Current Server Time: 10:28 (Central Europe)

#apertus IRC Channel Logs

2014/10/16

Timezone: UTC


01:04
pizthewiz
left the channel
01:18
intracube
the Metabones inverse-teleconverters look really interesting
01:19
intracube
heard about them on here for the first time today and also by chance on a youtube photography video
01:19
intracube
the universe works in mysterious ways :)
01:20
intracube
and they seem to support fast FF lenses too! woot
01:24
baldand
left the channel
01:26
baldand
joined the channel
02:40
baldand
left the channel
02:41
baldand
joined the channel
02:44
__anton__
he he, perhaps everybody except from me knew it already, but I've just discovered
02:44
__anton__
this part of the CMV12000 datasheet is a actually a very important one
02:44
__anton__
http://octoray.co.uk/axiom/misc/target-surface.png
02:44
__anton__
it does show us that this is in fact the right way to maintain flunge distance with cmv12000 and there here there are exactly 4 points of contact
02:44
__anton__
https://www.apertus.org/sites/default/files/styles/teaser/public/field/image/alpha_prototype.jpg?itok=28KvV2eS
02:45
__anton__
also Daniel suggested at some point that prosumer cameras here have got washers here to ensure that sensor is parallel to mount focal plane
02:48
__anton__
left the channel
02:57
wescotte
left the channel
03:10
AndroUser2
joined the channel
03:19
dmjnova
left the channel
04:19
Bertl
off to bed now ... have a good one everyone!
04:19
Bertl
changed nick to: Bertl_zZ
05:01
wescotte
joined the channel
06:00
botlab
left the channel
06:01
botlab
joined the channel
06:06
Rebelj12a
joined the channel
06:07
Rebelj12a
Hmm
06:08
Rebelj12a
Any mechanical engineers? Metal tolerance and all that?
06:11
aombk
left the channel
06:21
Rebelj12a
Nobody is here.. I'll leave a message after the beep.
07:02
wescotte
beep
07:02
wescotte
left the channel
07:03
wescotte
joined the channel
07:28
sebix
joined the channel
07:39
sebix
left the channel
07:41
baldand
left the channel
07:43
baldand
joined the channel
07:45
Rebelj12a
http://www.thephoblographer.com/2014/10/15/camera-manufacturers-speak-future-lenses/
08:32
g3gg0
joined the channel
08:45
philippej
joined the channel
08:45
AndroUser2
left the channel
08:51
botlab
sebastian created T12: test this feature. http://lab.apertus.org/T12
08:52
botlab
sebastian moved T12: test this feature to Unknown Object (Project Column) on the Restricted Project workboard. http://lab.apertus.org/T12
08:55
botlab
sebastian created M1: Review this.
08:56
botlab
philippej updated subscribers of T12: test this feature. http://lab.apertus.org/T12
08:59
Rebelj12a
Thank you for the update botlab mind if I call you HAL?
09:03
philippej
Botlab will be renamed soon :-)
09:05
botlab
philippej added a comment to T6: Research best practice for live video broadcasting community meetings. http://lab.apertus.org/T6
09:09
botlab
sebastian added a comment to T6: Research best practice for live video broadcasting community meetings. http://lab.apertus.org/T6
09:14
philippej
people here don't worry, we'll quickly tune the bot to not bother us that mush
09:16
botlab
philippej added a comment to T6: Research best practice for live video broadcasting community meetings. http://lab.apertus.org/T6
09:19
botlab
sebastian added a comment to T6: Research best practice for live video broadcasting community meetings. http://lab.apertus.org/T6
09:20
se6astian|away
changed nick to: se6astian
09:21
philippej
I'm considering removing the bot notifications for now, and only keep the system that post the url to a task when it is mentioned, sounds good ?
09:22
se6astian
fine for me :)
09:23
se6astian
HAL might indeed be a good name though :D
09:23
philippej
it must be available
09:23
philippej
let me try in a separate session
09:24
philippej
nope
09:24
philippej
already taken
09:31
botlab
left the channel
09:32
lab-bot-1
joined the channel
09:34
lab-bot-1
philippej updated subscribers of T2: Setup auto start of phabricator bot. http://lab.apertus.org/T2
09:43
lab-bot-1
left the channel
09:43
lab-bot
joined the channel
09:45
philippej
notifications disabled for now, the bot will be much mre quiet
09:46
se6astian
great
09:46
se6astian
unless I say T9!
09:46
lab-bot
T9: add AXIOM logo files as EPS/PNG download to logo page - http://lab.apertus.org/T9
09:46
se6astian
haha
09:46
philippej
of course
09:46
philippej
but this sounds good enough imho
09:46
se6astian
agreed
09:47
philippej
btw where are those logo files? I can take care of it if you send them to me
09:48
se6astian
in the threa
09:48
se6astian
thread
09:48
philippej
damn me, I didn't even read
09:50
se6astian
bot, whip philippe for not reading the thread! :P
09:50
philippej
:-)
09:54
philippej
have you noticed that if you start a comment somewhere, it is kept even if you switch computer or reopen a session, very clever.
09:56
se6astian
nice, I didnt notice yet!
10:26
jucar
left the channel
10:28
jucar
joined the channel
10:41
skoezzie
joined the channel
10:44
LiohAu_
joined the channel
10:59
philippej
changed nick to: philippej_away
11:02
Phil
joined the channel
11:02
Phil
changed nick to: Guest96405
11:03
philippej_away
left the channel
11:04
Guest96405
hi @all..
11:18
mars_
hi Guest96405
11:19
Legendin
joined the channel
11:23
Guest96405
should change name.. just got a day off and no beta to play with.. damn, since thr great campaign allmost all my patience is gone.. ;)
11:23
Guest96405
hi mars..
11:24
Guest96405
changed nick to: PhilC
11:25
PhilC
changed nick to: PhilCee
11:50
Bertl_zZ
changed nick to: Bertl
11:50
Bertl
morning folks!
12:07
PhilCee
morning bertl.
12:51
wescotte
left the channel
13:02
comradekingu
left the channel
13:04
PhilCee
left the channel
13:13
danieel
alexML: the aptina and the A7s are rolling shutter only... there is no doubt that anything with RS will have good low light... especially when the readout is 25fps
13:16
alexML
of course
13:17
alexML
but I bet many users will want low light capability instead of global shutter
13:17
Bertl
the campaign results would indicate otherwise (KAC vs CMV)
13:27
wescotte
joined the channel
13:27
alexML
CMV has 300fps and is larger, so I guess that's why everybody wanted it
13:36
philippej
joined the channel
14:08
philippej
left the channel
15:00
se6astian
gotta go
15:00
se6astian
bbl
15:00
se6astian
changed nick to: se6astian|away
15:16
sebix
joined the channel
16:04
sebix
left the channel
16:06
sebix
joined the channel
16:32
philippej
joined the channel
17:28
wescotte
left the channel
18:10
Rebelj12a
Wakey wakey eggs and bacey... Er bakey? Didn't think that one through.. Morning!
18:10
Bertl
morning Rebelj12a!
18:12
Rebelj12a
Nothing like the smell of solder flux in the morning...
18:19
surami
joined the channel
18:20
surami
evening guys!
18:20
Bertl
evening surami!
18:20
sebix
left the channel
18:22
surami
i just spining again on the 4K RAW thing :)
18:22
Rebelj12a
Spinning?
18:23
surami
yep
18:23
Rebelj12a
I imagine being frustrated and spinning in my office chair pretending not to care.
18:23
surami
few days ago I asked about the possibilies
18:23
Rebelj12a
Wheee
18:24
surami
that how to get continous 4K RAW, maybe MLV file on something
18:24
Rebelj12a
I thought it might be posaible?
18:24
Rebelj12a
Bertl I sent you a sort of email on that.
18:24
surami
some body wrote to me
18:25
surami
that there is PCIe on picoboard
18:25
surami
but is it closed?
18:26
surami
and somebody should write a core for it or how is this?
18:26
surami
technically i don't know what is possible
18:26
Rebelj12a
Pcie would be a good bet. That stuff is beyond me.
18:27
surami
http://opencores.org/project,pcie_mini
18:28
surami
somebody could take a look on this, maybe this guy could jump into the project?
18:28
surami
to write the core?
18:28
surami
for 7 series?
18:29
Bertl
the PicoZed comes in 7015 and 7013 varian
18:29
Bertl
+t
18:29
Bertl
both provide up to 4 Multi Gigabit connections
18:29
Bertl
the available IP cores for PCIe are proprietary
18:31
philippej
left the channel
18:31
Bertl
there is further the question how to connect it to a PCIe host/slave
18:33
surami
so the GTX or GTP should be connected to a PCIe host?
18:33
Bertl
normally it is designed to be used for a PCIe card, yes
18:34
Bertl
i.e. the FPGA is _on_ that PCIe card which gets plugged _into_ a PC
18:34
surami
so the PC is the host
18:34
surami
i mean the "driver"
18:35
surami
could you write the information flow?
18:35
surami
i mean so: light -> sensor -> etc..
18:36
Bertl
light -> sensor -> FPGA -> PC ?
18:37
surami
ok :) so simple it is :D
18:37
surami
http://www.hitechglobal.com/Accessories/PCIExpress_Test_Board.htm
18:38
Bertl
this is a board to test PCIe cards
18:39
Bertl
it provides a clock and power and a huge number of SMA connectors to connect the lanes
18:40
surami
so we have to solve the GTX -> PCIe host/slave somehow?
18:41
surami
i don't understand
18:41
Bertl
this is one problem, yes, but another problem is the physical connection
18:41
Bertl
you probably don't want to 'plug in' your Beta into a slot on your PC, no?
18:41
surami
yes i don't want it
18:42
surami
i just thinking in direction m2 ssd
18:42
surami
things
18:42
surami
http://www.addonics.com/products/adm2px4.php
18:42
Bertl
so PCIe is designed for short distances, like 10-15 inch
18:43
Bertl
if the Zynq implements a root complex (i.e. the part a PC/CPU usually provides)
18:43
surami
i think on something cardrige solution
18:43
Bertl
then you can directly connect it to a PCIe based SSD
18:45
surami
i think on that, if a m2 ssd cardridge (unique design, enclosure)
18:45
surami
could be connected somehow to the Beta
18:46
surami
then we need another thing oin the computer (something that i linked)
18:46
surami
but with open back plate
18:46
surami
and the m2 ssd cartridge could be connected to that adapter
18:46
Bertl
now you've lost me
18:47
Bertl
can you draw a picture what you have in mind?
18:47
surami
:) yes
18:47
surami
i can try
18:48
surami
but the question is again the host thing...
18:48
surami
how to write the data to the connected m2 ssd drive
18:49
surami
what connection?
18:49
surami
etc...
18:49
surami
ehhh
18:49
surami
not simple
18:51
surami
if we are that situation, that the information is on the m2 ssd
18:52
surami
then with the adapter we can read in the data on the PC
18:52
surami
but the adapter should be a unique design
18:53
surami
too
18:53
Bertl
well, the SDD won't last very long I guess, the connectors are not really designed to be used that often
18:53
Bertl
(if you are thinking to remove the PCIe based SSD after each shoot)
18:57
surami
yes i thought that
18:57
surami
what connection could be a good solution?
18:57
surami
direct cable?
18:59
Bertl
a cable might be a good solution to transfer the data, but it creates a new problem if you put the SSD into the camera
18:59
Bertl
you need more tranceivers, those for the SSD and those for the cable connection
19:00
surami
with one more board?
19:00
surami
maybe?
19:00
Bertl
you mean another PicoZed?
19:00
surami
yes
19:00
surami
and run parellely
19:00
Bertl
unfortunately that doesn't really help
19:01
Bertl
you would have to transfer the data between the PicoZeds
19:01
Bertl
which again would either require more tranceivers or would consume I/O pins, which are already used for the sensor
19:02
Bertl
so it's all a little tricky there :)
19:03
se6astian|away
changed nick to: se6astian
19:03
surami
so till now only the HDMI out ports are the solution....
19:04
Bertl
it seems so
19:04
se6astian
back
19:04
Bertl
wb se6astian!
19:04
surami
hey!
19:06
LiohAu_
left the channel
19:07
Rebelj12a
left the channel
19:11
surami
and the GTX -> USB 3 connection somehow?
19:12
Bertl
USB 3 is possible but a software nightmare
19:12
Bertl
i.e. you would require a complete USB 3 stack in the FPGA
19:17
surami
this guys maybe?
19:17
surami
http://opencores.org/project,usb3
19:18
Bertl
well, they started september 2011, haven't checked in any files and are in the planning phase
19:18
Bertl
so if all goes well, around 2025 maybe?
19:19
surami
:)
19:29
surami
i can't believe we can only get 4K RAW only through HDMI
19:29
surami
...
19:30
surami
-only
19:35
sebix
joined the channel
19:36
sebix
changed nick to: niemand
19:37
surami
if the physical connenction could be solved somehow, maybe this guys have to be asked: http://opencores.com/project,pcie_sg_dma
19:37
surami
i go now
19:37
surami
cheers all
19:38
surami
left the channel
19:40
dmjnova
joined the channel
19:44
Bertl
just for reference, the linked PCIe scatter gather controller uses the proprietary xilinx softcores
19:59
seku
joined the channel
19:59
seku
evenings :)
20:00
seku
well, i can understand the pinning for pcie or USB3 ... but i feel HDMI can be damn fine, and more than enough for the beta.
20:00
seku
transmitters are cheap, bandwidth is large...
20:01
seku
i suppose :)
20:02
seku
on the PC recording side i would guess it to be something like HDMI in, then buffered in RAM, and written directly to SSD(s), maybe wrapped in MLV?
20:03
dmjnova
Bertl: it seems there are open pcie implementations for fpga
20:04
dmjnova
though I've not tested any
20:04
Bertl
yes, we will have to investigate those
20:04
Bertl
please collect the links somewhere, preferably on the wiki
20:06
miroslav
joined the channel
20:10
jucar
left the channel
20:22
Rebelj12a
joined the channel
20:28
Rebelj12a
Did Hal-9000 get taken down? if not HAL-9000 as a name, GlaDOS would be welcome as well. XD
20:39
se6astian
not sure how many people played portal here
20:39
se6astian
and "HAL" was unfortunately already taken as nick :)
20:39
Rebelj12a
blast well HAL-9000 is one people should know.
20:40
Rebelj12a
Well its official is HAL-9000 heh
20:46
LiohAu
joined the channel
20:48
Bertl
welcome LiohAu!
20:48
LiohAu
thanks :)
20:56
jucar
joined the channel
21:01
slikdigit
joined the channel
21:01
slikdigit
left the channel
21:01
slikdigit
joined the channel
21:09
Rebelj12a
Anybody have the timeline for Beta Release?
21:24
Bertl
what would you like to know?
21:25
Rebelj12a
Bertl the general development timeline, when this are expected or need to be finished etc leading up to beta shipping.
21:26
Bertl
'when this are expected or needed?'
21:26
Bertl
we plan to have the first batch ready on April
21:26
Rebelj12a
Ok
21:26
Rebelj12a
Like the individual parts indepth of development.
21:27
Rebelj12a
Obviously working on sensor right now, others are working on HDMI recording etc.
21:27
Rebelj12a
Deadlines for those parts in order to be ready for batches.
21:27
Rebelj12a
parts - as in, software and hardware, pieces of the beta
21:27
Bertl
we will create some kind of timeline soon, at the moment we are busy processing the input from the campaign
21:27
Rebelj12a
Ah ok
21:27
Rebelj12a
Just wondering, and gathering data for the wiki.
21:27
Bertl
software will not have priority for us
21:27
Rebelj12a
Oh it wont?
21:28
Bertl
many folks will be able to help there (e.g. ML folks) once the hardware is available
21:28
Rebelj12a
Ah ok
21:28
Bertl
so for obvious reasons the hardware has top priority
21:29
Rebelj12a
Makes sense, didnt know. I meant more, sensor works with the OS type software. you know. Everything is read correctly.
21:29
Bertl
we already got the CMV12k running on the alpha, so I do not expect sudden problems there
21:29
Rebelj12a
Ok
21:30
jucar
left the channel
21:31
jucar
joined the channel
21:31
Bertl
we plan to do early Betas
21:31
Bertl
i.e. preliminary hardware which doesn't have all features or isn't final
21:31
Rebelj12a
Ah
21:32
Bertl
and we will offer them to interested folks who are willing to contribute software wise
21:32
Rebelj12a
Oh really?
21:33
Rebelj12a
Thats interesting
21:43
philippej
joined the channel
21:46
niemand
left the channel
21:46
jucar
left the channel
21:47
jucar
joined the channel
21:54
jucar
left the channel
21:56
jucar
joined the channel
21:59
LiohAu
left the channel
22:02
Legendin
left the channel
22:09
jucar
left the channel
22:10
jucar
joined the channel
22:22
philippej
left the channel
22:29
surami
joined the channel
22:38
surami
left the channel
22:39
Rebelj12a
I need a better editing machine than a Laptop D:
22:58
miroslav
left the channel
23:07
se6astian
time for bed
23:07
se6astian
changed nick to: se6astian|away
23:09
pizthewiz
joined the channel
23:11
skoezzie
left the channel
00:11
pizthewiz
left the channel
00:38
ctag_
left the channel
00:38
ctag_
joined the channel
00:59
g3gg0
left the channel