Yahoo Groups archive
Thread
2004-11-29 by digtalfreak
I am using the LPC2129 in an application and I need to know if the analog port will be damaged if the port gets more than V3A at any time. The manual says that the voltage must not be over V3A or the analog input readings will be invalid. Does that mean that they will be invalid forever or only for the moment where the voltage is over V3A?
2004-11-29 by Anton Erasmus
On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > I am using the LPC2129 in an application and I need to know if the > analog port will be damaged if the port gets more than V3A at any > time. The manual says that the voltage must not be over V3A or the > analog input readings will be invalid. Does that mean that they will > be invalid forever or only for the moment where the voltage is over > V3A? > The pins are supposed to be 5V tolerant, hence my assumption is that they should be able to handle up to 5V. The A/D converter is designed, so that it's full range is only V3A volts, hence an input above this voltage will give invalid A/D results. Regards Anton Erasmus > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus
2004-11-29 by Richard
Note: 5V on any pin that CAN be configured as an ADC input will corrupt the ADC conversion values. Richard --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > > > > > I am using the LPC2129 in an application and I need to know if the > > analog port will be damaged if the port gets more than V3A at any > > time. The manual says that the voltage must not be over V3A or the > > analog input readings will be invalid. Does that mean that they will > > be invalid forever or only for the moment where the voltage is over > > V3A? > > > > The pins are supposed to be 5V tolerant, hence my assumption is that they should > be able to handle up to 5V. The A/D converter is designed, so that it's full range is > only V3A volts, hence an input above this voltage will give invalid A/D results. > > Regards > Anton Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > ----------------------------------------------------------------- ---~-
> > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > -- > A J Erasmus
2004-11-29 by Anton Erasmus
On 29 Nov 2004 at 17:53, Richard wrote: > > > Note: 5V on any pin that CAN be configured as an ADC input will > corrupt the ADC conversion values. > At least now I know why my attempt to use only one A/D channel, and the rest as 5V digital inputs corrupted the A/D readings on my analog input. I hope Philips will update the user manual to spel this sort of behaviour more clearly. Finding this out after one has manufactured a 10 layer PCB is not the most convenient. Regards Anton Erasmus > --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > > On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > > > > > I am using > the LPC2129 in an application and I need to know if the > > analog > port will be damaged if the port gets more than V3A at any > > time. > The manual says that the voltage must not be over V3A or the > > > analog input readings will be invalid. Does that mean that they will > > > be invalid forever or only for the moment where the voltage is over > > > V3A? > > > > The pins are supposed to be 5V tolerant, hence my > assumption is that they should > be able to handle up to 5V. The A/D > converter is designed, so that it's full range is > only V3A volts, > hence an input above this voltage will give invalid A/D results. > > > Regards > Anton Erasmus > > > > > > ------------------------ Yahoo! > Groups Sponsor > > --------------------~--> Make a clean sweep of > pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get > it for free! > > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > > ----------------------------------------------------------------- > ---~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > -- > A J Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> $9.95 domain names from Yahoo!. Register > anything. http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus
2004-11-29 by Lasse Madsen
Hi Richard From what specification do you concoct that information? Regards Lasse
-----Original Message----- From: Richard [mailto:richas@...] Sent: 29. november 2004 18:54 To: lpc2000@yahoogroups.com Subject: [lpc2000] Re: LPC2129, analog port question. Note: 5V on any pin that CAN be configured as an ADC input will corrupt the ADC conversion values. Richard --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > > > > > I am using the LPC2129 in an application and I need to know if the > > analog port will be damaged if the port gets more than V3A at any > > time. The manual says that the voltage must not be over V3A or the > > analog input readings will be invalid. Does that mean that they will > > be invalid forever or only for the moment where the voltage is over > > V3A? > > > > The pins are supposed to be 5V tolerant, hence my assumption is that they should > be able to handle up to 5V. The A/D converter is designed, so that it's full range is > only V3A volts, hence an input above this voltage will give invalid A/D results. > > Regards > Anton Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > ----------------------------------------------------------------- ---~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > -- > A J Erasmus Yahoo! Groups Links
2004-11-29 by Lasse Madsen
I think that we ought to make a list of all the "mistakes" that is generally made with these controllers. There are a lot of people (my self included) that fell for all the nice features but when you look closer and pay the amount of money for PCB's you find out where the sun don't shine ... Would it be possible to summarize a list on the group page of the common mistakes etc? For instance: SPI master = Hardware High on SS pin! And very important Bootloader behaviour! What does it do and how can it be worked around as someone has talked about and what these issues mean from a hardware point of view etc. And apparently (maybe) There's another thing that the ADC can't stand 5V on a pin that Is related to it (What kind of sick cr*p is that?) I fancy these controllers don't get me wrong but one can easily be _very_ upset using _alot_ of time and money making PCB's and then finding out all the beginner mistakes ... Lets make a list of beginner mistakes etc. I'll help people help them selves. Who's in and who's out :) Regards Lasse
-----Original Message----- From: Anton Erasmus [mailto:antone@...] Sent: 29. november 2004 21:44 To: lpc2000@yahoogroups.com Subject: Re: [lpc2000] Re: LPC2129, analog port question. On 29 Nov 2004 at 17:53, Richard wrote: > > > Note: 5V on any pin that CAN be configured as an ADC input will > corrupt the ADC conversion values. > At least now I know why my attempt to use only one A/D channel, and the rest as 5V digital inputs corrupted the A/D readings on my analog input. I hope Philips will update the user manual to spel this sort of behaviour more clearly. Finding this out after one has manufactured a 10 layer PCB is not the most convenient. Regards Anton Erasmus > --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > > On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > > > > > I am using > the LPC2129 in an application and I need to know if the > > analog > port will be damaged if the port gets more than V3A at any > > time. > The manual says that the voltage must not be over V3A or the > > > analog input readings will be invalid. Does that mean that they will > > > be invalid forever or only for the moment where the voltage is over > > > V3A? > > > > The pins are supposed to be 5V tolerant, hence my > assumption is that they should > be able to handle up to 5V. The A/D > converter is designed, so that it's full range is > only V3A volts, > hence an input above this voltage will give invalid A/D results. > > > Regards > Anton Erasmus > > > > > > ------------------------ Yahoo! > Groups Sponsor > > --------------------~--> Make a clean sweep of > pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get > it for free! > > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > > ----------------------------------------------------------------- > ---~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > -- > A J Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> $9.95 domain names from Yahoo!. Register > anything. http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus Yahoo! Groups Links
2004-11-29 by Lasse Madsen
Gramma mistake: >Lets make a list of beginner mistakes etc. I'll help people help them >selves. Should have been: Let's make a list of beginner mistakes etc. It'll help people, help them Selves. No need to be too cocky hehe :) Regards lasse
-----Original Message----- From: Lasse Madsen [mailto:lasse.madsen@...] Sent: 29. november 2004 21:59 To: lpc2000@yahoogroups.com Subject: [lpc2000] LPC Series - Good things gone bad... I think that we ought to make a list of all the "mistakes" that is generally made with these controllers. There are a lot of people (my self included) that fell for all the nice features but when you look closer and pay the amount of money for PCB's you find out where the sun don't shine ... Would it be possible to summarize a list on the group page of the common mistakes etc? For instance: SPI master = Hardware High on SS pin! And very important Bootloader behaviour! What does it do and how can it be worked around as someone has talked about and what these issues mean from a hardware point of view etc. And apparently (maybe) There's another thing that the ADC can't stand 5V on a pin that Is related to it (What kind of sick cr*p is that?) I fancy these controllers don't get me wrong but one can easily be _very_ upset using _alot_ of time and money making PCB's and then finding out all the beginner mistakes ... Lets make a list of beginner mistakes etc. I'll help people help them selves. Who's in and who's out :) Regards Lasse -----Original Message----- From: Anton Erasmus [mailto:antone@...] Sent: 29. november 2004 21:44 To: lpc2000@yahoogroups.com Subject: Re: [lpc2000] Re: LPC2129, analog port question. On 29 Nov 2004 at 17:53, Richard wrote: > > > Note: 5V on any pin that CAN be configured as an ADC input will > corrupt the ADC conversion values. > At least now I know why my attempt to use only one A/D channel, and the rest as 5V digital inputs corrupted the A/D readings on my analog input. I hope Philips will update the user manual to spel this sort of behaviour more clearly. Finding this out after one has manufactured a 10 layer PCB is not the most convenient. Regards Anton Erasmus > --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > > On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > > > > > I am using > the LPC2129 in an application and I need to know if the > > analog > port will be damaged if the port gets more than V3A at any > > time. > The manual says that the voltage must not be over V3A or the > > > analog input readings will be invalid. Does that mean that they will > > > be invalid forever or only for the moment where the voltage is over > > > V3A? > > > > The pins are supposed to be 5V tolerant, hence my > assumption is that they should > be able to handle up to 5V. The A/D > converter is designed, so that it's full range is > only V3A volts, > hence an input above this voltage will give invalid A/D results. > > > Regards > Anton Erasmus > > > > > > ------------------------ Yahoo! > Groups Sponsor > > --------------------~--> Make a clean sweep of > pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get > it for free! > > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > > ----------------------------------------------------------------- > ---~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > -- > A J Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> $9.95 domain names from Yahoo!. Register > anything. http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus Yahoo! Groups Links Yahoo! Groups Links
2004-11-29 by Robert Adsett
At 09:58 PM 11/29/04 +0100, you wrote:
>I think that we ought to make a list of all the "mistakes" that is generally
>made with these controllers.
>
>There are a lot of people (my self included) that fell for all the nice
>features but when you look closer and pay the amount of money for PCB's you
>find out where the sun don't shine ...
>
>Would it be possible to summarize a list on the group page of the common
>mistakes etc?
A FAQ/ useful notes list like this is a good idea. I'd suggest the ARMuC
wiki if onlly because it would be an easy way to collaborate on such a list.
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-29 by Messal, Art
Hi Lasse, What is the "SPI master = Hardware High on SS pin" bug that you mentioned? thanks, -Art- _____
From: Lasse Madsen [mailto:lasse.madsen@...] Sent: Monday, November 29, 2004 1:59 PM To: lpc2000@yahoogroups.com Subject: [lpc2000] LPC Series - Good things gone bad... I think that we ought to make a list of all the "mistakes" that is generally made with these controllers. There are a lot of people (my self included) that fell for all the nice features but when you look closer and pay the amount of money for PCB's you find out where the sun don't shine ... Would it be possible to summarize a list on the group page of the common mistakes etc? For instance: SPI master = Hardware High on SS pin! And very important Bootloader behaviour! What does it do and how can it be worked around as someone has talked about and what these issues mean from a hardware point of view etc. And apparently (maybe) There's another thing that the ADC can't stand 5V on a pin that Is related to it (What kind of sick cr*p is that?) I fancy these controllers don't get me wrong but one can easily be _very_ upset using _alot_ of time and money making PCB's and then finding out all the beginner mistakes ... Lets make a list of beginner mistakes etc. I'll help people help them selves. Who's in and who's out :) Regards Lasse -----Original Message----- From: Anton Erasmus [mailto:antone@...] Sent: 29. november 2004 21:44 To: lpc2000@yahoogroups.com Subject: Re: [lpc2000] Re: LPC2129, analog port question. On 29 Nov 2004 at 17:53, Richard wrote: > > > Note: 5V on any pin that CAN be configured as an ADC input will > corrupt the ADC conversion values. > At least now I know why my attempt to use only one A/D channel, and the rest as 5V digital inputs corrupted the A/D readings on my analog input. I hope Philips will update the user manual to spel this sort of behaviour more clearly. Finding this out after one has manufactured a 10 layer PCB is not the most convenient. Regards Anton Erasmus > --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > > On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > > > > > I am using > the LPC2129 in an application and I need to know if the > > analog > port will be damaged if the port gets more than V3A at any > > time. > The manual says that the voltage must not be over V3A or the > > > analog input readings will be invalid. Does that mean that they will > > > be invalid forever or only for the moment where the voltage is over > > > V3A? > > > > The pins are supposed to be 5V tolerant, hence my > assumption is that they should > be able to handle up to 5V. The A/D > converter is designed, so that it's full range is > only V3A volts, > hence an input above this voltage will give invalid A/D results. > > > Regards > Anton Erasmus > > > > > > ------------------------ Yahoo! > Groups Sponsor > > --------------------~--> Make a clean sweep of > pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get > it for free! > > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM <http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM> > > > ----------------------------------------------------------------- > ---~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > -- > A J Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> $9.95 domain names from Yahoo!. Register > anything. http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM <http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM> > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus Yahoo! Groups Links Yahoo! Groups Sponsor ADVERTISEMENT <http://us.ard.yahoo.com/SIG=129h9etg0/M=298184.5639630.6699735.3001176/D=gr oups/S=1706554205:HM/EXP=1101848354/A=2434971/R=0/SIG=11eeoolb0/*http://www. netflix.com/Default?mqso=60185400> click here <http://us.adserver.yahoo.com/l?M=298184.5639630.6699735.3001176/D=groups/S= :HM/A=2434971/rand=708020608> _____ Yahoo! Groups Links * To visit your group on the web, go to: http://groups.yahoo.com/group/lpc2000/ <http://groups.yahoo.com/group/lpc2000/> * To unsubscribe from this group, send an email to: lpc2000-unsubscribe@yahoogroups.com <mailto:lpc2000-unsubscribe@yahoogroups.com?subject=Unsubscribe> * Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service <http://docs.yahoo.com/info/terms/> . [Non-text portions of this message have been removed]
2004-11-30 by James Dabbs
I'm trying to squeeze a few more cycles out of one my LPC, and I'm looking for an optimized way to divide by 10 and leave the quotient and remainder. I.e., a faster way to do this: #define DIVIDE_BY_10(D, Q, R) Q=D/10; R=D%10; // D is unsigned long [It's to support a prehistoric data collection system, writing LOTS of BCD ASCII data to a SRAM cartridge.] Any info would be appreciated. Thanks.
2004-11-30 by Lasse Madsen
Hi Art Well it's not a bug it's just a basic mistake... Normally one would take the /SS pin for an output to drive Slave Select lines but this is not the case when the processor is configured as a master then this pin must be forced high by hardware requiring another pin for /SS activity... I've was an inch close of ordering a prototype PCB with that mistake Money don't grow on trees but great advice from other users helps the tree not being picked to hard :) Regards Lasse M.
-----Original Message----- From: Messal, Art [mailto:art.messal@...] Sent: 29. november 2004 22:45 To: 'lpc2000@yahoogroups.com' Subject: RE: [lpc2000] LPC Series - Good things gone bad... Hi Lasse, What is the "SPI master = Hardware High on SS pin" bug that you mentioned? thanks, -Art- _____ From: Lasse Madsen [mailto:lasse.madsen@...] Sent: Monday, November 29, 2004 1:59 PM To: lpc2000@yahoogroups.com Subject: [lpc2000] LPC Series - Good things gone bad... I think that we ought to make a list of all the "mistakes" that is generally made with these controllers. There are a lot of people (my self included) that fell for all the nice features but when you look closer and pay the amount of money for PCB's you find out where the sun don't shine ... Would it be possible to summarize a list on the group page of the common mistakes etc? For instance: SPI master = Hardware High on SS pin! And very important Bootloader behaviour! What does it do and how can it be worked around as someone has talked about and what these issues mean from a hardware point of view etc. And apparently (maybe) There's another thing that the ADC can't stand 5V on a pin that Is related to it (What kind of sick cr*p is that?) I fancy these controllers don't get me wrong but one can easily be _very_ upset using _alot_ of time and money making PCB's and then finding out all the beginner mistakes ... Lets make a list of beginner mistakes etc. I'll help people help them selves. Who's in and who's out :) Regards Lasse -----Original Message----- From: Anton Erasmus [mailto:antone@...] Sent: 29. november 2004 21:44 To: lpc2000@yahoogroups.com Subject: Re: [lpc2000] Re: LPC2129, analog port question. On 29 Nov 2004 at 17:53, Richard wrote: > > > Note: 5V on any pin that CAN be configured as an ADC input will > corrupt the ADC conversion values. > At least now I know why my attempt to use only one A/D channel, and the rest as 5V digital inputs corrupted the A/D readings on my analog input. I hope Philips will update the user manual to spel this sort of behaviour more clearly. Finding this out after one has manufactured a 10 layer PCB is not the most convenient. Regards Anton Erasmus > --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > > On 29 Nov 2004 at 12:29, digtalfreak wrote: > > > > > > > I am using > the LPC2129 in an application and I need to know if the > > analog > port will be damaged if the port gets more than V3A at any > > time. > The manual says that the voltage must not be over V3A or the > > > analog input readings will be invalid. Does that mean that they will > > > be invalid forever or only for the moment where the voltage is over > > > V3A? > > > > The pins are supposed to be 5V tolerant, hence my > assumption is that they should > be able to handle up to 5V. The A/D > converter is designed, so that it's full range is > only V3A volts, > hence an input above this voltage will give invalid A/D results. > > > Regards > Anton Erasmus > > > > > > ------------------------ Yahoo! > Groups Sponsor > > --------------------~--> Make a clean sweep of > pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get > it for free! > > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM <http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM> > > > ----------------------------------------------------------------- > ---~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > -- > A J Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> $9.95 domain names from Yahoo!. Register > anything. http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM <http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM> > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus Yahoo! Groups Links Yahoo! Groups Sponsor ADVERTISEMENT <http://us.ard.yahoo.com/SIG=129h9etg0/M=298184.5639630.6699735.3001176/D=gr oups/S=1706554205:HM/EXP=1101848354/A=2434971/R=0/SIG=11eeoolb0/*http://www. netflix.com/Default?mqso=60185400> click here <http://us.adserver.yahoo.com/l?M=298184.5639630.6699735.3001176/D=groups/S= :HM/A=2434971/rand=708020608> _____ Yahoo! Groups Links * To visit your group on the web, go to: http://groups.yahoo.com/group/lpc2000/ <http://groups.yahoo.com/group/lpc2000/> * To unsubscribe from this group, send an email to: lpc2000-unsubscribe@yahoogroups.com <mailto:lpc2000-unsubscribe@yahoogroups.com?subject=Unsubscribe> * Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service <http://docs.yahoo.com/info/terms/> . [Non-text portions of this message have been removed] Yahoo! Groups Links
2004-11-30 by Karl Olsen
--- In lpc2000@yahoogroups.com, "James Dabbs" <jdabbs@t...> wrote: > > I'm trying to squeeze a few more cycles out of one my LPC, and I'm > looking for an optimized way to divide by 10 and leave the quotient and > remainder. I.e., a faster way to do this: > > #define DIVIDE_BY_10(D, Q, R) Q=D/10; R=D%10; // D is unsigned long Check out "How to optimize for the Pentium family of microprocessors" http://www.agner.org/assem/pentopt.pdf The section "Division (all processors)" has a smart algorithm for converting division by a constant to a cheaper multiplication and shift. Karl Olsen
2004-11-30 by Paul Curtis
All, The division by multiplication technique is well known, but does not generate a remainder--you need to compute the remainder by multiplying and subtracting. In essence, you're multiplying by the reciprocal of 10 expressed (for instance) as 2^32/10 or 2^16/10, but it *does* require that you can multiply two n-bit numbers to produce a 2n-bit product and then take the high n bits of that product--and you need to account for the error term. Some processors don't do this in hardware. Alternatively, Karl, try using the types div_t, ldiv_t and the functions div and ldiv that deliver both quotient and remainder for you. Note that the operation of div_t is consistent across all implementations of C, whereas / and % are *not*. -- Paul.
> -----Original Message----- > From: Karl Olsen [mailto:kro@...] > Sent: 30 November 2004 08:01 > To: lpc2000@yahoogroups.com > Subject: [lpc2000] Re: Slightly OT -- Fast Divide By 10 > > > > > --- In lpc2000@yahoogroups.com, "James Dabbs" <jdabbs@t...> wrote: > > > > I'm trying to squeeze a few more cycles out of one my LPC, and I'm > > looking for an optimized way to divide by 10 and leave the quotient > and > > remainder. I.e., a faster way to do this: > > > > #define DIVIDE_BY_10(D, Q, R) Q=D/10; R=D%10; // D is unsigned long > > > Check out "How to optimize for the Pentium family of microprocessors" > http://www.agner.org/assem/pentopt.pdf > > The section "Division (all processors)" has a smart algorithm for > converting division by a constant to a cheaper multiplication and > shift. > > Karl Olsen > > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> > $9.95 domain names from Yahoo!. Register anything. > http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM > -------------------------------------------------------------- > ------~-> > > > Yahoo! Groups Links > > > > > > > >
2004-11-30 by Paul Curtis
...and, just to say, this is also in the book "Hacker's Delight" (of which I have a copy). There is a slightly different version of this code in Hacker's Delight. It's quite a nice book for bit-twiddlers, but many of the examples in it expect that multiplication and shifting can be done very quickly, and it's not always true that that happens on some microprocessors. The ARM is, of course, lucky to be able to shift quickly. The code for the book is here: http://www.hackersdelight.org/HDcode.htm Take a look at "Figs. 10-2 and 10-3. Computing the magic number for unsigned division." Or just browse around, it's interesting. -- Paul Curtis, Rowley Associates Ltd http://www.rowley.co.uk CrossWorks for MSP430, ARM, and (soon) Atmel AVR processors
> -----Original Message----- > From: Karl Olsen [mailto:kro@post3.tele.dk] > Sent: 30 November 2004 08:01 > To: lpc2000@yahoogroups.com > Subject: [lpc2000] Re: Slightly OT -- Fast Divide By 10 > > > > --- In lpc2000@yahoogroups.com, "James Dabbs" <jdabbs@t...> wrote: > > > > I'm trying to squeeze a few more cycles out of one my LPC, and I'm > > looking for an optimized way to divide by 10 and leave the quotient > and > > remainder. I.e., a faster way to do this: > > > > #define DIVIDE_BY_10(D, Q, R) Q=D/10; R=D%10; // D is unsigned long > > > Check out "How to optimize for the Pentium family of microprocessors" > http://www.agner.org/assem/pentopt.pdf > > The section "Division (all processors)" has a smart algorithm > for converting division by a constant to a cheaper > multiplication and shift. > > Karl Olsen > > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> > $9.95 domain names from Yahoo!. Register anything. > http://us.click.yahoo.com/J8kdrA/y20IAA/yQLSAA/dN_tlB/TM > -------------------------------------------------------------- > ------~-> > > > Yahoo! Groups Links > > > > > > > > >
2004-11-30 by R M
You must be new to the low-cost micro controller scene because with most low-end micros (in my experience) the SS# pin is not normally driven by the master - it, or any other GPIO for that matter is driven by software to select a slave. This is not a Philips LPC thing - the AVR and many other micros act the same way. If you wanted a hardware controlled SS# peripheral then look at a more complex/expensive device like the Coldfire 5282 which has a multi channel SPI peripheral that bangs the SS# pins for you. However in my opinion, it is better to have software control over the SS# pin since it would otherwise preclude multiple slaves when only one SS# pin is provided (as is the case with low-end micros). As for the ADC on the LPC, I don't have any experience with this peripheral. However it is not uncommon for the inputs to not be 5V tolerant (if that what your rant was about) in mixed signal devices, or at least in peripherals/pins that interface with mixed signal silicon. What exactly is wrong with the boot-loader? I like the standard and easy to use API it provides to the Flash. I also like the ability to have a fresh part on a PCB programmable without the need for JTAG (as do my verification and production peers).
2004-11-30 by Bill Knight
The problem (annoyance) with the LPC is that if you want to operate the SPI in Master mode only, the SS pin must still be configured for SPI use and tied high. It cannot be configured as a GPIO pin and used for a different purpose. Regards -Bill Knight http://www.theARMPatch.com On Tue, 30 Nov 2004 17:43:07 -0000, R M wrote: You must be new to the low-cost micro controller scene because with most low-end micros (in my experience) the SS# pin is not normally driven by the master - it, or any other GPIO for that matter is driven by software to select a slave. This is not a Philips LPC thing - the AVR and many other micros act the same way. If you wanted a hardware controlled SS# peripheral then look at a more complex/expensive device like the Coldfire 5282 which has a multi channel SPI peripheral that bangs the SS# pins for you. However in my opinion, it is better to have software control over the SS# pin since it would otherwise preclude multiple slaves when only one SS# pin is provided (as is the case with low-end micros). As for the ADC on the LPC, I don't have any experience with this peripheral. However it is not uncommon for the inputs to not be 5V tolerant (if that what your rant was about) in mixed signal devices, or at least in peripherals/pins that interface with mixed signal silicon. What exactly is wrong with the boot-loader? I like the standard and easy to use API it provides to the Flash. I also like the ability to have a fresh part on a PCB programmable without the need for JTAG (as do my verification and production peers).
2004-11-30 by R M
I wasn't aware of that limitation - that is an interesting issue. --- In lpc2000@yahoogroups.com, "Bill Knight" <BillK@t...> wrote:
> The problem (annoyance) with the LPC is that if you want to operate the > SPI in Master mode only, the SS pin must still be configured for SPI use > and tied high. It cannot be configured as a GPIO pin and used for a > different purpose. > > Regards > -Bill Knight > http://www.theARMPatch.com
2004-11-30 by Lasse Madsen
Hi, >You must be new to the low-cost micro controller scene because with >most low-end micros (in my experience) the SS# pin is not normally >driven by the master - it, or any other GPIO for that matter is driven >by software to select a slave. This is not a Philips LPC thing - the >AVR and many other micros act the same way. I've been working with AVR for several years and I can tell you that the AVR controllers does NOT need the pin to be forced high by hardware to act as a master. What you are describing is not possible for the LPC2106... the /SS pin *must* be tied high on hardware for the processor to act as a master then you will need another pin to act as a /SS pin where in an AVR and a lot of other processors you can control the actual /SS pin in software like you mentioned. >However in my opinion, it is better to have software control over the >SS# pin since it would otherwise preclude multiple slaves when only >one SS# pin is provided (as is the case with low-end micros). Yes but not the case on Philips LPC2106 ! >As for the ADC on the LPC, I don't have any experience with this >peripheral. However it is not uncommon for the inputs to not be 5V >tolerant (if that what your rant was about) in mixed signal devices, >or at least in peripherals/pins that interface with mixed signal >silicon. Philip write that every GPIO pin is 5V tolerant if what I've been hearing is correct then if you have some >3V digital TTL input on a pin that shares the ADC your result on another ADC pin will be incorrect and if that's true well then Philips should stop making their apprentice design their processors ! >What exactly is wrong with the boot-loader? I like the standard and >easy to use API it provides to the Flash. I also like the ability to >have a fresh part on a PCB programmable without the need for JTAG (as >do my verification and production peers). I didn't say that there was something wrong with the bootloader but I've been hearing a lot of bootloader related problems lately especially security related postings and I would just like to see this explained more thoroughly in a document instead of browsing the message archives... That's why I suggested that we should make a list that new comers could refer to. Best regards Lasse
-----Original Message----- From: R M [mailto:rodlist@...] Sent: 30. november 2004 18:43 To: lpc2000@yahoogroups.com Subject: [lpc2000] Re: LPC Series - Good things gone bad... If you wanted a hardware controlled SS# peripheral then look at a more complex/expensive device like the Coldfire 5282 which has a multi channel SPI peripheral that bangs the SS# pins for you. Yahoo! Groups Links
2004-11-30 by Lasse Madsen
Hi R.M. >I wasn't aware of that limitation - that is an interesting issue. Now that's exactly why I proposed a list! Because there are many "limitations" on this processor and one can spend a fortune finding them manually. Regards Lasse Madsen
2004-11-30 by R M
> I've been working with AVR for several years and I can tell you that the AVR > controllers does NOT need the pin to be forced high by hardware to act as a > master. > > What you are describing is not possible for the LPC2106... the /SS pin > *must* be tied high on hardware for the processor to act as a master then > you will need another pin to act as a /SS pin where in an AVR and a lot of > other processors you can control the actual /SS pin in software like you > mentioned. Bill just enlightened me with the issue you were describing. I wasn't aware that the SS# pin had to be used irrelevant if there was a slave that needed to be selected. Yes this is indeed an issue. > Philip write that every GPIO pin is 5V tolerant if what I've been hearing is > correct then if you have some >3V digital TTL input on a pin that shares the > ADC your result on another ADC pin will be incorrect and if that's true well > then Philips should stop making their apprentice design their processors ! Interesting - I wasn't aware of that problem either (again I haven't used the ADC on the LPC series).
2004-11-30 by Scott Newell
At 12:07 PM 11/30/2004 , Lasse Madsen wrote: > >Hi R.M. > >>I wasn't aware of that limitation - that is an interesting issue. > >Now that's exactly why I proposed a list! > >Because there are many "limitations" on this processor and one can spend a >fortune finding them manually. Does the LPC2k UART support an interrupt on transmit complete? That's really nice to have if you need driver control for RS-485. I got stung on the H8/Tiny--it has a TC interrupt, but it fires on the last data bit, not the stop bit. -- newell
2004-11-30 by Robert Adsett
At 12:28 PM 11/30/04 -0600, you wrote:
>At 12:07 PM 11/30/2004 , Lasse Madsen wrote:
> >
> >Hi R.M.
> >
> >>I wasn't aware of that limitation - that is an interesting issue.
> >
> >Now that's exactly why I proposed a list!
> >
> >Because there are many "limitations" on this processor and one can spend a
> >fortune finding them manually.
>
>Does the LPC2k UART support an interrupt on transmit complete? That's
>really nice to have if you need driver control for RS-485. I got stung on
>the H8/Tiny--it has a TC interrupt, but it fires on the last data bit, not
>the stop bit.
It just implements the standard 16550 interrupts so it only interrupts on
buffer empty. You can get the same timing by listening to your own
transmission. The receive interrupt will occur on transmit complete. Hmm,
CTI timing will come into play so maybe it only works if you can tolerate
that delay before turning the line around. Messy, messy. I've always used
an interpacket delay with a fixed minimum value.
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-30 by Robert Adsett
At 06:04 PM 11/30/04 +0000, you wrote: >I wasn't aware of that limitation - that is an interesting issue. > >--- In lpc2000@yahoogroups.com, "Bill Knight" <BillK@t...> wrote: > > The problem (annoyance) with the LPC is that if you want to operate the > > SPI in Master mode only, the SS pin must still be configured for SPI use > > and tied high. It cannot be configured as a GPIO pin and used for a > > different purpose. > > > > Regards > > -Bill Knight > > http://www.theARMPatch.com That one is actually mentioned in Philips own FAQ. Like most manufacturers FAQ its mostly marketing fluff rather than real information but sometimes marketing flubs and lets a real answer slip through ;) Robert
2004-11-30 by Robert Adsett
At 07:04 PM 11/30/04 +0100, you wrote:
> >What exactly is wrong with the boot-loader? I like the standard and
> >easy to use API it provides to the Flash. I also like the ability to
> >have a fresh part on a PCB programmable without the need for JTAG (as
> >do my verification and production peers).
>
>I didn't say that there was something wrong with the bootloader but I've
>been hearing a lot of bootloader related problems lately especially security
>related postings and I would just like to see this explained more thoroughly
>in a document instead of browsing the message archives...
I'm not sure I'd call that a bootloader issue, although the bootloader does
relate to it is also vulnerable to JTAG investigation.
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-30 by Robert Adsett
At 07:07 PM 11/30/04 +0100, you wrote:
>Hi R.M.
>
> >I wasn't aware of that limitation - that is an interesting issue.
>
>Now that's exactly why I proposed a list!
>
>Because there are many "limitations" on this processor and one can spend a
>fortune finding them manually.
I got tired of suggesting someone else start the darn thing :)
I've set up a page on the ARMuC Wiki (that's what it's there for after all)
and added a few items. Please add to and edit this I'm hardly the be-all
and know-all of the various ins and outs and if nothing else could use the
help just getting the typing done.
http://www.open-research.org.uk/ARMuC/index.cgi?LPC2100Tips
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-30 by microbit
Hi Scott, > Does the LPC2k UART support an interrupt on transmit complete? That's > really nice to have if you need driver control for RS-485. I got stung on > the H8/Tiny--it has a TC interrupt, but it fires on the last data bit, not > the stop bit. Or RF amongst other media for that matter. This can be an important issue at times. A completely transparent INT driven media transaction can be ruined by having to "poll" for the last stop bit having been shifted out before you turn off the media .... (That H8/Tiny must have been fun and games on RS485, collison galore !) Personally I dislike vectors that share Rx and TX - still, could be worse, you could be on a PIC with 1 ISR only :-) -- Kris
2004-12-01 by microbit
Ah, OK. I thought it might have caused bad timing issues on the multi drop, depends on how fast your RS485 was running I guess, and also the specfic app. > It wasn't bad at all--left alone, it would shut the driver down in the stop > bit. The bias resistors forced the line into an idle state anyway, which > hid the bug for a little while. > > -- > newell > > > Yahoo! Groups Sponsor > ADVERTISEMENT > > > > > > -------------------------------------------------------------------------- ------
> Yahoo! Groups Links > > a.. To visit your group on the web, go to: > http://groups.yahoo.com/group/lpc2000/ > > b.. To unsubscribe from this group, send an email to: > lpc2000-unsubscribe@yahoogroups.com > > c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. > >
2004-12-01 by Scott Newell
At 10:09 AM 12/1/2004 +1100, microbit wrote: > >This can be an important issue at times. A completely transparent INT driven >media transaction can be ruined by having to "poll" for the last stop bit >having >been shifted out before you turn off the media .... >(That H8/Tiny must have been fun and games on RS485, collison galore !) It wasn't bad at all--left alone, it would shut the driver down in the stop bit. The bias resistors forced the line into an idle state anyway, which hid the bug for a little while. -- newell
2004-12-01 by Rick Collins
--- In lpc2000@yahoogroups.com, Robert Adsett <subscriptions@a...> wrote: > A FAQ/ useful notes list like this is a good idea. I'd suggest the ARMuC > wiki if onlly because it would be an easy way to collaborate on such a list. There is a wiki at http://www.open-research.org.uk/ARMuC/ It covers many ARM MCUs including the LPC devices. Please visit and add to it.
2004-12-01 by Clyde Stubbs
On Mon, Nov 29, 2004 at 09:15:49PM -0500, James Dabbs wrote: > I'm trying to squeeze a few more cycles out of one my LPC, and I'm > looking for an optimized way to divide by 10 and leave the quotient and > remainder. I.e., a faster way to do this: Take a look at the library functions div() and ldiv(). Clyde -- Clyde Stubbs | HI-TECH Software Email: clyde@... | Phone Fax WWW: http://www.htsoft.com/ | USA: (408) 490 2885 (408) 490 2885 PGP: finger clyde@... | AUS: +61 7 3552 7777 +61 7 3552 7778 --------------------------------------------------------------------------- HI-TECH C: compiling the real world.
2004-12-23 by lp2000c
The LPC213x has 16 pins which CAN be configured as an ADC input. Does this mean that I can't apply more than 3.3V to any of these pins, without corrupting ADC operation? --- In lpc2000@yahoogroups.com, "Richard" <richas@y...> wrote:
> > Note: 5V on any pin that CAN be configured as an ADC input will > corrupt the ADC conversion values. > > Richard
2004-12-26 by Anton Erasmus
On 23 Dec 2004 at 23:13, lp2000c wrote: > > > The LPC213x has 16 pins which CAN be configured as an ADC input. Does > this mean that I can't apply more than 3.3V to any of these pins, > without corrupting ADC operation? On a design I did, where I only wanted to use one of the ADC inputs, and connected the other lines to 5V devices, I got absolutely garbage readings even though the actual AD channel I was using never went above 3.3V. So it appears that the statement below is correct. Regards Anton Erasmus > --- In lpc2000@yahoogroups.com, "Richard" <richas@y...> wrote: > > > > Note: 5V on any pin that CAN be configured as an ADC input will > > corrupt the ADC conversion values. > > > > Richard > > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus
2004-12-27 by lpc2100_fan
Hi as far as I know the previous LPC devices had this problem but the LPC2130 series does not. I heard it somewhere but have not had the change to verify it. Bob --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > On 23 Dec 2004 at 23:13, lp2000c wrote: > > > > > > > The LPC213x has 16 pins which CAN be configured as an ADC input. Does > > this mean that I can't apply more than 3.3V to any of these pins, > > without corrupting ADC operation? > > On a design I did, where I only wanted to use one of the ADC inputs, and connected > the other lines to 5V devices, I got absolutely garbage readings even though the actual > AD channel I was using never went above 3.3V. So it appears that the statement below
> is correct. > > Regards > Anton Erasmus > > > > --- In lpc2000@yahoogroups.com, "Richard" <richas@y...> wrote: > > > > > > Note: 5V on any pin that CAN be configured as an ADC input will > > > corrupt the ADC conversion values. > > > > > > Richard > > > > > > > > > > > > > > ------------------------ Yahoo! Groups Sponsor > > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > --------------------------------------------------------------------~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > -- > A J Erasmus
2004-12-27 by Lasse Madsen
I would be nice to see the actual ADC/IO "Schematic" so one could see what on earth is going on... Regards lasse
-----Original Message----- From: lpc2100_fan [mailto:lpc2100_fan@...] Sent: 27. december 2004 02:03 To: lpc2000@yahoogroups.com Subject: [lpc2000] Re: LPC213x, analog port question. Hi as far as I know the previous LPC devices had this problem but the LPC2130 series does not. I heard it somewhere but have not had the change to verify it. Bob --- In lpc2000@yahoogroups.com, "Anton Erasmus" <antone@s...> wrote: > On 23 Dec 2004 at 23:13, lp2000c wrote: > > > > > > > The LPC213x has 16 pins which CAN be configured as an ADC input. Does > > this mean that I can't apply more than 3.3V to any of these pins, > > without corrupting ADC operation? > > On a design I did, where I only wanted to use one of the ADC inputs, and connected > the other lines to 5V devices, I got absolutely garbage readings even though the actual > AD channel I was using never went above 3.3V. So it appears that the statement below > is correct. > > Regards > Anton Erasmus > > > > --- In lpc2000@yahoogroups.com, "Richard" <richas@y...> wrote: > > > > > > Note: 5V on any pin that CAN be configured as an ADC input will > > > corrupt the ADC conversion values. > > > > > > Richard > > > > > > > > > > > > > > ------------------------ Yahoo! Groups Sponsor > > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > --------------------------------------------------------------------~- > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > -- > A J Erasmus Yahoo! Groups Links
2004-12-27 by lp2000c
What chip was that? We know the problem exists on 2114/2124. The question is whether it exists on 213x. > On a design I did, where I only wanted to use one of the ADC inputs, and connected > the other lines to 5V devices, I got absolutely garbage readings even though the actual > AD channel I was using never went above 3.3V. So it appears that the statement below > is correct. > > Regards > Anton Erasmus > > > > --- In lpc2000@yahoogroups.com, "Richard" <richas@y...> wrote: > > > > > > Note: 5V on any pin that CAN be configured as an ADC input will > > > corrupt the ADC conversion values. > > > > > > Richard > > > > > > > > > > > > > > ------------------------ Yahoo! Groups Sponsor > > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > ------------------------------------------------------------------ --~-
> > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > -- > A J Erasmus
2004-12-28 by Anton Erasmus
On 27 Dec 2004 at 15:23, lp2000c wrote: > What chip was that? We know the problem exists on 2114/2124. The > question is whether it exists on 213x. This was using an LPC2292 chip. > > On a design I did, where I only wanted to use one of the ADC > inputs, and connected > > the other lines to 5V devices, I got absolutely garbage readings > even though the actual > > AD channel I was using never went above 3.3V. So it appears that > the statement below > > is correct. > > > > Regards > > Anton Erasmus > > > > > > > --- In lpc2000@yahoogroups.com, "Richard" <richas@y...> wrote: > > > > > > > > Note: 5V on any pin that CAN be configured as an ADC input will > > > > corrupt the ADC conversion values. > > > > > > > > Richard > > > > > > > > > > > > > > > > > > > > > ------------------------ Yahoo! Groups Sponsor > > > --------------------~--> Make a clean sweep of pop-up ads. Yahoo! > > > Companion Toolbar. Now with Pop-Up Blocker. Get it for free! > > > http://us.click.yahoo.com/L5YrjA/eSIIAA/yQLSAA/dN_tlB/TM > > > ------------------------------------------------------------------ > --~- > > > > > > > > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > > > > > > > > > -- > > A J Erasmus > > > > > > ------------------------ Yahoo! Groups Sponsor > --------------------~--> $4.98 domain names from Yahoo!. Register > anything. http://us.click.yahoo.com/Q7_YsB/neXJAA/yQLSAA/dN_tlB/TM > --------------------------------------------------------------------~- > > > > > Yahoo! Groups Links > > > > > > > -- A J Erasmus