1.29.2010

Can the prime lending rate be only prime numbers?
----------------------------------------------------------------
Sent by Text Messaging from a mobile device.
Envoye par messagerie texte d'un appareil mobile.
----------------------------------------------------------------

1.25.2010

First day we have any weather, the train is late.
----------------------------------------------------------------
Sent by Text Messaging from a mobile device.
Envoye par messagerie texte d'un appareil mobile.
----------------------------------------------------------------

1.20.2010

If Jimmy cracks corn, and no one cares, why do I have a song about it stuck in my head?
----------------------------------------------------------------
Sent by Text Messaging from a mobile device.
Envoye par messagerie texte d'un appareil mobile.
----------------------------------------------------------------

1.18.2010

Start of what should be a good week.

So, as I alluded to on Saturday night, Friday I managed to get my firmware and software running on my FPGA, and got it to talk to my PC. A big issue was the communications interface between the FPGA board and my PC.

Communication is attained by writing text to a serial port on the FPGA board and then that signal is read into a PC using Hyperterminal. Here, the problem was two-fold. Firstly, we needed what is known as a null-modem adapter, a gadget which essentially takes the recieve and transmit pins on the connector and swaps them (so transmit on one device points to recieve on the other, and vice versa). The second was that one of my USB ports wasn't supplying enough power to the serial adapter in order for these signals to be read in (which was a real pain to debug considering I've been able to power things off it before). Anyways, it's done now, the hard part, that is. 

Right now, I'm just going to clean up the data acquisition software at which point I'm going to see if I can get the Ethernet running, a daunting task for sure, but at least now I've a proof-of-concept to work from.




1.16.2010

I need only write gigabit ethernet drivers now, and if I am lucky the default drivers will work fine.
----------------------------------------------------------------
Sent by Text Messaging from a mobile device.
Envoye par messagerie texte d'un appareil mobile.
----------------------------------------------------------------
It took two weeks, but now we have a slow (RS232) detector at my disposal. I am so excited.
----------------------------------------------------------------
Sent by Text Messaging from a mobile device.
Envoye par messagerie texte d'un appareil mobile.
----------------------------------------------------------------
I forgot to mention that I managed to complete my prototype detector yesterday. More to come on this.
----------------------------------------------------------------
Sent by Text Messaging from a mobile device.
Envoye par messagerie texte d'un appareil mobile.
----------------------------------------------------------------

1.08.2010

Incredibly productive week.

So, this week I noticed a huge bug in my FPGA compiling code. In a nutshell, it was removing the fact that my firmware was looking at 7 pins (corresponding to the 7 pixels of my camera) due to a typo I had made in the firmware code (literally, I forgot the word 'in' in two spots), and the firmware compiler didn't mind this, it figured it out on its own, but the higher level compiler for the embedded processing part of it.

That's fixed, and now I'm updating my netbook with the same compiler version as the powerhouse I've been using to make the bitstreams for the FPGA. Once that's done, I should be able to run downstairs and do a low-speed test (readout over Serial connection). That'll be my proof-of-concept.

Today is also the day of the Physics Department's talent show (ie, mostly good music, free beer). "The Redshift Blues Band", comprised of my supervisor (David Hanna on drums), the Chair of the department (Charles Gale), and the Dean of Science (Martin Grant). They are fantastic musicians, and so it'll definitely be a great end to the day.


1.04.2010

Fairly certain this isn't true.

In an attempt to rescue my failed file server, I came across this gem. 

Back from Boston, New Year's Day (or the lack thereof), beer making, first day back in the office.

So, Wednesday of last week we went to the Institute of Contemporary Art in Boston, which is a really interesting (the same meaning as in 'this food is interesting') place; no pictures allowed though, so I'll have to leave it at the fact that the neatest exhibit was a VW Beetle which had all its parts exploded out in three dimensions so you could see how it was made. Later that day we went walking around, saw the Harvard campus, and then we visited a local homebrew shop. 

While I was there, I picked up a Cornelius Keg, for use in my homebrewing beer. While I don't have a CO2 tank, pressure regulator, or any other supplies needed for kegging properly, the kegs there were about 40% cheaper then buying one in the Montreal region, so it was worth it to pick one up (and in retrospect, I almost wish I had gotten two). 

We didn't go to New York for New Year's finally, though, it was too damn cold. Instead, we drove home Thursday and took it easy that evening. Not being a very big fan of New Year's in general, I didn't mind not doing anything this year. 

Saturday was a pretty big day. I brewed two batches of beer, one Coffee Stout (which, should it come out well, I will christen "Dark Energy"), and my second batch of a coriander hefeweizen ("White Dwarf") which was great the first time I made it. They're bubbling away nicely in the brewery (I'll attach a picture as soon as I get home), and should be ready for bottling in a week or two. I do have one other batch, an "almost stout" (Porter?) which'll be ready for taste testing in a few days. Typically, bottle conditioning lasts 2 weeks. 

As for work today, I've put the finishing touches on my latest build code for my FPGA project. At slow speeds (serial) I should be able to count and read out the number of signals on one of 7 channels (corresponding to the 7 pixels of my camera). If this works, then I only have to pipe the stored values into the board's onboard DDR, and then fire it off in large bursts to my readout computer. This corresponds to the memory buffer being filled up and then read out (and wiped) via gigabit Ethernet in one burst. 

Anyways, I'm off to read about memory controllers and such. Tomorrow morning I'm going to do my first program tests, I want to have as much functionality ready as possible before I try working with the hardware.