USB programmable Open Source Hardware
in [FPGA]
|
From: Antti.Lukats on 28 Sep 2009 08:57 > > logic-U-Plus is GENERIC FX2 board that can be configured > > to emulate either Locic fromwww.saleae.com > > or > > USBee AX Pro fromwww.usbee.com > > Those are not cheap, nor are they chips - and cannot pgm the FPGA > out of the box... - as this thread is about FPGA boards, my comments > were in that context. > > ie what can do the PGM pathway _and_ give other runtime information, > almost for free. > In such single unit products, the incremental chip price variation is > not > material. > - Digikey shows the saving to be a princely $1 ;) > > So, give the user a choice of much faster download times, and the > potential? for a free Logic/Frequency HW pathway, all for $1 more, > and > what would they choose ? Jim, 1) FT2232H prices have gone down a great deal, I wasnt even aware well, if Cy would 1) remove the CPLD from the design this would be GOOD 2) replace FT245 with FT2232H this would be GOOD also but [2] is optional :) Antti
From: rickman on 28 Sep 2009 11:40 On Sep 28, 7:58 am, Uwe Bonnes <b...(a)elektron.ikp.physik.tu- darmstadt.de> wrote: > rickman <gnu...(a)gmail.com> wrote: > > On Sep 27, 12:42 pm, "Antti.Luk...(a)googlemail.com" > > <antti.luk...(a)googlemail.com> wrote: > > > On Sep 26, 8:39 pm, rickman <gnu...(a)gmail.com> wrote: > > > > > > > I enjoy your responses they are to the bone, but valid. The right > > <zillions of line of senseless quote delete> > > Do people ever use archived news and get angry about unrelated quote meaning > unrelated search hits? > > > Why a two layer board? I would expect any decent design to use at > > least four layers just so it can have a ground/power plane for noise > > reduction. Especially when you don't know what someone will be doing > > with it, best is to provide a bit of overkill. I'm not interested in > > saving every last penny on a development board like this. > > With programmable pin helping the layouter to get a planar layout, the > bottom layer can be made quite a continous groundplane. Yes, that is true, but not my point. A single ground plane does nothing to reduce noise on the power rails. The capacitor that is formed by parallel planes spaced 10 mil in a PWB is the best power supply decoupling device you can provide. Even on a very small board, these planes provide significant noise elimination, both in terms of minimizing the effect on the chips and also in terms of reducing EMI. I realize that many designs just don't have a need for this, but my point was that a general purpose development/eval board needs to consider a wide range of designs including ones that push the speed of the device and have a number of outputs switching at high edge rates. Capacitors alone will not normally address the problem adequately in these cases. Rick
From: rickman on 28 Sep 2009 11:45 On Sep 28, 8:23 am, "Antti.Luk...(a)googlemail.com" <antti.luk...(a)googlemail.com> wrote: > > Do people ever use archived news and get angry about unrelated quote meaning > > unrelated search hits? > > > > Why a two layer board? I would expect any decent design to use at > > > least four layers just so it can have a ground/power plane for noise > > > reduction. Especially when you don't know what someone will be doing > > > with it, best is to provide a bit of overkill. I'm not interested in > > > saving every last penny on a development board like this. > > > With programmable pin helping the layouter to get a planar layout, the > > bottom layer can be made quite a continous groundplane. > > > .. > > > Bye > > -- > > Uwe Bonnes b...(a)elektron.ikp.physik.tu-darmstadt.de > > > Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt > > --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- > > please complain to the GOOGLE (and saying that use something > else isnt an option always, sorry... I use what is the easiest way) > when replying using google news its not easy to delete the rubish > as it not even seen on screen > > well, i see the thread is reached status where people start to > complain > about quoting issues, :( If you are using Google Groups, I hope you are clicking the spam and reporting it. Google seems to continue to make this easier. The last time I noticed, you had to open the "opions" and click "report the message", then type at least "spam" into the edit box. Now I see that there is a link at the bottom of the post where you can just click once. It even confirms that the report has been accepted! How much easier can it get? Rick
From: Antti.Lukats on 28 Sep 2009 13:31 On Sep 28, 6:45 pm, rickman <gnu...(a)gmail.com> wrote: > On Sep 28, 8:23 am, "Antti.Luk...(a)googlemail.com" > > > > <antti.luk...(a)googlemail.com> wrote: > > > Do people ever use archived news and get angry about unrelated quote meaning > > > unrelated search hits? > > > > > Why a two layer board? I would expect any decent design to use at > > > > least four layers just so it can have a ground/power plane for noise > > > > reduction. Especially when you don't know what someone will be doing > > > > with it, best is to provide a bit of overkill. I'm not interested in > > > > saving every last penny on a development board like this. > > > > With programmable pin helping the layouter to get a planar layout, the > > > bottom layer can be made quite a continous groundplane. > > > > .. > > > > Bye > > > -- > > > Uwe Bonnes b...(a)elektron.ikp.physik.tu-darmstadt.de > > > > Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt > > > --------- Tel. 06151 162516 -------- Fax. 06151 164321 ---------- > > > please complain to the GOOGLE (and saying that use something > > else isnt an option always, sorry... I use what is the easiest way) > > when replying using google news its not easy to delete the rubish > > as it not even seen on screen > > > well, i see the thread is reached status where people start to > > complain > > about quoting issues, :( > > If you are using Google Groups, I hope you are clicking the spam and > reporting it. Google seems to continue to make this easier. The last > time I noticed, you had to open the "opions" and click "report the > message", then type at least "spam" into the edit box. Now I see that > there is a link at the bottom of the post where you can just click > once. It even confirms that the report has been accepted! > > How much easier can it get? > > Rick I do report all spam messages, if you wanted to know that Antti
From: -jg on 28 Sep 2009 17:34
On Sep 29, 3:40 am, rickman <gnu...(a)gmail.com> wrote: > Yes, that is true, but not my point. A single ground plane does > nothing to reduce noise on the power rails. The capacitor that is > formed by parallel planes spaced 10 mil in a PWB is the best power > supply decoupling device you can provide. Even on a very small board, > these planes provide significant noise elimination, both in terms of > minimizing the effect on the chips and also in terms of reducing > EMI. > > I realize that many designs just don't have a need for this, but my > point was that a general purpose development/eval board needs to > consider a wide range of designs including ones that push the speed of > the device and have a number of outputs switching at high edge rates. > Capacitors alone will not normally address the problem adequately in > these cases. I'd generally agree - a demoboard such as this, should not be a 'minefield for the unwary', but the design itself should be paste-able into someone's project. (unless it is some highly specific FPGA subset, that only uses half a dozen IOs, but that's a different type of demoboard... ) A better place to drive the PCB cost down, is simply to reduce the area. -jg |