@Leon: Interesting! Do you have more info about this on your site?
@waltc: thanks for the correction ... that's more reasonable indeed!
@Matt (mctrivia): Is crystal clocking the normal approach for these devices when you make your own boards?
@benty: looking forward to that LED flashing! ... considering what these chips can do, that's like carring a shoe box in a lorry :^) - but we have to start somewhere, eh?
Depends on speed of clock needed. A pll may be required for higher frequencies.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
propmod_us and propmod_1x1 are in stock. Only $30. PCB available for $5
Want to make projects and have Gadget Gangster sell them for you? propmod-us_ps_sd and propmod-1x1 are now available for use in your Gadget Gangster Projects.
Need to upload large images or movies for use in the forum. you can do so at uploader.propmodule.com for free.
My bad the Max II doesn't have the 100 write cycle limit its UFM that does. Some how I transposed the two when I was researching if I could use the MaxII as a EEPROM - you can but only about a 100 times.
After re-reading the MaxII device handbook I don't think there is a limit to how many times one can reprogram the chip.
This is wrong. The write limit is real and it is for all. The CFM and the UFM are two distinct and separately accessible areas of the same flash
In the CFM is stored the configuration so you can modify your circuit up to 100 times.
For the UFM, when erased it is preset to FFFF, when writing you are just changing "1" to "0": tho go back to "1" you need to erase. But every location can be written two times, of course the second write can change to "0" only the bits that remains as "1" during the first write cycle.
@Leon: thanks much for providing the blaster files. I do appreciate these helpful things you've made available for everyone!
RE Write cycles: OK - thanks for the clarification dMajo and waltc --- Is 100 writes pretty common for all these devices (not just MAX II's) ?
Do the more expensive, later families have more write cycles?
On the other hand, maybe using Xilinx XC95xx for experimenting is a better idea as at least one comes in PLCC - so once it's toast, pop it out for a new one?
This doesn't apply to FPGA's as they load their configuration each start up, but do they have some internal write limitations too?
Clearly, in production, you'd not *need* many cycles, but for experimenting, it seems lousy to burn up a demo board that fast. Maybe *that's* why the demo boards are so inexpensive?? Do you guys who do this for a living just buy them batches at a time?
Found this link from the UNM website, the course covers introduction to VHDL using CPLD/FPGA, has tutorials and everything. If you go into the "Past Versions" link, the 2004 course covers the Coolrunner II while the newer stuff covers the Spartan. Looks like it might be helpful
CounterRotatingProps said...
@Leon: thanks much for providing the blaster files. I do appreciate these helpful things you've made available for everyone!
RE Write cycles: OK - thanks for the clarification dMajo and waltc --- Is 100 writes pretty common for all these devices (not just MAX II's) ?
Do the more expensive, later families have more write cycles?
On the other hand, maybe using Xilinx XC95xx for experimenting is a better idea as at least one comes in PLCC - so once it's toast, pop it out for a new one?
This doesn't apply to FPGA's as they load their configuration each start up, but do they have some internal write limitations too?
Clearly, in production, you'd not *need* many cycles, but for experimenting, it seems lousy to burn up a demo board that fast. Maybe *that's* why the demo boards are so inexpensive?? Do you guys who do this for a living just buy them batches at a time?
cheers
If in your design you have eg registers, small sram ... you can write them how many times you want. The EPM is mainly a flash, sram and oscillator in the same package. Your circuit is saved in the CFM area of the flash. On power up your circuit setup is transferred from the flash to the sram. The only part that have a wear is the flash (so you cannot change the design more than 100 times) when running in the sram your circuit parts can be written how many times you want. Think of a counter: at every clock tick the value is updated (written) - not so good if after 100 clock periods your circuit doesn't work any more.
100 reprogram cycles are more than enough·for the product lifetime (future firmware upgrades) but I agree that is not so much during development.
Guys that do this for a living (not me[noparse]:)[/noparse] usually are (or became) skilled and know how to operate properly with the IDE so simulations and timing analyses are done and many troubles are solved virtually.
thanks, mate, for that link ... I am studying it now ... nice that got both coolrunner and spartan lessons.
@dMajo - OK that's helpful to understand the difference between CFM and EPM.
Anyone reading this thread that wants more advanced things to consider, from the practical aspects, might want to look at Matt's (mctrivia's) work on his "Super Prop":
thanks, mate, for that link ... I am studying it now ... nice that got both coolrunner and spartan lessons.
@dMajo - OK that's helpful to understand the difference between CFM and EPM.
Anyone reading this thread that wants more advanced things to consider, from the practical aspects, might want to look at Matt's (mctrivia's) work on his "Super Prop":
The EPM is the cpld (eg. epm240, epm570, epm1270...) containing a flash, a ram and the oscillator. The CFM is part of the flash (the other is the UFM) that, on startup, is clocked into the sram by the oscillator to configure it into your circuit.
If you do not mind the stand-by current a XC3S50AN costs some 7 €uro and have built-in flash. They are not instant-on but way cheaper and versatile. Collrunners are great but the pricetag may not yield, I think.
Comments
I use a USB Blaster now.
Leon
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle
Post Edited (Leon) : 8/24/2009 9:32:17 AM GMT
@waltc: thanks for the correction ... that's more reasonable indeed!
@Matt (mctrivia): Is crystal clocking the normal approach for these devices when you make your own boards?
@benty: looking forward to that LED flashing! ... considering what these chips can do, that's like carring a shoe box in a lorry :^) - but we have to start somewhere, eh?
thanks all,
- H
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
propmod_us and propmod_1x1 are in stock. Only $30. PCB available for $5
Want to make projects and have Gadget Gangster sell them for you? propmod-us_ps_sd and propmod-1x1 are now available for use in your Gadget Gangster Projects.
Need to upload large images or movies for use in the forum. you can do so at uploader.propmodule.com for free.
The files are available here:
www.leonheller.com/BB/Blaster.zip
Leon
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle
Post Edited (Leon) : 8/26/2009 6:28:40 PM GMT
In the CFM is stored the configuration so you can modify your circuit up to 100 times.
For the UFM, when erased it is preset to FFFF, when writing you are just changing "1" to "0": tho go back to "1" you need to erase. But every location can be written two times, of course the second write can change to "0" only the bits that remains as "1" during the first write cycle.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· Propeller Object Exchange (last Publications / Updates)
Post Edited (dMajo) : 8/27/2009 1:11:29 PM GMT
Thanks, I was right to begin with in assuming that the MaxII could only be modified about 100 times.
RE Write cycles: OK - thanks for the clarification dMajo and waltc --- Is 100 writes pretty common for all these devices (not just MAX II's) ?
Do the more expensive, later families have more write cycles?
On the other hand, maybe using Xilinx XC95xx for experimenting is a better idea as at least one comes in PLCC - so once it's toast, pop it out for a new one?
This doesn't apply to FPGA's as they load their configuration each start up, but do they have some internal write limitations too?
Clearly, in production, you'd not *need* many cycles, but for experimenting, it seems lousy to burn up a demo board that fast. Maybe *that's* why the demo boards are so inexpensive?? Do you guys who do this for a living just buy them batches at a time?
cheers
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
http://www.eece.unm.edu/vhdl/
100 reprogram cycles are more than enough·for the product lifetime (future firmware upgrades) but I agree that is not so much during development.
Guys that do this for a living (not me[noparse]:)[/noparse] usually are (or became) skilled and know how to operate properly with the IDE so simulations and timing analyses are done and many troubles are solved virtually.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· Propeller Object Exchange (last Publications / Updates)
Post Edited (dMajo) : 8/28/2009 7:09:09 AM GMT
thanks, mate, for that link ... I am studying it now ... nice that got both coolrunner and spartan lessons.
@dMajo - OK that's helpful to understand the difference between CFM and EPM.
Anyone reading this thread that wants more advanced things to consider, from the practical aspects, might want to look at Matt's (mctrivia's) work on his "Super Prop":
Open Source High Speed SRAM Module(AKA Super Prop)
http://forums.parallax.com/forums/default.aspx?f=15&m=375446
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Scroll to page bottom:
https://www.digilentinc.com/Products/Catalog.cfm?NavPath=2,400&Cat=10
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
· Propeller Object Exchange (last Publications / Updates)
"Spartan FPGA first device on [noparse][[/noparse]Altium's] NanoBoard 3000"
http://www.pldesignline.com/showArticle.jhtml?articleID=220000400
Cool looking board!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Visit some of my articles at Propeller Wiki:
MATH on the propeller propeller.wikispaces.com/MATH
pPropQL: propeller.wikispaces.com/pPropQL
pPropQL020: propeller.wikispaces.com/pPropQL020
OMU for the pPropQL/020 propeller.wikispaces.com/OMU