Focusrite Scarlett 2i2 (4th Gen) Interface Review

[jkim]

This thread is very interesting. I am currently testing two new, cost-effective audio interfaces: SSL 2 MKii and Topping E2x2.

This unanticipated journey started by my motivation for replacing a dead USB audio interface---EMU 0404 USB---that I've used for audio gear measurement for 16 years.

I narrowed down to these two products and ordered them---in fact, ended up ordering E1DA Cosmos ADC along with Scaler. But I still wanted to evaluate the SSL and Topping.

My main interest was in their ADC performance. The issue was with their unbalanced inputs through TS plugs. ADC of their balanced input was fine, although I was later spoiled by E1DA's performance

Initially I scratched my head, "what is wrong with unbalanced inputs?" I first suspected the cable quality, which was ruled out quickly. I was puzzled: How come unbalanced input ADC quality is much, much worse than balanced input ADC on both devices? The EMU 0404 USB had no such issue.

But then I recalledI had been using the 'ground lift' switch on 0404 USB without remembering its effect for so long. And searched this forum and found this thread. Turns out it is not just my experience.

I have been testing their ADC performance using a single laptop PC which is also connected to a DAC (Topping D50 III) used for signal generation. What would be the best way to break a ground loop in this case? I could use two separate laptops (battery-powered), one for generating tones and the other for measuring. Or use a USB isolator? Or use impedance balancing for unbalanced to balanced connection?

 

[jkim]

I resolved a ground loop issue in my particular setup. Now, the noise floor of ADC with unbalanced input nearly matches that of ADC with balanced input. Just a tiny bit worse.

But have another question.

I keep seeing a clearly greater 2nd harmonic product from unbalanced input than from balanced input. And somewhat higher 3rd harmonic from balanced input than from unbalanced input.

I mean in exactly the same setup except for balanced vs. unbalanced connection. Why? I think I read a related comment somewhere but cannot locate it..

 

[AnalogSteph]

I keep seeing a clearly greater 2nd harmonic product from unbalanced input than from balanced input. And somewhat higher 3rd harmonic from balanced input than from unbalanced input.

This is expected.

Even-order distortion (H2, H4, ...) is asymmetrical. With H2, a sinewave might get rounded off at the top but sharper at the bottom.

For a moment, let's a treat a balanced input like a pair of unbalanced ones, one of which is subtracted from the other. (Side note: That actually works decently if you need a mono balanced input in a pinch but unbalanced stereo is all you have. It often has to be done manually, but e.g. REW supports this for both inputs and outputs.)

Let's say they both distort the signal in such a way that +1 comes out as +0.9 and -1 comes out as -1.1.

Now we'll feed this input with a BTL signal: +1 becomes +1 on the hot leg and -1 on the cold leg. In an ideal distortion-free input, we would expect a differential of +1 -(-1) = +2.

+ input (hot): +1 in --> +0.9 out
- input (cold): -1 in --> -1.1 out
-----------------------------------
Differential: +0.9 - (-1.1) = +2.

The distortion has effectively been cancelled!

The same does not apply to odd-order distortion like H3, which you might see as symmetrical rounding of the sinewave tops. When both +1 and -1 become +0.9 and -0.9 equally, there is nothing the input can do about this. So you can only hope for H3 to not increase at best. It may go up the balanced signal is at higher absolute levels, including after the balanced receiver.

Note that while I have lumped together all the distortion in the input in this example, in general it's actually the product of the entire analog signal chain.

 

[jkim]

This is expected.

Even-order distortion (H2, H4, ...) is asymmetrical. With H2, a sinewave might get rounded off at the top but sharper at the bottom.

Thank you so much for the enlightening. This makes perfect sense.

As you mentioned, this phenomenon should occur in the entire signal chain. In this particular case (DAC to ADC), the balanced design within a DAC w/ two DAC chips should be the main contributor because I'm comparing the DAC (D50 III)'s balanced (XLR) and unbalanced (RCA) outputs.

Speaking of a two-dac-chip design, based on what I've learned from your post, I think it's mostly a marketing gimmick. The major part of a DAC's performance should be from a good chip architecture and surrounding circuit design, not mainly from the use of two DAC chips differentially. Sure, using two chips may produce a tad lower noise floor and 3rd harmonics, lowering measured SINAD. But the difference shouldn't be audible. Anyway, current DAC technology way passed the limit of audibility..

 

[Miguelón]

Overall I'd say this is likely a small upgrade over the previous generation even if the DAC is slightly, and likely inaudibly worse. The headphone amp, though still junk, is actually improved and possibly audibly so.

With that said, I'm surprised this interface remains so popular. My one attempt at a Focusrite interface ended after about 2 months: they stopped updating the drivers at the pace of the operating system and I couldn't use my interface. I assume they do better now. I couldn't give it away. I had a similar experience with EMU FWIW.

Good news is that I replaced the Focusrite with an Apogee Duet which gave me 10 good years and didn't have a driver issue until there was a processor typology revolution at Apple. The Apogee had considerably better preamps (dead silent and tons of gain) and headphone amp (lots of power), but was 6X the price. The Apogee was also far less user friendly with its breakout cable and it required you to have a window open to really see what you were controlling: a massive design fail in my book, but the sound and support kept me happy enough.

I could never buy another Focusrite product though. That lack of support ruined them for me. At least the Apogee worked and sounded fantastic.

Now I've got an IK Media Duet. It works perfectly and sounds good enough with the right headphones--MEZE Noir with Brainwavs ear pads which are very easy to drive. Its preamps are easily good enough for condenser mics and really good enough for dynamics. Not as outstanding as the Apogee, but good enough if you're careful with your gain settings. Best part is that it has real MIDI inputs/outputs.

If you're a hobbyist musician, I'm sure there are several interfaces that outperform this one in every way.

One of the success of Focusrite is simply the price, another maybe marketing: they acquired the reputation to be one of the best entry level interface for beginning the activity, and save some money for other gear…

As I see in @amirm list, Motu M2 is on the happy side, will be my next choice.

AFAIC the Scarlett is not audible in performances, but I record and play just for pleasure, perhaps in high-exigence professional world is not acceptable.

YouTube, Spotify and other internet platforms changed the edition and mixing world, now many people start a professional activity at home studio setups, and should be prudent when investing their money.

I’m curious for the RME babyface pro, do you know if has been measured?

 

[Blumlein 88]

I’m curious for the RME babyface pro, do you know if has been measured?

RME Babyface Pro FS Portable Interface Review

This is a review and detailed measurements of the RME Babyface Pro FS Portable professional audio interface (ADC, DAC, headphone and digital I/O). It was kindly sent to me by the company. The Babyface Pro FS Costs US $899. It is an upgrade to the non-FS version of the same product. The...

www.audiosciencereview.com

 

[Miguelón]

RME Babyface Pro FS Portable Interface Review

This is a review and detailed measurements of the RME Babyface Pro FS Portable professional audio interface (ADC, DAC, headphone and digital I/O). It was kindly sent to me by the company. The Babyface Pro FS Costs US $899. It is an upgrade to the non-FS version of the same product. The...

www.audiosciencereview.com

Thanks!

POST EDITED: very surprised reading Motu M2 values over RME babyface, that save me a lot of money since I was considering putting some money apart to the latest

 

[tomchr]

One of the success of Focusrite is simply the price, another maybe marketing: they acquired the reputation to be one of the best entry level interface for beginning the activity, and save some money for other gear…

Their biggest selling point, for me, was that they made specs available. I had a Focusrite Saffire 2x4 a number of years back. That was a solid piece of hardware.

The sad truth is just that corners will need to be cut when the world expects a piece of electronic hardware to cost less than $100.

I stil think you get remarkable value with the Focusrite products. They work well when operated within their limits. The single-ended performance of the 4th gen a noteworthy exception, but from the comments Focusrite dropped on my YouTube video, it sounds like they're working on a fix.

Tom

 

[Miguelón]

Their biggest selling point, for me, was that they made specs available. I had a Focusrite Saffire 2x4 a number of years back. That was a solid piece of hardware.

The sad truth is just that corners will need to be cut when the world expects a piece of electronic hardware to cost less than $100.

I stil think you get remarkable value with the Focusrite products. They work well when operated within their limits. The single-ended performance of the 4th gen a noteworthy exception, but from the comments Focusrite dropped on my YouTube video, it sounds like they're working on a fix.

Tom

Yes, for the amount of things it does is a very good value for the money.

From my subjective point of view, nothing to regret because I’m not a professional and bought it in a time where struggled with connecting my studio monitors to a dongle DAC plugged to an ipad… any voltage or impedance problems maid sound super noisy, or just the dongle was quite bad (probably the last one, Dragonfly Red; pushed over 60% volume has a tone of distortion, on the audible range).

So I bought the audio interface since it appeared to me that the “mysterious” bad sound (good speakers, 8030C) should come from a “mysterious” voltage or impedance value that only the natural thing to plug into studio monitors should have…

The result was good, even today I use the interface from record myself on a digital piano or to do measurements.

I don’t know if professionals can do good recordings with Focusrite 2i2 4, today many broadcast come from youtube or spotify so probably is ok to that standards.

 

[SMJH]

Long story short I don't feel like dealing with imports again, and the only local DAC options available to me are audio interfaces. Specifically Focusrite gen 4 and UA Volt.

What should I get for my Adam T8V monitors?
Gen 4 Solo or Gen 4 2i2
UA Volt 1 or 176 (If I want a fancy enclosure)

Gen 4 uses a modern DAC chip (CS43198) with supposedly better performance (and a couple forum members that swear it sounds much better to them), but as described here, has volume knob issues at least on the 2i2, plus potentially audible clicks on transients. The Volt's only downside is its dated ~2005 DAC chip (CS4272).

PS. Headphone output performance doesn't matter to me, I just want the best balanced out for speakers. And since I will be using Jriver's internal volume for music, the hardware's volume knob will only be used for non-critical tasks such as YouTube/Work/Games.

 

[bennybbbx]

but as described here, has volume knob issues at least on the 2i2,

I think all with analog stereo pots have this problems soon when change volume more times per day. I test with my replaced mackie big knob and do the volume changes in range 10 to 2 o clock. after 1000 movements(guess 30 per day 30 days =900 reach in 1 month) i get diffrence of more than 0,7 db between left and right in that range. when look in the datasheet of alps or alpha pot(mackie big knob have large alpha Pot) they write tolerance of +/-20% and 15.000 moves. there are some reports that left right diffrences are only on 8-9 o clock on the pots. this i have on my pots too, and i notice after test the reason is because the knob is used in this range. for the mackie test in range 10 upto 2 o clock i reduce in DAW the main output level 12 db, so i can make on the knob louder.

I really hear music and change volume ~30 times per day because music i want best sound possible and lowest ear wear and tear and neighbours need hear. I hear with ~75 db i also make louder on silent music parts and reduce on loud parts. My ear are very sensibly with loud hearing. > 82 db give no good feeling. I often make music many hours. best feeling i have with 74-78 db of music. help that i hear good and i near 60 . maybe other people that do not change volume can live with analog pots. for example when TV hear on my parents TV when i was very young volume was i guess 50-60 db and so never make it louder or more silent. TV i hear also in this volume. On youtube videos i need change volume alot between cinema actor videos or music only videos

All is clear for me. I never use since i notice this a stereo device which have stereo analog pots. this happen mostly in studio equipment. hifi equipment(which offer remote control) and have no motor driven pot, have digital control. this is good. On my apollo audio solo there is no diffrence between left and right channel measurable. REW limit is 0,1 db . I hear left right diffrence more than 0,3 db at ~75 db volume. as bad stereo image and ear tire when hear longer (as 1 hour)

I get answer from focurite about
durability and tolerance, after long write.

The Scarlett 8i6 (3rd Gen) pot has a manufacturer tested endurance lifecycle of 10,000 rotations.

in later mailthey answer, my english is not so good, but it seem they write that this diffrence from left to right channel is not on all pot positions only on full level. or how do native english speakers understand this ?

The tolerance on our maximum output level is 0.5dBu. This means that there could be a up to 1dBu difference left to right if the sides were at the opposite extremes of +/- 0.5dBu tolerance. It's worth noting that typically we are tighter than the 0.5dBu tolerance to the maximum output level specified for the product and this spec has provided adequate performance for the vast majority of our customers


With that said, if your product is within the warranty period and has a potentiometer that has fallen out side of the 0.5dBu tolerance, we'll be happy to provide a resolution.

If there are really devices with better stereo analog pots i did not believe because nobody do tests about that. maybe i am to senible about left right diffrences.

sad that no famous testers do durability tests what happen after 900 pot moves . everybody can test yourself if you hear 0,5 db or more diffrence between left and right when hear not so loud. when hear louder then diffrence are less hearable. this force people to hear louder than necessary because they maybe notice in subconscious feeling that it sound not so good. as months ago. I suggest every people who hear music with headphone or speakers to buy a cheap sound level meter(30$ is enough, i have 2 and i compare them ond they have 0,8 db diffrence ) and measure what you hear and use good devices so you get happy with the sound when it is less than 85 db peak. neighbours get also happy when they need not hear loud low boom. on headphones you can use it too. put the microphone to ear position

 

[ppataki]

Gen 4 uses a modern DAC chip (CS43198) with supposedly better performance (and a couple forum members that swear it sounds much better to them), but as described here, has volume knob issues at least on the 2i2, plus potentially audible clicks on transients. The Volt's only downside is its dated ~2005 DAC chip (CS4272).

I can only comment on the Gen4 4i4, one of friends has one, see this project: https://www.audiosciencereview.com/forum/index.php?threads/diy-2-2-living-room-project.57939/
I also had it for a few weeks to listen to it and I can say that I subjectively loved its sound - very detailed, more towards analytical but in a good sense
I guess the 2i2 would sound the same

I also use volume control upstream in Jriver so fully agreed that any issue with the volume knob would be a non-issue for you

 

[bennybbbx]

focusrite also write which devices are digital controlled. but they all have many inputs and outputs and are expensive. A low input low output version they do not plan.

they write

You had mentioned previously about us making an 8i6 (4th Gen). This isn't a product in our current line-up, however the latest 16i16 (4th Gen) is the equivalent SKU to the 8i6 (3rd Gen) in our latest generation and uses digital control for the monitor output, which as you mention would avoid differences in level due to any offset of pot wear. This is the same on the 18i16 (4th Gen) and 18i20 (4th Gen). The 18i20 (4th Gen) also has digital control on the headphone outputs.

digital control on headphone output is important too, i think even more important as with speakers.

 

[SMJH]

I can only comment on the Gen4 4i4, one of friends has one, see this project: https://www.audiosciencereview.com/forum/index.php?threads/diy-2-2-living-room-project.57939/
I also had it for a few weeks to listen to it and I can say that I subjectively loved its sound - very detailed, more towards analytical but in a good sense
I guess the 2i2 would sound the same

I also use volume control upstream in Jriver so fully agreed that any issue with the volume knob would be a non-issue for you

Really cool project! I did not know Focusrite allowed all outputs to run simultaneously, I might actually get a bigger model for future-proofing. Surround can be fun.

 

[SMJH]

focusrite also write which devices are digital controlled. but they all have many inputs and outputs and are expensive. A low input low output version they do not plan.

they write

You had mentioned previously about us making an 8i6 (4th Gen). This isn't a product in our current line-up, however the latest 16i16 (4th Gen) is the equivalent SKU to the 8i6 (3rd Gen) in our latest generation and uses digital control for the monitor output, which as you mention would avoid differences in level due to any offset of pot wear. This is the same on the 18i16 (4th Gen) and 18i20 (4th Gen). The 18i20 (4th Gen) also has digital control on the headphone outputs.

digital control on headphone output is important too, i think even more important as with speakers.

This is what I heard:
Focusrite Scarlett Gen 1, 2, 3 all have analog volume. But for Gen 4, the Solo is analog (maybe) and the rest (2i2 and up) are digitally controlled.
The volume knob problem discussed here is digital in nature. It has perfect left-right balance, but has distortion if you listen at lower volume using the knob.

 

[bennybbbx]

This is what I heard:
Focusrite Scarlett Gen 1, 2, 3 all have analog volume. But for Gen 4, the Solo is analog (maybe) and the rest (2i2 and up) are digitally controlled.
The volume knob problem discussed here is digital in nature. It has perfect left-right balance, but has distortion if you listen at lower volume using the knob.

The 2i2 and 4i4 4 th. generation are analog control. The 8i6o 3rd (i have) they do not plan do as 4.th generation currently as support write . there should buy the larger version.

the diffrence between left and right can also see in distortion a little. at 8 o clock, i get this distortion values on the 3 rd generation. because in ear get very loud at 8 o clock distortion at 1 khz and more can be possible hearable on headphones . At 12 oclock on headphone out (on line out not) is no measure mistake. it happen on both channels

I do test with cheap thoman cable and a 10 eur gold cable, and i notice no diffrences when plug them in out several ltimes. so cheap cables are also good

 

[bennybbbx]

I notice it is also important to show which is the output level the diffrence happen. when the speaker have more gain(depend on the trim pot) in speaker the diffrence get larger.

here is screenshot. the output value is 67.2 db on left channel and 66,4 db on right channel at 100 hz. thats 0.8 db diffrence .on speaker out when pot is at 9 oclock. thats 1.2 % diffrence between left and right. input gain of preamp i set to full left(mean 0 db). I use 1-2 month the focusrite pot and get this lots diffrence. On 100 db output and 1 db diffrence i guess is not hearable. In case of analog pots headpones with low sensitivity are better. same happen when reduce the input sensivity of the speakers with the gain knob as much can. but best is avoid analog volume control

 

[SMJH]

but best is avoid analog volume control

Yes, I will contact Focusrite for more 4th gen information and will share here.

 

[bennybbbx]

Yes, I will contact Focusrite for more 4th gen information and will share here.

you can do that, quick verify if it is digital control or not, you can see on photo. if there is no position mark on knob, it is a digital. can also see on scarlet 18i16. the volume and input knobs have no position mark only the headphone knobs have a position mark, so this mean headphone is analog. another problem the focusrite have i see no LED ring that show position of volume. this is important i think to see how much volume currently have . this scarlet show only a levelmeter. but if hear a video that have very lowm volume and you have set volme half up and later or next day you show a video again then you get horrible loud sound. with a level position you see before that this is too much level for most videos. here can see how it work on scarlet digital control

in the apollo audio solo you have a level display and a volume knob LED position

digital control on studio hardware is in children's shoes . the universal audio with only 2 inputs and 4 outputs can use together (LR for speaker+ LR for headphone cost 450 $) . if you want subwoofer or bass shaker too this can only do with "life hack" by use a headphone cable that have 2 male TLRS 6.3 outputs

hiere you can see how the LED ring work. every segment in the ring around the volume knob have also 3 brightness levels depend on volume so you see position of volume knob.

 

[SMJH]

@bennybbbx Thank you so much, I appreciate your insights. I certainly would not want an analog knob. The position marker is solid evidence. It stupefies me why "pro" audio is so cheaply made. Everything else no matter the price moved on to at least 24bit digital volume...