Pulsar II - clicks and noise...

An area for people to discuss Scope related problems, issues, etc.

Moderators: valis, garyb

Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

I was hoping someone may be able to help with a recent problem with my Pulsar II
card. I've moved the Pulsar II from my old Pentium III 450 machine into my
Carillon (P4 2.2GHz, 850 chipset, 512 RAM, Win XP - tweaked for audio), and
I'm now experiencing problems (all was fine in my old machine). After a while I start hearing clicks which
rapidly increase in frequency and end in continuous loud noise from the
Pulsar card. Also, sometimes the noise just kicks in full pelt with no
clicks first.

I can make this noise disappear but only if I disconnect the virtual cables
in the scope software from the inputs (sometimes this is just the Analog
ins, but sometimes also the ASIO or wave driver inputs) to the mixer in the
scope software (or exit scope entirely). If I reconnect the virtual cables,
the noise is still there.

I'm now using scope with Cubase SX (1.06) but the
problem happens whether I'm running Cubase or not.

I've already tried the card in all the available PCI slots, and I still get
the same problem.

Also, in wondering whether it might be an overheating problem, I tried
increasing the air flow by removing some foam from behind one of the air
vents in the Carillon, but this has made no difference either.

The other thing I was wondering is whether this is a known conflict with the
850 chipset, as the card worked fine on my old computer (BX chipset)?

Any information or suggestions as to how I solve the problem would be most
gratefully received.

Many thanks in advance,

Badger
User avatar
valis
Posts: 7684
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Post by valis »

I doubt its specifically a problem with the i850 chipset, too many people are using it.

Have you tried reinstalling into your old machine temporarily to see if the card encountered a problem during the move (physical defect)?
User avatar
interloper
Posts: 370
Joined: Sun Mar 25, 2001 4:00 pm
Location: amsterdam
Contact:

Post by interloper »

Or have you run the ASIO multimedia setup successfully with a SFP project open containing ASIO drivers? Did your system pass all the simulations?
Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

Thanks for the reply. No, I haven't tried moving it back to my old machine yet, but will give that a try as soon as I get a chance, just in case.

However, I hope that's not the problem as I've always been very careful when handling the card (strapped to earth, don't touch contacts, no forcing etc.) Anything's worth a try though. Will let you know the result ASAP.

Thanks again.
Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

Ah, interesting suggestion, Interloper - thanks. Will give that a try first.

Cheers
User avatar
interloper
Posts: 370
Joined: Sun Mar 25, 2001 4:00 pm
Location: amsterdam
Contact:

Post by interloper »

Just for reference, here's some steps you might already be aware of:

Don't start SX, load a SFP project with ASIO drivers and go to ASIO multimedia setup for SX, skip the initial test, select advanced, detect buffer size (output first) and then hit check buffers & sync.

Adjust settings until it works (it should work ok the first time, if you detected the buffer size).

Then repeat for the inputs (detect, check buffers & sync).

Save settings under a new name (DAW or something), then run simulation. That should do it...
Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

Thanks Interloper. It wouldn't have occured to me to try the ASIO Mulimedia setup as I'm using the ASIO Scope driver with Cubase SX.

However, I have now followed your instructions but unfortunately, the "Check Buffers" command only works on the Outputs, and then only if I have "16 bit" input selected (from the bottom right drop-down).

Otherwise I get I dialog box with the following error message: "Errorcode: FF-5-000C - Info: Device could not be opened". Running the main Setup (the one you said to skip) results in the following message:

"Not all the selected devices in the current configuration were successfully started with their selected options. If this card is currently configured to accept a digital signal, and no digital signal is present, please either reconfigure the card to use an 'internal' clock, or apply a digital signal, or if possible switch this port to accept an analog input".

But...Pulsar clock is set to "internal" and the Analog inputs are already connected and appear as Analog inputs in scope!

I forgot to mention that I also had Midi Yoke running and also have a Yamaha SW1000XG card installed, but ran the above tests both before and after disabling the drivers for both.

I'm still getting the click/noise problem intermittently. Any other ideas?

Cheers
User avatar
interloper
Posts: 370
Joined: Sun Mar 25, 2001 4:00 pm
Location: amsterdam
Contact:

Post by interloper »

The ASIO Multimedia setup is for the Scope drivers.

What is your ULLI setting? Set it to 13 ms.

What is your samplerate? I'm assuming 44.1 kHz.

I don't know whether the Yamaha card is affecting your system, because I don't have one.

Go to control panel > sounds & audio devices > audio.

You should have CW Play/Rec 1 selected for sound playback, and CW Play/Rec 1 selected for sound recording.

Then go to the hardware tab within the same properties and select your card, double-click > properties > audio devices > Pulsar II.

You should have the "Use audio features on this device" selected (you probably do).

When you say that you have the noise with or without SX open, does it just occur as soon as SFP is loaded? This may indicate it has nothing to do with the sequencer. But the fact that you can't check the buffers on the inputs is very strange.

Give us feedback on the above, then we might try some other things.
Guest

Post by Guest »

On 2003-10-07 15:41, interloper wrote:
The ASIO Multimedia setup is for the Scope drivers.

*****This refers to the wave drivers....




What is your ULLI setting? Set it to 13 ms.

What is your samplerate? I'm assuming 44.1 kHz.

I don't know whether the Yamaha card is affecting your system, because I don't have one.

Go to control panel > sounds & audio devices > audio.

You should have CW Play/Rec 1 selected for sound playback, and CW Play/Rec 1 selected for sound recording.

*****Again refers to wave drivers


Then go to the hardware tab within the same properties and select your card, double-click > properties > audio devices > Pulsar II.

You should have the "Use audio features on this device" selected (you probably do).

When you say that you have the noise with or without SX open, does it just occur as soon as SFP is loaded? This may indicate it has nothing to do with the sequencer. But the fact that you can't check the buffers on the inputs is very strange.
*****if the wave drivers are not loaded you will not be allowed to check buffers


Give us feedback on the above, then we might try some other things.


*******sorry for jumping in/out of the post I'm very busy at themoment

Paul
User avatar
interloper
Posts: 370
Joined: Sun Mar 25, 2001 4:00 pm
Location: amsterdam
Contact:

Post by interloper »

My mistake. Thanks for the clarification. Just trying to compare his system settings to mine, because I've had this problem before with SFP 2.01. It was somewhat intermittent and hard to localize. The problem was only present when the Multimedia Setup simulation did not run successfully. I should say also that the problem was only present during Cubase playback, which is why his scenario is strange.
Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

Well, thanks, I've now tried all that, regardless of whether it refers to the wave drivers or not (which is how I understood it - but I could've misunderstood and I'm grateful for your help in any case).

ULLI was set to 7ms, so I've reduced it to 13ms as suggested (maybe this helped - not terribly happy about it though, after all, what then, is the point of making the latency user-adjustable?). BTW, it was working fine on my old PIII 450 set to 7ms!

Yes, you were correct in assuming the sample rate to be 44.1KHz.

Following this, the ASIO MM setup response is exactly the same as before, but I'm not getting any clicks or noise this time. However, the problem was always intermittent, so I can't be sure yet whether the problem is solved.

No, the click/noise doesn't always start on opening the scope software (it almost never does, actually), but tends to kick in at some unpredictable time later (usually within half an hour or so of booting the computer, which was why I initially suspected overheating)

I suppose the next step is to try moving the Pulsar back into the old machine (as Valis suggested) to see whether I still get the problem there (never used to though).

Thanks again
User avatar
interloper
Posts: 370
Joined: Sun Mar 25, 2001 4:00 pm
Location: amsterdam
Contact:

Post by interloper »

Hmmm, what kind of video card do you have in the Carillon? I seem to recall that years ago when I had a video card with little memory and I was switching back between Cubase and SFP, the problem would manifest itself. Somehow, it was tied to the latency setting as well. Switching between the apps put a burden on the system, it seemed, and the snap, crackle, & pop started.

I would venture to guess that your card is not damaged, unless you grabbed it by the TDM or PCI terminations without grounding yourself when you swapped it.

<font size=-1>[ This Message was edited by: interloper on 2003-10-07 20:39 ]</font>
User avatar
valis
Posts: 7684
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Post by valis »

What Paul@Creamware was trying to say is that ASIO MM drivers are software ASIO drivers which will output to the WAV drivers in SFP, not what you want. Its a software asio layer installed with cubase SX for those without 'proper' ASIO support in their soundcards.

ASIO Scope is the driver you want and I'm not sure if the test app that Interloper pointed you to has anything to do with 'real' Asio drivers.

However you should easily be able to do 4ms (or less) in SFP with that PC configuration, I generally only switch to 7ms towards the end of a project for a bit extra cpu overhead (ie, recording is done and I'm just mixing).

Back to your problem...Initially when I read your post I assumed that this 'noise' and clicking issue was present in SFP itself, but is it the case that it only happens when using Cubase SX?


If so:


Before switching to Logic Audio (yes I know windows version is 'dead' :razz: ) I was a heavy Nuendo 1.x user. Cubase SX's audio architecture was inherited from Nuendo and suffers from having to process audio/plugins ALL the time regardless of whether there is input for the plugin to process or not. In addition Pentium4 chips are a lot more susceptible than previous p2/p3/athlon chips to a sitation referred to as 'denormaling'. As a consequence a P4 + Cubase SX (or nuendo) combination is VERY susceptible to 'denormaling' on plugins that don't check for the problem or include a 'fix'.

You can read more here:
http://phonophunk.phreakin.com/p4denormal.html

or here (a bit easier read, under "A brief overview...):
http://www.digitalfishphones.com/main.p ... &subItem=6

Digitalfishphones also offers a 'normalizing' plugin on that page which you can insert before any offending plugins, but I've found it best to simply stop using plugins which suffer from denormalizing issues.

The way to find the offending plugins: when you experience the problem go through your project and find every instance of 1 type plugin (waves L1 for example suffers HEAVILY from this) and disable every instance of it. If you don't see a change on your cpu meter and/our audio output then this plugin isn't the offender, go to the next plugin and disable every instance of THAT, etc etc etc.

Eventually you'll learn what plugins are susceptible to this problem, in many cases upgrading the plugin(s) will solve the issue. Waldorf D-pole was something I used to use a lot, tho I rarely use it anymore it was present in many old unfinished projects in Nuendo (same audio core as SX again). D-pole was upgraded to 1.6 and STILL had the problem, only with a subsequent 1.61 release was I able to load those projects without immediate problems.

In any case, if this ISN'T related to the problem you're having its still useful info I'm sure since you jumped from a p3 to a p4, its quite likely several plugins you're used to using will exhibit this issue.
User avatar
interloper
Posts: 370
Joined: Sun Mar 25, 2001 4:00 pm
Location: amsterdam
Contact:

Post by interloper »

Very interesting post. I was not aware that there was a device that helped eliminate the denormaling situation.

To give Badger a few more details on my specific situation, I initially had this problem with Cubase 3.7 and at the time I was running SFP 2.01 in a Pentium II 450 Mhz system with 256 MB of RAM (don't laugh, yes the machine was a joke). :razz:

This was the machine I first put a Pulsar II board into, and the problems seemed to be the same. Cubase playback after some random point would go into the land of "Rice Crispies" and the noise went from bad to worse. The video card at the time was an ATI with 8 MB on board (ok, you can laugh now).

As I mentioned before, when switching between Cubase and SFP, the problem either started or became more pronounced. Lowering the latency seemed to help, as apparently it freed up sufficient overhead for the machine to run better, thus eliminating the noise. This is why I asked about the amount of video memory. Common sense would dictate that your P4 should give you enough overhead, no matter what.

Regardless, when I would run the ASIO Multimedia Setup (which it turns out is associated with the wave drivers) :wink:, I would get errors in the simulation. Once the latency was cranked down, the errors in the simulation would go away. Now the kicker is that the crackles would go away as well. Now, I'm not saying that they are related, I'm just reporting it as it occured at the time.

Luckily, my P4 2.4 Ghz with 1 GB of RAM never manifested this problem.

Not sure if any of this helps, but who knows? Now, for the dumb question of the night: are you using the ASIO Scope drivers within SX's Device Setup?
User avatar
bassdude
Posts: 1004
Joined: Tue Jul 24, 2001 4:00 pm
Location: ACT, Australia

Post by bassdude »

I also found that if the ASIO Multimedia test didn't pass then there would be problems even when using the asio scope driver. No idea why.

So before installing cubase, I would just load the background/default SFP project.
Then the cubase install would go through the Asio multimedia test ok and everything would be fine.

The other thing I would check before installing cubase also was that the clock source was set correctly and locked (beit pulsar was master or slave).

<font size=-1>[ This Message was edited by: bassdude on 2003-10-08 04:48 ]</font>
Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

First of all, thanks to all of you (Interloper, Valis, Paul and Bassdude) for your comments/suggestions/info.

In answer to the various questions:

Interloper:

1. Video card in the Carillon is a Matrox 550(I'm using the DVI out into a TFT monitor). Ironically, my old machine (in which the Pulsar was working fine) still has an 8Mb ATI card like your old one installed - and I had no problems with it!

2. Yes, I am using the ASIO Scope drivers within SX's Device Setup.

Valis:

1. No, Cubase SX does not have to be open. The crackles happen whether I have Cubase SX running or not.

2.Thanks for the detailed info, though I was aware of the denormaling issue with the P4. This is definitely not the problem here though, as the noise happens regardless of whether I'm running any plug-ins or not (it can happen with a single MIDI track!). Also (if Cubase SX is running at the time) the CPU load doesn't normally exhibit peaking in the VST Performance meter prior to or at the point where the crackling/noise kicks in.

Bassdude:

1. Pulsar clock is definitely set to run at 44.1KHz as master.

Hmm. It has just occurred to me that I'm not sure the wave drivers were loaded into the project when I tried the ASIO multimedia test (might have been inputs only...) so will check this again to make sure.

And so it goes - thanks again all...
Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

Ah, tried loading in a wave destination module into my scope project (as well as the source module which was already there, even though I'm not using either) and this time all the ASIO MM testing procedure was successful - Hurrah!

Also, have now been running the Carillon for more than two hours without a single snap, crackle or pop in sight (or rather, sound)...

Could the lack of wave dest modules in the scope project/ASIO MM testing have been the problem? Seems very weird to me, especially as I'm using ASIO Scope and not the ASIO Multimedia driver in SX devices - but hey, I'm happy if it solves the problem...

Will report back after more extensive use - Fingers crossed...

Cheers all
User avatar
valis
Posts: 7684
Joined: Sun Sep 23, 2001 4:00 pm
Location: West Coast USA
Contact:

Post by valis »

That's an interesting interaction in SX....
Badger
Posts: 44
Joined: Sun Oct 05, 2003 4:00 pm

Post by Badger »

I'm not sure whether or not it has anything to do with SX. As I said before, the crackles/noise happened with the scope software running without SX.

I wonder whether maybe the scope software (or the pulsar II card itself) isn't happy unless each (at least wave) "source" module has an equivalent "destination" module loaded...?

<font size=-1>[ This Message was edited by: Badger on 2003-10-08 12:21 ]</font>
User avatar
interloper
Posts: 370
Joined: Sun Mar 25, 2001 4:00 pm
Location: amsterdam
Contact:

Post by interloper »

Keep us posted.
Post Reply