[Coco] DE-1 Clarification Please
Dave Philipsen
dave at davebiz.com
Sat Oct 28 23:04:53 EDT 2017
I think there are a few variables that come into play:
1) The Terasic control panel software leaves a lot to be desired. As I recall, it used a word-based addressing scheme as opposed to byte-based addressing which can be confusing.
2) Some people may not realize that the flash chip shouldn’t be erased multiple times. You can erase it once and then burn information into several different locations at different times. The control panel software only supports bulk erases and you can’t erase selective sectors.
3) The control panel software is very finicky in some cases. I have seen problems where the USB connection gets mangled and I’ve experienced issues with getting the USB driver to even work at all. I have one computer which works pretty much flawlessly and two others that simply refuse to work. I think it has to do with the Windows-installed driver.
That is why I was prompted to write the flash uploader awhile back which does not rely upon the USB connection or the control panel software for the transfer of data. Unfortunately, it wasn’t 100% debugged and it has some issues too although, in my opinion, seems to be a more stable approach than Terasic’s method.
Dave
> On Oct 28, 2017, at 3:37 PM, tim franklinlabs.com <tim at franklinlabs.com> wrote:
>
> I will program one of my resent acquisitions to see if it works. I know
> there's a document on the Yahoo page that precedes the website you
> mentioned. There may have been some changes. I know there was some
> discrepancies about the position of the ROM load address. I'll let you
> know later tonight or tomorrow.
>
> -Tim
>
> On October 28, 2017 at 3:21 PM James Ross wrote:
> Interesting Tim. Thanks for the*.sof advice, I did see a blurb about
> that in the manual. Wikipedia talks about most flash devices lasting
> ~ 100,000 cycles - but I've seen other places state as low as 10,000
> ... Hopefully these DE1's are closer to the 100k!
> Thanks for your previous answer on the programming order. I am going
> to play around w/ it soon and see if I can confirm your statement
> that it does not matter. I have no idea what was making my first 3
> or 4 tries fail. I am fairly certain that the guide at
> [1]http://www.brianholman.com/retrocompute/files/coco3fpga.html if
> followed verbatim does NOT result in a working system.
> Thanks again for the info.
> James
> ________________________________________
> From: Coco on behalf of tim franklinlabs.com
> Sent: Saturday, October 28, 2017 7:56 PM
> To: CoCoList for Color Computer Enthusiasts
> Subject: Re: [Coco] DE-1 Clarification Please
> ...Oh one more thing... Pay close attention on the number of times
> you
> program the flash *POF or ROM). Flash has a limit on the number of
> times you can write to it before it dies. It's write capability is
> very
> large but it does have a limit. Use SOF files if you are
> experimenting.
> Most flash cards do whats called "wear leveling" to increase the
> write
> cycles. I'm not sure if the DE1 flash supports wear leveling.
> -Tim
> On October 28, 2017 at 1:58 PM James Ross wrote:
> Hey Tim,
> Thanks for this clarification. Nice to know!!
> Since the day I got my DE1 to boot up to the 3 amigos, I have not
> had time to play w/ it again. However, I really want to get the
> process down pat, since I want to go back-and-forth re-programming
> it with tutorials and the CoCo3FPGA.
> Here is my question: does the order in which you program DE1 matter?
> For example, the last few messages on the Yahoo group a couple of us
> in that thread have determined that you must:
> 1) First program the FLASH w/ the ROMS
> 2) Next program the CoCo3 FPGA *.pof file
> Since if you reverse that order, programming the FLASH second, it
> will clobber the *.pof file?
> Since apparently to program the FLASH, regardless which control
> panel you use, the process must load a *.pof file -- w/ the older
> control panel you have to do it manually, w/ the newer control panel
> it does it automatically.
> *OR* am I/we incorrect in that assessment?
> James
> ________________________________________
> From: Coco on behalf of tim franklinlabs.com
> Sent: Tuesday, October 24, 2017 9:58 PM
> To: CoCoList for Color Computer Enthusiasts
> Subject: Re: [Coco] DE-1 Clarification Please
> The PROG/RUN switch enabled the write capabilities of the on board
> FPGA
> flash device. NOT the FLASH RAM. The FPGA doesn't retain anything
> when
> power is removed. You can send the progtam directly to the FPGA
> (.sof)
> file and run it using the programming tool. When power is removed,
> the
> FPGA forgets the program. There's a "Config Flash" chip connected to
> the FPGA that programs the FPGA automatically at power up. This
> flash
> is programed by placing the RUN/PROG switch in PROG mode and sending
> the POF file the same way you send the SOF file This is done with
> the
> programmer tool (not the control panel).
> The control panel allows you to write to the on board external
> flash. A
> totally seperate chip. This chip is NOT connected to the RUN/PROG
> switch therefore the switch has no affect on writing to it.
> On October 24, 2017 at 4:22 PM rcrislip wrote:
> Trying to follow the various procedures to flash/program the DE-1 is
> a
> little confusing for me. An answer to this question will clear the
> fog
> tremendously. When using the Control Panel, is the "prog/run" switch
> set to prog or run, likewise when using Quartus, what position is
> the
> switch to be in? I think I know, but conformation from you folks
> will help a lot. TIA
> RECrislip
> --
> Coco mailing list
> [1][1][2]Coco at maltedmedia.com
> [2][2][3]https://pairlist5.pair.net/mailman/listinfo/coco
> References 1. mailto:[3][4]Coco at maltedmedia.com 2.
> [4][5]https://pairlist5.pair.net/mailman/listinfo/coco
> --
> Coco mailing list
> [5][6]Coco at maltedmedia.com
> [6][7]https://pairlist5.pair.net/mailman/listinfo/coco
> --
> Coco mailing list
> [7][8]Coco at maltedmedia.com
> [8][9]https://pairlist5.pair.net/mailman/listinfo/coco
> References
> 1. mailto:[10]Coco at maltedmedia.com 2.
> [11]https://pairlist5.pair.net/mailman/listinfo/coco 3.
> mailto:[12]Coco at maltedmedia.com 4.
> [13]https://pairlist5.pair.net/mailman/listinfo/coco 5.
> mailto:[14]Coco at maltedmedia.com 6.
> [15]https://pairlist5.pair.net/mailman/listinfo/coco 7.
> mailto:[16]Coco at maltedmedia.com 8.
> [17]https://pairlist5.pair.net/mailman/listinfo/coco
> --
> Coco mailing list
> [18]Coco at maltedmedia.com
> [19]https://pairlist5.pair.net/mailman/listinfo/coco
> --
> Coco mailing list
> [20]Coco at maltedmedia.com
> [21]https://pairlist5.pair.net/mailman/listinfo/coco
>
> References
>
> 1. http://www.brianholman.com/retrocompute/files/coco3fpga.html
> 2. mailto:Coco at maltedmedia.com
> 3. https://pairlist5.pair.net/mailman/listinfo/coco
> 4. mailto:Coco at maltedmedia.com
> 5. https://pairlist5.pair.net/mailman/listinfo/coco
> 6. mailto:Coco at maltedmedia.com
> 7. https://pairlist5.pair.net/mailman/listinfo/coco
> 8. mailto:Coco at maltedmedia.com
> 9. https://pairlist5.pair.net/mailman/listinfo/coco
> 10. mailto:Coco at maltedmedia.com
> 11. https://pairlist5.pair.net/mailman/listinfo/coco
> 12. mailto:Coco at maltedmedia.com
> 13. https://pairlist5.pair.net/mailman/listinfo/coco
> 14. mailto:Coco at maltedmedia.com
> 15. https://pairlist5.pair.net/mailman/listinfo/coco
> 16. mailto:Coco at maltedmedia.com
> 17. https://pairlist5.pair.net/mailman/listinfo/coco
> 18. mailto:Coco at maltedmedia.com
> 19. https://pairlist5.pair.net/mailman/listinfo/coco
> 20. mailto:Coco at maltedmedia.com
> 21. https://pairlist5.pair.net/mailman/listinfo/coco
>
> --
> Coco mailing list
> Coco at maltedmedia.com
> https://pairlist5.pair.net/mailman/listinfo/coco
More information about the Coco
mailing list