1 Topic by lastninjagaijin (edited by lastninjagaijin 2026-01-31 20:24:33)

  • Registered: 2026-01-31
  • Posts: 3

Topic: RME Hammerfall 9652 Digi and Alesis Ai-3 ADAT strangeness

Hi, I really need some help.
I'm a novice at sound cards so I never really tested out this card I got a long time ago.
But after doing a recording of a DJ mix yesterday, that I needed to re-record a section for an splice in, I noticed that using Autosync with my Alesis Ai-3 ADAT is producing timing errors. For instance instead of 176bpm, the recording might be 176.2, or 177.5, or on ADAT3 input on my sound card, up to 200bpm.. I have swapped cables and am using new cables now, still same issues.

Also, it always says 44.1kHz frequency for the sync when autosync'd. I'm not sure if this is correct, I'm trying to record at 48kHz through the sound card.

If I switch to master sync, the timing is fine, and 48kHz frequency is shown in sync.

However another problem pops up.

I have a been recording simultaneously off my mixer into both the RME sound card, as well as my Rane SL3 interface. Every recording through the RME is too bright. I'm recording an electronic track, then comparing the recording to the original track, and is way brighter than it should be. Doesn't matter what setting - 44.1 - 16, 44.1 - 24, 48 - 16, 48 - 24. Have tried 32 bit too. I've tried all three 9652 ADAT1-3 inputs.

I'm recording with Steingberg Wavelab 6.11.

I can post spek spectrals and audio clips if that helps (spectrals are easiest to notice if you flip back and forth between them to see the RME recordings are pushed up in the frequency range).

Does anyone have any idea what's going on?

Also listening to the recordings now, I'm also noticing more noise floor than I'd expect, but maybe that's due to recording at a low volume, through my analog Xone 92 DJ mixer. I thought I recorded at around -16dB though, it shouldn't have that much noise floor I'm assuming.

Here are some audio clips to A/B for anyone curious what I'm talking about: https://drive.google.com/drive/folders/ … sp=sharing

I am using a balanced 2xTRS output from my Allen & Heath Xone 92 mix 2 output, running into Alesis Ai-3 ADAT, optical out to RME Hammerfall 9652 Digi.
I am also using the RCA record output to run into my Rane SL3 and recording on my laptop, and the recording is identical in tone to the source material. So it's not mixer colouration or anything like that. There's no EQ or other setting applied on the mixer.

2 Reply by waedi (edited by waedi 2026-02-01 06:31:49)

  • waedi
  • DAW Mastermind
  • Offline
  • From: Swiss Zürich
  • Registered: 2020-12-08
  • Posts: 7,376

Re: RME Hammerfall 9652 Digi and Alesis Ai-3 ADAT strangeness

Hello Lastninjagaijin

I want download all your tracks for loading into Reaper and compare.

I downloaded all tracks.
What a mess.
4 of 5 tracks have the same file name.
I had to delete them and redownload one after one and give a number to the name to have this under control.
Then imported into Reaper.
What a mess.
The source-file and recording 01 look the same, ok, but the rest is complete different.
this is not valid for a comparison.

https://i.ibb.co/BHT6vNtb/adat.png

See 01 and source are looking identical, this tells me the soundcard works perfectly fine.
The other recordings are different because you did different Dj mix on the DJ-mixer.

Recording thru digital input can not make the recording brighter or more bass-boom or whatsoever.
signal is coming or not coming.
Signal is what you deliver to the interface, it records exactly that, nothing else.

Do you use 2 toslink cables to the DJ-mixer ?

If you have only one cable from DJ-mixer to the interface, then the clock-source setting shall be clock-mode external optical (slave), that is also how I would connect :
1 toslink cable from DJ-mixer output to Babyface input.
In the Fireface USB settings clock-mode external optical input (slave)
I would let out the Alesis.
If the DJ-mixer has analog outputs only then connect these to the Babyface inputs 3/4 or 1/2
In this case : In the Fireface USB settings clock-mode internal  (master)

Record the Dj-mixer output in any DAW

M1-Tahoe, Madiface Pro, Digiface USB, Babyface silver and blue

3 Reply by lastninjagaijin (edited by lastninjagaijin 2026-02-07 12:55:57)

  • Registered: 2026-01-31
  • Posts: 3

Re: RME Hammerfall 9652 Digi and Alesis Ai-3 ADAT strangeness

lastninjagaijin wrote:

Hi, I really need some help.
I'm a novice at sound cards so I never really tested out this card I got a long time ago.
But after doing a recording of a DJ mix yesterday, that I needed to re-record a section for an splice in, I noticed that using Autosync with my Alesis Ai-3 ADAT is producing timing errors. For instance instead of 176bpm, the recording might be 176.2, or 177.5, or on ADAT3 input on my sound card, up to 200bpm.. I have swapped cables and am using new cables now, still same issues.

Also, it always says 44.1kHz frequency for the sync when autosync'd. I'm not sure if this is correct, I'm trying to record at 48kHz through the sound card.

If I switch to master sync, the timing is fine, and 48kHz frequency is shown in sync.

However another problem pops up.

I have a been recording simultaneously off my mixer into both the RME sound card, as well as my Rane SL3 interface. Every recording through the RME is too bright. I'm recording an electronic track, then comparing the recording to the original track, and is way brighter than it should be. Doesn't matter what setting - 44.1 - 16, 44.1 - 24, 48 - 16, 48 - 24. Have tried 32 bit too. I've tried all three 9652 ADAT1-3 inputs.

I'm recording with Steingberg Wavelab 6.11.

I can post spek spectrals and audio clips if that helps (spectrals are easiest to notice if you flip back and forth between them to see the RME recordings are pushed up in the frequency range).

Does anyone have any idea what's going on?

Also listening to the recordings now, I'm also noticing more noise floor than I'd expect, but maybe that's due to recording at a low volume, through my analog Xone 92 DJ mixer. I thought I recorded at around -16dB though, it shouldn't have that much noise floor I'm assuming.

Here are some audio clips to A/B for anyone curious what I'm talking about: https://drive.google.com/drive/folders/ … sp=sharing

I am using a balanced 2xTRS output from my Allen & Heath Xone 92 mix 2 output, running into Alesis Ai-3 ADAT, optical out to RME Hammerfall 9652 Digi.
I am also using the RCA record output to run into my Rane SL3 and recording on my laptop, and the recording is identical in tone to the source material. So it's not mixer colouration or anything like that. There's no EQ or other setting applied on the mixer.

Hi, firstly thank you for your reply. I appreciate any help with this matter. I'm sorry for the mess of files - the file names have different names - they are 44/16, 44/24, 48/16 and 48/24 recordings as shown in the file names. But yes, I normalised each file and recorded at different parts of the track, and was also blending the track, so it was a poor way to try to show my issue. I apologise for that. I was originally trying different recordings frequencies + bitrates to see if that had an impact on the artificially bright recordings, it didn't. I've redone my test in a way that should be very clear to be able to hear (if not see with analysis tools) that there is added brightness somehow creeping in the chain.

It is not an issue with my mixer I am sure - I don't get this added brightness when I use the mix 1 balanced output for my monitors. I am using mix 2 balanced output to record to the RME, they are basically the same output circuitry apart from the connector type (mix 1 is XLR, mix 2 is TRS). And I am using the RCA record output to run to the SL3 sound card.

Please check my new uploads, they are gain matched: 1. source track 2. SL3 recording 3. Alesis Ai-3 + RME Hammerfall 9652 Digi recording.

https://drive.google.com/drive/folders/ … drive_link

I have no toslink cables from the mixer, it's an Allen & Heath Xone 92 which is an analog mixer.

I don't have a Babyface, I'm using an RME Hammerfall 9652 Digi and Alesis Ai-3 ADAT as specified in the topic title.

Here is my lineage for the test recording:

* Dell laptop playing WEB FLAC file in Serato DJ Pro v2.57
* Oyaide D+ USB A to B cable from laptop to Rane SL3 sound card
* RCA cable from Rane SL3 to Allen & Heath Xone 92 line input channel, no EQ or gain applied

Output 1: RCA recording output of mixer into Rane SL3 channel 3, recorded on same laptop as playback

Output 2: Short (1m) Balanced/TRS cables to Alesis Ai-3 ADAT +4dB input
Toslink cable output with new toslink cable, to RME Hammerfall 9652 Digi ADAT2 input
RME settings: Buffer 256 (6ms), Clock mode Master, Pref Sync Ref ADAT2 In. SyncCheck ADAT2 In = Sync. Sync Ref Inp. Master Freq. 48kHz.
Recorded with Wavelab 6.11 - 48kHz 24bit.

Gain matched files by peak - obviously source file is a little bit louder, but still you can clearly hear the difference between the SL3 and RME recordings - the RME is artificially adding brightness. Not sure how this is happening.

  • waedi
  • DAW Mastermind
  • Offline
  • From: Swiss Zürich
  • Registered: 2020-12-08
  • Posts: 7,376

Re: RME Hammerfall 9652 Digi and Alesis Ai-3 ADAT strangeness

Track 3 is shorter, it was played faster this could sound high pitched and you think its added highs ?

Frequency spectrum looks all the same, no added brightness visible to me.

https://i.ibb.co/21m4XZR4/shorter.png

M1-Tahoe, Madiface Pro, Digiface USB, Babyface silver and blue

5 Reply by lastninjagaijin (edited by lastninjagaijin 2026-02-07 17:45:26)

  • Registered: 2026-01-31
  • Posts: 3

Re: RME Hammerfall 9652 Digi and Alesis Ai-3 ADAT strangeness

waedi wrote:

Track 3 is shorter, it was played faster this could sound high pitched and you think its added highs ?

Frequency spectrum looks all the same, no added brightness visible to me.

You're right about the brightness. It's strange, I was playing the tracks back through my Asus X570 TUF GAMING PLUS mobo onboard sound, and the treble was really apparent. But now playing through my Focusrite Scarlet 2i2 3rd Gen, it is true that there is no added brightness audible. This is really confounding me.

Still there is the issue with timing.. I was having huge timing/sync issues when using autosync setting - this 180bpm track was being recorded at anywhere from 0.2bpm to 20bpm difference in speed, always being pitched up.

For this newest test, I recorded into the RME's ADAT2 input - I had forgotten that during my last test, ADAT3 had ended up having the lowest timing issues (but I had thought they were all similar, I thought that my bpm analyser saying 179.98 was margin of error but maybe not now in retrospect). This current RME recording on ADAT2 is also showing up as 179.98 bpm, compared to 180.00 for the SL3 recording. I can try the test again with ADAT3 incase ADAT1 + 2 are just dying. But do you think this could be an issue with the whole sound card, or somehow the Alesis Ai-3 ADAT? The sound quality of the RME is noticeably better but I can't handle these timing issues.

I have another set of toslink cables I can test with also but I don't think it's the new toslink cables causing this issue.

It's also very bizarre that my motherboard onboard sound card was adding treble to these recordings - it makes no sense. I've never experienced this issue with my motherboard audio before. But you are right, there's no added treble when I listen through my Focusrite. This is driving me a little insane. The spek spectrums for the original recordings where I noticed the timing issues did also have frequencies pushed up in the RME recording (both recordings done at 48kHz). That was with autosync set on the RME settings though - and autosync was displaying 44.1kHz on the driver settings - compared to 48kHz when I've tested with the master sync since. So maybe it was only inputting 44.1kHz audio, yet recording at 48kHz - could this somehow 'push up' the frequencies in the recording? I wouldn't have thought so.

So it seems like somehow there is at least timing and pitch issues. No idea how my mobo onboard sound card was 'flavouring' these files though. I'll have to test that onboard sound again just so I know I'm not going crazy. I am listening to these two files from my recording last week, side by side, through the Focusrite, and as you say there's no 'super harsh treble' like it was when listening through my motherboard onboard audio. So it looks like the 'harsh treble' is some oddity I've never encountered before on my onboard motherboard audio (why only these recordings, and no other files??) - but the timing issues are definitely something wrong with either my RME sound card, and/or Alesis Ai-3 ADAT.

I'm guessing my next steps to try to figure out what device is causing the issue is to a) try recording again with ADAT3 input, b) I have a cheap crappy little RCA to toslink ADC - try running that direct into my RME toslink input/s instead of the ADAT. I also could swap my TRS cables running into the ADAT as well as trying a third set of toslink cables, but I don't think that either of those will do much, but may as well give it a go while I'm messing around. Is this looking like a jitter issue?

Here are speks for the original recording.

If you save these two images and flick back and forth, you can see the RME recording reaches higher frequencies, and some of the colour bands in the spectral are higher in the RME than in the SL3 recording. I'm not very good at analysing spectrals but it appears to me that the frequencies are pitched up visibly, 100-200Hz I'd guess? Presume this is from the timing issue. Not enough to be 'super harsh' treble like what I heard at first though, which is apparent with playback through my Focusrite.

RME: i.imgur.com/dqjc54e.jpeg
https://i.imgur.com/dqjc54e.jpeg

SL3: i.imgur.com/AJ3Mkbd.jpeg
https://i.imgur.com/AJ3Mkbd.jpeg

6 Reply by clive022lloyd (edited by clive022lloyd 2026-03-27 06:02:50)

  • Registered: 2026-03-25
  • Posts: 1

Re: RME Hammerfall 9652 Digi and Alesis Ai-3 ADAT strangeness

Hello,
When you use "Autosync," the RME is trying to follow the Alesis, but they aren't talking correctly, which is why your BPM is warping. Since setting the RME to Master fixes the timing, keep it that way! Just make sure you have an ADAT cable running from the RME Out back into the Alesis In so the Alesis can follow the RME’s 48kHz clock.

As for the brightness and noise, that's likely a level mismatch. The Xone 92 puts out a very hot professional signal. If the Alesis is expecting a lower consumer level, it can mess with the high frequencies genesis fs card services and leave you with a high noise floor.

Keep the RME as Master at 48kHz.

Turn up the "Mix 2" knob on your Xone to send a stronger signal; this should help bury that noise floor.

Double-check that TotalMix doesn't have any hidden gain or EQ active on those inputs.