This is a serious problem. In my case I bought the Waves 3 card because technically it is supposed to be able to transmit up to 128 channels at 96 Khz to record. This is a very attractive feature, because in the market there are currently not many options with these characteristics in a similar price range. Of course it is not cheap, but if it does what it promises it seems a good investment.
I bought my Waves 3 card in June 2017, mainly I use it to have assisted dsp in the dLive. Frankly, it’s not that the dLive needs external dsp, but some Waves things are interesting and attractive to me.
I must say that to date I will have made approximately 250 concerts with a dLive C3500, CDM48 system and I have never had a clock problem within the dLive system.
Multirack SoundGrid has always worked well, I have not had any serious problems, although on two or three occasional occasions I lost the synchronicity of the Waves 3 card, with the consequent problem that implies. Still to this day I do not know what was the reason for such lost whimsical.
I have recorded up to 40 tracks at 96 Khz in a Mac book pro with OS X El Capitan and up to now everything has always gone well. I have not had problems with drops, clicks or any type of digital artifact. But when I have tried large sessions the thing has changed substantially.
I recently made a recording of a live concert with a dLive C2500, CDM 48, DX168 system. The recording was 70 tracks, at 96 Khz (which incidentally, that’s the fun of having a Waves 3 card), all Cat 6 cables, new and high quality, a NetGear GS-108 switch (the advised by Waves) to do the split and two (to have redundancy in the recording) new PCs with Windows 10 Pro N, OS disks SSD and dedicated discs for recording type M.2 SSD Samsung Evo Plus (currently I do not know anything faster in both reading and writing).
That day the Waves 3 card only provided digital split for recording, there was no assisted dsp processing for the dLive with Multirack.
In the SoundGrid Studio settings, the WSG buffer adjusted to the maximum, this is 192 samples and the buffer driver adjusted to the maximum, 1088.
It is simply impossible. (If someone has succeeded, please explain how they have achieved it).
The result is that recording does record, but there are drops (they are like micro cuts in the signal) multiplied in all the tracks in a totally random way without adhering to any type of concrete pattern. It is an absolute failure and totally inexcusable.
I do not know what the causes of all this may be, but the Waves 3 card should not lose the sync so lightly. It is not a product to pass the time.
Some of you included the technical staff of Allen & Heath can tell me how to explain to the client that after hiring a recording service with 70 tracks in high resolution (96 Khz, 24 bits) including a backup system, the result has been an authentic failure and it has not worked ???
Sincerely, it makes you want to catch a plane to Laos…!
My main indignation is that in a product of this category should not have any problem with the transmission of data at a massive level because otherwise it loses all its value and effectiveness, plus it is not possible to deal with a multitrack recording service in high resolution (96 Khz, 24 bits) with guarantees of success.
I got in touch with Waves’ support (I must say that they are very well served and they are in for the job), but we have not yet managed to find out where the problem lies. I have been suggested to make a list of several tests, for which I need to almost replicate the situation. Currently, due to my work, I do not have time to do tests and experiments. It will take some time until I can do it.