Set the ASIO buffer size to 48-samples. Compare performance (running heavy load) when connected via integrated USB-2 vs. connected to a 3rd-party USB-3 port. You can run heavier loads (glitch-free) when connected to the integrated USB-2 controller.

+1

Tested and confirmed also, with the Quad.  In fact after testing I now have my USB3.0 capability turned off until I find something worthwhile to plug into it.

For further clarity too Freddie the -95dBu level you quoted for that unit is residual noise. IOW you don't hear any...  When comparing stated figures it is always best to quote like for like...

This makes zero engineering sense. We have a brand new ASUS board with 2(!) PCI slots taking up valuable real estate and Lord knows what else. The OP is trying to reliably stream a total of 352,800 bytes through his 1010 card (44100 * 4 bytes * 2 channels). That's 344Kbytes/sec, or ~ 0.37 Mbytes/sec (out of a theoretical max of 133 Mbytes/sec). Are we saying that this brand ASUS motherboard can't support a continuous load of 0.37 Mbytes/sec of data from one of the 2(!) PCI slots?!? I simply don't believe that. If it's true, then nobody should buy ASUS, because they obviously don't know what they're doing. Like the 2(!) useless PCI slots were who's idea, exactly? If there are specific technical reasons (from ASUS, Intel, whoever), then I'd be interested in knowing what those are -- because from my perspectiove the 2(!) useless PCI slots represents an Epic Engineering Fail. It'd be like buying a new car with a 500W surround system, but the radio itself only does AM… so you need to go out and buy a satellite radio setup. :^/ Which is great if you wanted to upgrade to a satellite setup anyway, not so great if you expected the freaking FM receiver to work.

 
Hi Keith,
 
This situation isn't limited to Asus.  It's across the board...
 
This scenario is analogous to Firewire - the associated issues when using "rogue" (crap) chipset controllers vs a quality TI chipset unit.  The *bandwidth* is the same... but compatibility issue/s impede performance with rogue units. 
Same situation with 3rd-party USB-3 controllers...
 
Note that some combinations work just fine.  We have numerous clients using RME, M-Audio, and Echo PCI units with SandyBridge based DAWs.
In principal, yes... all should work as expected.  But that's not the case... and (again) it's not limited to Asus.
BTW, I'm not defending PCI bridge chips. 
The situation is what it is...

Jonbouy

In principal, yes... all should work as expected. But that's not the case... and (again) it's not limited to Asus. BTW, I'm not defending PCI bridge chips. The situation is what it is...

Bang on again.

For some strange reason PCI works well on my particular board.  Sadly though getting updated drivers for PCI stuff is becoming increasingly more difficult.

Just another reason why PCI has become legacy as far as I'm concerned.

Just another reason why PCI has become legacy as far as I'm concerned.....and it  has been replaced with PCI express interface that do work across the board. PCI express have much better bandwide then both firewire, USB2 or PCI.
 
So to get the best of all worlds. Use a RME PCI-express ADAT I/O-card to connect to a external converters like SSL Alpha.

Sorry my suggestion started an argument......I am not sure what Cakewalk is currently/officially recommending regarding Core Parking, however, I responded with the suggestion as I had the exact same symptoms as the OP, whereby my system was playing for short periods of 2-20 seconds and then stopping.   I was also getting pops and crackles but the main problem was that the transport was stopping.  This issues started after updating my machine.   
 
I have a RME FF 800, a separate Fireware Card (TI Chipset) and a ASUS ASUS M5A99X EVO AMD 990X AM3+ Motherboard - AMD Phenom II 1090T Black Edition Six Core Processor - 3.20GHz,  running Windows 7 64 bit (though I run the 32 bit version of Sonar)  After trying several other "fixes" including driver updates and system optimization, I ran across the Core Parking thread, made the changes to the Registry and the problem was resolved.   I had this problem on both Sonar X1 and Sonar 8.5.3.  As it is a 30 second fix I thought it might be worthwhile to give it a try.   I don't really see any downside turning off Core Parking on a Desktop machine, it is really designed for power conservation and more relevent for laptops...
 
Regards

markno999


Sorry my suggestion started an argument......I am not sure what Cakewalk is currently/officially recommending regarding Core Parking, however, I responded with the suggestion as I had the exact same symptoms as the OP, whereby my system was playing for short periods of 2-20 seconds and then stopping.   I was also getting pops and crackles but the main problem was that the transport was stopping.  This issues started after updating my machine.   
 
I have a RME FF 800, a separate Fireware Card (TI Chipset) and a ASUS ASUS M5A99X EVO AMD 990X AM3+ Motherboard - AMD Phenom II 1090T Black Edition Six Core Processor - 3.20GHz,  running Windows 7 64 bit (though I run the 32 bit version of Sonar)  After trying several other "fixes" including driver updates and system optimization, I ran across the Core Parking thread, made the changes to the Registry and the problem was resolved.   I had this problem on both Sonar X1 and Sonar 8.5.3.  As it is a 30 second fix I thought it might be worthwhile to give it a try.   I don't really see any downside turning off Core Parking on a Desktop machine, it is really designed for power conservation and more relevent for laptops...
 
Regards

Like I said, it is probably worthwhile on some machines and not others.  For example on my Sandybridge machine CPU core/thread utilization 'appears' to be more evenly distributed with it left on, whatever the reason I get a few clicks and pops with it turned off which go away when I leave it on. In those circumstances what would you do?

My suggestion was is that it is worth testing on your particular machine (worth doing if you've just dropped $$$ on the hardware) to get the best out of it.  John has some personal vendetta toward me which is already well-documented so I apologise if it looked like I'd dragged your comment into what looked like an argument. btw have you notice John's comment on the thread you linked to, it just shows we have to work out what particular John we are dealing with on different days...

http://forum.cakewalk.com/fb.ashx?m=1902837

I have asked for some evidence where this idea of turning off core parking on 2nd gen boards is a 'no brainer' and as yet nobody has provided any, but I'm willing to see if there is a situation where the benefits are conclusive.

Till then I'm good to go on 48 samples buffer without glitching or 64 samples for heavier loads (which is where I generally leave it).

Turbo boost in bios is the 1st thing that pops up.. turn it off.....

Gotta love the internet 'experts'...

StarTekh


JonBouy ...Nice comment but as always... No Solution ! Do you do this at the Pro Tools fourm too !!

Evidence.

The best I got is a comment direct from Scott stating to keep TurboBoost ON on a Sandybridge (2nd gen) board.  TBH though I didn't take his word for it I tested.

Again the mere fact of the figures on my working (test) machine bears it out in practice.  I base all my findings on actuals rather than tons of conflicting articles I may have read.

Again, supply the evidence. btw I don't use Pro Tools or have occasion to visit their forum.

I'm not trying to argue here, I just want to see where this stuff can make my machine run better than it does already.  Hang on though 48 samples is as low as I can go. 100% reliability is also as good as I can get it.  Bummer.  No improvements for me then...

All these things are simple to test for yourself.  Recommended! That's the solution I provided and you seem to have missed that in providing yet another unsubstantiated 'claim'.

"Jonbouy"
The best I got is a comment direct from Scott stating to keep TurboBoost ON on a Sandybridge (2nd gen) board. TBH though I didn't take his word for it I tested.

Speed Step/Turbo Boost is overclocking. Smart overclocking? Maybe. But again these technologies are aimed at laptop users, where the benefits are the greatest (heat management requirements etc.). I'm starting to think that it may be a good idea to make a distinction between desktop and laptop machines when talking about tweaks.

I just dumped my Delta 1010 when I Got my new System a few months ago.  I picked up a Motu 2408 MK3 (PCI-e) and it's been perfect.. no issues and is cost effective.  ; )

Do you have the Asus AI Suite installed?  This will cause audio glitches and drop-outs guaranteed.

Before launching Sonar, look for the Asus AI Suite notification icon in your system tray, right click it and exit out of it.  Then try launching Sonar and see how you go.

Despite all the comments about PCI sound cards - you should only have a problem if you have another PCI device saturating the bus.

saaricom
I cannot get the song to play without consistent (every 15 seconds but never at the same timeframe) dropouts on playback. This PC was literally built last night with only Windows 7 and Sonar.

This may or may not be relevant, but got similar performance after installing Windows 7 for the first time on an existing dual-boot system.  It turned out that disk mode in the BIOS had been set to IDE instead of AHCI and this is unsuitable for Windows 7.  It's worth checking.

Kev999

saaricom
I cannot get the song to play without consistent (every 15 seconds but never at the same timeframe) dropouts on playback. This PC was literally built last night with only Windows 7 and Sonar.

This may or may not be relevant, but got similar performance after installing Windows 7 for the first time on an existing dual-boot system.  It turned out that disk mode in the BIOS had been set to IDE instead of AHCI and this is unsuitable for Windows 7.  It's worth checking.

It's not unsuitable for Windows 7 per se SATA drives will run just fine set at IDE mode at a good sustained transfer rate but if you want hot swapping etc you'll need AHCI turned on.  Preferably before the OS goes on.

Again if you want to test it out and you are stuck with an error message at boot time after you try to change to AHCI mode in the bios see here first.

http://support.microsoft.com/kb/922976

soundtweaker

The other cool thing about USB3 is that it provides up to 100 watts of power to devices. That should be enough to power larger multichannel soundcards without the need for an AC adapter.

Or powering 3 or so sata devices w/out the need for wall warts and power bricks and extra power cables out of the PC… :)  


Seriously, I was just looking the other day for a single ~100W dual channel supply that I could use to power 2-3 external drives in my own enclosure. They don't exist. And if they do they're hard to find for the average consumer. Like you have ATX power supplies, then… ???.  


So I could use the USB3 cable to power my enclosure, and do the actual data transmission over eSATA… Hmmm… :)
 

Hmmm.... This thread is obviously a subliminal messaging plot for the cerial industry!

You haven't noticed??? Snaps, crackles, pops, Kellogg (Seth), and now "cheerioss"?!

Hijackers! Where's the forum police when ya need 'em?

... eating breakfast!

I just upgraded from my old Delta 1010 to the M-Audio 2626 with a similar rig (SSD, 12 GB, Win 7 Pro S2, Sonar X1 Expanded) and it works fine. My inputs had gone bad on my Delta 1010 so hence the upgraded - outputs had gone as well, but I fixed those myself with new caps - wasn't sure on the inputs so I just upgraded. Been a blast. Wish you the best. Joel