The study above does not impress me. There was no quantitive assessment of the jitter of the equipment under test. High jitter there could mask smaller jitter simulated. And then we need to hope that simulation of jitter is accurate enough and source material does not have high jitter.
Regarding ASRC, I may not fully understand the benefit yet. Jitter on source would result in a doet of harmonica effect of ASRC. Only when combined with long buffers and very slow PLL I could imagine it would surpress jitter... !?
I would not be too concerned. The HFD software is actually pretty easy to learn. When I say it is slow and cumbersome, that comes from my perspective of developing more than 80 DSP filters using HFD. Even with all this experience, I have not found any shortcuts to entering a filter. click this, click that, enter numbers here, here, here and here, now click this again... It's a lot of keyboard and mouse work compared to modern apps.
Imagine if you had to work with a 20 year old PC running Windows 2000, Office 97, and Netscape browser... It would seem slow, clumsy, and annoying. But you would be able to figure it out pretty quickly.
To be fair, I should say that HFD is better than most of the engineering analysis software I have used. Early versions of Labview, and Matlab were very cryptic.
j.
"Harmonica effect" only on the ASRC input. The ASRC makes the output clock independent of the input clock (and any possible PLL).
While I haven't used it myself I came across the presentation of the new JBL Pro SRX 900 series - it once again seems Pro audio is way ahead also on the GUI side of DSP config and system overview. Also would be nice if DIRAC measurements and setup in miniDSP would provide a similar transparency to the user. But that is probably not going to happen anytime soon. 
I hav removed a direct link to not violate any forum rules.
But look it up on youtube (JBL Professional SRX900 Series: Technical Overview)
and start at 5:35 (it's just the h-control part that relates to this topic, so you can ignore the buzz words and stop at 7:40) I was unable to find a better short overview of the GUI, but here it is for inspiration
I hav removed a direct link to not violate any forum rules.
But look it up on youtube (JBL Professional SRX900 Series: Technical Overview)
and start at 5:35 (it's just the h-control part that relates to this topic, so you can ignore the buzz words and stop at 7:40) I was unable to find a better short overview of the GUI, but here it is for inspiration
Last edited:
How would an harmonica effect be avoided then? My assumption on ASRC approach would be that input rate is estimated with PLL. Then the samplerate convertor would be configured with ratio between estimated input and known output rate. In case of (high) jitter, estimated input may vary and hence ratio would vary over time. If this would occur, I wonder what kind of aliasing effects this would have and how this relates to effects of normal jitter...
OK, now I get what you mean by "harmonica effect". Yes, that is the "A" part. as in output clock is independent of input clock. Data comes in at whatever jittery rate the source provides, output is clocked at output clock rate. You don't need any PLL "estimation" - if the buffer is getting emptier, you change the conversion rate slightly, but because of the buffer, the rate of change can be very slow.
Indeed. I think still some sort of control algorithm is required, and buffer fill state can perfectly be used as input for this. But in the end, the questions are:
1. How often is the (sample rate conversion) ratio changed (after run-in)?
2. At what level does jitter affect how often the ratio is changed?
3. What are the audible effects of ratio changes?
4. How do these effects relate to normal jitter effects?
I would think that ratio changes could give somewhat similar problems as jitter, though the update is at (much) lower frequencies. Or perhaps it is comparable with low frequency jitter / phase noise.
Hi all, does anyone know or has tried what happens when the SP/DIF Output of the Fusion master module is just splitted to two individual slave modules via Y-adapter instead of daisychaining? Will everything - especially volume control - work this way though this mode is out of specification? That may help to avoid some meters of cables....
Thanks!
Best Regards
Peter
Thanks!
Best Regards
Peter
Y-adapter is not very good idea as SPDIF interface is made for 75 ohm impedance cables/lines. When Y-adapter is on the line, impedance cannot be any more 75 ohm. But SPDIF receivers can be so sensible (usually they are) that signal will go thru and will not be lost, only "little" jitter will be added. It means you newer konw without testing. If SPDIF generally will work (sound goes thru) then also volume control will work as it is implemented on the SPDIF software layer.
Hi all, how to calculate gain for a channel in the Hypex Studio? I have FA123, I looked for an example and found project
http://www.seas.no/index.php?option...=583:seas-kingro4y-mk-iii&catid=66&Itemid=250 (accidentally also coax with subwoofer). There are gains: 0,5dB for subwoofer, 2dB for mid and 3dB for tweeter. How comes it does not clip?
http://www.seas.no/index.php?option...=583:seas-kingro4y-mk-iii&catid=66&Itemid=250 (accidentally also coax with subwoofer). There are gains: 0,5dB for subwoofer, 2dB for mid and 3dB for tweeter. How comes it does not clip?
I'm not sure I understand.
Are you asking?
- Fusion amp
- Subwoofer drivers crossed and driven by fusion amp
- Analogue output crossed by the fusion amp
- Mains amplifier
- Mains speaker
- Mains amplifier
Its not how the amp is meant to be used. If you have a separate sub for each channel and use a 2-channel or more fusion amp it can, however, be done but you'll have to do some modding to the fusion amp cables. You'll have to redirect the analogue input to one of the fusion amp channels and use it as an output. In this case I would also run the fusion amp module bridged to not waste output power. It is not a plug and play fix though since you will have to analyze the PDF and create your own PCB header cables.
Thanks. Yes that’s what I would like to do.
It would be a great solution, but I don’t think I have that kind of skill…it’s a pity because it would be great to have this feature and control all the speakers from the fusion amp.
(I think dayton amps have this function but no the same sound quality)
Also your workaround doubles the output power, is there any drawbacks about this?
It would be a great solution, but I don’t think I have that kind of skill…it’s a pity because it would be great to have this feature and control all the speakers from the fusion amp.
(I think dayton amps have this function but no the same sound quality)
Also your workaround doubles the output power, is there any drawbacks about this?
The obvious more simple way to do about that is to skip the mains amplifier and use the fusion amp instead.
Then you won't have to fiddle with the PCB headers but can just add speaker level outputs to the sub for channel 2 (and 3 if you use the 3 channel modules)
You still need a separate fusion amp for each channel of course.
In general, all preamps can be used - but of course you have to check for correct input resistance, gains and voltage output on the lines. I haven´t checked in the case of the Hypex FA123, but I remember a lot of cases where home-audio equipment was used to drive pro-audio equipment and the voltage outputs just were too low... So better check in advance If the FA123 has Cinch Input, you are omst probably fine. Also, check if you have balanced out/in or not..
If the FA123 has Cinch Input, you are omst probably fine. Also, check if you have balanced out/in or not..