Do different DAWs sound different?

[T Boog]

Just thought this was an interesting video... 

 

[Mark Morgon-Shaw]

Pro tip:

The best-sounding DAW is the one you actually finish music in.

[Starship Krupa]

Very controversial subject on this forum.

I have plenty to say about it.

First, I really, really want it to be the case that they all sound the same, or at least the same considering differences in such things as default pan law. It's one less troubling thing for my over-focused ADHD brain to grind on. Unfortunately, I have in the past seemed to subjectively perceive differences. The first Cakewalk project I did when I was testing CbB was from audio tracks first recorded in Mick's Craft. Cakewalk immediately sounded "creamier," easier on my ears in a way.

As someone who used to be in the software biz, I know that there are so many ways to code a DAW that it seems unlikely to me that playback would sound identical. So many choices of how to talk to plug-ins, how to mix audio streams together, etc. To imagine that the results of all of these hundreds or thousands of programmers' efforts would result in identical results....it's just not possible for me. For instance, so many things can introduce jitter, and past a certain amount, jitter can be audible.

It's easier for me to believe that the recording of RAW audio sounds identical, but that's not the only thing you do with a DAW. You use the DAW to mix that audio together, process it with effects, etc.

Second, with any DAW, there are two different things that they do to produce audio, and people never consider this: playing audio and rendering audio are different processes. So it opens the possibility that there may be a difference between what you hear when you are in the DAW and hit Play, and what you hear when you export the audio and play it in your favorite media player. There's a compelling reason to make the playback engine easier on system resources, which is that the less CPU it takes, the better the latency. With a more streamlined playback process, you'll be able to stack more plug-ins before getting glitches and dropouts.

Third, there are resources on the web that show that not all audio programs at least do sample rate conversion perfectly. Most notable is the Infinitewave SRC test project, which is unfortunately region-excluded to disallow computers in the United States to access it. The person who runs the site is Canadian, and I'll leave it up to you to ponder why a Canadian might start blocking Americans from accessing his labor of love site.

Also, in the past several years, multiple DAW manufacturers have claimed that their products had new, better sounding audio engines. Ableton, MAGIX, and Acoustica claimed this about Live!, Samplitude/Music Maker/Sequoia, and Acoustica about Mixcraft. This can't be something that their marketing people came up with out of thin air. To make this claim is to implicitly say that the new engine sounds different from the old one. The Infinitewave SRC graphs confirm that newer revisions of the same program can have more accurate sample rate conversion even when rendered. BTW, their graphs for SONAR were at the top of the accuracy chart. Not alone at the top, but solidly there.

If the programming choices made when coding sample rate conversion can be so different, is it not possible that other critical processes might also be different?

I ran my own tests a couple of years ago, comparing Que Kwouk, Re: Purr, Whey Form, and Mick's Craft. The dataset was very difficult to set up. I used a project with a soft synth. 50% L/R pan. Short version, the exports didn't null perfectly, but they nulled better than I expected. However, I didn't make them using a physical loopback to test the DAWs' playback engines, they were all rendered before comparison. It's a LOT of work to set up these tests.

There are also the visuals to consider. When you stimulate one human sense, the other human senses that you are using are affected. So something that is more pleasing to the eye will sound more pleasing to the ear. Of course this can't be proven with such things as null tests, but it is very real. Two different UI's will sound different, if you consider that "sound" is something interpreted by a brain. The old tree falling in the woods question. So if when I first tried CbB the UI looked more pleasing to me than Mick's Craft's (it did), it would have affected my perception of its playback. Just as if someone had been stroking my hair while I was trying CbB vs. pulling it when listening to Mick's Craft. "How did you like the play Mrs. Lincoln?"

And of course, when we are actually working, we're not listening to simple sine waves. We're also using our favorite plug-ins, some of which might be the DAW's native ones.

It's terribly tricky. I have on my to do list to come up with a way to null test through physical loopback. Someday.....

[Starship Krupa]

53 minutes ago, Mark Morgon-Shaw said:

Pro tip:

The best-sounding DAW is the one you actually finish music in.

I'd say that's the ONLY "sounding" DAW.😄

[Mark Morgon-Shaw]

22 minutes ago, Starship Krupa said:

It's terribly tricky. I have on my to do list to come up with a way to null test through physical loopback. Someday.....

Watch the video , they already did that

[Starship Krupa]

6 minutes ago, Mark Morgon-Shaw said:

Watch the video , they already did that

I can't find where he did. He did the same thing I did, which is bounce/render/export and then null test the resulting files. That tests the rendering engine, not the playback engine. While that's ultimately the "sound" of the DAW, it isn't necessarily the sound we hear during playback.

What I mean by physical loopback is recording, via cables, the output of my interface. I've tried doing comparison tests between different players using my interface's internal loopback, and I got impossible results. It must have routed it farther upstream than I expected. So it has to be physical.

[T Boog]

11 hours ago, Starship Krupa said:

So something that is more pleasing to the eye will sound more pleasing to the ear

Cheers Krupa. CbB was my first daw and I find it does sound fat & creamy. I also tried Mick's craft and I thought it sounded kinda like it looked.... Toyish. I found it was a bit midrange-y & lacking in richness. However, I didnt test it in any clinical way.

In the last few nonths I've also experimented with Weeper and I find it sounds good though slightly diff than CbB/Sonar. Anyway, I don't know how much of it is in my head and my eyes but I do believe a person has to be inspired. Regardless of what the "science" says, if ur not feeling something, u should def use something u are feeling.  🤙

[Shane_B.]

13 hours ago, Starship Krupa said:

That tests the rendering engine, not the playback engine. While that's ultimately the "sound" of the DAW, it isn't necessarily the sound we hear during playback.

Exactly. I did a very in depth test one time on sampling frequencies. Iirc, on the soundcard I was using at the time, 48khz sounded best with the least amount if artifacts when rendering fx. This is especially evident when soloing reverb and listening to the trail end/fadeout of it.

The forum crashed shortly after I posted it on the old forum and all the posts for that day were lost. I never reposted it but I wish I did.

I had sound samples and listed the different daws I used. I got different results between Sonar, Reaper, and S1 Thats still internal.

IRT playback/monitoring real-time ... I can hear a difference in how they produce the sound you are hearing. Internally we can null test this and show there is no difference, but with real-time live monitoring there sure seems to be to my ears.

That said, I don't think I've ever heard a bad sounding daw, but I can hear a difference during monitoring/playback.

Sonar has always been my favorite. I only switched due to stability problems that apparently only applied to me (cough cough). All of my gear is boxed up. If I ever get back into music I may land back on the Sonar bandwagon.

Edited 12 hours ago by Shane_B.

[Mark Morgon-Shaw]

18 hours ago, Starship Krupa said:

I can't find where he did. He did the same thing I did, which is bounce/render/export and then null test the resulting files. That tests the rendering engine, not the playback engine. While that's ultimately the "sound" of the DAW, it isn't necessarily the sound we hear during playback.

What I mean by physical loopback is recording, via cables, the output of my interface. I've tried doing comparison tests between different players using my interface's internal loopback, and I got impossible results. It must have routed it farther upstream than I expected. So it has to be physical.

RME has an internal Loopback so he uses that

[Mr No Name]

a lot of them sound different on playback to me, same as media players, but they seem to render the same according to the test, but some don't according to the test ?

did he do the test wrong?

 

some sound  a bit "tinnier" than others to me.

 

go figure.

[Mr No Name]

there's only really one man to prove or disprove this theory conclusively.

 

[Mark Morgon-Shaw]

You can't tell which DAW was used just by listening, so I guess it proves it doesn't matter which one it is - more how you use it. 

[bitflipper]

1,023.

That's how many angels can dance on the head of a pin. Prove me wrong.

[Starship Krupa]

3 hours ago, Shane_B. said:

48khz sounded best with the least amount if artifacts when rendering fx. This is especially evident when soloing reverb and listening to the trail end/fadeout of it.

Yes! Reverb tails. I really hear them too. When I finally found a great reverb, which was iZo/Expo Phoenix when they first put it on sale for $9 a seat, it might have been the single purchase that's made the most difference to my mixdowns.

When I try to introduce people to mindful/critical listening, one of the things I tell them is "listen to the reverb tails." For whatever reason, reverse  reverb also reveals repro quality. My favorite piece of music for sonic eval is Radiohead's "Everything in its Right Place." It uses reverse reverb all over the place and they throw in a bunch of barely audible ear candy stuff throughout the song.

I also find that more accurate repro can make music affect me more emotionally. "Everything" is good for that because of his plaintive vocal style and the open-ended meaning (lessness) of the lyrics. If there's bad resampling or jitter or whatever, I'll sit and listen, but my reaction will be all in my head. When the little details get through, and I can hear slight breath intakes and vocal emphasis, I'll get a slight chill or my heart will feel different. The audiophools call this the "tearjerker" effect. Neil Young mentioned it when he was first pitching Pono.

When I first figured out how to do bit perfect transfers, I burned my girlfriend a CD direct from the DAT master of her album. Played it back on the studio monitors to see if she could tell the difference and tears started rolling down her face. It had made me cry before because she had been away on a trip and I had figured it out while she was away. It was like having a hologram of her between my speakers vs. just a lead vocal.

It's funny, some people seem to be able to hear this while others kind of look at me funny or ignore what I'm saying. You fellers seem to be in the former group. It might be what things are like for people who have experienced UFO encounters. Most people think they're delusional. So excuse me if I fill the screen....

3 hours ago, Shane_B. said:

The forum crashed shortly after I posted it on the old forum and all the posts for that day were lost. I never reposted it but I wish I did.

I had sound samples and listed the different daws I used. I got different results between Sonar, Reaper, and S1

Noooooooo! A few years ago I got raked over the coals on this very forum for asking the same question that Boog did. Supposedly someone on the old forum had done rigorous tests and posted their results showing that everything nulled, therefore it was a canonical fact and I was beating a dead horse. Although I think the lynch mob was mostly that one guy who used to try to jerk my chain, can't remember his handle at the moment. He split from here not long after the Great Licensing Reveal. We had some fun teasing each other.

If I had had your test results as opposing evidence, it would have gone much better. For me it's not whether different audio programs can sound different, but why. And I'm with you, it's not so much better/worse it's just....different. If that's because of CbB's pretty 3-D channel strips, so be it. But without knowing how he was going to describe it, I'd describe my subjective impression of Mick's Craft's playback sound vs. CbB's almost exactly the same way TBoog did: "creamy" was the exact word that popped into my head 7 years ago. Mick's Craft was still sporting what I called the "Fisher-Price Mixer" view; although I got some good work done in Mick's Craft I was starting to feel restricted by its capabilities. CbB looked and sounded "cozier" somehow.

3 hours ago, Shane_B. said:

Sonar has always been my favorite. I only switched due to stability problems that apparently only applied to me (cough cough). All of my gear is boxed up. If I ever get back into music I may land back on the Sonar bandwagon.

People on Reddit went from sharpening pitchforks to thanking the devs when the most recent Sonar release dropped, so you might want to give it a shot. Software never has any issues until the developers find them and fix them.🙄

[Starship Krupa]

11 minutes ago, bitflipper said:

That's how many angels can dance on the head of a pin. Prove me wrong.

Do you believe that every  topic that you've already made up your mind about is unworthy of discussion? Or just this specific one?😄

[Shane_B.]

1 hour ago, bitflipper said:

1,023.

That's how many angels can dance on the head of a pin. Prove me wrong.

Depends on the size of the pin I guess. 

I mean, Angels were giants. That whole Nephilum thing.

[Starship Krupa]

45 minutes ago, Shane_B. said:

Depends on the size of the pin I guess.

And whether the pin uses data compression to fit more of them on it.