I’m just using the DAW performance meter for Live 11 and Nuendo 11, nothing crazy scientific, I’m a musicianUrs wrote: ↑Sun Mar 21, 2021 7:45 pmOk, weird. So how do you measure CPU usage? - Asking because, we would like to try ourselves, but we won't know where too start if we don't know how you got there.simmo75 wrote: ↑Sun Mar 21, 2021 7:41 pmThanks Urs,Urs wrote: ↑Sun Mar 21, 2021 6:19 pmIt should actually use less CPU than the previous version, particularly with the UI open. How did you measure that?
Anyhow, pretty much all versions of everything up to now are available for download here:
https://dl.u-he.com/release-archive/
I was really excited but it uses more cpu than most of my plugins now.
Hive 2.1: new sonic territory
-
- KVRAF
- 4506 posts since 25 Mar, 2016 from Seattle
- u-he
- Topic Starter
- 28065 posts since 8 Aug, 2002 from Berlin
-
Funkybot's Evil Twin Funkybot's Evil Twin https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=116627
- KVRAF
- 11519 posts since 16 Aug, 2006
What patch? Under what conditions (how many sustained notes, etc.)? What version of Hive are you comparing to when you say 15% more? GUI open or closed make a difference?
I get that you're not doing anything crazy scientific but you're saying it's using 15% more than the old one but it also sounds like you haven't done a direct comparison of the two. So is it really using 15% more playing the same notes on the same patch or do you just think it's using more without having done more extensive comparison? Details are very important.
Older versions should be here: https://dl.u-he.com/release-archive/hive
Try a direct compare (if you haven't already) and report back.
I get that you're not doing anything crazy scientific but you're saying it's using 15% more than the old one but it also sounds like you haven't done a direct comparison of the two. So is it really using 15% more playing the same notes on the same patch or do you just think it's using more without having done more extensive comparison? Details are very important.
Older versions should be here: https://dl.u-he.com/release-archive/hive
Try a direct compare (if you haven't already) and report back.
-
- KVRAF
- 4506 posts since 25 Mar, 2016 from Seattle
Hive has always been super low cpu hit for me, one instance used approx 2-3% cpu in Live 11.
Now I’ve updated I’m seeing 15 to 17%.
It now causes Live 11’s GUI to glitch out and Hive 2.1’s GUI glitches out too and is really sluggish.
Maybe it could be something to do with my iMac Pro’s graphics card.
Now I’ve updated I’m seeing 15 to 17%.
It now causes Live 11’s GUI to glitch out and Hive 2.1’s GUI glitches out too and is really sluggish.
Maybe it could be something to do with my iMac Pro’s graphics card.
- KVRAF
- 4123 posts since 23 May, 2004 from Bad Vilbel, Germany
@simmo75: You ran both Hive versions in Live 11? I'll tell the guys in Berlin…
-
- KVRAF
- 9133 posts since 6 Oct, 2004
Spent some time comparing sounds among the linux releases, and Hive 2.1 seems
to be using more cpu than Zebra and ACE, in the ballpark with Repro 1, and probably less
than Repro 5 if considering polyphony, and a bit less than Diva on similar sounds.
But those are quick impressions, not in spreadsheet detail.
I would expect that result, because Hive sounds so good. And I suspect that sound design
skills improving over time, have a lot to do with lowering cpu use over the same time.
And brilliant effects can be deployed that bump the cpu readings back up again.
I hope miracles of optimization are discovered, that make an entire product line
use drastically less resources to produce comparable sounds, but even then,
it's not like fitting for new shoes, where a few different sizes are all that's needed.
Cheers
to be using more cpu than Zebra and ACE, in the ballpark with Repro 1, and probably less
than Repro 5 if considering polyphony, and a bit less than Diva on similar sounds.
But those are quick impressions, not in spreadsheet detail.
I would expect that result, because Hive sounds so good. And I suspect that sound design
skills improving over time, have a lot to do with lowering cpu use over the same time.
And brilliant effects can be deployed that bump the cpu readings back up again.
I hope miracles of optimization are discovered, that make an entire product line
use drastically less resources to produce comparable sounds, but even then,
it's not like fitting for new shoes, where a few different sizes are all that's needed.
Cheers
- u-he
- Topic Starter
- 28065 posts since 8 Aug, 2002 from Berlin
Here are our findings, using Live's CPU meter and comparing to top in the command line.
Live solo (no plugin loaded):
Live 10: top 12-14%, Live average: 1-2%
Live 11: top 12-14%, Live average: 1%
9709 (2.0.1)
10: Hive GUI open, not playing: top 28%, Live average 8-9%
10: Hive GUI open, playing startup preset: top 46%, Live average 18-20%
10: Hive GUI closed, not playing: top 22%, Live average 10-12%
10: Hive GUI closed, playing: top 33%, Live average 18-20%
11: Hive GUI open, not playing: top 58-60%, Live average 8-9%, current 15-30%
11: Hive GUI open, playing startup preset: top 76%, Live average 18-20%, current 20-40%
11: Hive GUI closed, not playing: top 22%, average 10-12%, current 20-30%
11: Hive GUI closed, playing: top 33%, average 18-20%, current 20-40%
11037 (2.1)
10: Hive GUI open, not playing: top 58-60% , Live average 8-9%
10: Hive GUI open, playing startup preset: top 75%, Live average 17-20%
10: Hive GUI closed, not playing: top 22%, Live average 8-10%
10: Hive GUI closed, playing: top 32%, Live average 18-20%
11: Hive GUI open, not playing: top 58-60%, Live average 8-9%, current 15-30%
11: Hive GUI open, playing startup preset: top 74%, Live average 15-18%, current 20-40%
11: Hive GUI closed, not playing: top 22%, average 8-9%, current 20-30%
11: Hive GUI closed, playing: top 30%, average 17-18%, current 20-40%
Note that the biggest difference is in Live's CPU metering is happening from Live 10 -> Live 11, but overall Hive 2.1 performs a tad better on our testing system.
An interesting tidbit is that our old Hive UI which was behaving "sluggish" was using less CPU when used in Live 10 than it does in Live 11. Our new "non-sluggish" GUI uses the same CPU in the background, according to top, but simply isn't sluggish anymore.
Live solo (no plugin loaded):
Live 10: top 12-14%, Live average: 1-2%
Live 11: top 12-14%, Live average: 1%
9709 (2.0.1)
10: Hive GUI open, not playing: top 28%, Live average 8-9%
10: Hive GUI open, playing startup preset: top 46%, Live average 18-20%
10: Hive GUI closed, not playing: top 22%, Live average 10-12%
10: Hive GUI closed, playing: top 33%, Live average 18-20%
11: Hive GUI open, not playing: top 58-60%, Live average 8-9%, current 15-30%
11: Hive GUI open, playing startup preset: top 76%, Live average 18-20%, current 20-40%
11: Hive GUI closed, not playing: top 22%, average 10-12%, current 20-30%
11: Hive GUI closed, playing: top 33%, average 18-20%, current 20-40%
11037 (2.1)
10: Hive GUI open, not playing: top 58-60% , Live average 8-9%
10: Hive GUI open, playing startup preset: top 75%, Live average 17-20%
10: Hive GUI closed, not playing: top 22%, Live average 8-10%
10: Hive GUI closed, playing: top 32%, Live average 18-20%
11: Hive GUI open, not playing: top 58-60%, Live average 8-9%, current 15-30%
11: Hive GUI open, playing startup preset: top 74%, Live average 15-18%, current 20-40%
11: Hive GUI closed, not playing: top 22%, average 8-9%, current 20-30%
11: Hive GUI closed, playing: top 30%, average 17-18%, current 20-40%
Note that the biggest difference is in Live's CPU metering is happening from Live 10 -> Live 11, but overall Hive 2.1 performs a tad better on our testing system.
An interesting tidbit is that our old Hive UI which was behaving "sluggish" was using less CPU when used in Live 10 than it does in Live 11. Our new "non-sluggish" GUI uses the same CPU in the background, according to top, but simply isn't sluggish anymore.
- u-he
- Topic Starter
- 28065 posts since 8 Aug, 2002 from Berlin
Hehehe, check this out, too:
https://oddsound.com
Now that it's official, Hive 2.1 is also our first product to support it. All other plug-ins following shortly.
https://oddsound.com
Now that it's official, Hive 2.1 is also our first product to support it. All other plug-ins following shortly.
- KVRAF
- 25445 posts since 3 Feb, 2005 from in the wilds
Interesting... will check it out! (thanks for the heads-up)Urs wrote: ↑Mon Mar 29, 2021 12:15 pm Hehehe, check this out, too:
https://oddsound.com
Now that it's official, Hive 2.1 is also our first product to support it. All other plug-ins following shortly.
-
- KVRian
- 678 posts since 15 Feb, 2012 from France
-
- KVRist
- 275 posts since 31 May, 2017
-
- KVRist
- 275 posts since 31 May, 2017
Serious? Or joke?
- u-he
- Topic Starter
- 28065 posts since 8 Aug, 2002 from Berlin
-
- KVRAF
- 2897 posts since 3 Mar, 2006
Hi Urs, is there a way to turn this feature off if I don't want hive to pull tuning data from the oddsound master plugin but I am using the oddsound stuff elsewhere?Urs wrote: ↑Mon Mar 29, 2021 12:15 pm Hehehe, check this out, too:
https://oddsound.com
Now that it's official, Hive 2.1 is also our first product to support it. All other plug-ins following shortly.