• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

How much have AD and DA chips advanced in RME units from Fireface 800 to current lineup?

GearNostalgia

Member
Joined
Dec 18, 2022
Messages
66
Likes
20
I am not good at comparing chips and specs. But in general terms it seems that equipment jumps drastically from time to time. How much better have RMEs converters progressed from Fireface 800, to Fireface 802 and to todays lineup with Fireface UFX III? Is there a clear audible advance worth and uppgrade?
 
Last edited:
RME is one of a few for whom one can trust their specs. Given that I'd say there are a few db improvement in general conversion capability. Not enough you would hear a clear audible improvement I don't think. Might hear none I would expect. What you would gain is something that isn't Firewire based. You can see basic specs at the link below.

 
I assume they have changed chips and from reading in other threads about DAC from Topping and others it seems like there have been a lot of improvemnts from one generation to another. So I am curious if RME line have improved in a similar fashion.
 
The ADC side hasn't seen as large an improvement as the DAC side. The Fireface 800 was pretty far up the ladder of what was possible then. That ladder hasn't extended all that much for most devices. The RME ADI 2 ADC which has line in only is about the best ADC short of a Cosmos. It is only a few db better. Plenty of not cheap interfaces aren't better than the Fireface 800.

The Fireface 800 listed SINAD of -104 db for the ADC with -100 db for the DAC. The UFX III has -110 db for the ADC and -110 db for the DAC. Now measures by third parties might be a touch different, but again this only claims a 6 db improvement on the ADC side. The ADI-2 ADC claims -116 db while Amir measured -117 db. Which as far as I know tops his list. The Babyface came in at -108 db on the ADC side. The Topping ADC is an overachiever getting -110 db on the ADC side. A Motu Ultra light mk5 is -114 db on the ADC side. Especially once microphone preamps are involved they usually swamp the ADC anyway.
 
I actually have an 802 now, considering if it is worth getting an UFX III or not.
 
The chips don't tell the whole story... The final product can be worse, or in some cases possibly tweaked to be better. Where I work we make a couple of products with DACs & ADCs and we use software calibration to improve the performance (and so we can meet the specs of our product). It's not an audio application, I think they are only 12-bits, and they run much slower than audio sample rates so there's plenty of time for the software to apply a correction. I don't think that's practical (maybe not possible) to calibrate/correct audio DACs & ADCs because with the higher precision the limitation is usually noise/instability and inconsistent errors can't be calibrated-out. But there still may be "tweaks" to improve performance.

There are part-to-part variations/tolerances and different ways of measuring, especially when it comes to noise.

If you are recording with microphones, the 1st concern is usually acoustic noise. Preamp noise is 2nd.
The actual DACs & ADCs are usually better than human hearing ("transparent"). If you can find independent-comparable measurements, it's probably worth checking the noise measurement on the preamp inputs. But of course there will be unit-to-unit variations so unless one unit measures significanlty better than another it may not be meaninful.
 
RME has an active forum. https://forum.rme-audio.de/

I believe I saw there a detailed rundown of the chips in some of their products.
 
I use my interface to record instruments using line inputs 99% of the time. I use the mixed mic/line jacks also to get more inputs, but I am doing electronica/symphonic music so recording with mic is rare thing. I have been at RME-forum from time to time as well, but I am unsure if it is a fan biased forum or not. The measurements here from Amir seems more fair to me.
 
In addition to SINAD/SQ type specs, modern converters also have less latency. Not going to make or break any session, but lower latency is generally preferred - especially when it's "Free" at the Converter (and not a CPU/Compute penalty like running 32-sample ASIO buffers can impose). Bigtime RME fanboi over here (see my sig) - because they have proven themselves time and time again. Solid all-around. That said - I'd also agree looking to the 802 or newer is recommended as the FF800 is basically a "Classic" at this point (good luck with FW on a modern PC/Mac!)...
 
In addition to SINAD/SQ type specs, modern converters also have less latency. Not going to make or break any session, but lower latency is generally preferred - especially when it's "Free" at the Converter (and not a CPU/Compute penalty like running 32-sample ASIO buffers can impose). Bigtime RME fanboi over here (see my sig) - because they have proven themselves time and time again. Solid all-around. That said - I'd also agree looking to the 802 or newer is recommended as the FF800 is basically a "Classic" at this point (good luck with FW on a modern PC/Mac!)...

I think I was not clear. I was asking for fact about how much performance have increased from 800 to 802 and to current UFX III. I am on 802 now and wonder if it is worth trading up for current line up or not.
 
So you have USB. There are less than 10 db difference vs the newest one. I see no reason to upgrade. What would you expect to gain? I didn't delve into whether latency is less on the newer UFXIII, but RME generally has been pretty low latency anyway even in the 802.
 
Isn't the biggest thing in UFX III MADI? If you have all the i/o you need in 802 I'm not sure what you're looking for? The performance will be great with both and cited specs are accurate.
 
Isn't the biggest thing in UFX III MADI?
MADI, digitally-controlled preamps, digitally-controlled main and HP outputs (that allow for precise setting and saving the values in TotalMix snapshots), DUrec, easier standalone operation due to fully-featured onboard controls with display, etc.

In short, a decision to upgrade to UFX III should be made based on its features, not on converter specs.
 
I need more line inputs, my 802 is maxed out (adat inputs filled as well) so I have started to investigate upgrade options. One way is to just get another usb/thundebolt device and merge them with Aggregate device in max. Or sell off what I have and go over to some madi/dante unit and have more expandability. Read and heard about the Apollo X latest gen sounding clearly better than previous the gen before so I am an curious about it there is also a clear difference in RME line.
 
Ah, ok, then just get UFX III. It seems to have it all and measured performance is as good as it gets.

Edit: something sounding clearly better implies there was something wrong with the previous gen. There is nothing wrong with RME so what's there to gain? Also, I'd be cautious with any claims of "better" sound with these things anyway.
 
"Sounding clearly better" is not likely unless going back 20 years. Being clearly more capable is absolutely likely - and is what you should be after IMO. That's what gets work done efficiently and effectively.

UFX III is a solid pick that you will be unlikely to exhaust. It has oodles of DSP compute for TotalMixFX (and Room EQ's now in TM). MADI is awesome IMO, but has a "cost of entry" getting something to connect to it (ADAT + SPDIF AD/DA's are a dime a dozen, MADI AD/DA's get much more costly or something like the ADI648 MADI-to-ADAT converter to allow connections to ADAT).
 
Can not recall where I heard it, but some Youtuber mentioned that DSP power is maxed out in the 802/ufx range and that is why it can not handle more or better effects and why speaker calibration is a bit limited. I am pretty ok with not having more effects in my soundcard so that does not matter much to me now, but it may matter to others.
 
Back
Top Bottom