Mult-track Support for Microsound Workstations

[admin]

How will MTU implement mult-track support - This has been stated for many months at our http://mtu.com/basics/news.htm page. There are three facets of support.

[li]Microeditor Project Multi-track assembly - I believe we have close to the ultimate system for this already. I don't see any major changes in the near future.

[li]Multi-track Input Channels - We are expecting to offer an AddOn Interface Program for multi-track I/O. It will use the industry standard ASIO software interface to connect to most of the multi-track I/O cards currently available. This optional program will link into Microeditor, appearing as a new way to record and play. For record, it will show a multi-track record screen, and allow creating segments with/without group assignments. Our target is 8, 16, and 24 tracks. We don't know yet if we must pre-create the track lengths before recording. If not, the track samples will be interleaved with each other on the disc. This must be evaluated. Our first target is 8 tracks to allow Surround Sound (not to state MTU is coding for Surround Sound yet).

[li]Multi-track Output Channels - There is no easy way to instantly play multi-track out of Microeditor. The first approach we will take is an Play Groups as Tracks command. All unmuted groups will be saved to the hard drive as separate wave files, with silence padding in front to insure each file starts playing in sync with the others. There will be a delay while the files are output, then they can be played out a multi-track ASIO control program. We are expecting to make this Play program part of the Multi-track Record program. Ideally, the "output" command will be a Play command that computes and stores the wave files, then automatically plays them.

[geezer]

----I am assuming that the multitrack play/record detailed in this thread will achieve I/O via a third party audio card (a la MOTU 2408). True?

For most of the functionality I'm seeking, FILE import/export with time stamping or time alignment is more important than actual multi-stream I/O, however.

[JTatum]

I am currently using Steinberg Cubase and a MOTU 2048 to handle my tape dumps, dumping 24 tracks into Cubase and then creating segments from the .wav files. This works pretty well, but is also a pain because of two things. First is the requirement to do the dump in a second software package. This is merely a minor annoyance. Second and more serious is the issue of latency. After working with Microeditor, I have gotten used to the convenience of zero latency. Even running dual PIII CPU's and Cheetah's configured RAID 5 on a new ultra 160 controller (800MB/Sec throughput), I can only get latency down to 3 ms. This means that if I have more than 24 tracks (most of my projects are 32 or more) syncronising the second layer of tracks is a major pain.

The Tascam (and clone?) solution seems to offer a zero latency solution which would be the best way to go. On the other hand, ASIO based hardware is cheap and a great way to get in the door to truly great mastering capability, so I would like to see this supported as well.

Microeditor has always been about being free from constraints so that we can concentrate on being creative. I would like to see this philosophy extend to hardware as well.

[pottersclay]

i would like more control over the live recordings and would like to see you implement multitrack recording. This would allow me to have more control on the overall recording. Thanks.

Also...do you know where I can find a rackmountable computer case.

thanx

[JTatum]

I am implementing a new solution here. I will be using an iZ Radar 24 and connecting via ethernet. From there it is simply a matter of converting the file format. The same will hold true on the Tascam MX-2424, the Mackie HDR 2496, and the Alesis ADAT HD24.

Ethernet will make imports much simpler and less time consuming. Simply connect to the unit in question via plain old cat 5 cable and use the create segment (which would have to become "create segment from" .wav, .bwav, .aiff, .etc.) and away you go. This should also be easier and less costly to implement than drivers as well as files are easier to deal with than hardware.

[admin]

There are some very good ideas here. I'm glad to see you working with us on ideas.

Larry is close to finishing up the coding of the new msnd32.dll that controls Krystal. This is one of those hated "no income" upgrades that are required from time to time to keep the product wheels turning (and selling).

I'm expecting the next thing he will be on is to redo MicroCD as an Export CD command to probably add into Microeditor's File menu. It will require our Microstudio software to write to the CD. MicroCD and Microstudio both have code to control the CDR drive, and we can't afford to maintain both. It makes far more sense to make MicroCD a Microeditor optional command that exports files for Microstudio to open and burn. Also, you could burn a CD while using Microeditor, which is not possible now. This should be a "relaxation" project for Larry, as the current work has been very hectic... converting all the bus mastering DMA assembly code in Krystal's downloaded software.

The project above will also give us some time to get the .bwav spec (geezer?), and a good direction on where to go with multi-track. I will need your inputs as to what/how to tackle this. It is the last "must have" on my list from 1997. John Tatum, your concept of Ethernet linking is very intriguing. I sure we would all appreciate more details on what devices now have Ethernet ports. That's like a SCSI port, but probably better behaved! Since Ethernet is a "conflict managment" network, versus a synchronized transfer net, there may be synchronizing problems with multi-tracks over this net. What has been your experience so far and how many channels and bit rates?

[JTatum]

Hard disk based recorders store sound as files, just like DAW's. Some hard disk based recorders (RADAR 24 for one) also support more than one file format. Using the ethernet port you can access these files over the network via tcp/ip just like you do any other file. At this point you can copy the file to the filesystem on the local DAW or open the file and convert it to another format. There are no real issues with ethernet except occasional packet loss. Packet loss, however, only tends to occur on large networks of 40 or more users or where routers are involved. Basically, I can use .wav files quite readily. The process I use will be simple. I will copy them across the network from the RADAR 24, create segments, and voila.

I have tested this out with a borrowed mx-2424 and it works real well.

[geezer]

---The ethernet import (as well as removable drive) "solutions" listed above have been my first logical line of approach in thinking this thing through, as well...However, to really be able to use the editing/mixing capabilities of MTU without really having digital multitrack OUTPUT, which seems to be the sticking point from earlier posts, requires being able to take the edited, mixed segments (presumably in groups combined as tracks or some such) from MicroEditor and export/carry them back to the other digital multitrack device in a format that can be read by that device so that you do have a playable, exportable multitrack package....Open TL seems to be one solution that now can be read not only by the MX2424, but Nuendo and a number of other popular DAWs as well....

ANY way of doing this would be fine, but it doesn't seem to me that just exporting .wav is going to work without some track location and time location information attached.

[admin]

With an "Export Groups as Waves" command, we would pad each exported group with silence so they all "start playing" at the same point. This would be from the beginning of the project. That way, there is no need for timecodes, human interface to set them or file format to embed them.

Also, this would work with MOTU, Event, etc. products that use wave files. I can't justify doing this custom for the Tascam. We just don't have the client interest to pay back the development.

[clawson]

Admin writes:

> we would pad each
> exported group with silence so they all "start playing" at > the same point.
> This would be from the beginning of the project.

I have to say that this scares me to death. Many of my groups don't even begin until nearly an hour into a project. This would mean (if I'm reading you correctly) that the exported WAV would have nearly an hour of silence in it and would be around 600MB in size. This would be worse than useless to me, I'm sorry to say. Please tell me that I am misunderstanding you.

Chas.

--
Charles Lawson
Senior Recording/Production Engineer
WETA Radio & Television

[admin]

Chas, this is for multi-track output. There is no other way I am aware of that we can sync multiple "tracks" to an external software program to play out a multi-track card. I am less than enthusiastic about doing custom programming for the Tascam MX2424, which is way more than we sell an entire workstation for today. The number of folks who have voted for this I believe is 4.

I understand your problem, but simply put, Microeditor was not designed as a multi-track I/O system (beyond 4). What I have suggested will work and is streight forward for Larry to code.

I am open to suggestions.

[clawson]

Well, I'll think about it, Dave.

However, it looks like we're about to go Tascam MX-2424 for our multi-track stuff. (The hard drive recorder with integrated DVD backup along with the I/O options make this the best choice for us at this time.) I guess that you could add me as a fifth vote for OpenTL support.

I'll still be able to mix on Microsound using the MX-2424 as a "front end," but I'm not sure how long that will be viable. I guess Nuendo (or maybe the Merging system) is going to creep in there for the heavy-duty music stuff.

In the meantime, Microeditor still makes great radio shows!

I'll forward any brainstorms as I get them...

Chas.