Ascend Acoustics new Klippel optimized speaker

[617]

For sure. Very cool. Especially to see one as "small scale" as Ascend. There are definitely other companies who own the NFS to design their speakers. A few of them make this public knowledge. Others do not. I've talked with a few of them and know of (2) that have been purchased after talking to me about how the NFS works. In fact, I'm having a Zoom meeting next Monday with another (smaller scale) manufacturer who wants to ask me some questions about my experience with the NFS because they're leaning hard toward buying one and just want a bit of extra info.

And makes it that much crazier that people are buying the NFS simply to review speakers...

What's crazy is that the big audio magazines don't use the NFS. You and Amir have this ridiculous CNC acoustic platform in your garage. Meanwhile Stereophile relies on John Atkinson, some burning sage, a twenty year old copy of CLIO and some excel spreadsheets.

 

[Ericglo]

No domestic cabs for these.
I don't think any part of these is made in the USA at all. (Which is understandable these days though nobody should be buying these thinking they are made here in the US)
If domestic cabs become available for these that would be cool, they look better to me with the magnetic grills. Upcharge on the Sierra is about $200 for them.

Is it just these speakers or all of Ascend speakers? IIRC on AVS a couple of years ago, Dave said they make the cabinets in Southern Cal.

 

[TurtlePaul]

I agree about not making a big deal out of it. I just think it is neat that Ascend was willing to spend the money to invest in a NFS. They could have just as easily sent it out to have it measured, which is what most companies should do if they don't want to purchase a NFS.

But it goes back to the point that having a NFS can drastically impact a speaker designers workflow. Dave said he ran the NFS at least once a day.

He can do a run, then see that the crossover frequency maybe could be adjusted 200 hz, or there is a driver breakup/resonance to be trapped, or it needs a steeper filter to adjust the off-axis. After a days work simulating crossover circuit tweaks the new version is back in the NFS. You can't do that if you send it away.

Similarly, Ascend has a good relationship with Seas. I would imagine that Dave has probably tried 5-6 Seas woofer and 5-6 tweeters in different combos until he found the right versions that made it into the final project. If you are sending the speakers out for measurements, you need to make a lot of these driver choices before having the data in hand. No way you can send two dozen different combos.

One could easily argue that a speaker wasn't designed with an NFS if the speaker designer didn't have direct access to an NFS. If a speaker was only on the NFS once, I personally wouldn't consider it to be fully "designed with an NFS."

 

[ROOSKIE]

Is it just these speakers or all of Ascend speakers? IIRC on AVS a couple of years ago, Dave said they make the cabinets in Southern Cal.

I am not sure as I was only considering this model.
Deff imported cabs for this set only and local cabs available for the Sierra at an understandable upcharge. Maybe local cabs as an option on the LX in the furure? IDK?
Maybe ask Dave or check their website and forum for info on other models.

 

[hardisj]

But it goes back to the point that having a NFS can drastically impact a speaker designers workflow. Dave said he ran the NFS at least once a day.

He can do a run, then see that the crossover frequency maybe could be adjusted 200 hz, or there is a driver breakup/resonance to be trapped, or it needs a steeper filter to adjust the off-axis. After a days work simulating crossover circuit tweaks the new version is back in the NFS. You can't do that if you send it away.

Similarly, Ascend has a good relationship with Seas. I would imagine that Dave has probably tried 5-6 Seas woofer and 5-6 tweeters in different combos until he found the right versions that made it into the final project. If you are sending the speakers out for measurements, you need to make a lot of these driver choices before having the data in hand. No way you can send two dozen different combos.

Good points. If you are in the business of making speakers, it seems to me the NFS is a solid investment. I certainly understand the apprehension and cost-of-entry, though.

Very cool that Ascend did this, too. And we can see not just the benefit in their product but - and this is more personal to me - that there is another proponent of data. To expand on the last part: While I wasn't involved in and had absolutely nothing to do with it, the initial Ascend product tested here showed some resonance issues in the midrange. Typical issue from most manufacturers who perform gated measurements because their gates simply aren't long enough to permit the resolution necessary to show mid-to-high Q resonances in these frequencies. IIRC, there was a lot of pushback from Ascend and thus the Ascend faithful on the measurements. I remember some grumblings over on AVSforum. I didn't perform those measurements, but as someone who does measure and review products, that negative feedback impacts me as well. So, when I saw that Dave had decided to do some further investigation and ultimately found that the NFS measurements were, indeed, correct, ... well that is one less "bump in the road" toward people trusting and understanding the usefulness of proper measurements.

 

[Shazb0t]

Hey Erin,

Yes and no.

The complex data export module for the NFS functions a bit differently than the basic text export. Pierre requested phase data, which can only be exported using this module (a near $10k add on) Pierre provided me with few details, I did not know he was going to be basing sensitivity on the export files I provided him.

This NFS export data module defaults to a 2m radius and 1v output. It is a different set of options compared to the far field visualization settings. I don’t recall what the reference voltage was even set for, as I didn’t think it would matter since I thought he was just using the raw data for generating spins. I was not aware he was going to be providing various specifications. My brief look at his site showed calculated spin graphs, I should have dug a bit deeper (entirely my fault)

That stated, I have provided him with new export data, properly referenced to 2.83v and a 1 meter radius (mic distance) so he can *properly* calculate sensitivity.

Obviously, the sensitivity spec he listed is off by quite a bit I am sure he will correct this promptly.

Hey Dave! Nice that you've joined in on the conversation over here at ASR. I took a look at @pierre Spinorama collection for Ascend products and it's pretty neat! Would you be able to also share with him the Klippel data for your Sierra RAAL Towers that you've posted graphs from on your forum? As an owner, I think it would be helpful to compare against some of the other offerings, particularly the LX, using his tool, thanks.

 

[pierre]

Dave, FWIW, when I send @pierre our data, I do so by just providing the SPL Horizontal and SPL Vertical data provided from the CEA-2034 module. Right click, save as text. I do this because that's what Amir posts for them and the guys wanted the data in the same format.

Not that it matters now. Just letting you know for future reference.

Hi Erin,

Dave send me the data with complex values which is great. I can generate the spin in the same way but I can also build better EQ.
As you know, when the group delay is not flat, IIR filters are not working properly. What we will see I believe is that EQ will be less efficient
at flattening the curves esp around the crossver(s) where they would be the most useful.

P.

 

[MZKM]

You just change the reference position and run it. I've mentioned this in other reviews on this forum when the reference was at the wrong spot (i.e., at the tweeter rather than between the waveguide/midwoofer, etc).

I know that you can change the reference point, but it looks like sorta built-in feature by how it’s all labeled by the degrees and in the same native Klippel chart.

 

[hardisj]

I know that you can change the reference point, but it looks like sorta built-in feature by how it’s all labeled by the degrees and in the same native Klippel chart.

I don't know if it's a copy/paste from a separate run but there is a way you can set up the module to run at different reference points. It's not covered much anywhere but you can see pointers to it in their "multi-source" talk. Section 10.5 in this link.

The interesting thing that I would like to explore is their simulated EQ/filter and time delay stuff. I need to try to remember to check this out.

 

[hardisj]

Hi Erin,

Dave send me the data with complex values which is great. I can generate the spin in the same way but I can also build better EQ.
As you know, when the group delay is not flat, IIR filters are not working properly. What we will see I believe is that EQ will be less efficient
at flattening the curves esp around the crossver(s) where they would be the most useful.

P.

I wasn't aware you wanted anything more than what I was already given. If you need, I can give you the complex data export (going forward). Just let me know.

 

[Byrdsmaniac]

Damn, that looks excellent. Probably sounds pretty good.

I have a pair of the Sierra 2ex. They are superb. I was skeptical that the ribbon tweeter could have such wide even dispersion, but it does and reproduction of delicate subtleties in the highs is very appealing. Dialogue intelligibility on movies is so good that I am able to use lower volume without missing anything that is said. Dynamic contrasts are often startling in their realism, as is timbral accuracy. When listening to my old Byrds or Beatles records there is never any doubt as to which of them is singing in the background.

 

[TurtlePaul]

@MZKM Is yhere enough data for your pref score and directionality plots?

 

[617]

Good points. If you are in the business of making speakers, it seems to me the NFS is a solid investment. I certainly understand the apprehension and cost-of-entry, though.

Very cool that Ascend did this, too. And we can see not just the benefit in their product but - and this is more personal to me - that there is another proponent of data. To expand on the last part: While I wasn't involved in and had absolutely nothing to do with it, the initial Ascend product tested here showed some resonance issues in the midrange. Typical issue from most manufacturers who perform gated measurements because their gates simply aren't long enough to permit the resolution necessary to show mid-to-high Q resonances in these frequencies. IIRC, there was a lot of pushback from Ascend and thus the Ascend faithful on the measurements. I remember some grumblings over on AVSforum. I didn't perform those measurements, but as someone who does measure and review products, that negative feedback impacts me as well. So, when I saw that Dave had decided to do some further investigation and ultimately found that the NFS measurements were, indeed, correct, ... well that is one less "bump in the road" toward people trusting and understanding the usefulness of proper measurements.

The resolution in the bass and midbass has been the #1 revelation for me with these NFS measurements. I assumed high q resonances didn't exist in these regions, which in retrospect was incredibly dumb. Years of T/S simulation and gated measurements trained me to believe that nothing crazy could happen below 300hz or so, but I was wrong.

I really like the idea of a company like Ascend using the NFS. Ascend has a very traditional hifi approach, favoring wide dispersion and high quality drivers. They are making the speakers I would be making if I had an NFS. Well, I'd be making 3 ways but close enough.

 

[Head_Unit]

the room will contribute to output. Do you include that in your sensitivity specification? I'd say ideally not.

Ha ha Klipsch and some others appear to disagree with you (I on the other hand vehemently agree with you)

 

[MZKM]

@MZKM Is yhere enough data for your pref score and directionality plots?

Pierre already has done it on his site:

A collection of loudspeakers measurements

pierreaubert.github.io

 

[abdo123]

Good points. If you are in the business of making speakers, it seems to me the NFS is a solid investment. I certainly understand the apprehension and cost-of-entry, though.

Very cool that Ascend did this, too. And we can see not just the benefit in their product but - and this is more personal to me - that there is another proponent of data. To expand on the last part: While I wasn't involved in and had absolutely nothing to do with it, the initial Ascend product tested here showed some resonance issues in the midrange. Typical issue from most manufacturers who perform gated measurements because their gates simply aren't long enough to permit the resolution necessary to show mid-to-high Q resonances in these frequencies. IIRC, there was a lot of pushback from Ascend and thus the Ascend faithful on the measurements. I remember some grumblings over on AVSforum. I didn't perform those measurements, but as someone who does measure and review products, that negative feedback impacts me as well. So, when I saw that Dave had decided to do some further investigation and ultimately found that the NFS measurements were, indeed, correct, ... well that is one less "bump in the road" toward people trusting and understanding the usefulness of proper measurements.

What i really don’t understand is why Ground plane measurements are not much more common for this exact reason.

Sure 1/24 oct resolution (NFS) won’t be achieved but something close to 1/12 should br possible.

 

[TurtlePaul]

What i really don’t understand is why Ground plane measurements are not much more common for this exact reason.

Sure 1/24 oct resolution (NFS) won’t be achieved but something close to 1/12 should br possible.

It is a lot of work and probably not the fun part of making speakers. I also commend those who do speaker on a tall pole measurements.

 

[Ciobi69]

Hey guys do you think that it's worth to sell my kef R3 for these? I think they might be more adapt for.me for the power handling, as i would like to use them for movies too,i think that they can handle more than my R3, but i don't know about the directivity factor, i like that the kef are more direct but i don't know hou much could change between them, thanks

 

[abdo123]

Hey guys do you think that it's worth to sell my kef R3 for these? I think they might be more adapt for.me for the power handling, as i would like to use them for movies too,i think that they can handle more than my R3, but i don't know about the directivity factor, i like that the kef are more direct but i don't know hou much could change between them, thanks

They're almost identical.

 

[Ciobi69]

They're almost identical.

View attachment 196825

Damn what a great low extension,i am lucky that i use a subwoofer,but even without it even the kef r3 has great bass,i can't imagine those