- Messages
- 4,743
So I duplicated a mix I did in Reaper, as closely as I could in both Cubase Pro 13 and Studio One version 5.0 (which is the only version I have right now)
Now the song is fairly dense. A mixture of audio recordings, effects processing, and a VST instrument for drums. It is hopefully fairly typical of a songwriter session you'd come across. There are around 155 plugins. There is a selection of these across the entire project:
I also used whatever bog standard compressor came with each DAW, as well as whatever bog standard trim/mixtool utility type of attenuation plugin comes with them too.
In Reaper I used ReaTune for some vocals. In Cubase 13 Pro I used the built in VariAudio pitch correction. In Studio One I didn't pitch correct.
The track count is the same in all projects, and each track uses the same plugins and presets within the plugins. But the sessions aren't quite 1:1 because of some small differences like the pitch correction I mentioned. There's also differences in terms of the automation of effects in each DAW.
All DAW's were at 48kHz sample-rate, and a buffer size of 64.
In Reaper the audio thread priority is set to time critical, the thread priority is set to highest, with the default behaviour, and anticipative FX processing is enabled.
So here's the Reaper session:
As you can see, I'm getting CPU usage around 35%, with the CPU running at 4.27GHz. I get very few audio dropouts, and can playback the project smoothly from start to finish. On the rare occasion I get audio dropouts, it tends to be within the first few seconds after hitting play.
Here is the Cubase 13 Pro session:
Now in Cubase, ASIOGUARD is enabled and set to low. I get a CPU usage of around 19% at 4.14GHz. However if I disable ASIOGUARD, the CPU goes up to 40%, but the audio output is extremely glitchy - very unusable without ASIOGUARD. I don't know why. I should also mention that the Steinberg power scheme is enabled. Disabling it makes no difference.
Here is the Studio One v5 session:
I'm getting around 23% at 4.35GHz. Dropout protection is set to minimum. However if I set dropout protection to ANY of the other levels, then the project is unlistenable, with similar glitching as I get with Cubase. But at minimum, playback is very smooth and there are no concerns.
This isn't exactly a scientific test. But I was curious what my experience would be. Mostly because I cannot stand to look at any more crappy Reaper themes that simply don't (and indeed cannot) address the workflow issues I have with it.
I'd been working towards moving over to Cubase 13 for most of the year, but haven't done any large projects in it yet. So this was interesting. Another interesting thing to me was, I haven't used Studio One in anger since 2021 when I mixed my bands last album. But I launched it, and was zipping around and using the program without a care in the world - no confusion, no moments of forgetfulness, I still remembered exactly how to do everything. I think the workflow in Studio One is actually excellent. Really excellent. More excellent than excellent.
Anyway......
I don't fully understand what the differences are in playback engines between Reaper, Studio One, and Cubase 13. I find it very puzzling that in Cubase 13 I simply cannot disable ASIOGUARD without experiencing bad performance. But contrary to that, in Studio One, using "minimum" as my dropout protection setting yielded very good performance. But any other setting put me back into bad performance land. These two experiences don't exactly align - I don't understand why I get opposite experiences across these two DAW's.
I also don't fully understand what the Reaper equivalent of these settings would be, if indeed it exists at all.
On the whole, I'm yet to make a determination about which DAW offers me the best performance for my particular workflow.
My system specs:
CPU: AMD Ryzen 5950X
Motherboard: Gigabyte b550 Vision-D (with Thunderbolt)
Drives: A selection of Samsung and Intel SSD drives, with two M2 drives also.
GPU: MSI GeForce RTX 4090 VENTUS 3X OC 24G
RAM: 32GB Crucial Ballistix RGB DDR4
Audio Interface: Presonus Quantum Thunderbolt
Audio Interface: Antelope Audio Discrete 8 Pro SC (WC'd and ADAT'd into the Quantum)
My GPU utilises the maximum power setting. Windows is setup to use the high performance power plan, with tweaks to ensure that USB selective suspend is disabled, and that the mininum and maximum processor states are 100%. I've also fully disabled c-states in my BIOS.
My machine is connected to the internet via ethernet. But I also sometimes share a hotspot with my laptop via WIFI. But this was turned off at the time of the tests (although the network device was not disabled)
I'm going to be replacing the two audio interfaces for an RME UFXII or UFXIII. So it's good to do these tests to establish a benchmark.
I'm curious what people think about this?
Now the song is fairly dense. A mixture of audio recordings, effects processing, and a VST instrument for drums. It is hopefully fairly typical of a songwriter session you'd come across. There are around 155 plugins. There is a selection of these across the entire project:
- EZDrummer3
- T-Racks 5 Black 76
- T-Racks 5 Comprexxor
- T-Racks 5 Saturator X
- DMG Audio TrackComp
- Line6 Helix Native
- Fractal Audio Cab-Lab 3 or 4
- Fabfilter Pro Q-3
- Boz Big Clipper
- ValhallaDSP Vintage Verb
- ValhallaDSP Room
- ValhallaDSP Delay
- ValhallaDSP Supermassive
- Izotope RX-7 De-esser
I also used whatever bog standard compressor came with each DAW, as well as whatever bog standard trim/mixtool utility type of attenuation plugin comes with them too.
In Reaper I used ReaTune for some vocals. In Cubase 13 Pro I used the built in VariAudio pitch correction. In Studio One I didn't pitch correct.
The track count is the same in all projects, and each track uses the same plugins and presets within the plugins. But the sessions aren't quite 1:1 because of some small differences like the pitch correction I mentioned. There's also differences in terms of the automation of effects in each DAW.
All DAW's were at 48kHz sample-rate, and a buffer size of 64.
In Reaper the audio thread priority is set to time critical, the thread priority is set to highest, with the default behaviour, and anticipative FX processing is enabled.
So here's the Reaper session:
As you can see, I'm getting CPU usage around 35%, with the CPU running at 4.27GHz. I get very few audio dropouts, and can playback the project smoothly from start to finish. On the rare occasion I get audio dropouts, it tends to be within the first few seconds after hitting play.
Here is the Cubase 13 Pro session:
Now in Cubase, ASIOGUARD is enabled and set to low. I get a CPU usage of around 19% at 4.14GHz. However if I disable ASIOGUARD, the CPU goes up to 40%, but the audio output is extremely glitchy - very unusable without ASIOGUARD. I don't know why. I should also mention that the Steinberg power scheme is enabled. Disabling it makes no difference.
Here is the Studio One v5 session:
I'm getting around 23% at 4.35GHz. Dropout protection is set to minimum. However if I set dropout protection to ANY of the other levels, then the project is unlistenable, with similar glitching as I get with Cubase. But at minimum, playback is very smooth and there are no concerns.
This isn't exactly a scientific test. But I was curious what my experience would be. Mostly because I cannot stand to look at any more crappy Reaper themes that simply don't (and indeed cannot) address the workflow issues I have with it.
I'd been working towards moving over to Cubase 13 for most of the year, but haven't done any large projects in it yet. So this was interesting. Another interesting thing to me was, I haven't used Studio One in anger since 2021 when I mixed my bands last album. But I launched it, and was zipping around and using the program without a care in the world - no confusion, no moments of forgetfulness, I still remembered exactly how to do everything. I think the workflow in Studio One is actually excellent. Really excellent. More excellent than excellent.
Anyway......
I don't fully understand what the differences are in playback engines between Reaper, Studio One, and Cubase 13. I find it very puzzling that in Cubase 13 I simply cannot disable ASIOGUARD without experiencing bad performance. But contrary to that, in Studio One, using "minimum" as my dropout protection setting yielded very good performance. But any other setting put me back into bad performance land. These two experiences don't exactly align - I don't understand why I get opposite experiences across these two DAW's.
I also don't fully understand what the Reaper equivalent of these settings would be, if indeed it exists at all.
On the whole, I'm yet to make a determination about which DAW offers me the best performance for my particular workflow.
My system specs:
CPU: AMD Ryzen 5950X
Motherboard: Gigabyte b550 Vision-D (with Thunderbolt)
Drives: A selection of Samsung and Intel SSD drives, with two M2 drives also.
GPU: MSI GeForce RTX 4090 VENTUS 3X OC 24G
RAM: 32GB Crucial Ballistix RGB DDR4
Audio Interface: Presonus Quantum Thunderbolt
Audio Interface: Antelope Audio Discrete 8 Pro SC (WC'd and ADAT'd into the Quantum)
My GPU utilises the maximum power setting. Windows is setup to use the high performance power plan, with tweaks to ensure that USB selective suspend is disabled, and that the mininum and maximum processor states are 100%. I've also fully disabled c-states in my BIOS.
My machine is connected to the internet via ethernet. But I also sometimes share a hotspot with my laptop via WIFI. But this was turned off at the time of the tests (although the network device was not disabled)
I'm going to be replacing the two audio interfaces for an RME UFXII or UFXIII. So it's good to do these tests to establish a benchmark.
I'm curious what people think about this?
![[Nathan]](/data/avatars/m/0/613.jpg?1678404614){kind=avatar}
