Performance Test C 10.0.20 vs Logic Pro 10.4.5 vs Studio One 4.5

If I recall and I do, it was recommended to use a secondary drive for all Cubase projects since way back when. Im guessing this is an old code area in Cubase

This points to the fact that you are more than likely correct, Cubase is not optimized well to run off an internal HD. I bet an SSD would fair better?

I have and use all three DAWS you are quoting there. I haven’t had any reason to test them though as I never use the internal drive here for any Projects

Carry on

It would be interesting for you to do the test using the Mac sound card and not the Steinberg interface just to see what ya get there.

I don’t know, I find this interesting. I’ve recently built a new PC DAW and even though I’ve upped my memory, my cores, moved from Win7 to Win10, and now run all SSDs, (internally), I do have a better Cubase experience and more ‘ease of use’ with the added power - but - eh, the Average Processing Load bar still reacts more than I had hoped for. Sure, with my 4 core Win7 DAW the Processing Load bar did go higher than with my new 6 core processor but plugins like Ozone 8 Advanced still suck some juice. I think this test is accurate.

You have 37 tracks in your cubase session, but only 2 audio files, is that right? While I don’t know what your problem is or why only Cubase is showing the i/o overload, I can tell you it is unusual for Cubase to have a disk i/o overload on a session like that and I think that’s probably why you haven’t received many responses to help you solve your problem. It would not be unusual however, for an 8 year old hard drive to have problems.

Yes, and everyone knows that hard drives will first start to perform particularly bad with Cubase before the effects show up elsewhere - most hard drive diagnostics run a version of Cubase for this very purpose… right? :slight_smile:

Nope. Not on my system… :wink:

maybe you do not understand the CPU meters correctly, and you can use Asio Guard for better performance.

please read this article:

Another nice Video “Asio vs. Performance Metering”:

Here on my OSX:
Cubase vs. Logic very close results (when you use the same third-party Plugins!)
But If I use Asio Guard, Cubase 10 runs a bit better than Logic 10.


C.

But it is not a CPU issue . Its a disk reading issue! the orgianal post clearly shows a problem with disk overload.

You’re asking if there could be a problem with one particular file on a hard drive? Yes. Could this only affect Cubase? Yes, if the other tests were done with copies of that file instead of the same file. I take your point that this seems unlikely, but without more details about the OP’s problem and how the tests were performed, that’s probably as close to an explanation for what he’s observing as he’s going to get.

You made a comment that it IS unusual for Cubase to be at fault, but NOT unusual for a hard drive to be a fault, i’m pointing out that it’s most likely to NOT be a hard drive failure so they don’t go spending out on a hard disk based on some janky advice.

On my system cubase performs worse (heavier on cpu) than Reaper or StudioOne. Tested both on Mac Book and on desktop (specs in the signature)

Here’s the facts:

3 different DAWS

exact same project.

2 audio tracks duplicated as many times possible with one instance of Scheps Omni channel on each track

1 Groove Agent vst track

Optimum settings selected to produce maximum performance on each DAW

ALL DAWs using the same internal 7200rpm drive.

Results quite different.

Cubase 10.0.30 ( just updated ) experiences HD overloads while the other 2 DAWs do not and can run quite a few more audio tracks than Cubase.

Here are videos:

Logic Pro 10.4.5

Studio One 4.5

Cubase 10.0.30

Sure anyone can simply suggest I buy a newer computer or swap out the 7200rpm hd for an SSD but that’s not the question in this test.

Do you have automatic hitpoint detection enabled ?
Tried using the backup function in C10 to make a fresh copy in another location?
Tried copying the project to an external drive?
Tried different preload settings in Cubase?
Defragmenting the drive?
How is the memory situation when running the various DAWs?
No backup/update or other task running in the background while testing? has happened to me :slight_smile:

Can you post the project? I don’t dispute the idea that you’re getting poorer performance in Cubase, but I do suspect something else is going on. For instance, when you import audio into Cubase, it puts the audio tracks in “Musical” mode by default, so basically enables time-stretching. I don’t see that little quarter note icon on the audio clips in your screen capture, but you may want to double check that (looks like your audio clips may have been trimmed - not sure if that hides the icon).

So if it were something like that, and someone wasn’t paying attention, or just doesn’t know about that, they may think, “hey, Cubase uses a ton more CPU playing back 20 audio tracks than Reaper!” Then after turning off musical mode they realize, “oh wait, Cubase uses about as much CPU doing the same thing as my other two DAWs.” This exact thing happened on KVR a few days ago. Someone did a compare, Cubase was eating a ton of CPU with 20 audio stems, and the other DAWs weren’t. They had musical mode enabled. Turned it off, Cubase was right there with the other DAWs. I only suggested it to that poster because I recently imported stems and had to turn Musical/time stretching mode off.

Now, I’m not saying that’s definitely what you’re seeing, but what I’m suggesting is, if you post the project, other people can try to replicate your settings and double check your work. Kind of like peer-reviewed studies. Sometimes having another set of eyes doing the same thing can lead to different results or find an issue.

I have multiple DAWs on my computer, Pro Tools, Cubase, Logic, Studio One. I don’t notice any less performance in Cubase. I actually notice a little less performance in Pro Tools on some sessions with lots of virtual instruments. In saying that, it’s a pretty fast computer.

I noticed in your videos that the cpu bar was maxed out in Logic, higher in Studio One than in Cubase. It might be a muti-threading issue. Do you have hypertheading enabled? I remember that was an issue on really old CPUs with Cubase but I’m not sure if it’s still a problem in the newer versions of Cubase. Also, the UR824 interface you have has extra functionality in Cubase. That is a variable. You can delete a file to break the functionality in Cubase for the UR824. I forget which file it is but you might be able to search online to find it.

This is a ridiculous thread, first off, you never run a professional DAW on a single hard drive, Protools for example won’t run properly at all on a single hard drive yet something like Reaper or Mixcraft will do much better.

Would you then say Mixcraft and Reaper are “superior” to Protools?

If you are going to do a test like this, please do it on a professional DAW with at least 2 preferably 3 hard drives. One for OS and programs/plugins, one for samples and one for recording/projects.

It’s not unusual at all for professional DAW programs to stumble when only one hard drive is being used.

I don’t think loading two wave files into multiple tracks warrants a dual drive setup lol. Also, Pro Tools runs fine on a single hard drive for a task like this - i just tried it, have some perspective rather than spreading third party fallacies please as it’s not helpful for anyone - particularly when you’re telling others they’re being ‘ridiculous’.

The second hard drive for Pro Tools is mainly recommended for recording high track counts at once and it’s more controllable putting content on a secondary. Majority of pro-tools users that record do so with multi-mic setups (Drums etc.). That is totally different to this performance test which is CPU focused and not recording any audio whatsoever.

Large majority of people running Cubase are doing so on 2-4 input audio interfaces, i presume you using some much larger amount of inputs to class this as ‘ridiculous’?

I think the main diff here is S1 and Logic both use a dual buffer system for input and output of audio, where as Cubase does not.

You can even use a diff interface in S1 for input and output.

It’s not very well documented, but yes, Cubase, when asio guard is turned on, does use a hybrid buffer system like Logic. That’s why Cubase and Logic generally show similar results in performance tests.

You completely miss my point, the “test” is ridiculous, please don’t claim I said others were ridiculous, that’s not what I said. Just because you have a 2 or 4 input audio device does not mean you run only 2 tracks, if you believe this then you really don’t understand much at all about recording music. Multiple drives isn’t just about recording, your wrong there also. It’s about breaking up the data flow load, so you can have samples coming from one disk, recording going on another and your OS and programs doing what they need to do. Your also wrong thinking that loading up a CPU in the way you have done will tell you anything, because as has already been explained to you, the disk in/out issue cannot be ignored as being the bottleneck when using a single hard drive. You don’t seem to understand much about music recording.

The point you made about needing to run multiple drives is a strong recommendation from a technical standpoint when RECORDING higher quantities of input sources, that’s why Avid etc. recommend it.

That’s why I said unless you’re using high quantity input interfaces that the advice for Pro Tools sensibly applies, please try to understand. Halting playback is one thing, breaks in recording is far more damaging for a professional wanting to minimise downtime and errors.

Multiple drives isn’t just about recording, your wrong there also. It’s about breaking up the data flow load, so you can have samples coming from one disk, recording going on another and your OS and programs doing what they need to do.

Then show me at what point a single drive gets overloaded then, as you seem to think this is such a necessity to break ‘the data flow’, I’ll start you off:-

Average budget drive runs at approx 400MB/sec.
16 bit 44k audio @ 176k/sec = 2,272 tracks

You must have some seriously big projects to be making these ‘ridiculous’ remarks so flippantly?. Loading two audio tracks is not a massive strain on the bus, stop being lead by fallacies that you’ve read somewhere and passing it off as your own expert advice.

So appreciate if you could back these claims up with some objective facts, instead of your childish little assumptions as below…

You don’t seem to understand much about music recording.

Clearly not, maybe an expert such as yourself cares to enlighten me further with these ‘data flow loads’ and why iMac, MacBook Pro, Mac Mini and Laptop users MUST use an external thunderbolt/USB3 drive or else they face problems? Because this is big news for many (currently) happy users!

I’ve only had 25+ years experience as a (failed :confused: ) musician, audio hardware service agent and audio app developer, so I’m all ears for you to explain the three drive requirement from a factual basis, and not just internet hearsay.

If you cant get it, then I fear you do not have the capacity to understand. All you are doing is again, restating false information. Your stated qualifications don’t seem to match to your actual understanding of audio recording and you don’t seem to understand what it is I am saying, your replies don’t match my points. There is no point in me continuing a conversation with you and nobody else in this thread is interested in our spat. So I am happy to leave it there with you.

My point is, to the average user, distributing the data load is more important than running a single hard drive and obsessing about the CPU. A single SSD over a HDD may make that easier, I don’t know. I would like to know how well a mac with a single SSD would handle a typical session, perhaps it is possible.

Typically, when anyone is using a laptop with a single hard drive, all may go smoothly but as they increase their track and plugin count and up VST usage and samples used they face bottlenecks and get juddery performance, poor latency etc. The way to deal with this is not to increase the CPU size but to spread the data load in different ways. Those who use large sample libraries and high track counts/plugins already know this, which is why they tend to switch to a desktop situation over a laptop. Yes you can attach external drives to a laptop and that may be a solution but laptops then have other problems with heat etc.

My typical session might be 15-20 VST instruments, some Kontakt and other libraries, 30 or so tracks and maybe up to 40 plugins or so and then of course I also want to record and playback smoothly with ZERO latency. So this is what I do with a puny 4th generation i5 and 8 gigs of ram. I tried it on a single 7200rpm HDD absolutely not possible, juddery playback, freezing, crashes, unreliable recording and poor latency. Didn’t like it at all (I would like to know if this is possible using a single SSD).

Once I had the sample libraries on one HDD (just upgraded to SSD), the recording/projects on another 7200 HDD and the OS and programs on another SSD then everything went smoothly, using the same CPU and ram. I further achieved zero latency recording by getting the UR44 and using the onboard DSP direct monitoring chip to do that.

I don’t think my system is anything out of the ordinary, it’s the same setup many others use in all different DAW forums. It’s just a typical 3 drive setup used by most people who are using big sample libraries and higher track counts.

My point is, for me with this system, the CPU is not that important. I think if you are going to test how different DAW’s perform against each other, then you would want to eliminate the other bottlenecks that arise in music production related to data load. It’s not just about CPU performance.

I could see though, that if you are more into a lot of synth style electronic music and not using large sample libraries or doing lots of audio tracks then 1 hard drive might be fine and then the CPU performance might be more important so I guess it depends on what you do.