01:10 <_whitenotifier-3> [m-labs/nmigen-boards] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/fjVY4

01:10 <_whitenotifier-3> [m-labs/nmigen-boards] Fatsie 8a40fb6 - Convert DOS line endings.

01:10 <_whitenotifier-3> [nmigen-boards] whitequark closed pull request #5: Convert DOS line endings. - https://git.io/fjVIB

02:33 <_whitenotifier-3> [nmigen-boards] whitequark reviewed pull request #6 commit - https://git.io/fjVYj

02:34 <_whitenotifier-3> [nmigen-boards] whitequark reviewed pull request #6 commit - https://git.io/fjVOe

02:35 <_whitenotifier-3> [nmigen-boards] whitequark reviewed pull request #6 commit - https://git.io/fjVOv

06:09 <_whitenotifier-3> [m-labs/nmigen] whitequark pushed 1 commit to master [+0/-0/±1] https://git.io/fjV3p

06:09 <_whitenotifier-3> [m-labs/nmigen] whitequark 23ed888 - vendor.lattice_ice40: use different --package for 4k devices.

06:10 <_whitenotifier-3> [nmigen-boards] whitequark reviewed pull request #6 commit - https://git.io/fjV3h

06:21 <_whitenotifier-3> [nmigen] Success. The Travis CI build passed - https://travis-ci.org/m-labs/nmigen/builds/547556983?utm_source=github_status&utm_medium=notification

06:21 <_whitenotifier-3> [nmigen] Success. 80.71% remains the same compared to b3c5ff7 - https://codecov.io/gh/m-labs/nmigen/commit/23ed888857ea7989a560a27bf0e6996e76e3fb8e

06:21 <_whitenotifier-3> [nmigen] Success. Coverage not affected when comparing b3c5ff7...23ed888 - https://codecov.io/gh/m-labs/nmigen/commit/23ed888857ea7989a560a27bf0e6996e76e3fb8e

09:10 <sb0> rjo: what is LMU style?

09:11 <rjo> lmu munich. where those specific ecdl designs originate from.

09:26 <rjo> or there the design toptica used originated. meanwhile they have revised it a couple years ago.

09:54 <sb0> rjo: btw at least some SFPs appear to have some infrared filter on the photodiode. do you have any info about removing that?

10:10 <rjo> the CWDM ones and the BiDi will definitely do filtering. but probably deeply embedded in the package.

10:25 <sb0> it could be other problems, but I have a 850nm one that doesn't seem to like 633nm

10:25 <sb0> not bidi, not cwdm

10:39 <_whitenotifier-3> [nmigen-boards] Fatsie synchronize pull request #7: mystorm BlackIceII support - https://git.io/fjVIu

10:39 <_whitenotifier-3> [nmigen-boards] Fatsie synchronize pull request #6: mystorm BlackIce support - https://git.io/fjVIR

10:42 <_whitenotifier-3> [nmigen-boards] whitequark closed pull request #6: mystorm BlackIce support - https://git.io/fjVIR

10:42 <_whitenotifier-3> [m-labs/nmigen-boards] whitequark pushed 1 commit to master [+1/-0/±0] https://git.io/fjVc8

10:42 <_whitenotifier-3> [m-labs/nmigen-boards] Fatsie 5ec0d1f - mystorm BlackIce support

10:43 <_whitenotifier-3> [m-labs/nmigen-boards] whitequark pushed 1 commit to master [+1/-0/±0] https://git.io/fjVc4

10:43 <_whitenotifier-3> [m-labs/nmigen-boards] Fatsie 5227037 - mystorm BlackIceII support

10:43 <_whitenotifier-3> [nmigen-boards] whitequark closed pull request #7: mystorm BlackIceII support - https://git.io/fjVIu

10:43 <_whitenotifier-3> [nmigen-boards] Fatsie reviewed pull request #6 commit - https://git.io/fjVcR

10:43 <_whitenotifier-3> [nmigen-boards] Fatsie reviewed pull request #6 commit - https://git.io/fjVc0

11:14 <rjo> i could imagine that they filter those as well.

11:16 <rjo> just noticed that even telco wavelength people are worried about epoxy contamination in ROSA/TOSAs to the extent that they pride themselves in completely exluding it from the optical paths. it's not just blue/uv ion people.

13:55 <sb0> https://www.finisar.com/sites/default/files/downloads/hfd6x80-418_10_gbps_850nm_pin-preamp_sc_rosa_package_product_specification_revb00_022015.pdf

13:55 <sb0> this one says "input wavelength 830 to 870nm"

14:04 <sb0> those bare ROSA are cheaper than SFPs and easier to take apart...

14:10 <sb0> ah but they have EAR99 bullshit

16:32 <mtrbot-ml_> [mattermost] <sb10q> @astro are you using the zc706 right now?

17:02 <sb0> ah, of course you should never ask thorlabs about using USPS to ship parts instead of their expensive options, otherwise they bother you about "personal use"

17:02 <sb0> unfortunately the two parts I want haven't been satisfactorily copied by the chinese yet

17:23 <whitequark> what do they even have against personal use?

17:35 <mtrbot-ml> [mattermost] <astro> @sb10q nope, I'll disconnect from it...

17:40 <mtrbot-ml> [mattermost] <astro> now that I can work on eth RX/TX I'll do that on the Cora Z7-10 board here because I can plug the rj45 into my laptop

19:00 <TD-Linux> any tips on debugging nmigen simulation vs synthesis mismatch? write a testbench around yosys's generated verilog?

19:08 <whitequark> i think so

19:08 <whitequark> any memories or multiclock?

19:50 <TD-Linux> one SyncFIFO

19:50 <TD-Linux> (all one clock domain)

20:01 <whitequark> very odd

20:01 <whitequark> and very concerning

20:06 <TD-Linux> yeah I'm going to try to narrow it down enough to post an issue