Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[paulkeeley]

As the previous M-Audio FW thread (https://macos9lives.com/smforum/index.php?topic=6942) has now been locked, I'm continuing my saga here.

I've now been able to get both a FW Audiophile and a FW 410 downgraded to run the latest available OS9 drivers and have them recognized on my PowerBook Pismo 400.

Here's the process which works for both devices (requires a G4 running OSX, max 10.4.11):

• First, I ran the FireWire_OSX_1_6_Build_1174.dmg (attached here) on my Aluminum PowerBook G4 running OSX 10.4.11.
• When complete, I shut down and plugged in the audio interface (per the post-install prompt).
• Upon booting up the G4, the interface flashed its blue status light until the OS was fully loaded, upon which time, the light went solid blue. The FW Audiophile stayed solid blue, wile the FW 410 flashed each of its front panel lights in succession before resting at the solid blue status. At this point, each of the interfaces were successfully downgraded to build 1174.
• Next, on the Pismo, I installed the latest OS9 drivers for each interface. 1.0.2 for the FW Audiophile, and 1.0.4 for the FW 410.
• After installation, I shut down, plugged in the interface and booted up. The same flashing-light-to-solid-blue process occurred, after which each interface was fully recognized by the system. Successful OS9 downgrade was now completed successfully, with each device's control panel showing the correct version number and working as expected(ish).

However...
Here's the issue I have experienced running OS 9.2.1 on my PowerBook G3 Pismo 400:

The audio output has regular periodic glitchy audio spikes every second. This happens regardless of whether I use Firewire bus power or mains power, regardless of which FireWire port I use on the back, and regardless of which FireWire 400 cable I use - I have tried multiple known-good cables. I have tried using the Macintosh audio system and ASIO drivers. No matter what, the same glitches occur - even on a fresh OS install.


So, it appears that both of these devices are a no-go on the Pismo 400, at least using the built-in FireWire ports. I have yet to try a PCMCIA FireWire card to see if that works (trying to source a compatible one for cheap). If anyone has had any successful experience with this particular configuration, please do let me know! It would be great to be able to do some multi-channel recording on the little tuxedo. As it stands, I now have two very stylish paperweights... although they do at least work on the PowerBook G4…

[aBc]

As both the M-Audio FW410… and the M-Audio Audiophile note the same basic system requirements of a 500 MHz CPU, could this possibly be the root cause of the problem you’re having? Initially I considered (as did you) the possibility of a bad cable — but then upon reading that you encountered no problems with the same configuration on the PowerBook G4, it seemed a sole problem with the Pismo 400.

[Knezzen mentioned previously that he’d had prior success with a Pismo 500.]

Perhaps trimming all unnecessary extensions, control panels, etc. from your Pismo 400 might possibly alleviate or reduce the problems somewhat and allow you improved performance with both the FW 410 and the Audiophile? Anything to reduce the primary basic demands on your Pismo’s CPU — which might then enable more suitable results with the Pismo 400.

What CPU present in your G4 Powerbook? 500 MHz or greater?


Thanks for the FireWire_OSX_1_6_Build_1174.zip attachment!

[paulkeeley]

Aha - I was not aware of the 500mhz minimum requirement! That would certainly explain the issue, since I don't have the 500Mhz upgrade in my Pismo 400. Hard to imagine 100mhz would make that much of a difference, but doing a comparison on everymac, it seems that the cache bus speed is also higher on the 500: 200Mhz vs 160Mhz. So, those two factors may well be enough.

Guess I'll pause my pursuit of FW audio on the Pismo until I can find a speedier one!

[ssp3]

Thanks for the FireWire_OSX_1_6_Build_1174.zip attachment!

One has to wonder where did he get it? 
https://macos9lives.com/smforum/index.php?topic=6942.msg58696#msg58696

[indibil]

Aha - I was not aware of the 500mhz minimum requirement! That would certainly explain the issue, since I don't have the 500Mhz upgrade in my Pismo 400. Hard to imagine 100mhz would make that much of a difference, but doing a comparison on everymac, it seems that the cache bus speed is also higher on the 500: 200Mhz vs 160Mhz. So, those two factors may well be enough.

Guess I'll pause my pursuit of FW audio on the Pismo until I can find a speedier one!

https://68kmla.org/bb/index.php?threads/just-overclocked-my-pismo.23546/

https://web.archive.org/web/20080224053124/http://www.bekkoame.ne.jp/~t-imai/pbg3e3.html

[paulkeeley]

Coming back to this after taking a break to pursue other activities that required less hair pulling…

Since apparently I can't leave things alone, I decided to connect the Firewire410 again (to get OMS MIDI working to test MIDI timing in other DAWs, which is another topic entirely). After getting the same glitching audio result as before, while running Ableton Live 4 I decided I'd try to load up the M-Audio Firewire control panel to change buffer size, bit depth etc while Live was still running to see if it made any difference. While going down to 44.1khz/16-bit at 2048 buffer setting made things slightly better, the periodic audio crunching/static persisted. Of course, running both at the same time resulted in the M-Audio control panel crashing and Finder telling me it needed attention, but Live kept playing.

And here's the AHA moment:

The app switcher icon in the top right was flashing between Live and Finder (due to the crash and Finder needing my attention) at the EXACT same interval as the audio static bursts. I tried stopping/starting playback to be sure and it always lined up. Turns out it's happening exactly every second.

So, it's very clearly an OS-level issue, and very clearly tied to the system clock somehow, since it's bang-on every second. I don't know if that's the interval that OS9 is polling applications to perform its multitasking, or if that's just a quirk of the Pismo.

I managed to stumble upon an old post on 440forums.com of someone with a 500mhz Pismo having the same problem and apparently M-Audio was unable to solve. Note: I tried disconnecting the power supply as suggested. It did not make any difference:
https://en.440forums.com/forums/?site440=forums&showtopic=17341&pid=111548&mode=threaded&start=


In summary, it seems like it doesn’t matter how fast the CPU is - the Pismo running OS 9.2.2 does not play nice with M-Audio firewire interfaces. Does it work with other devices? That’s something for someone else braver than I to answer.

[smilesdavis]

TiBooks will do just fine

[paulkeeley]

Maybe! I'll have to test that theory out if/when I'm tempted enough to add a TI to the collection. Just a shame that the Pismo has firewire ports but isn't powerful enough to properly use them for audio in OS9. I imagine for external HDDs and interfacing with video devices for file transfer it works great, which I suppose was the intended purpose.

I suspect, as with most things audio-wise, it's better under OSX, but where's the fun in that?

[paulkeeley]

Just adding a quick post to report that under OS 10.2.8, Firewire works fine, despite the M-Audio installation software stating that 500mhz is the minimum supported CPU.

So, a Pismo (400) does indeed work, just not under OS9...

One last thing I want to try is to find a PCMCIA Firewire card and see if it works. The hunch is that the Firewire and IDE bus are sharing the same OS polling interrupt and conflicting with each other somehow.