The sound of software......

[bonneybear]

Has anybody done a clinical a/b test of different daw software..
Using an external clock and coverter ? I'm curious if Pro-tools le sounds like Pro-tools tdm . and what does that sound like compared to Nuendo....etc.....etc...
Im using Radar but plan on dumping into PT le for editing, dont want to muck up the sound though...any thoughts

[simonv]

yeah, I've been reading posts about the sound quality of different daws... to me it doesnt make any sense... And I've never heard any sound difference between Vegas and Acid, but who knows...

[RainbowStorm]

I have done it, I have tested Cakewalk Sonar, Cubase SX and Nuendo in 16-bit, 24-bit and 32-bit float as well as in 44,1, 96KHz and 192KHz, to find out if the DAW software itself can have a sound, like some guys in Nashville say. I recorded idle noise and applied effects in different ways and measured the noise floor level. I have also used my ears when comparing music through different DAW software. The result was that there are differences but they are on the atomic level, so no differences between Sonar, SX and Nuendo. But if the software operates internally with more bits there will be a noticable difference after the 16-bit dithering if the project is heavy. For instance dithering has a much bigger impact on the sound than the choice of DAW software. I have not tested with different clocks and converters, in that way the difference is of course noticable, but it might be hard to know if it's the hardware or software or both that makes the difference that depends totally on the quality of the drivers and converter circuit. But the DAW software in itself doesn't affect the conversion-sound quality, it just communicates with the drivers according to the standard. The same with the operating system, as long as it operates on the same bit depth. The biggest difference is found in the latency, which is because different operating systems don't consume the computer resources equally efficiently and they are also loaded to different degree.

[Boby Stoica]

You won't find any differencies between Vegas and Acid. It's true for CubaseSX and Nuendo as well. But there are differencies between programs from different manufacturers, but - as it was said before, at an "atomic level".

[nathanvacha]

Originally Posted by RainbowStorm

....But if the software operates internally with more bits there will be a noticable difference after the 16-bit dithering if the project is heavy. For instance dithering has a much bigger impact on the sound than the choice of DAW software......

Has it ever been speculated that dither is part of the underlying reason for the popularity in the past couple of years of o.t.b. summing? I really never even considered that as a factor until reading this, just wondered if anyone had...

[astronmr20]

Dither is part of that... sort of. Summing OTB is more for "adding" sound to the mix buss, basically. I seem to think it really shouldent be attempted unless you have really really good converters and clock, but that's opinion.

I'm one who "mostly" beleives that DAW's DO have their own "sound."


I am noticing differences between Logic and Ableton... but I may be psycing myself out due to very different metering in the two DAW's. Visually, I'm a lot more accustomed to Logic for mixing, and that may have something to do with it sounding a bit "better" to me... so... I THINK it sounds better, but that may be psycological.

Also, there was talk in another forum of something called "pan laws" that are setup in a DAW. This I really don't understand, but proportedly can affect how the sound is perceived. In some DAW's you can mess with these "pan laws," But they don't change the "tone" or color per se... just how the stereo image is generated.

Anyone else?

[midigod]

For thirty bucks you can find out for yourself...

http://www.3daudioinc.com/catalog/pr...f634c0713e3301

[Fredrik]

Originally Posted by bonneybear

Has anybody done a clinical a/b test of different daw software..
Using an external clock and coverter ? I'm curious if Pro-tools le sounds like Pro-tools tdm . and what does that sound like compared to Nuendo....etc.....etc...
Im using Radar but plan on dumping into PT le for editing, dont want to muck up the sound though...any thoughts

Check this out, should be some interesting read. As I understand you can also order a CD with different plattforms to decide for yourself.

For me though, it´s a road that leads to making less music.

http://www.3daudioinc.com/3db/forumdisplay.php?f=15

[Fredrik]

Originally Posted by midigod

For thirty bucks you can find out for yourself...

http://www.3daudioinc.com/catalog/pr...f634c0713e3301

AH! you got me

[midigod]

Originally Posted by Fredrik

For me though, it´s a road that leads to making less music.

What did you mean? Do you mean you think one might spend way too much time noodling over minutiae? If so, I agree. Or did you mean something else?

[Fredrik]

Originally Posted by midigod

Do you mean you think one might spend way too much time noodling over minutiae?

Thats exactly what I ment

[RainbowStorm]

Originally Posted by nathanvacha

Has it ever been speculated that dither is part of the underlying reason for the popularity in the past couple of years of o.t.b. summing? I really never even considered that as a factor until reading this, just wondered if anyone had...

Exactly. Let's just sum it up a little and then target the problem. Digital processing is about applying algorithms that process the material by executing calculations on it and then returning the result. These calculations are internally typically at single precision float (32-bit or 4 bytes) due to the operating systems typically operating on that bit depth (float as datatype). With the new 64-bit operating systems that have been released, like for instance Windows XP 64-bit, for instance Steinberg has now added support for double precision (64-bit or 8 bytes) operations in their DAW software (double as datatype). This is crucial, since the rounding errors become smaller and smaller when the amount of bits available for storing the result of each calculation increases.

But summing of tracks is a completely different story. Since each track should be merged into one stereo track devided into two channels, the damage due to calculation errors increases with the number of tracks to be merged. Now imagine what happens when you have a project with approx 10 - 20 tracks playing at the same time, for instance bass, drums, guitars, piano, background vocals and a few overdubs on the main vocals. Typically these tracks also contain a few effects applied on them so each track coming into the merge process is typically already quite blurry due to calculation errors on each iteration in each effect process. What actually happens is that the amount of signal damage now can become really serious. It might be something as little as putting the wrong effect on the mix output bus and all of a sudden the signal is very damaged! Layering many effects on the output bus can also spell disaster... A typical nightmare scenario would be to put a bad limiter on the mix output bus and limit hard after other effects on the mix output bus, since that would really destroy the sonic quality on the whole mix. (something called digital harshness) This is why effect application in the digital domain should have a high isolation level and this is also why it's much better to do summing OTB if the amount of tracks is high (if you have good converters) and convert two times than to convert once and do the summing ITB. That will result in much better instrument definition and create a more balanced result. (even though the damage caused by low D/A quality also affects the instrument definition much, since the engineer can't pan the tracks very efficiently) Something that some professionals practise is to apply effects only on one channel (L or R) on each stereo track. In that way the signal is not as much damaged and the result still might sound pretty much the same and sometimes even better, but much better after the dithering process. So it's generally not a good idea to put any effects on the mix output bus ITB or apply many effects on group or FX channels, but this problem is by far not as serious in DAW software that has support for double precision operations.

Also remember that the merged project then goes through a downsample process in order to fit the CD format at 44,1KHz@16-bit, which means it takes the already blurry material and makes it even more blurry. A good dithering algorithm can then correct this problem to some degree by adding noise to the inaudible frequencies and in that way get more available bits for the important frequencies.

[Rossman]

In my humble opinion all your other gear (mics,pres,comps,converters,etc.) will have more of an impact on your sound than the software. Even though if I could record to tape I would.

[RainbowStorm]

Originally Posted by Rossman

In my humble opinion all your other gear (mics,pres,comps,converters,etc.) will have more of an impact on your sound than the software. Even though if I could record to tape I would.

That's right, that's why for instance the clock is so important in the digital domain, since an unstable clock will have deep impact on the final 16-bit product and make it impossible to do reroutes in order to get passed bad digital processes. Mic quality and pre-amp quality is of course much more important and critical in terms of sound quality, since it determines how much of the natural details is left after the initial sound capture. But a good summing process in combination with a high quality D/A converter and a stable clock aids in keeping the instrument definition clear, which in practise is very important to get an overall good sounding mix.

[SnakeCained]

I have recently been running a lot of different DAW software on the same mac runnig Digi HD hardware. There is not a massive difference in quality of sound when using them as mutlitracks and mixing on an analog console.

However when mixing ITB using plugs then it gets more noticeable. Some plugs on all platforms suck. Some are quite good. There seem to be more and better plugs in VST format IMHO, so it is better value.

I think every DAW has one or two plugs that sound fantastic and occasionaly will load a track into it specificaly for bouncing down "that" sound

I could probably be happy with one of 3-4 different systems but as always you feel more comfortable with the one you use the most. But I have't been too concerned with differing qualities of sound

[RainbowStorm]

Originally Posted by SnakeCained

I have recently been running a lot of different DAW software on the same mac runnig Digi HD hardware. There is not a massive difference in quality of sound when using them as mutlitracks and mixing on an analog console.

However when mixing ITB using plugs then it gets more noticeable. Some plugs on all platforms suck. Some are quite good. There seem to be more and better plugs in VST format IMHO, so it is better value.

I think every DAW has one or two plugs that sound fantastic and occasionaly will load a track into it specificaly for bouncing down "that" sound

I could probably be happy with one of 3-4 different systems but as always you feel more comfortable with the one you use the most. But I have't been too concerned with differing qualities of sound

Yep, much of this phenomenon can be explained by the chaos like law in digital processing, the fact that very small variations for instance in algorithm efficieny will have a big impact on the final result. That's basically due to process iterations where inefficiency happens during each iteration. It's also determined by the quality difference between the consumer's gear and the manufacturer's gear (for instance the clock). This is one weak point about digital recording. So being selective and keeping it simple is very important in the digital domain. I think many out there still prefer to do as much as possible OTB for this very reason. Limiting is really one of the processes when you don't want to do anything in the wrong way...!

[heathen]

I noticed quite a large difference immediately between sound forge and sequoia ,sequoia sounds quite a bit better to my ears, smoother in some respects. So yes different software does sound slightly different, in my opinion.