|
|
Post by mobeach on Nov 4, 2014 10:20:55 GMT -6
I think it would be very interesting to hear tracks recorded on different programs using the same instrument through the same interface by the same musician. Onto ProTools, Logic, Reaper, Sonar etc.. anyone aware of such a thing?
|
|
|
|
Post by svart on Nov 4, 2014 11:23:47 GMT -6
There's a site that compares daws, but i don't know the link. I'll see if i can find it.
|
|
|
|
Post by popmann on Nov 4, 2014 13:50:42 GMT -6
No difference, other than poor delay compensation on the part of some, like (native) ProTools. Any app oh trusts the developer, will have shit compensation. Want to test how well it works in your daw? Download Voxenego's latency compensation plug....if you can successfully delay the track with it, your DAW fails, because the basis of how it works is by intentionally lying to the host about how much latency it needs and then internally varying the delay "manually". If that plug works, you have an issue--know it or not.
I have extensively compared....floating point DAWs sound the same with the compensation working. Digital is not digital--fixed bit hardware mixers will absolutely NOT null...but, if you're going to work in a DAW, you should take my word from experience and simply not waste time.
Now, if you're using the built in plug ins, theres literally no way to accurately compare. But, I think Logic is king by a long shot....Reaper's built in are pretty darn good--awful UI aside. Mostly, like the mixer itself--they're all using the same (IMO flawed) code blocks.
|
|
|
|
Post by mobeach on Nov 4, 2014 19:20:37 GMT -6
I'm using 64 bit Reaper, I'm curious how it stacks up against the rest as far as the average ear goes. I actually like the UI in Reaper, it seems natural to me. Sonar is one I DO NOT like.
|
|
|
|
Post by wiz on Nov 4, 2014 19:40:30 GMT -6
No difference, other than poor delay compensation on the part of some, like (native) ProTools. Any app oh trusts the developer, will have shit compensation. Want to test how well it works in your daw? Download Voxenego's latency compensation plug....if you can successfully delay the track with it, your DAW fails, because the basis of how it works is by intentionally lying to the host about how much latency it needs and then internally varying the delay "manually". If that plug works, you have an issue--know it or not. I have extensively compared....floating point DAWs sound the same with the compensation working. Digital is not digital--fixed bit hardware mixers will absolutely NOT null...but, if you're going to work in a DAW, you should take my word from experience and simply not waste time. Now, if you're using the built in plug ins, theres literally no way to accurately compare. But, I think Logic is king by a long shot....Reaper's built in are pretty darn good--awful UI aside. Mostly, like the mixer itself--they're all using the same (IMO flawed) code blocks. Hey popmanncan you expand a little on this topic please. Eg: are you saying that I could have plugins, e.g. say IK or waves or slate plug ins, that may not report their latency to the host correctly (in my case Logic Pro X) and cause phase issues? And that this plug in by voxengo, would allow me to test that by downloading it and inserting it? How do you test an individual plug in? cheers Wiz |
|
|
|
Post by popmann on Nov 4, 2014 21:35:04 GMT -6
Testing an individual plug in is mostly an uphill battle. I mean....you're gonna have a very sharp click of an audio track....bounce it to another track....insert the plug in....bounce it to a third track...and when I say bounce--I actually mean record the audio in real time. Here's the back story--the last version of Nuendo I tried to use many years ago was N2....it had the issue. When I contacted Steinberg, they claimed it was because the plug in reported incorrectly and there was "nothing they could do about it"....I wrote them a nice little politically correct "bullshit" email. Wherein I told them that IMO, the ONLY job of a host to record/playback audio in correct time--if they allowed a third party's "reported latency" screw that up, it's THEIR fault--not the company misreporting. Well, apparently, they took that to heart, because next time I tried audio mixing in their host (Cubase 4)--they'd taken my advice and tested every plug in when you turn it on. Which is another aspect--when you turn it on....it may not have the same latency as if you enable a feature--for example, in ProQ, if you engage some linear phase mode, it needs more CPU time (thus latency)....so, you need to disable and enable it after you change said feature. Or next time you open the project, Cubase will check it itself and correct it...but, obviously you want it done NOW.... free plug in www.voxengo.com/product/latencydelay/ -there is an audio unit....but, I don't know that there's a 64bit one for LogicX. The purpose of the plug was to DO compensation back when hosts weren't doing it right....so, umm--maybe you could use in (software) ProTools today.  But, it works on a simple principle--it tells the host it needs some huge amount of latency--say a full second (whatever it's max is)....and then internally allows you to adjust from 0-1sec in milliseconds. It's intentionally "lying" to the host. If you put it into Cubase today, it will not work--because Cubase won't accept their "lie in reporting"--it will test it and compensate for whatever you set the latency to--ie, the actual latency of the plug in. The compensation being off cause TIME issues....phase issues would result from time issues on multimic'd sources, as it' relational. The time issues have another effect, which is that the whole mix gets "softer"....sloppier--not by enough that you "hear" like "ooh--that's hitting late and that other track is too early, but it's cumulative effect is it just all gets....looser feeling. Logic is an interesting beast with regards to it's compensation. It actually SLOWLY moves tracks to the right place in time--rather than do what Cubase does, which is stop playback (breifly) when you insert a plug in anywhere....it keeps playing back and if you use a really latent plug, you can hear it start wrong and start to MOVE back where it needs to be. No idea how accurate it is--feel free to test it--I just mention the phenomenon so you test something longer than like a second, because it takes it a bit to catch up.... But, yes--third party plug ins reported wrong for YEARS....I have no reason to believe that changed. I would not have switched audio production to software had Steinberg not stopped trusting them. It just had too much cumulative damage. I actually, for years, couldn't believe that people couldn't hear this and tolerated it....ProTools 10 was terrible. It was like N2 all over again. They really fundamentally ARE where Steiny was then--being focused on an actual native DAW, they are the absolute noobs on the block. Immature code. I have a feeling that's what the ground up 11 rewrite was really about. They were trying to convince people to buy HDX system, which are native with plug in DSP support--thus effectively it's PTLE with some DSP cards--unlike TDM which was a hardware mixer AND DSP. |
|
|
|
Post by wiz on Nov 4, 2014 22:01:35 GMT -6
thanks bud, for taking the time to write that reply...
great info
cheers
Wiz
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 4, 2014 22:43:56 GMT -6
That's the fear keeping me from PT11
|
|
|
|
Post by scumbum on Nov 4, 2014 23:58:52 GMT -6
Testing an individual plug in is mostly an uphill battle. I mean....you're gonna have a very sharp click of an audio track....bounce it to another track....insert the plug in....bounce it to a third track...and when I say bounce--I actually mean record the audio in real time. Here's the back story--the last version of Nuendo I tried to use many years ago was N2....it had the issue. When I contacted Steinberg, they claimed it was because the plug in reported incorrectly and there was "nothing they could do about it"....I wrote them a nice little politically correct "bullshit" email. Wherein I told them that IMO, the ONLY job of a host to record/playback audio in correct time--if they allowed a third party's "reported latency" screw that up, it's THEIR fault--not the company misreporting. Well, apparently, they took that to heart, because next time I tried audio mixing in their host (Cubase 4)--they'd taken my advice and tested every plug in when you turn it on. Which is another aspect--when you turn it on....it may not have the same latency as if you enable a feature--for example, in ProQ, if you engage some linear phase mode, it needs more CPU time (thus latency)....so, you need to disable and enable it after you change said feature. Or next time you open the project, Cubase will check it itself and correct it...but, obviously you want it done NOW.... free plug in www.voxengo.com/product/latencydelay/ -there is an audio unit....but, I don't know that there's a 64bit one for LogicX. The purpose of the plug was to DO compensation back when hosts weren't doing it right....so, umm--maybe you could use in (software) ProTools today. But, it works on a simple principle--it tells the host it needs some huge amount of latency--say a full second (whatever it's max is)....and then internally allows you to adjust from 0-1sec in milliseconds. It's intentionally "lying" to the host. If you put it into Cubase today, it will not work--because Cubase won't accept their "lie in reporting"--it will test it and compensate for whatever you set the latency to--ie, the actual latency of the plug in. The compensation being off cause TIME issues....phase issues would result from time issues on multimic'd sources, as it' relational. The time issues have another effect, which is that the whole mix gets "softer"....sloppier--not by enough that you "hear" like "ooh--that's hitting late and that other track is too early, but it's cumulative effect is it just all gets....looser feeling. Logic is an interesting beast with regards to it's compensation. It actually SLOWLY moves tracks to the right place in time--rather than do what Cubase does, which is stop playback (breifly) when you insert a plug in anywhere....it keeps playing back and if you use a really latent plug, you can hear it start wrong and start to MOVE back where it needs to be. No idea how accurate it is--feel free to test it--I just mention the phenomenon so you test something longer than like a second, because it takes it a bit to catch up.... But, yes--third party plug ins reported wrong for YEARS....I have no reason to believe that changed. I would not have switched audio production to software had Steinberg not stopped trusting them. It just had too much cumulative damage. I actually, for years, couldn't believe that people couldn't hear this and tolerated it....ProTools 10 was terrible. It was like N2 all over again. They really fundamentally ARE where Steiny was then--being focused on an actual native DAW, they are the absolute noobs on the block. Immature code. I have a feeling that's what the ground up 11 rewrite was really about. They were trying to convince people to buy HDX system, which are native with plug in DSP support--thus effectively it's PTLE with some DSP cards--unlike TDM which was a hardware mixer AND DSP. Hey popman I use 7.4 LE and only use the stock plugins or the Sony Oxford bundle . Those should be accurate reporting any delay ? |
|
|
|
Post by popmann on Nov 5, 2014 7:35:46 GMT -6
Stock plugs will IME always report correctly. I would not count on any third party reporting correctly all the time. It's not a function of them doing something wrong....like Sonnox is "a good company" therefore they will work in very host and version....if they are latent (and not all are, fwiw)....there's a chance of miscommunication.
But, honestly....if you've not noticed in how many years? How much of an issue can it be?
|
|
|
|
Post by svart on Nov 5, 2014 8:22:36 GMT -6
I searched high and low for the website that had very scientific tests for the audio engines in each of the most famous DAWs. I can't seem to find it now, it may be gone.
Anyway, the short story is that scientifically there isn't much difference between the DAWs and their outputs. Some of the differences between the DAWs were mostly about the quality of their digital Nyquist filters, but the audio band was almost identical.
The internet is filled with those who wish to sway opinions into the "yes they are different" or into the " there is no difference", but the truth is that the code for each DAW is different, therefor the process is different, and that means that the result must be different.
Now, that doesn't mean the result is different enough to matter though.
And I have a hard time believing that any company shelling out millions of dollars to develop and maintain software like this would allow serious or fatal flaws to happen. I'd say pick one that has the features you like and work with it.
Otherwise, you could do a bunch of nulling tests between the DAWs available to you.
|
|
|
|
Post by mobeach on Nov 5, 2014 10:56:16 GMT -6
I think stability of the program is huge, the less expensive DAW's tend to freeze up a lot more. Or Mixcraft 6 for example, it's a decent DAW for beginners but you have to re scan your plugins just about every time you fire it up. Which takes about 4 minutes.
|
|
|
|
Post by clonewar on Nov 9, 2014 23:40:20 GMT -6
I searched high and low for the website that had very scientific tests for the audio engines in each of the most famous DAWs. I can't seem to find it now, it may be gone. You might be thinking of the 3DAudioInc site? www.3daudioinc.com/3db/forumdisplay.php?15-Awesome-DAW-SUM-ComparisonI used to wonder about this and ran my own tests. I found that when the same tracks are brought into different DAWs, and they're carefully setup the same way (faders, pan, pan law, etc), and bounced with no dither, that the bounces from each DAW are bit identical and null to infinity. I tested with PT, Logic, Cubase, Sonar, Studio One, and SAW Studio. When you start adding plugins, automation, etc, things can change, and delay compensation has already been talked about. |
|
|
|
Post by M57 on Nov 10, 2014 6:05:25 GMT -6
This is semi-OT, but I miss using my Paris system. It looked and sounded great and the future was bright and shiny.. Now all DAWs sound the same.
|
|
|
|
Post by svart on Nov 10, 2014 8:19:46 GMT -6
I searched high and low for the website that had very scientific tests for the audio engines in each of the most famous DAWs. I can't seem to find it now, it may be gone. You might be thinking of the 3DAudioInc site? www.3daudioinc.com/3db/forumdisplay.php?15-Awesome-DAW-SUM-ComparisonI used to wonder about this and ran my own tests. I found that when the same tracks are brought into different DAWs, and they're carefully setup the same way (faders, pan, pan law, etc), and bounced with no dither, that the bounces from each DAW are bit identical and null to infinity. I tested with PT, Logic, Cubase, Sonar, Studio One, and SAW Studio. When you start adding plugins, automation, etc, things can change, and delay compensation has already been talked about. Nah, that wasn't it. it was a site where you had pull down menus with different DAWs and settings and you could compare charts/graphs of frequency sweeps, group delays, etc. It was pretty technical and mostly centering around the audio engine performances. I'll look again. |
|
|
|
Post by clonewar on Nov 11, 2014 23:18:04 GMT -6
You might be thinking of the 3DAudioInc site? www.3daudioinc.com/3db/forumdisplay.php?15-Awesome-DAW-SUM-ComparisonI used to wonder about this and ran my own tests. I found that when the same tracks are brought into different DAWs, and they're carefully setup the same way (faders, pan, pan law, etc), and bounced with no dither, that the bounces from each DAW are bit identical and null to infinity. I tested with PT, Logic, Cubase, Sonar, Studio One, and SAW Studio. When you start adding plugins, automation, etc, things can change, and delay compensation has already been talked about. Nah, that wasn't it. it was a site where you had pull down menus with different DAWs and settings and you could compare charts/graphs of frequency sweeps, group delays, etc. It was pretty technical and mostly centering around the audio engine performances. I'll look again. Interesting, I don't think I've ever heard of that site. |
|
|
|
Post by clonewar on Nov 11, 2014 23:27:10 GMT -6
This is semi-OT, but I miss using my Paris system. It looked and sounded great and the future was bright and shiny.. Now all DAWs sound the same. Back-ish on topic. I did a session with Doug Sax a couple of years ago and he basically said the same thing that clonewar is saying. The same audio used in any given DAW unprocessed is pretty much going to be the exact same... it's how they handle certain plugins, the math of doing fades, etc. that causes some differences. I have heard (both read and actually heard) that Pro Tools does sound better than Ableton but that's most likely due to the fact that Ableton is usually perpetually in time-warping mode so there are artifacts. Don't use it but saw a demo which seemed to confirm that. You mentioned the math of doing fades and that reminded me that there can be differences in fader taper too. That's why it's really important to go by the DB reading of track faders (or leave all faders at zero) when you're comparing DAW engines. I know of at least one DAW (SAW Studio) that let's you customize the fader taper, so you could set the 'meaty' part of the fader, say 3/4 the way down from zero DB, to only cover a few DB, and then taper off really fast from there down. Ableton Live is kind if a special case because of its audio warping engine like you mentioned. |
|
|
|
Post by clonewar on Nov 11, 2014 23:32:51 GMT -6
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 12, 2014 0:58:51 GMT -6
You might be thinking of the 3DAudioInc site? www.3daudioinc.com/3db/forumdisplay.php?15-Awesome-DAW-SUM-ComparisonI used to wonder about this and ran my own tests. I found that when the same tracks are brought into different DAWs, and they're carefully setup the same way (faders, pan, pan law, etc), and bounced with no dither, that the bounces from each DAW are bit identical and null to infinity. I tested with PT, Logic, Cubase, Sonar, Studio One, and SAW Studio. When you start adding plugins, automation, etc, things can change, and delay compensation has already been talked about. Nah, that wasn't it. it was a site where you had pull down menus with different DAWs and settings and you could compare charts/graphs of frequency sweeps, group delays, etc. It was pretty technical and mostly centering around the audio engine performances. I'll look again. Svart, could it be you mix it up with the SRC comparison at src.infinitewave.ca/? |
|
|
|
Post by M57 on Nov 12, 2014 5:56:02 GMT -6
Sweet.. Actually, I'm still sitting on all of my Paris Gear after some 10 years of not using it. I know it's an active community, but out of the box Paris is all but unusable, and you have to be super-geeky to get it to work with the upgrades and plugs people are coming up with. I started on that journey with limited success, but unfortunately, it turns out I'm just a geek-wannabe. I really should sell it ..or perhaps figure out how to use it for tracking. |
|
But, it works on a simple principle--it tells the host it needs some huge amount of latency--say a full second (whatever it's max is)....and then internally allows you to adjust from 0-1sec in milliseconds. It's intentionally "lying" to the host. If you put it into Cubase today, it will not work--because Cubase won't accept their "lie in reporting"--it will test it and compensate for whatever you set the latency to--ie, the actual latency of the plug in.
