64-bit processing- the feature I'm going to pay for

You mean the entire piano sounds transposed or is out of tune between notes?? How far out is it?

Well, maybe I was not precise. There is no problem with my piano track. I just wanted to point out, that 64 bit mixing gives your sound much more details, and if something needs correcting in your recording, it’s easer to hear it. Like good studio monitors- they should not only give you beautiful sound but TELL THE TRUTH.

“Please run export twice from 9.5 and run the null test and report back the results. If you hear any difference then you just invalidated your 9 to 9.5 comparison.”
Just checked. First- for excluding the confusion of some “randomising” effects- I switched them off.

Then exported the same mix form C9 and C9,5- twice from each.

2 mixes from one version give null sound when 2nd track phase- inverted. Joining C9 and C9,5 shows some difference.

I don’t understand why you have run them from different Cubase versions which leaves room for error in settings. Just switching the 64bit processing on or off in the same version is a fair test.

Honestly I’m not going to say there can’t be any difference using 64bit engine but saying the 64bit is suddenly “audiophile” or that there is “much more detail” is really overstating what is possible. Any difference should be incredibly subtle if not completely inaudible.

Otherwise you’re telling me we’ve all been using a crippled DAW all these years while other DAWs already had 64bit processing…I’m sure the reason Steinberg haven’t rushed to catch up is that they know it makes little to no difference to the final result.

From Philippe, developer of Steinberg WaveLab:

The interest of 64 bit float is not about “headroom” / “dynamic range”…

pro:

no need to convert between 32 bit and 64 bit float: 64 bit is needed by some plugins for their internal computations. 64 bit then means: small performance gain and no precision lost between succeeding 64 bit plugins.
Better audio precision when mixing audio signals. I explain this at the end of this message.
If audio devices ever go beyond 24 bit precision, 64 bit float will be needed (because 32 bit float means, in fact, 24 bit precision)

con:

requires more memory, which can mean a performance lost (more memory to move). But as soon as a sophisticated plugin is used, this one is likely to become the bottleneck, compared to the memory overhead. Therefore, this is a “relative con”.
64 bit CPU instructions are as fast as 32 bit instructions, because the CPUs are 64 bit today. But certain rare instructions are faster with 32 bit float, because the CPU can conjugate 2 of them while in the same time, only one 64 bit instruction is performed (SIMD).

Now, an explanation about 32 bit float vs 64 bit float, for mixing.
While 32 bit float means in fact 24 bit precision, 64 bit float means in fact 48 bit precision. This means, far more precision.
I can illustrate this difference with elementary school maths (this is an analogy of what happens in reality).

Let’s say samples can have only values 0, 1, 2, 3, 4, 5,…
Let’s start with a sample that has value “3”
An audio gain of “divide by 2”, is applied. We get the value “1.5”, but this value is not allowed hence must be rounded, eg. the new value becomes 1.
Later another gain “multiply by 2” is applied. The new sample becomes “2”.

Consequence: we started from value “3” and ended up with value “2”, while the two gains should have cancelled each other.

When this kind of loss is performed multiple of times (complex mixing), then errors stack up.
The consequence is not dramatic, because some errors are (randomly) compensated by others (round-down / round-up), but this compensation actually means “digital fog” aka noise.

64 bit float processing pushes the digital fog far from the 24 bit domain. Hence a cleaner result at the end of the audio chain.

The difference 32/64 is therefore about “audio definition”, if your ears can sensible enough. But that’s another topic!
Philippe

Exactly! I am musician only and couldn’t explain it in such sophisticated way :wink:

I was absolutely not going to say that any older Cubase versions were “crippled”, but emphasize its new great feature. (In the other thread I am criticizing some other feature a lot :wink:

It doesn’t matter if I used old C9 for comparison or just switched C95 from 64 to 32 bit. (You have to restart it after changes anyway).

You can’t hear any difference on one track, but when it comes to calculate some less or more complex mix- more bits you use for calculation- more detailed mixdown you get. Seems simple :wink:

It does if you’re wanting to compare 32bit to 64 bit and not maybe a difference between other things that may have changed between 9 and 9.5.

Kind of like how the new 120Hz TVs show the flaws in movie special effects, because the images are so clear? :smiley:

The allways thinking users that are techies will think 64 bit audio must be way better then 32bit. I can see it, i can hear it.
No man, 200 Hz or even 500 Hz TV’s! WOW it has such a clear picture! :laughing: :laughing: :laughing: :laughing: :laughing: :laughing: :smiley: :smiley: :slight_smile: :slight_smile: :wink: :wink: :wink: :open_mouth: :open_mouth: :open_mouth: :open_mouth: :open_mouth: :confused: :confused: :confused: :astonished: :astonished:

To be fair:

Televisions are an endpoint for viewing. Whatever the original signal is (including its resolution and frame rate) will be displayed utilizing whatever upscaling/frameblending/etc. technology is built into the television - but the capabilities of the television really only can affect the final output to the user. If you’re dealing with a 720p 29.97fps signal, the television can only work to enhance the signal. Yet, newer televisions tend to do a better job at this than older ones due to the sophisticated algorithms built-in. But I digress.

The bit depth capabilities of an audio engine, on the other hand, enable every step of the processing and mixing process to be performed with higher resolution. In many projects, there are dozens or even hundreds of tracks, each being effected and transformed perhaps several or even dozens of times, depending on DAW and plugin architecture. Then the full mix itself (the main buss) is often also effected and transformed. In the aggregate of all this processing, precision certainly matters.

If you took, say, 500 digital televisions from 2005 and could find a way to run a video signal through them in series, you might see a detectable difference in the output at the last television in the series. Depending on the televisions’ error correction and signal loss schemes, there might be greater loss of resolution than in a series of 500 120Hz televisions, perhaps at least partially due to the faster refresh (and therefore, depending on the technologies built into the television, the ability to perform higher-quality error correction and signal amplification). This is a more fair comparison than just comparing a single television to a DAW.

“Techies” aren’t thinking that 64 must be better than 32 because of a brute numeral comparison. When you aggregate the precision benefits of 64-bit processing in a mix, accuracy is improved. It’s hard for me to understand why an audio professional - indeed, a purveyor of audio software - cannot understand this basic fact.

I believe the issue is with the phrase “Accuracy is improved”… the math certainly looks great on paper, but in reality, is unlikely to be perceived by human ears. If someone sold a high-end light bulb with a 1000Thz band output, when humans can only perceive up to 770, I’m sure they’d be “ocularphiles” claiming that it lifts the “fog” or enhances the light somehow. :unamused:

In short, I have yet to find a reputable, unbiased source claim that 64 bit audio sounds better than 32 bit (summing or otherwise).

I think many are missing the point here.

The argument doesn’t rely whatsoever on the idea that there is something beyond human perception that certain people are claiming to somehow hear. This isn’t the point, and makes this conversation different from those sorts of conversations.

The concept is not that a single stream of data is being improved on its own in isolation (although theoretically that could occur).
The concept is not about signal vs. noise floor.
The concept is not about hearing a perceived difference in a single stream of 32-bit data vs. 64-bit data.

The point is that a well-designed 64-bit audio engine is inherently capable of higher precision (and thus resultant accuracy) than a 32-bit audio engine. Thus, the results of mixing in the 64-bit audio engine have the inherent capability of sounding “better,” depending on the processing done and the taste of the person doing the mixing. The results can, in fact, be audibly different. It’s not a mystery.

The claim isn’t the brute idea that “64-bit is inherently better sounding than 32-bit.” That’s definitely not what I am saying. What I am saying is that the capability to perform accurate audio processing is greatly enhanced - as a mathematical fact - by a 64-bit engine vs. a 32-bit engine (assuming the engines are otherwise equivalent and well-designed).

The high-end light bulb example is flawed because what you’re trying to describe is a single output mechanism emanating in spectrum beyond human perception capabilities. That’s not what we’re talking about.

What we are talking about is the additive and multiplicative effect of numerous calculations in a chain, wherein the precision of calculation at each point in the chain has a direct impact on later calculations and overall quality. We’re not talking about whether a single element might be benefited by 32-bit vs. 64-bit (and hence the perceived difference might be nominal, if not null). We’re talking about the collection of a variety of complex calculations, some of which can amplify error numerous times (think about wide variances in gain staging, extreme saturation processing, etc.), and the result of combining numerous effected signals via mixing.

The light bulb example is flawed because the light bulb isn’t performing numerous light calculations, effecting and mixing perhaps hundreds of streams of light to form its output. It’s simply creating a single stream of output.

In a DAW, accuracy will benefit from 64-bit processing because, literally, greater precision is mathematically possible. Arguing against this is somewhat silly because it’s true. Whether or not you would prefer the results is up to you, and whether the results are audibly different depends on how the material is being processed. With 64-bit processing, more precision is possible, and thus higher resultant accuracy can be achieved than with 32-bit processing. Depending upon the nature of the processing you do with a 64-bit engine vs. 32-bit, you may or may not hear or notice differences. It depends on how you use the engine.

Think about a 2017 Ferrari. If you only drive it 30mph in town, will you notice many differences between it and a Honda Civic? Acceleration, handling, etc. sure, but at that speed, a car’s a car. But when you rev it up to 180mph, you’ll definitely be able to tell the difference between the Ferrari and the Civic. Pushed to limits, the Civic simply can’t keep up with a well-tuned Ferrari. In a DAW, the differences in accuracy in processing between 64-bit and 32-bit could be more readily heard if you “rev it up” in processing algorithms and scenarios where the 32-bit engine can’t “keep up” due to errors like the 64-bit engine can. When signal chains and the mix itself is highly effected in ways that multiply/amplify the errors, a difference can be more readily heard. It’s just math, and math is real. It has a real effect. I’m not saying you’ll always hear a clear difference between 64-bit and 32-bit processing; I’m saying the potential is inherently available in a 64-bit engine to produce higher-accuracy results.

Also, if you don’t think Philippe Goutier is a reputable source on this topic, you may be in trouble finding a reputable source.

It is really not that complicated, turn on 64 bit float engine, restart Cubase, load your project.
Look at the performance meter in Cubase and on your computers cpu usage.
Is there any difference? For me about the same, some projects have even better performance.

Are the data calculations done more precise YES
Can it possibly improve the sound quality, something subtle that I don’t hear (yet) YES
Why would I ever go back to using the 32bit float engine?
Just the potential of sound improvements without any cost, makes it worth flipping that switch IMHO

The cost is small, but it is there, in terms of more storage needed, and possibly more CPU power used. I believe that’s true, please disabuse me of my misinterpretation if otherwise. Thank you!

Confirmation bias…google it…

Yes there is a cost, it will use more memory and cpu cycles.
but potentially less cpu cycles if all plugins are running 64Bit float so no converting To and from 32bit float has to be done.
But we are talking 3-5%, well that is my experience, not a massive difference.

There should not be an increased storage requirement, as you can process exactly the same files in a 64-bit engine as a 32-bit engine. (Unless you’re talking about the storage for code of Cubase itself, which should be considered trivial in this context.)

CPU and memory usage will vary according to what you’re doing with the 64-bit engine. In most cases I’ve encountered so far, and as peakae mentioned, CPU and memory usage has actually benefited from the 64-bit engine. If you reduce the amount of conversion work Cubase has to do by keeping everything in 64-bit processing, CPU and memory usage will be lower.

Oh, Thanks God, our discussion again turned into professional level! At the very beginning I said that I compared files MIXED on 32 and 64 bits (files were 16 bit ones). Never tried to claim that I hear the difference between 32 and 64 bits single files.

Cubase has just provided us with more precise tool for MAKING music, not just LISTENING. We can manage without it and make interesting things even on Windows Movie Maker :wink: but- at professional level- better and more precise tools we have- more chance we get to create a real masterpiece. At the price of 60 Euro (upgrating from 9) and a little more processing power, it is worth to try!

You guys… this is a very, very old discussion. It has been beaten to death with every incremental change in audio technology.
The future is 64-bit processing, it is true. We will all get there some day.

Of course, incredible records can be made on 32-bit systems…or 16-bit systems…or tape…or 4-track…or 2-track…or even direct-to-acetylene.

Mixing with the new engine will not make your mixes better. And kids these days are just gonna encode it to mp3 and listen in the car through earbuds… so it really doesn’t matter.

All that said- I am as much a gear-sloot as the next guy. Bring on the new stuff! :slight_smile:

Pity you didn’t start the discussion professionally rather than with hyperbole such as how your mixes were suddenly “Audiophile” because of the 64bit engine. :laughing: