Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarApr2017

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search

Messages from 129075

Article: 129075
Subject: Re: ML505 with Petalinux
From: John Williams <jwilliams@itee.uq.edu.au>
Date: Thu, 14 Feb 2008 11:25:53 +1000
Links: << >>  << T >>  << A >>
Hi,

muthusnv@gmail.com wrote:
> Has anyone successfully ported Petalinux in Microblaze (ML505) and

There's an ML506 reference design in PetaLinux v0.30, which should be 
fairly simple to modify to run on a ML505.

> used the SATA AHCI driver?
> Is the AHCI driver from Petalinux distribution working OK?

I'm not aware of anyone using it - it is jsut the same driver from a 
standard 2.6.20 kernel.  Assuming you've got a working PCI layer up and 
running, I wouldn't expect great difficulties with the AHCI driver.

Regards,

John



Article: 129076
Subject: Re: setup time not met in Quartus
From: Mike Treseler <mike_treseler@comcast.net>
Date: Wed, 13 Feb 2008 18:31:09 -0800
Links: << >>  << T >>  << A >>
C-M, Chang wrote:

> But unfortunately again, my boss deny to change design 

In that case, you are done.

            -- Mike Treseler

Article: 129077
Subject: Re: setup time not met in Quartus
From: Peter Alfke <alfke@sbcglobal.net>
Date: Wed, 13 Feb 2008 19:18:44 -0800 (PST)
Links: << >>  << T >>  << A >>
Great adventures in Engineering:

"Fix this, but do not change anything !"

Peter Alfke


On Feb 13, 6:31=A0pm, Mike Treseler <mike_trese...@comcast.net> wrote:
> C-M, Chang wrote:
> > But unfortunately again, my boss deny to change design
>
> In that case, you are done.
>
> =A0 =A0 =A0 =A0 =A0 =A0 -- Mike Treseler


Article: 129078
Subject: Virtex-4 input pad failures
From: "RobJ" <rob@abc.net>
Date: Wed, 13 Feb 2008 20:00:56 -0800
Links: << >>  << T >>  << A >>
Out of a lot of 12 boards we've had two now that have experienced input pad 
failures (shorts to VCCO). These are LVDCI_25 IBUFs with VCCO = 2.5V. They 
are being driven by 2.5V LVCMOS output buffers from a Marvell Phy (RGMII 
interface). The Phy and Virtex-4 I/O are powered by the same 2.5V supply. 
The traces are in the 1.5" to 2" range and are unterminated. The engineer 
responsible for the board suspects overshoot, and says he has observed 
overshoot of up to 500mV on these lines (which happens to be the absolute 
max for Virtex-4 2.5V inputs).

These traces and pins are not exposed so ESD is almost impossible as a cause 
of the failures. The Virtex-4 inputs are pretty beefy (well protected), and 
I personally have never seen an input buffer fail on any IC due to overshoot 
from a compatible driver (as opposed to being overvoltaged by an 
incompatible driver). So I have my doubts about overshoot causing the 
failures and want to make sure we understand this before building more 
boards. The current plan for the respin is to add series terminators to damp 
the signals. Even if the problem was not overshoot, the resistors are a good 
idea. I'm just looking for ideas and experiences on this.

Thanks, boys (and girls!).

Rob 



Article: 129079
Subject: Re: When are FPGAs the right choice?
From: Sky465nm@trline5.org
Date: Thu, 14 Feb 2008 05:58:34 +0100 (CET)
Links: << >>  << T >>  << A >>
>It's an interesting idea, and when mainstream processors start having
>>16 cores, it will also be interesting to see when people start
>"throwing away" things like x86 cores in a similar fashion.

Let's hope they throw away the whole x86 architecture. It's flawed from the
start.


Article: 129080
Subject: Erratic Behavior of Virtex 4 FPGA
From: Vijayan <vicchemical@gmail.com>
Date: Wed, 13 Feb 2008 21:22:02 -0800 (PST)
Links: << >>  << T >>  << A >>
Hi all,

I am using  Virtex 4 FPGA (XC4VFX100) in our application. In this, we
find some erratic behaiour of FPGA.

Whenever the bit file is loaded into the FPGA, there is a drop in core
voltage(power module getting shutdown).Bit file gives such random
behaviour uses particular three banks for its logic. Here we thought
like this, it may be due to any internal damage in FPGA IOBs around
that three banks.

Here is the few points added,

* We tried with a bit file that has high logic utilization (logic is
not on that faulty banks. it uses other banks). It works fine.
* Also we tried some other bit files with low logic utilization using
that banks is working properly (power won't disturbed).
* Whatever signals from that faulty banks are not shorted with VCC/
GND. We probed it.
* Good core voltage power solution.

Pls suggest me the solution to identify the exact source of that
behaviour.

Regards,
Vijayan

Article: 129081
Subject: Re: Erratic Behavior of Virtex 4 FPGA
From: mh <moazzamhussain@gmail.com>
Date: Wed, 13 Feb 2008 21:56:28 -0800 (PST)
Links: << >>  << T >>  << A >>
On Feb 14, 10:22 am, Vijayan <vicchemi...@gmail.com> wrote:
> Hi all,
>
> I am using  Virtex 4 FPGA (XC4VFX100) in our application. In this, we
> find some erratic behaiour of FPGA.
>
> Whenever the bit file is loaded into the FPGA, there is a drop in core
> voltage(power module getting shutdown).Bit file gives such random
> behaviour uses particular three banks for its logic. Here we thought
> like this, it may be due to any internal damage in FPGA IOBs around
> that three banks.
>
> Here is the few points added,
>
> * We tried with a bit file that has high logic utilization (logic is
> not on that faulty banks. it uses other banks). It works fine.
> * Also we tried some other bit files with low logic utilization using
> that banks is working properly (power won't disturbed).
> * Whatever signals from that faulty banks are not shorted with VCC/
> GND. We probed it.
> * Good core voltage power solution.
>
> Pls suggest me the solution to identify the exact source of that
> behaviour.
>
> Regards,
> Vijayan

Vijayan,
I think that there is roughly a 10~20 % rise in current during
configuration process,

1-  What is the current being drawn from source during the
configuration process in both working and failing bitstreams and is
the voltage regulator capable of providing enough power and provide
good voltage regulation under transients?
2-  Have you seen X-ray inspection of BGA of short circuiting in those
specific banks ?

Just random thoughts, Hope some one with more experience would come up
with better ideas.

/MH





Article: 129082
Subject: i need ur help
From: rabbiaqamar@yahoo.com
Date: Wed, 13 Feb 2008 21:58:44 -0800 (PST)
Links: << >>  << T >>  << A >>
 i want information about fpga boards. for my application i required a
fpga board with pcie interface and 10 Giga bit /s interface . i have
searched a lot but i didn't find anyone. how can i find it.
another question i want to ask is i am studying the manual of startix
II GX PCI express development board .does it have 10 Gig interface. it
has HSMC port. to use this card which daughter card will be used.but
remember i need 10 Gig interface.

Article: 129083
Subject: i need fpga board with 10 Gig interface and pcie interface
From: rabbiaqamar@yahoo.com
Date: Wed, 13 Feb 2008 22:14:03 -0800 (PST)
Links: << >>  << T >>  << A >>
hi all, for my application i need a fpga board which has PCI express
interface and 10 Gig ethernet or fibre channel interface. plz tell me
if there is any board.
one more thing startix II GX PCI Express development board of Altera
has10 Gig interface or not
if not which daughter card can be used to make it for 10 Gig.

Article: 129084
Subject: Re: i need fpga board with 10 Gig interface and pcie interface
From: Allan Herriman <allanherriman@hotmail.com>
Date: Thu, 14 Feb 2008 17:35:10 +1100
Links: << >>  << T >>  << A >>
On Wed, 13 Feb 2008 22:14:03 -0800 (PST), rabbiaqamar@yahoo.com wrote:

>hi all, for my application i need a fpga board which has PCI express
>interface and 10 Gig ethernet or fibre channel interface. plz tell me
>if there is any board.

>one more thing startix II GX PCI Express development board of Altera
>has10 Gig interface or not

If it's the development board I'm thinking of, it doesn't do 10Gb/s.
The onboard StratixIIGX SERDES only works to about 6Gb/s.  You could
do 4G Fibrechannel or 1G Ethernet using the SFP sockets supplied.


It's not too hard to roll your own 10Gb/s board though, but you'll
need to use external SERDES parts if you want it to work.  See AMCC,
Vitesse, etc.
A couple of years ago I would have said that XFP was the only 10Gb/s
serial interface to use; now I would consider SFP+, as SFP+ will
probably be much cheaper than XFP over the lifetime of products
currently being designed.

Regards,
Allan

Article: 129085
Subject: Re: State machine outputs and tri-state
From: "Grumps" <grumpsnothere@hotmail.com>
Date: Thu, 14 Feb 2008 07:22:31 -0000
Links: << >>  << T >>  << A >>
<bobster.thelobster@yahoo.co.nz> wrote in message 
news:6e5dd37a-c594-479e-95a5-19d6136073bc@m34g2000hsf.googlegroups.com...
> Sorry only half the post went :-) here it is again
>
> Hi Grumps,
>
> You say you・re a beginner, so please allow me to offer the one piece
> of advice that you should never waver from.
> You are designing hardware, its not software, if you ignore that point
> you・ll join the ranks of people that are giving the industry a bad
> name with their unstable designs.
> Always try to imagine the circuit you・re describing, and you won・t go
> far wrong.
>
> I・m saying that only because you said you were a beginner, not because
> of you code .
>
>
> Anyway look at a state machine, it・s usually a mishmash of flipflops
> and logic gates.
> A tristate is usually a buffer on a bus. In hardware that・s 2
> different jobs.
>
> So do something like this
>
> 1st describe a tri-state buffer something like
>
>
> signal MY_BUFFER_OUT, MY_BUFFER_IN, MY_BUFFER_CONTROL : std_logic;
>
> MY_BUFFER_OUT <= MY_BUFFER_IN when MY_BUFFER_CONTROL = .1・ else .Z・;
>
> Assign MY_BUFFER_CONTROL from your state machine, only assign 1 and 0
> to RDY, connect MY_BUFFER_OUT to your port and MY_BUFFER_IN to the
> equivalent of  RDY.
>
> This looks exactly the same if you think of it in software terms but
> it・s a different beast altogether if you take a hardware view.
> I・d just like to add, I・m glad to see your using ISE, unlike the more
> expensive tools it wont accept nonsense, my advice is stick with ISE
> until your competent, then don・t lose those good coding habbits.
>
> Let me know how it goes Bobsterthelobster

Thanks for the advice and suggestions.
Your code is (similar to) what I ended up doing.

It's funny, I've been a hardware engineer for ages, and done FPGA design 
with Viewdraw schematics many years ago. We now have a VHDL department, so 
that's our preferred FPGA design route now. But this little project I 
thought I'd do myself. It works, and only takes 22 slices of a spartan3 
(there is a little more to it than this 9-state state machine).

I'm sure I'll be back here with more questions in the future.
All the best. 



Article: 129086
Subject: Re: Erratic Behavior of Virtex 4 FPGA
From: "Grumps" <grumpsnothere@hotmail.com>
Date: Thu, 14 Feb 2008 07:25:59 -0000
Links: << >>  << T >>  << A >>
"Vijayan" <vicchemical@gmail.com> wrote in message 
news:52a8d239-b0af-416a-8838-67ce47f90129@s13g2000prd.googlegroups.com...
> Hi all,
>
> I am using  Virtex 4 FPGA (XC4VFX100) in our application. In this, we
> find some erratic behaiour of FPGA.
>
> Whenever the bit file is loaded into the FPGA, there is a drop in core
> voltage(power module getting shutdown).Bit file gives such random
> behaviour uses particular three banks for its logic. Here we thought
> like this, it may be due to any internal damage in FPGA IOBs around
> that three banks.

Does the core vltg drop during configuration or after (or both) ?



Article: 129087
Subject: Re: State machine outputs and tri-state
From: backhus <nix@nirgends.xyz>
Date: Thu, 14 Feb 2008 08:31:28 +0100
Links: << >>  << T >>  << A >>
Grumps schrieb:
> [I posted this to comp.lang.vhdl, but maybe you FPGA experts know the 
> answer.]
> 
> Hi
> I'm not a VHDL expert, just learning, so please don't shout.
> 
> I'm using Xilinx ISE9.2sp4 and have the following code as part of a state
> machine:
> CP_IN_OUTPUT_DECODE: process (state_cp_in)
>   begin
>     if state_cp_in = sta_idle then
>     RDY <= 'Z';
>     BUSY <= '0';
>   end if;
> 
>   if state_cp_in = sta_1 then
>     RDY <= '0';
>     BUSY <= '0';
>   end if;
> 
>   if state_cp_in = sta_2 then
>     RDY <= '1';
>     BUSY <= '1';
>   end if;
> ...
> ...etc
> 
> The state machine goes from sta_idle, then to sta_1, then to sta_2, etc.
> RDY is a pin on the device.
> During operation, I can see RDY go low in sta_1, but not high in sta_2. I
> know it gets to sta_2 as I can observe BUSY. I don't think RDY is tri-stated
> in sta_2 as there is an external pull-up; it just stays low. Gray encoding
> is used.
> 
> If I change the RDY to rdyi (signal) and then have:
> RDY <= 'Z' when state_cp_in = sta_idle else rdyi;
> outside of the decode process then it all behaves itself.
> 
> Apart from lack of experience, what mistake(s) have I made?
> Thanks.
> 
Hi Grumps,
you did no simulation? so how can you be sure that your FSM ever enters 
sta_2 ?

If you are using an if chain instead of a case for Output decoding you 
must keep in mind, that the last valid condition sets the output. even a 
typo can bring you in deep trouble.  A case on the other hand has a more 
straightforward behavior. And the synthesis tool would complain when 
conditions appear twice.

If RDY is a port there should be no problem with the 'Z' assignment.

I strongly recommend to do a simulation.

and if you like, you can try to rewrite your code like this (if applicable):

case state_cp_in is
   when sta_idle => RDY <= 'Z';
                    BUSY <= '0';
   when sta_1    => RDY <= '0';
                    BUSY <= '0';
   when sta_2    => RDY <= '1';
                    BUSY <= '1';
...
end case;

Have a nice synthesis
   Eilert

Article: 129088
Subject: Re: State machine outputs and tri-state
From: "Grumps" <grumpsnothere@hotmail.com>
Date: Thu, 14 Feb 2008 08:14:55 -0000
Links: << >>  << T >>  << A >>
"backhus" <nix@nirgends.xyz> wrote in message 
news:fp0qmq$ebt$1@news.hs-bremen.de...
> Grumps schrieb:
>> [I posted this to comp.lang.vhdl, but maybe you FPGA experts know the 
>> answer.]
>>
>> Hi
>> I'm not a VHDL expert, just learning, so please don't shout.
>>
>> I'm using Xilinx ISE9.2sp4 and have the following code as part of a state
>> machine:
>> CP_IN_OUTPUT_DECODE: process (state_cp_in)
>>   begin
>>     if state_cp_in = sta_idle then
>>     RDY <= 'Z';
>>     BUSY <= '0';
>>   end if;
>>
>>   if state_cp_in = sta_1 then
>>     RDY <= '0';
>>     BUSY <= '0';
>>   end if;
>>
>>   if state_cp_in = sta_2 then
>>     RDY <= '1';
>>     BUSY <= '1';
>>   end if;
>> ...
>> ...etc
>>
>> The state machine goes from sta_idle, then to sta_1, then to sta_2, etc.
>> RDY is a pin on the device.
>> During operation, I can see RDY go low in sta_1, but not high in sta_2. I
>> know it gets to sta_2 as I can observe BUSY. I don't think RDY is 
>> tri-stated
>> in sta_2 as there is an external pull-up; it just stays low. Gray 
>> encoding
>> is used.
>>
>> If I change the RDY to rdyi (signal) and then have:
>> RDY <= 'Z' when state_cp_in = sta_idle else rdyi;
>> outside of the decode process then it all behaves itself.
>>
>> Apart from lack of experience, what mistake(s) have I made?
>> Thanks.
>>
> Hi Grumps,
> you did no simulation? so how can you be sure that your FSM ever enters 
> sta_2 ?

Just a guess! All other state outputs seemed to follow my design. Only the 
RDY (which was the only one with the 'Z') was errant. What I posted was a 
cut-down version to make the issue clearer.

> If you are using an if chain instead of a case for Output decoding you 
> must keep in mind, that the last valid condition sets the output. even a 
> typo can bring you in deep trouble.  A case on the other hand has a more 
> straightforward behavior. And the synthesis tool would complain when 
> conditions appear twice.

Yes, using an if chain. I can't see any typos, and all state conditions are 
tested.

> If RDY is a port there should be no problem with the 'Z' assignment.

RDY's an inout port.

> I strongly recommend to do a simulation.

I will. Not to check my design, which now works, but to see how the 
simulator works.

> and if you like, you can try to rewrite your code like this (if 
> applicable):
>
> case state_cp_in is
>   when sta_idle => RDY <= 'Z';
>                    BUSY <= '0';
>   when sta_1    => RDY <= '0';
>                    BUSY <= '0';
>   when sta_2    => RDY <= '1';
>                    BUSY <= '1';
> ...
> end case;

I will try that. Should only take a few minutes.

> Have a nice synthesis

Thanks :)



Article: 129089
Subject: Re: microblaze firmware + UART handshaking blues
From: David Brown <david@westcontrol.removethisbit.com>
Date: Thu, 14 Feb 2008 09:56:15 +0100
Links: << >>  << T >>  << A >>
chrisdekoh@gmail.com wrote:
> Hi,
> 
> 1) Does a microprocessor in debug mode run on a slower clock? I din
> know that

Sometimes cpus will run slower when you are using a debugger (for 
example, the debugger might have tracing features that require the cpu's 
cache to be disabled, or it might "simplify" the processor in other 
ways).  However, I think what Alex meant was that the debug compile 
flags will generate slower code than when you pick flags for more 
optimisation.  When using debug flags, the compiler will try to keep the 
structure of your program intact (such as its loops and functions), to 
make it easier to follow with a debugger.  When doing full optimisation, 
the compiler will do more inlining and loop unrolling if that makes the 
code smaller and/or faster.

> 2) under what circumstances should what variables be volatile?
> 

A variable should be "volatile" to indicate that its value might be read 
or changed without the compiler knowing about it (including being 
accessed from another C function if the compiler does not know that it 
may also run at any time, such as an interrupt function or a separate 
thread).

mvh.,

David


> Chris
> 
> 
> On Feb 14, 5:58 am, Alex Freed <al...@mirrow.com> wrote:
>> GMM50 wrote:
>>> Hi All:
>>> This may indeed be a problem but how does debug make it all work.
>>> george
>> In my experience the most common scenario is a lack of "volatile"
>> keyword. Debug version turns optimization off and release version
>> optimizes a read of a hardware register out, 'cause it doesn't know it
>> is hardware and nothing in the code changes it's value.
>>
>> The second possibility is a race condition. Dubug version may execute
>> more slowly.
>>
>> -Alex.
> 

Article: 129090
Subject: Re: State machine outputs and tri-state
From: "Grumps" <grumpsnothere@hotmail.com>
Date: Thu, 14 Feb 2008 09:33:09 -0000
Links: << >>  << T >>  << A >>
"Grumps" <grumpsnothere@hotmail.com> wrote in message 
news:61ibkcF1vc3kkU1@mid.individual.net...
> "backhus" <nix@nirgends.xyz> wrote in message 
> news:fp0qmq$ebt$1@news.hs-bremen.de...
>> Grumps schrieb:
[ snipped ]
>> and if you like, you can try to rewrite your code like this (if 
>> applicable):
>>
>> case state_cp_in is
>>   when sta_idle => RDY <= 'Z';
>>                    BUSY <= '0';
>>   when sta_1    => RDY <= '0';
>>                    BUSY <= '0';
>>   when sta_2    => RDY <= '1';
>>                    BUSY <= '1';
>> ...
>> end case;
>
> I will try that. Should only take a few minutes.

I replaced my original process based on lots of if <state> = <blah>, with 
the case method (above). It works!
I don't fully understand why though.
I simply replaced the "if state_cp_in = " with "when" in my original code so 
any typos would be preserved.



Article: 129091
Subject: Re: Virtex-4 input pad failures
From: "Symon" <symon_brewer@hotmail.com>
Date: Thu, 14 Feb 2008 09:56:08 -0000
Links: << >>  << T >>  << A >>
RobJ wrote:
> Out of a lot of 12 boards we've had two now that have experienced
> input pad failures (shorts to VCCO). These are LVDCI_25 IBUFs with
> VCCO = 2.5V. They are being driven by 2.5V LVCMOS output buffers from
> a Marvell Phy (RGMII interface). The Phy and Virtex-4 I/O are powered
> by the same 2.5V supply. The traces are in the 1.5" to 2" range and
> are unterminated. The engineer responsible for the board suspects
> overshoot, and says he has observed overshoot of up to 500mV on these
> lines (which happens to be the absolute max for Virtex-4 2.5V inputs).
>
> These traces and pins are not exposed so ESD is almost impossible as
> a cause of the failures. The Virtex-4 inputs are pretty beefy (well
> protected), and I personally have never seen an input buffer fail on
> any IC due to overshoot from a compatible driver (as opposed to being
> overvoltaged by an incompatible driver). So I have my doubts about
> overshoot causing the failures and want to make sure we understand
> this before building more boards. The current plan for the respin is
> to add series terminators to damp the signals. Even if the problem
> was not overshoot, the resistors are a good idea. I'm just looking
> for ideas and experiences on this.
> Thanks, boys (and girls!).
>
> Rob

Hi Rob,
I also doubt that the overshoot is to blame, and your caution is 
appropriate. Here's some questions, I apologise if some of them are teaching 
you to suck eggs!

How do you know the I/P pad failure is short to VCCO?
Did each board failed the same single FPGA pin?
You've got a ground plane, right?
What's the rise time of the signals?
Did replacing the FPGA on the board fix the problem?
How did your mate measure the overshoot? (I.e. can he work a high speed 
'scope properly?)
Does he wear an earth strap or nylon pants when probing? (Ahem!)
Any 12V traces nearby on the board that a probe can short to these signal 
lines?
Have you tried simulating with Hyperlynx or somesuch?
Did you ask Xilinx to take a look at the broken parts?
Are the two failures from the same batch?
Are you sure the 2.5V power supply can't overvoltage?

HTH. and good luck, Syms.



Article: 129092
Subject: Re: i need fpga board with 10 Gig interface and pcie interface
From: "Symon" <symon_brewer@hotmail.com>
Date: Thu, 14 Feb 2008 10:01:19 -0000
Links: << >>  << T >>  << A >>
Allan Herriman wrote:
> A couple of years ago I would have said that XFP was the only 10Gb/s
> serial interface to use; now I would consider SFP+, as SFP+ will
> probably be much cheaper than XFP over the lifetime of products
> currently being designed.
>
> Regards,
> Allan

Hi Allan,
Thanks for your post. Have you used SFP+ yourself yet? I'd be interested to 
hear your opinions and experiences.
Thanks, Symon. 



Article: 129093
Subject: Re: Is a FPGA the solution ?
From: "RCIngham" <robert.ingham@gmail.com>
Date: Thu, 14 Feb 2008 06:15:26 -0600
Links: << >>  << T >>  << A >>
>I would like to know if a FPGA is suited for this application: sound
>sources localization and separation.  The algorithm comes in 3 steps:
>sources detection using a steered beamformer (in frequency-domain), a
>particle filter to track the sources and a combination of a geometric
>source separation algorithm and a non-linear post-filter for source
>separation.
>
>Also, the audio stream comes from an array of 8 microphones, and all
>possible microphone pairs must be considered during the detection
>phase.
>
>This algorithm is already implemented of a floating point DSP and I
>would like to know if a FPGA could be a good choice to improve
>performance.

What aspect(s) of "performance" are you trying to improve?

>
>I heard that a FPGA wouldn't beat a DSP in floating point calculation,
>what do you think about that?  Still, I know that an implementation on
>a FPGA will require a lot of work to translate the code anyway, so I
>suppose it wouldn't be harder to rethink it directly in fixed point.

Mostly true. FPGAs with hardware multipliers are essentially fixed-point,
but can be extended to floating-point with sufficient "skill in the art".
But if the algorithm will run fine on a fixed-point DSP, that might
improve "performance" enough, anyway.

>
>The FPGA could be used to improve only a fraction of the algorithm
>too.
>Thanks for your comments.
>
>L-C
>



Article: 129094
Subject: signal generation in VHDL on FPGA.... Check my code please
From: rossalbi <rossalbi@hotmail.com>
Date: Thu, 14 Feb 2008 04:50:13 -0800 (PST)
Links: << >>  << T >>  << A >>
hi, i have writen a pice of code which should impliment a value on
the
LEDs of my FPGA development board as the signal 'count' increases.
However it is going strait to the ' when others => LEDs <=
"00000000";
' value.

any help would be much appreciated...


    -- Declare signals
    signal CLK : std_logic;
    signal RST : std_logic;
    signal Count : std_logic_vector(21 downto 0);
    signal LEDs : std_logic_vector(7 downto 0);
    signal LEDVal : std_logic_vector(7 downto 0);
    signal Dir : std_logic;


begin


    -- Tie unused signals
    User_Signals <= "ZZZZZZZZ";
    IO_CLK_N <= 'Z';
    IO_CLK_P <= 'Z';
    IO <= (0=>LEDs(0), 1=>LEDs(3), 41=>LEDs(4), 42=>LEDs(1),
43=>LEDs(4),
                          44=>LEDs(5), 45=>LEDs(2), 46=>LEDs(7),
others => 'Z');


    -- Clock divider
    process (CLK, RST)
    begin
        if (RST='1') then
            Count <= (others=>'0');
        elsif (CLK'event and CLK='1') then
            Count <= Count + 1;
        end if;
    end process;


    process (CLK, RST)
    begin


                  case Count is
                when "0000000000000000000000" => LEDs <= "00000001";
                when "0000000000000000000001" => LEDs <= "00000010";
                when "0000000000000000000010" => LEDs <= "00000100";
                when "0000000000000000000011" => LEDs <= "00001000";
                when "0000000000000000000100" => LEDs <= "00001001";
                                                            .
                                                            .
                when "0000000000000001111110" => LEDs <= "01000000";
                when "0000000000000001111111" => LEDs <= "10000000";


                when others => LEDs <= "00000000";
                  end case;


    end process;



Article: 129095
Subject: Re: setup time not met in Quartus
From: KJ <kkjennings@sbcglobal.net>
Date: Thu, 14 Feb 2008 05:04:10 -0800 (PST)
Links: << >>  << T >>  << A >>
On Feb 13, 11:33 am, "C-M, Chang" <cmchan...@gmail.com> wrote:
> On 2$B7n(B13$BF|(B, $B2<8a(B9$B;~(B42$BJ,(B, KJ <kkjenni...@sbcglobal.net> wrote:
>
> Thanks KJ for anser and told my question clear.
> But unfortunately again, my boss deny to change design (coding from
> somewhere).
> In my project, I think that ck125 and ck25 are set as global clock
> network.  I guess a global clock network need 4ns delay, so my ck125
> has 4ns delay, and ck25 has 8ns delay. (4ns from ck125 and 4ns itself
> global clock network).
> Does Quartus have any option let ck25 just has 4ns delay.
> Or can i use something like:
> set_instance_assignment -name MAX_DELAY 0.5ns -from regB -to regA, (i
> think this no work)
> or
> set max_clock_arrival_skew ,min_data_arrival_skew or
> SETUP_RELATIONSHIP
> can someone told me in the situatuin which choise is better, and is
> right,
> Because i can not change code.
> Thanks.
>

Just to make things worse, I think you might also only be looking at
half of the problem.  If you tell Quartus to also a fast timing model
analysis you might find that you have hold time issues on some of the
clock domain crossings.  But even if you don't have any now, if you
were to find some way to control the delay of ck25 to get rid of the
setup problem then it is even more likely that you'd see the hold time
problem.

Bottom line....there is no way to generate a clock signal as the
output of a flip flop clocked by some other clock and not create skew
between those two clocks.  That is the root of the problem that is
showing up for you as setup issues (and possibly hold time issues as
well).  It's not a Quartus issue, or even an FPGA issue, it's nothing
that you can add some magic constraint to engineer a fix.  That's why
there are PLLs/DLLs inside the more modern FPGAs because they can
generate derived clocks from a source clock in a completely different
manner and by doing so, they can precisely manage the phase
relationship of the source and derived clocks.

On a more positive note, it's possible that the original designers
know all about the clock domain crossings and designed things such
that when ck125 is generating a signal that is to be sampled by ck25
that they do not allow that signal to change around the time that the
rising edge of ck25 is going to occur.  Also any signal that gets
clocked by ck25 and then gets sampled by ck125 does not get sampled
around the time that ck25's rising edge has just occurred.

If that's the case for each and every signal, then these clock domain
transfers are being handled properly by design and all you have to do
to get rid of the setup (and or hold) violations is to specify a multi-
cycle constraint on those signals.  Although a multi-cycle constraint
'could' be used by the synthesis engine to possibly optomize
something, what it is mostly doing is saying that these signals are
allowed to take more than one clock cycle to settle out so when
performing the timing analysis the 8 ns clock period of ck125 is
relaxed to say 16 ns or 24 ns (depending on what value you set
multi_cycle to to correspond to 2 or 3 clock cycles).

The problem is that it is far too easy to convince yourself that
something will be stable for more than one clock cycle and not
changing when you don't want it to be when in fact it isn't.  Adding
the multi_cycle constraint in some sense is simply saying that you
don't want to be bothered with reports of supposed violations, the
danger being that you analyzed it incorrectly.  To analyze it though,
you need to do a thorough simulation and inspection/understanding of
the code to see if these assumptions are correct...and then re-run and
re-check if the design changes in any way because that change may be
done in such a way that it now violates the required multi cycle
assumption.

Lastly, if none of the above applies and the boss insists on no design
changes then there isn't much of anything you can do other than change
the random number seed, cross your fingers and hope....that isn't
engineering and it won't be a robust solution but you might get lucky
because luck will be your only ally.

Kevin Jennings

Article: 129096
Subject: Re: Virtex-4 input pad failures
From: AugustoEinsfeldt <aee@terra.com.br>
Date: Thu, 14 Feb 2008 05:22:40 -0800 (PST)
Links: << >>  << T >>  << A >>
Hi Rob,
Like Symon had quoted you have to provide more information, like:
- Are the affected pins nearby to 2.5V pins or traces? I have seen
board short circuit that happens only after assembly process due to
thermal or mechanical stress.
- Are you sure it is not related to board assembling process?
- Did you extracted the Phy or FPGA to know which device is causing
the short to 2.5V?
- If the short doesn't happen in the moment the board is energyzed how
long it takes to happen? Does the board/device increase in temperature
before it happens?
- Freezing the board changes the behavior?
- Only when all possible board related causes can be discarded then
the stress could be reason and only then the talk about overshoot or
silicon behavior should arise.

Good luck,
-Augusto

Article: 129097
Subject: Cyclone flash configuration data
From: Frank Buss <fb@frank-buss.de>
Date: Thu, 14 Feb 2008 14:23:25 +0100
Links: << >>  << T >>  << A >>
I've implemented a flash programmer for the EPCS1 serial flash, attached to
a Cyclone, using the Serial Flash Loader entity. Programming and reading
the flash works, but how do I create the data for programming it? I've
programmed it with a Byte Blaster and a JIC file from within Quartus and
comparing the content which was flashed with a RBF file, it looks like
every byte is just mirrored (from LSB to MSB instead of MSB to LSB) and
some bits at the beginning are different. Is it safe to use this RBF file
for flashing the flash?

Another question: I would like to compare the flash content with some
external file whenever the system starts, if an update is needed. Would be
good, if I could just read some checksum or date signature inside the flash
(currently I'm writing my own checksum at the last block of the flash). Is
there a documentation of the RBF file format?

-- 
Frank Buss, fb@frank-buss.de
http://www.frank-buss.de, http://www.it4-systems.de

Article: 129098
Subject: Re: signal generation in VHDL on FPGA.... Check my code please
From: Brian Drummond <brian_drummond@btconnect.com>
Date: Thu, 14 Feb 2008 13:25:54 +0000
Links: << >>  << T >>  << A >>
On Thu, 14 Feb 2008 04:50:13 -0800 (PST), rossalbi <rossalbi@hotmail.com> wrote:

>hi, i have writen a pice of code which should impliment a value on
>the
>LEDs of my FPGA development board as the signal 'count' increases.
>However it is going strait to the ' when others => LEDs <=
>"00000000";

How do you know it is going straight to that state? 

You  don't tell us what the CLK signal is ... 1 Hz? 80 MHz? A push button?

- Brian


Article: 129099
Subject: Re: Is a FPGA the solution ?
From: Martin Thompson <martin.j.thompson@trw.com>
Date: Thu, 14 Feb 2008 13:36:38 +0000
Links: << >>  << T >>  << A >>
L-C <Louis-Charles.Caron@usherbrooke.ca> writes:

> I would like to know if a FPGA is suited for this application: sound
> sources localization and separation.  The algorithm comes in 3 steps:
> sources detection using a steered beamformer (in frequency-domain), a
> particle filter to track the sources and a combination of a geometric
> source separation algorithm and a non-linear post-filter for source
> separation.
>
> Also, the audio stream comes from an array of 8 microphones, and all
> possible microphone pairs must be considered during the detection
> phase.
>
> This algorithm is already implemented of a floating point DSP and I
> would like to know if a FPGA could be a good choice to improve
> performance.

In that case, profile your current implementatio.  Find out which
piece of code is the causing you most problems.  Focus on that, you
might find it amenable to FPGA implementation (but don't discount the
overhead of moving data from your DSP to the FPGA).  I wouldn't
attempt to move the whole thing into hardware, although you could put
the bits that remain software into an FPGA processor, in which case
the "acceleration " hardware can be much more closely coupled with the
SW.

>
> I heard that a FPGA wouldn't beat a DSP in floating point calculation,
> what do you think about that?  

Depends - a big FPGA can do a *lot* of FP in parallel, if your
algorithms are amenable.

HTH,

Cheers,
Martin

-- 
martin.j.thompson@trw.com 
TRW Conekt - Consultancy in Engineering, Knowledge and Technology
http://www.conekt.net/electronics.html



Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarApr2017

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search