Minidsp sampling frequency.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

I am interested in purchasing a minidsp plus miniamp set up. I would like to know one thing first. It seems the sampling frequency of the minidsp is 48KHz. So what happens when I input in a spdif from a cd player. Does it convert the 16 bit 44.1KHz signal to a 28 bit/48KHz signal first before it starts messing around with the signal?

Oon
 
Hi,

A few more questions.

1) If I use the USB streamer board for a minidsp+amp, I don't need the minidigi board anymore right? It can receive input from USB and SPDIF, correct?

2) Most of my material will be from a CD player. Can the Minidsp run at the native frequency or at a higher sampling rate 96KHz. From a hifi perspective that is less inaccuracy to the signal.

3) The Miniamp has a PWM out for subwoofer. Do you have an application circuit on how to utilise the PWM switching signal out? External transistors perhaps?

Thank you.

Oon
 
Oon,

In answer to your questions,
- the miniDIGI indeed will convert any digital input signal it receives (up to 192kHz) to 48kHz thanks to the Asynchronous Sample Rate Converter (ASRC) that is provided on the input. No need to worry about bits. :)
- Note that the miniStreamer isn't meant to really integrate to the miniDSP + miniDIGI. While a ministreamer + miniAMP will work easily, a ministreamer to MiniDSP configuration isn't as plug&play and not recommended without a miniDIGI (taking care of the ASRC).
- The miniDSP only runs at 48kHz I'm afraid.
- We don't have any app note on how to use the PWM output. You would need an output power stage to get that PWM output to power anything. Not some resistors.. We actually never got that request to implement the PWM out... we could look into it later on next week.

Hoping this info helps,

DevTeam
 
thanks for the prompt reply. I have some slight concerns on the resampling. If the resampling is done at a much higher frequency, I would be more comfortable. I suppose the system has some form of interpolation when it is resampling to guess the values in between the dots. Bypassing the dsp is also disturbing, since I have no way of implementing volume control.

Having an external board with the suitable transistors, similar to that from a tripath ta 3020 would be nice too. Then I can easily use it in biamping arrangement with the stereo pwm subwoofer out for bass.

Oon

Oon

Having a

Sent from my Milestone using Tapatalk
 
will mini dsp ever have a 192khz sampling rate ? i was thinking of using it as a digital crossover with my cambridge audio 740c cd player , spdif is 192khz.
We could indeed get it to operate to 192kHz with a new plug-in on the miniDSP 2x8. Whether it will happen or not is a bit early to say...

Only the 2x8 platform currently has enough power to process data @ 192kHz though. And if we ever decide to do it, we would only be left with a 2x4 configuration (1/2 of the current 96kHz)... Anyway, we just released this platform, so gotta give it some time. :)
 
Hi,

I purchased the minidigi/minidsp/miniamp combo recently. I am quite happy with the sound so far.

however, I feel if 96KHz can be implemented at a sacrfice of some funtionality via a plug in, you should create such a plug in. I have the 2.1 plug in and I feel some of the functionality is overkill, peq before and after crossover... for example. One is sufficient.

by not putting it in , you are alienating a lot of hifi buff. It would be the same reason. Would you buy 48KHz dac? You would buy an nos sampling dac (44.1 KHz) because it is the original data. You buy a 96KHz dac because it is upsampled. But if you buy a 48KHz dac, you would have lost original data due to resampling but it is not a significant higher sample rate.

Oon
 
Hi minidsp dev team,

I had a thought, if you could just change the sampling rate to 96KHz by just a plug in, simplifiying the there is a lot of possibility in this. You just need to relabel it as a DAC with built in DSP and crossover. There is a lot of new possibilities with this. A DAC like this will allow us to use Valve amps on the high pass and class D on the low pass. But you need to keep the perception of quality high, a 48KHz won't cut it though.

Oon
 
FYI, the topic of higher of higher sampling rate has "widely" been covered before and I don't think that we're going to say more than what we initially said before.. :)

96kHz would mean 1/2 the processing power of what you have and end up with 1 EQ band, no 48dB slopes etc...So it's just not a solution for 90% of our users who are asking for more processing, not less..:) If you can't live without 96kHz, then look at the miniDSP 2x8 which was designed with a bigger processor for that matter. That's all we can say.

DevTeam.
 
Its not the higher sampling frequency per se, its the corruption of the data when the data is resampled from 44.1KHz to 48KHz that is bugging me. When a data is resampled at frequencies which is not multiples of the original, data is interpolated to form new sets of data and the original data is thrown away.

The datas are calculated by either linear regression or some form of polynomial fit. The new data point is now guessed from this fitting. Since the original data are of a different sample frequency than the new one, they are discarded. However if you ran at a much higher sampling frequency, the datapoints are much closer together, so although you loose the original datapoints, the new datapoints are of much closer to the original ones and you have the added benefit of having a lot more datapoints. You have to remember that at frequencies near 20KHz there is only 2 datapoints, one on the positive cycle and one on the negative cycle, it is already very bad. It's okay if you are resampling frequencies at 2KHz, where there are tens of datapoints, I don't think you could hear the difference. And if you are resampling the data, that means the data is completely guess work. So here you are spending all the money to get a good CD player with all the bells and whistle, stable mechanism, to ensure the integrity of the data, and then you throw it all away at the DSP. The whole idea of going this route is I dislike the amount of losses that occurs from D to A then all the lossy analog stuff along the way such as volume control etc, then it gets converted back to digital in class D. For all I know, I can't hear the effect of the resampling. But you will always have the nagging feeling you have already corrupted the data so much. Incidentally what algorithm does minidigi use to resample the data from 44.1 to 48 anyway?

I don't mind if it can be run at 44.1KHz. I wouldn't mind at all if CDs are run at 48KHz. But they are not, if you run Digital audio tape, you would get 48KHz sampling rate, then I wouldn't be making any noise. 96KHz is a compromise.

I am not suggesting you change all plug-ins for everyone to 96KHz. I am suggesting you create a new alternative plug in for people who don't need so complicated a process to allow them to use 96KHz. Give the users a choice. There can be plug in stereo crossover 2.1 standard and stereo crossover 2.1 96KHz. I actually only want a stereo 2 way crossover with 24 dB slopes. A PEQ with 3 band would be nice, but not a necessity.

Oon
 
I can't hear the difference between 48 and 44.1 kHz sampling. If I did I'd be disappointed that I bought to soon. None the less, as they have said they cannot/do not want to make such a plug in and that you can purchase the 2x8 which has all your looking for. It seems more of a neurosis, this dwelling on what "might be but isn't heard" obsession, than it does a real concern especially since you admit you cannot hear the difference. Sorry if I crossed a line but its the psych nurse in me coming out.
 
Whether one can hear the effects of resampling or not, depends on the rest of one's equipment, as well as one's ears and of course how you conduct the tests. As anyone will know, speakers make the most differences. If you are using a low end budget british speaker such mordaunt short, mission etc or if you are using a pair of mini monitors which costs a few thousands. Not to imply that you have crap equipment,but there is wide range of equipment on this forum, from college students to people who's other set costs a 100k and sits in its own room. So I don't know rest of your equipment and hence I can't comment... The difference will of course not be as obvious as a change in peq setting like a 6db boost at 55 hertz. It might result in suttle change of sound like, warmer, improved imaging, loss of detail, wider soundstage etc. For example I have a dac that can have the oversampling turned off or on, and the sound is different. One has nicer highs, more crisp, the other has more robust mids, better detail. Many people attribute it to the brickwall low pass filter of dacs, which messes around with the signal too much. That is why NOS (non oversampling dacs) are still so popular with the audiophiles

Oon
Sent from my Milestone using Tapatalk
 
Dear Theresa,

Don't take what I said in the post earlier the wrong way, but I have encountered all too many case of people saying they can't tell the difference between 2 amplifiers on their 50 dollars Sony speakers. :eek:

The thing is I don't know anything about the set up you had use to evaluate the criteria, I cannot gauge the accuracy of the statement. And I have no idea your level in hifism, that I have to take everything with a pinch of salt, since you can be anywhere between a college student to a reviewer from sixmoons.com. But of course if you told me that you are a reviewer from sixmoons.com than it would have been a completely different story, I would definitely take your word for it. But my experience of DACs tells me they generally sound a little different, and I guess that is why there are so many models out there and a lot of it has to do with the DSP along the way. So I find it a bit surprising that you find no audible difference between a resampled and non resampled sound.

Incidentally how did you do the evaluation, and the signal was switched between passing the minidsp and without passing the minidsp? was it an indepedent DAC followed by amplifier and speakers. I can't figure out how to do this for an AB comparison, since I can't get it to go 44.1 KHz to begin with...:(

Oon
 
Member
Joined 2007
Paid Member
ah look up how asyncronous sample rate converters work and see that there is no difference between using 96Khz and 48Khz sample rates from 44.1KHz*. ;) as for number of samples per cycle >=2 fully represents the signal please look up:
Nyquist?Shannon sampling theorem - Wikipedia, the free encyclopedia

rember corectly applied maths dosen't lie but ears do!


*The process applied is an optimisation of upsampling to multi GHz sample frequancies and then downsampling
 
My equipment includes an Emotiva UMC-1 Pre/Pro, 2 unbalanced miniDSPs, 1 balanced miniDSP, Emotiva XPA-3, Emotiva UPA-5, Emotiva UPA-2, Behringer EP4000 for 2 subs, etc. Speakers include actively crossed over Etons and ScanSpeak Revelators (both bought as "kits") and ERD-1 Emotiva surrounds. No, its not high end but is very good regardless. When someone claims they can hear what science says they can't then I account that to psychology, not to objective fact, unless they have objective test results agreeing with their assertions. In my case what I do and don't hear agrees with the science. I'm certain that if I leave some lighted candles and incense burning while listening it will make the sound more enjoyable and accessible, but this is a psychological effect (which is perfectly valid) not an objective one.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.