Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Focus Fidelity Filter Designer
#11
It is perhaps worth mentioning that the idea behind the Focus Fidelity software is to attenuate resonant modes and improve time domain, looking at impulse responses, step response, phase etc in the room. So whilst a given speaker may have issues with not being time aligned or with phase, ultimately the software is correcting for a lot more than that once the influence of the room is taken into account.

As for the sounding "dead" issue, I shall reserve judgement on that one for now, I simply have not had enough listening time to be able to be sure.

One good thing running the convolutions in Roon, it is very easy to switch between convolutions, parametric EQ and uncorrected whist listening. I have some a basic PEQ set up in Roon that I think sounds good and is essentially just correcting the worst of the bass resonance modes. It will be interesting to compare this to running uncorrected and the Focus Fidelity convolutions.

I'll try this when I get some free time and report further.

As a final thought. There are a lot of parameters that you can play with in focus fidelity. The target curve is pretty much fully adjustable, you can limit the amount of correction, both positive and negative, plus many other things. Quite how all of this will work out in my system remains to be seen, but there is quite a lot that can be customised to suit a given system, and personal preference.

For anyone interested in investigating further, the user manual is available on the Focus Fidelity website, well worth a quick look.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
#12
I have been running the Focus Fidelity convolutions for over a month now.

The short version of this post is that I am absolutely delighted by the results. I am hearing a significant increase in clarity, the improvement in sound staging is stunning, plus in very many ways everything is just plain nicer to listen to. Listening fatigue is pretty much banished, when I start listening now it is getting very difficult to stop. I feel more like I have completely upgraded my system, improved amp, better speakers and so on, rather than adding a software tweak.

So what are the negatives? There were mentions earlier in this thread of room correction DSP giving a "dead" sound, in the past I have seen many similar comments about RC "sucking the life out of music", reducing dynamics, reducing detail, killing the bass, and so on. Indeed, in the past I have made comments like this myself. I have dabbled with a number of RC solutions but had always given up on then, preferring the more purist approach of playing direct and being DSP free.

With Focus Fidelity I not experiencing any of these negatives. With the bass cleaner than ever, dynamics are subjectively improved. I'm hearing details in the music I have never heard before, again resulting from the ultimate clarity I think. Bass itself is superb, and I am finding more life than ever in music.

It is perhaps worth mentioning bass a little more. With my previous RC experiments, I always found that I lost something with bass. Maybe bass was clearer, some room resonances eliminated, and so on. So some positives but it was always a trade off. Yes, there were improvements, but the end result tended to end up sounding a bit thin, somehow less satisfying and enjoyable to listen to. This is not the case with the Focus Fidelity convolutions. I stuck with my plan to get a convolution that worked OK, then spend a few weeks just listening. This I did for about a month, after which I made a set of six new convolutions, fine tuning the roll off, level of bass, this kind of thing. It is very early days with me trying these slightly revised convolutions, but I have a couple of convolution options now where there is simply too much bass. Of the six, I can see that there is one I prefer over the others. If this view holds out over the next month or so of listening, I guess I could fine tune some new convolutions, similar to my preferred choice. That said, I am more than happy listening to what I have now.

In terms of the value proposition here, last week I tried a little experiment. I tried running my system with the convolutions, but in Roon ready mode direct to the Devialet, so taking the Mutec and SOtM kit out of the chain. I then compared this to listening with my preferred method of HQPLayer / SOtM / Mutec, but switching the convolutions off. Of these two configurations, Roon Ready direct to Devialet with DSP was by far the best to listen to. So in a way, £185 worth of software and a little effort beats £7k's worth of hardware in terms of what you can add in front of a Devialet amp. Although it is not quite that simple, the sound quality improvements offered by the SOtM / Mutec kit are different to that offered by the DSP, and ultimately the best results are obtained by using both together. But if it came to it, and I had to choose one or the other, I would take the Focus Fidelity software and lose the SOtM / Mutec kit.

As a final point - customer service. I had posted on AS about my plan to stick with the initial convolutions I made, then see how things sounded for a few weeks of long term listening, then try fine tuning the convolutions in whatever direction I thought appropriate. A couple of weeks ago I received an email from Focus Fidelity, advising that if I was about to try making new convolutions then it might be an idea to wait a couple of weeks, the reason being that they were maybe one or two weeks away from issuing an updated version of the software. I received that email on June 20th, I was sent a link to the new software June 26th. A nice touch I thought.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
#13
Hi Confused,
I could not resist after your first post here and bought a license as well.
The results were really good but I had some issues: I could not enter numbers with any decimal places in the target designer and I also had an issue that in some bass region is had removed some bass where I did not want it to be removed. I told the support but never got any response...
But with the new 1.5 version this bass problem seems to be solved.
With the old version 1.2x I liked my Sweet Room configuration more, with the new version this could have been changed.
I will do some more listening this evening and report back. But what I can say already is that this is the best RC I tried so far.
Reply
#14
Do you setup the Devialet with the Sweet Room abs then in addition with Focus Fidelity. Or just Focus Fidelity? Is the measurement module which was mentioned to be available in the meantime ready?
Reply
#15
@markush - The Focus Fidelity measurement module is not available yet. Back in March they advised that it was "several months away", which I would interpret as being maybe some time towards the end of 2021.

I think your Sweet Room question was aimed at @Vivialet , but what I can say is that Sweet Room and the Focus Fidelity convolutions are very different things. Sweet Room allows basic parametric equalisation, where as Focus Fidelity is using 65536 tap length FIR filters, and hence can perform "time domain" phase correction etc. So the Focus Fidelity filters need to be used in the playback software such as Roon, JRiver, HQplayer, plus some others. I guess that it would be possible to run the convolutions in the playback software and have some options to "fine tune" (maybe album specific) in Sweet Room, but this is perhaps an overly complex approach. One thing is for sure, you cannot use the Focus Fidelity convolutions in Sweet Room. (which is a shame, it would be great if you could do this, then have the convolutions available for all inputs, including vinyl etc., but I do not think the Devialet has the required processing power even if Devialet wanted to allow this as a feature?)

@Vivialet - It is strange that you did not get a response from the support email address. Checking back, I did send a couple of basic questions myself, and in each case I received a response the next day.

Anyway - I am pleased to see at least one other person is trying this with a Devialet. The results from room correction will be highly defendant on the specific room and system, and to a degree the listeners ears, so it is great to have another reference point here. Now that your earlier issues seem to be resolved, it will be interesting to hear how you get on with your new "1.5" convolutions.

I had no issues with bass with the earlier version, but I managed to get some listening in over the weekend with my new "1.5" convolutions, and yes, I would say that the bass I am getting is better than ever now, I am very pleased.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
#16
Hi Markus,
of course I'm user either one or the other. FocusFidelity has been improved with 1.5, but still it is not as universal as my sweet room correction. With some recordings it is better, with others it is worse. My simple sweet room filters are sounding good with each kind of music.
I'm still trying to find a better target curve for focus fidelity. One problem with all automated corrections is that they also increase the gain while my sweet room configuration only lowers the gain for some peaks in the bass region. I have to increase the volume by around 7db with ff to get the same volume as with sr.
A good option would be if I could limit the gain to less than the currently possible 4db limit. Even with 4db I get some unprecise bass response with some recordings.
Reply
#17
Yes maybe posting again in the Sweet Room thread at the website would be a good idea as Mathieau recently posted there. Status of bug fixes or new features is long overdue.
I read at Focus Fidelity it does “ reversing loudspeaker phase and impulse distortions” - doesn’t SAM do something similar?
When measuring for room correction - so you have SAM on and at what percentage? Guess this also has a reasonable impact on the overall outcome.
Reply
#18
My advice would be to keep Sweet Room EQ (and SAM if there is a profile for your speakers) enabled and do measurements for FF on top. FF will perceive this as "how the speakers work in the room" and do corrections accordingly. I always had SAM enabled for my speakers when I did measurements for Audiolense. Audiolense alone could never match the combination of SAM+Audiolense.
*
Devialetless!
Roon, ROCK/Audiolense XO/Music on NAS/EtherRegen/RoPieee/USPCB/ISORegen/USPCB/Sound Devices USBPre2/Tannoy GOLD 8
250 Pro CI, MicroRendu(1.4), Mutec MC-3+USB
Reply
#19
(06-Jul-2021, 00:07)markush Wrote: doesn’t SAM do something similar?

It does, but SAM is a pure electrical correction made from laser feedback vs. a correction made from acoustical measurements in a room.
*
Devialetless!
Roon, ROCK/Audiolense XO/Music on NAS/EtherRegen/RoPieee/USPCB/ISORegen/USPCB/Sound Devices USBPre2/Tannoy GOLD 8
250 Pro CI, MicroRendu(1.4), Mutec MC-3+USB
Reply
#20
(06-Jul-2021, 09:19)ogs Wrote: My advice would be to keep Sweet Room EQ (and SAM if there is a profile for your speakers) enabled and do measurements for FF on top. FF will perceive this as "how the speakers work in the room" and do corrections accordingly. I always had SAM enabled for my speakers when I did measurements for Audiolense. Audiolense alone could never match the combination of SAM+Audiolense.

This is an interesting point.  I debated this one with myself before producing the measurements that I used in Focus Fidelity.  In the end I opted to take the measurements with SAM enabled, but set to 0%.  The idea being that I could then use SAM and FF together, and I could use SAM% to dial in a little more bass if I felt that a particular track or album needed it.  I wonder if SAM 50% might be a better option, then allowing adjustment up or down.  To be honest, I think both methods would work, albeit slightly differently.  I am happy with the approach I have taken so far, so maybe something to try later if I have some free time.

@Vivialet - Your comment regarding the 4dB limit is interesting, and indeed makes me wonder why there is this restriction in the settings.  Whilst this has not been an issue for myself, I can imagine that this might be problematic with some systems and rooms.  I think the end result would be dependent on the speakers, at what frequency the key nulls manifest themselves, and so on.  So I was wondering, have you asked Focus Fidelity why there is this 4dB limit imposed in the settings?  For me, I would have thought that if 4dB can be set, then why not 2dB, or even zero?  Maybe a "feature request" for Focus Fidelity to consider in the future?  Unless there is a specific technical reason within the software why the limit cannot go below 4dB?  Which is of course a question for them to answer.

I am thinking that there might be some workarounds in the Target settings screen, you could manually adjust the curve to follow the problematic bass null, and hence mitigate the gain issue.  But this is obviously far from ideal, hence why I would be interested in the comments of Focus Fidelity in this regard.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)