• Registered: 2019-02-14
  • Posts: 148

Topic: 768kHz, 192kHz & crippled devices

Howdy folks.

So I have a few topics that should be addressed and publicly discussed.

768kHz & 192kHz and crippled RME devices, meaning those RME products, that have more I/O than what those resolutions can handle (with the crippled I/O options currently available)

One example out of countless:

UCX II = can't run 768kHz, can only run 2 channels at 192Khz per port, meaning not all I/O can be utilized at the same time, at that resolution, to connect to other RME devices, a computer or other. Same story with UFX II - not all analog I/O can be used simultaneously to and from other devices at max resolution due to the ports removed from previous versions.


Likewise, the ADI-2 devices can run 768kHz through USB, but not through the digital I/O, which is capped at 192kHz @ only 2 channels - this renders the 768kHz feature obsolete for my use case, as I need to rely on ultra low latency ultra low CPU so I cannot use USB at all ! Some of us still use the PCI-E devices for our tracking, since we actually need the low latency & some of our systems actually need the low cpu usage that PCI-E provides at small buffer settings, as opposed to USB, which is always a CPU hog at low latencies, comparatively speaking (this is not part of the discussion, it is simply a fact, even with RME self-written drivers - no point arguing about facts)

Now, on to the discussion:

Which RME devices are truly & fully NOT crippled?

Meaning fully capable of running all on board I/O at max resolution (192kHz) from one device to another, ie from analog to digital and vice versa (without using USB, obviously) ?

Can any RME product run 768kHz outside of using USB ? If so, which & how ?

Might there be an updated version of the MADI FX in the pipeline?

There was an updated version of the AIO some years ago, and now there is also a smaller MADI form factor, so maybe it's time for a fresh MADI FX PCI-e card with the latest features such as the "ADI-2" EQ & SteadyClock FS?
I'm sure that would be a winner on the Pro market, and possible room for 3x MADI on one card due to the smaller form factor ?!


Food for though - you're welcome smile

  • mkok
  • DAW Mastermind
  • Offline
  • Registered: 2020-11-04
  • Posts: 632

Re: 768kHz, 192kHz & crippled devices

Just read the manuals. They tell you exactly what is supported. It’s not just RME. What’s the use case for 768khz? I can’t see many using it.

Babyface Pro Fs, Behringer ADA8200, win 10/11 PCs, Cubase/Wavelab, Adam A7X monitors.

3 Reply by ramses (edited by ramses 2024-05-06 20:26:18)

  • ramses
  • DAW Mastermind
  • Online
  • From: UTC+1
  • Registered: 2012-10-14
  • Posts: 13,182

Re: 768kHz, 192kHz & crippled devices

torbenscharling wrote:

Howdy folks.

So I have a few topics that should be addressed and publicly discussed.

768kHz & 192kHz and crippled RME devices, meaning those RME products, that have more I/O than what those resolutions can handle (with the crippled I/O options currently available)

One example out of countless:

UCX II = can't run 768kHz, can only run 2 channels at 192Khz per port, meaning not all I/O can be utilized at the same time, at that resolution, to connect to other RME devices, a computer or other. Same story with UFX II - not all analog I/O can be used simultaneously to and from other devices at max resolution due to the ports removed from previous versions.


Likewise, the ADI-2 devices can run 768kHz through USB, but not through the digital I/O, which is capped at 192kHz @ only 2 channels - this renders the 768kHz feature obsolete for my use case, as I need to rely on ultra low latency ultra low CPU so I cannot use USB at all ! Some of us still use the PCI-E devices for our tracking, since we actually need the low latency & some of our systems actually need the low cpu usage that PCI-E provides at small buffer settings, as opposed to USB, which is always a CPU hog at low latencies, comparatively speaking (this is not part of the discussion, it is simply a fact, even with RME self-written drivers - no point arguing about facts)

Now, on to the discussion:

Which RME devices are truly & fully NOT crippled?

Meaning fully capable of running all on board I/O at max resolution (192kHz) from one device to another, ie from analog to digital and vice versa (without using USB, obviously) ?

Can any RME product run 768kHz outside of using USB ? If so, which & how ?

Might there be an updated version of the MADI FX in the pipeline?

There was an updated version of the AIO some years ago, and now there is also a smaller MADI form factor, so maybe it's time for a fresh MADI FX PCI-e card with the latest features such as the "ADI-2" EQ & SteadyClock FS?
I'm sure that would be a winner on the Pro market, and possible room for 3x MADI on one card due to the smaller form factor ?!


Food for though - you're welcome smile

Only to a few points …

As you can read, the ADI-2 Pro series of devices is also advertised as AD/DA front end for measurement systems with up to 768 kHz sampling rate. Such high sample rates for mixing, mastering are pointless.

For recording, mixing and mastering, double speed is already enough because the AD/DA filter work then already linear in the audible range up to 20 kHz. If you think you need 192 kHz, fair enough, possible. But you're not doing yourself any favors with every system because a lot more data has to be processed in time. Some systems can so, some not. It also depends on the number of channels.

USB drivers and CPU Hog. In my world RME USB driver work very efficient, see e.g. this example:
https://forum.rme-audio.de/viewtopic.ph … 92#p210092
and
https://www.tonstudio-forum.de/blog/Ent … cks-de-en/

Before speaking of crippled systems, one should carefully consider whether such a statement is justified.
Sorry, but your post doesn't give me that impression at all.

As mkok already mentioned. All vendors are compliant to the same usual standards in terms of digital transport of audio, no matter whether we speak about ADAT, SPDIF, AES, MADI.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14
  • Registered: 2016-06-14
  • Posts: 1,142

Re: 768kHz, 192kHz & crippled devices

torbenscharling wrote:

UCX II = can't run 768kHz, can only run 2 channels at 192Khz per port, meaning not all I/O can be utilized at the same time, at that resolution, to connect to other RME devices, a computer or other. Same story with UFX II - not all analog I/O can be used simultaneously to and from other devices at max resolution due to the ports removed from previous versions.

I don’t understand:
Of course these devices have some additional standalone functionality, but mainly they are audio interfaces for the use on a computer. So why should they be able to convert every single analog input and send it out over their own available digital output ports @192kHz (and vice versa)? For this use there are some standalone AD/DA converters on the market.

UCX - FF 400 - Babyface pro - Digiface USB - ADI-2 (original)
Mac mini M1 - Macbook pro - iPad Air2
  • Registered: 2019-02-14
  • Posts: 148

Re: 768kHz, 192kHz & crippled devices

oli77sch wrote:
torbenscharling wrote:

UCX II = can't run 768kHz, can only run 2 channels at 192Khz per port, meaning not all I/O can be utilized at the same time, at that resolution, to connect to other RME devices, a computer or other. Same story with UFX II - not all analog I/O can be used simultaneously to and from other devices at max resolution due to the ports removed from previous versions.

I don’t understand:
Of course these devices have some additional standalone functionality, but mainly they are audio interfaces for the use on a computer. So why should they be able to convert every single analog input and send it out over their own available digital output ports @192kHz (and vice versa)? For this use there are some standalone AD/DA converters on the market.

Most of their AD/DA converters are also crippled. There you go...Why shouldn't they be able to convert every single analog input and send it over their own available digital output ports @192kHz (and vice versa)? Your guys seem to love avoiding the hard questions wink

  • Registered: 2019-02-14
  • Posts: 148

Re: 768kHz, 192kHz & crippled devices

ramses wrote:
torbenscharling wrote:

Howdy folks.

So I have a few topics that should be addressed and publicly discussed.

768kHz & 192kHz and crippled RME devices, meaning those RME products, that have more I/O than what those resolutions can handle (with the crippled I/O options currently available)

One example out of countless:

UCX II = can't run 768kHz, can only run 2 channels at 192Khz per port, meaning not all I/O can be utilized at the same time, at that resolution, to connect to other RME devices, a computer or other. Same story with UFX II - not all analog I/O can be used simultaneously to and from other devices at max resolution due to the ports removed from previous versions.


Likewise, the ADI-2 devices can run 768kHz through USB, but not through the digital I/O, which is capped at 192kHz @ only 2 channels - this renders the 768kHz feature obsolete for my use case, as I need to rely on ultra low latency ultra low CPU so I cannot use USB at all ! Some of us still use the PCI-E devices for our tracking, since we actually need the low latency & some of our systems actually need the low cpu usage that PCI-E provides at small buffer settings, as opposed to USB, which is always a CPU hog at low latencies, comparatively speaking (this is not part of the discussion, it is simply a fact, even with RME self-written drivers - no point arguing about facts)

Now, on to the discussion:

Which RME devices are truly & fully NOT crippled?

Meaning fully capable of running all on board I/O at max resolution (192kHz) from one device to another, ie from analog to digital and vice versa (without using USB, obviously) ?

Can any RME product run 768kHz outside of using USB ? If so, which & how ?

Might there be an updated version of the MADI FX in the pipeline?

There was an updated version of the AIO some years ago, and now there is also a smaller MADI form factor, so maybe it's time for a fresh MADI FX PCI-e card with the latest features such as the "ADI-2" EQ & SteadyClock FS?
I'm sure that would be a winner on the Pro market, and possible room for 3x MADI on one card due to the smaller form factor ?!


Food for though - you're welcome smile

Only to a few points …

As you can read, the ADI-2 Pro series of devices is also advertised as AD/DA front end for measurement systems with up to 768 kHz sampling rate. Such high sample rates for mixing, mastering are pointless.

For recording, mixing and mastering, double speed is already enough because the AD/DA filter work then already linear in the audible range up to 20 kHz. If you think you need 192 kHz, fair enough, possible. But you're not doing yourself any favors with every system because a lot more data has to be processed in time. Some systems can so, some not. It also depends on the number of channels.

USB drivers and CPU Hog. In my world RME USB driver work very efficient, see e.g. this example:
https://forum.rme-audio.de/viewtopic.ph … 92#p210092
and
https://www.tonstudio-forum.de/blog/Ent … cks-de-en/

Before speaking of crippled systems, one should carefully consider whether such a statement is justified.
Sorry, but your post doesn't give me that impression at all.

As mkok already mentioned. All vendors are compliant to the same usual standards in terms of digital transport of audio, no matter whether we speak about ADAT, SPDIF, AES, MADI.

This topic is not about whatever you assume people use the devices for.

This topic is not about a speculative theory around the importance of high kHz.

If you have concerns about using high kHz or whatever, you can start your own post, just click the New Post button. Next time, consider carefully if your commentary is at all needed and justified. So far you comments don't give me that impression at all.

Let's keep things on topic please.

As already explained, this is not about being compliant to the usual standards in terms of digital transport of audio.

As you may not be aware, during production, a choice is made on how many digital ports to implement in the device being produced.

The choice of which digital ports to use, and how many, determine how crippled or not said device will be in terms of "full functionality" at the highest quality setting.

It is about pinpointing exactly which RME devices are crippled, how much, why, and how to improve that in future products. Obviously we all have an interest in making sure our fellow RME customers make the right choice, make it easier for them, and you, to understand how reality works, and make other people happy with the choice they made.

This information will also spread awareness of the current flaws & bad judgements made, when introducing new iterations, that further cripple the devices (removing important ports such as Thunderbolt, not offering MADI or other options to gain "full functionality" etc.)

7 Reply by MC

  • MC
  • MC
  • Administrator
  • Offline
  • From: Germany/Thailand
  • Registered: 2007-04-06
  • Posts: 25,582

Re: 768kHz, 192kHz & crippled devices

So far the only thing crippled here are your derogatory statements in this thread. I have no clue if this is a language issue or if you have some strange agenda, but this thread serves no purpose and stops right here. If you want to discuss general hardware design outside of existing formats and interface standards (as done by any company existing right now) please do this in one of the many discussion forums available somewhere else.

Regards
Matthias Carstens
RME

MC's Website

  • From: Germany
  • Registered: 2007-04-10
  • Posts: 9,770

Re: 768kHz, 192kHz & crippled devices

torbenscharling wrote:

Likewise, the ADI-2 devices can run 768kHz through USB, but not through the digital I/O, which is capped at 192kHz @ only 2 channels - this renders the 768kHz feature obsolete for my use case, as I need to rely on ultra low latency ultra low CPU so I cannot use USB at all !

Your lack of knowledge of the simple fact that SPDIF as such is limited to 192k renders the entire point of your "discussion" moot. Your "hard questions" are not based on technical facts and can therefore be discarded as irrelevant.

Regards
Daniel Fuchs
RME

RME Support's Website