Menu
▾
▴
motufw-devel — This is the discussion forum for motufw developers
You can subscribe to this list here.
| 2006 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(6) |
Jun
|
Jul
|
Aug
|
Sep
(2) |
Oct
(18) |
Nov
|
Dec
|
|
From: Mathias L. <mat...@go...> - 2007-10-14 08:51:17
|
Hi Francois, thank you. BTW, your website pasd.fr shows phpinfo() output! Kind regards Nudge On Samstag, 13. Oktober 2007, Francois.ernoult wrote: > Hi guys, > > I finally found a couple of minutes to write it. The attached patch makes > the Main-out and Mic-in ports declared before the others. Now they are > first in the Jack's port list and clients that automatically connect > themselves use them. > > Regards, > Francois -- Votum Solvit Libens Laetus Merito Mathias Lieber <mat...@go...> |
|
From: Francois.ernoult <fra...@pa...> - 2007-10-13 01:10:40
|
Hi guys, I finally found a couple of minutes to write it. The attached patch makes the Main-out and Mic-in ports declared before the others. Now they are first in the Jack's port list and clients that automatically connect themselves use them. Regards, Francois |
|
From: Jonathan W. <jw...@ph...> - 2007-10-03 00:26:20
|
Hi Francois Ah, you're still up. :-) > > > It's difficult to say. With a complete Ardour project full of plugs, I have > > > to set buffers to 1024, but it's the same with Cubase/Windows. The only > > > difference is that FFADO looses the sync and rarely recovers it where the > > > ASIO driver just clicks. > > > > Are you saying that there are audible glitches under windoze with this > > setting? > > With a project full of plugins, yes there are. But every config does. > So it's more a processor charge problem, and the amount of proc time > available for plugins to do the job. This is good to know. > Windows has nothing to do with a realtime OS You can say that again. > When I just received my 828mkII, the current ASIO driver was loosing sync > in that case (with the well known high pitched noise). The problem was > solved after a few revisions of the driver; the noise was replaced by > silence and the driver now manages to recover the sync in a very short > time. Again, all good to know. Thanks for this. > But most of the time under windows, when the number of plugins starts to be > too high for the machine they are audible glitches, which is solved by > setting bigger buffers. This is the normal comportment for a DAW. True. Regards jonathan |
|
From: Francois.ernoult <fra...@pa...> - 2007-10-03 00:14:03
|
Hi Jonathan, > > It's difficult to say. With a complete Ardour project full of plugs, I > have > > to set buffers to 1024, but it's the same with Cubase/Windows. The only > > difference is that FFADO looses the sync and rarely recovers it where > the > > ASIO driver just clicks. > > Are you saying that there are audible glitches under windoze with this > setting? With a project full of plugins, yes there are. But every config does. So it's more a processor charge problem, and the amount of proc time available for plugins to do the job. Windows has nothing to do with a realtime OS, so the driver also has problems when the proc charge grows. When I just received my 828mkII, the current ASIO driver was loosing sync in that case (with the well known high pitched noise). The problem was solved after a few revisions of the driver; the noise was replaced by silence and the driver now manages to recover the sync in a very short time. But most of the time under windows, when the number of plugins starts to be too high for the machine they are audible glitches, which is solved by setting bigger buffers. This is the normal comportment for a DAW. Regards, Francois |
|
From: Jonathan W. <jw...@ph...> - 2007-10-02 23:41:04
|
Hi guys > > did you experience that the numbers of glitches or unsyncs grow > > with the load/channels or was it independent from them? (Thinking > > about how stable the device might be at home and then in the studio). > > It's difficult to say. With a complete Ardour project full of plugs, I have > to set buffers to 1024, but it's the same with Cubase/Windows. The only > difference is that FFADO looses the sync and rarely recovers it where the > ASIO driver just clicks. Are you saying that there are audible glitches under windoze with this setting? The long-term intention is that ffado will also be capable of recovering sync in these circumstances. While it works sometimes it's not reliable, but this has more to do with the underlying streaming infrastructure than the MOTU code itself. Having said that there are some minor issues in the MOTU code which I'm still working through. > > What the thing about the 1kHz tone? Is it necessary for keeping > > up the stream or is it for test purposes only until the stable libffado is > > out? > > It's just a test signal to see if it works without using a jack client to > generate one. It's here for a long time ago, when the motufw driver wasn't > part of FreeBob/FFADO. While not needing to start a jack client is one benefit, the main reason it was done this way was to facilitate testing with ffado's streaming test applications. These are separate to jack, so jack clients don't help. The other reason I use it is because it's far easier to hear non-fatal sync glitches with a sine signal than with music. Regards jonathan |
|
From: Jonathan W. <jw...@ph...> - 2007-10-02 23:35:54
|
Hi Nudge > did you experience that the numbers of glitches or unsyncs grow > with the load/channels or was it independent from them? That's an interesting one - it's actually been rather difficult to nail this one down. I *think* loading has some bearing on things but I suspect that loading is not the root issue. There is something in the ffado streaming subsystem which is sub-optimal and that once this is addressed I think all ffado systems will become far more stable. I have seen that the period size (jackd's "-p" flag) and the number of buffers (jackd's "-n" flag) can affect the ease of triggering a sync loss. I use "-p 1024 -n 4" when doing live event recording and this appears to be quite stable. The latency is very high though. I have found "-p 256 -n 3" to still be usable on my system. "-p 128 -n 3" is flaky, and any period size below that is unusable (sync is lost almost immediately). This is on my system - if your system is slower then the numbers at which the behaviours kick in will most likely be different. Having said all that, the above behaviour is not limited to the MOTU devices. BeBoB devices suffer a similar problem, which is why the current theory in FFADO-land is that there's an underlying issue with the streaming infrastructure. Pieter is looking into this as time allows. > What the thing about the 1kHz tone? Is it necessary for keeping up the > stream or is it for test purposes only until the stable libffado is out? > It is silenced out when playing/recording through analog 1&2? The 1 kHz test tone is added by the driver to facilitate my testing and debugging of the streaming layer. It is for testing purposes only and will not be present in any stable release. It is not silenced when playing a signal to analog 1&2 - it will replace any signal sent to analog 1 but analog 2 will be unaffected. The tone is also only a *playback* tone - if you record analog 1&2 you will record whatever signal is sent to those channels, not the test tone. I'm thinking of moving the test tone to another channel (like analog 8 for example) but there are a couple of things related to my debugging setup that I need to deal with before I can do that. Regards jonathan |
|
From: Jonathan W. <jw...@ph...> - 2007-10-02 23:26:53
|
Hi Nudge > I have used full preemption while compiling. Cool. > For audio use, I made a small start script that shuts down any > unnecessary service like mysql, network and so on, makes an xhost local > and then runs jackd as root. So usually I do nothing special when > starting jackd but giving -R and running it as root. In my eyes, this should be > enough regarding job priority. It depends on the machine - how fast is it? Sometimes it's necessary to elevate the priority of the IRQ thread servicing the ieee1394 card and then select a suitable jackd priority (via the "-P" option). This way the ieee1394 IRQ service gets priority over anything else in the system. Usually I'll run the IRQ at a priority of 12 above the jackd process which in my setup means 72. I use "-P 60" with jackd. A full discussion of this stuff can be found at http://tapas.affenbande.org/wordpress/?page_id=40 > Afterwards, when jackd is running, I start qjackctl to see the connections. > I always wondered why qjackctl never lights up the RT in the display. > Do I have to start qjackctl again with some rtlimits specified or whatever? I can't help you there - I don't currently use qjackctl. However, there could be an issue if you run qjackctl as a different user to the jackd process - I know that can cause issues. If you start jackd as root then I think qjackctl (and in fact all jack applications) also must run as root. Since this is a really bad idea from the security and convenience standpoint it's usually better to allow your user access to RT scheduling (via PAM), or something like set_rlimits (found on my homepage at http://www.physics.adelaide.edu.au/~jwoithe/). Personally I use set_rlimits since my system (slackware 12) doesn't use PAM (for which I am eternally grateful). At one point PAM required a patch to let it set rt access, but I don't know what the situation is now. > I think I should use the options Francois mentioned and > recompile libffado + jack to see whether it helps my situation. Starting with the same (or similar) configuration would allow us to get a better handle on what might be going on. Regards jonathan |
|
From: Francois.ernoult <fra...@pa...> - 2007-10-02 16:15:51
|
Hi Nudge, > did you experience that the numbers of glitches or unsyncs grow > with the load/channels or was it independent from them? (Thinking > about how stable the device might be at home and then in the studio). It's difficult to say. With a complete Ardour project full of plugs, I have to set buffers to 1024, but it's the same with Cubase/Windows. The only difference is that FFADO looses the sync and rarely recovers it where the ASIO driver just clicks. > What the thing about the 1kHz tone? Is it necessary for keeping > up the stream or is it for test purposes only until the stable libffado is > out? It's just a test signal to see if it works without using a jack client to generate one. It's here for a long time ago, when the motufw driver wasn't part of FreeBob/FFADO. > It is silenced out when playing/recording through analog 1&2? No, it has nothing to do with Jack. A sinus is copied to the buffer just before it's sent to the interface, so it hardly erases the normal out. In production, I turn it off. Just change the #define TESTTONE 1 in src/libstreaming/MotuStreamProcessor.cpp into #define TESTTONE 0. Only Analog playback 1 is affected, Analog record works normally. > Last question: Where can I find your patch? No where for now. The patch includes untested code that I have to remove before sending it to Jonathan (and you if you want). I'll probably do that tonight if I have time. Regards, Francois |
|
From: Mathias L. <mat...@go...> - 2007-10-02 15:44:25
|
SGkgRnJhbmNvaXMsCgpkaWQgeW91IGV4cGVyaWVuY2UgdGhhdCB0aGUgbnVtYmVycyBvZiBnbGl0 Y2hlcyBvciB1bnN5bmNzIGdyb3cKd2l0aCB0aGUgbG9hZC9jaGFubmVscyBvciB3YXMgaXQgaW5k ZXBlbmRlbnQgZnJvbSB0aGVtPyAoVGhpbmtpbmcKYWJvdXQgaG93IHN0YWJsZSB0aGUgZGV2aWNl IG1pZ2h0IGJlIGF0IGhvbWUgYW5kIHRoZW4gaW4gdGhlIHN0dWRpbykuCgpXaGF0IHRoZSB0aGlu ZyBhYm91dCB0aGUgMWtIeiB0b25lPyBJcyBpdCBuZWNlc3NhcnkgZm9yIGtlZXBpbmcKdXAgdGhl IHN0cmVhbSBvciBpcyBpdCBmb3IgdGVzdCBwdXJwb3NlcyBvbmx5IHVudGlsIHRoZSBzdGFibGUg bGliZmZhZG8gaXMgb3V0PwpJdCBpcyBzaWxlbmNlZCBvdXQgd2hlbiBwbGF5aW5nL3JlY29yZGlu ZyB0aHJvdWdoIGFuYWxvZyAxJjI/CgpMYXN0IHF1ZXN0aW9uOiBXaGVyZSBjYW4gSSBmaW5kIHlv dXIgcGF0Y2g/CjotKQoKUmVnYXJkcwpOdWRnZQoKCgpPbiAxMC8xLzA3LCBGcmFuY29pcy5lcm5v dWx0IDxmcmFuY29pcy5lcm5vdWx0QHBhc2QuZnI+IHdyb3RlOgo+IEhpIE1hdGhpYXMsCj4KPiBJ dCBzZWVtcyBteSBleHBlcmllbmNlIHdhcyBiZXR0ZXIgdGhhbiB5b3Vycy4KPiAtIERlYmlhbiBs ZW5ueS9zaWQKPiAtIGtlcm5lbCAyLjYuMjMtcmMyLXJ0Mgo+IC0gbGlicmF3MTM5NCAxLjMuMCBm cm9tIHN2biAoci4xNzIpCj4gLSBqYWNrIGZyb20gc3ZuIChyLjEwNTMpCj4gLSBsaWJmZmFkbyBm cm9tIHN2biAoci42MTUpCj4gLSBxamFja2N0bCAwLjMuMWEKPiAtIDgyOG1rSUkKPgo+IEZGQURP IGp1c3QgY29tcGlsZSBmaW5lIHdpdGggRU5BQkxFX0JFQk9CPUZhbHNlIEVOQUJMRV9GSVJFV09S S1M9RmFsc2UKPiBFTkFCTEVfTU9UVT1UcnVlLgo+IFRoZSBzYW1lIGZvciBKYWNrIHdpdGggLS1k aXNhYmxlLW9zcyAtLWVuYWJsZS1vcHRpbWl6ZS4KPgo+IFFqYWNrY3RsIGxhdW5jaCBqYWNrIHdp dGg6Cj4gL3Vzci9sb2NhbC9iaW4vamFja2QgLVIgLVA2OCAtZGZpcndpcmUgLXI0ODAwMCAtcDUx MiAtbjMKPgo+IC1QNjggaXMgc2V0IGFjY29yZGluZyBteSBpZWVlMTM5NCBpbnRlcnJ1cHQgcnRw cmlvLgo+IEkgaGF2ZSB2ZXJ5IGZldyBzeW5jIGxvb3NlIChodW0gYW5kIGhpc3MpLiBJIG1hZGUg YSB0d28gZGF5cyByZWNvcmQgc2Vzc2lvbgo+IHdpdGggYSB2ZXJ5IHNpbWlsYXIgY29uZmlndXJh dGlvbiAoYSBmZXcgbW9udGhzIGFnbywgc28gd2l0aCBlYXJsaWVyCj4gdmVyc2lvbnMpIGFuZCBp dCB3YXMgc3RhYmxlIGVub3VnaCB0byB3b3JrIGNvcnJlY3RseS4KPiBJIGFsc28gcmVjb3JkZWQg YSBjb25jZXJ0IGluIG11bHRpLXRyYWNrIHdpdGggdGhlIHNhbWUgY29uZmlndXJhdGlvbi4KPgo+ IFRoZSBjaGFubmVscyBuYW1lIChjYXB0dXJlX3ggYW5kIHBsYXliYWNrX3gpIGFyZSBhIG5ldyBm ZWF0dXJlIG9mIGphY2tkCj4gKHIuMTAzMikuIE9yaWdpbmFsIG5hbWVzIGFyZSBjb3BpZWQgYXMg YWxpYXNlcyBhbmQgcmVhbCBuYW1lcyBhcmUgcmVwbGFjZWQKPiBieSBzdGFuZGFyZHMgb25lcy4K PiBJIGRvbid0IHJlYWxseSB1bmRlcnN0YW5kIHdoeSBpdCdzIGJldHRlciBsaWtlIHRoYXQuIEkg YXNrZWQgZm9yIHByZWNpc2lvbnMKPiBvbiB0aGUgW2phY2tpdC1kZXZlbF0gbWFpbGluZyBsaXN0 IDMgd2Vla3MgYWdvIGJ1dCBJIGRpZG4ndCBoYXZlIGFueSBhbnN3ZXIuCj4gSSBjb21tZW50ZWQg b3V0IHRoZSBjb2RlIGRvaW5nIHRoaXMgaW4gamFja2Qgc28gdGhhdCBJIGhhdmUgdXNhYmxlIG5h bWVzLgo+Cj4gSSBjaGFuZ2VkIHRoZSBwb3J0cyBjcmVhdGlvbiBvcmRlciBpbiBsaWJmZmFkby9z cmMvbW90dS9tb3R1X2F2ZGV2aWNlLmNwcCBzbwo+IHRoYXQgdGhlIE1haW4tb3V0IGlzIHRoZSBm aXJzdCBvdXQgdXNlZCBieSB0aGUgYXV0by1jb25uZWN0IGZlYXR1cmUgb2YKPiBjbGllbnRzLiBU aGlzIGlzIHZlcnkgdXNlZnVsIGlmIHlvdSB3YW50IGxpc3RlbmluZyBtcDMgb3IgcmFkaW8gZWFz aWx5Lgo+Cj4gVGhlIGFuYWxvZyBvdXQgMSBsZWQgc2hvdWxkIGJlIE9OLCBiZWNhdXNlIGEgMWtI eiB0ZXN0IHRvbmUgaXMgc2VuZCB0byBpdCBieQo+IEZGQURPLiBJZiBpdCdzIGJsaW5raW5nLCB0 aGUgc2luayBpcyBwcm9iYWJseSBsb29zZWQuCj4KPiBGaW5hbGx5LCBJIGRvbid0IGhhdmUgYW55 IHByb2JsZW0gd2l0aCB0aGUgY29udHJvbCBwYW5lbC4KPgo+IFJlZ2FyZHMsCj4gRnJhbsOnb2lz Cj4KPiA+IC0tLS0tTWVzc2FnZSBkJ29yaWdpbmUtLS0tLQo+ID4gRGU6IG1vdHVmdy1kZXZlbC1i b3VuY2VzQGxpc3RzLnNvdXJjZWZvcmdlLm5ldCBbbWFpbHRvOm1vdHVmdy1kZXZlbC0KPiA+IGJv dW5jZXNAbGlzdHMuc291cmNlZm9yZ2UubmV0XSBEZSBsYSBwYXJ0IGRlIE1hdGhpYXMgTGllYmVy Cj4gPiBFbnZvecOpOiBsdW5kaSAxIG9jdG9icmUgMjAwNyAxMzo0OQo+ID4gw4A6IG1vdHVmdy1k ZXZlbEBsaXN0cy5zb3VyY2Vmb3JnZS5uZXQKPiA+IE9iamV0OiBbbW90dWZ3LWRldmVsXSBleHBl cmllbmNlIHdpdGggZmZhZG8gZHJpdmVyCj4gPgo+ID4gSGkgZm9sa3MsCj4gPgo+ID4gSSBoYXZl IGZpbmFsbHkgZm91bmQgdGhlIHRpbWUgdG8gY29tcGlsZSBhbmQgdGVzdCBsaWJmZmFkbyBvbiBt eSA4MjguCj4gPgo+ID4gWW91IGNhbiBmaW5kIGEgbW9yZSBvciBsZXNzIGRldGFpbGVkIHJlcG9y dCBvZiBteSBleHBlcmllbmNlcyBvbiB0aGUKPiA+IGZmYWRvIE1PVFUgODI4IG1rSUkgcGFnZSBv biBmZmFkby5vcmcgdW5kZXIKPiA+IGh0dHA6Ly93d3cuZmZhZG8ub3JnLz9xPW5vZGUvNjMsIHdo ZXJlIEkgaGF2ZSBhZGRlZCBpdCBhcyBhIGNvbW1lbnQuCj4gPiBVbmZvcnR1bmF0ZWx5LCBpdCBp cyBvbmx5IHZpc2libGUgd2hpbGUgbG9nZ2VkIGluLgo+ID4KPiA+IEtpbmQgcmVnYXJkcwo+ID4g TnVkZ2UKPiA+Cj4gPiAtLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tCj4gPiBUaGlzIFNGLm5ldCBlbWFpbCBpcyBz cG9uc29yZWQgYnk6IE1pY3Jvc29mdAo+ID4gRGVmeSBhbGwgY2hhbGxlbmdlcy4gTWljcm9zb2Z0 KFIpIFZpc3VhbCBTdHVkaW8gMjAwNS4KPiA+IGh0dHA6Ly9jbGsuYXRkbXQuY29tL01SVC9nby92 c2UwMTIwMDAwMDcwbXJ0L2RpcmVjdC8wMS8KPiA+IF9fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fCj4gPiBtb3R1ZnctZGV2ZWwgbWFpbGluZyBsaXN0Cj4gPiBt b3R1ZnctZGV2ZWxAbGlzdHMuc291cmNlZm9yZ2UubmV0Cj4gPiBodHRwczovL2xpc3RzLnNvdXJj ZWZvcmdlLm5ldC9saXN0cy9saXN0aW5mby9tb3R1ZnctZGV2ZWwKPgo+Cj4gLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLQo+IFRoaXMgU0YubmV0IGVtYWlsIGlzIHNwb25zb3JlZCBieTogTWljcm9zb2Z0Cj4gRGVm eSBhbGwgY2hhbGxlbmdlcy4gTWljcm9zb2Z0KFIpIFZpc3VhbCBTdHVkaW8gMjAwNS4KPiBodHRw Oi8vY2xrLmF0ZG10LmNvbS9NUlQvZ28vdnNlMDEyMDAwMDA3MG1ydC9kaXJlY3QvMDEvCj4gX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KPiBtb3R1ZnctZGV2 ZWwgbWFpbGluZyBsaXN0Cj4gbW90dWZ3LWRldmVsQGxpc3RzLnNvdXJjZWZvcmdlLm5ldAo+IGh0 dHBzOi8vbGlzdHMuc291cmNlZm9yZ2UubmV0L2xpc3RzL2xpc3RpbmZvL21vdHVmdy1kZXZlbAo+ Cg== |
|
From: Mathias L. <mat...@go...> - 2007-10-02 15:36:26
|
Hi Jonathan, I have used full preemption while compiling. For audio use, I made a small start script that shuts down any unnecessary service like mysql, network and so on, makes an xhost local and then runs jackd as root. So usually I do nothing special when starting jackd but giving -R and running it as root. In my eyes, this should be enough regarding job priority. Afterwards, when jackd is running, I start qjackctl to see the connections. I always wondered why qjackctl never lights up the RT in the display. Do I have to start qjackctl again with some rtlimits specified or whatever? I think I should use the options Francois mentioned and recompile libffado + jack to see whether it helps my situation. Kind regards Nudge On 10/2/07, Jonathan Woithe <jw...@ph...> wrote: > Hi Nudge > > > I have finally found the time to compile and test libffado on my 828. > > You can find a more or less detailed report of my experiences on the > > ffado MOTU 828 mkII page on ffado.org under > > http://www.ffado.org/?q=node/63, where I have added it as a comment. > > I see it - thanks for taking the time to write about your experiences. What > follows are brief comments on what you have there - I don't have time to go > into more detail right at the moment. > > > After doing some work on it, some strange noise on a very high > > pitch appeared. > > Yes, this is due to a loss of synchronisation. This in turn is probably due > to ffado not getting a packet off to the interface in enough time. While > there are some things you can do to minimise this happening, I still have > some more work to do to make the system less susceptable to the problem. > The latest thought is that it could be something in the base-level streaming > part of ffado, but this is yet to be proved. > > With your kernel, what preemption setting did you use? To gain the benefits > of the RT kernels you need to select either the "complete preemption" or > "low latency desktop" low latency desktop" preemption settings when > configuring your kernel. > > > Turning volume of Main Outs or Headphone up and down on the device, the > > display stays with the Message "Clock Source: Internal", so the faders > > cannot be seen anymore. > > Odd - this must be something peculiar to the MkII. On the Traveler all > cuemix controls can be manipulated and seen when the computer is using the > unit. > > > Though in the logs the channels had names like Analog7 and Analog8 - the > > names one is used to from the WDM drivers, all channels are named from > > capture_1 to capture_14 and playback_1 to playback_14. > > Yes, Francois mentioned this in his email over the weekend. I'm not > entirely sure I understand why the JACK guys have gone down this naming path > myself. For an interface such as the MOTUs it's most irritating because > the available channels change with sample rate. I haven't looked into this > in great detail yet but I too am after a workable solution. > > Thanks once again for your report and your interest in the FFADO MOTU > driver. > > Regards > jonathan > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > motufw-devel mailing list > mot...@li... > https://lists.sourceforge.net/lists/listinfo/motufw-devel > |
|
From: Francois.ernoult <fra...@pa...> - 2007-10-02 01:46:39
|
> [...] Feel free to send me just the ordering portion of > your patch earlier if you like. In fact I would prefer it if the patch > were > split into separate patches anyway as it makes application and maintenance > somewhat easier. [...] Sure, I'll do that this week. Regards, Francois |
|
From: Francois.ernoult <fra...@pa...> - 2007-10-02 01:35:50
|
Hi guys,
> > Turning volume of Main Outs or Headphone up and down on the device, the
> > display stays with the Message "Clock Source: Internal", so the faders
> > cannot be seen anymore.
>
> Odd - this must be something peculiar to the MkII. On the Traveler all
> cuemix controls can be manipulated and seen when the computer is using the
> unit.
My 828mkII doesn't have this problem.
> > Though in the logs the channels had names like Analog7 and Analog8 - the
> > names one is used to from the WDM drivers, all channels are named from
> > capture_1 to capture_14 and playback_1 to playback_14.
>
> Yes, Francois mentioned this in his email over the weekend. I'm not
> entirely sure I understand why the JACK guys have gone down this naming
> path
> myself. For an interface such as the MOTUs it's most irritating because
> the available channels change with sample rate. I haven't looked into
> this
> in great detail yet but I too am after a workable solution.
For now, I just replaced lines 974 and 982 of jackd/engine.c by 'if (0) {'
so that it leaves initial names alone.
Regards,
Francois
|
|
From: Jonathan W. <jw...@ph...> - 2007-10-02 01:28:13
|
Hi Francois > > > I changed the ports creation order in libffado/src/motu/motu_avdevice.cpp so > > > that the Main-out is the first out used by the auto-connect feature of > > > clients. This is very useful if you want listening mp3 or radio easily. > > > > That's an interesting point. The reason it's where it is in the main > > FFADO codebase is because the position used for "main out" on the MkII > > is used by AES/EBU on the Traveler (my interface) and having the defacto > > default channel being AES/EBU is a bit of a pain in the field. > > > > Having said that, I could look into moving things around a bit in the > > light of this issue to make things easier for MkII owners. However, > > this might not occur over the next few weeks because I'm doing audio for > > a theatre show and therefore don't expect to have too much free time > > over that time. > > Don't spent some time to do something I already done. I made a patch, the > reason I didn't sent it to you already is that it includes experimental > support for the 8Pre and the UltraLite. Be sure I'll do as soon as the code > will be validated on at last one of those interfaces. Sounds fair. However, feel free to send me just the ordering portion of your patch earlier if you like. In fact I would prefer it if the patch were split into separate patches anyway as it makes application and maintenance somewhat easier. For example: Part 1 = ordering for MkII Part 2 = 8Pre support Part 3 = UltraLite support If this is submitted as one large patch it makes it hard to see which changes applied to which feature. This also means that new features (like the ordering) can be applied now without having to wait for other things which aren't really related to it. Regards jonathan |
|
From: Francois.ernoult <fra...@pa...> - 2007-10-02 01:18:55
|
Hi Jonathan, > > I changed the ports creation order in > libffado/src/motu/motu_avdevice.cpp so > > that the Main-out is the first out used by the auto-connect feature of > > clients. This is very useful if you want listening mp3 or radio easily. > > That's an interesting point. The reason it's where it is in the main > FFADO > codebase is because the position used for "main out" on the MkII is used > by AES/EBU on the Traveler (my interface) and having the defacto default > channel being AES/EBU is a bit of a pain in the field. > > Having said that, I could look into moving things around a bit in the > light > of this issue to make things easier for MkII owners. However, this might > not occur over the next few weeks because I'm doing audio for a theatre > show > and therefore don't expect to have too much free time over that time. Don't spent some time to do something I already done. I made a patch, the reason I didn't sent it to you already is that it includes experimental support for the 8Pre and the UltraLite. Be sure I'll do as soon as the code will be validated on at last one of those interfaces. Regards, Francois |
|
From: Jonathan W. <jw...@ph...> - 2007-10-02 01:10:01
|
Hi Nudge > I have finally found the time to compile and test libffado on my 828. > You can find a more or less detailed report of my experiences on the > ffado MOTU 828 mkII page on ffado.org under > http://www.ffado.org/?q=node/63, where I have added it as a comment. I see it - thanks for taking the time to write about your experiences. What follows are brief comments on what you have there - I don't have time to go into more detail right at the moment. > After doing some work on it, some strange noise on a very high > pitch appeared. Yes, this is due to a loss of synchronisation. This in turn is probably due to ffado not getting a packet off to the interface in enough time. While there are some things you can do to minimise this happening, I still have some more work to do to make the system less susceptable to the problem. The latest thought is that it could be something in the base-level streaming part of ffado, but this is yet to be proved. With your kernel, what preemption setting did you use? To gain the benefits of the RT kernels you need to select either the "complete preemption" or "low latency desktop" low latency desktop" preemption settings when configuring your kernel. > Turning volume of Main Outs or Headphone up and down on the device, the > display stays with the Message "Clock Source: Internal", so the faders > cannot be seen anymore. Odd - this must be something peculiar to the MkII. On the Traveler all cuemix controls can be manipulated and seen when the computer is using the unit. > Though in the logs the channels had names like Analog7 and Analog8 - the > names one is used to from the WDM drivers, all channels are named from > capture_1 to capture_14 and playback_1 to playback_14. Yes, Francois mentioned this in his email over the weekend. I'm not entirely sure I understand why the JACK guys have gone down this naming path myself. For an interface such as the MOTUs it's most irritating because the available channels change with sample rate. I haven't looked into this in great detail yet but I too am after a workable solution. Thanks once again for your report and your interest in the FFADO MOTU driver. Regards jonathan |
|
From: Jonathan W. <jw...@ph...> - 2007-10-02 00:40:16
|
Hi guys Francois wrote: > > I changed the ports creation order in libffado/src/motu/motu_avdevice.cpp so > that the Main-out is the first out used by the auto-connect feature of > clients. This is very useful if you want listening mp3 or radio easily. That's an interesting point. The reason it's where it is in the main FFADO codebase is because the position used for "main out" on the MkII is used by AES/EBU on the Traveler (my interface) and having the defacto default channel being AES/EBU is a bit of a pain in the field. Having said that, I could look into moving things around a bit in the light of this issue to make things easier for MkII owners. However, this might not occur over the next few weeks because I'm doing audio for a theatre show and therefore don't expect to have too much free time over that time. > The analog out 1 led should be ON, because a 1kHz test tone is send to it > by FFADO. If it's blinking, the sink is probably loosed. True. If it's blinking it means the test tone output is intermittant. Since a continuous tone is generated the blinking does indicate that something is amiss. Having said that, a loss of sync often causes the LEDs of other channels to flash intermittantly as data ends up going to the wrong channels. Regards jonathan |
|
From: Francois.ernoult <fra...@pa...> - 2007-10-01 14:23:17
|
Hi Mathias, It seems my experience was better than yours. - Debian lenny/sid - kernel 2.6.23-rc2-rt2 - libraw1394 1.3.0 from svn (r.172) - jack from svn (r.1053) - libffado from svn (r.615) - qjackctl 0.3.1a - 828mkII FFADO just compile fine with ENABLE_BEBOB=3DFalse = ENABLE_FIREWORKS=3DFalse ENABLE_MOTU=3DTrue. The same for Jack with --disable-oss --enable-optimize. Qjackctl launch jack with: /usr/local/bin/jackd -R -P68 -dfirwire -r48000 -p512 -n3 -P68 is set according my ieee1394 interrupt rtprio. I have very few sync loose (hum and hiss). I made a two days record = session with a very similar configuration (a few months ago, so with earlier versions) and it was stable enough to work correctly. I also recorded a concert in multi-track with the same configuration. The channels name (capture_x and playback_x) are a new feature of jackd (r.1032). Original names are copied as aliases and real names are = replaced by standards ones. I don't really understand why it's better like that. I asked for = precisions on the [jackit-devel] mailing list 3 weeks ago but I didn't have any = answer. I commented out the code doing this in jackd so that I have usable = names. I changed the ports creation order in = libffado/src/motu/motu_avdevice.cpp so that the Main-out is the first out used by the auto-connect feature of clients. This is very useful if you want listening mp3 or radio easily. The analog out 1 led should be ON, because a 1kHz test tone is send to = it by FFADO. If it's blinking, the sink is probably loosed. Finally, I don't have any problem with the control panel. Regards, Fran=E7ois =20 > -----Message d'origine----- > De=A0: mot...@li... = [mailto:motufw-devel- > bo...@li...] De la part de Mathias Lieber > Envoy=E9=A0: lundi 1 octobre 2007 13:49 > =C0=A0: mot...@li... > Objet=A0: [motufw-devel] experience with ffado driver >=20 > Hi folks, >=20 > I have finally found the time to compile and test libffado on my 828. >=20 > You can find a more or less detailed report of my experiences on the > ffado MOTU 828 mkII page on ffado.org under > http://www.ffado.org/?q=3Dnode/63, where I have added it as a comment. > Unfortunately, it is only visible while logged in. >=20 > Kind regards > Nudge >=20 > = -------------------------------------------------------------------------= > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > motufw-devel mailing list > mot...@li... > https://lists.sourceforge.net/lists/listinfo/motufw-devel |
|
From: Mathias L. <mat...@go...> - 2007-10-01 12:48:45
|
Hi folks, I have finally found the time to compile and test libffado on my 828. You can find a more or less detailed report of my experiences on the ffado MOTU 828 mkII page on ffado.org under http://www.ffado.org/?q=node/63, where I have added it as a comment. Unfortunately, it is only visible while logged in. Kind regards Nudge |
|
From: Joseph H. <jos...@ho...> - 2007-09-28 21:31:02
|
Ben, don't feel bad, you're not the only one - My 828 MKII backlight blew = out also. Sucks when you want to use your MOTU without a computer for no-l= atency monitoring. I'm a bit skeptical about trying to replace the LED sin= ce I might end up rendering it useless... But other than that it's been working fine. I don't have the hum problem t= hough. Looking foward to FFADO support so I can ditch Windows entirely. > From: mot...@li... > Subject: motufw-devel Digest, Vol 1, Issue 2 > To: mot...@li... > Date: Tue, 25 Sep 2007 12:35:40 -0700 >=20 > Send motufw-devel mailing list submissions to > mot...@li... >=20 > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.sourceforge.net/lists/listinfo/motufw-devel > or, via email, send a message with subject or body 'help' to > mot...@li... >=20 > You can reach the person managing the list at > mot...@li... >=20 > When replying, please edit your Subject line so it is more specific > than "Re: Contents of motufw-devel digest..." >=20 >=20 > Today's Topics: >=20 > 1. owner, maybe i can help (Ben Powers) > 2. Re: owner, maybe i can help (Jonathan Woithe) > 3. Re: owner, maybe i can help (Ben Powers) > 4. Re: owner, maybe i can help (Jonathan Woithe) > 5. Re: owner, maybe i can help (Francois.ernoult) > 6. Re: owner, maybe i can help (Jonathan Woithe) > 7. bennyp wants to share sites with you... (be...@ou...) >=20 >=20 > ---------------------------------------------------------------------- >=20 > Message: 1 > Date: Wed, 23 May 2007 10:37:28 -0400 > From: "Ben Powers" <ben...@gm...> > Subject: [motufw-devel] owner, maybe i can help > To: mot...@li... > Message-ID: > <8c9...@ma...> > Content-Type: text/plain; charset=3D"iso-8859-1" >=20 > Hello, I have a semi-broken mk2 (backlight blew out, and there's a bit of= a > hum on it) but If i can help, i'd be happy to. I have a macbook. >=20 > --=20 > READ CAREFULLY. By [accepting this material|accepting this payment|accept= ing > this business-card|viewing this t-shirt|reading this sticker] you agree, = on > behalf of your employer, to release me from all obligations and waivers > arising from any and all NON-NEGOTIATED agreements, licenses, > terms-of-service, shrinkwrap, clickwrap, browsewrap, confidentiality, > non-disclosure, non-compete and acceptable use policies ("BOGUS AGREEMENT= S") > that I have entered into with your employer, its partners, licensors, age= nts > and assigns, in perpetuity, without prejudice to my ongoing rights and > privileges. You further represent that you have the authority to release = me > from any BOGUS AGREEMENTS on behalf of your employer. > -------------- next part -------------- > An HTML attachment was scrubbed... >=20 > ------------------------------ >=20 > Message: 2 > Date: Mon, 28 May 2007 11:21:04 +0930 (CST) > From: Jonathan Woithe <jw...@ph...> > Subject: Re: [motufw-devel] owner, maybe i can help > To: mot...@li... > Cc: Jonathan Woithe <jw...@ph...> > Message-ID: > <200...@tu...> > Content-Type: text/plain; charset=3Dus-ascii >=20 > Hi Ben >=20 > > Hello, I have a semi-broken mk2 (backlight blew out, and there's a bit = of > > a hum on it) but If i can help, i'd be happy to. I have a macbook. >=20 > Hmm, no backlight; that would make things a bit interesting. How did the > backlight blow out, or did it just happen? >=20 > There are a couple of things. Firstly, the motufw project is, as you can > tell, fairly quiet. While it was very useful as a "proof of concept" > project, it turned out that for the "real" driver it was better to integr= ate > our work into the FreeBob project (which has since been renamed to FFADO)= . > Therefore all the real work on supporting these MOTU interfaces under Lin= ux > is now being done within FFADO. >=20 > At this stage I am fairly sure that we don't have too many FFADO testers > with macbooks, so I expect that having someone willing to test FFADO > on this architecture would be rather handy. >=20 > By way of introduction you might like to hop across to >=20 > http://www.ffado.org/ >=20 > and take a look around. If you are interested in helping with testing an= d > can get Firewire working on your macbook under Linux I think it would be > most helpful. This is especially so over the next couple of months since > I'll have some time to return to FFADO programming again (I've been busy > with DVD productions for the past 4 months). >=20 > Best regards > jonathan >=20 >=20 >=20 > ------------------------------ >=20 > Message: 3 > Date: Mon, 28 May 2007 00:11:49 -0400 > From: "Ben Powers" <ben...@gm...> > Subject: Re: [motufw-devel] owner, maybe i can help > To: mot...@li... > Message-ID: > <8c9...@ma...> > Content-Type: text/plain; charset=3D"iso-8859-1" >=20 > Thanks Jonathan, >=20 > The backlight just blew out in a puff of mysterious blue smoke one day wh= en > i powered on, for no explainable reason. Hurray! Other than that the unit > works acceptably (bit of a buzz tho). I have right now a windows XP dual > boot on my macbook. a quick question before I start, can I just wipe the > winxp partition, install some distro and then install lilo and refit and = be > off to the races? >=20 > -B >=20 > On 5/27/07, Jonathan Woithe <jw...@ph...> wrote: > > > > Hi Ben > > > > > Hello, I have a semi-broken mk2 (backlight blew out, and there's a bi= t > > of > > > a hum on it) but If i can help, i'd be happy to. I have a macbook. > > > > Hmm, no backlight; that would make things a bit interesting. How did t= he > > backlight blow out, or did it just happen? > > > > There are a couple of things. Firstly, the motufw project is, as you c= an > > tell, fairly quiet. While it was very useful as a "proof of concept" > > project, it turned out that for the "real" driver it was better to > > integrate > > our work into the FreeBob project (which has since been renamed to FFAD= O). > > Therefore all the real work on supporting these MOTU interfaces under > > Linux > > is now being done within FFADO. > > > > At this stage I am fairly sure that we don't have too many FFADO tester= s > > with macbooks, so I expect that having someone willing to test FFADO > > on this architecture would be rather handy. > > > > By way of introduction you might like to hop across to > > > > http://www.ffado.org/ > > > > and take a look around. If you are interested in helping with testing = and > > can get Firewire working on your macbook under Linux I think it would b= e > > most helpful. This is especially so over the next couple of months sin= ce > > I'll have some time to return to FFADO programming again (I've been bus= y > > with DVD productions for the past 4 months). > > > > Best regards > > jonathan > > > > -----------------------------------------------------------------------= -- > > This SF.net email is sponsored by DB2 Express > > Download DB2 Express C - the FREE version of DB2 express and take > > control of your XML. No limits. Just data. Click to get it now. > > http://sourceforge.net/powerbar/db2/ > > _______________________________________________ > > motufw-devel mailing list > > mot...@li... > > https://lists.sourceforge.net/lists/listinfo/motufw-devel > > >=20 >=20 >=20 > --=20 > READ CAREFULLY. By [accepting this material|accepting this payment|accept= ing > this business-card|viewing this t-shirt|reading this sticker] you agree, = on > behalf of your employer, to release me from all obligations and waivers > arising from any and all NON-NEGOTIATED agreements, licenses, > terms-of-service, shrinkwrap, clickwrap, browsewrap, confidentiality, > non-disclosure, non-compete and acceptable use policies ("BOGUS AGREEMENT= S") > that I have entered into with your employer, its partners, licensors, age= nts > and assigns, in perpetuity, without prejudice to my ongoing rights and > privileges. You further represent that you have the authority to release = me > from any BOGUS AGREEMENTS on behalf of your employer. > -------------- next part -------------- > An HTML attachment was scrubbed... >=20 > ------------------------------ >=20 > Message: 4 > Date: Mon, 28 May 2007 14:16:36 +0930 (CST) > From: Jonathan Woithe <jw...@ph...> > Subject: Re: [motufw-devel] owner, maybe i can help > To: mot...@li..., ben...@gm... > Cc: Jonathan Woithe <jw...@ph...> > Message-ID: > <200...@tu...> > Content-Type: text/plain; charset=3Dus-ascii >=20 > Hi Ben >=20 > > The backlight just blew out in a puff of mysterious blue smoke one day = when > > i powered on, for no explainable reason. Hurray! >=20 > Curious. I sincerely hope that doesn't happen to my unit. >=20 > > I have right now a windows XP dual boot on my macbook. a quick question > > before I start, can I just wipe the winxp partition, install some distr= o > > and then install lilo and refit and be off to the races? >=20 > That I'm unsure of simply because I haven't had any direct experience wit= h > macbooks. It sounds like it should work but like I said, I've never trie= d > it because I don't have a macbook. I assume the other operating system > you want to have on there is MacOSX? If so, my main question would be > whether lilo can boot MacOSX. >=20 > Regards > jonathan >=20 >=20 >=20 > ------------------------------ >=20 > Message: 5 > Date: Mon, 28 May 2007 11:16:21 +0200 > From: "Francois.ernoult" <fra...@pa...> > Subject: Re: [motufw-devel] owner, maybe i can help > To: <mot...@li...>, <ben...@gm...> > Message-ID: <000901c7a108$db897ee0$0364a8c0@PASD> > Content-Type: text/plain; charset=3D"iso-8859-1" >=20 > Hi Ben, >=20 > Many people had the same problem with the backlight, mainly with the firs= t > version's bleu screen but it can happen with the green one. I know it's > possible to open the screen and replace the LED (some people did report i= t > on a French forum). > The hum from the power supply inside the 828mkII is also a known problem. > It's present on mostly all ones but it's very low. >=20 > Jack runs pretty well on OSX, it could be interesting to know if it's > possible to build it with FFADO support. libraw1394 seems to work on all > POSIX platforms. >=20 > Regards, > Fran?ois > =20 > > -----Message d'origine----- > > De?: mot...@li... [mailto:motufw-devel- > > bo...@li...] De la part de Jonathan Woithe > > Envoy??: lundi 28 mai 2007 05:47 > > ??: mot...@li...; ben...@gm... > > Cc?: Jonathan Woithe > > Objet?: Re: [motufw-devel] owner, maybe i can help > >=20 > > Hi Ben > >=20 > > > The backlight just blew out in a puff of mysterious blue smoke one da= y > > when > > > i powered on, for no explainable reason. Hurray! > >=20 > > Curious. I sincerely hope that doesn't happen to my unit. > >=20 > > > I have right now a windows XP dual boot on my macbook. a quick questi= on > > > before I start, can I just wipe the winxp partition, install some dis= tro > > > and then install lilo and refit and be off to the races? > >=20 > > That I'm unsure of simply because I haven't had any direct experience w= ith > > macbooks. It sounds like it should work but like I said, I've never tr= ied > > it because I don't have a macbook. I assume the other operating system > > you want to have on there is MacOSX? If so, my main question would be > > whether lilo can boot MacOSX. > >=20 > > Regards > > jonathan > >=20 > > -----------------------------------------------------------------------= -- > > This SF.net email is sponsored by DB2 Express > > Download DB2 Express C - the FREE version of DB2 express and take > > control of your XML. No limits. Just data. Click to get it now. > > http://sourceforge.net/powerbar/db2/ > > _______________________________________________ > > motufw-devel mailing list > > mot...@li... > > https://lists.sourceforge.net/lists/listinfo/motufw-devel >=20 >=20 >=20 >=20 > ------------------------------ >=20 > Message: 6 > Date: Tue, 29 May 2007 08:51:11 +0930 (CST) > From: Jonathan Woithe <jw...@ph...> > Subject: Re: [motufw-devel] owner, maybe i can help > To: mot...@li... > Cc: fra...@pa..., Jonathan Woithe > <jw...@ph...> > Message-ID: > <200...@tu...> > Content-Type: text/plain; charset=3Dus-ascii >=20 > Hi guys >=20 > > Many people had the same problem with the backlight, mainly with the fi= rst > > version's bleu screen but it can happen with the green one. I know it's > > possible to open the screen and replace the LED (some people did report= it > > on a French forum). >=20 > Thanks for the info - I'll keep it in mind. At least it *is* replaceable > and isn't some obscure lighting device. >=20 > > The hum from the power supply inside the 828mkII is also a known proble= m. > > It's present on mostly all ones but it's very low. >=20 > That may be part of the reason they opted for external power with the > Traveler (that and the desire to keep size and weight down). >=20 > > Jack runs pretty well on OSX, it could be interesting to know if it's > > possible to build it with FFADO support. libraw1394 seems to work on al= l > > POSIX platforms. >=20 > As it stands FFADO (through libraw1394) requires the Linux kernel's firew= ire > stack, or more specifically the raw1394 API provided by the Linux kernel.= As > a result it's almost certainly not going to run under MacOSX unless someo= ne > decides to add MacOSX support to libraw1394. Of course running Linux on = the > machine is a different matter; in this case it should be possible to get > things working (assuming they're not already). >=20 > I know that there have been problems in the past on the PPC but they rela= te > mainly to endianness issues and AFAIK they are being dealt with. Being a= n > intel-based machine though, the Macbook won't suffer from those problems. >=20 > Regards > jonathan >=20 >=20 >=20 > ------------------------------ >=20 > Message: 7 > Date: Tue, 25 Sep 2007 12:35:27 -0700 > From: be...@ou... > Subject: [motufw-devel] bennyp wants to share sites with you... > To: mot...@li... > Message-ID: <093...@sm...> > Content-Type: text/plain; charset=3D"iso-8859-1" >=20 >=20 > bennyp (be...@ou...) has invited you > to StumbleUpon! >=20 > You can see my other favorites here: > http://bennyp.stumbleupon.com >=20 > Thanks, > bennyp >=20 > --- > StumbleUpon lets you discover great sites with a > single click. Give it a try at: > http://www.stumbleupon.com/join.php?friend=3D520051&emailcode=3Dvign9fw5a= 17g7g36 > =09 >=20 > -------------- next part -------------- > An HTML attachment was scrubbed... >=20 > ------------------------------ >=20 > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2005. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ >=20 > ------------------------------ >=20 > _______________________________________________ > motufw-devel mailing list > mot...@li... > https://lists.sourceforge.net/lists/listinfo/motufw-devel >=20 >=20 > End of motufw-devel Digest, Vol 1, Issue 2 > ****************************************** _________________________________________________________________ Connect to the next generation of MSN Messenger=A0 http://imagine-msn.com/messenger/launch80/default.aspx?locale=3Den-us&sourc= e=3Dwlmailtagline= |
|
From: <be...@ou...> - 2007-09-25 19:35:36
|
bennyp (be...@ou...) has invited you to StumbleUpon! You can see my other favorites here: http://bennyp.stumbleupon.com Thanks, bennyp --- StumbleUpon lets you discover great sites with a single click. Give it a try at: http://www.stumbleupon.com/join.php?friend=520051&emailcode=vign9fw5a17g7g36 |
|
From: Jonathan W. <jw...@ph...> - 2007-05-28 23:21:24
|
Hi guys > Many people had the same problem with the backlight, mainly with the first > version's bleu screen but it can happen with the green one. I know it's > possible to open the screen and replace the LED (some people did report it > on a French forum). Thanks for the info - I'll keep it in mind. At least it *is* replaceable and isn't some obscure lighting device. > The hum from the power supply inside the 828mkII is also a known problem. > It's present on mostly all ones but it's very low. That may be part of the reason they opted for external power with the Traveler (that and the desire to keep size and weight down). > Jack runs pretty well on OSX, it could be interesting to know if it's > possible to build it with FFADO support. libraw1394 seems to work on all > POSIX platforms. As it stands FFADO (through libraw1394) requires the Linux kernel's firewire stack, or more specifically the raw1394 API provided by the Linux kernel. As a result it's almost certainly not going to run under MacOSX unless someone decides to add MacOSX support to libraw1394. Of course running Linux on the machine is a different matter; in this case it should be possible to get things working (assuming they're not already). I know that there have been problems in the past on the PPC but they relate mainly to endianness issues and AFAIK they are being dealt with. Being an intel-based machine though, the Macbook won't suffer from those problems. Regards jonathan |
|
From: Francois.ernoult <fra...@pa...> - 2007-05-28 09:16:11
|
Hi Ben, Many people had the same problem with the backlight, mainly with the = first version's bleu screen but it can happen with the green one. I know it's possible to open the screen and replace the LED (some people did report = it on a French forum). The hum from the power supply inside the 828mkII is also a known = problem. It's present on mostly all ones but it's very low. Jack runs pretty well on OSX, it could be interesting to know if it's possible to build it with FFADO support. libraw1394 seems to work on all POSIX platforms. Regards, Fran=E7ois =20 > -----Message d'origine----- > De=A0: mot...@li... = [mailto:motufw-devel- > bo...@li...] De la part de Jonathan Woithe > Envoy=E9=A0: lundi 28 mai 2007 05:47 > =C0=A0: mot...@li...; ben...@gm... > Cc=A0: Jonathan Woithe > Objet=A0: Re: [motufw-devel] owner, maybe i can help >=20 > Hi Ben >=20 > > The backlight just blew out in a puff of mysterious blue smoke one = day > when > > i powered on, for no explainable reason. Hurray! >=20 > Curious. I sincerely hope that doesn't happen to my unit. >=20 > > I have right now a windows XP dual boot on my macbook. a quick = question > > before I start, can I just wipe the winxp partition, install some = distro > > and then install lilo and refit and be off to the races? >=20 > That I'm unsure of simply because I haven't had any direct experience = with > macbooks. It sounds like it should work but like I said, I've never = tried > it because I don't have a macbook. I assume the other operating = system > you want to have on there is MacOSX? If so, my main question would be > whether lilo can boot MacOSX. >=20 > Regards > jonathan >=20 > = -------------------------------------------------------------------------= > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > _______________________________________________ > motufw-devel mailing list > mot...@li... > https://lists.sourceforge.net/lists/listinfo/motufw-devel |
|
From: Jonathan W. <jw...@ph...> - 2007-05-28 04:46:54
|
Hi Ben > The backlight just blew out in a puff of mysterious blue smoke one day when > i powered on, for no explainable reason. Hurray! Curious. I sincerely hope that doesn't happen to my unit. > I have right now a windows XP dual boot on my macbook. a quick question > before I start, can I just wipe the winxp partition, install some distro > and then install lilo and refit and be off to the races? That I'm unsure of simply because I haven't had any direct experience with macbooks. It sounds like it should work but like I said, I've never tried it because I don't have a macbook. I assume the other operating system you want to have on there is MacOSX? If so, my main question would be whether lilo can boot MacOSX. Regards jonathan |
|
From: Ben P. <ben...@gm...> - 2007-05-28 04:11:53
|
Thanks Jonathan, The backlight just blew out in a puff of mysterious blue smoke one day when i powered on, for no explainable reason. Hurray! Other than that the unit works acceptably (bit of a buzz tho). I have right now a windows XP dual boot on my macbook. a quick question before I start, can I just wipe the winxp partition, install some distro and then install lilo and refit and be off to the races? -B On 5/27/07, Jonathan Woithe <jw...@ph...> wrote: > > Hi Ben > > > Hello, I have a semi-broken mk2 (backlight blew out, and there's a bit > of > > a hum on it) but If i can help, i'd be happy to. I have a macbook. > > Hmm, no backlight; that would make things a bit interesting. How did the > backlight blow out, or did it just happen? > > There are a couple of things. Firstly, the motufw project is, as you can > tell, fairly quiet. While it was very useful as a "proof of concept" > project, it turned out that for the "real" driver it was better to > integrate > our work into the FreeBob project (which has since been renamed to FFADO). > Therefore all the real work on supporting these MOTU interfaces under > Linux > is now being done within FFADO. > > At this stage I am fairly sure that we don't have too many FFADO testers > with macbooks, so I expect that having someone willing to test FFADO > on this architecture would be rather handy. > > By way of introduction you might like to hop across to > > http://www.ffado.org/ > > and take a look around. If you are interested in helping with testing and > can get Firewire working on your macbook under Linux I think it would be > most helpful. This is especially so over the next couple of months since > I'll have some time to return to FFADO programming again (I've been busy > with DVD productions for the past 4 months). > > Best regards > jonathan > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > _______________________________________________ > motufw-devel mailing list > mot...@li... > https://lists.sourceforge.net/lists/listinfo/motufw-devel > -- READ CAREFULLY. By [accepting this material|accepting this payment|accepting this business-card|viewing this t-shirt|reading this sticker] you agree, on behalf of your employer, to release me from all obligations and waivers arising from any and all NON-NEGOTIATED agreements, licenses, terms-of-service, shrinkwrap, clickwrap, browsewrap, confidentiality, non-disclosure, non-compete and acceptable use policies ("BOGUS AGREEMENTS") that I have entered into with your employer, its partners, licensors, agents and assigns, in perpetuity, without prejudice to my ongoing rights and privileges. You further represent that you have the authority to release me from any BOGUS AGREEMENTS on behalf of your employer. |
|
From: Jonathan W. <jw...@ph...> - 2007-05-28 01:51:21
|
Hi Ben > Hello, I have a semi-broken mk2 (backlight blew out, and there's a bit of > a hum on it) but If i can help, i'd be happy to. I have a macbook. Hmm, no backlight; that would make things a bit interesting. How did the backlight blow out, or did it just happen? There are a couple of things. Firstly, the motufw project is, as you can tell, fairly quiet. While it was very useful as a "proof of concept" project, it turned out that for the "real" driver it was better to integrate our work into the FreeBob project (which has since been renamed to FFADO). Therefore all the real work on supporting these MOTU interfaces under Linux is now being done within FFADO. At this stage I am fairly sure that we don't have too many FFADO testers with macbooks, so I expect that having someone willing to test FFADO on this architecture would be rather handy. By way of introduction you might like to hop across to http://www.ffado.org/ and take a look around. If you are interested in helping with testing and can get Firewire working on your macbook under Linux I think it would be most helpful. This is especially so over the next couple of months since I'll have some time to return to FFADO programming again (I've been busy with DVD productions for the past 4 months). Best regards jonathan |