening test with which to detect it, I'm all ears, so to speak.

I also have an ADI-2 Pro FS R BE arriving tomorrow (!) and can take some measurements with it at some point with the AB switch.

 

[wisechoice]

I just noticed that @amirm's crosstalk measurement here starts at 1kHz and I'm not sure why. Since I might be measuring a few other aspects my signal chain, I wonder if someone might be able to illuminate that. I'm very new to all of this.

 

[Rja4000]

he higher noise floor shows up as a constant as we sweep the input level:

@amirm
I'm not sure I understand this plot.

Is the Pre90 used with a fixed level of 0dB and you vary the ingoing signal level ?
(I read "Pre90 0dB" in the legend)
Or are you using the Pre90 volume control to vary the level ?
(Like you do with the Volume8)
And then what's the test signal level ?

 

[wisechoice]

Compare with the final graph (channel separation) to see what a manual adjustment looks like when it’s being measured through the range of volume attenuation.

[EDIT: I've been corrected about this by @Rja4000 -- previous and following posts deleted to avoid needless confusion]

 

[Rja4000]

Compare with the final graph (channel separation) to see what a manual adjustment looks like when it’s being measured through the range of volume attenuation.

That gives you no idea of SINAD vs manual level

 

[Rja4000]

Compare with the final graph (channel separation) to see what a manual adjustment looks like when it’s being measured through the range of volume attenuation.

This last graph gives you attenuation over time.
So, of course, it is somehow chaotic, since it includes Amir's manual interaction.

The one I questioned is using volume level (Vrms) as the X axis. I don't expect to clearly see the impact of the manual interaction here.
I've done it myself, and, as you can see in below plot from my own measurements of the Volume2, it's hardly visible

Here is a trend of THD+N and THD vs volume setting

---

.

 

[wisechoice]

This last graph gives you attenuation over time.
So, of course, it is somehow chaotic, since it includes Amir's manual interaction.

The one I questioned is using volume level (Vrms) as the X axis. I don't expect to clearly see the impact of the manual interaction here.
I've done it myself, and, as you can see in below plot from my own measurements of the Volume2, it's hardly visible

Here is a trend of THD+N and THD vs volume setting

---

.

Thanks, that makes sense. I now realize your question was only about the Pre90 measurement.

Sorry for the extra confusion.

 

[wisechoice]

The legend is less readable on the new version because you’ve used squares and not lines.

You’ve labelled the vertical axis, which is good.

Your horizontal axis is better labelled now, but it would be clearer if you made it “measured output level.”

Also, you have a typo in “output leveld.”

 

[Rja4000]

The legend is less readable on the new version because you’ve used squares and not lines.

You’ve labelled the vertical axis, which is good.

Your horizontal axis is better labelled now, but it would be clearer if you made it “measured output level.”

Also, you have a typo in “output leveld.”

So which one of those 3 do you prefer ?
NB: I removed it above because the data doesn't match my previous post
This new graph is probably using wrong data.
But it doesn't matter to judge the readability.

1.

2.

3.

 

[wisechoice]

The first one is the most readable in terms of grid spacing and shading. But I'm not even sure you need this level of detail on the grid, tbh. Because the numbers on the vertical axis are so much bigger than the spaces between lines, it defeats the purpose of the very fine distinctions that you would make about, e.g., where maximum THD+N can be found, the steepness of the slope, etc. I can tell where the measurement lands on the line, but it's pretty hard to trace that back to an actual number.

 

[Rja4000]

So, isn't that easier with this one, after all ?

 

[wisechoice]

So, isn't that easier with this one, after all ?

View attachment 222623

Probably, although I might ditch the grey shading. Perhaps you misinterpreted my comment about squares vs. lines, though? I just meant in the legend.