The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
FireWire Audio Heat Problem (Macbook Pro, Late 2008) VIDEO proof Audio Interfaces
Old 12th June 2010
  #1
Gear Nut
 
SPiTFiREgr's Avatar
 

Exclamation FireWire Audio Heat Problem (Macbook Pro, Late 2008) VIDEO proof

YouTube - Macbook Pro Unibody (Late 2008 2.53Ghz) - FireWire Audio Heat Problem

After a lot of hours of trying to figure out WHY my macbook pro heats up (a lot) when i connect my newly bought Apogee Duet firewire audio interface, i found out that actually what was happening is that there are CPU spikes of extremely short period (so short that Activity Monitor and iStat could not catch) constantly spiking the CPU frequency from 1596Mhz (Idle) to 2572Mhz (Maximum). Apart from the frequency spike there is also a VOLTAGE spike from 1.0v to 1.15.

I know that this is normal due to speedstep technologies and such BUT the rapid spiking and alterations of frequency and voltage ALWAYS happening when i have my Firewire audio plugged is not normal to me. There is nothing else running, everything's idle and i only plug the firewire interface.

Also this is not happening when i restrict OSX from loading AppleFWAudio.kext (either by deleting, moving, renaming it). I connect the firewire audio interface, the leds are illuminating but no heat problem (thats to show you that is not because the interface is self-powered)

Also, my macbook is using Lucent / Agere (LSI) firewire chipset (the particular and known revision 6) but i dont think that this is the cause since Apogee didnt have any problems with the chipset

Im posting this at Apple's Forums (+ im gonna inform Apogee about the issue) as well as any other audio related forum i know

Info: The app im using to monitor the Frequency and Voltage is CoolBook, can be found here: CoolBook

EDIT: Summing up later info

Also, id like to notice that its not a manufacturer's driver fault because even with the drivers uninstalled, OSX loads up AppleFWAudio and the spikes are the same, regardless if the manufacturer's drivers are installed or not

Quote:
I wanted to monitor the power that was required for Duet to run and to see if the temperature rise was due to that, so:
  • With Apogee drivers uninstalled (just to be sure its not their drivers related, which is not)
  • MBP on battery power with System Profiler opened (to monitor the mA required by the computer from System Profiler -> Power)
  • CoolBook opened to monitor the temps
  • Terminal opened to load/unload the AppleFWAudio
  • GeekBench on desktop to monitor the CPU performance

I did the following tests:

1st Test:

Idling machine. Duet is unplugged, AppleFWAudio.kext is not loaded

Results:

Idle CPU temps: ~40c :: Power Consumption: ~ -900mA :: GeekBench Score: ~3550

2nd Test:

Im plugging in Duet. AppleFWAudio gets loaded. Duet outputs audio

Results:

Idle CPU temps: ~50c :: Power Consumption: ~ -1850mA :: GeekBench Score: ~3450

3rd Test:

With Duet plugged in. I kextunload AppleFWAudio. Duet cannot longer output. The device is disappeared from System Preferences > Audio

Results:

Idle CPU temps: ~41c :: Power Consumption: ~ -1300mA :: GeekBench Score: ~3550 (like 1st Test)

4th Test:

I unplug the Duet. Wait for a minute

Results:

Idle CPU temps: ~41c :: Power Consumption: ~ -900mA :: GeekBench Score: ~3550

5th Test:

I plug the Duet again. The AppleFWAudio is not auto loaded since i unloaded previously. Duet cannot output

Results:

Idle CPU temps: ~42c :: Power Consumption: ~ -1300mA :: GeekBench Score: ~3550

6th Test:

With Duet plugged in, i kextload the ApppleFWAudio again. Duet can output again

Results:

Idle CPU temps: ~51c :: Power Consumption: ~ -1800mA :: GeekBench Score: ~3450 (100 points drop again)

If anyone else with similar problems has the time to do the same tests, it would be nice to have the results.
The commands for the terminal:

sudo kextstat -> Lists the currently loaded kexts
sudo kextunload -b com.apple.driver.AppleFWAudio -> Unloads the AppleFWAudio (dont be bothered about the classes report after unloaded) will disable output
sudo kextload "/System/Library/Extensions/AppleFWAudio.kext" -> Loads the AppleFWAudio and will enable output
By disabling IntelCPUPowerManagement either by deleting the kext or by installing NullCPUPowerManagement i get a constant ~2.50W CPU power either with or without the interface plugged in.

That means the IntelCPUPowerManagement sets the CPU to draw more power for some reason when the interface is plugged. By disabling it the CPU power remains the same as when idle without the interface but because it's disabled, there is no power management for the CPU and the ~2.50W are above the required for powering an idling CPU


With IntelCPUPowerManagement enabled i get 0.95W to 1.30W when idling but this goes up to 3.91W when connecting the interface which is why the CPU produces more heat.

That seems to make sense. AppleFWAudio is coded to do some realtime never ending task for providing glitch free audio. And that's may be the reason of its CPU usage and the side effect of CPU drawing power.

But still then, the power management is not ideal. I cant leave AppleIntelCPUPowerManagement enabled and my firewire interface attached and live with the CPU drawing ~4W of power and 0.97v constantly causing an overall heat increase (since i consider my setup to be mobile, and be able to carry my interface with me for external recordings and be on battery power) when by disconnecting the firewire interface, the CPU power need drops to 0.97 - 1.00 Watts and voltage at about 0.87v (idling conditions). The voltage with firewire attached is even higher than my 0.95v setting i have with CoolBook for the idling frequency (* that maybe also an indication that there is still a rapid speedstep swapping because i've set the higher frequency voltage at 1.00v and the lower at 0.95. So normally, with the CPU at its idling frequency the voltage shouldn't been allowed to go over the 0.95v setting of CoolBook as that is the task of the CoolBook overrider)

Quote:
Originally Posted by SPiTFiREgr View Post
Apple Engineering responded to my Bug Report:

25-Jun-2010 12:01 AM ***** ***** :
Engineering has provided the following information:

This is a maxbusstall issue. FireWire isoch services call requiremaxbusstall(). This is expected.

We are now closing this bug since our engineers are aware of the issue and will continue to track it offline. A different resolution to this issue may be considered in the future.

-----

So anyone can explain exactly what that means? What's maxbusstall exactly? He says that FireWire Isochronous services call requiremaxbusstall(). But what is maxbusstall exactly? Does that means that the stream between the bus and the audio interface never idles?
Guys, i advice you to install iStat Menus and monitor your CPU Wattage requirement. This is the cause of the CPU overheating. CPU Wattage increase when FireWire audio is present and IntelCPUPowerManagement manages the CPU Power. You'll see that when you plug in your FW audio interface, the wattage increases BUT with AppleIntelCPUPowerManagement disabled the wattage stays steady but at a higher level than idle with AppleIntelCPUPowerManagement enabled. Check my previous posts!

For future references:

Without Coolbook or any other speedstep control app/kext:

Idling without Firewire Audio interface connected :: Idling with Firewire Audio interface connected



Anyway i've come to a temporary better situation solution myself:

As i've said before, by disabling AppleIntelCPUPowerManagement the power consumption of the CPU remains steady at ~2.50W either with or without firewire audio attached. This is better than ~3.50 - 4.00 caused with AppleIntelCPUPowerManagement enabled. The only problem i had with disabling it was sleep not working and resulting in kernel panic.

So i remembered that back in the days when i was using a hackintosh, there was another kext that i was installing with the NullCPUPowerManagement and that was SleepEnabler that does the exact work. Enables sleep when there is no AppleIntelCPUPowerManagement

I seeked for it and indeed there is a 10.6.4 compatible version of it.

So right now I am with: CoolBook to have the control of the CPU voltages and disable speedstep on demand (if i want to do live performance audio) + NullCPUPowerManagement + SleepEnabler for 10.6.4 and im at least cooler than before with ~2.50Watts of CPU power consumption and i know that im not wearing off the CPU as much as before

The thing you have to watch out if you wish to follow my temporary workarround is that with every OSX update, it is adviced to enable the AppleIntelCPUPowerManagement and delete SleepEnabler as it may kernel panic your system. But worry not, the Hackintosh community is active as hell and they make any changes required on hackintosh kexts to be compatible with the latest osx updates.

Who could have said this to me.. that i bought a bloody macbook pro that costs 2 months of work to stop hassling with a Hackint0sh setup only to come back to Hackint0sh community for a temporary workarround of a problem that is an Apple software engineering dependent one for their own macbook PRO line and their reflex for a solution to the issue are more slow than a turtle..
Old 12th June 2010
  #2
Gear Nut
 
SPiTFiREgr's Avatar
 

Also, id like to notice that its not a manufacturer's driver fault because even with the drivers uninstalled, OSX loads up AppleFWAudio and the spikes are the same, regardless if the manufacturer's drivers are installed or not
Old 12th June 2010
  #3
Gear Head
 

What temp does it eventually get up to?
Old 12th June 2010
  #4
Gear Nut
 
SPiTFiREgr's Avatar
 

Firewire Audio disconnected / or AppleFWAudio.kext unloaded = Idle temp arround 50 celcius (iStat)

Firewire Audio Connected with AppleFWAudio.kext loaded = Idle temp arround 60 celcius (iStat)

If theres a possibility that this is due to bus powered firewire device (cause i have an external FW hard drive that is self-powered and temps are not rising when i connect it) then AppleFWAudio.kext maybe manages something with bus powered firewire.

Maybe when i have the interface connected but the kext not loaded the mA sucked by the interface are too little for causing a temp rise but by the time the kext is loaded and the interface is converting audio the mA required tops up so the temp rise comes from the mA load on the firewire bus and not the cpu itself.

Even if that's the case, the rapid spikes of throttling on the cpu by the time the Firewire Audio interface is connected bothers me. This isn't something that should normaly happen on an idling machine.
Old 12th June 2010
  #5
Lives for gear
 
valis's Avatar
Looks to me like the cpu has a core getting interrupt calls to service the firewire buffer, and that brings the cpu up out of lower sleep states. 60C isn't a concern on a laptop, and speedstepping is a natural function of cpu's. The real question is, are you able to find 2008/2009 Macbook pro's that can have a firewire audio card connected to CoreAudio without seeing speedstepping affected? If so then I might look into it more, otherwise...
Old 12th June 2010
  #6
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by valis View Post
Looks to me like the cpu has a core getting interrupt calls to service the firewire buffer, and that brings the cpu up out of lower sleep states. 60C isn't a concern on a laptop, and speedstepping is a natural function of cpu's. The real question is, are you able to find 2008/2009 Macbook pro's that can have a firewire audio card connected to CoreAudio without seeing speedstepping affected? If so then I might look into it more, otherwise...
Yeah i might be able.

Also, i made my macbook's speedstepping unaffected of FireWire audio by installing CoolBook and overriding the default speedstep settings (IntelCPUPowerManagement.kext i think) with my own settings-steps. No more rapidly speedstep change even if i leave the default settings but have the original kext overriden by the CoolBook's kext. The thing is that this didn't change the idling temp with the audio interface connected but took away the speedstep rapid changes
Old 13th June 2010
  #7
Lives for gear
 
valis's Avatar
Why are you specifically worried about the speedstep changes?

Aside from that, heat is always generated by the cpu, and macbook pro 'unibody' models with upper speed tiers are renowned for generating a lot of heat. Anytime you raise the voltage/internal clock to work on a higher workload there's going to be a corresponding change in heat output. So to 'fix' the thermal 'issue' you're having you would instead need to create and fix the cpu to a LOWER voltage/clock stepping instead of higher.

If you want to increase cooling you can either use a 3rd party notebook cooler that's capable of transferring heat away (ie, metal--I personally find an older Lian Li cooler to be a good fit for my 17" mbp and since it's aluminum that might be even more so for newer models.) Or you can use something like iStat menus 3 to define a few fan profiles (beyond default/medium) and manually change fan speeds yourself to offset the heat. This is actually a good idea for older mbp's with 8-series gpu's (8600m/8400m) due to well documented failures. But cpu failures are very far from common (at least from my experience) so in your case I wouldn't worry about heat unless it's a personal comfort issue (hot to the touch, lap etc.)
Old 13th June 2010
  #8
Here for the gear
 

I'm having the same problem I think. I'm on a macbook pro 3,1 and everytime I plug my duet into it the fans go crazy, even tho the whole computer is idle. I can't really see a big temp change though.

I have spoken to apogee about this but they have no answers.

I'm very intrigued to hear how they respond to your findings.
Old 13th June 2010
  #9
Gear Nut
 
SPiTFiREgr's Avatar
 

Im worried about the speedstep changes because they are rapid (im monitoring 30times/s and there is a clear indication of constantly changing speedstep setting) and THIS was the problem of MacPro users after the famous MacPro Audio Update 1. Everyone applied that updated and connected a firewire audio interface to their macpro they were getting a constant sine at about 8Khz right out from their MacPro (Google MacPro Audio Update 1). After investigation from apple they figgured out that this was because of constant speedstep changes that was forced by a (or some) kext/kexts. MacPro Audio Update 1.1 was released to fix this issue. Actually if you download and see the update package you're gonna figgure out that the problematic kexts was IntelCPUPowerManagement and an SMC related kext (IOPluginFamily). After that update, no constant rapid changes of speedstep when idling

This is a first clear indication that this is not normal to happen also to my macbook

The second one is that this rapid speedstep change happens ONLY when i connect the firewire audio interface (same thing with the macpro problem). You can watch it happen at the video i posted at the first post

The third one is that after i installed CoolBook which is an app that overrides the speedstep settings by overriding the IntelCPUPowerManagement.kext and uses its own kext the rapid changes when i have the firewire audio interface connected are GONE

This is about the speedstep changes

--

As for the temperature rise, as i said, the rise remains the same even after the CoolBook override of speedstep driver. So that isnt "rapid speedstep change" related.

The one thing for sure is that by attaching my bus powered Firewire audio interface the temps are rising 10c when the AppleFWAudio driver is loaded

It maybe because of the bus powered current required by the card to operate (and i think it needs a lot of mA)

I would be sure if i tested with a firewire audio interface that uses AppleFWAudio but is selfpowered. Or with a selfpowered Firewire hub.. I'll try to test

And yes temp rise is bothering me because of hot to lap actually. My palms are sweating. Its hot here in Greece
Old 13th June 2010
  #10
Lives for gear
 
valis's Avatar
On Core2 era cpu's you won't have the same issues that the Mac Pro 'fix' was released for. Nehelem based cpu's have a separate control mechanism for speedstepping/EIST entirely, the "PCU" (power control unit) which is on the cpu itself and does more than simply clock voltages and multiplier up & down (it turns cores entirely off.) So 2010 era laptop owners might have to worry about something similar, but not 2009.

Speedstepping will affect performance for audio uses somewhat, but again locking the cpu' to a higher (or the highest) setting won't reduce heat. To do that you need to increase cooling to the laptop. And yes bus powering firewire devices may create additional heat by passing more power through the laptop. So I wasn't doubting you there at all, it's just best to separate the issues and deal with each properly.
Old 13th June 2010
  #11
Gear Nut
 
SPiTFiREgr's Avatar
 

Yeah of course there is no PCU on the c2d inside my macbook that its going to whistle at 8Khz like MacPro users experienced but this doesn't cancel the fact that this behavior is an abnormality that needs to be fixed right? I think we agree that its an abnormality right?

Its not happening anymore to me since i installed and activated the CoolBook driver plus i actually undervolted the steps (i have configured 4 steps) to get a little bit cooler CPU. So im ok with it since it wont affect me anyhow badly.

When you say "will affect performance for audio uses" what do you mean? Speedstepping changes in general is bad for audio uses? If that, then lets imagine how bad is the rapid speedstep changes i experienced without the CoolBook driver override! stike
Old 13th June 2010
  #12
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by MMerovingian View Post
I'm having the same problem I think. I'm on a macbook pro 3,1 and everytime I plug my duet into it the fans go crazy, even tho the whole computer is idle. I can't really see a big temp change though.

I have spoken to apogee about this but they have no answers.

I'm very intrigued to hear how they respond to your findings.
Do the fans kick in right after you connect the duet? Download CoolBook to monitor a temp change. Maybe it rises so fast that it forces the fans to kick in fast. Do the fans slow down after some time?

Also with CoolBook monitor the frequency and voltage settings when you connect the Duet. Set an update rate of ~15/s to see if there are any rapid frequency/voltage changes before and after you connect the Duet
Old 13th June 2010
  #13
Lives for gear
 
miro's Avatar
 

ha! i posted something similar longer ago but noone could help. i have the same problem ever since with an echo audiofire. as soon as it's plugged my MBP (late 2006/early 2007) runs hot (fans "loving it" to run even if barely doing something)

less than 1 year ago i gave it to repair. they cleaned the thermal paste-overload (seems that was a common problem at that time...the chips were "drowning" in that paste) and it "seemed" to help me for a while.

anyway...now i still notice the problem as soon as my small soundcard is plugged - which pisses me off!! (especially since i can't just unplug without shutting down before doing so....super annoying!)

oh and i bought coolbook myself (back then) not sure if it really helped.
Old 13th June 2010
  #14
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by miro View Post
ha! i posted something similar longer ago but noone could help. i have the same problem ever since with an echo audiofire. as soon as it's plugged my MBP (late 2006/early 2007) runs hot (fans "loving it" to run even if barely doing something)

less than 1 year ago i gave it to repair. they cleaned the thermal paste-overload (seems that was a common problem at that time...the chips were "drowning" in that paste) and it "seemed" to help me for a while.

anyway...now i still notice the problem as soon as my small soundcard is plugged - which pisses me off!! (especially since i can't just unplug without shutting down before doing so....super annoying!)

oh and i bought coolbook myself (back then) not sure if it really helped.
Is audiofire bus-powered? Whats the idling temperature and whats the idling temperature with the card conencted?

Can you also check with CoolBook if there is a rapid switching with speedstep when the firewire interface is connected? Also, do OSX load AppleFWAudio when you connect it? You can check it out by giving terminal command -> sudo kextstat
Old 13th June 2010
  #15
Lives for gear
 
valis's Avatar
Look at it this way, when the system is idle or doing very little work the firewire driver is causing just enough interrupts to the cpu to kick it up a notch, as if it thinks it might start to 'fall behind' a tad in the queue of buffers its delivering to the firewire chip so speedstepping/eist kicks the cpu up a multiplier/voltage step. There's nothing 'broken' with that per se...the heat exists for you even at the higher permanent setting regardless. So all you're really doing when doing that for idle conditions (assuming it doesn't have issues when hitting start in an audio app) is forcing the cpu to create more heat than it otherwise would.

What I meant about speedstepping causing issues, is for applications that don't properly handle their priority level for the audio engine sometimes find themselves in a similar position. You're causing 2 cores to do work that's near a threshold for the OS to kick the multiplier/voltage 'stepping' up a tad, at which point the workload is finished and the cpu kicks it back down. However in this case you're actually DOING SOMETHING and not just monitoring a graph under idle conditions. So Ableton Live users in particular (or max/msp, Bidule etc) users sometimes find that using the processor prefpane to disable speedstepping (or using coolbook as you did to lock to the highest setting) helps smooth performance when doing live sets.

However the offset is that they need to consider cooling even more, especially in a crowded/hot venue. I prefer setting my mbp's fans to a custom preset that is NOT maxed out and using a separate metal cooler with underneath fans that is driven by a USB 'power' port that I connect to a longer usb cable & a typical mains plug that feeds a single usb port for power alone (cheap device to recharge portable usb devices basically.) Keeps the power strain off my mbp's usb port and helps cool a bit more effectively, even worked in desert heat conditions a few times where other laptops were failing.

However again I wouldn't even bother worrying about a driver kicking the cpu up a speedstepping state just because a peripheral is connected. Deal with your heat concerns, which exist regardless and then see if you have problems during actual use. This is no different than setting any Windows machine to High Performance mode during production work to disable similar features (and again the post-core nehelem hardware actually has the same PCU issues, so some disable enhanced EIST in the bios completely.) There's a lengthy thread or two here on that but you're not even in the generation of machines to worry about that atm. So plenty of documentation on the subject you're looking at imo...
Old 13th June 2010
  #16
Gear Nut
 
SPiTFiREgr's Avatar
 

Yeah i've got what you're saying.

Im gonna actually try to test other macbooks and self powered firewire interface to see if this speedstepping behavior exists so to be sure that is not something that occurs only to me.

I have also thoughts on buying a fan-cooling stand + i probably wont use a firewire interface on live performance scenarios but rather a USB

Thanks for your interest on my issue, valis
Old 14th June 2010
  #17
Gear Nut
 
SPiTFiREgr's Avatar
 

Back to the temperature issue: Its exactly 10c degrees raise when i connect my audio interface..

Is there anyone out there that he has 10 degrees temp raise when he connects audio interface via firewire (if anyone with Duet, would be better)? On a unibody mbp? Late 2008 or not?
Old 14th June 2010
  #18
Lives for gear
 

Quote:
Originally Posted by SPiTFiREgr View Post
Back to the temperature issue: Its exactly 10c degrees raise when i connect my audio interface..

Is there anyone out there that he has 10 degrees temp raise when he connects audio interface via firewire (if anyone with Duet, would be better)? On a unibody mbp? Late 2008 or not?
Yes. I've posted about this before here, but without generating any interest. Mine is a 2010 13" unibody Macbook Pro, and the interface in question is a Mackie Onyx Satellite. When I connect it, I am seeing CPU temps rise 20-25 degrees C above idle levels. I'm not terribly happy about it, but it hasn't caused me any problems per se.
Old 14th June 2010
  #19
Gear Nut
 
SPiTFiREgr's Avatar
 

I also tested with a friend now, he has a 10 degrees temp rise on 2009 macbook pro with bus powered firewire audio.

It seems theres a lot current flowing to these cards. Enough for generating a 10c and above thermal product
Old 14th June 2010
  #20
Lives for gear
 
valis's Avatar
You can always use an external power supply and avoid the added power load to the Mac. Question is if the cpu temps rise regardless (which is possible) but the overall heat generated in the laptop should still be less.

Using external power supplies without breaking bus power has led to ground noise issues for me in the past, so if you experience that you can break the bus power connection from the PC to fix that (6 pin to 4 pin and back again).
Old 15th June 2010
  #21
Here for the gear
 

Quote:
Originally Posted by SPiTFiREgr View Post
Do the fans kick in right after you connect the duet? Download CoolBook to monitor a temp change. Maybe it rises so fast that it forces the fans to kick in fast. Do the fans slow down after some time?

Also with CoolBook monitor the frequency and voltage settings when you connect the Duet. Set an update rate of ~15/s to see if there are any rapid frequency/voltage changes before and after you connect the Duet
give it about maybe a minute.. it varies, but the fans got to 6000 and take some time to slow down after I unplug, and only after I unplug. Never see the cpu get hotter than 60 degrees during this time. Very weird problem.

I'll try the coolbook thingy tomorrow.. no time now.
Old 15th June 2010
  #22
Here for the gear
 

I've got a similar problem: The fan(s) of my MBP (Late 2006) become loud after connecting the Apogee Duet. Speed rises from 2000 to (sometimes) 3600 rpm. This also happens when daisy-chaining the Duet with a 3.5" hard disc that has its own power supply. So maybe it isn't a bus power issue...? (I don't know)

I didn't notice this strange behaviour on MacOSX 10.4. I think it began after my 10.6 upgrade. According to the Apogee Support it's a (Snow) Leopard issue. I read about somebody who reinstalled Tiger - and the problem disappeared. That's why I think this is something that can (and should) be fixed by Apple.
Old 15th June 2010
  #23
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by Tropical View Post
This also happens when daisy-chaining the Duet with a 3.5" hard disc that has its own power supply. So maybe it isn't a bus power issue...? (I don't know)
Yeah this stands for me also. But it doesn't cancel that its a bus power cause. Try unplugging the power supply from the hard drive. The duet will continue to work, so even if on a self powered hard disk daisychained, Duet is sucking power from the firewire port.

Quote:
Originally Posted by Tropical View Post
I didn't notice this strange behaviour on MacOSX 10.4. I think it began after my 10.6 upgrade. According to the Apogee Support it's a (Snow) Leopard issue. I read about somebody who reinstalled Tiger - and the problem disappeared. That's why I think this is something that can (and should) be fixed by Apple.
Where does Apogee state this? I have contacted with their support 3-4 times and they dont know the issue
Old 15th June 2010
  #24
Here for the gear
 

Quote:
Originally Posted by SPiTFiREgr View Post
Where does Apogee state this? I have contacted with their support 3-4 times and they dont know the issue
I hope my post about Tiger/(Snow) Leopard wasn't mistakable. On the one hand I was talking about my own experience and the Apogee support, on the other hand I was talking about a guy who switched back to Tiger which apparently solved his problem.

Concerning the latter: I've read his post in a forum (I'll tell you if I find it again. I'm searching for it.)

Concerning the Apogee support: I don't want to quote the E-Mail that they sent personally to me (I think that's not permitted). But I tell you that the man I was dealing with knew something about a FireWire fan noise issue - though he didn't talk about technical details. He didn't actually blame the OS Leopard/Snow Leopard (I only read this between the lines). But he told me to contact Apple because it's not just a Duet issue but a problem with a lot of FireWire (audio) devices.

I'm wondering if Apple knows about it... I've sent a feedback to Apple. I called them. But I don't know how many people did this as well.

Your discovery (CPU spikes et cetera) is not bad. I just noticed the strange behaviour and didn't find any reason (I only watched the activity monitor and iStat Pro).

As you've told (or will tell) Apogee about all this, maybe they will contact Apple, and Apple will fix it immediately and provide an OS update or a firmware update... (oh, I'm just dreaming)
Old 16th June 2010
  #25
Gear Nut
 
SPiTFiREgr's Avatar
 

What i've just found:

Run Geekbench latest version with firewire interface unplugged: ~3580 score
Run Geekbench latest version with firewire interface plugged: ~3480 score

There is a 100 points decrease when my firewire interface is connected.

Tested with and without apogee drivers, with and without CoolBook enabled, with 10.6.3 and with 10.6.4. Same results

Info: 10.6.4 updates AppleFWAudio.kext but doesn't seem to change anything
Old 16th June 2010
  #26
Lives for gear
 
valis's Avatar
Polling the firewire bus eats a small amount of cpu time, hence the 100 pt difference, and also the previous 'results' of seeing the voltage tick up occasionally (since the cpu was being kicked out of idle each time the firewire bus is polled.) This is how computers work, there's nothing *wrong* here.

It's possible that Tiger kicks in different power management schemes as well btw.
Old 17th June 2010
  #27
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by valis View Post
Polling the firewire bus eats a small amount of cpu time, hence the 100 pt difference, and also the previous 'results' of seeing the voltage tick up occasionally (since the cpu was being kicked out of idle each time the firewire bus is polled.) This is how computers work, there's nothing *wrong* here.
So every firewire audio interface causes a small CPU performance drop (100pt on geekbench ?)? Can anyone else A/B the benchmark? I told a friend to do it with a m-audio firewire interface and there was no points difference. But he also have the temperature issues...
Old 17th June 2010
  #28
Lives for gear
 
valis's Avatar
Potentially they do, it depends on what exactly is being serviced when the device is polled (ie, feeding 8-12 channels of bidirectional audio is going to show a marked difference to a simple stereo firewire device.) Fwiw *every* computer peripheral, onboard or otherwise, potentially eats a small slice of cpu time if it's not powered down. This is just how computers work.

The 'heating' issue still seems to be related to bus power to me, have you tried using a 4 pin adapter to remove bus power and power externally, then see if the laptop still 'heats up'?
Old 17th June 2010
  #29
Gear Nut
 
SPiTFiREgr's Avatar
 

Quote:
Originally Posted by valis View Post
Potentially they do, it depends on what exactly is being serviced when the device is polled (ie, feeding 8-12 channels of bidirectional audio is going to show a marked difference to a simple stereo firewire device.) Fwiw *every* computer peripheral, onboard or otherwise, potentially eats a small slice of cpu time if it's not powered down. This is just how computers work.

The 'heating' issue still seems to be related to bus power to me, have you tried using a 4 pin adapter to remove bus power and power externally, then see if the laptop still 'heats up'?
No, can i do this somehow without the need of a firewire hub? Using a 4 pin adapter will not transfer bus power but how can i then feed the interface with power?
Old 17th June 2010
  #30
Here for the gear
 

I think this is the one I was talking about (TC Electronic Forum). I don't post the link but you will find it if you look for it. It's not really up-to-date (2008-06-04/09) and it's not about the Duet, but it's interesting.

Re: Fan speed on a Macbook
I have the same problem with the konnekt 24d. It´s really annoyning, the fan speeds up just by puting in the firewire cable. By disconnecting the firewire cable, the fan speeds down... After some research in the internet, there might be a problem with the macBook firewire chipset, but not for sure....


5 days later he writes:
...well, I unistalled Mac OX 10.5 Leopard and installed 10.4 (Tiger?) just to see what happen. And it works...no increasing temperature and no nervy fan noises. It´s seems like a leopard problem. Why are there problems with the newest version, but not with the old version?
MacBook 2.16 ghz, 2007, 2.5 gb ram



Quote:
Originally Posted by SPiTFiREgr View Post
Also, my macbook is using Lucent / Agere (LSI) firewire chipset (the particular and known revision 6) but i dont think that this is the cause
Seems to be right. There are two people on MacRumors ("MacBook and FireWire Audio issues") with TI chipsets having similar problems - not with the Duet but with different audio interfaces (at least the second one):

1)Aug 28, 2008
(...) I checked and I have the TI chipset and my Audio interface works fine but I still get high fan speeds when it's connected. Any info about this issue?

2)Mar 31, 2010
I too have the TI chipset and whenever I connect my m-audio Firewire 410, the fanspeed goes 6000 rpm within few minutes and stays there...


And here's someone on LogicProHelp:
Posted: Sun Mar 15, 2009
Post subject: Focusrite Saffire Pro 10 and CPU temperature/fan speed
Hi All,

I have been battling with fan noise on my 2 year old MBP 2.33 running Leopard 10.5.6 ever since upgrading to the newest operating system.

I have a Focusrite Saffire Pro10 that is always connected to the FW400 port. I have nothing connected to the FW800 port (two eSata drives are connected via the ExpressCard slot and a Sonnet card).

Recently I have noticed that with the Saffire switched off, the MBP idles well at 2000rpm and a reasonable 55-60 degrees C "measured" by iStat unless under fairly normal use.

However, as soon as I simply switch the Saffire on, the CPU temp jumps to 70 C and within 5 mins the fans are churning away at a very audible 4500rpm+ with a 70+ temp.

Switching the Saffire off again, fan speed and temperatures are back to normal within 5 mins. Throughout this basic test, CPU usage never steps beyond 8%.

I realise that the Firewire bus and the Audio engine will use some CPU cycles, but why should it bump CPU temp and hence fan speed so much without a significant CPU usage increase?

It's a concern, as the fan noise is considerable and obviously I prefer that the CPU is as cool as it can be and that I'm not causing needless wear of the fans.

I wasn't aware of this problem when running the Saffire under Tiger. My MBP has the TI chipset, so that shouldn't be a problem ...

Is there an issue between the Saffires and Leopard? I don't want to have to give up on all my shiny Leopard features simply to run the Saffire (I'm also considering upgrading to a Mac Pro in the near future and so would prefer to keep the Saffire ...)
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
echoromeo / Music Computers
11
ISedlacek / Remote Possibilities in Acoustic Music and Location Recording
12
Timur Born / Music Computers
1
laperlestudio / Music Computers
3

Forum Jump
Forum Jump