#milkymist IRC log for Monday, 2013-05-13

GitHub144[milkymist-ng] sbourdeauducq pushed 2 new commits to master: http://git.io/iO5KAw07:38
GitHub144milkymist-ng/master 301008c Yann Sionneau: Some Makefile love07:38
GitHub144milkymist-ng/master 6bf6477 S├ębastien Bourdeauducq: Merge pull request #1 from fallen/makefile-001...07:38
GitHub151[milkymist] sbourdeauducq pushed 1 new commit to master: https://github.com/milkymist/milkymist/commit/711b9894d3bbec894f2834a70d8baeb35b77160608:14
GitHub151milkymist/master 711b989 Larry Doolittle: pfpu/test: keep up with pipeline length changes08:14
lekernelpots are working :)11:38
Fallenouthe slider?11:55
lekernel_slider and blackout, yes13:25
Fallenoulekernel_: (slider) yeah! a real working mixer now :)13:43
Fallenouwhat's the blackout?13:43
Fallenoua switch to turn off one input?13:43
lekernel_"intensity" of the output13:43
lekernel_black when at min, no change when at max13:43
GitHub158[milkymist-ng] sbourdeauducq pushed 2 new commits to master: http://git.io/zVHAyA13:47
GitHub158milkymist-ng/master 32c478a Sebastien Bourdeauducq: top: integrate ADC for pots13:47
GitHub158milkymist-ng/master 9f81afb Sebastien Bourdeauducq: software/videomixer: connect crossfade and blackout pots13:47
lekernelany idea about what precisely can cause DVI signal errors when the SDRAM is read?14:13
lekerneltried rewiring the DVI clock to a GCLK pin (with coax and to the SD card pin, so quite shielded from the SDRAM traces) - no change14:14
GitHub34[milkymist-ng] sbourdeauducq pushed 1 new commit to master: http://git.io/NSWy8g15:20
GitHub34milkymist-ng/master 581cf5b Sebastien Bourdeauducq: timer: atomic reads15:20
lekernelhmm, vivado in command line looks usable16:00
lekernelit's actually much faster than ISE16:03
wpwrakdidn't someone (lars ?) mention some horrible slowdown ?16:06
lekernelI did, but this was with the GUI16:08
lekerneland an older version16:08
wpwrakoh, a speed penalty for gui users. nice :)16:10
lekernelit does synthesis + p&r of the whole lm32 core under a minute16:11
wpwrakhow long did it take on your system before ?16:11
lekernelnot sure about the exact figure... but I remember waiting ages16:15
kristianpauldoes vivado still using the xilinx description language or they wrote a new one?17:02
lekernelXDL? I don't think so...17:09
lekernelbut not sure17:09
lekerneltry :)17:09
kristianpauloh, 104usd for a zynq soc (XC7Z010)17:11
larscyea pretty expensive :/17:13
larscwhich is funny that the paralella sells for 99$ and has a zynq on the board17:14
davidc__For those that don't know - digikey prices for Xilinx stuff bear no relation to actual real-world volume prices17:29
wpwrakdavidc__: is there a conversion function ?17:30
wpwrakor a collection of data points ?17:31
davidc__wpwrak: nah; just call up Avnet/$YOUR_DISTY, ask for pricing at $Xkunits, call em back a week later saying that you think you're going with Altera and ask if they can do any better17:32
davidc__"Its a really nice part and we like developing with ISE (*snort*), but the pricing just doesn't work for our design"17:33
wpwrakdirty :)17:34
davidc__wpwrak: I've see GigE switch silicon go from $35 /pc down to $7.25/pc17:35
davidc__Just by making two chipmakers duke it out with each other. [I'd rather have given them bricks and shut them in a closet, but the purchasing guy overruled me]17:36
davidc__wpwrak: keep in mind though that if you don't actually buy $X kunits, they're not going to give you good pricing the next time round17:37
davidc__so use actual real volume estimates17:37
wpwrakdavidc__: perhaps we need a "priceleaks" where people can anonymously post chips, quantity, and price. randomize and filter the data a little to make it harder to trace it back to a specific customer18:03
davidc__Meh, but good luck getting the distys/mfgs to give you those prices. Besides, if you really want to get depressed about chip prices; get a native speaker in China to get quotes from legit distys for the same parts18:07

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!