Old 04-27-2006, 05:25 PM   #1
Hedgehog
Human being with feelings
 
Hedgehog's Avatar
 
Join Date: Apr 2006
Posts: 110
Default Poll: How many effects can you run?

I'm playing around with Reaper seeing what I can get out of it, how stable it is ect ect.


Running at 44.1 with live monitoring of one channel I added effects until I start getting clicks and pops.

All good up until 50% Total CPU.

Most effects seem to run between about 1.5% and 3% with a few like Ambience running as high as 18% so that means I can handle around fifteen or so effects on average if I don't run any monster hogs.

This is on a 3Ghz Sony P4 that has 2G of RAM running XP.



Wondering if this is similar to what others are getting to work or what?

Also what is the deal exactly with selected and total CPU % at the bottom of the effects window? Don't think I follow exactly.
Hedgehog is offline   Reply With Quote
Old 04-27-2006, 06:37 PM   #2
RokkD
Human being with feelings
 
Join Date: Jan 2006
Posts: 1,089
Default

Quote:
Originally Posted by Hedgehog
Also what is the deal exactly with selected and total CPU % at the bottom of the effects window? Don't think I follow exactly.
If you have say, 3 effects in the window, all turned on, highlighting one will give you the its cpu%, plus the total for all three.
RokkD is offline   Reply With Quote
Old 04-27-2006, 07:13 PM   #3
Art Evans
Mortal
 
Join Date: Feb 2006
Posts: 6,656
Default

For any meaningful comparison you need to post at least a summary of the hardware used (which you have), and specify not only the effect in use but also the preset, plus whether the worker threads for FX are enabled. Also the bit depth and sample rate of the audio.

Here using a P4 3gHz 1Gb memory system I can smooothly replay 5 instances of Ambience reverb using the Amoebe Hall preset with the FX worker threads enabled, or 4 disabled, on 44.1/16 bit stereo files.
Art Evans is offline   Reply With Quote
Old 04-27-2006, 08:03 PM   #4
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by Art Evans

Here using a P4 3gHz 1Gb memory system I can smooothly replay 5 instances of Ambience reverb using the Amoebe Hall preset with the FX worker threads enabled, or 4 disabled, on 44.1/16 bit stereo files.

that's a ridiculously small number for that machine - I'll have to check with my machine if there's something wrong with Reaper...
jens is offline   Reply With Quote
Old 04-27-2006, 08:20 PM   #5
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

yeah - it really could be better - six instances here and the playback starts to crackle occasionally - seven is the maximum possible number but Reaper starts to become really slooooooooooow... (almost doesn't react anymore at all)

With eXT etc. (theoretically there shouldn't be any significant difference between different hosts) it's seven instances without any serious problems and a maximum of eight instances.

(Xp2600+ (Barton), 768mb, Echo Mia running asio4all @336 samples of latency)
jens is offline   Reply With Quote
Old 04-27-2006, 08:29 PM   #6
Hedgehog
Human being with feelings
 
Hedgehog's Avatar
 
Join Date: Apr 2006
Posts: 110
Default

Quote:
Originally Posted by jens
that's a ridiculously small number for that machine - I'll have to check with my machine if there's something wrong with Reaper...

Are familiar with Ambience?

At full power it uses several times more juice than most effects.

I can only run two instances of Ambience and a couple other effects but I'm running 24 bit 44.1k and I was live monitoring.
Hedgehog is offline   Reply With Quote
Old 04-27-2006, 09:05 PM   #7
Art Evans
Mortal
 
Join Date: Feb 2006
Posts: 6,656
Default

Reaper does better with Ambience than another I have... yes, it's a hog, which is why it's quick to test with!
Art Evans is offline   Reply With Quote
Old 04-28-2006, 04:54 AM   #8
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by Hedgehog
Are familiar with Ambience?

At full power it uses several times more juice than most effects.
for sure I am - what makes you think I might not be? - I use it for number-crunching tests for years now. ;-)


Perhaps you should have read my second post as well... ;-)
jens is offline   Reply With Quote
Old 04-28-2006, 04:59 AM   #9
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by Art Evans
Reaper does better with Ambience than another I have... yes, it's a hog, which is why it's quick to test with!
hm...? Is this other host perhaps Live?

as I said before there actually shouldn't be any real difference between the different hosts so you should get pretty similar results with all of them (Live is not worth comparing with) as by far the most cpu-cycles are consumed by the dsp-code of the plugin with such a test.


Still your machine should be more powerful than mine.

Last edited by jens; 04-28-2006 at 05:02 AM.
jens is offline   Reply With Quote
Old 04-28-2006, 06:25 AM   #10
RokkD
Human being with feelings
 
Join Date: Jan 2006
Posts: 1,089
Default

Quote:
Originally Posted by jens
(Xp2600+ (Barton), 768mb, Echo Mia running asio4all @336 samples of latency)
Another Echo mia user I see, but using Asio4all, so Im curious, how does Asio4all compare to the Echo 6.08 drivers, and does Asio4all install alongside the Echo 6.08, or should I uninstall the Echo drivers first?

Pc Winxp home P4 768mb Echo mia midi
RokkD is offline   Reply With Quote
Old 04-28-2006, 06:43 AM   #11
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by RokkD
Another Echo mia user I see, but using Asio4all, so Im curious, how does Asio4all compare to the Echo 6.08 drivers
I have found the performance to be usually better - also you have a much better control over the buffer-setting


Quote:
and does Asio4all install alongside the Echo 6.08, or should I uninstall the Echo drivers first?

you need the Echo-drivers in order to use asio4all with your Mia - the asio4all driver is a wdm-to-asio converter.

And this is the only problem about it:

sometimes when an application crashes while accessing the wdm-driver the driver remains unavailable for the rest of this Windows-session so you need to restart Windows for being able to continue using it (but you can still switch to the Echo asio-drivers then).
jens is offline   Reply With Quote
Old 04-28-2006, 07:25 AM   #12
Hedgehog
Human being with feelings
 
Hedgehog's Avatar
 
Join Date: Apr 2006
Posts: 110
Default

Quote:
Originally Posted by jens
for sure I am - what makes you think I might not be? - I use it for number-crunching tests for years now. ;-)


Perhaps you should have read my second post as well... ;-)

I didn't see your second post until after I posted myself.


Anyway, I was surprised you thought Art's four or five instances of Ambience was poor considering his computer so I just wanted to clarify that you understood Ambience uses up several times the power compared to many other effects.


How many instances of Ambience can you run or how many do you think Art should be able to?


My initial test was running live monitoring 24bit 44.1k, I'm not sure exactly of the latency but fast enough not to notice it, on one mono track. This only allowed for two instances of Ambience, well two Ambience and five Classic eq's actually.


A second test of just playback, rather than live, allowed four instances of Ambience with no trouble. Otherwise the same conditions as my original live test were used.


I have twice the RAM Art has (2G rather than 1G) and I would say I'm doing about as well as he is. His test was at 16 bit with a stereo track while mine was 24 bit with a mono track, I would think that would be similar as 24 bit is like twice the information as 16 bit.
Hedgehog is offline   Reply With Quote
Old 04-28-2006, 07:27 AM   #13
RokkD
Human being with feelings
 
Join Date: Jan 2006
Posts: 1,089
Default

Thanks for the info, I like having choices....will consider this one. Im okay with performance, but Im always looking for an edge with audio fidelity.
RokkD is offline   Reply With Quote
Old 04-28-2006, 07:28 AM   #14
Hedgehog
Human being with feelings
 
Hedgehog's Avatar
 
Join Date: Apr 2006
Posts: 110
Default

I didn't mention earlier but I'm using an Edirol FA101 firewire box.

Jens: It's interesting what you are saying about the Echo drivers vs this ASIOforall driver.

The Edirol driver has pretty vague controls as well, buffer size is like five choices but no mention of the actual latency in samples.

I would think it's safer (better?) to go with the drivers that were made by the manufacturer of the interface but maybe not?

Last edited by Hedgehog; 04-28-2006 at 07:35 AM.
Hedgehog is offline   Reply With Quote
Old 04-28-2006, 09:23 AM   #15
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by Hedgehog


How many instances of Ambience can you run or how many do you think Art should be able to?

seven or eight as I already have mentioned in this thread - considering Art's cpu and considering that he has 256mb more of ram than me (Ambience's ram-consumption is pretty excessive as well) he should at least be able to run nine instances at the same time(in hosts with normal performance such as eXT, Tracktion, Sonar, Podium, Samplitude, etc. that is)
jens is offline   Reply With Quote
Old 04-28-2006, 09:26 AM   #16
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by RokkD
Im okay with performance, but Im always looking for an edge with audio fidelity.

the asio-driver has no influence on the sound
jens is offline   Reply With Quote
Old 04-28-2006, 09:29 AM   #17
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by Hedgehog
s.

I would think it's safer (better?) to go with the drivers that were made by the manufacturer of the interface but maybe not?
the asio4all driver (just as the kx-driver for 10k1/k2 cards) is much better than most native drivers...
jens is offline   Reply With Quote
Old 04-28-2006, 10:17 AM   #18
Hedgehog
Human being with feelings
 
Hedgehog's Avatar
 
Join Date: Apr 2006
Posts: 110
Default

Quote:
Originally Posted by jens
the asio4all driver is much better than most native drivers...

Really... very interesting!


Ok, I downloaded asio4all and it is offered in Reaper now for my computer's internal soundblaster but not for My Edirol FA101.

Do I need to get rid of the Edirol driver and if so is that all I have to do to get asio4all to see my FA101 inputs and outputs?
Hedgehog is offline   Reply With Quote
Old 04-28-2006, 10:55 AM   #19
Hedgehog
Human being with feelings
 
Hedgehog's Avatar
 
Join Date: Apr 2006
Posts: 110
Default

Jens, YOU ROCK!!!


Switching to asio4all has doubled what I can run for effects.


I'm so glad I started this discussion!


I guess the Edirol driver pretty much blows chunks.


Thanks.
Hedgehog is offline   Reply With Quote
Old 04-28-2006, 11:00 AM   #20
pipelineaudio
Mortal
 
pipelineaudio's Avatar
 
Join Date: Jan 2006
Location: Wickenburg, Arizona
Posts: 13,384
Default

Justin bore witness to me running so many tracks full of effects (once SMP was enabled)that I lost count. Im running a pretty uber computer, cost me around 1500 bucks minus the soundcards.

It seems like there are some combos where reaper is less efficient than other apps in dealing with fx, but in this case you can run so many effects that you truly dont consider their cpu use! Im not kidding, first time ever I could work this way. Wanna add an effect? Add it. Even Magnus' Ambience and MJ Compressors.
pipelineaudio is offline   Reply With Quote
Old 04-28-2006, 11:10 AM   #21
RokkD
Human being with feelings
 
Join Date: Jan 2006
Posts: 1,089
Default

Quote:
Originally Posted by jens
the asio-driver has no influence on the sound
I guess it would be the soundcard its self that influences the sound.
RokkD is offline   Reply With Quote
Old 04-28-2006, 01:37 PM   #22
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by RokkD
I guess it would be the soundcard its self that influences the sound.

yes, indeed - the D/A converters (and for recording analog sources the A/D converters)
jens is offline   Reply With Quote
Old 04-28-2006, 01:39 PM   #23
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by Hedgehog

Switching to asio4all has doubled what I can run for effects.

hey, that's really great :-D - so your Edirol soundcard has been recognized by asio4all by now?
jens is offline   Reply With Quote
Old 04-28-2006, 02:06 PM   #24
Hedgehog
Human being with feelings
 
Hedgehog's Avatar
 
Join Date: Apr 2006
Posts: 110
Default

I didn't even remove the Edirol driver, just installed asio4all and selected it in the Reaper preferences.

Then all I had to do was in asio4all choose the Edirol, and it was recognizing all the Edirol I/O !

Now my computer is handling twice the effects with live monitoring and about one and a half times with playback compared to before with the Edirol asio driver.
Hedgehog is offline   Reply With Quote
Old 04-28-2006, 02:50 PM   #25
fluffy
Human being with feelings
 
fluffy's Avatar
 
Join Date: Jan 2006
Location: Port Neches, Texas
Posts: 1,949
Default

I was working the other day on a project with 72 tracks. I have no idea how many plugins I had going, but it was a lot. A LOT!

Athlon xp1700+
1gig Ram .
XP Home
__________________
IT'S A TROMBONE !!!
fluffy is offline   Reply With Quote
Old 04-28-2006, 05:31 PM   #26
manning1
Human being with feelings
 
Join Date: Apr 2006
Posts: 2,959
Default

fluff..
running fine here on various amd's durons and athlons.
let me clarify from a engineering standpoint.
the only thing impacting sound quality is convertor quality in the soundcard.
i'm gonna find a decked out dual opteron next to see what justins fine creation can do.
i'm not surprised asio4all works better than factory drivers.
factory drivers are sometimes pretty poor in my experience.
its cos the good software engineers ....lol ..are either building their own products like justin or working on geeky linux stuff or networks cos pay is often peanuts for good computer engr's these days.
couple of weeks back a co asked me if i wanted to come out of retirement. the veep was crying he couldnt get computer techs.
i told him i wasnt surprised with his pay scales.
i told him you pay peanuts you get monkeys...lol...
needless to say we didnt see eye to eye...lol
the problem is snr mgt are greedy and wont pay the techs.
glad i'm retired !!
manning1 is offline   Reply With Quote
Old 04-28-2006, 05:56 PM   #27
Art Evans
Mortal
 
Join Date: Feb 2006
Posts: 6,656
Default

Here my biggest issue with Reaper on the performance front is that it won't smoothly replay the same number of no-FX tracks that (say) Audition will play, using the same files. Last time I tried, Reaper could only manage about half. Then I accidentally wiped the set of files when beta testing another program (!) and haven't tested this aspect again for a few weeks - must revisit it.

When it comes to VST FX, in my experience there are quite significant differences between host applications in the number that can be run. Also, some hosts (eg Reaper) only process items on tracks, not whole tracks, whereas others might process the whole track even though there is only one tiny item on it. So in real-life situations where items are scattered around tracks, one program might have no difficulty in processing loadsa effects because at any one time, only a small number of items are actually being simultaneously played and therefore processed, whereas another app might be laboriously processing the silence between items across all used tracks, and therefore run out of steam much earlier.
Art Evans is offline   Reply With Quote
Old 04-28-2006, 06:12 PM   #28
jens
Human being with feelings
 
jens's Avatar
 
Join Date: Feb 2006
Location: Basel, Switzerland
Posts: 4,633
Default

Quote:
Originally Posted by Art Evans
, some hosts (eg Reaper) only process items on tracks, not whole tracks, whereas others might process the whole track even though there is only one tiny item on it.

currently Reaper does not disable track-inserts while they aren't in use...




edit: it does *oops* - cool :-D - just not Jesusonic plugins as it seems... oO

Last edited by jens; 04-28-2006 at 06:27 PM.
jens is offline   Reply With Quote
Old 04-28-2006, 07:10 PM   #29
Art Evans
Mortal
 
Join Date: Feb 2006
Posts: 6,656
Default

Easy to test - have ten tracks each with Ambience VST running, have two minute items on each track, but stagger their starts by say ten secs - plays fine to begin with but by the time you get to the end, I doubt whether Reaper will still be playing glitch free!

But some other apps would choke right at the start.
Art Evans is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 01:15 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2019, vBulletin Solutions Inc.