Skip to main content

We finally recieved the computer for the new studio, and smiled as we thought, "A couple more connections and we will be up and running." Suffice it to say, things just got way more complicated.

Setup and Questions:

We are running 24 channels from the studio into channels 1-24 on a Sony DMX-r100 console. This board (capable of 48 channels) is sending those channels out to the SIU-100/100T MADI system interface unit. This has six TDIF cards capable of 48i/o channels (total), which we have run to all six TDIF banks of two MOTU 2408MK3's. These are, of course, connected to the 424 card in the computer. We originally intended to send 24 channels to/from the computer (recording in Cubase) and 24 channels just FROM the computer. These additional 24 would be saved for virtual instruments, overdubs, and effects, etc... In addition to not being able to get any of it to even resemble working, we have all six banks registering a hard-clipping digital pulse of some sort when it is all connected (maybe 60bpm, not sure) The Madi interface reports "ext. sync 44.1" but also says "no word." This seems like lying to me.

My only hypothesis thus far is that sending a digital signal through 24 channels to the computer and back into those same 24 channels just isn't going to happen the way we want it too, or at all. It would appear that we just have a massive feedback loop, but since none of the manuals overlap each other (in regards to other manufacturers) it has been exceedingly difficult to glean an answer aside from intuition and common sense as to whether we even have it all hooked up correctly. Especially when every DMX forum has disappeared off the face of the earth.

The other issue is the word clock syncing. We do not have a dedicated outboard master clock in order to sync everything on a star pattern as I have seen recommended. So once we daisy-chain all of these coax cables, what would be the best source for word-clock. We were thinking the PCI424's internal clock, but does the way-more-expensive Sony possibly have a more stable clock?

I know this may qualify as answering my own question, but is it likely that we will have to settle for 24 channels to/from the computer so that we can still sum OTB on channels 25-48? And if so, WTF is the point of having 48i/o?

Any help would be appreciated, we have our first session this weekend.

Thanks,

*post edited to remove stupid errors*

Comments

ghellquist Sat, 10/01/2005 - 09:37

Hi, seems you have a bit of work ahead to get things working.

I would think I should attack this thing methodically, one thing at a time. That is not trying to get every path to work at the same time. Start with one direction.

I would start with word-clock. When recording you would if possible want to have the internal clock in the AD-converter as clock. Generally, although a lot of sales people wants you to believe otherwise, you get less errors using the internal clock close to the AD instead of going through a lot of critical circuits.

So set the Sony to Master. Connect one end of the coaxial cable (make sure it is 75 Ohms) on the WC OUT. Connect the other end to you slaved equipment WC IN. The correct way to do this (forget the sales people again) is to have a T-connector on the WC IN connector. Now daisy-chain all these inputs with T-connectors and short WC cables. On the last unit, on the open end of the T-connector you place a 75 Ohm BNC termination resistor (when you buy that small shiny thing it says 75 on it). You could probably get things to work anyway, but it is better to use equipment the way it was designed.

Once this is in place and you can see that it works, I would connect the path from the Sony to the computer. Leave the path from computer to Sony open so far. Verify that you can record to the computer.

Finally connect the path from computer to Sony. Hopefully this will not be a problem, but if it is you know when it happens. To further understand the reason of the problem in that case, you might want to pull out the recording path. If output works then, it gives you one more clue to where the problem might be.

Best of luck

Gunnar

schizojames Tue, 10/04/2005 - 09:17

Ok, so a lot of progress has been made since the day before yesterday.

I somewhat took the advice given (thanks again) in that I ran the ref. word output (not the word sync out found on the MADI card...should it matter?) to the ref in on a video/sync card in the SIU. There is a switch to select between Ref Word and Video Sync. This card has a ref word "thru" which I ran to the first MOTU, and then ran that MOTU's out to the other MOTU's in. Now I can set the mixer, the SIU (auto slaves), the MOTUs (via software), and the Cubase project all to 48kHz. Pfew!

I am getting really excited at this point so I quickly set up a 24 mono-track project in Cubase and record a few bars of the output of my Mp3 player that I happen to have plugged in to channels 11 and 12 of the console. To my satisfaction I see a waveform appear and get more excited. Then, to get the audio back out I assign MADI inputs 1-24 to channels 25-48 on the mixer and in Cubase route 11 and 12 out of 11 and 12 which are set up up as the left and right of stereo submix 6. When I switch over to faders 25-48 and bring the volume up, I find that the recording sounds like it has been digitally mutilated. Not an occasional pop or click, and not clipped anywhere I could tell in the gain structure, it sounded like some kind of evil digital distortion unit, and I was at a loss for an answer at this point.

I'm getting to it, I promise...

The only possible cause (that I could see) for this problem was found in the i/o status page under the slot 4 MADI information box. It has a lot of info on each MADI channel that I do not understand such as DI PLL, CRC, and EMPHASIS. The one that concerns me is FS, as it is the only place in any of the equipment that still says 44.1kHz after everything else has been changed, and it is not selectable. I am not going back until tomorrow morning, and thus have been re-reading the manual and typing all this nonsense. The only other place where I have seen this "FS" is in the Title Manager page, and being at home makes it hard to see if I can change this from here. I would likely bet on it though, as I found this WHILE typing this up....that's how these things work.

As far as putting i and o on the same faders in the routing matrix I found that you can select both analog and MADI inputs to the first 24 faders, just not at the same time. So it seems that pipe dream is down the pot. Anyway, I'm gonna read Tape Op for a while to get my mind off all this digital nonsense.

So... very... close....

x

User login