True Bypass on Delta 1010

Audio_Drivel

New member
Hello there,
I recently posted in regards to running my tape machine (tascam 48 1/2" 8 track) in parallel with the delta 1010, I was wondering, I'll probably put the delta as the first in the chain, does it have a true bypass, where the monitored signal goes straight through to its outs without digital conversion? or does it have to run through a DAC/ADC and all?
 
I have one and I opened the manual and looked it up. Here's a quote from the manual.

"The ‘Mixer Inputs’ are inputs to the monitor mixer. These inputs accept hardware audio streams (directly from the Delta’s analog and digital input ports) and software audio streams (digital audio generated in software to be output). This combination of streams makes the monitor mixer extremely flexible. Each mixer input channel has its own level fader and may be panned anywhere in the left/right stereo field. Each input also has its own peak meter. The peak meters indicate the incoming "pre-fader" levels of the incoming audio and are therefore not affected by the fader settings. However, the input faders do affect the levels of the signals exiting the mixer and you will see the affect of the input faders on the output "Master Volume" peak meters."

Although they don't say it explicitly, I believe that all monitoring goes through the conversion process. Keep in mind that the internal precision is 36 bit so quality of the outgoing signals should be fairly clean. Also, in the 1010LT manual they mention that the monitor mixer function actually happens on the processor on the board, which means to me that it has to be digital. The 1010 and 1010LT are slightly different animals, however, so my assumptions may not be correct.

I think I know where you're going with this. If you are wanting to go the purist route, and you don't want any conversion happening in the signal path to the tape, you will have to bypass the 1010 manually (i.e. repatch).

Here's the rest of the manual if you're interested.

http://www.m-audio.com/support/manuals/pdf/DELTA1010_manual.pdf

I would also be very interested in knowing if anyone else has a more detailed knowledge of the signal paths of the 1010.
 
No, it has "zero latency monitoring" which means that monitoring is PRE conversion.

For instance, I plug my guitar into a DI and the DI into the delta, and I play without any latency, unless I'm using live input processing which is software-dependent.

This is why what you hear while you're monitoring doesn't sound exactly like what you recorded...

Slackmaster 2000
 
Are you sure about that? I thought all the monitor mixer functions were implemented via the DSP which deals with digital signals (ie post conversion). The zero latency monitoring is part of the capabilities of the DSP (and does not involve the operating system allowing 'zero' latency).

To have a hard bypass there would have to be a physical switch that routes the analog input to the output (you would hear a 'click' when you select input monitoring). On the other hand I agree that what you hear when monitoring seems different from what is recorded so I'm not sure what is actually happening.
 
Last edited:
There is no A/D/A conversion in this case. As far as hearing "clicks", have you ever heard of "soft switching"? Been around for some time and is implemented in many devices where routing without hearing artifacts of making a physical contact are needed.

So while you don't need to worry about any A/D/A conversions while possibly sending what comes to the inputs to the outputs on the Delta card, you MIGHT pay attention to possibly what the analog part of the circuit is doing. MAudio is funny stuff, and not nearly as "professional" as they hype it to be. Your best bet if you are looking to record to the computer and the analog machine at the same time is to use patchbay connections to do so.

Good luck.

Ed
 
Do you know where/how the analog switching/routing is implemented? It must be happening outside of the VIA Envy DSP chip but it cant be a unique design by M-Audio as you are able to use ST-Audio's drivers on the 1010 (ST Audio cards also use the Via Envy DSP). A block diagram of the Envy DSP can be found on this page.

http://www.xbitlabs.com/articles/multimedia/display/terratec-aureon71_3.html

Obviously it only handles digital signals being a DSP. I originally thought that the zero latency input monitoring was handled by the digital mixer (far left of block diagram) routing the signal from the ADC directly to the DAC without going into the PC. But this cant be how it is implemented if analog switching/routing is happening. Are you sure that the input monitoring is not being handled by this digital mixer?

ps I guess I should have said 'might hear a click' in my previous post (I thought it might help confirm the fact that switching was going on if they heard something, if M-Audio had not used a 'soft' switch).
 
Last edited:
There are soft switched devices, I agree. However, if the signals are routed directly, at what point is the monitor mixer in the chain? The mixer is always there, you can't disable it, which means to me that the signal is always going through the conversion. I dont beleive that the 1010 is using digital potentiometers for the mixer section, which is the only way for it to control an analog signal without conversion.

I would really like to know the real answer to this, so I emailed M Audio support today, and asked them. I'm waiting for them to respond, and I will pass along whatever they tell me.
 
Yes, I am completely sure. Been using Delta products for about 2 years solid now. Lots of products use this same technology.

I've even done testing....input to output latency, regardless of buffer settings or software being used, is less than 1ms constant.

Slackmaster 2000
 
Having no latency doesnt mean no conversion is occuring. As you can see on the block diagram of the DSP the digital mixer can route the signal from the ADC directly to the DAC - this would occur with no noticeable latency. Latency only obvious when the signal has to go from the ADC to the CPU via the digital mixer and PCI bus, be processed by the CPU and come back again. Buffer settings and software would have no impact on the ability/speed of the DSP's digital mixer to route the signal directly from ADC to DAC for input monitoring.

I am interested in what M-Audio say.
 
alfalfa said:
Having no latency doesnt mean no conversion is occuring.

I am interested in what M-Audio say.

Try this.

Set the rate to 8Khz. On the applet for the mixer, that is.

Monitor your "direct no A-D-A" signal.

Notice how it sucks.

Now click to 44KHz or higher.

Fixed?

All of the monitoring on the delta 44 IS absolutely 100% through the card. It is zero latency, because it's in dedicated hardware and not passing through buffers and stuff into the PC. I'd expect the 66 and 1010 to be the same architecture.

The faders for the monitor mixer are digitally implemented. There's very little on the analog side of the card at all.

Try reading the datasheets for the chipset on the boards, they are very detailed. If you are sufficiently bored :)

Mike.

Mike.
 
So is the digital conversion definitely happening on your delta 1010 Slackmaster2K? It is happening on my Echo Mia.

Maybe the soundcard Sonusman uses does things differently and does have a 'hard' bypass. What is the soundcard you use Sonusman? What happens when you follow mjbee's experiment?
 
Last edited:
Slackmaster2K said:
Interesting.

So how come the direct monitored signal sounds so different from the recorded signal?

Slackmaster 2000

Can't answer that one, as mine don't (sound different).

What I hear coming through the monitor bit of the Delta is the same as what comes back on playback. How good this sound depends on the clocking rate of the card. So it could be same-same, or same-but-crap-both-times. If you see what I mean.

If I was using some analogue monitoring thing, like say an old style Soundblaster, I'd expect (for low sample rates) the monitoring to sound great, and the recording to sound band-limited when played back.
 
Back
Top