00:03 <azonenberg> awygle: Musing about something i thought about a while back. It's now feasible

00:03 <awygle> an artful response, completely devoid of information :p

00:04 <azonenberg> both an ecp5 and an esp32 are less than 12mm on the shorter side, and the sum of them uses less than 1/1.5W

00:04 <azonenberg> Meaning it is likely technically feasible to build a SFP/SFP+ module that bridges 1000base-X to 802.11n

00:04 <azonenberg> with a SMA/RP-SMA on the front side of the "optic" instead of an LC connector

00:06 <azonenberg> i have no immediate plans to do it, but it's no longer "probably impossible"

01:10 <futarisIRCcloud> azonenberg: I suppose it'd be cheap... But slow over 802.11n.

01:11 <azonenberg> well the point would be to just get a free AP anywhere you have a switch

01:11 <azonenberg> no cabling or separate power needed

03:39 <TD-Linux> reading the quicklogic datasheet and it claims to be "Android 6.0 compliant"

03:48 <whitequark> TD-Linux: there was something about sensors in android 6 right?

03:48 <whitequark> so you have android 6 compliant accelerometers and such

03:49 <TD-Linux> ah I guess the idea is with their software stack you could implement an android 6 compliant imu

04:50 <whitequark> awygle: lattice relented

04:50 <whitequark> https://www.linkedin.com/posts/lattice-semiconductor_lattice-propel-license-activity-6674864964295114752-fe-5

04:52 <azonenberg> "thank you for pointing out a new bitstream usage restriction"

04:52 <azonenberg> um

04:53 <whitequark> reads like a (somewhat unsuccessful) attempt to save face to me

04:53 <TD-Linux> is it fixed? I don't want to log into their website

04:53 <azonenberg> it appears to be basically saying "hey, you read the eula"

04:55 <whitequark> TD-Linux: yes, they removed that part

04:55 <azonenberg> so they actually edited the license?

04:55 <azonenberg> Sounds like mission accomplished then, they backtracked

04:56 <whitequark> yep

04:56 <azonenberg> although i have to wonder, if they've decided not to fight bitstream RE, why don't they release the spec?

04:56 <azonenberg> they obviously have an internal spec for the tool writers

04:56 <azonenberg> it seems like the only two reasonable positions are to open the spec or to sue anyone who tries to reverse it

04:56 <whitequark> in my experience, the reason they do not release docs is "there is no business case to spend resources on that"

04:57 <azonenberg> That actually sounds somewhat like my experience with silego's programming spec

04:57 <azonenberg> they dont want to support users of it

04:57 <TD-Linux> there are a lot of companies in that boat. e.g. amd gpus

04:57 <whitequark> azonenberg: its not even about support

04:57 <TD-Linux> even intel cpus if you count agner fog's microarchitecture reverse engineering

04:57 <whitequark> i asked some questions privately about some of their primitives

04:57 <whitequark> they just dont want to spend *any* time answering those

04:58 <whitequark> my understanding is that lattice's position regarding OSS tools is currently "we're happy with someone else doing them and we are absolutely not going to spend any engineer time towards that direction"

04:59 <azonenberg> Yeah, but that seems like you could potentially get the same thing i got wrt the silego spec

04:59 <azonenberg> i.e. released as-is under "frienda", don't publish it because they don't want to support users of it

04:59 <azonenberg> but you can use the information contained in it as you see fit

05:00 <whitequark> i've tried and was unsuccessful

05:00 <azonenberg> my only guess is that there might be third party IP data in said docs

05:00 <azonenberg> and they dont want to spend the time to scrub it

05:00 <azonenberg> e.g. serdes ip might not be in house

05:01 <whitequark> yep

05:02 <azonenberg> in any case "not actively hostile" beats the alternatives

05:07 <awygle> oh missed the ping, so good they did remove that clause then? Excellent

05:09 <whitequark> yep

05:11 <awygle> genuinely the best news I've had all week

05:12 <azonenberg> awygle: you say that like it's a high bar

05:12 <azonenberg> this is 2020 :p

05:12 <azonenberg> "I didn't fall out of bed and break my neck this morning" qualifies as good news