Custom Search
Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive
Compare FPGA features and resources
Threads starting:
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
>hi all. >im currently learning FPGAs and having big interesting in it. but the >only place i can practice is the uni lab. i wanna practice at home >with my own laptop as well. but i dont know which software should i >download as im using ubuntu10.4 as my OS. any advices and tips? thx :D > Whose FPGAs are used in the lab? --------------------------------------- Posted through http://www.FPGARelated.comArticle: 152826
On Oct 26, 8:02=A0am, Paul Gotch <pa...@at-cantab-dot.net> wrote: > In comp.arch.embedded fatalist <simfid...@gmail.com> wrote: > > > Believe it or not, but we need "first to file": it just makes life > > simpler for everybody, including small guys > > I'm arguing for first to file. > > > I can predict exactly what happens: PTO will be flooded with junk > > Um rather than the system being flooded with junk patents which are > massively costly and time consuming to invalidate afterwards? I've much > rather the patents were never awarded in the first place even it it > does mean the PTO have to do a lot of sifting of paper. > > > And who are those evil "trolls" ? Have you seen one ? > > Yes. However clearly I can't go into details of specific cases. The > system at the moment is setup on the basis of mutally assured > destruction. > > A sues B over infringement > B counter sues A over infringment > > A and B both produce widgets and their businesses will be adversely > affected if they have to stop shipping widgets. Therfore A and B settle > out of court and sign a cross licensing agreement. > > Now C, a 'Patent Troll' doesn't have anything widgets to counter sue > over all they do is litigate until either their patent is invalidated > or they win. The amount of time an money spent on this is such that > for many companies it is cheaper to settle and pay royalties on an > invalid patent than it is to get it overturned. > > This is then used as precident to attack other people using the same > patent 'look they licensed it it must be valid' pay up or else. > > > And why treble damages for willful =A0infringement should be removed ? > > Because the actual effect of this is that you build something and you > *do not* do any kind of search to see if you need to license anything > and need to pay royalties. If you do do a search and miss something > then it's impossibly hard to prove that you didn't find the patent in > question and aren't willfully infringing it. > > -p > -- > Paul Gotch > -------------------------------------------------------------------- "Therfore A and B settle out of court and sign a cross licensing agreement." Not anymore... Big corporate monstrosities are now beating the shit out of each other in patent courts and ITC, trying to get *injunctions* against each others products (e.g. Apple vs.Samsung, Apple vs. HTC etc etc etc) Also, many of them collect huge patent royalties on someone else's products - e.g. Microsoft lining its pockets with patent royalties for Android - Google's flagship product. (Is Microsoft a patent troll or what ?) Don't you love it ? I certainly do :)Article: 152827
Hi Is there any way of getting modelsim not to use the widoz registry for settings. I would prefer if it would use my .modelsim file. I find it impossible to express the extent of my disgust at having to temporally use such an abhorrent operating system for development!!!Article: 152828
In comp.arch.embedded fatalist <simfidude@gmail.com> wrote: > Big corporate monstrosities are now beating the shit out of each other > in patent courts and ITC, trying to get *injunctions* against each > others products (e.g. Apple vs.Samsung, Apple vs. HTC etc etc etc) That's atypical. Steve Jobs was philosophically opposed to Android and would rather litigate than settle. For example Apple vs Nokia settled eventually. > Also, many of them collect huge patent royalties on someone else's > products - e.g. Microsoft lining its pockets with patent royalties for > Android - Google's flagship product. (Is Microsoft a patent troll or > what ?) The MS situation is annoying because MS are licensing 'unspecified' IP. There are almost certainly valid patents in there but there are also probably lots of invalid ones. The most litigated patent of the lot in there is the VFAT long file names one. MS managed to get FAT32 used as the default file system on many flash devices and then started litiging afterwards. This is morally dubious but trolling per say. What you don't see is that a normal handset has IP from perhaps 30 companies inside it and the majority of the software cost is license fees for patents. All you see are the ones that actually end up in court due to brinkmanship on one or other side. -p -- Paul Gotch --------------------------------------------------------------------Article: 152829
On 26/10/2011 13:49, dgreig wrote: > Hi > > Is there any way of getting modelsim not to use the widoz registry for > settings. No only Unix/Linux uses the .modelsim file. Hans. www.ht-lab.com > I would prefer if it would use my .modelsim file. > > I find it impossible to express the extent of my disgust at having to > temporally use such an abhorrent operating system for development!!!Article: 152830
On 26/10/2011 14:36, fatalist wrote: > > "Therfore A and B settle out of court and sign a cross licensing > agreement." > > Not anymore... > > Big corporate monstrosities are now beating the shit out of each other > in patent courts and ITC, trying to get *injunctions* against each > others products (e.g. Apple vs.Samsung, Apple vs. HTC etc etc etc) > Also, many of them collect huge patent royalties on someone else's > products - e.g. Microsoft lining its pockets with patent royalties for > Android - Google's flagship product. (Is Microsoft a patent troll or > what ?) > Yes, increasingly MS is a patent troll. They own a selection of dubious patents, and use standard troll racketeering techniques to squeeze protection money out of Android phone and tablet manufacturers. The fact that the patents involved are not revealed is classic patent troll behaviour. These manufacturers don't have many patents of their own to fight back, or for cross-licensing deals. Even if they have patents that MS arguably infringes on in its own phones, why would they care? No one buys them anyway, so they have nothing significant to lose. <http://www.gottabemobile.com/2011/08/05/microsofts-android-business-brings-in-3-times-as-much-money-as-windows-phone/> Of course, patent trolling is not MS's core business - but it is an increasing part of their business. > Don't you love it ? I certainly do :)Article: 152831
http://www.bbc.co.uk/news/technology-15461732 Completely barmy. There is definitely something very, very wrong with software patents. RupertArticle: 152832
fatalist wrote: > Big corporate monstrosities are now beating the shit out of each other > in patent courts and ITC, trying to get *injunctions* against each > others products (e.g. Apple vs.Samsung, Apple vs. HTC etc etc etc) Well, that's because Apple invented the smart phone, along with the wheel, and hot water; *how* *dare* these copycats swoop in and steal our preciousss IP! (I can definitely picture the Apple board taking turns playing Gollum.) Patents and copyright are a net loss for society.Article: 152833
Paul Gotch wrote: > The MS situation is annoying because MS are licensing 'unspecified' IP. > There are almost certainly valid patents in there but there are also > probably lots of invalid ones. The MPEG-LA, and associated parasites, have no problem requiring payment of royalties for expired or unrelated patents. > The most litigated patent of the lot in > there is the VFAT long file names one. MS managed to get FAT32 used as > the default file system on many flash devices and then started litigating > afterwards. This is morally dubious but trolling per say. Drug dealers should have patented that business model years ago.Article: 152834
On 26 Oct 2011 13:02:45 +0100 (BST), Paul Gotch <paulg@at-cantab-dot.net> wrote: > >> And why treble damages for willful infringement should be removed ? > >Because the actual effect of this is that you build something and you >*do not* do any kind of search to see if you need to license anything >and need to pay royalties. If you do do a search and miss something >then it's impossibly hard to prove that you didn't find the patent in >question and aren't willfully infringing it. > This can't be fixed since it is not practical for a small inventor to do an exhaustive search of existing IP prior to filing, and doing so would hopelessly contaminate the inventor, anyway. The idea of "willful" rather than inadvertent infringement has legs, I think, if it can be shown proven that the infringement was willful. Otherwise the default position is almost necessarily that one does not do a prior art search prior to filing. Removing treble damages for willful infringement just makes it easier for those who really do infringe willfully. Eric Jacobsen Anchor Hill Communications www.anchorhill.comArticle: 152835
In article <4ea8235d.66344947@www.eternal-september.org>, Eric Jacobsen <eric.jacobsen@ieee.org> wrote: >This can't be fixed since it is not practical for a small inventor to >do an exhaustive search of existing IP prior to filing, and doing so >would hopelessly contaminate the inventor, anyway. By which you mean 'would tell the inventor what works and what doesn't, meaning that he just needs to licence it rather than invent it'. Six months in the lab can save you an afternoon in the library and all that. > The idea of "willful" rather than inadvertent infringement has legs, > I think, if it can be shown proven that the infringement was > willful. > Otherwise the default position is almost necessarily that one does > not do a prior art search prior to filing. Surely it is prior to *inventing* that you should do the prior art search, so that you know which patents you could licence instead of doing the tedious invention. That would be easier if it were compulsory to licence all patents under reasonable non-discriminatory terms, and *that* is the direction in which patent reform should go - mostly people infringe patents because the original inventor isn't willing to licence them on generally-reasonable terms. TomArticle: 152836
On Oct 26, 12:16=A0pm, Thomas Womack <twom...@chiark.greenend.org.uk> wrote: > In article <4ea8235d.66344...@www.eternal-september.org>, > > Eric Jacobsen <eric.jacob...@ieee.org> wrote: > >This can't be fixed since it is not practical for a small inventor to > >do an exhaustive search of existing IP prior to filing, and doing so > >would hopelessly contaminate the inventor, anyway. > > By which you mean 'would tell the inventor what works and what > doesn't, meaning that he just needs to licence it rather than invent > it'. =A0Six months in the lab can save you an afternoon in the library > and all that. > > > The idea of "willful" rather than inadvertent infringement has legs, > > I think, if it can be shown proven that the infringement was > > willful. > > Otherwise the default position is almost necessarily that one does > > not do a prior art search prior to filing. > > Surely it is prior to *inventing* that you should do the prior art > search, so that you know which patents you could licence instead of > doing the tedious invention. =A0That would be easier if it were > compulsory to licence all patents under reasonable non-discriminatory > terms, and *that* is the direction in which patent reform should go - > mostly people infringe patents because the original inventor isn't > willing to licence them on generally-reasonable terms. > > Tom "mostly people infringe patents because the original inventor isn't willing to licence them on generally-reasonable terms. " Whaaaaaat ??????Article: 152837
I've got a project going on a Cyclone III, and have hit an issue that seems like it has a simple solution if only I already knew it. I've got a 125 MHz input clock (CLK125). I've got an ADC that takes in an LVDS 250 MHz clock (CLKOUT), and outputs 250 Msps parallel LVDS data, changing on the rising edge of a regenerated 250 MHz clock (CLKFB). The phase relationship of my FPGA to anything other than the ADC doesn't matter. Jitter on CLKOUT isn't the femtosecond sensitivities than you might expect for a 250 Msps ADC; the application turns out not mind a bit of spectral smear. So my goals are to generate from a 125 MHz reference: * A 250 MHz clock going out to the ADC * An internal 250 MHz clock; properly phase aligned to latch the ADC data in the center of the valid window. * An internal 125 MHz clock for processing that can't be done at 250 MHz, phase aligned to the internal 125 MHz This seems like it should be pretty straightforward, that there's some obvious application of the PLL's source synchronous mode that makes this all just work, but I'm not seeing it. I suppose I could use one PLL to turn CLK125 into CLKOUT, then source the second PLL from CLKFB, but that seems like it's going to leave me with an internal clock that's the result of cascading PLLs, and at 250 MHz that seems like the jitter might seriously cut into my timing margins. Anyone have any thoughts? Thanks, Rob -- Rob Gaddi, Highland Technology -- www.highlandtechnology.com Email address domain is currently out of order. See above to fix.Article: 152838
On Oct 26, 2:12=A0pm, HT-Lab <han...@htminuslab.com> wrote:
> On 26/10/2011 13:49, dgreig wrote:
>
> > Hi
>
> > Is there any way of getting modelsim not to use the widoz registry for
> > settings.
>
> No only Unix/Linux uses the .modelsim file.
>
> Hans.www.ht-lab.com
>
>
>
>
>
>
>
> > I would prefer if it would use my .modelsim file.
>
> > I find it impossible to express the extent of my disgust at having to
> > temporally use such an abhorrent operating system for development!!!
Although I had set the MODELSIM_PREFERENCES environment variable for
both user and sysyem tp point to a file ("c:\.modelsim") it was not
being updated when closing modelsim.
It now is being updated...
Something to do with windoz, perhaps? Ho humm.
Article: 152839On Oct 26, 7:13=A0am, "rupertlssm...@googlemail.com" <rupertlssm...@googlemail.com> wrote: > http://www.bbc.co.uk/news/technology-15461732 > > Completely barmy. There is definitely something very, very wrong with > software patents. > > Rupert One of the requirement for a US patent are ~not being obvious to someone versed in the art. I am hugely biased. The existence/invention of the mouse at PARC and devices like touch pads and drawing pads pretty much means just about anything done since is pretty obvious to the point of being derivative. *BUT* this kind of stuff has been going on for years. A specific example would be Atari patented using 4 bits to map an 8 position joystick for the 2600 VCS. Nintendo was smart enough to design their own system but Atari successfully sued Sega and IIRC won $10s of millions years after the fact. RickArticle: 152840
In comp.arch.fpga Thomas Womack <twomack@chiark.greenend.org.uk> wrote: (snip, someone wrote) >> Otherwise the default position is almost necessarily that one does >> not do a prior art search prior to filing. > Surely it is prior to *inventing* that you should do the prior art > search, so that you know which patents you could licence instead of > doing the tedious invention. That would be easier if it were > compulsory to licence all patents under reasonable non-discriminatory > terms, and *that* is the direction in which patent reform should go - > mostly people infringe patents because the original inventor isn't > willing to licence them on generally-reasonable terms. I suppose for some, like Thomas Edison, that might make sense. It seems to me more usual that someone has a problem to solve and searches for solutions to the problem. That could be done looking through patents, or one might just reinvent something. Now, the real problem it seems to me is that the PTO isn't so good at detecting "obvious" inventions. Also, some patents are unnecessarily broad, covering things that really haven't yet been invented. For both those reasons, one might find something already patented when it shouldn't be. -- glenArticle: 152841
In comp.arch.fpga fatalist <simfidude@gmail.com> wrote: (snip, someone wrote) > "mostly people infringe patents because the original inventor isn't > willing to licence them on generally-reasonable terms. " > Whaaaaaat ?????? I would guess that more often it is because one doesn't realize that something is covered by a patent. Either one doesn't know at all, or assumes that a known patent doesn't cover the specific case. One well known infringement case was the Kodak instant camera. Kodak, in trying to compete with Polaroid, designed a camera that, I would guess, they believed didn't infringe. It is pretty hard to believe that they didn't know about Polariod. (Actually, Kodak made the negative material used by Polaroid, so they pretty much definitely knew.) Polariod sued and, eventually, Kodak lost. (You still see Kodak instant cameras in thrift stores.) To avoid generating a mirror image, Polaroid cameras reflect the light off a mirror before it hits the film. Kodak, instead, designed a system that exposes through the back with the print visible from the front. I presume they thought that difference would avoid the Polaroid patent. On the other hand, cobalt-doped ferric oxide magnetic tape was specifically developed to avoid the patent on chromium dioxide. Tape that worked with recorders designed for CrO2 tape could be made, cheaper and maybe even better. As far as I know, that was never considered infringing. -- glenArticle: 152842
Sad news. I didn't knew him personally, but I remember great articles from him, like "Unusual Clock Dividers": http://www2.fiu.edu/~vjaya002/vlsi%20BOOKS/clock%20dividers.pdf and many interesting articles in this newsgroup, which helped me a lot when I was a FPGA newbie. -- Frank Buss, http://www.frank-buss.de piano and more: http://www.youtube.com/user/frankbussArticle: 152843
On 26 Oct 2011 17:16:02 +0100 (BST), Thomas Womack <twomack@chiark.greenend.org.uk> wrote: >In article <4ea8235d.66344947@www.eternal-september.org>, >Eric Jacobsen <eric.jacobsen@ieee.org> wrote: > >>This can't be fixed since it is not practical for a small inventor to >>do an exhaustive search of existing IP prior to filing, and doing so >>would hopelessly contaminate the inventor, anyway. > >By which you mean 'would tell the inventor what works and what >doesn't, meaning that he just needs to licence it rather than invent >it'. Six months in the lab can save you an afternoon in the library >and all that. I can tell you've never tried this. Many patents aren't for anything useful, or are for such a small niche that one might think it useful for a particular application when it isn't. This assumes that one can even sort out what a patent is trying to say, since many, if not most, are written to be deliberately obscure. Just because it's patented doesn't mean it works, or even if it works that it's a good way to do it. And in many, if not most, cases, there is such a huge number of patents that will turn up searching for a particular topic, or NOT turn up when they should, that searches are pretty useless to start with. IMHO, anyway. >> The idea of "willful" rather than inadvertent infringement has legs, >> I think, if it can be shown proven that the infringement was >> willful. > >> Otherwise the default position is almost necessarily that one does >> not do a prior art search prior to filing. > >Surely it is prior to *inventing* that you should do the prior art >search, so that you know which patents you could licence instead of >doing the tedious invention. As an engineer your job is often to invent, and if you work in a lab or an IP company your job is to create useful IP. In this case, there may be a strong incentive to NEVER do patent searches because you may be "contaminated" by what you read, i.e., unable to avoid using ideas that you got from reading patents. You can spend your time trying to solve a problem in an efficient way with the state-of-the-art components available to you, or you can spend your time slogging through patents that may or may not be useful, or even work properly, that were developed with technology that is probably at least five years old (which is typical time for a patent to grant from application). In my experience what you're proposing just doesn't make sense in most practical environments. > That would be easier if it were >compulsory to licence all patents under reasonable non-discriminatory >terms, and *that* is the direction in which patent reform should go - >mostly people infringe patents because the original inventor isn't >willing to licence them on generally-reasonable terms. > >Tom I don't know of very many cases like that. In my experience most infringement cases come from not knowing that the infringing patent existed, because they can be quite difficult and time consuming to identify. Just looking at the economics of how the game works patent searches generally don't make any sense to perform. There is more risk in doing patent searches, due to contamination, potential for subsequent "willful" infringement, and just plain wasting time trying to find something relevant. Again, just IMHO. Patents are a very strange game, and I don't know that the "reforms" have made it any better. Many patent portfolios are defensive, and not intended to be asserted against anyone. Eric Jacobsen Anchor Hill Communications www.anchorhill.comArticle: 152844
In comp.arch.fpga Eric Jacobsen <eric.jacobsen@ieee.org> wrote: (snip) > I can tell you've never tried this. Many patents aren't for anything > useful, or are for such a small niche that one might think it useful > for a particular application when it isn't. This assumes that one > can even sort out what a patent is trying to say, since many, if not > most, are written to be deliberately obscure. > Just because it's patented doesn't mean it works, or even if it works > that it's a good way to do it. But it might be patented anyway. One that I know about is the patent on analysis of mass spectroscopy for protein sequencing. They patented all possible algorithms for mass spectrometry sequence analysis, even ones not yet devised. It should be possible to overturn, but it will take money from someone who really wants to do it. -- glenArticle: 152845
Guys, Do you know if it is possible to get a complete pinout report from the Actel compilation flow? I want something complete like the CSV file that comes out of the Xilinx ISE compilation process. I use the PHDL language to design my printed circuit boards (phdl.sourceforge.net). I use the Xilinx CSV file to autogenerate the device declaration in my board design.The Xilinx CSV file contains all the pins of the package including the power, ground, special purpose pins, used I/O and unused pins. I wrote a Xilinx2PHDL java program that parses the CSV file into a device declaration so that I can instantiate even large parts onto my board with very little typing. I would like to write a similar translator for Actel FPGAs but I cannot find a way to get a complete pinout report from my compiled FPGA design. Any help is greatly appreciated.Article: 152846
Rick <richardcortese@gmail.com> wrote: >One of the requirement for a US patent are ~not being obvious to >someone versed in the art. Hah. SteveArticle: 152847
On Oct 26, 12:44=A0pm, Rob Gaddi <rga...@technologyhighland.invalid> wrote: > > * An internal 125 MHz clock for processing that can't be done at 250 > MHz, phase aligned to the internal 125 MHz > > This seems like it should be pretty straightforward, that there's some > obvious application of the PLL's source synchronous mode that makes this > all just work, but I'm not seeing it. =A0I suppose I could use one PLL to > turn CLK125 into CLKOUT, then source the second PLL from CLKFB, but that > seems like it's going to leave me with an internal clock that's the > result of cascading PLLs, and at 250 MHz that seems like the jitter > might seriously cut into my timing margins. =A0Anyone have any thoughts? > Why not just bring the ADC data into a dual clock FIFO clocked on one side by CLKFB? You may also be able to widen the data prior to putting it into the FIFO so that the output side can be clocked by CLK125. KJArticle: 152848
On Oct 26, 10:13=A0am, "rupertlssm...@googlemail.com" <rupertlssm...@googlemail.com> wrote: > http://www.bbc.co.uk/news/technology-15461732 > > Completely barmy. There is definitely something very, very wrong with > software patents. > > Rupert I looked at this and I think it is a perfect example of how poor the patent examination process is. If your primary user interface is a touch screen and you want to lock the device, how else would you unlock the device than through a touch screen "gesture"? How bleeding obvious does the invention have to be to be unpatentable? I guess Google could claim they aren't using a touch screen gesture but rather they are presenting the user with a virtual control which the user operates... and patent that! RickArticle: 152849
On 26 Okt., 21:39, glen herrmannsfeldt <g...@ugcs.caltech.edu> wrote: > In comp.arch.fpga fatalist <simfid...@gmail.com> wrote: > > (snip, someone wrote) > > > "mostly people infringe patents because the original inventor isn't > > willing to licence them on generally-reasonable terms. " > > Whaaaaaat ?????? > > I would guess that more often it is because one doesn't realize > that something is covered by a patent. =A0Either one doesn't know > at all, or assumes that a known patent doesn't cover the specific case. > I seem to remember hearing about companies where reading or searching patents wasn't allowed, I assume as a defence in case they accidently infringed on some patent > snip -Lasse
Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive
Compare FPGA features and resources
Threads starting:
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