Roon & Home Audio Fidelity (Room Correction / convolution filter creation)


(Martin Ingvarsen) #266

This is an interesting question. I don’t have the answer but would be interested to hear if anybody knows how important it is playing the mesurement sounds through Roon. I don’t think it matters if it’s just through Roon directly to an usb dac but what if it’s a networked endpoint like the digiOne or MicroRendu. Would the positive chacnges they make to the sound be big or significant enough to the convolution process or is it a detail that makes no difference in convolution?


#267

Totally agree :slight_smile:


#268

In general I don’t think many DAC’s has made a good USB input. And I don’t have one. I’m using a modified Singxer-SU1 getting its signal from UltraRendu which again gets it signal from a SonicTransporter.

The key to computer audio is not to use a computer :slight_smile:

Anyway I also got some files from the French guy on email. I will test those as well.

Still why do i need to do Pink Noise ?

And more important is those settings.


#269

You don’t need pink noise, only the sweep files. The pink noise files related to @magnus REW room correction guide, not the HAF filters.


#270

No, I’ve no idea whether it would make an audible difference either - presumably depends very much on individual factors. It certainly seems plausible that it might, which is why I went that route (for a networked endpoint).


#271

OK. These output settings does not matter as of cause the output is the files given from @alec_eiffel in his post.
I’ve got similar files from Thierry, but I’ve used Alec’s files. They are almost no difference. Probably same source. (There is a DR difference by 1)

image

I did my measurements with HQPlayer in the chain. Probably a mistake. I’m waiting for Thierry respond.

As @dhusky said, not much use of the PN. I used it just to set volume to 75.

To me it make sence to use the propper real chain. I guess that’s also why also Dirac is incorporated direct into to those pre’s that use Dirac.

Next month I will be able to compare HAF (I guess that mean HomeAudioFidelity) to Dirac in my Theta Casablanca 4a

Still not sure why REV stats 16 down left when file is 24 bit. MAybe REV can’t measure in 24 bit ? Probably not matter at all…

I must say my measurements look totally different than others. Hope I’ve got this right.


(Magnus) #272

The measurement looks fairly normal for a not-to-well treated room I would say. You will hear a big difference once you get the correction done (and will probably think it sounds a little thin and boring compared to now).

But I think you did the output part a little wrong. First of all, set output to the DAC if possible (or whatever audio endpoint you play through). It also looks like you play through both speakers, and play timing reference through left speaker. Typically you would want the opposite.


#273

Probably not. The output settings is related to if you use the PC as source. I added the files to my Roon library. As shown here:

This is also confirmed by Thirry, unless I misunderstood his email to me.

Here is his exact words:
“As the measurement process will not use the sweep signal generated by REW this setting can be left as it is as it will have no impact.”


(Magnus) #274

Ok, then it should work!


#275

Got my filters today.
Tierry says it’s OK to do the measurements through HQPlayer.

Sounds great of cause :slight_smile:

More listening tonight.


#276

Cool, enjoy the music :grinning:


#277

Were you doing upsampling in hqp? I couldnt get useable rew measurements for Thierry when going through hqp


#278

Yes to 24/192.

But I played through Roon, not REV.

I have no idea how this influence or not, or if I can use the given files without HQPlayer.

I actually plan to ask Thierry if the given files is ok without HQPlayer.

Edit:
If you DSD, REV seems not or be useful:

If HAF care about DSD or not, I don’t know.


(Oliver P) #279

What a difference…
I would like to share my impression with the HAF filters using an pictorial representation:

This is before:

…and this is after:

After measuring with the “sweep files” and a first round with two demo tracks I bought the premium version. Thierry provided me with the standard and x-talk filters. After a first round of critical listening I asked him for a “second round” since I had to correct my listening position.
He created his new beta x-talk filters (regarding my head’s geometry) and …wow…

In my case his filters don’t change the sound colours, they “just” get clarity, saturation and spatial resolution. Using my analogy with the pictures: Picture 1 is staled and fading, picture 2 generates more details, clarity and happiness (for me).

I also used Dirac live in my Arcam amplifier, but Dirac can’t win against HAF. The filters are treating the tracks in a different way.
Sometimes the changes in the listening experience are small, so I don’t hear them, but feel them. Sometimes the changes are stunning…

I clearly recommend to try out the demo version of Thierrys filters, he is very helpful also after the purchase.


(Mark Allen) #280

Great analogy Oliver! So very true.


(Larry Megugorac) #281

What I think potential HAF customers need to know is…after they take their REW sweeps…that Theirry can take one of their Reference tracks and modify the track with his filters so you know what the improvements will be before you pay for them.

I would also agree with Oliver’s picture analogy above…I also have a Mini DSP DDRC-88A that does 8 channels of RC but the hardware does another A>D and then back D>A …undoing the beauty my DAC just did…I also use Theirry’s X-Talk filters although have not demo’d the Face version…Still have a hard time believing that (that small of measurements) can make a difference…But really enjoying his work otherwise…!!!


(Larry Post) #282

Tierry made my filters, both normal and non-beta x-talk, about 6 weeks ago.

Since then I’ve enjoyed listening to all 3 ‘modes’, including without. At first, I stuck with the x-talk as it (they) struck me as the most positive change. I don’t feel ‘better’ fits as none are bad, only different.

3 weeks ago, I made a very significant DAC swap. I had a Oppo UDP-205 via Ethernet fed by 2012 Lenovo small desktop PC running ROCK, replaced by PS Audio DirectStream FPGA DAC, all balanced to their Stellar gain cell preamp to S300 amp > Ascend Acoustics Sierra 2 or Senn HD580 via preamp headphone amp.

I also had a DirectStream Jr. for about a week in there but liked it so much I forked over more for the Sr. The analog output stage is the most significant change. It runs much cooler too as the cast alum heatsink case is larger.

Now I find I like either no convolution filter or non-xtalk. Xtalk now sounds closed-in as if the speakers are way exaggerated toed-in placing the image too far in front of me; instead of enveloping me on all sides, top and bottom.

The room, speaker placement, has not changed. I did make a couple DIY power cables but the jury is still out if these are making any difference. Power was already in pretty good shape being regenerated, and only running low ripple SMPS throughout short of the linear transformers in the DAC/pre/amp and subwoofers amps.

I will have Thierry update my filters shortly and the exploration will begin again. I’ll gladly pay the modest price to have this much fun again, perhaps once a quarter. PS Audio (Ted Smith) updates the FPGA code a couple times a year too…and I have all the older (~8) versions to try.

I wholeheartedly recommend a home eval DirectStream DAC if you’re in a position to do so. Paul and company are very smart paying all costs for a 30 day in home demo of everything they make in Boulder, CO. I think non-USA has to go through a ‘local’ dealer. Lo siento.

Larry


(Mark Allen) #283

Interesting read. I’m also in the market for a new DAC. I can’t imagine listening w/out Thierry’s filters. I’ll likely purchase a new DAC listen for awhile w/out filters just to adjust my ears. Then I’d likely reach out to Thierry and purchase updated filters(I’ve done this before once I purchased a microphone stand)


(John) #284

Is there any additional benefit using the HAF Filters within HQPlayer, running under Roon.

Has anyone tried the HAF Filters just via Roon, and also via HQPlayer, but still managed within Roon?


(Larry Post) #286

I don’t see why it wouldn’t work. It really shouldn’t matter where the convolution filters run, either in Roon or HQP, might as well let the computer with the most horsepower do it or split it up and see if there is any sonic benefit realized by splitting the load of convolution and upsampling.

I let Roon convol as my DAC upsamples by design (DirectStream)