|
Post by Quint on Oct 11, 2024 15:05:45 GMT -6
If I was starting over I’d be hard pressed to not use Studio One. They got a lot of things right with that DAW. About 3 years ago I did a deep extended trial of S1…I was very close to switching but at the time there were still 1 or 2 things it didn’t do well that were deal breakers for me. What were those 1 or 2 things?
|
|
|
Post by Quint on Oct 11, 2024 15:09:28 GMT -6
I've been holding out for the last couple of years or so, always thinking that UA would eventually release a new Apollo with ARM. But UA has now shown us their cards with this new weak sauce Apollo. Message received UA. Fool me once... I may not be making the move today, or even tomorrow. I'm gonna take my time with this decision. But, sooner or later, I'm probably dumping Luna and the Apollos. I'm intrigued by the dual buffer thing, particularly in S1. It more or less offers the same benefits of using hardware monitoring, but without having to be tied to any one company's interface or any one company's plugins or any one company's marketing and bullshit. Plus, I would no longer be waiting on UA to implement features I want/need in Luna. For me, getting to this point is like some huge and painful pimple just popped. 😉 I'm just so tired of UA's nonsense. Maybe we should have dumped UA sooner, but at least UA has now given us a clear message on where they are headed, so we can make decisions accordingly. I mean, UA will never SAY that this is where they are headed, because they know no other way than to obfuscate. But their actions are speaking very loudly for them. I kind of hate to give up on the DSP monitoring thing, but UA is making it untenable. I'll be curious to see what you think about S1. My version of this is MOTU supporting AVB. They say that they're going to continue to support it but their most recent release doesn't and all their AVB units are getting long in the tooth (and most are out of stock). That's THE reason I'm on MOTU. So I hear you. If DSP isn't a thing for Apollo, then what's the point? Yeah, I was really kicking around going with Motu AVB a few years ago. It had a lot to like. I still have no idea why they went away from it. AVB would have been the biggest reason I would have gone with Motu. Yeah, if UA is abandoning DSP, I'm not about to continue to put up with their bullshit and all of the other compromises. It was the only thing keeping me around.
|
|
|
Post by gravesnumber9 on Oct 11, 2024 15:14:54 GMT -6
Well the MOTU rep who does Drew's job (way better than Drew of course) on GS basically said that they are moving away from AVB on the 828 (duh, they already did) but that they plan to continue expanding their pro-line AVB offerings.
I'm in wait and see on that though. I'm sure that's their intent. But it's been many years since they've released anything for the AVB line.
However, (man, this is just like what you're saying with Apollo) it's also been many years since they've updated their pro lines at all. So the big turning point for me on MOTU will be when they release the next pro level upgrade will it support AVB. If not, that's the writing on the wall.
As it stands, MOTU AVB is a pretty insane cost/value prop.
|
|
|
Post by sirthought on Oct 11, 2024 15:16:31 GMT -6
I think 'untenable' is a bit of an overstretch. I'm upset about the constant pricing gymnastics and switching of policies that makes me consider dumping them. But the Apollo tracking with DSP still works great in its original design.
I don't know which plugins you feel are necessary on the way in, but you can probably track with the legacy EQs and compressors and still get excellent tracks. There are a good number of Unison pres now. If you're into those, you have choices.
It's beyond stupid that things like the newer Waterfall rotary speaker, amp sims, and verve analog machines aren't UAD-2.
But I was tracking bands fine without those before. And I could probably never buy a UAD plugin again and it would still work for years.
|
|
|
Post by gravesnumber9 on Oct 11, 2024 15:21:29 GMT -6
I think 'untenable' is a bit of an overstretch. I'm upset about the constant pricing gymnastics and switching of policies that makes me consider dumping them. But the Apollo tracking with DSP still works great in its original design. I don't know which plugins you feel are necessary on the way in, but you can probably track with the legacy EQs and compressors and still get excellent tracks. There are a good number of Unison pres now. If you're into those, you have choices. It's beyond stupid that things like the newer Waterfall rotary speaker, amp sims, and verve analog machines aren't UAD-2. But I was tracking bands fine without those before. And I could probably never buy a UAD plugin again and it would still work for years. When Waterfall and the rest of the new stuff like that was coming out it almost had me tempted to go back to Apollo (I'm a keyboard guy) and then I saw... wait, huh? Big whiff on that UAD.
|
|
|
Post by sirthought on Oct 11, 2024 15:33:04 GMT -6
When Waterfall and the rest of the new stuff like that was coming out it almost had me tempted to go back to Apollo (I'm a keyboard guy) and then I saw... wait, huh? Big whiff on that UAD. Big whiff is right. It's exactly what Quint means with a lack of dedication to the DSP format. And these new interfaces have a couple of noteworthy features, but things people have been asking for for years is completely ignored and that lack of innovation and development is disappointing. Especially considering they made a whole line of other interfaces that have nothing to do with the approach they built their name on (DSP and chaining units with Thunderbolt). I'm more interested in improvements of converters and sound, and more professional connection options.
|
|
|
Post by Quint on Oct 11, 2024 15:42:25 GMT -6
I think 'untenable' is a bit of an overstretch. I'm upset about the constant pricing gymnastics and switching of policies that makes me consider dumping them. But the Apollo tracking with DSP still works great in its original design. I don't know which plugins you feel are necessary on the way in, but you can probably track with the legacy EQs and compressors and still get excellent tracks. There are a good number of Unison pres now. If you're into those, you have choices. It's beyond stupid that things like the newer Waterfall rotary speaker, amp sims, and verve analog machines aren't UAD-2. But I was tracking bands fine without those before. And I could probably never buy a UAD plugin again and it would still work for years. Untenable for ME. I should have been more clear about that. If we were just talking about using UA plugins for tracking, then yeah, I may not care as much whether the quality of some of the plugins I'm using during tracking are top quality, because I'd theoretically be using something else during mixing to replace them. And a lot of those UAD DSP-only plugins are pretty old and bested by other plugins these days, so I would be looking to replace them in that scenario . However, the appeal for me with Luna/Apollos was the auto DSP switching. Mixing is tracking. Tracking is mixing. It's the same plugin for both scenarios. You fluidly move between tracking and mixing, and start shaping, using the same plugins from start to finish, provided you use those particular plugins which UA has made available in both DSP and native format. Great workflow. And, yes, this will continue to work for me for now. But what I don't want to happen is to be in the same boat as AAX plugins are now, where development is dead. I don't want to stick with a system where no further development is going to happen. I don't think anyone would argue that the AAX plugins available today can generally compete with the latest generation of plugins. AAX stopped advancing years ago. So, for the same reason that people mixing natively now wouldn't want to be stuck using only old native plugins, five years from now I don't want to be stuck using old UAD DSP plugins which stopped advancing five or more years ago. We're all always looking around for plugins which have raised the bar to another level. This is no different. Maybe I'm a bit of an edge case. I don't know. But for how I wanted to work, which is why I went to Luna, this new Apollo release was a clear line in the sand that things had changed. I don't trust UA at all at this point, and I'm not terribly interested in sticking around to see what other promises they might break or what other policies they might change mid stream.
|
|
|
Post by Quint on Oct 11, 2024 15:54:23 GMT -6
When Waterfall and the rest of the new stuff like that was coming out it almost had me tempted to go back to Apollo (I'm a keyboard guy) and then I saw... wait, huh? Big whiff on that UAD. Big whiff is right. It's exactly what Quint means with a lack of dedication to the DSP format. And these new interfaces have a couple of noteworthy features, but things people have been asking for for years is completely ignored and that lack of innovation and development is disappointing. Especially considering they made a whole line of other interfaces that have nothing to do with the approach they built their name on (DSP and chaining units with Thunderbolt). I'm more interested in improvements of converters and sound, and more professional connection options. Routing, or lack thereof was a big one for me. That was always unnecessarily too limited. I was hoping UA would make improvements there for this new Apollo release, but they didn't.
|
|
|
Post by BenjaminAshlin on Oct 11, 2024 15:56:53 GMT -6
There is some misunderstanding about how DAWs and buffering work in low latency.
S1, Logic, and Cubase all utilize hybrid buffers, but each does it differently.
While hardware does control buffer size, DAWs manage this by pre-processing unarmed tracks to align with armed/monitored tracks and therefore filling the buffer at the lowest latency.
Additionally, both S1 and Cubase offer an option to constrain delay compensation (known as Z monitoring in S1), which disables high-latency plugins during recording. Giving you the lowest latency monitoring path possible.
In these scenarios, busses with delay and reverb should not contribute any additional latency to the audio path. Cubase has a feature in the mixer to display the latency of each track in milliseconds so you can easily identify tracks with high latency.
Also, many are unaware that DSP plugins like those from Apollo add latency. For example, basic/older UAD plugins introduce 33-55 samples of latency, while the precision multiband plugin adds around 1500 samples.
|
|
|
Post by Tbone81 on Oct 11, 2024 16:54:40 GMT -6
If I was starting over I’d be hard pressed to not use Studio One. They got a lot of things right with that DAW. About 3 years ago I did a deep extended trial of S1…I was very close to switching but at the time there were still 1 or 2 things it didn’t do well that were deal breakers for me. What were those 1 or 2 things? 1) I couldn’t figure out a good way to setup a monitor only bus. Cubase has a dedicated monitor bus where you can throw on plugins like Arc or Sonarworks, control volume, headphone mixes etc. There might be a why to setup the routing in S1 to do the same but I couldn’t get it to work like I wanted. 2) The way S1 handled “rendering in place” was weird to me. I forget exactly how they handle it but if you render an audio file in place or render a midi file as audio it would hide the original track in a way that made it hard to go back. In Cubase it’s super easy to render either Audio or MIDI and then if you want to make changes you still have the archived old track to go back to. There were a few other minor things I don’t quite remember, related to automation and mixing in general, but those two were the biggest. It would’ve been a non issue if I had started on S1 but I’ve been using Cubase for 20 years so….im pretty comfortable in my ways.
|
|
|
Post by Quint on Oct 11, 2024 19:40:39 GMT -6
What were those 1 or 2 things? 1) I couldn’t figure out a good way to setup a monitor only bus. Cubase has a dedicated monitor bus where you can throw on plugins like Arc or Sonarworks, control volume, headphone mixes etc. There might be a why to setup the routing in S1 to do the same but I couldn’t get it to work like I wanted. 2) The way S1 handled “rendering in place” was weird to me. I forget exactly how they handle it but if you render an audio file in place or render a midi file as audio it would hide the original track in a way that made it hard to go back. In Cubase it’s super easy to render either Audio or MIDI and then if you want to make changes you still have the archived old track to go back to. There were a few other minor things I don’t quite remember, related to automation and mixing in general, but those two were the biggest. It would’ve been a non issue if I had started on S1 but I’ve been using Cubase for 20 years so….im pretty comfortable in my ways. What did you like better about S1 versus Cubase?
|
|
|
Post by benqbasic on Oct 11, 2024 20:54:39 GMT -6
1) I couldn’t figure out a good way to setup a monitor only bus. Cubase has a dedicated monitor bus where you can throw on plugins like Arc or Sonarworks, control volume, headphone mixes etc. There might be a why to setup the routing in S1 to do the same but I couldn’t get it to work like I wanted. 2) The way S1 handled “rendering in place” was weird to me. I forget exactly how they handle it but if you render an audio file in place or render a midi file as audio it would hide the original track in a way that made it hard to go back. In Cubase it’s super easy to render either Audio or MIDI and then if you want to make changes you still have the archived old track to go back to. There were a few other minor things I don’t quite remember, related to automation and mixing in general, but those two were the biggest. It would’ve been a non issue if I had started on S1 but I’ve been using Cubase for 20 years so….im pretty comfortable in my ways. What did you like better about S1 versus Cubase? I use both, I wish s1 was as fully fledged as cubase. the pitch correction, time alignment, spectalsound, and other small things are better. Cubase is also better CPU usage. Both are great though.
|
|
|
Post by thehightenor on Oct 12, 2024 2:08:00 GMT -6
Yes, Cubase has a dual buffer. I leave my buffers at 128 and Cubase takes care of the rest - I never touch them again no matter how big a mix gets. I could probably run at 64 with my 24 core 13900K workstation but 128 covers me for everything and for tracking I use an analog monitoring mixer for true zero latency tracking.
|
|
|
Post by BenjaminAshlin on Oct 12, 2024 2:24:31 GMT -6
Yes, Cubase has a dual buffer. I leave my buffers at 128 and Cubase takes care of the rest - I never touch them again no matter how big a mix gets. I could probably run at 64 with my 24 core 13900K workstation but 128 covers me for everything and for tracking I use an analog monitoring mixer for true zero latency tracking. Same, i set it to 64 and never think about it again.
|
|
|
Post by Quint on Oct 12, 2024 9:39:13 GMT -6
Yes, Cubase has a dual buffer. I leave my buffers at 128 and Cubase takes care of the rest - I never touch them again no matter how big a mix gets. I could probably run at 64 with my 24 core 13900K workstation but 128 covers me for everything and for tracking I use an analog monitoring mixer for true zero latency tracking. So are you not using software monitoring during tracking? You mentioned that you use analog. I guess I'm confused. Am I understanding you to say that, yes Cubase has dual buffers, but no you don't use them, because you use analog monitoring?
|
|
|
Post by svart on Oct 12, 2024 10:44:21 GMT -6
There is some misunderstanding about how DAWs and buffering work in low latency. S1, Logic, and Cubase all utilize hybrid buffers, but each does it differently. While hardware does control buffer size, DAWs manage this by pre-processing unarmed tracks to align with armed/monitored tracks and therefore filling the buffer at the lowest latency. Additionally, both S1 and Cubase offer an option to constrain delay compensation (known as Z monitoring in S1), which disables high-latency plugins during recording. Giving you the lowest latency monitoring path possible. In these scenarios, busses with delay and reverb should not contribute any additional latency to the audio path. Cubase has a feature in the mixer to display the latency of each track in milliseconds so you can easily identify tracks with high latency. Also, many are unaware that DSP plugins like those from Apollo add latency. For example, basic/older UAD plugins introduce 33-55 samples of latency, while the precision multiband plugin adds around 1500 samples. This is what I was taking about earlier, that hardware monitoring isn't "zero latency". But I guess what I tried to convey is that other than certain plugins, most of them don't add appreciable latency in native mode and I don't think hardware monitoring is necessary in most situations, especially if "dual buffering" is simply disabling plugs with high latency.
|
|
|
Post by Quint on Oct 12, 2024 10:55:33 GMT -6
There is some misunderstanding about how DAWs and buffering work in low latency. S1, Logic, and Cubase all utilize hybrid buffers, but each does it differently. While hardware does control buffer size, DAWs manage this by pre-processing unarmed tracks to align with armed/monitored tracks and therefore filling the buffer at the lowest latency. Additionally, both S1 and Cubase offer an option to constrain delay compensation (known as Z monitoring in S1), which disables high-latency plugins during recording. Giving you the lowest latency monitoring path possible. In these scenarios, busses with delay and reverb should not contribute any additional latency to the audio path. Cubase has a feature in the mixer to display the latency of each track in milliseconds so you can easily identify tracks with high latency. Also, many are unaware that DSP plugins like those from Apollo add latency. For example, basic/older UAD plugins introduce 33-55 samples of latency, while the precision multiband plugin adds around 1500 samples. This is what I was taking about earlier, that hardware monitoring isn't "zero latency". But I guess what I tried to convey is that other than certain plugins, most of them don't add appreciable latency in native mode and I don't think hardware monitoring is necessary in most situations, especially if "dual buffering" is simply disabling plugs with high latency. I don't think anyone was questioning that zero latency is actually truly zero latency though. I think we all know that it just means "low" latency. Though I would agree that it is an unfortunate use of the term, that could be confusing to sime people. But at least in this thread, I think it's clear that we are just talking about low latency. As for dual buffering, there does still seem to be some confusion here about what that means. It's not just automatic disabling of certain plugins. There is more to it, from what I've seen. I'm willing to be convinced otherwise, as I'll admit to the idea of dual buffers being something I'm just now starting explore in earnest. That said, from what I'm reading/watching about dual buffers, the input (recording) input is in fact a different faster path, because of a separate lower buffer, than the playback (if you set it up that way). This is why this approach has some benefit over one single buffer for everything. You're not asking all tracks to try to playback at a super low buffer, just those that are record enabled. The playback tracks can run at a slower, more stable buffer. Win, win. Watch some videos on S1. Also, I have no reason to disbelieve what Gravesnumber has been saying on the subject. He uses it all the time. www.soundonsound.com/techniques/studio-one-buffers-low-latency
|
|
|
Post by Johnkenn on Oct 12, 2024 12:26:51 GMT -6
|
|
|
Post by Quint on Oct 12, 2024 15:46:52 GMT -6
Alright. I'm downloading the S1 demo now. We'll see how it goes.
|
|
|
Post by popmann on Oct 12, 2024 17:36:24 GMT -6
I was summoned? Is it jsut "Doesn't Cubase have dual buffers?" No..it has three now. I'm just not sure what actually happens on the one you set as the hardware buffer size. I leave mine on 512@96khz. Full time. Cubase has always run a fraction (think I estimated 1/4 once?) for input enabled instruments and audio. So my setting the hardware buffer to 512 means input buffer is 128@96khz. But, in recent versions ASIO Guard gets involved which is a HUGE multiple processing buffer. Set to sm/med/lg maybe? I keep it on Large which I'm pretty sure I read was 4096, but it might end up being a multiple of the set buffer--I don't know.
Once you reduce the process buffer to 128 and lower, the input buffer goes away. So, someone running it at 128....has more latency on an instrument/input than someone running 256(because it will run closer to 64sample on an input)....AND having half the CPU DSP power.
NOTHING I'm saying ever makes you press for the constrain delay compensation of "low latency mode" as Logic calls it...In fact, those all, IME, are a way to ensure your track's recording compensation is fucked. Albeit not by a large scale...but, that isn't a workflow solution. That was always intended as a solution for when you have a full mix of DSP going--so the compensation is incurring a lot of latency...and you have to add a last minute overdub...they all work a little differently--but, they're basically killing the compensation accuracy (and sometimes disabling DSP all together) in order to BREIFLY get you low latency on an input. It was pointed out above that it's not the same thing--I want to not just second that, but fucking hit that note HARD. THAT is a jury rig that's supposed to save your ass...NOT a way to track all overdubs with lower latency.
If there was something specific I missed...I mean was it about analog monitoring? Hardware digital? Using software reverbs to monitor using analog OR digital hardware mixers on your interface?
|
|
|
Post by gravesnumber9 on Oct 12, 2024 18:30:26 GMT -6
What were those 1 or 2 things? 1) I couldn’t figure out a good way to setup a monitor only bus. Cubase has a dedicated monitor bus where you can throw on plugins like Arc or Sonarworks, control volume, headphone mixes etc. There might be a why to setup the routing in S1 to do the same but I couldn’t get it to work like I wanted. 2) The way S1 handled “rendering in place” was weird to me. I forget exactly how they handle it but if you render an audio file in place or render a midi file as audio it would hide the original track in a way that made it hard to go back. In Cubase it’s super easy to render either Audio or MIDI and then if you want to make changes you still have the archived old track to go back to. There were a few other minor things I don’t quite remember, related to automation and mixing in general, but those two were the biggest. It would’ve been a non issue if I had started on S1 but I’ve been using Cubase for 20 years so….im pretty comfortable in my ways. 1) has been update maybe? There is now a Cubase style listen bus. 2) definitely weird, I just record out to a new track
|
|
|
Post by veggieryan on Oct 12, 2024 19:18:58 GMT -6
As I understand it, “dual buffer” means you have one large buffer for playback and another small buffer for low latency monitoring on enabled audio tracks and virtual instruments. That also means that the system might have to disable and bypass certain plugins on the tracks you are live monitoring if they have extremely long latency. Disabling plugins that have high latency is what Logic refers to as “Low latency mode” but this does not mean that Logic has “dual buffers” like Studio One. I am currently demoing Studio One version 7 and I think all DAWs should work this way, there’s not much downside except for the difficulty of coding it. As for Cubase it seems like it has something similar for windows called ASIO-Guard but not Mac but I need to demo it on Mac to see for sure because their documentation is not so great on this subject. The cubase docs mention something called “Constrain Delay Compensation” that says some VST3 plugins have a “Live” button which is a low latency mode for that plugin. So if you activate constrain delay compensation it automatically enables the live mode for *some plugins.
Anyways, it’s 2024 and latency is still a confusing problem. Sigh.
|
|
|
Post by Quint on Oct 12, 2024 19:49:12 GMT -6
As I understand it, “dual buffer” means you have one large buffer for playback and another small buffer for low latency monitoring on enabled audio tracks and virtual instruments. That also means that the system might have to disable and bypass certain plugins on the tracks you are live monitoring if they have extremely long latency. Disabling plugins that have high latency is what Logic refers to as “Low latency mode” but this does not mean that Logic has “dual buffers” like Studio One. I am currently demoing Studio One version 7 and I think all DAWs should work this way, there’s not much downside except for the difficulty of coding it. As for Cubase it seems like it has something similar for windows called ASIO-Guard but not Mac but I need to demo it on Mac to see for sure because their documentation is not so great on this subject. The cubase docs mention something called “Constrain Delay Compensation” that says some VST3 plugins have a “Live” button which is a low latency mode for that plugin. So if you activate constrain delay compensation it automatically enables the live mode for *some plugins. Anyways, it’s 2024 and latency is still a confusing problem. Sigh. This is also my understanding of dual buffers. I hadn't looked into Logic yet, but if that's how low latency mode actually works in Logic, I'm probably not interested in Logic, as long as S1 is an option.
|
|
|
Post by Tbone81 on Oct 12, 2024 20:30:49 GMT -6
1) I couldn’t figure out a good way to setup a monitor only bus. Cubase has a dedicated monitor bus where you can throw on plugins like Arc or Sonarworks, control volume, headphone mixes etc. There might be a why to setup the routing in S1 to do the same but I couldn’t get it to work like I wanted. 2) The way S1 handled “rendering in place” was weird to me. I forget exactly how they handle it but if you render an audio file in place or render a midi file as audio it would hide the original track in a way that made it hard to go back. In Cubase it’s super easy to render either Audio or MIDI and then if you want to make changes you still have the archived old track to go back to. There were a few other minor things I don’t quite remember, related to automation and mixing in general, but those two were the biggest. It would’ve been a non issue if I had started on S1 but I’ve been using Cubase for 20 years so….im pretty comfortable in my ways. What did you like better about S1 versus Cubase? I loved the drag n drop nature of everything. The midi implementation was great (but not necessarily better than Cubase), being able to drag a midi file to an audio track and it automatically renders, having a drum sequencer built into the midi drum edit window, vst instruments were solid as were the stock plugins. Some DAWs, like PT, are more geared towards recording/mixing/editing/post production etc Real engineering work. Others, like Ableton, are really geared towards performance and songwriting. Logic and Cubase are in the middle, and I feel like S1 is in a sweet spot on that same continuum.
|
|
|
Post by Quint on Oct 12, 2024 22:59:14 GMT -6
What were those 1 or 2 things? 1) I couldn’t figure out a good way to setup a monitor only bus. Cubase has a dedicated monitor bus where you can throw on plugins like Arc or Sonarworks, control volume, headphone mixes etc. There might be a why to setup the routing in S1 to do the same but I couldn’t get it to work like I wanted. 2) The way S1 handled “rendering in place” was weird to me. I forget exactly how they handle it but if you render an audio file in place or render a midi file as audio it would hide the original track in a way that made it hard to go back. In Cubase it’s super easy to render either Audio or MIDI and then if you want to make changes you still have the archived old track to go back to. There were a few other minor things I don’t quite remember, related to automation and mixing in general, but those two were the biggest. It would’ve been a non issue if I had started on S1 but I’ve been using Cubase for 20 years so….im pretty comfortable in my ways. Hey, as I've been learning about S1, I think what you're maybe seeking for a monitoring bus is called a Listen Bus in S1.
|
|