hi howard,
you can help us hunt for ET
by downloading the s...@home screen saver at
http://seti.berkeley.edu
best wishes,
dan
Dan Werthimer
SETI Program Director
University of California, Berkeley
Howard Relles wrote:
Sir/Madam,
Can you tell me how I can help evaluate/process incoming
hi jan,
the ADC time domain data in your plots looks
like it might not be time ordered correctly.
time is perhaps swapped in groups of four:
eg: ADC outputs samples t0, t1, t2, t3, t4, t5
and you might be displaying and analysing in this order:
t3,t2,t1,t0,t7,t6,t5,t4,t11,t10,t9,t8
hi andrew,
i'm ccing aaron parsons on this, designer of the original fft blocks.
aaron might be able to comment on your email.
as far as i know, no one has ever used the overflow outputs
from the FFT blocks, so you may be the first person to debug
this.
best wishes,
dan
Andrew Martens
hi jonathan,
we expect roach boards will be about $2500 each, plus FPGA.
you'll need to add DRAM ($50 min), ATX enclosure ($100), and
ATX power supply ($50).
there are several ADC boards you can use, but only two ADC
board that can sample 1 Gsps currently available:
a) the Atmel/E2V dual
I did get strange results when my sync pulse was too short, but that is
likely just the block resetting before the reorder finishes.
as laura points out, the FFT's minimum sync pulse period is tricky:
see casper memo #24
http://casper.berkeley.edu/memos/sync_memo_v1.pdf
btw, many people have
dear casper collaborators,
we now have five adc boards that can
plug into ibob's and roach's, and there
is some confusion about what to call them.
we had a discussion about ADC nomenclature yesterday,
and we suggest the following names:
ADC2x1000-8
is the casper dual iADC: 2 inputs,
wan,
i strongly advising the standard bee_xps tools
that everyone in the casper collaboration
uses. if you use a different set of tools then
people won't be able to assist you, and your design
won't be able to be used by other people.
best wishes,
dan
wan.ch...@csiro.au wrote:
Hi Henry:
on the modules I will use in my
design if I could understand your library development process. But
unfortunately, there is less documents talking about the casper library
development process. I hope I could get some help at this point.
Thanks
Wan
-Original Message-
From: Dan Werthimer
hi wan,
sync should be zero on every clock except
it should go high for one clock every N cycles,
where N is in henry chen's memo on sync pulses.
for example simulink spectrometer designs and
examples on how to use sync pulse,
see design gallery at:
hi wan,
regarding your question on roach I/O:
in about a month, roach will be running BORPH,
a linux operating system for fpga's
(the operating system we use on the BEE2).
in BORPH, registers, block rams and fifo's
in your simulink design appear as linux files
which you can read and write.
G Jones wrote:
Hi John,
I have brought this up a few time on this and the BEE2 mailing lists
(archives???).
hi glenn,
be careful what you say on these mailings:
billy made a search-able archive of the casper mailing, and
he will put it on the web site soon.
dan
if it works with one ibob board (no chaining),
i suggest buying (or asking xilinx to donate)
four USB download cables.
dan
Rurik Primiani wrote:
Hi everyone,
Thanks for the replies! To clarify a little more, our system is now on
the summit of Mauna Kea so physically checking jumpers and
hi wan,
in case you haven't seen this,
there are some sync pulse guidelines in casper memo 24:
Sync Pulse Usage - July 2008 (Henry Chen, et al)
http://casper.berkeley.edu/wiki/Memos
best wishes,
dan
wan.ch...@csiro.au wrote:
Hi:
After reading the memo of sync pulse, I still feel a little
hi jason zheng,
as you probably know,
a 2^20 point FFT won't fit on an fpga,
as it requires 2^19 times 36 bits of memory.
the largest single fft that we've ever used
on an FPGA is 2^15 points.
to implement higher resolution spectrum analyzers
one needs to implement analysis in
two stages
support
multi-million point FFTs.
~Jason Zheng
On Mon, Jul 27, 2009 at 9:09 AM, Dan Werthimer
d...@ssl.berkeley.edu mailto:d...@ssl.berkeley.edu wrote:
hi jason zheng,
as you probably know,
a 2^20 point FFT won't fit on an fpga,
as it requires 2^19
hi shilpa, glen, and casper collaborators
i'm forwarding your roach questions to casper@lists.berkeley.edu
(roach collaborators - please see email chain appended below -
can you help shilpa and glen?)
in general if anyone has hardware, software, tools, or library problems,
it's best to send
dear casper collaborators,
can anybody please advise ron on his question
appended below about problems with 32K point
wideband real transforms??
ron - does your 32K wideband real FFT work
correctly in simulation?
we've build instruments doing 32K complex transforms,
but these FFT's are
the masks to find that parameter and setting it so that the value is
greater than the number of stages in the transform to see if that fixes it.
Glenn
On Thu, Aug 27, 2009 at 5:16 PM, Dan Werthimer d...@ssl.berkeley.edu
mailto:d...@ssl.berkeley.edu wrote:
dear casper collaborators
to connect the actual iBOB clock signal to the SMA
GPIO in the model file?
Jason
--
Richard Armstrong
ʞn˙ɔɐ˙xo˙oɹʇ...@ƃuoɹʇsɯɹɐ˙pɹɐɥɔıɹ
+44 (0) 79 0682 9979 (UK mobile)
+44 (0) 1865 273597 (Office)
--
Dan Werthimer
Space Sciences Lab and Berkeley Wireless Research Center
University
hi casper collaborators,
can anyone help answer ted's questions below?
thanks,
dan
On 08/29/2009 04:45 PM, Ted Jaeger wrote:
Hi Dan,
Is there a
tutorial anywhere on setting up the software needed for a fully
functional iBOB development system. I have been following the wiki
articles on
hi marco,
it would be great if you and others can acknowledge
the CASPER collaboration, but there are no requirements
or guidelines.
andrew or others - can you email marco a logo?
dan
Hi everyone,
Is there any guideline about how to cite the CASPER, writing
presentations and articles?
hi wan,
i don't know of anyone who has roach ethernet
problems at 100 Mbit/sec.
i'm cc'ing casper community to see if anyone has any ideas.
in general, it's good to post questions to cas...@lists,
so that everyone can help answer, and everyone can see the answers,
and the info will be
hi andrew,
the information in your email below will be helpful
to all instrument developers using FFT block.
is there an easy way to display this information in the
fft parameter selection window?
eg: click here for guidance on the choice of these parameters
or see fft documentation
hi john,
glenn jones is working with terry filiba today on improving the speed
of the corner turn in the FFT.stay tuned.
dan
Hi John
Some pointers on improving timing generally in ROACH.
Decrease the space your design occupies. Implement delays on data
inputs to multiplier cores
mo ohady m...@digicom.org
On 11/18/2009 10:46 AM, Karl Warnick wrote:
Hi all,
I'd like to get in touch with Mo at Digicom to see if there are any
boards currently available. Could someone provide contact information
(or information on available boards from the most recent run)?
Thanks,
of this board).
Best Wishes,
Dan
Tom Kuiper wrote:
Dan Werthimer wrote:
Can you transmit the RF or IF through analog fibers, and then locate
the digitizers in
an electronics lab, where they are easy to service, upgrade, keep
cool, shield from the receivers, ...?
We can, in principal
Hi Tom
We use hardware time tagging in some of the spectrometers we've built,
accurate to a few nS, described below:
We've implemented time tagging in spectrometers and correlators (we can
send you some spectrometer
designs if you want), by running the ADC sample clock from a
synthesizer
hi wan,
what are the levels of this output?
if they are the least significant bits,
then this is likely from round off noise.
best wishes,
dan
On 1/12/2010 6:45 PM, wan.ch...@csiro.au wrote:
Hi All:
I use a matlab sine wave generator to generate a sine wave as an input
to PFB and FFT.
or
minimize the steps on noise floor?
Thanks
Wan
*From:* Dan Werthimer [mailto:d...@ssl.berkeley.edu]
*Sent:* Wednesday, 13 January 2010 2:37 PM
*To:* casper@lists.berkeley.edu
*Subject:* Re: [casper] DC part of FFT output
all different rounding options, I found round +/- inf is
the best, even better than round even or odd. Is this normal?
For wider bit width in FFT, I remember over 18 bits could not work
properly. So now it is fixed and it can work upto 25 bits?
Thanks, Dan.
Wan
From: Dan Werthimer [mailto:d
hi danny,
i think your numbers are about right.
for a noise like signal (eg: from a radio telescope), we typically
drive the adc at levels around -20 to -15 dBm.
best wishes,
dan
On 1/14/2010 8:29 PM, Danny Price wrote:
Hi all, was hoping to get some clarification on real-world ADC bit
, but it important for long FFT's.
best wishes,
dan
On 1/17/2010 4:43 PM, Dan Werthimer wrote:
hi wan,
the noise floor is independent of shifter settings -
the noise RMS is typically about 1 or 2 LSB's
(depending on the length of the FFT), due
to quantization noise and round offer errors.
so you'd like
hi tom,
there's a lot of current work in the areas you asked about:
terry filiba recently ported the ibob based pulsar instrumentation to roach,
(peter mcmahon and she developed this for parkes pulsar work).
jonathan kocz and mathew bailes are working on roach porting as well.
see peter's
wei-chung,
jason is in the karoo deploying a new PAPER correlator so i'm not sure
when he'll be able to read your email.
do you have 1 pulse per second sync signal going into your iADC boards?
you need to drive one of the two iADC boards on each iBOB with a 1 PPS
signal
so all the iBOB's
dear casper collaborators,
our collaboration is growing and
xilinx requested we aggregate most of our donation
requests so xilinx doesn't have to process as many requests.
instead of each university or organization requesting chips,
xilinx proposed they would give us a bulk donation of chips,
hi randy,
the sync outputs from the adc yellow block are a copy of
the signal that is injected into the adc's sync input SMA connector.
(in your case, the 1 PPS signal).all four syncs are identical.
to know which adc sample is taken on the 1 second tick, one needs to
calibrate by looking
i don't think there's a reason, except perhaps decorative.
henry - is there a reason for four sync's?
dan
On 3/4/2010 11:52 AM, Paul Demorest wrote:
Hi Dan,
I have to ask.. if all four syncs are the same, why are there four of
them? ;)
-Paul
On Thu, 4 Mar 2010, Dan Werthimer wrote
hi jason, dave and paul,
regarding syncing up one or two adc's with 1 PPS:
at boot up, the iADC (also called ADC2x1000-8),
yellow block software/gateware aligns two adc's that are plugged into
roach or ibob.
the code does this by sampling the relative phase of the two clocks that
emerge
hi dana,
matt's emails indicates full scale input of an iADC1.1 board is
-2 dBm for a 630 MHz sine wave.
you also asked about maximum input voltage before damage:
we haven't intentionally tried to destroy an ADC board, but we've
accidently injected
strong signals into these boards by
hi rick,
the 10GBe block incorporates multiple clock domain logic already:
you can send your data into the 10Ggbe block at any fpga clock rate
(since the 10Gbe block has fifo's inside) and the 10Gbe block will
transmit
your data using it's internal 156.25 MHz clock. you must take
care
hi gregory,
our group has only tested myricom and chelsio 10Gbe NIC's.
there are a few other NIC boards that have been recommended,
but intel is not on the list.
if you want to test a different NIC, i suggest you
try to find a NIC that features an on board processor
that can process UDP
hi laura,
are you trying to clock your FGPA at 200 or 400 MHz?
if you configure the ADC yellow block to input 4 samples per clock,
and sampling at 1600 Msps, then you will need to clock your
fpga at 400 MHz, which is pretty near impossible for virtex 5
and you'll get lots of timing errors.
if
hi shilpa,
the early CX4 10Gbe spec didn't supply optional power
through the connector, and the first revision bee2's and
ibob's we built didn't have powered connectors.
but the spec changed several years ago, so i would have
guessed that any modern NIC board would have powered
connectors.
ports, and think it's a bad idea. He
says buy the SFP+ NIC, which can also be used with Direct Attach
copper cables, for short runs (but I don't know what that means,
really). Since we are about to buy NIC and Zarlink cable, I'm now
concerned they won't work together. Any thoughts?
Rick
Dan Werthimer
chosen.
dan
On 05/20/2010 10:01 AM, Rick Raffanti wrote:
So I reckon I'll just buy the Chelsio board N310E-CXA- it's the same
price, and specifically states Active Cable Compatible. I'll buy some
5m copper cables and some of the 15m Zarlink. Any reason that won't work?
Rick
Dan Werthimer wrote
hi dean, cc casper lists
here are the doc's we have on the 3Gsps board:
1) schematic
2) pcb layout
3) simulink blocks for interfacing
this adc to casper dsp libraries
and other instruments.
4) verilog and vhdl interface code
(under the simulink block).
5) various
hi jon,
there are a couple of possibilities that might explain
the frequency response you are seeing:
1) the ADC2x1000-8 board has poor termination.
(the input impedance is not 50 ohms at all frequencies).
this causes reflections, which depend on cable length
and frequency.
hi homin,
the 16 input FFT takes twice as many resources as an 8 input FFT.
(the 16 input FFT has 16 parallel inputs so it can compute fft's at 16
times the fpga clock rate).
this factor of two expansion in fpga resources is largely independent of
the input
data bit width, because almost
I'm very sad to report that Don Backer died this afternoon.
Don collapsed just outside his home, probably from a heart attack.
Don had a great life and was a wonderful man.
We will all miss him.
hi tom,
if you don't need much more than 10 Mbit/sec, then your system will
be a lot cheaper if you use the 1Gbit roach ports. you'll have to change
your gateware and software a bit to do this, but you'll save a lot of money
in 10Gbe switches, cables, and nic boards.
if you want to stick
hi andrea,
you might be interested in other roach based seti systems:
laura spitler ported the 128 million channel bee2 seti spectrometer
to roach, and terry filiba developed a roach/GPU seti spectrometer,
where roach digitizes data, breaks it up into sub-bands using a polyphase
filter bank,
hi danny,
i suggest trying jason's items first (below),
and then if you are still having trouble, talk to suraj gowda,
who got got the PFB and FFT to work at about 375 MHz
by using xilinx's plan-ahead and calling out more explicit
utilization of DSP48 slices for the FFT butterflies.
suraj will
end.
I'm using the Chelsio with 1kB packets, haven't figured out how to do jumbo
packets yet (which are a maximum of something like 10kB anyway).
Rick
-Original Message-
From: Dan Werthimer d...@ssl.berkeley.edu
Sent: Sep 24, 2010 4:32 PM
To: rick raffanti rik...@earthlink.net
Cc
hi jason,
we don't have a design to write 3 Gsps ADC data into DRAM,
but if it's useful, suraj and billy have a roach designs that
write data from a pair of 3 Gsps ADC's to block ram
for subsequent readout via borph.
best wishes,
dan
On 9/30/2010 8:10 AM, Jason Ray wrote:
At 09:16 PM
hi jonthan,
you can purchase 48 port SFP+ switches for about $250/port.
you can purchase 20 port CX4 switches for about $500/port.
(eg: Fujitsu XG2000C)
regarding connectors - here's what i've gleaned from switch
vendors and others:
a) CX4 is being phased out.
b) SFP+ is what most vendors
hi laura,
yes, digicom has done this revision on all ibob's since we
learned about the problem.
dan
Laura Spitler wrote:
Hello,
I finally went and checked my iBOB, and it appears that there are
already nylon and metal washers under the CX4 screws. Did Digicom
start implementing this fix in
hi John,
you asked about SPEAD FPGA implementations:
Jason added SPEAD to the packetized correlator.
i'm appending Jason's November 26 update on correlators,
which includes a section on SPEAD.
also, i think a recent pocket correlator, developed by Zaki,
uses SPEAD. Zaki and Aaron have also
to Digicom?
Thanks,
Dale
--
Dan Werthimer
Space Sciences Lab and Berkeley Wireless Research Center
University of Calfornia, Berkeley
hi jonathan,
some ideas for your correlator:
1)
300 MHz is a good target, especially for V6.
suraj has shown how to achieve 375 MHz for V5
by using floor planning and auto-placing.
suraj or i can send you his draft paper on this if you'd like.
2)
you might want to consider FFX instead of
6 GHz BW demux 32 case suggested in
response to Suraj still requires 400 MHz FPGA clock, thus not so
practical. Can one gain a factor of 2 in demux doing quadrature
sampling, and having I and Q inputs to a complex input PFB each at 1/2
the rate?
Jonathan
On Dec 23, 2010, at 5:24 PM, Dan
analog filters.
Apparently they have not worried about complete spectral
coverage with the multitap off-line cross correlator.
I wonder if there are other solutions to this problem.
Regards,
Bob Wilson
On Fri, 24 Dec 2010, Dan Werthimer wrote:
hi jason, jonathan,
regarding jason's concerns
hi jonathan,
as you pointed out, for a phased array feed, where the correlator is
used only to calibrate the beamformer coefficients,
you don't need to compute all the possible cross correlations,
and you don't have to do the computation it in real time.
the calibration time scale requirement
hi ben,
i'm not sure which FFT you are using.
if you are using the biplex FFT, this computes
two complex FFT's at once:
two complex data streams go in,
each with 2^N complex points,
and two 2^N complex point streams come out. (not 2^(N-1)).
i think the frequency ordering of the biplex FFT
hi tom,
i suggest you purchase a myricom 10Gbe CX4 NIC board,
rather than a chelsio board. the myricom boards work
well and jason wrote a casper memo with test results on myricom nics.
myricom has drivers optimized for UDP packets,
and drivers that skip over kernel memory and transfer data
hi tom,
one more note:
if you use fiber optic CX4 cables,
please see the warning at
http://casper.berkeley.edu/wiki/Recommended_10_GbE_Hardware
not all NIC boards have built in power to support
fiber optic cables. check with myricom.
the ibob/bee2/roach boards have built in power.
dan
.
Rick
On 1/28/2011 5:53 PM, Dan Werthimer wrote:
hi tom,
one more note:
if you use fiber optic CX4 cables,
please see the warning at
http://casper.berkeley.edu/wiki/Recommended_10_GbE_Hardware
not all NIC boards have
Dear Casperites,
If you are interested in attending the mini-casper
workshops in Taipei (March 23-25) or Beijing (March 28-30),
please contact the workshop organizers listed below.
We'll have talks about casper hardware, tools and libraries,
discussions about correlators, beamformers,
hi tom,
you could connect the TTL signal to the SMA connector
on the adc board, called sync in.
this is a spare digital input, 0 to 2.5 volts, 50 ohm.
but it would be easier to connect your signal to a printer port on your
computer,
or buy an inexpensive digital input to USB board.
), I'm ok with Ubuntu but with this OS I had some
issues making things work, if some of you has passed for this before,
I'd be happy to know how you resolve it.
Regards,
Daniel H.
--
Dan Werthimer
Space Sciences Lab and Berkeley Wireless Research Center
University of Calfornia, Berkeley
hi andy,
it's probably fine for experts like yourself to use non-supported OS's,
but i wouldn't recommend it for non experts:
we have run into some strange and subtle problems that i wouldn't have
guessed would be operating system related. for instance, some of the FFT's
in the simulink
hi miguel,
i recommend that you only use an operating system
supported by xilinx. we've encountered strange
errors and found instruments that produce incorrect
results when using non-supported operating systems.
and xilinx won't answer any questions or provide support
if you use a
hi danny, cc stuart rumley, valon technology
i agree with dana - if you use a switching power
supply to power a synthesizer, the phase noise
and jitter will likely be terrible and you'll get spurs
and degraded SNR.
i suggest you check with stuart rumley, the designer
of the valon
hi david,
homin jiang at ASIAA has developed an ADC board
with E2V's EV8AQ160 (8 bit),
but not the AQ190 (10 bit).
these two ADC's are very similar, almost
pin for pin compatible, so it might
be possible to modify homin's design.
best wishes,
dan
On 5/10/2011 2:26 PM, David Forbes wrote:
hi tom,
regarding pushing the KatADC from 2*1.5 Gsps to 4 Gsps:
it will be difficult, perhaps impossible, to
move 4 Gsps 8 bit ADC data through a single zdoc connector
into roach I. each of the 32 bit lvds lanes would
have to operate at 1 Gbit/sec, and we've never
tried that on Roach I.
hi mandana,
where you set the ADC input level depends on the
RFI environment. it's a balance between
setting the level high enough to minimize
quantization noise, yet low enough that
the input signal doesn't saturate the ADC very often.
if you don't have much RFI,
a good level for an 8 bit
hi luis,
i think mo ohady at digicom just bought 200 of these.
he can probably sell you some:
m...@digicom.org
best wishes,
dan
On 6/14/2011 1:57 PM, Luis Quintero wrote:
All,
I wonder if someone knows where to buy in US the Viking Modular DDR2
memories (1Rx8 PC2-5300R-555-13-F0)
Hi Glenn and others interested in frequency synthesizers:
The $295 Valon Technology 5007 dual synthesizer board
has two synthesizers that cover 138 MHz to 4400 MHz.
Phase noise and other specs at:
http://valontechnology.com/5007%20product%20data%20sheet%2011-04-10.pdf
Jitter at 1500 MHz is
hi jesus,
to follow up on rick and jack's comments:
to use the dual =1Gsps ADC at 1200 Msps,
then as you point out, you need to use the
ADC in interleave mode:
you need to supply a 600 MHz clock to the ADC board;
and the ADC will then supply a 150 MHz clock
to the FPGA (the ADC clock is
hi aziz,
can you use the standard casper libraries and tools?
if you can use the standard libraries and tools,
then you will be able to share your designs with
the casper community as well as use the casper libraries
and open source instrument designs.
you'll also be able to get a lot of help
hi gerry,
adc boards are listed at
https://casper.berkeley.edu/wiki/Hardware
but there's no price sheet. you can get quotes
for most of these open source boards
from m...@digicom.org
you might consider the
ADC1x5000-8https://casper.berkeley.edu/wiki/ADC1x5000-8,
which can
also be used as a dual
hi kaushal,
francois provided numbers for worst case power consumption of roach II.
they also have made measurements: about 65 watts typical for lightly
loaded fgpa
design, measured at the 220VAC input (includes the losses in power
supplies, etc).
for a heavily loaded fpga design using the
hi marcus,
there are several casper instruments that digitize data using
ADC's plugged into one or more FGPA board(s).
The FPGA boards packetize and time stamp the ADC data
and send it over 10Gbit ethernet
to one or more CPU's and GPU's.
here are a couple of examples.
the VEGAS instrument
hi david,
there are a few casper instruments that are similar to what you are
suggesting.
1) some of the VEGAS spectrometer modes use a roach to digitize the data and
send almost raw adc samples to a GPU where the spectroscopy is done.
2) some of the GUPPI pulsar machines use casper hardware to
hi joe,
can you leave the DC-DC converters on the roach board?
then you only need to supply the 12, 5 and 3.3 volts that
come from the ATX power supply.
best wishes,
dan
On Thu, Jan 19, 2012 at 10:28 AM, Joseph Greenberg jgree...@nrao.eduwrote:
We at NRAO are going to mount the Roach I
, 2012 at 10:53 AM, Joseph Greenberg jgree...@nrao.eduwrote:
**
Hi Dan,
Thanks for the reply.
Dan Werthimer wrote:
hi joe,
can you leave the DC-DC converters on the roach board?
Yes. Are they standard on all Roach I boards or do we have to specify
them?
then you only need to supply
eliminate using the on-board 3.3V. Too bad tthe Pico device
doesn't come in a 48V model.
Thanks for all the help,
Joe
Dan Werthimer wrote:
hi joe,
the DC-DC converters are standard on the roach boards,
so you only need to supply 3.3, 5 and 12 volts to the ATX connector.
if you use john ford's
hi billy,
the SX95T is the best Roach I chip for correlators, fft's and beamformers,
because it has a lot of DSP48 multiplier/accumulator blocks.
the LX110T has more slices but doesn't have as many DSP blocks.
best wishes,
dan
On Tue, Jan 31, 2012 at 6:27 PM, Barott, William Chauncey
hi tom,
beware of non-xilinx supported operating systems.
they can waste a lot of your time.
we've experience twisted and frustrating behavior with non-xilinx supported
operating systems. in one case, our design compiled without any error
or warning messages, but the FFT didn't produce the
hi tom,
if you are using kat-ADC's, you'll need to change
the adc yellow block in the tutorial.
dan
On Fri, Mar 2, 2012 at 3:57 PM, Kuiper, Thomas (3266)
kui...@jpl.nasa.govwrote:
I think that's it! I after replacing the file system, I made the
necessary changes from memory and general sys
hi billy,
info on copper and optical cables we've tested is at:
https://casper.berkeley.edu/wiki/Equipment_Cables
virtex2 fpga's (ibob, bee2) do not meet 10GbeCX4 specifications,
but virtex5 fpga's (roach) does meet the 10GbeX4 spec,
so 5 meter CX4 cables should work.
the longest cable we've
hi laura,
there are a couple of standard ways people stream data to the PPC and out
over the network:
ping pong memories
-
the FGPA write to memory A and then sets a flag saying memory A is full.
the PPC code waits for this flag to be set and then moves memory A to
hi laura,
i second john's remarks:
an inexpensive 10Gbit nic card in your computer
would make your data streaming task relatively easy,
as there are tutorials and several instrument designs
that stream data from a roach over 10Gbe into a computer.
there's also a lot of good open source
a...@luto.us wrote:
On Mar 14, 2012 3:26 PM, Dan Werthimer d...@ssl.berkeley.edu wrote:
hi laura,
i second john's remarks:
an inexpensive 10Gbit nic card in your computer
would make your data streaming task relatively easy,
as there are tutorials and several instrument designs
to.) He had suggested that
thishttp://www.cdw.com/shop/products/Myricom-Myri-10G-network-adapter/2034587.aspxmay
work, but I would be grateful for input from others that are currently
using a RedHat machine with a 10GbE NIC.
--Laura
On Thu, Mar 15, 2012 at 9:45 AM, Dan Werthimer d
hi louis,
i urge people to use only xilinx supported operating systems.
you might get other operating systems to work in the sense they might
not produce error messages and appear to compile your designs.
but we've encountered strange results when compiling with
non supported operating systems.
article on Xilinx new Vivado Design tools:
http://www.eejournal.com/archives/articles/20120501-bigdeal/
hi bill,
i think all the boards you mention have analog bandwidth out to 2 GHz,
so they should work well for your 1-2 GHz band.
the asiaa board is the least expensive, but this board does not have
programmable
attenuators like the Kat-ADC. the asiaa board can be used as a single
5 Gsps ADC, or
: Dan Werthimer d...@ssl.berkeley.edu
To: Bill Petrachenko wtpe...@yahoo.ca
Sent: Monday, May 28, 2012 11:21:40 PM
Subject: Re: [casper] 1-2 GHz sampler
the kat-adc has two different front end fiilter options.
make sure to order the filter option for high frequency.
dan
On Mon, May 28, 2012
hi ian,
the ibob uses the old virtex2Pro fpga's, which do not meet 10Gbit ethernet
spec,
so 3 meters is about as far as you can go with copper cables.
as dave mentioned, if you want to go 3 to 100 meters you can use optical
CX4
cables. there are some part numbers of optical and copper cables
hi alex,
sorry - you can't adapt the 10Gbe port to a 1Gbe port
as the CX4 and RJ45 signals are not electrically compatible.
there is, as you probably know, a 1Gbe port on the roach I,
but it's not directly connected to the FPGA -
the data from the FPGA must go through the CPU to the 1Gbe port,
1 - 100 of 313 matches
Mail list logo