So this thread is about voodoo instead facts… “My interface sounds so silky! That’s why I am paying 1000 euro for a fucking 5-dollar-DAC. I don’t know what’s going on, I just believe”
So this thread is about voodoo instead facts… “My interface sounds so silky! That’s why I am paying 1000 euro for a fucking 5-dollar-DAC. I don’t know what’s going on, I just believe”
YOu might not know …but a lot of money you pay for are the drivers ( research/development)…this is also true for nvidia quadro /gforce video cards …
The hardware spec’s are basically the same …only differences are the drivers …
Rme is known to code their in house usb drivers and use in house usb controllers , reason why they are rock solid …maybe the dac’s are not on par with metric halo or apogee …but they are good enough .
YOu might not know …but a lot of money you pay for are the drivers ( research/development)…this is also true for nvidia quadro /gforce video cards …
The hardware spec’s are basically the same …only differences are the drivers …
Rme is known to code their in house usb drivers and use in house usb controllers , reason why they are rock solid …maybe the dac’s are not on par with metric halo or apogee …but they are good enough .
Ok, that’s usual marketing hype about RME. I mean, “rock solid” nowadays is every single interface. I have read that RME drivers also bring benefits in performance, because coded in assembler or whatever. So are there any serious comparison charts?
I doubt the AD/DA converter chips by themselves are the decisive factor in a card’s ability to do its job, it’ll be like any other electronic device, the overall design and all components working together is what matters.
I love the RME but no, I don’t hear a major difference in sound quality between it and the UR22 I had prior, but I can see it on a spectrogram (less noise overall, lower noisefloor). The pres seem to handle harmonics better with big pad-type sounds than the UR did – a big string pad sound from the MKS-70 sounds nearly perfect through the RME without EQ, while through the UR22 the first few harmonics tended to glare too much. The RME also gives a much better sense of space and clarity, especially through headphones. I can’t say to what degree that’s confirmation bias on my part, but I did record some comparisons myself before selling the UR, and had a definite sense the RME handled big, wide sounds a lot better. Most of these differences are probably due to the pres and the circuitry around those rather than the AD/DA chips. Also, ability to handle lots of inputs at once without trouble, the number one reason I got it.
That said, I’d agree the price is a bit much – I love the device but don’t think it’s worth the full retail, you’re paying a bit of an “RME believes they’re an audio legend” premium. Should have been ~$200-300 less.
I have access to a high-end digital Soundcraft board at work, I don’t know the model but it’s like a $10000 device, and I don’t hear any difference between it and the RME. I have access to an analog board, too, but that’s not a very useful comparison if you’re committed to working in digital anyway, imo. However, for the big Renoise library I’m doing sampling outboard synths, I use the analog board when I can, more for fun than anything.
I feel like I remember seeing that AD/DA converter comparison you’re talking about, gentleclock, but if I had it should be i my bookmarks (it isn’t). Closest I’ve got is this:
Anyway, this is the RME UFX A/D: https://www.cirrus.com/en/pubs/proDatasheet/CS5368_F5.pdf
And D/A: http://www.ti.com/lit/ds/symlink/pcm4104.pdf
They’re not that bad! 
The surprising one to me is the use of the AK-series chips in older Fireface interfaces – the AK series was pretty good… in 2005, when I got the Edirol UA-25 I still have. But its pres are crap. My memory is rusty about this stuff though, I stopped worrying all that much about AD/DA years ago now…E
Edit: here’s a more recent compilation of the same list linked above: https://translate.google.ca/translate?hl=en&sl=ja&u=http://www.logsoku.com/r/2ch.sc/dtm/1429373163/&prev=search
Ok, that’s usual marketing hype about RME. I mean, “rock solid” nowadays is every single interface. I have read that RME drivers also bring benefits in performance, because coded in assembler or whatever. So are there any serious comparison charts?
Not it’s not …
Most usb audio interfaces use of the shelf usb controllers and standard asio drivers with their own GUI slapped on
., RME does NOT …whatever you want to believe
Might want to put off that tin foil hat of yours .
I am not saying others are bad , there are a lot of fantastic sounding , cheap interfaces on the market …choose what’s best for you
Best sound card is the one connected to the audio reproduction devices currently placed on my head or in front of me. The one with the silky sound. And the skullcrushing bass. It is the best, really, and if it was another brand and sound similar, it would still be the best.
Other than that for me is important the thing runs perfectly on linux. Not every device is so stable feature complete etc. for the tux.
Other point is minimum real world latency, in case you want to rock your guitar or mix with live dsp. Obviously pci cards usually can boast more power in this regard than usb, though I guess usb or firewire are - shielding wise and with more distance to lots of radiation inside the case - on the better side when it comes to ultra low noisefloor without any crap in it for your recording pleasures.
I did hear the difference between my current m-audio delta pci and the onboard crap, most notably in the high frequencies. Like the onboard card has slight problems with clarity there and counters it with some kind of harsh exciter device or so, but very subtle. And bass less defined somehow, but only vaguely. Meh. Production grade sound interface is probably the first best investment you can do for starting audio production, together with proper headphones/speakers that make the benefit audible.
Just moved into a new place and finally (finally!!!) have enough room for a dedicated studio space. Did a fair amount of research and decided to go with a Focusrite Scarlett Solo. Only been working with it for a couple of weeks, but so far, it’s been excellent. Very basic, 1 line/instrument in, 1 mic in, but it just “works” (which is what I was looking for). Latency is imperceptible - and what I LOVE about the unit is that there is a switch to direct monitor the inputs - so, in the event there was latency through the software, you could just flip over to listen to the signal directly. You get a nice visual feedback on the knobs which glow green when sound is coming in and levels are ok, and red if you are starting to clip. . . a really nice way of figuring out how to adjust the sound.
So far have run a Strat through the instrument input (worked fine - I know earlier versions had issues with guitars with “hot” pickups - no issues thusfar) and also run a turntable->Behringer VMX200->Scarlett Solo line in with no issues there. Haven’t tested the mic input yet but not expecting any bad surprises there. Sound quality is decent enough - not super-audiophile-amazeballs but pretty damn good for the cost.
Hard to tell how long this thing will last and, obviously, won’t be as good as a “production grade” interface, but for 99$ I have to say it’s pretty fucking boss if you’re just looking for a basic, solid audio I/O solution.
Ok replace my “production grade” with “musician grade”, ofc I didn’t want to tell anyone to buy sound chips that are suitable to making vinyl masters with, but rather ones that are designed for musicians producing stuff & sound better and more clean, direct, detailed and especially “honest” than the usual onboard or hype-up-the-3d-spatial-gaming-and-movies stuff.
Chiming in as we were discussing the Focusrite Scarlett 2nd gen on Linux not too long ago on the Linux Musicians board.
Apart from the marketing hype, it looks like they really did manage to reduce the latency in a measurable way between 1st & 2gen (as least for the 2i4)
You can expect to shave off ~2ms for most configs (I’m talking about measured round trip latency here, with same buffer, period & sample rate settings).
See: https://linuxmusicians.com/viewtopic.php?f=24&t=15809&start=30