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 225

Article: 225
Subject: (fwd) Postings sent as mail ???--------------
From: gt0361b@prism.gatech.edu (Benjamin Gene Cheung)
Date: 28 Sep 1994 00:23:14 -0500
Links: << >>  << T >>  << A >>

Article: 226
Subject: What do think about the Intel Flexlogic8160?
From: gt0361b@prism.gatech.edu (Benjamin Gene Cheung)
Date: 28 Sep 1994 01:24:48 -0400
Links: << >>  << T >>  << A >>
Has anyone sampled the ifx8160 from intel?  What do think of it?


-- 
Benjamin Gene Cheung
Computer Engineer
Georgia Institute of Technology
Internet: gt0361b@prism.gatech.edu


Article: 227
Subject: Re: Xilinx 4000
From: bobe@soul.tv.tek.com (Bob Elkind)
Date: 28 Sep 1994 17:17:25 GMT
Links: << >>  << T >>  << A >>
fliptron@netcom.com (Philip Freidin) writes:

     ...  much good info deleted, Phil obviously knows X4K well ...

:Here are some excerpts of Bob's mail with my comments.
:
:>2.  Don't use the -4 (fastest) speed grade.  Design to the -5 speed
:>    grade (or slower), and then specify -4 speed grade (or whatever
:>    the next fastest speed grade is) parts for production.
:
:This seems a little over precautious / pessamistic. The Xilinx
:speed files seem to be weighted to the pessamistic side already
:so designs that meet timing specs acording to XDELAY have always
:worked for me.

I stand by what I said in my original posting.  I've discussed the issues
involved with Phil, and perhaps his perspective may have shifted a bit,
but I don't want to put words in his mouth.

I get paid to design products that make money for our company, on schedule
and meeting design specs (including reliability and performance), and
with best possible manufacturing cost.  At this point in time, I would
*not* design for anything more aggressive than -5 performance, and I *would*
make sure that the FPGAs used in production were one speed grade faster
than my design criteria.

There is nothing more I can add to this subject.  Do as you see fit.

:>    I don't think the BUFGS/BUFGP delay differences are well
:>    characterised; they certainly aren't described well in the data books.
:
:The BUFGPs are about 1 nS faster than the BUFGSs.
:
:>    You will probably be happier if you stick to the BUFGPs for all
:>    clock distribution where clock skew is a concern.
:
:The clock skew for both is < 1 nS. I have measured them both and seen
:skews in the 300 to 500 pS range. Worst case skew is between an IO FF
:in the middle on the top or bottom edge , and a corner IO FF
:
:>  There are some at
:>    Xilinx who would debate this contention.  I have personal reasons
:>    for this conviction which relate to using the on-chip RAM.
:
:I have seen differences in the BUFGP to BUFGS skew too, but it is
:of the order of a few 100 pS, and less than 1 nS

I believe that there are differences in drive between the two types of
clock buffers, and resulting differences in rise/fall time which may
affect absolute timing over temperature, etc.  I can't prove this, however.

:>7.  Be careful about using both clock edges with tight timing tolerances.
:>    The clock buffers have TTL thresholds and asymmetric rise/fall times
:>    internally.  So 50% duty cycle clock waveforms externally don't
:>    necessarily translate into 50% duty cycle square waves internally.
:>    If you clock on both clock edges, you probably should provide at least
:>    4 nS of timing margin.
:
:The I/O input buffers have TTL thesholds (~ 1.5V), so a 5V square wave
:of 50% duty cycle will end up being seen as more clock high time and
:less clock low time, assuming equal rise and fall times. If the rise
:and fall times are fast ( < 3 nS), then the effect is about 1 nS.
:Slower edges will have more effect. Higher frequencies will make the
:effect a higher percentage of the cycle time. Internaly the FPGAs
:have their thresholds set to minimize clock stretching and shrinkage.
:4nS seems like a good safe number for clock margin, when doing things
:on both edges. Dont assume that CLK-BAR gets to the FFs later than CLK,
:as the clock might be distributed as CLK-BAR, not CLK.

I agree with Phil, if you are using a 5V square wave for external clock
signal.  Timing variation is greater if your clock signal is a typical
TTL signal which peaks around 2.5 or 3.0 V (check the 10125 data sheet for
max VOH).  The waveshape *around* (i.e. below, at, and above) the
"reference" voltage affects the transfer characteristic of the internal
buffer.  This is real-world consideration that not all semi vendors
appreciate in their characterisation processes.

Perhaps the best advice is to make your TTL clocks as sharp and
"full-swing" as possible.  External clock buffers will add skew,
however, but you get the idea.

:>8.  Don't be lulled by a batch of Xilinx FPGAs with excellent timing
:>    margins.  All FPGA vendors, including Xilinx, are aggressively pursuing
:>    smaller and smaller fab processes.  As the distribution of their
:>    parts with respect to timing performance changes, and as their order
:>    demand for different speed grades shift, you may get parts which
:>    greatly exceed minimum timing specs.  The next month you may get parts
:>    which aren't quite so fast.   So don't get over-confident from the
:>    excellent yield and/or margins on the first 50 boards.
:
:ABSOLUTELY. Thats what speed files and simulation are for: Make sure
:your design will work with the slowest silicon that meets the databook
:guarantees, at high temperature, and low VCC

The value of speed files, simulators, and simulation is limited by the
quality of the characterisation on which the speed files are based, the
competency of the simulator (or timing verifier/calculator), and the
applied skill of the user of the simulator.  There are a lot of IFs in
this equation, especially considering that characterisation of *ANY* FPGA
technology is a very, very complex matter.  I would suggest being very
conservative, and weigh the consequences of mis-guessing on the over-
aggressive side.

:I've designed with 4003, 4003A, 4005, 4008, 4010, and lots of XC3K

Phil's experience shows.

May the electrical force be with you.

Bob Elkind


Article: 228
Subject: Re: Xilinx 4000
From: david@fpga.demon.co.uk (David Pashley)
Date: Wed, 28 Sep 1994 17:24:49 +0000
Links: << >>  << T >>  << A >>
In article <CwqKLu.LFL@hpqmoea.sqf.hp.com> marting@hpqtdla.sqf.hp.com writes:

"Hi All,
"       After several years of 3000 Series Xilinx work, I'm considering a 4000
"device because of it's internal RAM.
"
"I was wondering if anyone with experience of the 4000 RAM could give some hints,"
"or just some general 4000 best practices??
"
"
"If people prefer to mail direct rather than post to the newsgroup, I'll post a
"summary of what I receive.
"
"                        Thanks
"
"                                Martin Curran-Gray
"
There's quite a lot to say in response to this. You'll find 4000 
series substantially different. There are more implementational 
details to take care of if you want to achieve decent clock speeds.

However, one thing I would strongly recommend is that you take a 
look at the AT&T ORCA devices, especially the 2CXX series. This 
device series offers good price/performance and has 16x4 potential 
RAM per logic block.

I'm not saying it's better than XC4000 - just that it's an 
alternative worth looking at. As always, it's your design which 
dictates which is the best device to use!

David Pashley
Direct Insight Ltd
The independent CPLD/FPGA experts
Tel: +44 280 700262
Fax: +44 280 700577



Article: 229
Subject: Re: Software costs (was Re: Lattice ISP software)
From: bbutler@netcom.com (Bryan Butler)
Date: Wed, 28 Sep 1994 19:56:51 GMT
Links: << >>  << T >>  << A >>
Richard Vireday (rvireday@pldote.intel.com) wrote:
> In article G26@netcom.com, bbutler@netcom.com (Bryan Butler) writes:
> >I always thought it was counter-productive to charge for software that is only
> >good for one vendors chips

> Well Bryan, when it can cost any company over 1 million dollars
> to get the software written for their hottest new device, you had better 
> do your part and buy a lot of chips to amoritise the cost.

> If a typical PLD/FPGA sells only 10,000 units in the first year
> of introduction (this is an industry average, ref. Dataquest), then you can tell
> your suppliers to tack on an extra $20 onto each device you buy!
> (Just assume there are four other people out there buying it too! :-)

> --Richard Vireday
> Formerly of Intels PLD & FPGA Business Unit 

So you used to work for Intel? One of the main reasons I started using Intel
FPGAs is because I could get the software for free. If I had to buy the
software, I probably wouldn't use them. (Yes I know they sold the FPGA line
to Altera. I hope Altera keeps up the free policy).

One thing you may be missing is that small sales of FPGAs now may result
in larger sales later. I work for a consulting company. I will choose and
FPGA based on the tools I have available. I probably won't purchase many
FPGAs, but my customers will. The customer probably won't make much use
of the design tools; he'll probably just program his parts with the JEDEC
file I give him. 

So, as the design engineer, I'm primarily concerned with the cost of 
developing the prototype, which means that I can't spend a lot of money
on design tools, particularly if I'm not sure whether the FPGA for which
the tools are targeted will even work in the design.

A customer might spring for a new design tool if it means a reduction in 
manufacturing costs. However, this tool now belongs to the customer, so I
won't be able to use it for future designs for other customers.

I would speculate (with no real evidence) that the major market for FPGAs
is for low volume production, prototype, or proof-of-concept research
projects. Large volume production would probably favor ASICs, with higher
NRE but lower per device costs. Thus, there is a relatively large number
of copies of the design software (desired) compared to the number of devices
sold. By selling the software for a low (preferably $0) cost means more
of these small-time users will try your chips, thus more chip sales.

Obviously, the cost of developing the software has to be recovered, which
means higher device costs. I would not balk too much at paying $10-20 more
for a device with free software. If I need a handful of devices, this cost
is negligible. If the alternative is to try to use a less appropriate device,
I might spend more time trying to make the design fit that the small savings
is worth.

For the production customer, a solution could be to offer steeper price breaks
for volume purchase. The price breaks could even be cummulative, so if I 
buy 100, then 100 more, I can get the quantity 200 price break on the second
purchase.

--
-------
Bryan Butler
bbutler@netcom.com


Article: 230
Subject: Re: Lattice ISP software: really good
From: tomv@arc.umn.edu (Tom Varghese)
Date: Thu, 29 Sep 1994 00:14:44 GMT
Links: << >>  << T >>  << A >>
mcgaugh@eng.iac.honeywell.com (Paul McGaugh) writes:

>: Not true, auto partitioning is included with PDS+

Exactly what is pDS+? I am using pDS, and have never heard of pDS+ from lattice

Tom Varghese
tomv@s1.arc.umn.edu




Article: 231
Subject: Re: Exemplar CORE experiences?
From: dany@netcom.com (Danny)
Date: Thu, 29 Sep 1994 01:53:15 GMT
Links: << >>  << T >>  << A >>
In article <CwqpnH.4qG@ips.cs.tu-bs.de>,
Andreas Koch <koch@eis.cs.tu-bs.de> wrote:
>We would be interested in hearing about any experiences with
>Exemplar's CORE/TopDown+ synthesis tool. It seems somewhat
>less hardware-hungry than the Synopsys suite. While our primary
>target would be Xilinx 4000 FPGAs, comments regarding other
>technologies (eg. standard-cell ASICs like ES2) would be
>useful.
>
>Thanks,
>	Andreas Koch
>-- 
>Andreas Koch                                   Email  : koch@eis.cs.tu-bs.de
>Institut f"ur theoretische Informatik          Phone  : x49-531-391-2384
>Abteilung Entwurf Integrierter Schaltungen     Phax   : x49-531-391-5840
>Gaussstr. 11, D-38106 Braunschweig, Germany    Telex  : 95 25 26


I did work on Exemplar and Synopsys for the XC4000 series.
Hardware requirements for exemplar are a tad less, but 
the FPGA compiler was consistantly performing better than
Exemplar(20%-35%) depending on the design. The Design compiler
is comparable to Exemplar.
Synopsys has a lot of options and tweaks to mess around with, to 
tune you design and optimize it. Exemplar is very limited.
Just my experience.

Danny.



Article: 232
Subject: Re: Lattice ISP software: really good
From: trev@ss11.wg.icl.co.uk (Trevor Hall)
Date: Thu, 29 Sep 1994 05:33:27 GMT
Links: << >>  << T >>  << A >>
tomv@arc.umn.edu (Tom Varghese) writes :-
>Exactly what is pDS+? I am using pDS, and have never heard of pDS+ from lattice

Try pages 6-15 onwards of the 94 databook.

Although I have never used PDS, I think the main difference is that PDS+ supports
design entry from other tools ;-
ABEL
Viewlogic schematic
ISDATA LOG/iC
VHDL
MINC (soon)
and probably others I have failed to mention.


Looking at the PDS data sheets it look a little crude wrt design entry (boolean only)
when compared to the likes of ABEL/VHDL/MINC etc..



T.Hall
Goldrush Project
ICL









Article: 233
Subject: Re: XC1765DPD8C
From: phill@dd.eng.bbc.co.uk (Phil Layton)
Date: Thu, 29 Sep 1994 13:10:11 GMT
Links: << >>  << T >>  << A >>
David Pashley (david@fpga.demon.co.uk) wrote:

: I think you are being led astray by whoever told you that you need 
: to use a jedec file. The jedec format is only applicable to 
: programmable logic devices, whereas the XC1765 is an EEPROM device. 
: Therefore (and certainly for all the leading programmer types), you 
: can present the data to the programmer an any supported PROM type 
: format, but definitely *not* JEDEC.

: Some older programmers have trouble with this particular device as 
: the susceptibility to clock pin noise is somewhat higher than the 
: norm.

: David Pashley
: Direct Insight Ltd
: The independent CPLD/FPGA experts
: Tel: +44 280 700262
: Fax: +44 280 700577

Unfortuanately there are programmers around - Stag 3000 in particular
- that for some unknown reason treat the Xilinx serial proms as PLDs 
and therefore insist on Jedec format.  Stag in the past have been prepared
to give out conversion utilities to customers who complain or it is fairly
straight forward to write your own.

--

 ******************************************************************************
 *         Phil Layton, Senior R & D Engineer                                 *
 *         BBC Research & Development Department                              *
 *         Kingswood Warren, Tadworth, Surrey KT20 6NP, UK                    *
 *                                                                            *
 * internet: phill@rd.bbc.co.uk                                               *
 * phone: 0737 836575 (+44 737 836575) fax: 0737 832336 (+44 737 832336)      *
 ******************************************************************************


Article: 234
Subject: Re: Xilinx 4000
From: ian@neocad.com (Ian McEwen)
Date: Thu, 29 Sep 1994 17:25:24 GMT
Links: << >>  << T >>  << A >>
David Pashley (david@fpga.demon.co.uk) wrote:
: ...
: However, one thing I would strongly recommend is that you take a 
: look at the AT&T ORCA devices, especially the 2CXX series. This 
: device series offers good price/performance and has 16x4 potential 
: RAM per logic block.

I echo this recommendation.  In my work I target a number of designs
to a variety of FPGA device architectures, including XC4000 and Orca.
The att2c family is excellent, from a number of standpoints.  First,
it's very easy to place and route, and as a result, the amount of
processor time spent in the place and route tools is significantly
reduced.  All of the Orca devices can handle very high utilization -- 
in fact, I have not yet seen an Orca design which we couldn't get to
route, even at 100% PFU utilization.  High TBUF usage will have the
biggest impact on routability, but will still be easier to route than
in an equivalent XC4000.  You won't sacrifice speed for this ease
of routing, either; in my experience designs will run at least as fast
in an Orca part as they will in Xilinx chips.  These are real customer
designs, too, not contrived test cases or PREPCO designs.

Also, if you're working with very large designs, you'll want to look
at the att2c15.  Every XC4013 we retargeted to the att2c15 fit and
routed without effort; conversely, there's no way a packed att2c15
design would fit (let alone route) in a XC4013.  In fact, in most
cases an att2c12 will handle a design which routed in a XC4013.

: I'm not saying it's better than XC4000 - just that it's an 
: alternative worth looking at. As always, it's your design which 
: dictates which is the best device to use!

Agreed, every architecture has its strong points.  But if you're
evaluating different FPGAs for a project and aren't tied to any
one vendor, definitely check out the Orca devices.

/////////////////////////////////////////////////////////////////////////
Ian McEwen              Opinions expressed here are a product of my
NeoCAD, Inc.            heredity and environment, and should not be 
2585 Central Ave.       considered the same as those of my employer.
Boulder, CO 80301

Internet:  ian@neocad.com
Phone:  (303)442-9121	FAX:  (303)442-9124
/////////////////////////////////////////////////////////////////////////


Article: 235
Subject: Re: What do think about the Intel Flexlogic8160?
From: bbutler@netcom.com (Bryan Butler)
Date: Thu, 29 Sep 1994 20:11:40 GMT
Links: << >>  << T >>  << A >>
Benjamin Gene Cheung (gt0361b@prism.gatech.edu) wrote:
> Has anyone sampled the ifx8160 from intel?  What do think of it?

Were they ever made available by Intel? Just before Intel sold to Altera
I was told they weren't available yet. Is Altera going to bring it out?


--
-------
Bryan Butler
bbutler@netcom.com


Article: 236
Subject: PhysComp 94 -- Advance Program
From: bhanu@seas.smu.edu (Bhanu Kapoor)
Date: Fri, 30 Sep 1994 00:18:10 GMT
Links: << >>  << T >>  << A >>

		Registration Information and Advance Program
	      Workshop on Physics and Computation, PhysComp '94
			   This Decade and Beyond
		    November 17 - 20, 1994, Dallas, Texas
		  Sponsored by Dallas IEEE Computer Society
			    Sponsored by ONR/ARPA
	      Corporate Sponsor: Texas Instruments Incorporated

Enclosed is  the registration  information for the  Workshop  on  Physics and
Computation, PhysComp '94,  to be held  in Dallas, TX,  November 17-20, 1994.
See  the    enclosed  preliminary  schedule  of  speakers.     The conference
registration information  must be sent in   by Monday Oct   17 to qualify for
early   registration  (see   registration).  Please   make    your  own hotel
reservations directly with the hotel by Tuesday November 1 (see hotel).

			    WORKSHOP REGISTRATION
			    ---------------------

To  confirm your plans to  attend the workshop, please  fill out the enclosed
registration form, and send it to one of the following addresses.

 date:  Early registration deadline is Monday October 17

 email: matzke@hc.ti.com
  
 mail:  Doug Matzke
        Texas Instruments Incorporated
        PO Box 655474, MS 446
        Dallas, TX 75265

 fax:   Doug Matzke @ (214) 995-6194

The early registration fee  will be $180,  if  registration is made and  paid
before Monday Oct 17.  We  can accept payment  for early registration ONLY BY
CHECK drawn from US banks or  US money order payable  to Dallas IEEE Computer
Society. After that  date the registration fee  is $200. We  will accept both
checks and Visa/Master Card at on site registration.  To aid in our planning,
PLEASE  REGISTER IN ADVANCE even if  you plan to pay your  fee at the on site
registration.

A registration table  will be set up  at the Wednesday evening reception, and
at the Thursday and Friday morning breakfast.  Please stop by to register and
to  pick up your   workshop packet, including  your copy  of the proceedings.
Additional  PhysComp 94 (and   PhysComp  92) proceedings can be   preordered.
Information about the location of the  reception and other workshop functions
will be available at the hotel desk at check-in.

		      HOTEL REGISTRATION/ACCOMMODATIONS
		      ---------------------------------

The Workshop will be  held at The  Harvey Hotel in  Addison, Texas.   We have
reserved a block of rooms at a discounted per night rate  of US$59 for single
or US$69 for double occupancy.   Please make your hotel reservation  directly
with The Harvey Hotel before Tuesday November 1 to assure accommodations.

 date:  Tuesday November 1

 mail:  The Harvey Hotel
	14315 Midway Road
	Addison, Texas 75244

 phone: (800) 922-9222 or (214) 980-8877

**WARNING: PLEASE REGISTER EARLY**.  Because  of attendance uncertainty, only
125 hotel rooms have been reserved at the very low PhysComp 94 rate.  If this
block of rooms is filled, then the Harvey Hotel will help you  to book a room
in the adjacent hotel at the HIGHER RATE of $89 per night. So register early.

Please  mention  the  "PHYSCOMP  or IEEE  or Texas  Instruments Workshop"  to
receive the discounted rate.  State your preference for a smoking/non-smoking
room when making  your reservation.   NO  SMOKING WILL  BE PERMITTED  at  any
workshop  functions.  If  you   have any other  special  requirements, please
indicate on the registration form.

Hotel facilities include a pool, an outdoor spa, and a weight room.  There is
a full service restaurant, a  50's style diner, and a  lobby bar.  The  hotel
also has  complimentary van  service available  for  transportation within  a
three   (3) mile radius   for restaurants,   shopping, etc.  Gratuities   are
accepted.

Roommate requests should be handled by  checking the box on registration form
and including email and phone numbers. At the end of each day, I will email a
copy of all people requesting roommates to all people on that email list, and
you  can  make your   own  arrangements.   You   and your roommate   are then
responsible for making your   own hotel reservations.   Please inform  me and
people on the roommate list to remove your name from the list.  Please do not
place roommate notices on the regular PhysComp email list.

				    MEALS
				    -----

The following meals  are included  with  your registration fee:   continental
breakfast on Thursday thru Sunday, lunch  on Thursday, Friday and Saturday, a
reception  on Wednesday evening, a  reception and banquet  dinner on Saturday
evening, and refreshments at breaks. Cash bar is available at receptions.

If you  have special dietary   needs (Vegetarian, Kosher, Low  Sodium, etc.),
please specify your requirements on the registration form.
          
			       TRANSPORTATION
			       --------------

The   Harvey House is within  easy  reach  of both  Dallas   Fort Worth (DFW)
International Airport  and Love Field  regional airport (service only to/from
states adjoining Texas).

Rental Cars:  All   of the major car   rental  companies  serve  both  Dallas
airports, although only Avis, National, Hertz, and Budget are on-site at DFW.
Ask for directions to The Harvey Hotel/Addison at the rental desk.

Airport/Hotel  Shuttles:  Super  Shuttle  provides van  shuttles  to/from the
Harvey Hotel.   The vans are  available outside  the terminals  on the  lower
level at DFW, and in front of the main  terminal at Love Field.  Phone number
(toll free in DFW area) is:

  Super Shuttle: (817) 329-2000; to/from DFW $16.00 per person; to/from Love
  Field $14.00 per person. Visa/MC/AMEX/Diners Club accepted.

Taxis: Taxis are available outside  each terminal on  the upper level at DFW,
and in front of the main terminal at Love Field.  The approximate fare to the
hotel $30 per person from/to DFW, and $25 per person from/to Love Field.

Estimated distance to the hotel from DFW is approximately 12-15 miles.  There
is free parking around the hotel and a covered parking garage at the hotel.

			   QUESTIONS OR ASSISTANCE
			   -----------------------
Please feel free to contact me if you need any assistance with arrangements.

Douglas Matzke                                  Net: matzke@hc.ti.com
Texas Instruments                               Tel: (214) 995-0787
PO Box 655474, MS 446                           Fax: (214) 995-6194
Dallas, TX 75265   				MSG: MTZK

Email is the best way to reach me. If you call, and I'm not at my desk, please
leave a message on my answering machine, and I'll call you back.
---------------------------------- cut here ----------------------------------

			      REGISTRATION FORM
	      WORKSHOP ON PHYSICS AND COMPUTATION, PHYSCOMP 94

NAME/TITLE:___________________________________________________________________

AFFILIATION:__________________________________________________________________

MAIL ADDRESS:_________________________________________________________________

MAIL CITY/STATE:______________________________________________________________

TELEPHONE:_____________________________FAX:___________________________________

EMAIL ADDRESS:________________________________________________________________

TOPICS OF INTEREST:___________________________________________________________

ARRIVAL DATE/TIME:____________________________________________________________

DEPARTURE DATE/TIME:__________________________________________________________

I PLAN TO ATTEND THE WORKSHOP:  YES____  NO____

I PLAN TO PRESENT MY PAPER AS SCHEDULED BELOW:  YES____  NO____

I PLAN TO ATTEND THE RECEPTION WEDNESDAY EVENING:  YES____  NO____

I PLAN TO STAY AT THE HARVEY HOTEL ADDSION:  YES____  NO____

ADD MY NAME TO ROOMMATE LIST FOR THE HARVEY HOTEL ADDSION:  YES____  NO____

SPECIAL DIETARY REQUIREMENTS:_________________________________________________

OTHER SPECIAL REQUIREMENTS:___________________________________________________

*****************************PrePayment Information*************************

Early Registration fee of $180 if paid with check before Oct 17.____________
or
Normal fee of $200 if paid after Oct 17 or using credit card.   ____________
add
Number of additional PhysComp 94 Banquet Tickets at $25.00 each ____________
add
Number of additional PhysComp 94 Proceedings at $25.00 each     ____________
add
Number of previous PhysComp 92 Proceedings at $35.00 each       ____________

                                                 TOTAL FEE DUE: ____________

                                     CHECK ENCLOSED FOR AMOUNT: ____________

I WILL PAY NORMAL FEE OF $200 AT THE WORKSHOP USING CREDIT CARD INFO BELOW:

CREDIT CARD NUMBER:___________________________________________________

CREDIT CARD TYPE: _________________  EXPIRATION DATE:_________________

             (receipts will be included in conference packets)
	 (all early registrations will be confirmed within 2-3 days)
---------------------------------- cut here ----------------------------------

			     Preliminary Program
		Workshop Physics and Computation, PhysComp 94
			    Harvey Hotel-Addison

====================================================================
====================  Wednesday pm, Nov 16, 1994 ===================

 6:00 -  9:00  Registration, Reception, and cash bar

====================================================================
====================  Thursday am, Nov 17, 1994  ===================

 7:30 -  8:30  --- Continental Breakfast ---

Session 1:     The Technologist's Perspective on Nanoelectronics
----------     Chair: Gary Frazier

 8:30 -  9:00  Integrated Circuits, Nanoelectronics, and
               21st Century Electronic Systems
               Bob Bate (Texas A&M University)

 9:00 -  9:30  The Life Expectancy of CMOS Technology
               Bob Doering (Texas Instruments)

 9:30 - 10:00  Research Toward Nanoelectronic Computing
               Technologies in Japan
               Rick Kiehl (Fujitsu)

10:00 - 10:30  --- BREAK ---

Session 2:     Computing With Quantum Devices
----------     Chair: Gary Frazier

10:30 - 11:00  Resonant Tunneling Quantum Devices and Circuits
               Alan Seabaugh (Texas Instruments)

11:00 - 11:30  Quantum Cellular Automata: The Physics os Computing
               with Arrays of Quantum Dot Molecules
               Craig S. Lent, P. Douglas Tougaw,
               and Wolfgang Porod (Univ. Notre Dame)

11:30 - 12:00  Results on Two-Bit Gate Design for Quantum Computers
               David DiVincenzo (IBM)

12:00 -  1:30  --- LUNCH ---

====================================================================
====================  Thursday pm  =================================

Session 3:     Architecture Issues in Nanoelectronics
----------     Chair: Wolfgang Porod

 1:30 -  2:00  Horizons of Parallel Computation
               Gianfranco Bilardi (University of Padova)

 2:00 -  2:30  Multiprocessor Architectures and Physical Law
               Paul Vitanyi (CWI / Univ. of Amsterdam)

 2:30 -  3:00  Algebras and Architectures for Nanoelectronics
               Gary Frazier (Texas Instruments)

 3:00 -  3:30  --- BREAK ---

Session 4:     Architecture Issues for Computation
----------     Chair: Doug Matzke

 3:30 -  4:00  The Latest in Adiabatic Computing
               John Denker (AT&T)

 4:00 -  4:30  Impact of Locality and Dimensionality Limits
               on Architecture Trends
               Doug Matzke (Texas Instruments)

 4:30 -  5:00  Space, Time, Logic, and Things
               Dick Shoup (Interval Research)

 5:00 -  5:30  Space and Time in Computation, Topology
               and Discrete Physics
               Louis H. Kauffman (Univ. Illinois at Chicago)

====================================================================
====================  FRIDAY am, Nov 18, 1994  =====================

 8:00 -  9:00  --- Continental Breakfast ---

Session 5:     KEYNOTE ADDRESS      
----------     General Chair: Doug Matzke

 9:00 - 10:00  Computation in Analog and Digital Physical Systems
               Carver Mead  (Cal Tech)

10:00 - 10:30  --- BREAK ---

Session 6:     QUANTUM COMPUTERS
----------     Chair: Bill Frensley

10:00 - 10:30  Is Quantum Mechanics Useful?
               Rolf Landauer  (IBM)

10:30 - 11:00  The Stabilisation of Quantum Computations
               Andre Berthiaume (Univ. Montreal), David Deutsch 
               (Univ. Oxford), and Richard Jozsa (Univ. Plymouth)

11:00 - 11:20  Can Quantum Computers Have Simple Hamiltonians?
               Michael Biafore  (MIT)

11:20 - 11:40  Quantum Oblivious Transfer Is Secure Against All
               Individual Measurements
               Dominic Mayers and Louis Salvail (Univ. Montreal)

11:40 -  1:30  --- LUNCH ---

====================================================================
====================  FRIDAY pm  ===================================

Session 7:     QUANTUM COMPUTATION
----------     Chair: Wolfgang Porod

 1:30 - 2:15   A Fast Algorithm for Factoring on Quantum Computers
               Peter Shor (AT&T)

 2:15 - 3:00   Decoherence and Quantum Computers- A Problem
               Bill Unruh  (Univ. B. C., Vancouver)

 3:00 - 3:30   --- BREAK ---

Session 8:     PHYSICS as COMBINATORIAL COMPUTATION
----------     Chair: Riley Jackson

 3:30 - 4:00   Physical Parallelism and Computation
               Keith Bowden (Univ. East London)

 4:00 - 4:30   Bit-String Physics: A Novel "Theory of Everything"
               H. Pierre Noyes  (Stanford Univ.)

 4:30 - 5:00   Toward an Information Mechanics
               Michael Manthey  (Aalborg Univ.)

====================================================================
====================  SATURDAY am, Nov 19, 1994 ====================

 7:30 -  8:30  --- Continental Breakfast ---

Session 9:     REVERSIBLE LOGIC
----------     Chair: Paul Vitanyi

 8:30 -  9:00  Reversible Logic Issues in Adiabatic CMOS
               Bill Athas and Lars Svensson  (USC)

 9:00 -  9:20  Thermal Logic Circuits
               J. G. Koller, W. C. Athas, and L. J. Svensson (USC)

 9:20 -  9:40  A Reversible Instruction Set Architecture and Algorithms
               J. Storrs Hall  (Rutgers Univ.)

 9:40 - 10:00  Encoded Arithmetic for Reversible Logic
               Akhilesh Tyagi  (Iowa State Univ.)

10:00 - 10:30  --- BREAK ---

Session 10:    CELLULAR AUTOMATA & REVERSIBLE CA
-----------    Chair: Andrew Ilachinski

10:30 - 10:50  Some Results on Invertible Cellular Automata
               Andrea Clementi, Patrizia Mentrasti (Univ. Roma), 
               and Pierluigi Pierini (MIT)

10:50 - 11:10  On the Average-Case Complexity of the Reversibility
               Problem for Finite Cellular Automata
               Andrea Clementi (Univ. Roma), Pierluigi Pierini (MIT), 
               and Russell Impagliazzo (UC San Diego)

11:10 - 11:30  Necessary and Sufficient Conditions for Reversibility in
               One Dimensional Cellular Automata
               Jose Alberto Baptista Tome (INESC Lisbon)

11:30 - 12:00  Coupling Computations Through Space
               Pedro P. B. de Oliveira (Nat'l Inst. Space Res.,  Brazil)

12:00 -  1:30  --- LUNCH ---

====================================================================
=========================  SATURDAY pm  ============================

Session 11a:   QUANTUM COMPUTATION 
------------   Chair: Bill Frensley

 1:30 - 1:50   Quantum Waveguide Structures and Devices
               Stephen M. Goodnick, A. Weisshaar, A. Ecker,
               and V. K. Tripathy (Oregon State Univ.)

 1:50 - 2:10   On a Method of Solving SAT Efficiently Using
               the Quantum Turing Machine
               Takashi Mihara and Tetsuro Nishino
               (Japan Adv. Inst. Sci. & Technol.)

 2:10 - 2:30   Chu Spaces: Automata with Quantum Aspects
               Vaughan Pratt  (Stanford Univ.)

Session 11b:   STATISTICAL MECHANICS and INFORMATION
------------   Chair: Riley Jackson

 2:30 - 2:50   Statistical Mechanics of Combinatorial Search
               Tad Hogg (Xerox PARC)

 2:50 - 3:10   Phase Transitions and Coarse-Grained Search
               Colin P. Williams and Tad Hogg (Xerox PARC)

 3:10 - 3:30   --- BREAK ---

Session 12:    ENTROPY and INFORMATION
-----------    Chair: Andrew Ilachinski

 3:30 - 4:00   The Boltzmann Entropy and Randomness Tests
               Peter Gacs  (Boston Univ.)

 4:00 - 4:20   Entropy Cost of Information
               Paul N. Fahn  (Stanford Univ.)

 4:20 - 4:40   The Complexity and Entropy of Turing Machines
               Paul A. Dufort and Charles J. Lumsden  (Univ. Toronto)

 4:40 - 5:00   A Fast Algorithm for Entropy Estimation of Grey-Level Images
               Salvatore D. Morgera and Jihad M. Hallik (McGill Univ.)

 6:30 - 9:00   --- Reception, cash bar, and Banquet  --- 
               Making waves with our troubadour
               Gilles Brassard (Universit\'e de Montr\'eal
               and Ecole Normale Sup\'erieure)

====================================================================
======================  SUNDAY am, Nov 20, 1994  ===================

 7:30 - 8:30   --- Continental Breakfast ---

Session 13:    PARALLEL COMPUTATION
-----------    Chair: Sharad Saxena

 8:30 -  9:00  Computational Spacetimes
               E. Theodore L. Omtzigt  (Intel)

 9:00 -  9:20  Evolution, Entropy, and Parallel Computation
               Kurt Thearling  (Thinking Machines)

 9:20 -  9:40  On Physical Models of Neural Computation and Their 
               Analog VLSI Implementation
               Andreas G. Andreou  (Johns Hopkins Univ.)

 9:40 - 10:00  Analog Computation with Continuous ODEs
               Michael S. Branicky  (MIT)

10:00 - 10:30  --- BREAK ---

Session 14:    Panel session on Physics and Computation
-----------    Chair: John Denker

10:30 - 12:00  Topic:  Will PhysComp make an impact? 
               
==============================================================================
==============================================================================

			   PLEASE POST AND FORWARD



Article: 237
Subject: AT&T ORCA FPGA
From: attmes@solomon.technet.sg (Koh Kim Huat)
Date: 30 Sep 1994 07:57:30 GMT
Links: << >>  << T >>  << A >>
Folks,

Look! AT&T has announced the biggest FPGA device, ORCA 2C26, and it seems 
very powerful. Any comments?

Simon.

*****************************************************************************

       AT&T MICROELECTRONICS ANNOUNCES 26,000 GATE ORCA FPGA

            AT&T Microelectronics announced the ATT2C26, a 26,000 gate device
       which is the highest density single chip FPGA to date.  Like the
       earlier introduced ATT2C15, the 2C26 is based on a half micron, three
       level metal CMOS process.

            These half micron devices deliver up to 150MHz clock rates.  Both
       the 2C15 and 2C26 meet the PCI bus specification requirement with a
       pin-to-pin, clock-to-out delay of less than 11 nanoseconds.  Using
       PREP benchmarks as a reference, the 2C26 delivers an average capacity
       of 104 instances and an average benchmark speed of 50MHz.  The average
       Fext is 40MHz.

            The 2C26 is available in two speed grades, -2 and -3, with logic
       delays of 2.5ns and 1.8ns respectively.  Offering users 384I/O's,
       36,864 bits of user RAM, and 2,304 flip-flop in the core logic cells,
       the 2C26 leads the industry as the highest density single chip FPGA.
       The various packages offered provide footprint compatibility with any
       other ORCA FPGA device so users can easily integrate more logic and
       upgrade to the 2C26.

            Compared to competitive devices using two level metal, ORCA Series
       2C devices exhibit 33 percent more routing tracks to sustain and even
       increase performance.  Additional routing resources are incorporated in
       the Series 2C architecture, permitting easier routing in highly complex
       designs, as well as higher speeds as a result of minimized interconnect
       delays.

       Price and Availability
            ORCA FPGA users can route designs in the 2C26 by utilizing the
       ORCA development system version 3.0 or the NeoCAD tool set version 6.1.
       Samples of the ATT2C26-2PS208 are currently available.  1000 unit price
       is $720.00.  The device will sell for $395.00 in volume by the middle
       of 1995.  The 2C26 is available in the following package options.

                               208-Pin Quad Flat Pack
                               240-Pin Quad Flat Pack
                               304-Pin Quad Flat Pack
                           429-Pin Ceramic Pin Grid Array




Article: 238
Subject: Re: Determine protected PAL's equations
From: david@fpga.demon.co.uk (David Pashley)
Date: Fri, 30 Sep 1994 09:16:47 +0000
Links: << >>  << T >>  << A >>
In article <9409271733.AA08796@dpi.inpe.br> celso@dpi.inpe.br writes:

"If you need to determine the equations of (or duplicate)
"protected PALs of the series 16L8, 20L8, 16R4/6/8, 
"20R4/6/8 and others, we may help you. For more information,
"contact us:
"
"C.R.Sonnenburg            or           C.L.Mendes
"sonnen@ieav.cta.br                     celso@inpe.dpi.br
"
I can think of at least one commercially available tool that does
this - and not just for simple PALs.

David Pashley
Direct Insight Ltd
The independent CPLD/FPGA experts
Tel: +44 280 700262
Fax: +44 280 700577
 


Article: 239
Subject: Area of a FPGA tile.
From: moran@die.upm.es (Javier Moran Carrera)
Date: 30 Sep 1994 06:40:55 -0500
Links: << >>  << T >>  << A >>
Hi:

I would like to compare FPGA technology against conventional VLSI approaches
(say gate-arrays, full custom, etc...). I have seen in the literature
some figures comparing the area and speed of the different approaches but
they are not useful enough for me. I want to compare the silicon area
used by our FPGA based Computing Machine, against the one used by a conventional
processor. Our system uses Xilinx XC4008 FPGAs.

1.- How much area has a tile on a XC4008-6 FPGA? 
    (A tile is a CLB with the adyacent routing network)

2.- How much area is devoted to the routing and how much to the CLB inside
    the tile?

3.- How much area is devoted to the lookup table inside the CLB?

4.- How much area (on a tile) is spent on making it programmable?
    (I mean the fraction of logic used just to program 
     the device, and not used by the application). 

5.- What technology is currently being used? (CMOS 0.8, two metal layers; I guess)

I do not know what information may be classified by the people of Xilinx.
However, I think that questions 1, 2 and 5 are too general to be sensitive, and
anyone could get this information just looking (very close) at the chip.

Any information from other manufacturers will also be appreciated.

Thank you.

Javier Moran.
Department of Electronic Engineering.
Technical University of Madrid. SPAIN.
email: moran@die.upm.es



Article: 240
Subject: Re: What do think about the Intel Flexlogic8160?
From: devb@char.vnet.net (David Van den Bout)
Date: 30 Sep 1994 08:39:55 -0400
Links: << >>  << T >>  << A >>
>Were they ever made available by Intel? Just before Intel sold to Altera
>I was told they weren't available yet. Is Altera going to bring it out?

The word from Altera is that they will definitely bring out the 8160
and also improved versions of the 780 and 740 (with flash instead of
EPROM).


-- 

||  Dave Van den Bout  ||
||  Xess Corporation   ||


Article: 241
Subject: Re: XC1765DPD8C
From: mjm@hpqtdzk.sqf.hp.com (Murdo Mckissock)
Date: Fri, 30 Sep 1994 13:10:14 GMT
Links: << >>  << T >>  << A >>
Tomas Whitlock (txw@festival.ed.ac.uk) wrote:

: I'm sure someone has already posted this, but I am new to this newsgroup.
: Is there a Xilinx FTP site with public domain software anywhere? Its
: address would be much appreciated.


No, all they currently offer is a billboard system accessed by Modem :-(
I hope they will soon provide something more convenient.

I do not know the uk phone number for the billboard.  Call Xilinx UK for this.


--Murdo McKissock (modem - whats that?)


Article: 242
Subject: ATT/ORCA references desired
From: bobe@soul.tv.tek.com (Bob Elkind)
Date: 30 Sep 1994 18:02:33 GMT
Links: << >>  << T >>  << A >>
Up until recently, Xilinx 4K has had the playing field pretty much to
themselves.  I 've heard from several designers that the ATT ORCA 2Cxx
family is real, and is worth considering.

I'm interested in your opinions of the 2Cxx technology, particularly if
you have real-world, production (i.e. shipped products) experience with
the stuff.

If your comments are necessarily confidential, I will respect your
confidentiality (please reply via private email!).

As time and work level permit, I will post a summary of the responses
to the net.

Keep in mind, selecting an FPGA technology is not a popularity contest.
There are many considerations with varying importance for differing
customers and applications.  Often we have the tendency to vilify the
beast with whom we've lived over the years because we are intimately
familiar with each others' shortcomings, and (as they say) the grass
appears somewhat more green on the other side of the fence.

As a designer, I'm performing due diligence responsibility in
considering the technology alternatives.  I've designed with Xilinx 3K
and 4K products for >4 years, and I have many successful designs "under
my belt".  I have a lot of respect for Xilinx, and their sales reflect
their having done their work very well.

Bob Elkind, Tek TV  bobe@tv.tv.tek.com


Article: 243
Subject: PCMCIA
From: Corby James <cjames@onramp.net>
Date: Fri, 30 Sep 94 19:30:29 PDT
Links: << >>  << T >>  << A >>

I'm looking for a description of the PCMCIA protocols (ie. bus timing and 
signals, standard pinout if applicable).  Can anyone point me in the right 
direction?

	Thanks in advance.



Article: 244
Subject: copy a GAL16V8B ?
From: tparker@netcom.com (Todd Parker)
Date: Sat, 1 Oct 1994 03:44:00 GMT
Links: << >>  << T >>  << A >>
I have a pc card that has a lattice 'gal16v8b 25lp'. I suspect the gal is 
used for address decoding, the card is a simple parallel-type interface, 
but not a lpt parallel port.

I would like to re-create the card on to another card (combining 3 cards 
to one for a slot-starved portable...)

I have no problem with removing this gal (it is in a socket) and putting 
it into the new card, but... what IF I let the smoke out?

I would be sunk. the card was mfg by a 3rd party, and the second party is
not going to tell me who/what/when or sell me the part itself (they will
sell me another card for us$150.00, but that defeats the project goal). 

Can the gal be duplicated without having the original .jed file?

--
[todd] Saving Virtual Trees; www file://ftp.netcom.com/pub/tparker/home.html


Article: 245
Subject: Re: What do think about the Intel Flexlogic8160?
From: gt0361b@prism.gatech.edu (Benjamin Gene Cheung)
Date: 1 Oct 1994 02:50:58 -0400
Links: << >>  << T >>  << A >>
Bryan Butler (bbutler@netcom.com) wrote:
: Benjamin Gene Cheung (gt0361b@prism.gatech.edu) wrote:
: > Has anyone sampled the ifx8160 from intel?  What do think of it?

: Were they ever made available by Intel? Just before Intel sold to Altera
: I was told they weren't available yet. Is Altera going to bring it out?

I think Intel is sampling them now.  Only Altera knows if they are going
to sell it or not.


-- 
Benjamin Gene Cheung
Computer Engineer
Georgia Institute of Technology
Internet: gt0361b@prism.gatech.edu


Article: 246
Subject: Re: What do think about the Intel Flexlogic8160?
From: gt0361b@prism.gatech.edu (Benjamin Gene Cheung)
Date: 1 Oct 1994 02:52:15 -0400
Links: << >>  << T >>  << A >>
David Van den Bout (devb@char.vnet.net) wrote:
: >Were they ever made available by Intel? Just before Intel sold to Altera
: >I was told they weren't available yet. Is Altera going to bring it out?

: The word from Altera is that they will definitely bring out the 8160
: and also improved versions of the 780 and 740 (with flash instead of
: EPROM).

I think intel is sampling them now.  Has anybody tried them out?

-- 
Benjamin Gene Cheung
Computer Engineer
Georgia Institute of Technology
Internet: gt0361b@prism.gatech.edu


Article: 247
Subject: new PLDshell FTP site
From: devb@char.vnet.net (David Van den Bout)
Date: 2 Oct 1994 12:10:00 -0400
Links: << >>  << T >>  << A >>
Since Intel will drop their FTP site for PLDshell on Oct 1, XESS Corp.
will begin providing an FTP site for PLDshell 3.1.  The site is

	ftp.vnet.net

Transfer to directory pub/xess and get the files pldsh.zip and install.txt.
pldsh.zip has the 3.1 version of PLDshell, the downloading software, and
all the design examples from the "FPGA Workout" text.  the install.txt
file describes how to unpack and install the software.  You will need
the PKUNZIP tool to do this.

XESS Corp. will maintain this FTP site until Altera can provide this
service directly.  We will upgrade to PLDshell version 4.0 once Altera
provides us with an officially released copy of the software.

-- 

||  Dave Van den Bout  ||
||  Xess Corporation   ||


Article: 248
Subject: Re: What do think about the Intel Flexlogic8160?
From: Eric@wolf359.exile.org (Eric Edwards)
Date: Sun, 02 Oct 1994 21:48:33 GMT
Links: << >>  << T >>  << A >>
In article <36h0ur$lpr@char.vnet.net>, David Van den Bout writes:

> >Were they ever made available by Intel? Just before Intel sold to Altera
> >I was told they weren't available yet. Is Altera going to bring it out?
> 
> The word from Altera is that they will definitely bring out the 8160
> and also improved versions of the 780 and 740 (with flash instead of
> EPROM).

Cool.  Any idea of time frame for the improved 780's and 740's?  This
would eliminate my biggest beef about the FlexLogic chips, the #$@ OTP
EPROM.

----
Eric Edwards: Bang= cg57.esnet.com!wolf359!eric Domain= eric@exile.org
Remember the home hobbyist computer: Born 1975, died April 29, 1994



Article: 249
Subject: PREP info needed
From: potral@e5sf.hweng.syr.ge.com
Date: Mon, 3 Oct 1994 16:45:06 GMT
Links: << >>  << T >>  << A >>

I'm interested in obtaining the PREP Benchmark Results, Version 1.3.
Does this document exist on-line?  Can it be mail ordered?

Any and all info is greatly appreciated !!

Thanks in advance ...
Rich Potral




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