How does Multicore works on U-He Synths? still a mystery

[Urs]

With 6 instruments, the project ran at around 13% total. I swapped out each instance of Hive for Spire, programmed each patch the exact same with the same effects and the project hit around 10% to

So, where does CPU friendly start then if 13% doesn't cut it? 12%? 11%? Or is Spire's 10% the ultimate benchmark for "CPU friendly" for the current and coming decades? The magic break off point, from where companies are allowed to advertise their achievements?

I hereby admit, we have not much benchmarked competing software to assess the attribute "CPU friendly". We've got plenty of our own to compare to, and we *know* what crazy shit we went through to make Hive happen. Surely there are ways to get nearly as good results with a lot less CPU, but why the hoohah if it wasn't for the conceptual work, the sound and the results?

[neuromod]

I don't really want to get involved in this, but I don't see any weird shit, fury, or wild accusations.

[pdxindy]

Bruh, you are on some weird shit with these wild accusations, even when I point out to you that you've made a false claim you still cling to the idea that there was truth to your assertion, it's like you WANT me to be the bad guy. There is no reasoning with you, so I'm not going to read through all that crazy fury you just posted. Please chill out.

I did some first hand tests... I did them fairly and reported the results. If Hive used a lot more CPU I would have reported it. Instead, it used a bit less in the specific tests I did.

Perhaps it would use more in a different test. The point is, there is not a big disparity in the cpu consumption of different wavetable synths as was claimed.

[Urs]

I don't really want to get involved in this, but I don't see any weird shit, fury, or wild accusations.

There are recurring themes that are quite tiresome. And the video that has been posted is the snotty culmination of a kind of view and method we don't share.

[pdxindy]

I don't really want to get involved in this, but I don't see any weird shit, fury, or wild accusations.

There are recurring themes that are quite tiresome. And the video that has been posted is the snotty culmination of a kind of view and method we don't share.

I only watched half that video cause it was so obnoxious.

[neuromod]

I don't really want to get involved in this, but I don't see any weird shit, fury, or wild accusations.

There are recurring themes that are quite tiresome. And the video that has been posted is the snotty culmination of a kind of view and method we don't share.

(I don't know if I was misunderstood - I was backing up pdxindy. If that was understood, ignore this post).

[plugmon]

I don't really wanna be involved in this kind of topic either but just for the honor of the synth I sincerely love :

https://www.youtube.com/watch?v=24_q8VMzI9I

This is a comparison of Hive/Serum/Vital, each with init settings and edited settings(unison, resonant filter, modulation and effects), so 6 tracks in total.
Hive seems to be smart in that its CPU usage does not much increase when many modules are activated and connected, although it may not be efficient on the very init state. (But this is just one experiment on one environment so I never say I prove anything)

The video posted early in this thread can easily make people overlook these property differences, because in the video Hive with edited settings is never demonstrated, despite that all other synths are played both on init & edited settings. Accidentally or intentionally, the video shows the bad part of Hive while hiding the good part of it. "Let's compare this to Hive with f**king 1 voice" is sensational but not a scientific manner of experiment -- The conditions must be as close as possible. I feel like I witnessed a clear example of how a fact can be distorted using only facts.

And of-course you shouldn't forget the value of the balance between sound quality and CPU performance. A "relatively CPU friendly" synth may well deserve a title of "CPU friendly synth" provided that its sound quality is superb.

[neuromod]

"a fact can be distorted using only facts" is a nice way of describing a widespread phenomena

[Urs]

I don't really want to get involved in this, but I don't see any weird shit, fury, or wild accusations.

There are recurring themes that are quite tiresome. And the video that has been posted is the snotty culmination of a kind of view and method we don't share.

(I don't know if I was misunderstood - I was backing up pdxindy. If that was understood, ignore this post).

(Sorry, I wasn't sure, it was me who first assumed that the person who posted the video was making its views their own)

[Urs]

I don't really wanna be involved in this kind of topic either but just for the honor of the synth I sincerely love :

Thank you so much!

Developers are always confronted with comparisons of their products to that of others. There is never a good answer to these, since pointing out strengths of one's own product might seem like bagging someone else's products. So we can't really make such a video ourselves.

I have actually become very careful in how I mention other synthesisers since I've been accused of bullying for saying out loud that I disliked the factory presets of a certain vintage hardware synthesiser (Alesis A6, if you need to know).

Anyhow, what you show is a good example of what I meant to say about optimisation. It is only really worth investing considerable effort when the result is obvious and foreseeable enough. If, say, an 8x unison oscillator uses, say, 10% CPU while 4x unison uses 5%, I would assume that 1x goes down below 2%. That's an optimisation I would certainly go for, since it scales proportionally. But if 8x uses 10% while 4x uses 8%, I'd assume that the step from 4x to 1x may be too minimal to pursue. In fact, the logistics behind the optimisation itself may make things worse.

Now for some technobabble: Optimizing Hive's oscillators for 1x/2x unison may involve templating and specialising the algorithm. This leads to duplication of code blocks with little variances. This in turn means that memory access for executable code multiplies, as multiple version of the algorithm may need to be loaded into memory. This may be good for laboratory settings, but in the real world it easily degrades performance overall.

To further fuel things, it is our experience that SSE-like parallelisation ("4 at once") really only ever halves CPU. That is, we "get 4 for the price of 2". Therefore, there may only be a path to ever optimise for 1x oscillators as 2x will already perform exactly like 4x, and the outcome will be nowhere near 1/4 of the CPU. The fact that unison does not scale proportionally means that there isn't only time spent on actual playback - there is also time spent on logistics behind shared parameters, such as tune, modulation and mixing. Factoring this in, I think a 1x Hive oscillator would probably consume 0.12% instead of 0.16% which is nowhere near the 0.04% that one would naively expect.

These are some of my thoughts in regards to further optimisation of Hive. Not processing an oscillator at all if it isn't selected in a filter input is a separate topic, and one that I might actually pursue.

[Urs]

Search this thread for "false advertisement" and tell me where that was said please.

Brilliant idea, but I searched Google instead and figured out that the person who made the video is the person who is an "artist formerly known as claude defaran". And then there's this in the video description:

Their claims that Hive is a low CPU synth should be considered false advertising.

[Howard]

Boom! Excellent detective work.

[zvenx]

Oh dear, indeed.


rsp

[MOD Edit: Deleting the two pics that contain user's real name, which is still visible in the video he himself posted, in case someone wishes to investigate]

[MOD Edit2: User does not want screenshots from his Youtube account shared either... hence deleting the 3rd pic as well]

[zvenx]

.....

(Q: Wasn't Dune 3 *always* processing on multiple cores, no matter what? I can't quite remember but I thought this was one fo its features?)

I think that was Legend (and probably Obsession too). For sure Dune 3 has a multithread on and off selection.
rsp

[pdxindy]

Search this thread for "false advertisement" and tell me where that was said please.

Brilliant idea, but I searched Google instead and figured out that the person who made the video is the person who is an "artist formerly known as claude defaran". And then there's this in the video description:

Their claims that Hive is a low CPU synth should be considered false advertising.

Busted! ... hehe... perfect