Joyous times :)
Baggers
Posts: 3,019
My good mate and co-prop-head Coley paid me a visit today as he was in the area working, and he had with him his DE2 setup with him
I showed him my setup, which would program and verify the DE2 and even show a Prop with Ctrl-G in PNut, but wouldn't run the demo ;(
We tried the simplest test first, swapping his add on board with mine, which also didn't work
We then tried his with my add on board, which did work, proving my add on board worked which also meant it was possibly the DE2 which was at fault. which would mean shipping back to the Ken for fixing.
So we tried downloading the latest code, as mine wasn't the November pof release, as a last resource before packing it to be shipped.
And by some miracle it worked!
Now I don't know if swapping the add-on boards may have awoken it or what, but all is now working and I'm back to being able to play with Prop2 so can re-convert my driver to the new timings and instructions.
Watch this space
It's good to be back, I just wish Coley had called a week before Xmas, as I had two weeks off, and could of had an even greater Xmas
I showed him my setup, which would program and verify the DE2 and even show a Prop with Ctrl-G in PNut, but wouldn't run the demo ;(
We tried the simplest test first, swapping his add on board with mine, which also didn't work
We then tried his with my add on board, which did work, proving my add on board worked which also meant it was possibly the DE2 which was at fault. which would mean shipping back to the Ken for fixing.
So we tried downloading the latest code, as mine wasn't the November pof release, as a last resource before packing it to be shipped.
And by some miracle it worked!
Now I don't know if swapping the add-on boards may have awoken it or what, but all is now working and I'm back to being able to play with Prop2 so can re-convert my driver to the new timings and instructions.
Watch this space
It's good to be back, I just wish Coley had called a week before Xmas, as I had two weeks off, and could of had an even greater Xmas
Comments
Good news, and a bit mysterious too. Encouraging to know that a couple of really smart guys like you and Coley also go for the high-level troubleshooting efforts first, like me (not an engineer). Replace the board, replace the cable, blah blah. Though I can talk with Chip all the time about FPGAs, I still know so little about what can and can't be done in them, which is why I keep asking if we can synthesize the I/O pin in the same blob (and move it out of manual layout).
As for the service of these boards, we can take care of that for people now. We've signed an agreement with Terasic to sell and support them, so I can always repair/replace whatever you guys decide to detonate. Even Chip casually crossed a few pieces of metal on a DE2-115 a while back - making a $600 mistake in a few milliseconds. So, go about your programming business and know that the only inconvenience will be down time while you wait for a cross-shipped board.
We have seven DE2-115s in stock, just not shown on the web yet.
We've also just started our own design using the Cyclone V A-7. This Parallax FPGA will be more customized for all of our uses around P1, P2 and P3.
Greetings to Coley on my behalf.
Ken Gracey
Anyway, congratulations on resurrecting your DE2-115 board!
David, I was using source files from the same zip install, which I find most bizarre that it didn't work.
Either way, it's working now, which is all that matters just need a time machine now to go back to pre-christmas for two weeks off haha
Joyous times indeed:)
Ken.
Let's get those puppies moved over to the website! We can only buy it if you agree to sell it:) I am wondering about the future board. I don't mind buying both. There are things that I could do with the DE2-115 after it is retired, but I am
CURIOUS about that new board.
Best wishes for a productive and profitable New Year.
Rich
You got it, Rich. Just like you said in the iPad thread - we'll get off the pot!
And Baggers, I don't think you're alone when it comes to mysterious FPGA behavior. Talking with Chip I know he sometimes encounters things he may not be able to explain. And we all know he hates indeterminism, unexplainable behavior, etc.
Ken Gracey
This is much better than all the lamenting that has been going on here recently.
I hope we can do some serious testing on the current design before it goes to silicon. And no more major changes. Or even small ones from a user perspective.
Good point - is there some post-power on, but pre-code means (eg LEDs?) to show what version of Core the FPGA has booted with ?
I read through that thread this morning… not sure exactly how you get off THAT pot.
My theory is that Baggers hit the F11 when he should have hit F10:) and then just gave up for a month.
Not even small changes to fix bugs ?
The opcodes should be pretty much locked down now, baring fixes, but the SerDes and Counters need to be FPGA exercised fully, and so I'd expect a few more FPGA builds yet.
+1 Heater. I hope to have 1,000 of you loading FPGAs with P2 cores before we go to production.
Ken Gracey
That would be test coverage even the big guys can only envy
I'll take the pledge. I won't ask for any more instruction set changes for P2.
Edit: Well, maybe I'll hold off on the pledge until I see the next iteration! :-)
More likely a loose solder connection. Put in an untouchable shrine and burn incense every day.
Glad you're up and running Baggers.
Sounds like some of my relatives!
So if I'm supposed to be to be working on Robotics only projects in 2014 (resolution), then if I want to play along again, I'll need to build a DE0 P2 emulation controlled robot? Yikes!
Jazzed, David,
Yep. A Dalek infestation can be very troublesome. In the old days they could not get up the stairs but it's a bit more tricky now.
As far as PropGCC goes, once we have a stable instruction encoding then PropGCC can be updated and will probably stay working even with additional core changes as long as they don't involve yet another instruction encoding change.
Here: http://en.wikipedia.org/wiki/Dalek we see old and new Daleks, including the first ones that could make it up the stairs.
Thing about the original Dr Who was that they had special video effects that were clearly impossible at the time. And the theme tune was something that the young Jean Michel Jarre must have been inspired by.
Then of course all that time travel and a telephone box that's bigger on the inside than the outside was great stuff for a seven year old mind in 1963.
It must have been your dodgy coding that was to blame
Good to see you again matey and like I said there is no excuse now !
We took my 9 year old to see "The Day of the Doctor" back in November.
Before the movie she told us we were geeks.
She ended up loving the movie and they got me a TARDIS Christmas tree ornament and the DVD set of season 7.
We watched two episodes a night until we had seen the entire season and she wants to buy more seasons now!
C.W.
You have to start at the beginning: Doctor Who Original: Episode One, An Unearthly Child, Part 1 circa 1963:
http://www.youtube.com/watch?v=SIwjLrhBLsw
I'm sure you can find the rest-
Seems the VCR went obsolete before I thought I needed one, and then the DVD. Not sure where Blue Ray is now but I guess I will outlive that without one as well.
Who actually has time to watch 50 years of Dr Who in a ten zillion disc set?
Potatohead, possibly, and swapping with Coley's may have dislodged it? who knows. I'm just glad it's working again!
Heater, yeah some good positive vibes for P2 again, I was a bit shocked at the lament thread to be honest, but I won't go into it in my positive P2 thread haha
Rich, I tried many times on both F10 and F11, it wasn't that
Jazzed, thanks me too! haha
Coley, cheers bud haha yeah probably my dodgy code!
All: Because of the restrictions in the current fpgas, things have been found that otherwise would never be found, and suggestions. Just look at the work ozpropdev did with the single cog. Then the discussion over thanksgiving ultimately gave us 256KB hub and following that we have hubexec coming. These two changes alone make it a P3 disguised as a P2.
Obviously there has been a lot more work required for the Onsemi process, but the improvements seem well worth the change/delay. The P2, whatever its ultimate incarnation, has to find a niche where other micros cannot compete. The more capability, flexibility, etc that the P2 can provide will give us more possible niches where the P2 can compete. It may well be that one of the hobby projects gives rise to one of these niches, or inspires commercial users to give it a try. Either way, I am sure there are plenty of niches where the P2 will shine.
I already have my first P2 project(s) sort of mapped out.
Turn the FPGA off, switch to program, turn FPGA on, load the programmer, set the mode to active serial programming, add TOF file, then set the three boxes for program, verify, blank check, then click start, wait for it to do it's magic, then turn FPGA off, switch to run, turn it on, load the spin file into the latest PNut and press F10
Cluso99, Yeah totally, I'm running again, but I can't help that niggling question in my mind "why?" and I can guarantee I'm not the only one here on this forum, if it happened, they would have the same question stuck in their head even after it working again!
I also agree, with the awesome updates and 256KB hub ram and HUBEXEC making it P3 disguised as a P2 as you say
Now to go through my driver making it work with the new instructions haha