I am using an HDSPe AIO Pro on Windows 11 with the current HDSPe driver.

I need a native 5.1 WDM Speaker endpoint for gaming, using the standard channel layout:
FL / FR / FC / LFE / SL / SR.

Currently, the Windows Speaker Configuration for the AIO Pro only exposes 7.1, not 5.1. I have also tried changing the Windows speaker configuration/metadata externally, but the driver continues to expose the endpoint as 7.1.

I found this older RME forum discussion:
https://forum.rme-audio.de/viewtopic.php?id=29777

In that thread, Matthias Carstens mentioned that the driver could use a fixed preselection, resulting in either 7.1 or 5.1 fixed (plus stereo).

Would it be possible to provide either:

a hidden/registry option for fixed 5.1,
a test HDSPe driver with fixed 5.1 support, or
an official driver option allowing the AIO Pro WDM Speaker endpoint to expose native 5.1?

I specifically need the game/application to see a true 5.1 endpoint. A 7.1 endpoint with software downmixing is not suitable because game engines render side and rear speaker positions differently.

I would be very happy to test a beta/test driver if one is available.

Thank you!

The problem may not necessarily be the RME card/driver, but hard to say exactly what it is so I'm approaching from different fronts.

I have two computers using HDSPe PCIe cards in Logic 12.3, and only one of them (mac mini M2) seems to be experiencing this issue.

The other computer is an m3 Mac Studio, running HDSPe MADI.  The mac mini m2 receives the MADI into Logic 12.3 (buffer of 64) and outputs via SPDIF from RAYDAT to monitors.  This has worked flawlessly for years on both this setup and older setups, until the logic 12.3 update.

After I updated to Logic 12.3 (in OS26.6)and experienced this issue of distortion, I noticed my RME driver was out of date by one version.  Updating the driver had no effect.

And while it may seem like updating Logic is the most likely cause, it didn't seem to affect my Mac Studio at all.  Restarting Logic on the mac mini fixes the issue temporarily (until it returns).

My HDSPe cards are running in Sonnet Express enclosures purchased at the same time, and these receive power from a Black Lion PG-2 power conditioner, they are clocked by an external wordclock (internally connected between RayDAT and MADI on the m2).

The problematic effect sounds like a bad clocking error, digital distortion to various degrees of awfulness.

My previous configuration was the MADI card with an SSL Alpha Link AX (ADAT) and using some older ADAT > Analogue cards was able to take advantage or more in/outs.

I had purchased a Alpha Link Live-R in hopes of using it with the AX and MADI through for more in/outs...
but alas I quickly realized it does not work that way...

The SSL AX developed a clocking issue...(I will try to change the crystals at some point to see if that helps) so I recently picked up a 2nd Live-R and figured I could run AES in/out from Live-R #1 to #2 and get 48 i/o and use some outs for custom HP mixes from Total Mix.

Well I finally got around to getting things hooked up, aaaaaannd... the RME DSP was not "SYNC'ing"...only "LOCK". After changing MADI cables, clocking via different methods, different Live-R's....it would not sync once ProTools or Mixbuss started up.... very frustrating.

On a whim, I figured maybe both Live-R's needed to run in a loop (similar to Digico consoles), I connected the RME HSDP out to Input on Live-R #1, then a single MADI cable from Out #1 to Input on unit 2, and Out on unit #2 back to RME HDSP....

Now SYNC when PT is running!!

I guess this is a clocking glitch between the units...

Hope this helps anyone with multiple MADI units.

To set the scene: I have an RME HDSPe MADI FX audio card installed on a Windows PC. I have connected an Antelope Orion 32+ and a Ferrofish A32 to the two optical MADI ports (inputs/outputs) respectively, giving me a total of 64 I/O channels.

My question is: I would like to connect the RME card's third MADI port—which is coaxial—to another Orion 32+ unit that uses optical MADI. I spent a whole day searching Google for an alternative to the SSL X-Logic DeltaLink MADI OptiCoax adapter, so I would like to ask if you know of any commercially available adapter that costs less than the SSL model.
I hope I have made myself clear.

Thank you for your help.
Sorry for my English...

Gian Marco

AIO with breakout XLR Cable and
AES Rev.1 with Tascam DB25 XLR breakout cable

I do external vocal processing in a MacMini M2 LiveProfessor VST plugin Host with the AIO, inserted on 2 different channels of an Avid S6L Mixing Console :

RME AES Out 1 -> S6L AES In 1 (ins. Return)
RME AES Out 2 -> S6L AES In2 (ins. Return)
S6L AES Out 1 (Ins. Send) -> RME AES In 1
S6L AES Out 2 (Ins. Send) -> RME AES In 2

The VST Host MacMini project is running @48K / RME AIO @48k
The Avid S6L is running @96khz
Everything is working flawlessly.

But if i swap from HDSPe AIO card to HDSPe AES card w/ db25 to try the same project i got random ugly audio glitches/artefacts/distortion on both inputs/outputs.

I tried to change the "double wire" stuff, i tried to change XLR in/out AND VST Host in/out but i'm really struggling to understand where goes the signal in which XLR in this type of configuration and how to get rid of the glitches.

I was thinking that the SRC would be as OK with the AES card than the AIO but i may be wrong or incorrect.

Any help would be much appreciated

I own a HDSPe MADI Fx PCIe card that is installed in a Mac Pro 5.1 with Mojave OS and it worked like a charm but i need to update my OS to newer that i can update my Protools and all audio software.
So i decided to install SEQUOIA Os using OpenCore Legacy patcher on my Mac Pro 5.1.

I worked like a charm then i tried to install the HDSPe MADI Fx PCIe Card, so i first update the card firmware like describe on this page https://rme-audio.de/driverkit-info-hdspe.html when i was on my MOJAVE Os.
Then i install the new driver for DriverKit on SEQUOIA Os,  also i set the security system of Sequoia to be able to accept the RME GmBh, so i follow all instruction from this page too https://rme-audio.de/login-items-extens … ia-en.html

Now my HDSPe MADI Fx card is well recognized but sound on OUTPUT or INPUT in Total Mix (that is configured as it was on my Mojave Os. I set a new protools session with a tone generator on all Madi OUTPUT but no sound or vu meter movement in Total MIX

And now i'm stuck because the HDSPe MADI Fx card don't work on SEQUOIA and can't work anymore on MOJAVE because of the Firmware update.

-Is there a way to downgrade the HDSPe MADI Fx Card firmware to be again Mojave compatible ?
-Is it a issue using Opencore legacy patcher on Macpro 5.1 with Sequoia with a updated HDSPe MADI Fx Card ?

Your help is much appreciated because i have to fix it as soon as possible cause lot of recording session is coming

Signal Chain
JRiver Media Center (bitstream or PCM, WASAPI HDMI out) → Trinnov AltitudeCI-10 (HDMI in, decodes bitstream and passes PCM to 10 channel AES out) → RME HDSPe AES (AES in, clocked to incoming signal, ASIO to AES outputs) → 6x Xilica XD4080 4in8out DSP processors → Benchmark AHB2 amplifiers → speakers and 8x JL Audio Fathom subwoofers. 29 total processed channels, with 5 bed speakers actively tri-amplified.

The Behavior
The RME Hammerfall DSP Settings window exposes sample rate and clock source as ComboBox controls, but the RME driver conservatively sets its sample rate to half the incoming AES rate rather than matching it directly. This is after I set Clock Source to follow "AES 1 In"... So for example, a 88.2kHz on AES 1 In from the Trinnov AES outputs results in the Hammerfall DSP running at 44.1kHz, and 192kHz on AES 1 In results in 96kHz for the clock source (despite the inputs at the bottom of the window syncing to the correct rate from the Trinnov). This means the RME sample rate must be explicitly set to match the correct output rate for each piece of content as it will not auto-follow the incoming AES clock. For now, I had to write a AHK script to do this, and it works, but I'm curious if this is a known bug.

Question for RME
Is this expected behavior, and if so is there a setting or API that would allow the HDSPe AES32 to automatically follow the incoming AES clock rate without requiring manual intervention each time the sample rate changes?

Previously, it clocked fine to MADI 1, but now it won't reliably lock to anything — it bounces between 'Lock' and 'Sync'. Audio passes, but there are random clicks and pops across all 96 channels at different times.

As part of troubleshooting, I updated the card firmware, drivers, and TotalMix, but the issue persists. Nothing else in the chain has changed.

Does anyone have ideas?

How this would Handle the HDCP Copyprotection at all

If you use an
Hyperion DPR-16 that has AES EBU Outs
or the Cheepest option

Nuprime-X H16-AES Only AES Outs aviable!

How this Devices feed the Stream to an AES DAC what happen to that Copypotection???

Is that Possibel to use that Card in between at all???

I guess all AES Out streams become PCM Format to an DAC right? And SACD may be DSD or DoP ?

But what happen to HDCP or Copyprotection than???

1.Since users might not have Windows sounds disabled yet you'll get a full blast notification sound - maybe related to https://forum.rme-audio.de/viewtopic.php?id=43300 - not sure if it was a success or error sound

2. You turn down the main volume (since not checked the default connection) and start playback of any fine

Result: Music playing on full blast since default main out is connected to "Phones" instead of "Analog 1/2"

Expected: Default main out is connected to "Analog 1/2" (which I think is the most used output) - with this I wouldn't have get "blasted"

Obviously there are quite a few variables I need to trouble shoot here, but I wanted to try reducing the number of inputs/outputs connecting to Cubase anyway... just because I only need 32 at the moment.

Not saying it's related, but still... just want to run through variables/tests.

Is there a reason I can only reduce down to 128, and not 64 or 32?

In my setup, the main and backup signals from two desks are routed to MADI A-1 and MADI B-2. It would be great if these could be remotely switched to inputs 1–64 via OSC.

Thank you in advance and best regards,
sewi