HQPlayer - Can it improve the audio performance of my system?

Audio Gear Talk HQ Player
21

I have tried many digital hardware solutions from the very inexpensive, to the expensive, as well as pretty much all the popular software solutions both free and paid.

HQPlayer embeded is the gold standard imo. I will not even consider a server unless HQPlayer embedded can be installed. Installing the embedded version is not something for the faint of heart, and Jussy does not provide support for non-professionals installing it outside of comments on forums.

HQPlayer plays extremely well with Roon. Everything Roon does with music (up sampling, volume, etc) HQPlayer does better. Having the choice of filters and dithering methods allows you to really fine tune the sound of your system and allows you to use a filter that plays the best with the reconstruction filter on your DAC. Assuming your PC is very low noise, the HQPlayer volume control is about as good as a digital volume can be, and will be better than most preamps until you get into expensive gear.

Treat the cost as being the cost of a component.
I strongly recommend going all in, and using a dedicated media server with HQPlayer embedded rather than a general duty PC. There are a few commercial offerings, the Sonictransporter is an excellent choice, If you know your way around Linux, you can install the embedded version.

3 Likes
22

There’s an even simpler option for Embedded.

HQP OS. Burn the image to USB stick and it runs off that. Or you can burn the image to SSD drive.

1 Like
23

I think you’ve missed my point. That is, what’s benefit is there in doing these things with HQP is the DAC resamples everything?

24

Few DACs do that. Usually if you feed a DAC that internally upsamples to 352.8/384k at those rates, it bypasses the upsampling part.

From upsampling point of view, point of HQPlayer is to squeeze best possible performance out of DAC by using the DAC with most optimal input format.

Bit-perfect DACs are the best choice for HQPlayer, such as Holo Audio or T+A through it’s DSD section. Or DACs based in TI/BB DAC chips are always bit-perfect using DSD. Also many AKM chip based DACs support DSD Direct mode of the DAC chip.

DACs that use some DAC chip and don’t have a separate DSP in front (most DACs on the market), we know how the chip in question behave and can choose best strategy for it. ESS chips bypass internal upsampling when input is 705.6k or higher, DSD streams are remodulated without rate conversions.

DACs that use some custom DSP in front, we can choose the strategy. For example Chord DACs bypass WTA1 when running at 705.6k or higher, but should not be fed with DSD.

Many people choose DAC based on what is best match for use with external upsampling such as HQPlayer.

This is still not taking into account all the benefits you get from things like running digital RIAA correction for vinyl sources, digital room/speaker correction, digital headphone correction, headphone cross-feed, dynamic loudness, etc.

And of course in any case you need some kind of player for playing your digital content. HQPlayer works fine as a standalone player, or as DSP playback engine for Roon.

2 Likes
25

Again, I was giving the OP the heads-up, as the DAC wasn’t specified.

But not always.

Hopefully, the OP has enough information to make an informed choice.

In my view, understanding the current “differences” rather than adding complexity should be the first step.

1 Like
26

That is very small minority of the DAC market. Of course, in such cases one can replace the DAC for a better match.

I don’t know what differences you are talking about. Nothing such in the original posting. But the HQPlayer part is easy to understand through digital domain analysis and measurements from the DAC output. Even if the DAC is doing some resampling even for 352.8/384k inputs, unless it is somehow horribly mangled.

27

Aaviiks own app sounds better than roon…

28

But that has nothing to do with HQPlayer which is the topic here.

I don’t know what kind of DSP processing some Aaviiks (what is that?) is doing. But in any case it is not relevant here.

1 Like
29

It has everything to do with it. It’s the only reason the OP is considering HQP. IMO, helping the OP understand why there is a difference is more important since there should be no difference.

1 Like
30

Why there shouldn’t be? Do you know what kind of DSP there was going on in these cases?

Did you even measure there were no differences before making such statement? And what measurements did you perform?

But maybe not the reason why HQPlayer would sound different. So you cannot draw any conclusions regarding HQPlayer from some other comparison.

In any case, easy to test and costs nothing but time and effort to try.

31

No. That’s the point.

You should read the thread, since you appear to be barking up the wrong tree.

32

So you cannot either claim that there shouldn’t be any difference.

Or maybe you, because you are trying to conclude something from three unknown things that a third unrelated thing which is topic of this thread is somehow related.

So I feel that you are barking up the wrong tree.

1 Like
33

This isn’t going anywhere. I’m not claiming anything.

34

Ok gentleman, let’s cool it a little please.
No need to let things get heated here.
Thank you for your patience and understanding.

3 Likes
37

Hi all,

To help steer the topic back on track I’ve changed the topic’s title.

HQPlayer - it worth the money?

This was way too general and hence it would be difficult to reach any satisfactory conclusion, beside I believe the @Frank_Helenius2 is seeking specific advice about his system … not a general comment about the value of HQP (that after all is a very subject topic).

Thus I’ve amended to read…

HQPlayer - Can it improve the audio performance of my system?

I agree with @Martin_Webster it is important for the OP to understand what the differences are between using Roon and Aavik, this would give a base to help steer any future system changes.

@Frank_Helenius2, unfortunately your opening post did not contain enough information to get any specific advice, but you I see your have since supplied more details on your system, thank you for that.

I had a cursory read of Aavik Website but it was not clear to me which products you are using, I think it would help people here if you could provide full system details including how everything is connected.

Going back to the HQP question, I think @andybob offered some wise advice, in that you your trial HQP on your system. HQP is extremely powerful, you may love it, you may not. It also extremely complex and you should be prepared to invest time to get the best from it.

2 Likes
39

I attempted to implement this solution. I must say that it is a Byzantine option that is far beyond the fund of knowledge for the vast majority of users, so I would encourage anyone without a robust understanding of digital audio processing to take heed of this.

At present I lack the proper instruments to test the claim that the DSP engine in HQPlayer does “everything better” than Roon’s own. I’m very adept at DSD upscaling and conversion in JRiver, and have spent endless hours tweaking a multitude of settings and filled a few hard drives full of the same album or tracks converted with all assortments of settings, without much in the way of audible results—other than volume variations or small dynamic range widths. In the vast majority of cases, JRiver’s automated settings produced a perfectly satisfactory and optimal result, and whatever tweaks I made produced imperceptible results.

Based on the features I’ve found in HQPlayer, I highly doubt this costly and labor intensive option offers any audible advantages over Roon’s own, very user-friendly algorithms. The latest delta sigma DACs, particularly those built on ES9038pro, ES9039pro, AK4499 and its recent successor chipsets, all offer extraordinarily sophisticated and optimized oversampling algorithms, and are capable of processing native DSD signals with no interim PCM conversion.

The analog signals these DACs are outputting are the closest approximation that can be achieved of the original master at the time it was digitally sampled, varying only in the degree of distortion, SNR and dynamic range produced—and for the past half decade at least, these factors are all well beyond what we’re capable of discerning. Therefore in the era of the “modern” delta sigma DAC, we’ve achieved perfection from a practical standpoint of the technology, and the quality of the sound we experience is primarily limited by the quality of the source material, with few exceptions.

I’m perfectly happy with the DSP engine that is available without surcharge in Roon. And I want to take strong exception that one should invest in an outlandishly costly Holo May or similar DAC to achieve the “benefits” of HQPlayer—I take exception to the purchase of these costly DACs in the first place. Digital audio technology continues to progress at sequentially lower costs, as is true of any computational device available to consumers today, from cell phones to portable computers and tablets. And with each passing year, it becomes possible to produce progressively more sophisticated wares affordably and cheaply. DACs are no exception.

Whether one purchases a $359 SMSL SU-9 Pro with the latest ES9038 Pro chipset or an eight year old $14,000 DAVE, the differences in sound quality between the DACs themselves have been proven to be beyond the range of human hearing. We’ve achieved the technological limits of what a DAC can do, based on the limitations of the human ear, and in 2023 the DAC is the component in our digital audio systems that should cost the least. Advances in tech have offered us the gift of saving our hard earned cash to invest in our headphones, where quality differences are likely to be audible, and the money spent worthwhile.

40

There are many points in the text above that I disagree with. It is presented as factual when it really is just opinion.

I would argue that the oversampling engines built into the list chips are not “extraordinarily sophisticated”. Rather, they are limited by design choices dictated by the modest processing power of the DAC chips themselves. HQPlayer is not limited in this way and is limited only by how much processing power the user chooses to provide for HQPlayer.

Pragmatic perfection? What is that? All I know is that the idea that DACs don’t affect the quality of what we hear is ludicrous.

If you are happy with the Roon DSP engine and a less expensive DAC, I say more power to you! That is certainly your choice and I wouldn’t suggest you should change a thing. After all, your opinion of what sounds best is the only one that should matter to you.

The SMSL SU-9 Pro sounds nothing like a DAVE or a Holo Audio May…or a PS Audio DirectStream or even a Schiit Yggdrasil. You can believe that all DACs sound basically the same if you want. But my ears tell me much different story. So do the ears of many music listeners. I most definitely trust my ears over the opinion you are presenting here as fact.

In my case HQPlayer with the Holo Audio May KTE gives me sound quality I have never experienced with a DAC using something like the ES9038 Pro chipset.

I would tell anyone looking for improved sound quality from their digital music to look hard at their DAC choice and to consider HQPlayer. It’s been nothing short of a revelation for me and my listening enjoyment!

2 Likes
41

Hey, you’re welcome to disagree, I welcome anyone’s opinion without any ego about it. That’s what I think these forums are for.

I clarified the language of my poorly-worded “pragmatic perfection” to illustrate my actual point, which was that we’ve achieved the technological limits of what our ears can process with modern DACs. That’s not opinion, that’s empirically proven fact. Any further improvements in DA conversion that increase dynamic range, lowers the noise floor or reduces distortion have been inaudible for the past several years—we’re incapable of hearing much beyond a 96db dynamic width, which means the best ears can’t discern much beyond 18 bit sound sources,

This doesn’t mean that I don’t enjoy having gear that can produce 768 kHz or DSD 512–I own such gear in the Neo Stream and my Topping d90se, and I like the fact that I have high resolution sound sources available to me, even if I can’t hear their differences.

However, your assertion that the Holo May or the DAVE’s DAC sections—meaning only the DAC itself and not any bundled preamp, amplifier, filters etc—sound different than far cheaper ESS or AKM based DS DACs is not borne out by the scientific or empirical data. In fact, we know that R2R ladder DACs, as an example of often expensive wares, have far lower dynamic ranges and far more noise and distortion than delta sigma DACs, and we’ve proven multiple times through rigorous measurements that the eight year old Chord DAVE performs much worse than an inexpensive option such as an SMSL SU-9 Pro at literally 1/30th of the cost.

Of course the DAVE is not a pure DAC; it is an integrated DAC/Amp, and I never said it would sound the same as the SMSL paired with an amp. But Chord can get away with saying it’s better to the tune of 30 times the cost, even though its technology is eight years old, because our ears can’t discern the differences between what its DAC outputs compared to the SU-9. That extra distortion and noise the DAVE produces isn’t audible, and in fact whatever distortion it does audibly produce might in fact have some components of even order harmonics that some folks like. That’s the principle of valves and vinyl—although digital distortion is much more likely to be odd order harmonics that are decidedly unpleasant. But who knows, maybe Watts got lucky.

The debate between those who believe in audio science and those who believe cost defines quality will continue to go on forever, and I’m at peace with that. But as someone who is a scientist by profession, and who has invested considerable effort to educate myself about audio technology for the very purpose of protecting my wallet from marketing hype, I feel a sense of responsibility and purpose to share what I learn with others in the hopes that they might not make the same costly mistakes I’ve made on my own journey. I don’t try to stop others from posing other arguments; it’s up to each person reading these forums to decide which points of view resonate with them and to make purchasing decisions accordingly.

That being said, if someone does make claim that a piece of gear that costs as much as a Holo May sounds superior to something that measures as exquisitely as a Topping d90se or an SMSL SU-9 Pro, both of which cost several orders of magnitude less, I’m going to ask them for objective proof. “I can hear a difference” is not an acceptable response, any more than it would be for any product for which one has already invested a fortune. And I’m particularly suspect of any “premium” manufacturer, such as Chord or PS Audio, who don’t release their own measurements. Shouldn’t anyone be, when being confronted with those asking prices?

Anyway, to get back to the topic, I’m open to hearing what specific advantages this HQPlayer offers over Roon’s native DSP engine. I see lots of complicated selectable tweaks and options, but no evidence that the end result is objectively, audibly improved enough through the effort of playing with them to justify the extra work and expense. If you have any data supporting these endpoints, I would welcome them! Peace…

2 Likes
42

Topping D90SE is pretty buggy with various awkward behavior with certain input signals (I have it and I have measured it in various ways). My worst 900€ spending so far, I still regret wasting that money.

Now I know even better, how buggy ES9038PRO is, but I also know some workarounds for it’s bugs too.

But since it has just the poor stock ESS digital filters, it can be significantly improved with better external filters.

AK4499 has DSD Direct mode where the DSD input (raw delta sigma data) bypasses all it’s on-chip DSP and goes straight to the analog conversion stage. Allowing much better external processing.

2 Likes
43

I have the Topping d90se and have found no bugs. I’ve experienced none of the negatives you’ve mentioned and I don’t have any objective evidence to support the negative claims you make about the device or its chipset. The objective evidence I have in favor of it is its published and multiple replications of its extraordinary performance on the bench—although I think there are newer options available that perform equally well at less cost.

Which one of our opinions is correct? And what does your ad hominem point have to do with the Sygnalist question I posed?

Correction—the Topping d90se does have some objective problems with dropouts when paired with certain devices through its coaxial output—however that issue has been attributed to its CS8416 input receiver, which affects all DACs that use it.