On 08/26/2016 06:17 PM, jonetsu(a)teksavvy.com wrote: Well, the 22MTK delivers 22 (23-24 are the stereo sum of the mix) inputs at a reliable and tolerable latency => -r48000 -p256 -n2 -Xseq -D -Chw:MTK -Phw:MTK I use my m-audio gear now mainly as external instrument outputs for my software instruments (on a different computer) that are fed into the 22MTK. The rest of the inputs are my guitars and basses. All are routed directly into ardour and recorded. To be honest, the thing has been extremely reliable and I have absolutely no complaints--except that the USB inputs are post gain, pre-eq/fader ... but I've gotten used to it.

On Fri, 26 Aug 2016 21:12:08 +0100 Will Godfrey <willgodfrey(a)musically.me.uk> wrote: Looks more and more like replacing the caps on the 1010LT (if that turns out to be the problem) ... and getting another one as backup. Our youngest son has one also. Maybe he's be interested in trading it for one of those modern USB interfaces.

On Fri, 26 Aug 2016 22:50:27 +0200 Ralf Mardorf <ralf.mardorf(a)alice-dsl.net> wrote: It's all about 432Hz, the universal frequency. That is THE truth. Universe aside, as I've written "if it turns out to be the problem", caps would not fail immediately. There would be a transition time I think. Or then maybe not. Got the idea manly from Tim here, although I thought about it. I still keep an eye on Bitwig. Every time it happens Bitwig runs. Might not be playing something at that time, but it runs. I should do a test and find something to do on Ardour/Mixbus only for the same length of time, possibly without using any pulseaudio application at all like firefox/Opera or vlc. The latter would not be so easy as I'm following lessons these days. Fazit, there is still no conclusion.

On Fri, 26 Aug 2016 17:51:25 -0400, jonetsu(a)teksavvy.com wrote: This could be possible. If they are dried out, then capacitance got lost. If the the oxide coating is broken, they short. Assuming that there is dark scum on the PCB around the capacitor, resp. on the rubber thingy on the bottom of the capacitor, then it's dried out. Assuming the top of the capacitor is bloated, then the capacitor is broken, too. If you can't see something, they could be broken, too. However, if you can't see that they are broken, then take their function into account. What is a capacitor used for, that could cause delayed trouble, if it's broken, instead of causing the trouble all the times? IIUC you hear the crackling only, if there is an audio signal, there's no crackling, without an audio signal. If the crackling happens, does show top something unusual? Maybe upowerd or something else turns nasty? What happens if you don't use two sound servers at the same time? Have you tested Firefox using plain ALSA, while neither pulseaudio, nor jackd are running? Regards, Ralf

On Fri, 26 Aug 2016 20:08:54 -0700 (PDT), Len Ovens wrote: Again guessing. This card is not an amplifier that requires a big reliable power supply, neither it's comparable to a CRT, that has got a high power consumption. I do not own this card, so I ask questions to perhaps provide more than just guessing. I didn't find a circuit layout, but already tried to read on a photo what is printed on the large TO-220 package component. I didn't find a photo where I could read what's printed on it, but while searching such a photo, I found this https://offog.org/notes/delta-1010lt-repair/ So the two capacitors close to the large TO-220 package component are reservoir capacitors. Note, my PCI Envy24 cards have a voltage regulator were the delta seems to have one, too. At the further end of the large TO-220 package component, there are two small capacitors and something that looks like a miniature edition of a TO-220 package. On my cards it's a 1117 voltage regulator. However, assuming that the provided power by the computer's power supply should be too fluctuating, to feat the stabilisation of the delta, I would suspect more dramatic effects. If the stabilisation should be the reason for the issue, then I would expect broken capacitors on the audio card. Regards, Ralf

Am Samstag, den 27.08.2016, 03:43 -0700 schrieb Tracey: There are still new mobos with chipsets that support PCI classic (without bridging it over PCIe), but you have to take AMD-based ones. I know, Intel offers more power, but you can at least use your old hardware. I hope Zen will support classic PCI, too. I want to use my RME PCI card (HDSP 9652) as long as I can, because I haven't used anything safe like this, ever. Keep my fingers crossed that I can upgrade my system for some more years... Greets! Mitsch

On 8/27/16, Tracey <traceyanne(a)sonic.net> wrote: It may not be needed after all :) AFAIK all PCI slots are connected now to the PCI-E bus with a bridge like the old PCI-ISA bridge and there's no way around it. Based on former reading and experimenting with a desktop graphics card attached to a laptop, I just searched for "external pci slot" and got exactly what I wanted: The PCI-E to PCI bridge and a sort of riser card. That means with the proper adapters or surgery, you could use a 1010 with a notebook. Just 1 lane of a PCI-E 1.0 bus is already faster than the entire shared bandwidth of a classic PCI bus so it's certainly not a problem. YMMV and would probably depend on the quality of that bridge IC. In any case this seems to require yet more delightfully affordable Chinese stuff and yet more testing. -- Rob

On Sat, 27 Aug 2016 12:56:18 -0700, Fernando Lopez-Lezcano wrote: Hi Fernando, IIUC there are different 1010 models out there and some are using external power supplies and those power supplies seem to be be very cheap. Regarding this http://offog.org/notes/delta-1010lt-repair/fixed_size1600.jpg picture and some reading this card, the "LT" model, has got no external power supply. It gets the power from the PCI slot. In front of the picture, on the left side and on the right side, there seem to be voltage regulators and a few caps that belong to the circuits around the voltage regulators. My TerryTec EWX 24/96 don't have something on the left side, IOW not such a TO-220 package thingy and no big caps, but on the right side the TerraTecs look similar to the Delta. Two small electrolytic capacitors and behind them is a 1117 voltage regulator on the TerraTecs Envy24 cards. The power comes from the PC's power supply. I don't know how easy it is for the OP to replace the caps. It could be possible to replace caps using a 25 W ERSA soldering iron without risking to damage vertical interconnect accesses, but sometimes removing the old caps even doesn't work with a Weller desolder station. Once a friend needed to desolder caps for me at work, using hot air. I couldn't remove them using my soldering iron and we couldn't do it using his desolder station. If the OP could simply replace the caps, then it doesn't harm to replace them, even if they are still ok, but as soon as it should be hard to remove the caps and there should be no visible leaking or bloat, it's better not to risk it, before it isn't ruled out, that it is not a software issue. I don't claim that the caps are ok, they could be borked, but a zombified sound server related process is good enough for reasonable doubts. Sure, we could suffer from cooties and fleas at the same time. 2 Cents, Ralf

On Sun, 28 Aug 2016 01:43:07 +0200 Ralf Mardorf <ralf.mardorf(a)alice-dsl.net> wrote: You probably don't want to know! I've seen some pretty nasty ways of doing this - 40106s with their outputs wired in parallel 3s as a push-pull oscillator was of the better ones :( -- Will J Godfrey http://www.musically.me.uk Say you have a poem and I have a tune. Exchange them and we can both have a poem, a tune, and a song.

On 08/27/2016 04:20 PM, Ralf Mardorf wrote: I commented on the plain 1010, that has an external "brick" power supply, a pci card and a rack mounted D/A A/D unit - that is where the bad capacitors usual are. I have never used a 1010lt so I know nothing about it... sorry... -- Fernando ...

Ralf Mardorf &lt;ralf.mardorf(a)alice-dsl.net&gt; writes: On my 1010LT, one of the filter caps was visibly leaking: http://offog.org/notes/delta-1010lt-repair/ That's a pretty rational reason for replacing them! -- Adam Sampson &lt;ats(a)offog.org&gt; <http://offog.org/>

Hi, yes, that's why I explained how to notice broken caps, by either leaking or if they are bloated. I only pointed out that if you can't see dirt from leaking or a bloated top, then it's important to understand what a cap is good for, before replacing it, without an idea if it could be related to the described issue. Btw. when searching for a circuit layout or at least a clear photo to read what's printed on some components, I found the same link and posted it. If you follow the thread, there never was mentioned that the caps are looking broken and even the described symptoms leave behind an impression, that it likely could be a software issue. Running two sound servers, while one sound server kept a zombie and runs by a second instance, not clearly lead to borked caps. I also didn't notice any comment what is shown by top, when the crackling appears. So my recommendation still remains, as long as the caps are looking good and the OP didn't run top and fixed the pulseaudio zombie issue, replacing the capacitors just by guessing isn't a good idea. Regards, Ralf

On Sat, 27 Aug 2016 12:02:48 -0400 &quot;jonetsu(a)teksavvy.com&quot; &lt;jonetsu(a)teksavvy.com&gt; wrote: And naturally Wikipedia has something about it, The Capacitor Plague: https://en.wikipedia.org/wiki/Capacitor_plague

On Sat, 27 Aug 2016 19:53:54 +0200, Ralf Mardorf wrote: PS: Not to mention that the more important bulk capacitors aren't electrolytic ;).

On Sat, 27 Aug 2016 11:10:12 -0400 termtech &lt;termtech(a)rogers.com&gt; wrote: Before doing anything, and especially opening the box, I will try to be certain as much as possible of the nature of the problem, in this case by excluding software totally, which is still not yet done. I still have an eye on Bitwig, by coincidence, and that pulseaudio zombie does not look good. At least for MIDI devices Bitwig is doing things different than Ardour which is, after all by Paul Davis who also made jackd. I would presume that the reference in using jackd comes from Ardour. Also, Bitwig 1.11 had major crash problems in playing audio files from the sampler library and was replaced by 1.12 in a very short time. One more thing, I have demonstrated elsewhere on KVR that moving a plug-in GUI from the same desktop where Bitwig runs to another desktop will cause Bitwig to not play fully an audio clip in the Clip Launcher. Why would there be an interaction between the GUI/Desktop and the capacity to fully play an audio clip ? It might be that Some Things are not that clear in Bitwig, and that could be understandable if they wanted initially to quickly get to the market. So, software is not yet entirely out of the picture. One thing for sure though, it might be a Good Thing to have another 1010LT card. Cheers.