What would you buy, LE with Apogee or TDM without?

[Dither16bit]

I actually am not asking what to literally do , I am really kind of asking what YOU would do in the hopes of drawing a comprehensive consensus of the great mind-pool on DUC, keep them coming smarties!

[mindnoise]

considering the prices for TDM plugs and upgrades I stick with LE.

just Beat Detective and new Auto Delay Compensation don´t make up for it. On the other Hand TDM got Massenburg and Echofarm.

No LE is just fine. Since high quality recording affords high quality equipment/frontend I can´t afford.

$.02
regs

[sdemott]

Get a new Mac with 002 or used TDM?

New G5 Mac & 002r. Then, down the road when things really get cranking, add an HD/Accel system to your G5 and grab a PowerBook for the 002r (for off-site editing & location recording, or just for a 2nd room).

One caveat: wait! If at all possible, hold off just a little bit longer, because the new G5s are right around the corner. Apple will have taken care of many of the issues with the current revision with these new ones...and rumour has it they finally figured out how to make these things quiet

That's what I would do in your situation.

[Stephen Bond]

Well I would get a TDM system but you knew that already!

If you post the same thread on the TDM section I'm sure you will get mainly 'get a TDM system' answers.

OK here are some other points to bear in mind.

considering the prices for TDM plugs and upgrades I stick with LE.

just Beat Detective and new Auto Delay Compensation don´t make up for it. On the other Hand TDM got Massenburg and Echofarm.

Sure TDM plugs sometimes cost more, often twice as much but unless you want to use them on Aux/Master faders then you can still use RTAS plugs in your TDM session. Also, there are quite a few plugs that are the same price for both TDM and RTAS.

You get a lot more with a TDM system that just BD and ADC. Check out your PT manual and look at all the 'TDM only' entries. Also there are a miriad of hardware options to expand your system which aren't available for a LE system.

[Chris Coleman]

You get a lot more with a TDM system than just BD and ADC.

ADC isn't even available yet (as far as I can tell, nobody's running 6.4 yet) - and do we know for sure that LE won't have it, too?

[O.G. Killa]

Hi,
I am a full time working, (and suprisingly busy for 2.5 years now), studio owner and producer in Seattle. I have been squeaking by on a 001 running on a G-4 400, (unchanged since purchased 1n '01, works GREAT with OS 9.1), with an Apogee Rosetta 2-channel front end A/D converter and have created some sonic masterpieces in my time on this rig, albeit I spend a g-damn LOT of time and energy bouncing to disk and audio-suiting tracks to achieve this. The time has finally come where I can afford a beefy G-5 and an 002. If I kill myself financially I may be able to swing a bare bones used TDM system and a handful of TDM plug-ins (do TDM plug-ins sound better than native?). Anyhoo, the main question is, do you think a TDM system, (using the 888 converters) would sound better than a 002 with the Apogee Rosetta converters?, (I would have to sell the Apogee to afford the used TDM, so I cannot have the TDM with Apogee, it's either or). What would you do in my shoes?..
Thanks,
Michael Castleman
Sanctuary Studios Seattle Wa.

IMHO, Mix systems are crap. a G5 running RTAS plugins will be much more powerful than a Mix+ or maybe even a Mix3. The 888 sound like !@#$ too. And because the Mix systems are TDM based, you will have tons of issues with phase (that's why everyone makes a big deal about ADC). Here, I'll give you an example...

I work in a studio on HD Accel and HD systems. I have an G4 iBook with an MBox at home. Here is a little test I did...

Test#1
-Open up a new session at 44.1KHz and import audio into it (a stereo track from a CD works great).
-Set the hardware buffer for the playback engine to 1024
-Create two stereo tracks and put the same audio file on both.
-Make sure they both start at the top of the session to ensure they will be exactly aligned.
-Mute one and hit play. While playing unmute the second track. Everytyhing just gets louder by a couple dB (as it should).
-Now hit stop, nudge the second later by 2 SAMPLES.
-Mute the second track, hit play, unmute the second track. Hear the comb filtering? That means there are phasing issues.

Test#2
-OK, now put the second track back where it was (nudge it 2 samples ahead).
-Put a plugin (BF LA2A works great, or a waves plugin) on the first track but don't adjust the settings at all. it shouldn't be doing anything to the audio, just taking up DSP
-Mute the second track, hit play, unmute the second track. What do you hear? (see results below)

Test#3
-Mute the Second audio track and get rid of any plugins you have on the first track
-Create a stereo Aux track
-set it's input to bus 1-2
-create an Aux SEND on track 1 with the fader set to 0. Set it's output to bus 1-2.
-mute the aux track, hit play, unmute the aux track. What did you hear? (see results below)

My results...

Test#1 = TDM-Heard Comb filtering/ LE-Heard comb filtering
(this isn't a test so much as an example so you know what to listen for)
Test#2 = TDM-heard comb filtering/ LE-DID NOT HEAR ANY COMB FILTERING
Test#3 = TDM-Heard comb filtering/ LE-DID NOT HEAR ANY COMB FILTERING

You see, people have been screaming for ADC for years...but LE has already had it. Now, granted it doesn't work as well as it should...if you start piling up plugins on a track it will eventually start to get delayed by a few samples. The TDM mixer doesn't do any compensation for the delay created by processing. This is why everyone complains about how protools sounds bad to them. (THAT IS JUST MY OPINION, BUT I STRONGLY FEEL THAT IS WHY TDM SYSTEMS DON'T SOUND AS GOOD).

When you talk to engineers who mix a lot on Protools, you'll find they will split out all tracks to a mixing board. Then they will either insert processing from the board using hardware EQs/Compressors/FX, or they will use HARDWARE INSERTS in protools. Not many people seem to be aware that you can insert an external analog piece of gear on an insert inside protools.

When recording at higher sampling rates, the latency is reduced. DSP chips (on Mix and HD cards) have latency inherently in them that is measured in samples. each chip may have anywhere from 3 to 9 samples of delay (this is without adding plugins...just using the chip for the mixbus). Then you add a Waves L2 which requires a lot of processing power and you can have a latency of up to 100 samples if not more. BUT!!! each sample at 44.1KHz sampling rate is 1/44,100th of a second. Each sample at 96KHz is 1/96,000th of a second. So a 3 sample latency at 44.1KHz would be 3/44,100ths of a second, at 96KHz it would be 3/96,000ths of a second.

Aaahhhh, the phase shift due to latency will be significantly less at higher sampling rates...this is why people think recording at 96KHz sounds better. Everyone always says, "The bass is tighter and punchier...it sounds fuller, and the high end is more defined..." All these things happen when you reduce the amount of phasing shifting.

DSP latency is also why everyone always says the RADAR sounds better than protools. The Radar (which i have also used extensively) HAS NO MIX BUS!!! You record a single track in, it plays it back out. you record 8 individual tracks in, it plays 8 individual tracks out. There are no plugins, no bussing to multiple destinations, no aux tracks or aux sends, no DSP, no mix engine. If you could use Protools|HD that way (bypassing the TDM cards and Mix Engine) it would sound very similar to a RADAR.

Sorry I'm going off on a tangent...myself and another engineer got in a big debate about this the other day (which is what sparked me doing the test listed above with him).

But anyway...getting back to why I went of on this tangent... If the only reason you would be getting a mix system is so that you don't have to AS and bounce to disk as often then DONT DO IT. A Dual G5 with a 002r/PTLE will have more horsepower than an old Mix system and sound better. If you were to ask the same question about an HD system and a 002r, then it would be a tougher call since the HD Accel systems are very powerful and recording at 96KHz/192KHz will help keep everything time aligned (well at least moreso than at 44.1/48) when mixing.

Anyway, I hope this helped...but even if it didn't, I feel much better now for getting it off my chest!

Oh, and by the way...this is, like I said, my opinion and what I have noticed when using the Protools. Other opinions are welcome. I'd be interested if other people have noticed this or if anyone has done more extensive research into this.

[Stephen Bond]

Sorry I'm going off on a tangent...myself and another engineer got in a big debate about this the other day (which is what sparked me doing the test listed above with him)

Sorry but I think this really summed things up! All digital processors from outboard to plugins have some latency but this is normally only an issue if you have lots of spill. How many times do you want to double a track and use different processing on both tracks, playing at the same time?

Also I don't think LE peforms in quite the way you are suggesting with regard to latency. Many plugs don't even report the processing delay.

[O.G. Killa]

Sorry but I think this really summed things up! All digital processors from outboard to plugins have some latency but this is normally only an issue if you have lots of spill. How many times do you want to double a track and use different processing on both tracks, playing at the same time?

I'm not really sure what you're getting but I think you're agreeing with me, correct? One thing I will say is that it's an issue AT ALL TIMES. I used a stereo audio file from a CD as an example. But like you said how often will you double a track and use different processing at the same time...BUT....How often do you record drumset with more than two microphones...or a Strings...or Horns???

Think about a drumset for example. A normal drumset mic configuration is 10-12 mics. You move all the mics around so that when all of them are on, the drumset sounds good. This means they are all in phase. You record the drumset, everything sounds great!

Now you mix. And the kick drum needs a little more umph. So you add an EQ and boost around 100Hz...no big deal right? WRONG!!!! In a TDM system using a Waves REN EQ you've just put 4 samples of delay on the kick drum by adding that plugin. In my little test, you could very easily hear comb filter after only nudging by 2 samples!!!! So now your kick starts to sound like crap. Then you decide to compress the snare and add a little EQ. Now the Snare track is shifted from all of the other mics that have picked up the snare (including the room mics). Not to mention the phsase shifting that naturally happens as you apply and EQ. As you can see, if you tweak all the tracks of the drumset the way you would on an analog console, the phase shifting between all the of tracks is going to be monumental!

This also can sometimes happen between instruments too. For example, kick drum and bass guitar. When you recorded the band the bass guitar and kick sat nicely together...but now as you mix, it's starting to get messy. The relationship in time between the bass drum and bass guitar effects how the two will sound together...you can have a kick and a bass guitar cancelling each other out if your not careful.

Also I've noticed this when processing double-tracked electric guitars. Weird things can start happening when you add plugins to one and not the other. Say you want make one a little brighter than the other so you can tell there is two separate tracks of guitar...once you start it starts to get smeared sounding and lacks definition.

Also I don't think LE peforms in quite the way you are suggesting with regard to latency. Many plugs don't even report the processing delay.

Actually, if the plugins aren't reporting any delay on an LE system then it is performing the way I suggested. If it is not reporting any delay, that means that the system is buffering the audio earlier by however many samples the plugin would delay it...then it processes it and when it comes out its perfectly in time with the rest of the session.

What can happen in LE systems, if you add A LOT of plugins to one channel...you'll noticed it will report that there is some delay (maybe 5 to 10 samples). For example, to get that telephone voice effect, you need to compress, EQ, compress and then limit. On a vocal track (in PTLE) add Autotune, a Waves RenComp, RenEQ, a BF LA2A, and then a Waves L2. Set them all so they are giving you that desired effect...now look at how much delay there is. You'll notice protools reports some although not very much (in a TDM system it would be dramatically higher). BUT, like I said, even two samples of delay will drastically effect the sound of your material.

[Dopamine]

If you need more than 32 tracks and are interested in beat detective get the used tdm rig. otherwise, by all means, go for the 002/G5 combo. That way you will be current for the next few yrs w/ PT updates. Don't sell you Apogee though!

[Stephen Bond]

Do a search before you buy-there are a lot of people with problems of noise using the G5 and 002/002r combos.

Also, as budget is limited maybe a fast dual G4 would be better and performance isn't hugely different to the G5.