That's certainly a possible longer term plan for me IF I ever get to the point where I feel the analog path in my 4x10HD is limiting things. Most days I doubt that will happen.
Please explain. I am somewhat worried that you are joining JC and RNM in dismissing things as Midfi now. Clearly they are rubbing off on you.
Why does a USB to spdif adapter even have asrc's? The host pc can handle those tasks in that application.
The Sigma Studio and Shark chips are very powerful and the ADI tools can do a lot. Unfortunately ADI won't let you expose them to users so Minidsp.had to encapsulate them and let don't have as much capability. It would be fun to add the flanger. . .
The Sigma Studio and Shark chips are very powerful and the ADI tools can do a lot. Unfortunately ADI won't let you expose them to users so Minidsp.had to encapsulate them and let don't have as much capability. It would be fun to add the flanger. . .
Hi Bill,
Remember that some "midfi" stuff has rather good performance. SOTA in disguise. 🙂
-Chris
Remember that some "midfi" stuff has rather good performance. SOTA in disguise. 🙂
-Chris
The only potential reason to use an ASRC there would be if the USB receiver operates in synchronous or isochronous mode, which is a design deficiency anyway.
I mean, if they are using it as a "jitter cleaner".
Last edited:
If you want to run it to another box you could make a small PCB with LVDS transceivers and add an RJ45 connector. Analog Devices also makes LVDS isolator ICs I think, if you felt the need to isolate.
ASRCs are on the SPDIF inputs. Can't control the clocks on external source devices, so have to accept whatever comes in.
Last edited:
Yeah, I don't blame you, it's probably not worth it IMO. Richard wants to push the limits though, and it is definitely possible to improve in the measurements department.
I don't usually use the term, midfi, but I would not purchase the minidsp given the choice of SRCs. I suppose I used the term out of disappointment. For short time I was thinking it might be nice thing to have.
It looks like the same part but might have a smaller LUT. I would rather have the SRC4392 yes, but I have my doubts that it's audible. It certainly bothers my inner perfectionist that they would use it to save a whole $3.
That is how I feel also. I just cant use a heavily compromised toy. If they dont want to put in better parts, better interface nd software, I will do it for them. Or just go find something else. This may be DIY, but I dont hear many saying I am going to use the cheapest part or opamp I can find for my project. We use the best we can each afford within our budgets. That isnt the attitude I find in these commercialized products, however. Its just the opposite, I seems.... too much emphasis on the budget side just takes the joy out it.
THx-RNMarsh
THx-RNMarsh
Last edited:
I did not realize the adapter was full duplex. Asrc's make sense at the input. I see the inputs as a patch to support older stuff. The host would be the main source normally.
These days virtually all the audio power amplifiers we use have DSP built in. American made ones are pretty much using stuff from Scott's old friends.
Never looked inside the lower cost ones made in Singapore. As the cost of the DSP audio power amplifiers is dropping bellow $500.00 stand alone boxes are are hard to justify.
Keep in mind economics have pretty much doomed small power amplifiers. By the time you get a case and power supply there is not much difference in the number of output transistors and heatsink.
Derfy,
Why do you think cables can't have directional properties? BTY it has nothing to do with semiconductors or diode action.
Never looked inside the lower cost ones made in Singapore. As the cost of the DSP audio power amplifiers is dropping bellow $500.00 stand alone boxes are are hard to justify.
Keep in mind economics have pretty much doomed small power amplifiers. By the time you get a case and power supply there is not much difference in the number of output transistors and heatsink.
Derfy,
Why do you think cables can't have directional properties? BTY it has nothing to do with semiconductors or diode action.
And for the application off USB in to SPDIF out that part isn't used, so what is the issue for that?
You not wanting it is fine, but I still fail to see the problem. Given you by admission listen at a level of around 65dBA (assuming that is for -20dBFS) and you room noise level will be at best 20dBA IF you are holding your breath whilst listening, then stuff at -124dB is so far below 0dBA as to be not worth worrying about.
But for USB to SPDIF it still appears to do what you want?
Bill,
You need a larger envelope to do those classic back of envelope calculations.
A symphony orchestra peaks 30 dB or more above the average spl level. One can also hear tones buried 20 dB below the noise floor.
Result is 95 dB of perception range maybe more.
So using a simple rule of thumb, double that to be sure! 190 dB!!!!
(Yes I know, that was humor, or my attempt at it.). 😉
You need a larger envelope to do those classic back of envelope calculations.
A symphony orchestra peaks 30 dB or more above the average spl level. One can also hear tones buried 20 dB below the noise floor.
Result is 95 dB of perception range maybe more.
So using a simple rule of thumb, double that to be sure! 190 dB!!!!
(Yes I know, that was humor, or my attempt at it.). 😉
Ed, do you mean peak as measured with an SPL meter or peak as in crest factor? I get confused easily. Also in hall the orchestra peaks that high, but can you find any examples of then not gain riding things in recording or mastering so we get the full experience?
100dB of perception range I could buy* if you listen at THX ultra levels. that still means the 'midfi' SRC has 25dB to spare for reproduction?
* With the usual caveats on masking and ear recovery time.
100dB of perception range I could buy* if you listen at THX ultra levels. that still means the 'midfi' SRC has 25dB to spare for reproduction?
* With the usual caveats on masking and ear recovery time.
Bill,
If I had the minidsp I would want to connect an ADC, maybe some people wouldn't care. If digitizing phono, some people might use a SPDIF ADC for that.
Regarding SRC4392 vs SRC4382, I have a fair amount of experience with SRC4392, and I hear some distortion that I don't know the exact cause of. After I noticed that I heard from someone else the same basic observation. In both cases it was using the part for SPDIF in and out, which is not the only way to use it. I am also aware that DAC-3 uses an SRC4392 and I am not aware of the same distortion in it, but I also don't know how they are using it. Given the distortion I have experienced, I would be reluctant to commit to SRC4382.
Also, distortion in SRC4382 is given as down at -125dB, down around where if I play around with distortion compensation with ES9038Q2M I hear some slight change in brightness of music. As I said before, I suspect what is audible with music is possibly IMD from a lot of mixed signals. Don't know about SRC chips with IMD and music, but my inclination would be to play it safe and opt to spend $3 per chip for lower rated distortion.
EDIT: I should probably add that my peak levels are not necessarily up at 0dB FS, or perhaps even very close to it. They would likely be at least at -20dB FS with an ADC to avoid any clipping, and the sound at that point would be uncompressed.
Last edited:
Bill when the standard sound level meter reads 65 dBa an instantaneous peak hold meter could show 95 dBa. Or to produce 1 watt average energy would require 1000 watt amplifiers.
So when the band plays louder that peak power requirement might go to 100,000 watts. However your actually don't require that because as the average level goes up your hearing perception shifts to a lower sensitivity.
So with a reasonable efficiency loudspeaker at what I would consider reasonable listening levels 100 watts or so is fine.
Even if you want to listen at 90 dBa average level then one's preference would be for more efficient loudspeakers and require about the same amplifier power.
So when the band plays louder that peak power requirement might go to 100,000 watts. However your actually don't require that because as the average level goes up your hearing perception shifts to a lower sensitivity.
So with a reasonable efficiency loudspeaker at what I would consider reasonable listening levels 100 watts or so is fine.
Even if you want to listen at 90 dBa average level then one's preference would be for more efficient loudspeakers and require about the same amplifier power.
- Status
- Not open for further replies.