 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
108.234.77.201
Anyone using this? I can do it at the file level, real time on the pc or a setting in the dac. Dac will do up to 384 IIRC. Content is 44.1 native. Any point to upsampling and which method would be best? which rate?
Cut to razor sounding violins
Follow Ups:
Ok. Well it was a fairly short test. 1st off I couldnt get files past 176khz to play. I made some at 352 and they played but were super slow. I used DBP power amp and R8brain to make the conversions and both results were the same. Dac and playback support over 192k but for some reason it just didn't play properly.
2. I would have to use USB for anything about 96k. I am fairly certain that doing that would sound worse than any benefit of upsampling.
3. I forgot that I switched to a sd card for the music library and it taps out at 128gig. So yikes no way I can do a whole collection in upsampled files.
Also I tried audio nirvana. Very weird. I had tried it about a year ago I suppose on a completely different computer. Like 3 computers later I installed it and it said the trial was expired. I seems like I just put the old hd on the new computers...but I didnt. I certainly didnt do any back ups and this computer was created for work so its crazy to me that it thinks I had this on this computer. Its possible I forgot but I didnt, this is a work computer and by design the music is on another pc. Weird.
Anyhow 3 strikes and dawnrazor is out...
Cut to razor sounding violins
"2. I would have to use USB for anything above 96k. I am fairly certain that doing that would sound worse than any benefit of upsampling. "
Why do you say that?
Basically because of this post and the other posts it mentions. If you look at the rankings to get usb to be on top you have to do all kind of crap.
https://www.head-fi.org/threads/usb-strikes-back-watch-out-aoip-usb-ethernet-chain-beats-all-at-least-for-me.829639/
More flexibility too. WIth dante I can just send sound to multiple devices without worry. Not sure I can do that with usb. Since I have 2 sources its way easier with dante of I want switch. Also I plan to do some recording and have a Dante architecture is just easier to incorporate that.
Also I can isolate the computer electrically from the dac with ethernet/ dante. Not so much with usb.
Cut to razor sounding violins
"If you look at the rankings to get usb to be on top you have to do all kind of crap. "
And you believe what you read on the internet from posts that are 6 years old? OK.
Its more than just that one post if you look there are multiple threads on different sites and I read most of them. And then I jumped in and well have been happy with the sq from these inexpensive studio internet devices. My recent experience seems to jive with those posts. Its its not like I just read something and said OK there you have it. If I did that I would be using a nondedicated computer with a cheap usb dac because its all bits is bits and all dacs sound the same blah blah blah.
Also it seems like usb is being subplanted with ethernet for audio anyhow so I am just being future proof too IMHO by avoiding usb.
And its just way more flexible with the network. I can put sound where I want it. With USB I can only send sound from one computer to the usb connection. WIth the ethernet I can send sound from either computer to any of my network dacs. That can come in handy occasionally if one computer is updating or something. Or if I want to compare 2 different dacs, I can feed them the same source, etc.
Cut to razor sounding violins
Synchronous upsampling is good, for music digitized at a sample rate at less than 50 kHz...... But it must be synchronous.... DSD is a form of synchronous conversion, but avoid variable rate "ASRC" chips, which embed noise artifacts into the signal. (At least the industry has abandoned asynchronous sample-rate conversion, in most part.)It would be better if one knew the nature of the type of "filter" being utilized in the upsampling..... Different types of music tend to be optimized best with different types of filtering..... I think simple folk music doesn't need upsampling at all at 44.1 kHz sample rate. I think long minimum phase filters are ideal for complex (orchestral) music...... A short minimum or linear phase filter works best for most music.
For sample rates of at least 88 kHz, I believe upsampling isn't necessary, because there is ample room between the 20 kHz and half the native sample rate, to use purely analog filters..... The simpler the digital conversion, the smaller its "RFI footprint" is.
Edits: 01/05/20
and I'm currently experimenting with it on my system.
But I'm only doubling the sample rate on 44.1 and 48 streams from QOBUZ.
Nothing else.
My DAC is set to '0' up-sampling where what it receives is what it processes.
I do with all my music that is stored on hard drives. I use jRiver Media Center 26 and upsample to DSD 128. I think it sounds slightly better over native rate.
My only advice is to use multiple of 2 upsampling, so 44.1 to 88.2 or 176.4 and 48 to 96, 192 or 384.
I do all of my upsampling in Audirvana using the SoX routines built into it. I've also tried PCM to DSD conversion but my DACs apparently don't all take DSD and convert it back to PCM. When I did a small test with the one DSD DAC I have I couldn't hear a difference between it and the non-DSD DAC playing upconverted DSD or just 24/96 native. I've since turned all upsampling off.
My own advice is to just sit back and enjoy.
I use JRiver and its upsampling of red book files seems to improve SQ in my set-up.
Thanks. I will hopefully feel like listening this weekend and see what happens.
Cut to razor sounding violins
It depends.There are very good reasons for doing upsampling.
But this depends
1. on the DAC you're using.
2. on the external resampler qualityThere are numerous DACs out there doing internal upsampling/oversampling.
They pretty much all come with reconstruction filters inside.Both parts the up-/oversampler and the filter have impact on the sound - positive and negative.
What you need to know is how your DAC is processing the data.
Some use internal DAC chip technology for resampling, some use ASRCs, others FPGAs... And there are also still NOS DACs out there.Without knowing how your DAC is processing the data you won't be able
to approach the subject properly. Of course you can try and see what
you end up with.Some DACs do internal resampling to 352k8/384k. There are DACs where the internal resampler gets bypassed at 352k8/384.
In this case you'd IMO have a very good chance to improve the sound by using an external highest quality upsampler. Highest Quality. That's the key of course!Further @ these high samplerates filters can be configured much less aggressive. This usually translates into much lower filter artifacts.
No filter is lossless, no resampler is lossless. And that's a reason why manufacturers apply these high samplingrates at all.
The DSP jobs inside the DAC are usually (quality) limited. They can't achieve highest quality DSP. That's a typical bottleneck.
With an excellent external tool you might be able to beat the DAC internal DSP/filters, if the DAC allows to bypass its bottlenecks.
On a lot of DACs (e.g. ESS Sabre) it simply makes no sense to feed 384k since they run internally much higher samplerates. And feeding 384k externally would mean more losses by the exteranl resampler and losses
by processing much higher rates in the chain (CPU load etc.) .It's not that simple.
Good luck.
Edits: 01/03/20
I liked the point you made about needing to understand how the DAC works in order to come up with an informed approach. With some combinations you could have overlapping filter transition bands with unpredictable results.
My one quibble is about high sample rate filters being less aggressive. You said: "Further @ these high samplerates filters can be configured much less aggressive." That's not really so. Filter slope can be set independently from sample rate.
I'm assuming the goal of software upsampling is to be in control of the reconstruction filter. If you have 44.1k in and 352.8k out to the DAC, that will take the DAC's reconstruction filter out of the equation in most cases. But the upsampling filter will require a stop band at 22.05k to be "ideal", and you'll make the same tradeoffs between stop band rejection and impulse response that you would if you're a DAC designer.
Thanks for throwing a ton at me!
I know very little about this dac. Its the Stello Da100MKII:
Now, April Music team is proud to introduce the most advanced D-A converter, stello DA100 MKII, which supports both DSD files (2.8M & 5.6M) and PCM files up to 384kHz/32Bits (via its USB port).
Driven by proprietary FPGA (Field Programmable Gate Array) module (AMAP : April Music Audio Processor), which is programmed with the specially designed algorithms by April Music team, DA100 MKII reproduces the digital signals into a very natural but still analog-like sounds.
Sound is tuned not only by the measurements but also by the ears of the many people in April Music's tuning team.
Connect your favorite digital sources like PC, CD Transport, Blueray/DVD players or TV etc, and enjoy the highest sound ever which can be equaled only by several times more expensive D-A converters in the market.
With DA100 MKII, digital signal is not just a combination of 0 and 1. It is a flow of natural signal that we call 'Music'.
DAC ES9018K2M Sabre32 Reference Audio DAC
Dynamic Range 127dB
THD+N -120dB
So some answers I suppose. There are all kinds of menu options:
1. Rate. Options are BYPASS 44.1 all the way to 384
2. Conversion- For converting PCM to DSD mode. You have 2 options, BYPASS and DSD.
3. F Filter- Sound processsing filter for PCM source. Options are Fast and SLow
4. High Frequency Cut Off Filter. Options are OFF, 50k,60k,70k
5. PHASE- 0/180
Lots of variables...
For a converter for creating upsampled files I was planning to use DBPoweramp.
I don't have a box that can do the conversion, and am not excited about real time in the computer.
Cut to razor sounding violins
That looks like a typical approach of eliminating the DAC chip filters
and adding their own stuff.All the info you have at hand now won't get you any closer to answer the actual question of how the DAC (FPGA + DAC chip) is handling the stream though. It won't get you any hint if Upsampling would have any impact.
Probably you simply try it. Take some of your reference files and convert them.
I'd use sox btw. It's also available on Windows. (I'd guess that's the system you're using.) sox IMO offers reference SRC quality.I wrote an article some years ago.
https://soundcheck-audio.blogspot.com/2011/04/resampling-if-you-cant-avoid-it.html
I keep it updated (especially my preferred settings) if necessary.
You'll find my preferred upsampling settings for sox over there.
Good luck.
Edits: 01/04/20 01/04/20
I like to up-sample PCM to DSD. My FiiO M11 Pro has the capability to do this and I've discovered that it takes the digital edge off of PCM when it's up-sampled to DSD. Anyway, it sounds better to me.
Best regards,
John Elison
I will try the dsd setting and see whats up.
Cut to razor sounding violins
There is no right or wrong here. It's a matter of personal preference only. Upsample and determine if YOU like it.Aside from some experimentation, I never upsample. I also tried on the fly PCM to DSD conversion and determined that for MY needs, I just play the file as is.... but that's just me.
Edits: 01/02/20
yeah I agree and do plan to listen for myself. I Used to upsample to 176 IIRC. Hardware changes made me just do 44.1. But now with the new dac it has more options. HD space can support upsampled files if I want. But the dac has DSD rates over 192k and I don't have any experience with that and was asking just to save me some time to narrow down the testing.
Cut to razor sounding violins
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
