03:36 <sb0> it's probably not worth the flight nor dealing with the money-hungry housing clusterfuck, but is anyone interested in exhibiting at sxsw? i have a free pass.

11:45 <sb0> whitequark, making the mass spectrometer not able to detect helium is not a good idea. I have yet to see a vacuum system that did not have at least one leak problem.

11:47 <sb0> essentially, if it cannot be used for leak testing, I have zero interest in it

12:20 <larsc> does anybody know what the threshold (and maybe other conditions) is for the elastic buffer to produce data?

12:25 <larsc> a always very clear answers... https://www.xilinx.com/support/answers/56066.html

12:36 <larsc> http://www.zju.edu.cn/jzus/oldversion/opentxt.php?doi=10.1631/jzus.C1300249 "High-speed, fixed-latency serial links with Xilinx FPGAs"

13:24 <larsc> there is a frightening correlation between RXPCOMMAALIGNEN begin toggled and a sample being dropped, both on assertion and deassertion, that's quite terrible

13:24 <larsc> s/begin/being

13:25 <larsc> 1/3 chance that this happens

13:25 <larsc> assertion is not that much of a problem

13:25 <larsc> but on de-assertion is not good

14:01 <larsc> sb0: you still do that probalistic comma-align right? haven't tried anything else?

14:04 <larsc> on the positive side doing comma align in fabric has less latency anyway

16:43 <GitHub> [artiq] cjbe commented on issue #566: I now also have this issue on several different machines with Artiq 2.2. Downgrading to icu 56.1 does not solve this, neither does re-installing pyqt from pip. Any suggestions? https://github.com/m-labs/artiq/issues/566#issuecomment-280385869

17:59 <GitHub> [artiq] jordens commented on issue #566: @r-srinivas @cjbe pretty sure you guys have a different issue. Please file a new one and provide information about what you have installed (context), what you did, and what your problem is (symptoms). https://github.com/m-labs/artiq/issues/566#issuecomment-280408258

22:12 <GitHub> [artiq] cjbe opened issue #670: Conda package for Artiq 2.2 broken on Windows: PyQt5 DLL load failed https://github.com/m-labs/artiq/issues/670

22:20 <GitHub> [artiq] jordens commented on issue #670: Do you have a previous conda environment around where this was working? If yes, please provide that `conda list`.... https://github.com/m-labs/artiq/issues/670#issuecomment-280480861

22:39 <rjo> sb0: you forgot to close the 2.2 milestone

22:41 <GitHub> [artiq] cjbe commented on issue #670: @jordens all of the packages in that environment were installed by running 'conda create -n artiq-main artiq-kc705-nist_clock'. ... https://github.com/m-labs/artiq/issues/670#issuecomment-280486554

22:48 <GitHub> [artiq] cjbe opened issue #671: Broadcasting large datasets fails https://github.com/m-labs/artiq/issues/671

22:53 <GitHub> [artiq] jordens commented on issue #670: Please try `conda install qt5=5.5.1=2 pyqt5=5.5.1=py35_0` in the non-working environment and paste the output.... https://github.com/m-labs/artiq/issues/670#issuecomment-280489767

23:01 <GitHub> [artiq] cjbe commented on issue #670: No joy - here is the output:... https://github.com/m-labs/artiq/issues/670#issuecomment-280491711

23:07 <GitHub> [artiq] jordens commented on issue #670: And now with `conda install icu=56.1`?... https://github.com/m-labs/artiq/issues/670#issuecomment-280493320

23:09 <GitHub> [artiq] jordens commented on issue #670: It also seems to ignore that icu dependency in `qt5=5.5.1-2`. https://github.com/m-labs/artiq/issues/670#issuecomment-280493651

23:09 <GitHub> [artiq] jordens commented on issue #670: It also seems to ignore that icu dependency in `qt5=5.5.1-2`. https://github.com/m-labs/artiq/issues/670#issuecomment-280493651

23:11 <GitHub> [artiq] cjbe commented on issue #670: Oops - I attached the wrong conda list last post. [This](https://github.com/m-labs/artiq/files/781780/list.txt) is the list after the last 'conda install'... https://github.com/m-labs/artiq/issues/670#issuecomment-280494112