43 thoughts on “Audient Launches EVO 16, A Compact 24-in, 24-out Audio Interface”

1. I almost got really excited. This device has 8 analog inputs and 8 analog outputs. Everything else is marketing fluff and ADAT jacks.

   1. Its a “standard” with any audio interface like “s/n in RMS” and “dynamic range in dBA”.
      Some would even call it a “48 channels audio interface” 🙂
      Digital i/o are also channels processed by the audio interface and you can add a/d/a converters so there is logic to it, just have some patient to read the specs before you get excited, I know I do.

   2. I agree! Can we start calling them by physical inputs. Okay, yeah it has digital inputs but lets say we connect two of these bad boys digitally – should we call the combined system 48-in 48-out? That’s too confusing. If we only used the physical inputs it would make buying much easier – go on any equipment vendor and try to filter for only interfaces that are 16 or more physical inputs – it’s not possible on any of the major vendors. Most of the time you have to click on every one and look at the photos or the tech specs to find the physical count. It’s false advertising.

      1. Digital are physical inputs too, I use them all the time and its crucial for me to know how many channels possible in total.
         I also prefer not having all analog channels in one rack since my instruments are not close together.
         Maybe it’s better people will educated themself about all types of in and outs and read the technical specifications, it takes 10 (maybe 15) seconds to understand the types of channels you get with a new product somebody work on for at least a year. Oh no! 🙂

   3. No, technically it’s 10 analog inputs, 12 analog outputs.

      1. Thanks, I realized that after I’d written my initial comment. My point still holds. It’s not a 24 in/24 out interface without a couple of ADAT expanders attached.

   4. ADAT is awesome.
      Long live ADAT!!!

2. There is no mention on their website, nor on the Audient Wikipedia page (though they mention workers in Bulgaria) as to where their products are manufactured.
   However, their ID44, Evo 8 and ASP400 all clearly say on the back: Made in China.
   Their mixers, ASP 4816 and 8024 carry the label ‘Manufactured in England’.

   China has an appalling record of crushing democracy, suppressing free speech and supporting Russia’s murderous war in Ukraine.
   https://www.theguardian.com/world/2022/may/10/china-pro-russia-propaganda-exposed-by-online-activists-ukraine
   China is expanding their influence and control of raw materials. They have a long-held policy to invade Taiwan. They are positioning themselves globally in a similar way that Russia has positioned itself with fossil fuels in Europe.
   https://www.theguardian.com/world/2022/may/26/deal-proposed-by-china-would-dramatically-expand-security-influence-in-pacific

   New evidence of China’s inhumane treatement of Uyghur people:
   https://www.bbc.co.uk/news/resources/idt-8df450b3-5d6d-4ed8-bdcc-bd99137eadc3

   Companies like Audient can cosy up to a brutal totalitarian regime because their customers don’t kick up a fuss.
   Embracing economic ties with China have done nothing to improve democracy and human rights in China, nor make the world a safer place. They have only served to make a totalitarian regime stronger and bolder.

   1. Sonic, what you say is serious stuff and important. But, as others have expressed, this is a website for synth lovers, not politics. Let’s have fun here, even if we are nasty SOB’s toward each other in our synth snobbery.
      Love

      1. I appreciate and understand your comment. Electronic music has been an enormous part of my life for many decades. It is fun but also serious.
         One should never sideline ethics, nor allow personal greed to cloud your judgements.
         We live in a deeply connected world that is at a turning point environmentally and politically.
         The typical hip marketing spiel that accompanies Audient’s EVO line promises to release your personality and creativity. In reality the new interfaces offer little more than the previous couple of generations, they have a high environmental burden and have extremely poor human rights.
         Feel free to pass over my comments. I’ll try to confine my posts about ethics to only new product announcements.
         Companies like Audient will continue to trade off environmental and human rights concerns in the name of profit until the consumers hold them to account. They’ll throw a little smoke and pixie dust marketing in there distract us like the fools we are.

         1. All of Audient new products are made in china, but 99% of pro audio interfaces manufactured in china, It will be easier to to name the one who don’t
            Like Prism, Lynx, RME (not included Baybface series), UAD (not included Apollo twin and below), Motu (not including the m-series), Antelope (don’t know if all) Maybe also Metric Halo and some other very exotic ones. But that is about it, I may forgot one or two brands but all others are made in china and the one who don’t are using many made in china parts.
            I for one trying to avoid “made in china” audio related products but it’s a hard task and i do that for completely different reasons than yours.

            1. Thanks for the insights. I’m disappointed that the Babyface range are also China made. RME are a progressive company with amazing driver support for even decade old products.

               1. Even two decades, they are the best but I guess they have to keep their business competitive against UAD twins and Apogee newer Duets

      2. You’d thing the perfect place to discuss the politics of synth manufacturing would be a synth website though, right?.

   2. The US invades any country it feels like and bombs schools and hospitals. Therefore, we should all stop buying Moogs.

      btw something like 70% of the world’s silicon is mined by Uyghurs. I think you should stop buying electronics entirely and smash the computer or smartphone you’re using to post here.

      1. “The US invades any country it feels like and bombs schools and hospitals.” Simply not true. You may feel that it is true and you can certainly find evidence that the US has invaded some countries. Apart from the Iraq wars, the US has mostly been containing itself to somewhat justifiable military actions in recent decades. I’m sure you’ll disagree.

         China silicon production is indeed very interesting and another example of why we should hold them to account. China have no technological nor geological advantage that enables them to produce more ferrosilicon or silicon metal than other countries. What they have is forced labour and government control. Ultimately that allows them to overproduce and dump at low prices.
         The Critial Raw Materials Alliance have this to say:
         “There is no level playing field between the EU and its main competing regions in terms of policy in the energy, climate and environment fields. European Silicon metal producers are faced with fierce and often unfair competition from third countries. The still existing Europe-based commodity production must be preserved if the EU wants to avoid exposing its main economic value chains and sectors to a total dependence vis-à-vis external raw materials supply.”

         Sheffield Hallam University produced an in-depth report on Uyghur forced labour being used in solar panel production:
         https://www.shu.ac.uk/helena-kennedy-centre-international-justice/research-and-projects/all-projects/in-broad-daylight

         “I think you should stop buying electronics entirely and smash the computer or smartphone you’re using to post here.”
         You’re probably correct on the former and entirely wrong on the latter. You need to learn from your mistakes not beat yourself up about them.

      2. “70% of the world’s silicon is mined by Uyghurs”
         No that I support such policies, but how much of that goes into the world’s semi-conductors?
         Only a fraction of all silicon goes into electronics.

         Maybe you should broaden your research about chip production in the big wide world.
         A rather important chip maker from Taiwan (democratic with freedom of expression):
         https://esg.tsmc.com/en/update/responsibleSupplyChain/caseStudy/8/index.html

         Stop being so defeatist.
         https://en.wikipedia.org/wiki/Learned_helplessness

3. As far as I can determine, this interface is the only 24-channel I/O interface that sells for well under $1k and that doesn’t require a Thunderbolt port. I need a 24-port interface, and until this, the only device that was available for under $1000 was the Presonus Quantum 2626 ($700). Currently, I am using a Focusrite Scarlett 18i20 with an OctoPre to give me 16 audio channels. This has required me to use a couple line-level mixers to accommodate 20 channels of inputs from my synths. Originally, my plan was to wait until the Zen-4 AMD processor release, which should occur in the late summer, and to build a new 32-thread computer to host my DAWs (assuming that at least one motherboard manufacturer will incorporate Thunderbolt-4 in their design). Now, it appears that I won’t have to wait as long to get an interface that will accommodate all of my synths directly. It’s about time somebody did this.

   1. If you have a desktop you can add a thunderbolt pcie expansion card,
      As for your new Ryzen system, TB4 is not needed/used by pro audio interfaces.
      For best speeds you can use PCIe interfaces with only digital i/o and connect external converters/preamps.
      But Steinberg UR824, MOTU 828 MK3, 828x, 828es, 896MK3 are all 24 i/o (or more) usb2.0 and cost below 1K$.

4. I’m wondering if you can only do 24 tracks at 16bit 44.1
   If it can go the 16 extra tracks thru ADAT at 24/96 I’m buying it next

   1. Assuming that these are standard ADAT interfaces, then the answer would have to be no. If you want to increase the frequency resolution to 44.1 or 48k you would need to have twice the number of available ADAT channels. The only way the ADAT interface can achieve those frequencies is to multiplex their channels. However, standard ADAT does allow 24 bit quantization. The overwhelming majority of people who think they can hear a difference between a signal encoded at 22k and 44k are only imagining it. Multiple controlled psychoacoustic studies have determined that the number of listeners that can actually do that is statistically insignificant. On the other hand, the difference in dynamic range provided by 24-bits over 16-bits is detectable by most people. So, the myth of 44/48k continues on, so it seems.

      1. Please post your sources for the listening tests you are referring to.

      2. Seems your answer is “half true” for half the sample rate 🙂
         I think you meant 44.1/48kHz vs 88.2/96/192kHz?

         “the difference in dynamic range provided by 24-bits over 16-bits is detectable by most people”
         No, it is scientifically proven human can’t detect 96dB (16bit) compered to 144dB (24bit) of dynamic range. The advantage of 24bit is only when recording/mixing.

         1. I’m not sure what you are saying. The total number of dB doesn’t have anything to do with it. It is the quantization interval that human hearing can detect. When you divide the total dynamic range by the differences in amplitude when the steps are smaller (e.g., as would be the case when comparing the quantization steps in 16-bit values to the steps in 24-bit values), the larger the scale, the smaller the steps. The extreme example of this is zipper noise when you quantize to 256 steps or below (e.g., MIDI). Human hearing is capable of some incredible time-related differences in sound but not so much to frequency. For example, when we are resolving spatial location of the origin of a sound, the brain can resolve very small differences between the arrival of the sound at one ear vs. the other. The speed of sound is constant, but the frequency of sound that can be heard can vary over a range from about 20Hz to about 14kHz for most people. When the frequency of the sound is low, it takes the complete wave longer to affect both cochleae (because of arrival time of the peak at the ear drum) so we loose the ability to localize low frequency sounds. Yet, for high frequency sounds we are able to resolve arrival differences of less than 1/250,000 sec. The problem with the theory that if we increase the sample frequency of a sound to preserve “more” of the high-frequency components of the sound, is that those changes that look awfully good on paper are lost to the transduction mechanism. What is the myth is that those higher frequencies that everybody agrees we can’t hear are lost to the transducer (i.e. the ear, the auditory canal, the mechanical amplification system of the middle ear, and the viscosity of the fluid in the inner ear as it supports the waves that bend the hair cells).

            1. A lot of big players spent a lot of time and money to decide what specs CD audio would be.
               They chose 16bit 44.1khz for very good reasons.

               Again I ask you to post your links to the research that says most people can tell the difference between 24bit and 16bit audio.

               The lack of paragraphs in your writing speaks volumes about how much you really want to communicate with others, rather than subject them to a monologue.

               1. “Again I ask you to post your links to the research that says most people can tell the difference between 24bit and 16bit audio”

                  +1

               2. I hope you mean, lack of paragraph separation not lack of paragraphs. One thing I’ve seldom been accused of is being terse in my writing. As for references, I don’t have them at my immediate disposal. Over the course of the last fifty years I have done a lot of writing in which I have cited well over a couple thousand papers. I can remember facts and factoids but not all of the references, unfortunately. If you want to find the references you can look for them. I believe that most credible audiology and psychophysics research would be in journals covered by PubMed, so that would be a good place to start. Otherwise you can just choose not to believe me. I’m not going to get into a pissing contest with you over my veracity, just like I’m not going to lecture this list with my political opinions. I’m not copping out, it is just that I’ve got more important things to do with my time. If you believe me to be wrong, you could always present some references that support what you believe to be true.

                  1. So you saying something that is completely the opposite of what is scientifically tested and known for many years:

                     “the difference in dynamic range provided by 24-bits over 16-bits is detectable by most people”

                     But don’t have any facts to back it up because you don’t remember what reference you based on?
                     Say, If it was so easy to detect by most people shouldn’t it be all over the web?
                     All of this poor people who listen to audio in 16 bit don’t know what they missing…
                     Ok…

               3. The CD audio standard had to accommodate early 1980s DACs and the mount of data that could be recorded on a single disk. 24-bit DACs running at 96kHz were simply not an option.

            2. “When you divide the total dynamic range by the differences in amplitude when the steps are smaller”

               Definitely not. Dynamic range is not a fix value, what you call “quantization interval” is discreet values that represent loudness levels. 24 bit gives you a max of 144dB of dynamic range or 16,777,216 values, This values added on top of the 65,536 discreet values of 16 bit. so you get higher headroom, higher amplitude without clipping or if you like larger bandwidth.

               As for all you wrote, Again half truth at best, I don’t know what you are based on but we most definitely want to different schools, in different universes.

               1. Not exactly. The total dynamic range (i.e., the difference between some low limit and some high limit of some physical entity that can be measured on an interval or ratio scale) is usually specified in terms of some reference item or device for which it is calculated (e.g., an instrument (musical, medical or otherwise), a recording device, a medium, etc.). It is constant only for the thing for which it is being referenced. Each “thing” that gets measured will have a different dynamic range, and any single “thing” will have a different dynamic range depending on how we measure it.. For example, when we are talking about the amplitude of sound produced by an acoustic guitar there is no argument that the lowest amplitude a guitar can can produce is 0 dB (which will be true for any acoustic sound emitter when it is not being excited), the maximum amplitude depends on how we measure it, which is why when you see a measurement in dB it usually will (or at least should) contain the qualifiers on how it was measured (e.g., for the acoustic guitar… How the strings were excited (strummed, plucked, fingered, etc.); AND the measuring device (manufacturer, device name, device model number); AND how the measuring device was calibrated A-Weighted, B Weighted; AND how the “sample” was taken, e.g., peak; AND a distance from the particular place on the guitar; AND the axis(es) of measurement). If you change any of those parameters, the guitar will be found to have a different dynamic range because the maximum amplitude of the sound will be measured differently. So, let’s say that you strum the guitar with some standardized amount of force with all of the strings open and you have a calibrated B&K precision microphone capsule positioned 100mm from the center of the soundhole on a direct axis with it and you are calibrated for a A-weighted peak measurement. Then, on the first strum you record 97dB, on the second, you record 100dB, on the third, 96db, for a total of 20 independent strums. The average of the dB measurements could them be used as good representation for the dynamic range of the guitar given those measurement parameters.

                  For the sake of argument, lets say it averages 97dB. Then, no matter how you quantize the measurement, the dynamic range of that guitar under those conditions is 97dB. If you measure it on a scale with only eight available measuring points (8 bits), or 24 measuring points (24 bits), the guitar is still going to have a dynamic range of 97dB in either case. It doesn’t matter whether you are measuring an acoustic guitar or a nuclear blast, the sound pressure is going to be the same no matter what you measure it with, and if the measuring device is calibrated for the measurement, the full 8 bits (for the 8-bit device) will be the same as the that for the full 24 bits (for the 24-bit device) and will reflect maximum recorded amplitude and the total dynamic range of the event. So, independently of what university we learned this in, that is the way it is. The difference (assuming the proper calibrations) is not in the maximum loudness (or any loudness, for that matter) that can be represented, it is the accuracy with which the loudness can be measured.

                  Above, I have been discussing static measurements (e.g., single strums, single bomb) measured for peak amplitude, because that’s how dynamic range for sound emitted by things is measured. However, you appear to be revealing your fundamental misunderstanding of how the dynamic range relates to what sound engineers refer to as “headroom” when recording an analog signal, repeatedly OVER TIME, as would be the case in digital audio sampling. Remember, at the ends of the sampling circuit there are A/D-D/A converters, and “amplifiers” that precede the circuit and follow the circuit that are analog devices. Except when discussing things like “quality of the preamps” frequency response has nothing to do with this. So, just assume the sampling rate is 40kHz. We will also be using the same analog paths on either end of the converter circuit, so the only thing we are changing is the amplitude quantization of the digitization (16 bits vs. 24 bits). The first thing we have to get to is your confusion about what the “dynamic range” difference of the two systems actually means. This measurement has nothing, at all, to do with loudness. It is simply the representation of the maximum difference that any two loudness extremes can be represented, the volumes don’t change in the real world, it is only the digitized representation of the amplitudes corresponding to those volumes that are of concern here. The 16-bits is capable of 96db of amplitude RESOLUTION because 6dB represents a doubling of amplitude and 16 doublings gives you 96dB. Accordingly, 24 doublings at 6dB per doubling gives you 144db. So, both systems are “seeing” identical analog representations of the signal, and in both systems, when all of the bits of the converter are high, the same amplitude signal will be decoded at the input to the D/A. At a reference point of 0dB (e.g., what the dB meter is reading on a recording input of the console), both systems are representing the maximum amplitude of the signal (and it is the same amplitude). Now, if there is no signal (i.e., a 0V input to the A/D converter) that signal on the 16-bit system will be 96dB lower than 0dB and that signal on the 24-bit system will be 144db lower in respect to 0dB. Notice, 0Volts is no signal. It is the same for both systems and it will be -96 dB for the 16-bit system and -144dB for the 24-bit system. Also, 0dB represents the same maximal signal amplitude for both systems. So, what actually changes, and how do we get to the concept of “headroom”?

                  [I hope I am providing enough paragraphs to accommodate Sonic, here]… Anyway, remember what I discussed in the section on how the dynamic range of “things” is determined. The number of bits in which you divide the total dynamic range of a signal reflects the accuracy of the amplitude measurement at that point in time. The more bits you have, the more accurate the representation. The goal in recording audio is to have the “best” representation of the signal possible (given the limits of human hearing, but I digress). Now, it is possible to make the 24-bit system be exactly as accurate as the 16-bit system. We could do that by allowing the 24-bit system to set what it is referring to as 0dB to 48dB lower than than the actual 0dB (i.e., 8 bits x6dB = -48dB). In this newly configured system, the maximum amplitude of the signal will record at -48dB (on the meter) as opposed to 0dB. Again, nothing changes on the analog input to the converter after the input “level” is adjusted. The guitar is still at 97dB and the maximum amplitude of that signal will be represented by all of the 16 bits WE HAVE MADE AVAILABLE going high for input into the D/A. Since we are back in the analog domain after the D/A, we can amplify the signal to whatever amplitude/loudness we desire. So, effectively, we have turned a 24-bit system into a 16-bit system, and this system will preform identically to a 16-bit system (everything being equal except the quantization of the converter). So, how does all of this relate to “headroom”?

                  Well, as I said “headroom” is a contrived word. What it describes is the decision that can be made on a 24-bit system to lower what is going to be considered 0db to a lower amplitude than true 0db. So, let’s say we decide that the maximum amplitude of a signal we will be recording will be set to occur when 20 bits of the 24-bit word are used (i.e., -24db — 6dB x 4 bits). We now have a system that will encode the incoming analog system at a much higher resolution (i.e., more accurately) than the 16-bit system and still give us an extra 24dB in case the signal should actually supersede what we thought was going to be the maximum signal when we set the levels. Moreover, we will still still record it accurately as opposed to clipping it, as would be the case in a true 16-bit system. This is what “headroom” is. It is that amount of “slop” that we want to build in to the recording process for accommodating signals that get hotter than we may have expected. It has nothing to do with loudness.

                  Sorry to be so verbose, but given your beliefs about what the terminology means, I just thought I might explain it so that others may also benefit.

                  1. I wished you spend half of this time to find the references and the test that proving what you wrote and what I and Sonic replay to you about (not what you want to talk about)

                     You wrote:
                     “the difference in dynamic range provided by 24-bits over 16-bits is detectable by most people”

                     Can you provide any reference/proof/test?

                     1. First of all, nobody can ever scientifically “prove” anything. That’s not the way science works. Proofs are something that exist in mathematics, and maybe in criminal trials. As for talking about what “I want to talk about”, isn’t this what this list is for? Just like most people here, I enjoy reading what others write. If i read something that I feel needs to be elaborated on, answered, or corrected, I reply. I tend to be verbose. It’s probably because I have been programmed to be that way. You made a number of statements challenging what I said about dynamic range and quantization in a much more terse statement I previously made. The verbose explanation of what I was saying was probably too professorial, but I’m sorry, that’s what I do in the real world as my day job. I infer from your writing that English isn’t your native language, and if that is causing me to misunderstand your criticisms, I’m truly sorry. The amplitude quantization issue is actually a minor one for the discussion here, and it appears that you, Sonic and I are the only people here that are really concerned about what constitutes “good enough for most people” vs. “can be discriminated by most people”. I think the more important thing, is that we agree on the frequency quantization issue, which is what got me to bring up amplitude quantization in the first place. Yes, in the time I spent writing here, I could have probably performed a substantial PubMed search to find the articles I was writing about. But, what would be the fun in that? 🙂

   2. Just count the ADAT ports, If you see only 2x In and 2x outs you can only use 16 ADAT channels x at 48kHz or 8x 96khz (if it support s/mux) or 4×192 (if it support s/mux4).

   3. ADAT can combine channels as mentioned by John Rossi and tash. It’s called SMUX (sample multiplexing).
      1 ADAT port runs 8 channels. By default each run at a maximum of 24bit 48khz.
      If your kit supports SMUX II then you can choose to have 4 channels at 24bit 96khz
      If your kit supports SMUX IV then you can choose to have 2 channels at 24bit 192khz

      From my understanding, all people can’t tell the difference between 48khz and 96khz when the audio stream is playing mixed music. Only when the quality to discern is ‘is it real or not’ when recording live acoustic sound does 96khz come into play. In such circumstances the quality of the playback system and environment are critical.

      1. “From my understanding, all people can’t tell the difference between 48khz and 96khz when the audio stream is playing mixed music. Only when the quality to discern is ‘is it real or not’ when recording live acoustic sound does 96khz come into play. In such circumstances the quality of the playback system and environment are critical.”

         Yes, except that in the case of what you are referring to as “mixed music”, statistically speaking, people CAN’T discern it. However, I think that you may be correct about which frequency resolution more closely approximates reality of a “live source”. However, to get to that conclusion, you are actually asking a different question than about “quality”. I realize is is a subtle differentiation, but the “reality” of a perception of a recorded event may be different that a detected difference between two instances of a recorded event in an A-B comparison. I have to concede that I’m not familiar with the research you are citing, but it makes sense the same way that looking at a single frame of a motion picture recorded at 24.9 FPS may not be perceived as being different than that same frame recorded in time at 120fps. But, in the movie film example (sorry, but again I don’t off-hand remember the reference, but I think it was one of the Italian-American directors)… when asked, people couldn’t articulate a difference between the footages but when the film was of first person riding on a a roller coaster, electrophysiological measurements unambiguously revealed that the people were “experiencing” a difference that could only have been explained by the film speed (i.e., sample rate).

         So, just to correct myself. What I was trying to say was, that statistically speaking, people can’t differentiate between the “quality” of a 44k sample rate vs. 96k in a randomized A-B comparison of two otherwise identical clips.

5. Another 24 I/O interface with only 8 physical line inputs, so you need to buy 3 of them to get the full input count. Might as well get yourself a MOTU 24 AI and combine it with your existing interface as an aggregate device to save some money and rack space.

   1. Only Apple computers support aggregate, don’t they?
      And even then I’ve heard they are a pain to sync without a word clock.

      The MOTU 24 AI is twice the price and will need breakouts for all the analogue inputs.

      Some folks like ADAT:
      Modular folks with Expert Sleepers ADAT to CV kit.
      Folks with computers that want to output 8 digital channels to another system with perfect quality and extreme low latency. Using analogue would require 8 cables, 8 physical ports on both machines, would be prone to noise and distortion and would have additional latency to the signal due to DAC then ADC conversion.
      ADAT gives you the choice about which converters you use. Give the drummer and bass player the shit ones and your singer the good ones :p

      Long live ADAT!

6. A Roland UA-1610 Studio Capture with 16 ins/10 outs is like ~$700 used right now.

   1. Yes, the UA-1610 has 16 physical inputs in one box. It is, however a two space rack device (which also seems to be no longer manufactured, according to Sweetwater), If all you need is 16-channels, for about the same price you can get a NEW Focusrite solution (18i20 & OctoPre) which also gives you the advantage of having at total of 16 presumably higher quality Focusrite mic preamps as opposed to the 12 on the Roland. Also, if you are more concerned with the total number of line inputs, and don’t want to argue about the quality of the mic preamps (as in my case), you really can’t beat the Behringer U-Phoria UMC 1820 coupled with a Behringer ADA 8200 for about $500, if all you need is 16 line-input channels. However, if you did want to argue about the quality of the mic preamps, then I would suggest that you listen to the Midas preamps on the Behringer stuff, for you particular application, before you draw conclusions based on your biases toward the parent company or the country they are manufactured in. You might be pleasantly surprised.

7. Its 2022… can we please stop using DAT connections on interfaces. Just put all the connectors on the box. Sourcing DAT Expansions anymore is a pain in the butt.

   1. I think you don’t understand the meaning of ADAT and why it is there. It’s not for “Sourcing DAT Expansions” (ADAT, Not DAT, DAT are the digital tapes, not related).
      Even if it was 32 analog i/o interface ADAT Connection are very helpful, especially for people who have ADAT Converters or old interfaces that can work as converters, Or like with the Antelope Orion, lot’s of i/o but shitty driver (or out of date driver) that you can be remedy with the ADAT connection and RME Digiface USB.
      At some point this interface will be useless without an ADAT.

   2. This is a bit extreme, but what if you needed 56, or even 112 inputs, and what if those inputs were coming from mics or instruments that were tens or hundreds of meters away? That would require about seven (for 56) or 14 (for 112) rack spaces of equipment in one place (i.e., the interface that would presumably plug into the recording device), and miles of shielded 3-conductor low-impedance cable. ADAT allows a unit with the mic or instrument inputs to be located closer to the microphones or instruments, and for the output back to the main interface it requires just one tiny fiber optic cable. ADAT technology is quite old, however, and dates back to the early 90s and Alesis’ revolutionary 8-channel recording system called ADAT that used low cost S-VHS tape as the recording medium (i.e., Alesis Digital Audio Tape). Since then, digital recording technology has come a long way and we now have newer systems, e.g. MADI, which allow 56 (in the case of MADI) or more channels of digital information to be transferred via one cable.