X48 results so far....

[LX3]

Originally Posted by muziekschuur

Major Update Adds Features to Best-selling 48-track Workstation

Montebello, CA (October 6, 2008): New for the TASCAM X-48 is software version 1.10, an update that adds several user-requested features to the best-selling 48-track digital audio workstation. Among the new features are a new "Big Meter Mode," which changes the VGA display into a meter bridge for live recording and tracking. Other added features include a new "fail safe recording" mode which saves the file every 5 seconds. TASCAM also made improvements to the user interface, automation system and minor bug fixes.... some of the top names in live recording use the X-48. We listened to their requests and version 1.10 is the result of their input."

YEE-HAAAAA! So, there *is* an update! I think we'd all started to wonder...

(Although, who are these users that have requested features? I have long been sitting on a big list of minor changes that would make a real difference to the useability of the machine... but no-one asked me . Hopefully Tascam have been secretly paying attention to TascamForums user requests. I'll be fascinated to see what has changed.)

X-48 version 1.10 is in final testing and will be available as a free update for owners later this year.

Ah... a bit too excited too soon.

I'm looking forward to those bug-fixes too. Tascam might be interested to know that I recently used my X-48 to record the new Siouxsie Sioux live DVD. Totally out of the blue, it crashed with a delightful Windows error message in the second set. So the DVD will be using the audio from the safety recorders.

PS regarding sound quality issues, I've been using an external master clock for my X-48 recently, and it seems much better than it was on internal, so I'm finally happy using the Tascam analog cards.

[muziekschuur]

Did you use the internal drive or an external firewire harddisk?

I still have two Roland VSR880's wich were synced at the time and wich recorded some bands.

Unfortunately it was impossible to playback the recordings and even Roland seemed unable to get the audioplayback going.

I still hope them audio is on those disks.

Why am I thinking about that? Related to the X48? If the audio is written to the internal harddisk, dropout may occur when the harddisk is allmost filled. Due to the nature of the windows filesystem performance goes down when a disk is more than halffull. Others will be better in explaining this particular phenomenon. This is why others allready have suggested to use an external (large) drive. But I guess you allready knew that.

Muziekschuur

[Chance]

I am curious about VST plug ins. Are you limited? Can you use VST "virtual instruments" within the DAW of the X-48?

[LX3]

I don't have any problem recording to the internal HD. Even when it's 85% full or more (although it's obviously better to have a bit more free space than that. I do wish that audio was recorded to an entirely separate partition on the internal HD - instead of just it's own folder - so you could totally re-format the audio partition without touching the OS).

As far as I can tell, it's the internal DSP mixer that crashes the X-48. Something random happens, it freaks ("Windows attempted to write to an illegal memory location" or something like that), and Windows crashes, requiring a complete reboot of the machine - which takes a good old while.

It doesn't happen a lot. But it's happened to me twice (on two different X-48s) during critical one-off recordings... at which point the backup recorders have saved my reputation.

Thankfully, with the acquisition of a new desk, I now hardly ever need to use the internal mixer.

Chance, regarding VST plugs and instruments - I don't think you can do VIs at all on the X-48. To be absolutely honest, I would not attempt to mix or do any DAW-type operations (e.g. editing, mixing) on the X-48. Although it's a handy facility to have on-board, you'll find that the lowliest PC or Mac-based DAW works a lot better if you need anything more than basic monitoring. The X-48 is based on a single Celeron and in my experience can only just keep up with routing/pan/level across 48 tracks of audio. A few EQs would push it over the edge... I hate to think what would happen at 96kHz.

But for what I'm doing, the X-48 is working great, as long as you're aware of the limitations. I just recorded four days of a live-music TV show on it, about eight hours a day, 48 tracks running the whole way thru, locked to timecode, without a glitch.

Paul

[justintime]

Hey everybody,

I've recently been working with the Tascam X48 doing some very basic editing of live recordings and the darn thing doesn't seem to like me very much. When I press play after about ten seconds or so the thing starts to skip back to a point in the audio a few seconds previous and 'loop' repeatedly. At this point the transport locks up and I have to press the stop button 3 or 4 times before it will stop. This happens quite often and is really, really, really, reeeeeeeeeeeeeeeeaaaaallly annoying . Any thoughts?

[Roland]

Yes, make sure you have your files backed up elsewhere and try doing your edit on something else like Pro-tools, Pyramix or even Nuendo and alike. I've heard many tales about instability with the X48, especially when trying to use the mixer during record, it's a shame for what promised to be a potentially great product.

Regards


Roland

[Remoteness]

He's not talking about using the mixer during recording or even during the mix.

He's saying that during simple playback or editing the audio starts to skip back to a point a few seconds previous starts to loop repeatedly. Then the transport locks up and needs to press the stop button aa number of times before it actually stops.

Not a cool scenario when you're trying to get work done.

[yrplace]

Originally Posted by justintime

Hey everybody,

I've recently been working with the Tascam X48 doing some very basic editing of live recordings and the darn thing doesn't seem to like me very much. When I press play after about ten seconds or so the thing starts to skip back to a point in the audio a few seconds previous and 'loop' repeatedly. At this point the transport locks up and I have to press the stop button 3 or 4 times before it will stop. This happens quite often and is really, really, really, reeeeeeeeeeeeeeeeaaaaallly annoying . Any thoughts?

Are you using the internal drive or an external firewire 400 one?

We've been using an X-48 as our backup machine for remotes for a year and a half with no problems and we always record to a good external drive.

Once I even turned on the mixer while doing a 48 track analog xfr session and monitored the digital out and it worked fine.

I should point out that we are recording at 48k and I have never tried any higher sample rates.

Strange that some are having a lot of trouble with the Tascam while others are not.

I wonder if this is a build problem or some software issue.... very odd.

Mark

[Roland]

Originally Posted by Remoteness

He's not talking about using the mixer during recording or even during the mix.

He's saying that during simple playback or editing the audio starts to skip back to a point a few seconds previous starts to loop repeatedly. Then the transport locks up and needs to press the stop button aa number of times before it actually stops.

Not a cool scenario when you're trying to get work done.

I understood that, the point I was making (though not very clearly) is that apart from strictly recording or playback there seem to be lots of reports of problems when other functionality is needed.

Regards


Roland

[Remoteness]

Gotcha...

Furthermore, Mark makes a good point -- why do some folks have technical problems and others have none?

Is it indeed a build problem or some sort of firmware / software issue?

[muziekschuur]

Any news besides 1.10 wich doesn't seem to be released?


Muziekschuur

[HughH]

Hello,

We're using X-48's for 96 tracks of record with a Nuendo dual-madi backup, and we have just had to resign ourselves to it's issues - gui instabilities, random consolidation of tracks if we chase code for too long, refusal to go into record when sitting too long, and very, very clumsy editing. So we work around it - we only use it for acquisition, we transfer to another device for even the simplest editing, and we reboot it more often than we should have to. We all wish it was a better device, and here's to hoping for 1.10 being an improvement, but no one knows if it will ever get there. That's why we have a backup.

Hugh

[Remoteness]

We feel your pain and hope for the best...

[bishopthomas]

I used a pair of X48's the other day and had boot failures on one machine. After way too long it displayed "Error loading host" and I had to reboot. Same thing several times until it finally decided to cooperate. Both machines had the same firmware, even purchased as a pair.

[LX3]

My X-48 has been hammering thru a live music TV show for the last six weeks. On balance, I'd have to say it's been pretty solid... although we don't use the on-board mixer any more.

A couple of new bugs in 1.04 that we've discovered along the way:

i ) the only way to stop copying audio in the file manager seems to be to pull the mains! Embarrassing when we suddenly needed to get into record again after we thought the show had wrapped one night.

ii) it drops out of record when timecode rolls past midnight. But it doesn't stop rolling - it just goes into play. Took us two minutes to notice. Oops.

But very few crashes or misbehaviours, which when you consider how hard we've been driving it on these shows was a pleasant surprise.

[huub]

What happens with the x48 wav files if power fails during recording?

huub

[Remoteness]

Originally Posted by huub

What happens with the x48 wav files if power fails during recording?

huub

You can set the machine to backup the files at a time point you can set.
I believe we have it backing up the files every ten minutes.

[yrplace]

And better yet get a UPS t least for the deck. If it saves your *ss once it will be well worth it. And trust me someday you will learn this lesson, we all do.

BTW regarding the post above..... timecode should never be allowed to go from 23:59;59 back to zero. Almost all machines will stop recording if this happens.

Best to set code for AM ie midnight is 12:00:00 not 24:00:00 to avoid this problem.
Easier to read anyway.

Mark

[LX3]

Originally Posted by yrplace

And better yet get a UPS t least for the deck. If it saves your *ss once it will be well worth it. And trust me someday you will learn this lesson, we all do.

BTW regarding the post above..... timecode should never be allowed to go from 23:59;59 back to zero. Almost all machines will stop recording if this happens.

Best to set code for AM ie midnight is 12:00:00 not 24:00:00 to avoid this problem.
Easier to read anyway.

Mark

Fine if you're recording in isolation, but in my case I have to be working to the same timecode as cameras, which are all time-of-day. But thanks for the tip... if I'm ever working with code but not with video, I'll remember that one.

Some mention of the issue in the manual would have helped. Had to find out the hard way.

I don't see any reason why any audio recorder should *have* to malfunction when you roll over midnight. It's just lazy programming and design, and in a machine designed for use in a broadcast environment the manufacturers should realise that people work late occasionally!

Another good reason for always having a safety recorder (as well as a UPS).

[yrplace]

We always ask video to use 12 hr code rather than time of day unless the show is going to run for more than 12 hrs, and most video trucks will do that anyway if its a concert that has any possibility of running past midnight.

I believe the problem with crossing midnight (24:00:00)and going back to zero hrs is a function of SMPTE not the machines, but I know most machines unless they are just recording code onto a track can't chase code past 23:59:59.

If this is still a problem you probably can also avoid it by setting the X-48 chase to infinite flywheel which is best anyway to prevent a tc dropout from causing the deck to stop.

If its a long show that needs to be in constant sync with video then you need to be slaved to video sync. If its a concert where there will be numerous pullups between songs you can get away without video sync as well.

Mark

[LX3]

Thanks for the suggestions about working to a 12-hour clock, I'll bear that in mind... except we're often starting recording in the AM, so it's not a solution on those occasions. This was just one of those unusual situations where we were in record at 11am, then recording past midnight as well. Lonnnggg daaaayyyy!

I'm locked to word clock that's synchronous with the time code. Works great. But will def look for an infinte flywheel mode. Seems sensible, since these machines don't really chase timecode the way an open-reel and a synchroniser does. So no advantage to continally "chasing code" as I can see it.

Worth noting of course that the crew's video cameras didn't suddenly drop out of record at midnight. I don't see why audio recorders should be any different. As I say, I think it's just lazy design. I can see there are a few issues to contend with, but there are several better ways a recorder could behave other than just dropping out of record.

I think Nuendo knows what day (as in "day 1", "day 2") it is in a timecoded project, which is quite cool.

For now, just knowing that the X-48 doesn't like midnight is a start. You can't compensate for it if you don't see it coming. Which I didn't. I suppose I should have, but as I say, it had been a very long day.

[yrplace]

I would run a test to see if you can record past 23:59:59 on the X-48 if it is in infinite flywheel mode.

But you are right using infinite mode is best when recording on digital decks or DAWs since there is no need to chase code the way you would with an analog recorder.

Mark

[huub]

Also, BWF only records a timestamp at the beginning of the file, so as long as the wordclock is synced to the video, you'll still be sample accurately sync until eternity..

huub

[HughH]

Hello,

The ability to record past midnight while chasing code is based on whether each software developer decided to implement a "method" to deal with it, in the sense that there is nothing past 23;59;59;29 in the SMPTE standard, so it's an add-on capability for the workstation.

You're correct, Nuendo has already done that by adding a "day" field to the tc information which is easy for Steinberg's implementation because all their project location data is sample based, so it's just more samples. Of course you can't put that into smpte code except in user bits. A number of film-style location recorders put the day information (along with take, scene, etc.) into the user bits. There are some loose standards but they're not completely codified yet.

Lazy programming? Well, I'd phrase it more like "not dealt with yet". But that's just me.

H

[LX3]

Thanks for all the suggestions folks... I'll run some tests when I get a chance. Yup, I can see how "infinte" flywheel might fix the problem, if it means the machine never bothers to check incoming timecode again after it first locks up.

To be honest, because I've been generating synchronous clock and timecode, I haven't been taking too much notice of my flywheel settings... All I know off the top of my head is that it's not set to zero, but having it set to infinite would have saved my ass on one other occasion (e.g. it's really easy in the mad rush of TV to forget that... even though you're supplying nice synchronous word clock... your X-48 might accidentally be running on internal clock )

Worth noting that "infinite" threshold was only added with v1.0.4 of the Tascam OS... (so it's not in the manual, only in the 1.0.4 release notes).

BTW, something else I'll check, but I *think* the X-48 timeline knows that an event at 1.00am might be AFTER an event at 11.30pm, and not earlier... because looking at the project in question the other day, the audio we recorded at 1am was further to the right on the timeline than the audio at 11.45pm. Even though I'll swear that when we were recording it seemed like the song re-wound itself all the way back to the start of the timeline (i.e started placing clips to the left of the already recorded audio).

I've noticed today that the Denecke timecode products add a flag in the output timecode user bits when they roll past midnight. So, you know, at least someone out there is thinking about the issue, even if there's no well-defined way of handling it.

[muziekschuur]

How's everybody doing so far?

No sign of 1.10. Allthough officially announced on october 2008

[Remoteness]

Originally Posted by muziekschuur

How's everybody doing so far?

No sign of 1.10. Allthough officially announced on october 2008

Sounds about right.

[HughH]

Hello,

We have two of them with v1.04 and have discovered a few more things to note -

The previously mentioned problem of unwanted file joining (and watched personally by Mr. Chang at Tascam) seems to be related to Destructive Mode. We have one operator who always wanted to use Destructive Mode projects. It appears that if you use Non-destructive projects then the files don't join. At least so far.

If you're chasing code and want to drop out of record, don't push "stop", push "play" which will drop out of record while continuing chasing. Occasionally pushing "stop" while recording will lock up the machine and the gui goes away and doesn't recover. The audio files are fine but the project file is unrecoverable.

Don't use any of the workstation functions. At this point in time it's not where it needs to be with the previously mentioned gui issues. We use it for acquisition and it's mostly fine.

We ALWAYS run a backup. We use the two X-48s for primary 96 track record and a Nuendo system for 128 track backup.

Hugh

[muziekschuur]

Hugh,

In the past you told us about a very complex way of recording. Is the X48 workable now, with the play button use while in chase.... ? Could the solving of this issue be the explanation for the delay of 1.10?

[LX3]

I had similar issues early on with "tape mode", but to avoid slandering the X-48 any more than I have done already let's just say, yes, destructive mode = bad, non-destructive = good.

I don't think there's any particular advantage to using destructive mode when recording live. If you lock to time-of-day (which is always handy) you never "rewind" anyway, so there's nothing to consolidate.