• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Firewire Audio interface latency too high

R

reasoN

Weaksauce
Joined
Dec 3, 2006
Messages
78
I really hope someone understands my problem and can help me with this.


My old system (Shuttle SP35P2 pro) allowed me to use low audio buffer size settings e.g. 256, 128, and even 96, without clicks and pops. My new system (in my signature) will not let me run my Audio interface with a buffer size lower than 512. I've been using 512 but my latency is at around 33ms which makes it hard to play on my midi controller and be creative. Does anyone know how I can improve my firewire connection? I even purchased a PCIe firewire card thinking it would help, but the latency is virtually the same as the onboard firewire. I may have to get a new system if I can't fix this issue.

I'm using a Motu Traveler Firewire Audio interface. My motherboard is a Gigabyte EP45-UD3P. The mobo has an onboard Texas instruments Firewire controller, the PCIe firewire card also has a Texas Instruments processor.
 
To help diagnose your problem, run DPC Latency checker:

http://www.thesycon.de/deu/latency_check.shtml

Ideally, you will see mostly green, with only occasional yellow/red spikes.

I used this to help troubleshoot audio dropouts on my MBP under Win7. For me, it turned out that the Apple Bootcamp utility that controls the LCD/KB back light is borked for multicore CPU's. Forcing that to run on one core fixed my issue . You might be able to track down a similar situation on your system (or eliminate software/drivers as being the issue). WiFi drivers are notoriously bad for causing latency spikes, BTW.
 
Thanks for the reply

My latency is inconsistent. I will continue trouble shooting.

dpclatencytest2.jpg


dpclatencytest.jpg
 
You're welcome.

You're second screen shot looks like a high-latency situation, although it's hard to tell in such a small time window. What's the trend over 4-5mins look like? It's also of value to boot into safemode and check your latency under minimal conditions.

I google'd your MB and I noticed that one of the features is labeled as "dynamic energy saver". Total shot in the dark, but can you disable that in the BIOS and test?
 
Hey Laird,

The DLC test over 5 minutes is pretty sporadic. Dynamic energy Saver" is a software utility, I never installed it. I will try a test in safe mode and will report back.

These are on top of my culprit list:
1. Motherboard (the motherboard in my previous Shuttle computer gave me flawless audio/video performance.
2. Windows 7 Professional (I used Ultimate on my previous shuttle with no problems)
3. My graphics card (disturbing bus traffic?)
 
Was your shuttle x64 as well?

My Motu would act really stupid until I changed my firewire driver (for the controller) to:
1394 OHCI Compliant Host Controller (Legacy)

It is part of win7's built in drivers. Worth giving it a shot if you're using a driver from Gigabyte or whoever built your pcie card.
 
Yes, I ran Win7 Ultimate x64 with my Shuttle. The onboard firewire and the pcie card are both Texas Insturments chips, windows found the driver automatically and says that I have the most up to date version.

I want to mention also that with this build I've also encountered problems when using Pro Tools LE. I get these buffer under-run and cpu halt errors. It's really disappointing as this build was supposed to be an upgrade from the Shuttle. I miss the solid performance, but I don't miss the noise.
 
Just to confirm, are you running these tests with your CPU OC'ed or at stock?

doctah's advice is worth trying, as well as testing using Windows standard VGA driver.
 
reasoN said:
... windows found the driver automatically and says that I have the most up to date version.
Click to expand...

Right...mine did the same. I had to manually switch it to the "legacy" driver.
 
I'm testing without overclocks, everything is at stock settings. I uninstalled my video drivers, that didn't help. I noticed that disabling "Automap Propellerhead" under "Sound video and game controllers" in the device manager made things more stable but I still get clicks/pops once in a while. I can't find the option to switch to legacy firewire drivers.

This build is really making me upset. Even if I solve my firewire latency issue, I will remain baffled with the instability of Pro Tools LE which doesn't even use firewire (original Mbox USB powered)

I will try my buddy's Motu Ultralite MK3 and see if the problem is with this old Motu Traveler. If that doesn't work, I will format my computer and go back to Windows 7 Ultimate. If problems persist after formatting, I might have to turn to Apple. EDIT: or another motherboard/cpu/ram combo.

This computer is completely stable with gaming, but hinders my creativity in Audio/Video/Music applications.
 
Last edited:
To test with the legacy FW driver go to device manager, expand the IEEE 1394 bus host controller entry, right click on your FW controller(s), select update driver -> browse my computer -> let me pick from a list of drivers.

It is interesting that you're getting dropouts with your MBox, too, though. Have you checked your DPC latency in safe mode? Sounds like it might be software/driver other than 1394.

Another area to check is whether your CPU is throttling, might want to try disabling C1E and EIST in the BIOS.

BTW, which audio host are you using with your MOTU?

Good luck, I understand your frustration with this problem!
 
Also, don't make the jump to Apple hardware too quickly if you're planning on running windows. They are using the Agere FW chipset and it's ability to play nice with sound cards under bootcamp is terrible. I ultimately had to trade in my MOTU Ultralite mk1 for the Ultralite Hybrid w/ FW & USB to get things stable on my 2009 MPB.
 
I still have to test DPC latency in safe mode. I'm not getting audio dropouts with the MBox, I'm getting errors that halt playback. Something like "CPU waited to long" and errors that told me to increase buffer size and then when I increased I'd get an error telling me to decrease buffer size, basically a run around.

I'm using Ableton Live mainly with Reason in rewire mode. I have a Novation 49SL MKII with Automap Pro software.

Hopefully my buddy can stop by with the Ultralite Hybrid soon for testing. What host do you use, and what do you set your Sampling frequency and Buffer size to? Also what is your total latency?

This is Ableton Live's latency when Sample rate is set to 48KHz and Buffer Size is at 128 samples:

Input Latency: 3.65ms
Output Latency: 5.50ms
Overall Latency 9.15ms

I'd like to add that after disabling "Automap Propellerhead" in device manager I am now experiencing minimal clicks/pops/dropouts in Ableton with the above settings. However, I'm guessing I can't use Automap pro with Reason now. I'd like to thank you again for showing me DPC latency tester.



I believe this was ONE of the 2 or 3 errors in protools:
"The operating system held off interrupts for too long. If this occurs frequently, try increasing the “HW/Buffer Size” in the Playback Engine panel"




EDIT: DPC latency checker would not open in safe mode.
 
Last edited:
I'm running Cubase 5 with the MOTU Traveller Hybrid, connected via FW to my Q6600/Asus P5K-E box. I'm currently running at:

44.1/24bit, 128 samples, input latency=4.1ms, output latency=6.2

which is equivalent to your latency at 48kHz.

I was working on a small project this morning, I left DPC latency checker running just for comparison, it stayed firmly in the green with latency peaking at ~360us.

I'm not up to speed with Pro Tools, so unfortunately I can't help there.

Any luck testing with CPU throttling disabled?
 
How about disabling processes/hardware one by one? A bit time consuming, I know, but should at least get to the bottom of whether it's software/driver problem or hardware problem.

And to get back to basics, have you run Memtest86?
 
I ran memtest and it reported no errors. I have to do the process of elimination still.
 
I have read of anecdotal evidence that certain FW chipsets and certain FW devices do not play well together. Note I said anecdotal, so take it with a grain of salt. Might be worth trying something with a different chipset than the TI.
 
You must log in or register to reply here.
Back
Top