ATC speakers / Monitors

[Joe852]

As good as it needs to be is the only bar. How good it needs to be is of course subjective and different for different people and purposes. Once an aspect is as good as it needs to be then attention goes to other areas. ATC measures every tweeter and matches pairs. They also store the measurement so if you blow one they will be able to send you one that matches as closely as possible. Does anybody else do that? How important is it? Do you think SINAD 107 is better than SINAD 105? Or maybe it's not a meaningful difference, 105 is more than good enough and even bringing this up as a difference is evidence of poor analytical skills.

I don't spend a lot of time here and I'll probably get suspended/banned and that's fine. I don't like making insinuating comments. I'd rather just say the thing. The dislike of ATC on this site is clearly closer to a personality conflict than anything objective. They aren't trend chasers. They clearly don't change how they do things if they don't think it's an improvement, don't think it's worth the expense, etc. They focus on making the best drivers they can and that has worked out for them.

 

[Chrise36]

You know, I'm using SCM25As today and they do image very well and sound just about accurate to me.

These are certainly not the best speakers I've ever heard but I'm sure not going to kick them out of bed for eating crackers. There are so many worse speakers throughout studios than these.

View attachment 460292

What are the best you have heard and in what way are those better?

 

[dfuller]

What are the best you have heard and in what way are those better?

In this form factor, the SCM25A MK2s are better. The KH310s are no slouch either and are actually surprisingly similar. PSI A23Ms aren't bad either.


Better than these include:
- SCM50ASL Pro
- KH420
- D&D 8C

 

[Mart68]

As good as it needs to be is the only bar. How good it needs to be is of course subjective and different for different people and purposes. Once an aspect is as good as it needs to be then attention goes to other areas. ATC measures every tweeter and matches pairs. They also store the measurement so if you blow one they will be able to send you one that matches as closely as possible. Does anybody else do that? How important is it? Do you think SINAD 107 is better than SINAD 105? Or maybe it's not a meaningful difference, 105 is more than good enough and even bringing this up as a difference is evidence of poor analytical skills.

I don't spend a lot of time here and I'll probably get suspended/banned and that's fine. I don't like making insinuating comments. I'd rather just say the thing. The dislike of ATC on this site is clearly closer to a personality conflict than anything objective. They aren't trend chasers. They clearly don't change how they do things if they don't think it's an improvement, don't think it's worth the expense, etc. They focus on making the best drivers they can and that has worked out for them.

I agree with you about ATC but don't think the SINAD thing is relevant at all.

I think the problem people are having with ATC is that they don't have a ruler flat FR and that this is a big problem. But we saw on the very long NS10M thread that some people have got the idea that if you make a recording with monitors that are not perfect, then the recording will mimic those imperfections. But it doesn't work like that!

Nothing is made using just one set of monitors, look at the tens of thousands of great recordings made using 'monitors' that are far less accurate than ATC. You cross check and you compensate in any case.

As you say ATC are good enough as regards accuracy and there are many other non-sonic aspects of professional use which have to be considered when the studio is buying new monitors.

 

[Ilkless]

I see "state of the art" (SOTA) mentioned a lot lately here in the Speaker section. Can you name the most significant breakthroughs in speaker development in the last 20 years?

A few notes that might help:

• Don't bother talking much about measurements and Spinorama that ASR is focused on. Floyd Toole's time at the NRCC was from 1965 to 1991 and his time at Harmon was 1991-2007, so this stuff has been known for a LONG time.
• Don't talk about waveguides, as they have been around since 1958

The basic tenets have been known for so long that it only reflects poorly on manufacturers who fail to attain even this minimum level.

Also, it is rather disingenuous to dismiss waveguides. The basic concept of directivity matching and the role of waveguides in this was discovered way after 1958.

Also, there have been continuing and to my mind significant advances in waveguide geometry through computer modelling -- to wit, GGNTKT, Ascilab, Neumann and Genelec with rather non-standard geometries from the early 2000s onwards. Just because biradial and other rudimentary geometries existed long ago does not therefore mean the "SOTA" has not shifted.

 

[a4eaudio]

...Also, it is rather disingenuous to dismiss waveguides. The basic concept of directivity matching and the role of waveguides in this was discovered way after 1958.

I didn't dismiss waveguides, I dismissed the idea that waveguides are related to "state of the art". But the near-obsession with waveguides in some ASR conversations disregards the fact that they are only necessary to match directivity with certain driver choices. A 1" dome, 8" woofer two-way NEEDS a waveguide for good directivity matching. But you don't need a waveguide with a 3" dome midwoofer.

Also, there have been continuing and to my mind significant advances in waveguide geometry through computer modelling -- to wit, GGNTKT, Ascilab, Neumann and Genelec with rather non-standard geometries from the early 2000s onwards. Just because biradial and other rudimentary geometries existed long ago does not therefore mean the "SOTA" has not shifted.

AKABAK..."Legacy Akabak is getting on in years. The first version was presented at the AES 1995 in Paris. It had 2.1 as version-number and was available for a long time." So 40-years ago anyone who wanted to had access to computer modelling to design waveguides.

Again, I'm not criticizing waveguides, simply that there is not much SOTA about them. Just because someone designed a good/better waveguide isn't SOTA, its just good engineering. Ascilab, Neumann, Genelec have created good speakers. They may very well be better than ATC. But that is due to design choices, nothing that has drastically changed regarding speaker know-how in the last 20 (or 40?) years. They are all boxes using tweeters, woofers (and maybe a midrange) made of aluminum, paper, etc. and magnets. The real criticism of ATC seems to be that they simply don't follow the design choices that some people want them to - this criticism is valid to the extent that the work of Toole, Olive, etc. is based on science. But I would argue ATC has followed the exact right design choices if it has lead to the maximum profits for its owners/shareholders.

So, ATC speakers may suck, but that is not because they aren't state of the art, but rather they choose a poor design philosophy. Otherwise SOTA just becomes a synonym for the latest trend/fad.

 

[Palmer]

I agree with you about ATC but don't think the SINAD thing is relevant at all.

I think the problem people are having with ATC is that they don't have a ruler flat FR and that this is a big problem. But we saw on the very long NS10M thread that some people have got the idea that if you make a recording with monitors that are not perfect, then the recording will mimic those imperfections. But it doesn't work like that!

Nothing is made using just one set of monitors, look at the tens of thousands of great recordings made using 'monitors' that are far less accurate than ATC. You cross check and you compensate in any case.

As you say ATC are good enough as regards accuracy and there are many other non-sonic aspects of professional use which have to be considered when the studio is buying new monitors.

I would like to add, the NS10M thread fell way short of inputs from people who, not too long ago - and sometimes even today, made money using the NS10M in the studios that they are still in. To this day. I mentioned in the same thread that I never liked them, I am more of a fan of the original Auratones and still have a pair. Never heard the "consumer" ATC models, but have been in environments professionally where ATC's were used and, again, not my favourite, but you have to get paid at the ed of the day even if it's Tannoys, so be it.

 

[Ktacos]

ATC measures every tweeter and matches pairs.

So their QC is bad then? Other companies can put out exceptionally well matched drivers in the thousands for really cheap. If you need to do matched pairs, it means your QC is bad or there's an issue with manufacturing somewhere. Matched pairs really doesn't imply what you think it does.

They also store the measurement so if you blow one they will be able to send you one that matches as closely as possible. Does anybody else do that?

Most companies have a golden sample they match their production runs to. Needing to store data for each speaker just tells me QC is bad.

The dislike of ATC on this site is clearly closer to a personality conflict than anything objective.

No it's really mostly about the objective performance and the companies reluctance to share any data. A lot of the issues pointed out here are actually objective, such as the dome mid to tweeters spacing and it's high xover point and diffraction issues for instance. ATC could clear all this up by sharing some measurements but they won't which is just suspicious in a world where many manufacturers are more than happy to share them. Alarms should generally go off in ones head when you see speakers costing tens of thousands of dollars with no data to prove that you're getting something good. If that doesn't set off alarms I don't really know what to do tell you.

So far we've had only one user share data from their bookshelf speakers that showed what I'd consider great response and dispersion characteristics, still not quite on par with genelec or neumann but nothing I'd complain about. Some user shared data from their large 3 way models but the data was so poor with hilariously misleading scaling that you couldn't gather any meaningful info out of it. When people asked them to fix that problem they got mad and left the forum. How embarassing.

I would like to add, the NS10M thread fell way short of inputs from people who, not too long ago - and sometimes even today, make money using the NS10M in the studios that they are still in.

I would be careful to attribute making money in audio fields with having a trustworthy opinion on things. I've worked with some impressively bad engineers that were getting paid a lot more than me. One event I worked was hosting a rock band and an orchestra and I was to mix the rock band and setup the board for the the orchestra, their in house mixing guy would mix the orchestra. He applied zero EQ to any channel. It's still one of the worst sounding performances I've ever heard and it was filled with feedback.

 

[Joe852]

So their QC is bad then? Other companies can put out exceptionally well matched drivers in the thousands for really cheap. If you need to do matched pairs, it means your QC is bad or there's an issue with manufacturing somewhere. Matched pairs really doesn't imply what you think it does.



Most companies have a golden sample they match their production runs to. Needing to store data for each speaker just tells me QC is bad.



No it's really mostly about the objective performance and the companies reluctance to share any data. A lot of the issues pointed out here are actually objective, such as the dome mid to tweeters spacing and it's high xover point and diffraction issues for instance. ATC could clear all this up by sharing some measurements but they won't which is just suspicious in a world where many manufacturers are more than happy to share them. Alarms should generally go off in ones head when you see speakers costing tens of thousands of dollars with no data to prove that you're getting something good. If that doesn't set off alarms I don't really know what to do tell you.

So far we've had only one user share data from their bookshelf speakers that showed what I'd consider great response and dispersion characteristics, still not quite on par with genelec or neumann but nothing I'd complain about. Some user shared data from their large 3 way models but the data was so poor with hilariously misleading scaling that you couldn't gather any meaningful info out of it. When people asked them to fix that problem they got mad and left the forum. How embarassing.




I would be careful to attribute making money in audio fields with having a trustworthy opinion on things. I've worked with some impressively bad engineers that were getting paid a lot more than me. One event I worked was hosting a rock band and an orchestra and I was to mix the rock band and setup the board for the the orchestra, their in house mixing guy would mix the orchestra. He applied zero EQ to any channel. It's still one of the worst sounding performances I've ever heard and it was filled with feedback.

"Other companies can put out exceptionally well matched drivers in the thousands for really cheap" and you know this for a fact right?

They don't need to share data, anyone is free to do their own measurements just as they're free to use their own ears. Do you believe specs that come from manufacturers? Really?....Really? I wonder if they don't publish a lot of specs so they can avoid customers with room temp IQs.

 

[Mart68]

I would be careful to attribute making money in audio fields with having a trustworthy opinion on things. I've worked with some impressively bad engineers that were getting paid a lot more than me. One event I worked was hosting a rock band and an orchestra and I was to mix the rock band and setup the board for the the orchestra, their in house mixing guy would mix the orchestra. He applied zero EQ to any channel. It's still one of the worst sounding performances I've ever heard and it was filled with feedback.

Running a front of house isn't the same as making recordings, not by a margin. I encountered more than one person running a desk who didn't really have a clue. I don't know if it's still the same now since I'm going back decades here.

 

[Palmer]

...
I would be careful to attribute making money in audio fields with having a trustworthy opinion on things. I've worked with some impressively bad engineers that were getting paid a lot more than me. One event I worked was hosting a rock band and an orchestra and I was to mix the rock band and setup the board for the the orchestra, their in house mixing guy would mix the orchestra. He applied zero EQ to any channel. It's still one of the worst sounding performances I've ever heard and it was filled with feedback.

While I certainly feel bad that you had that experience, I must say that I have more than 25 years behind me, about quarter of it in the recording studio, and the rest later in the Broadcast world. I can tell you that while I've gotten burned by people who, yes, made more money than me - your reputation lives and dies with trust, and yes, opinion.
Of course I am glad that I am pretty much out at this point, since there are way more fools out there nowadays, that get away with things of which in my days you would have been looking at another career.
Cheers.

 

[Ktacos]

They don't need to share data, anyone is free to do their own measurements just as they're free to use their own ears. Do you believe specs that come from manufacturers? Really?....Really?

Yeah, everyone's free to measure their 3 ways as long as they have $5000 for a single SCM25 lol. Curiously neumann was happy to send Amir a KH420, so what's ATC's hold up?

I do believe driver specs, within tolerances, also basing my words on my own measurements I've taken. I have some cheapo Dayton td20 tweeters that measure pretty much identically. If they can do that for $10 each, anyone should be able to do it.

Threads done and there's nothing new to add until someone gets a 3 way atc on a klippel.

 

[Ktacos]

Running a front of house isn't the same as making recordings, not by a margin.

Of course not, who said it was? I've done both anyway.

 

[Ilkless]

But you don't need a waveguide with a 3" dome midwoofer.

The issue is said 3" has a footprint closer to a 6" driver, compromising the directivity blend.

Also, the main foundational shifts include: DSP (incl. phase linearisation), computer-aided systems design with directivity matching as a goal which a 3" driver of the kind ATC does active impedes, low diffraction cabinets, not to mention driver advances like Purifi's neutral surround

 

[dfuller]

The issue is said 3" has a footprint closer to a 6" driver, compromising the directivity blend.

Also, the main foundational shifts include: DSP (incl. phase linearisation), computer-aided systems design with directivity matching as a goal which a 3" driver of the kind ATC does active impedes, low diffraction cabinets, not to mention driver advances like Purifi's neutral surround

I think if the Fc was dropped to 2.8-3k from 3.5k the directivity problems basically wouldn't be there. the dome (in its integral waveguide) radiates fairly widely to about 3-4k from every bare driver measurement I've seen. They also cross it maybe a bit closer to its only major breakup mode than I'd like. I'm not well enough versed in VituixCAD to sim this out, but I'm sure someone is.

As far as low diffraction baffles, there's not much you can do if you still have a large woofer in there without quite deep WGs (which will substantially alter the sound, and maybe not in a way that is desirable). The pro models at least include a fairly substantial roundover - half an inch or more, IIRC. It's not a replacement for a narrower cabinet but again, good luck doing that with a large woofer.

 

[Blockader]

As good as it needs to be is the only bar. How good it needs to be is of course subjective and different for different people and purposes. Once an aspect is as good as it needs to be then attention goes to other areas. ATC measures every tweeter and matches pairs. They also store the measurement so if you blow one they will be able to send you one that matches as closely as possible. Does anybody else do that? How important is it? Do you think SINAD 107 is better than SINAD 105? Or maybe it's not a meaningful difference, 105 is more than good enough and even bringing this up as a difference is evidence of poor analytical skills.

I don't spend a lot of time here and I'll probably get suspended/banned and that's fine. I don't like making insinuating comments. I'd rather just say the thing. The dislike of ATC on this site is clearly closer to a personality conflict than anything objective. They aren't trend chasers. They clearly don't change how they do things if they don't think it's an improvement, don't think it's worth the expense, etc. They focus on making the best drivers they can and that has worked out for them.

Both Neumann and Genelec factory match their monitors. Neumann’s KH 120 II for instance is calibrated to within ±0.5 dB between channels and that precise match is preserved whenever the speaker goes in for service.

When the left and right speakers are level matched to within about 0.5 dB, the phantom centre locks in, panning cues stay accurate and the stereo image remains stable across the listening area. With no level imbalance for your brain to correct, cognitive load drops and listening fatigue is kept to a minimum. DSP monitors with built in room correction software can fine tune each speaker so the pair is level matched even more tightly.

What ATC was doing may be spectacular in 80s but that's not the case anymore. We keep seeing people hype up what ATC does, even though every evidence driven speaker brand has left those tricks behind over the last decade. And they do it for incredibly cheaper.

 

[Joe852]

Both Neumann and Genelec factory match their monitors. Neumann’s KH 120 II for instance is calibrated to within ±0.5 dB between channels and that precise match is preserved whenever the speaker goes in for service.

When the left and right speakers are level matched to within about 0.5 dB, the phantom centre locks in, panning cues stay accurate and the stereo image remains stable across the listening area. With no level imbalance for your brain to correct, cognitive load drops and listening fatigue is kept to a minimum. DSP monitors with built in room correction software can fine tune each speaker so the pair is level matched even more tightly.

What ATC was doing may be spectacular in 80s but that's not the case anymore. We keep seeing people hype up what ATC does, even though every evidence driven speaker brand has left those tricks behind over the last decade. And they do it for incredibly cheaper.

"DSP monitors with built in room correction software can fine tune each speaker so the pair is level matched even more tightly." this is closing in on something. Since DSP can easily correct any imperfect frequency response it almost isn't a factor. Once it's within an EQable range why even bring it up?

 

[Blockader]

"DSP monitors with built in room correction software can fine tune each speaker so the pair is level matched even more tightly." this is closing in on something. Since DSP can easily correct any imperfect frequency response it almost isn't a factor. Once it's within an EQable range why even bring it up?

I write on the assumption that readers are acting in good faith and won’t cherry‑pick or twist my words just to win an argument.

There is no harm in EQing speakers below a certain frequency interval and that's below the room transition frequency. That's the area where the standing waves in a room still dominate the response. Speakers with built-in room correction can channel match speakers below the room transition frequency and even tweak them within the transition band to further upgrade the channel matching. So built-in DSP can be utilized to fix channel matching problems caused by the room.

 

[dfuller]

Both Neumann and Genelec factory match their monitors. Neumann’s KH 120 II for instance is calibrated to within ±0.5 dB between channels and that precise match is preserved whenever the speaker goes in for service.

When the left and right speakers are level matched to within about 0.5 dB, the phantom centre locks in, panning cues stay accurate and the stereo image remains stable across the listening area. With no level imbalance for your brain to correct, cognitive load drops and listening fatigue is kept to a minimum. DSP monitors with built in room correction software can fine tune each speaker so the pair is level matched even more tightly.

What ATC was doing may be spectacular in 80s but that's not the case anymore. We keep seeing people hype up what ATC does, even though every evidence driven speaker brand has left those tricks behind over the last decade. And they do it for incredibly cheaper.

FWIW, ATC to my understanding do at the very least pair match; what little measurement there is shows that they do match +/-0.5dB. I don't know if every single speaker is within that close of a tolerance, but we could ask. The actives have trimpots on the crossovers and are certainly matched closer than the passives can be.

It would be nice, of course, if someone from ATC would actually chime in for a change.

 

[Avp1]

I didn't dismiss waveguides, I dismissed the idea that waveguides are related to "state of the art". But the near-obsession with waveguides in some ASR conversations disregards the fact that they are only necessary to match directivity with certain driver choices. A 1" dome, 8" woofer two-way NEEDS a waveguide for good directivity matching. But you don't need a waveguide with a 3" dome midwoofer.


AKABAK..."Legacy Akabak is getting on in years. The first version was presented at the AES 1995 in Paris. It had 2.1 as version-number and was available for a long time." So 40-years ago anyone who wanted to had access to computer modelling to design waveguides.

Again, I'm not criticizing waveguides, simply that there is not much SOTA about them. Just because someone designed a good/better waveguide isn't SOTA, its just good engineering. Ascilab, Neumann, Genelec have created good speakers. They may very well be better than ATC. But that is due to design choices, nothing that has drastically changed regarding speaker know-how in the last 20 (or 40?) years. They are all boxes using tweeters, woofers (and maybe a midrange) made of aluminum, paper, etc. and magnets. The real criticism of ATC seems to be that they simply don't follow the design choices that some people want them to - this criticism is valid to the extent that the work of Toole, Olive, etc. is based on science. But I would argue ATC has followed the exact right design choices if it has lead to the maximum profits for its owners/shareholders.

So, ATC speakers may suck, but that is not because they aren't state of the art, but rather they choose a poor design philosophy. Otherwise SOTA just becomes a synonym for the latest trend/fad.

SOTA is not end of the world. One of companies which was SOTA in mechanical speaker design - MBL, recently went under. Their driver technology was unique, and it's execution top notch, still not enough to sustain business. I did hear their speakers on multiple occasions and was impressed with how good they played in untreated rooms. Other speaker makers may not be at cutting edge technologically, but still exist after half a century. They key here is to know customer needs and cater them accordingly, while being technically solid. I do like other "expensive" and very good sounding speakers, like Estelon. Not much of talks here about that brand. Somehow for many SOTA means software correction of physical problems. But this is not the only way to produce great speakers.