Blog is now 4 years old

No kidding.

Resume is now 2 for 2

In March I submitted a resume just for kicks and I made it clear I wasn’t going to graduate until May 2011. I got an interview request anyway.

On Sunday, my advisor e-mails me and tells me to send him my resume asap. I now have an interview scheduled for next Wednesday. What’s funny is I spent all this time preparing for technical interviews and I really believe this interview is going to behavioral. I don’t think the interview is going to be pretty.

A few months from now, my resume’s batting average is going to drop like a rock after I spam it everywhere.

Testing is a pain

On Monday afternoon I found out I was going to have to go into the RF lab on Tuesday and probably for the rest of the week. My first reaction was to go running right away or else I won’t be able to run for the rest of the week.

It turns out I got more from the RF lab than I bargained for. I’ve been standing in the RF lab for most of the last three days and my legs are killing me. Well, at least I ain’t dying soon.

I think I need another two days in the RF lab and it’s going to suck.

Friend: I did something stupid. Guess what I did.
Me: You just bought a spanking nice car.
Friend: No.
Me: You have a girlfriend now.
Friend: No.
Me: Hmm…you bought something.
Friend: Yes.
Me: Electronics?
Friend: Yes.
Me: You bought an iPod?
Friend: Close.
Me: Uh oh, you bought an iPhone.
Friend: Yes.
Me: Shakes head.

During lunch…

Me: I know you’re looking at your iPhone right now but you don’t have the face to hold it above table level for me to see because you’re so ashamed.
Friend: Yeah.
Me: Good, because I don’t want to see that piece of junk.

Why I hate smartphones in general and iPhones in particular…and why owning one is the very definition of stupid.

1. Four words: Dog on a leash. If you own a smartphone, you are a dog on a leash. You are a slave. You are always on call. The fact that I can contact you anytime via e-mail means that the world doesn’t revolve around you. Think you’re a bigshot CEO who needs a smartphone? You’re nothing. You’re accountable to the shareholders of the company. You are a dog on a leash. That’s why you have a smartphone.

2. Texting is stupid. You know when you’re sitting on a bus and the person next to you is firing off a text message every 45 seconds? Do you think that person is talking about anything important? Heck no. I can understand the need for an occasional text message, but smartphones are built with keyboards and touchscreens to enable you to text like mad. Is there a point? Absolutely not. Do something else with your idle time rather than texting someone.

3. There’s nothing interesting on the Internet anyway. “Oh look, I can get the Internet on my phone!” So what!? What are you going to do with it? Watch crap on YouTube? Check out the live news feed on Facebook? Read my retarded blog? Being a slave to someone else via e-mail is one thing, but being a slave to the Internet is just lame…and it isn’t even that great to begin with. I’ll admit that there can be situations where having Internet access on your phone is actually useful for something, but I bet those situations constitute less than 1% of the bandwidth that you use.

4. The ownership of a smartphone (and in particular an iPhone with a service contract) in 90% of cases represents the inability to distinguish between needs and wants. It is the very poster child of the excessive consumerism that got us into this economic mess to begin with. And that’s why I hate it. I understand that for some people their job requires them to have a smartphone and I’m fine with that. That’s a legitimate need…but they’re still a dog on a leash. Unfortunately, a good percentage of the unemployed in this country whining about being unable to find a job send off $70 a month to AT&T for their retarded iPhone. Yeah, I understand they’re on a two year contract, but their ownership of a smartphone is correlated with all the other stupid financial decisions that they made. If you’re unemployed, cut costs like the plague, and unless you’ve done that, don’t whine.

Lastly, here’s a survey on cell phone use, and while I would like to voice my opinions on the implications of this study, it is probably better that I keep it to myself.

Okay, I haven’t actually been that lazy. I’m referring to when I actually want to graduate. I previously mentioned that finishing in December is technically doable, but realistically speaking it’s more like February or March. Now I’m starting to think about taking it easy, finishing in May, and then starting a new job in August. Yeah, that’s right, stretch out the Ph.D. for a few more months and then sit on my butt and do nothing for another two months before I start working. For some reason, that actually sounds very appealing. I need to snap out of it.

Future Self Kicking

I’ve already been kicking myself for selling my US Treasuries too soon…in a few months I think I am going to kick myself for not shorting US Treasuries. Seriously, this chart can turn around fast.

Be nimble.

It’s not that I couldn’t handle the truth, I simply didn’t believe that the derivation was correct. Intuitively speaking, there’s no physical way the derivation could have ended up like that. My understanding of physics and the math simply did not match. Until the two match, you simply cannot believe the math (or the physics depending on your world view).

Unable to match the math with the physics, I went to the gym, came back to the office, bashed my head for a few hours, still couldn’t figure it out. I simply refused to believe my circuit would have more loss.

For dinner, I walked to Chipotle Mexican Grill. On the way there, I realized I had committed the ultimate noob mistake: power ratios are 10*log(ratio) and voltage ratios are 20*log(ratio). My derivation of my loss was from a power ratio and I correctly 10*logged that ratio; however my derivation of the alternatives were from a voltage ratio and I incorrectly 10*logged those ratios (resulting in lower loss that is not correct) when I should have 20*logged those ratios.

That was a pretty good scare but I’m glad it worked out okay. I’ll take a noob mistake that eats up half a day of work over the wrong design decision any day.

UPDATE the next day: Another scare today…

I wanted to see some old S-parameters in MATLAB today so I loaded the .m file containing the S-parameters and typed the following:

>> 20*log(abs(S(:,5,1)))

The S matrix was organized with the first index representing the frequency, the second index representing the first index of the S-parameter, and the 3rd index representing the second index of the S-parameter. So S(:,5,1) spits out S51 for all frequencies, I then take the absolute value, and then I 20*log it.

To my horror, the numbers made no sense. I checked the name of the file that I loaded to make sure it was the right one. It was the right one. Then why are my numbers all whacked up?

I went into a panic for about 2 minutes until I realized that log() in MATLAB is natural log and log10() is log base 10. Those punks at Mathworks need to call natural log ln() and log base 10 log().

So yeah, yesterday I was freaking out over 10*log and 20*log and today I was freaking out over the goofed up log() and log10() MATLAB functions. Man, I suck.

-27.2% month over month. That’s beyond bad. I’m afraid of checking bond market quotes. Every time I check I just want to cry.

The way things are looking now, I’ll graduate when the entire economy goes straight to hell.

• Discrete Time Fourier Series in chapter 3
• All of chapter 5 on the Discrete Time Fourier Transform
• Chapters 6-8 which have significant overlap with other things I will study
• All of chapter 10 on the Z-transform

In short, in the interest of time, I made sure I knew all my continuous time junk well and skimped on the discrete time garbage. When you’re an analog circuit designer and you deal with the analog domain, and you have limited time to prepare for an interview, then you ignore all that digital domain crap. Watch that calculation bite me in the future…if not during my interview, then during my job.

Seriously, in the continuous time version of things, you have to integrate stuff…although I suck at that, it’s manageable. Now the discrete time version of things, crap…I have to take a freakin’ infinite sum. I’m not going to waste my time and memorize tables for various infinite sums…and if it’s just for N samples, heck I’m not going to write each one out. That’s crazy, man…screw dat. Man, who the heck thinks in discrete time?

I hate signals and systems, but this week I just had to hold my nose and swallow it. And even so, I just know enough so as to not embarrass myself. I’m going to have to find some time to do a second pass in the future.

Now that I’ve gotten device physics and signals and systems out of the way, I feel I’ve gotten over the halfway point in terms of interview preparation and it shouldn’t be as painful because I’ll be going over very familiar circuits material. The whole point of going over device physics and signals and systems first is that when I go over the circuits material, I will have a better background understanding from which to analyze things. There’s no point in bashing your head over the temperature dependence of your circuit if you don’t know your device physics and how your transistors work. There’s no point in trying to understand a mixer if your math background isn’t up to par.

Being a Ph.D. candidate adds pressure to the job hunt. The longer you fart around in academia, the dumber you become; unfortunately, people expect more out of you during an interview. If I simply had a Masters degree and I didn’t know something, no big deal. But no, since I’m a Ph.D., they expect more out of me…I have to be up to standard. My research no longer matters. I could have the most awesome journal paper, but if I look like a fool during an interview, then no dice.

From 4/18…

The layoffs may have slowed, but it ain’t getting better. Weekly unemployment claims seems to be stuck between 450k and 500k. If it breaks over 500k, you know the game is over.

5% doesn’t seem like much until you realize what kind of crap interest you get in a savings account these days.

Dude man, that’s not cool. Seller’s remorse can be very painful sometimes. But it is what it is.

1. I went out to lunch with 11 other people to a Spanish restaurant. Three of them (a husband, a wife, and a kid) were new to the city and when we sat down at the restaurant, they offered to pay for everyone. Of course the other 9 of us refused to let them pay. Of course they insisted and we argued back and forth about it until the topic dissipated. After everyone finished eating, the waiter placed the check on the table directly in front of me and 1 of the 3 who wanted to pay for everyone. (It was the wife.) I could have snatched it instantly, but I wanted to see how she would react first. She slowly picked it up off of the table and started to look to her right towards her husband who was the one who was presumably going to pay…when it was 4 inches off the table, I put my hand on the other side of the black check holder…for about half second she didn’t seem to notice that my hand was on it. In that half second, I thought about holding it tightly instead of yanking the whole thing away, but recognizing the severity of the situation, I decided to yank it away. After I yanked it away, the 9 of us took turns looking at the check figuring out how much each one of us should put in, and only until we had the cash for the 9 of us did we let the other 3 see it. They ended up taking the cash and carding it.

The moral of the story…if you want to pay for everyone and you have an idea that the other people won’t let you, then you don’t tell people that you’re going to pay until you have the check in your hand…and that no other hands are on it.

2. After lunch I went to Starbucks with a friend…

Friend: I’m paying for you. I have a Starbucks card.
Me: No, you’re not paying for me.
Friend: You can order first.
Me: No, you go first because you’re not paying for me.
Friend: Okay, I’ll go first, but I can wait.
Me: You’re not waiting. You’re not paying for me.

Barista (to friend): What can I get for you?
Friend: Can I get an iced coffee?

Barista starts preparing his order. Friend gives his Starbucks card to the cashier. Friend motions to me as if he’s waiting for my order before the cashier processes the order.

Me (to cashier): Hey, he’s not paying for me.
Barista (to me): What do you want?
Me (to barista): Hold on, this guy is not paying for me. I’m not telling you my order until he pays for his order. Go on, go pay first.

Cashier swipes my friend’s Starbucks card.

Me (to barista): Okay, I’ll have an iced caffe mocha.

Cashier swipes my friend’s Starbucks card a second time.

Me: Doh!
Friend: Hey, I come here very often. They know me. They’ll listen to me. They won’t listen to you.

The moral of the story…if you don’t want someone to pay for you, then you should go first.

Oh well. You can’t win every battle. 1 for 2 ain’t bad and I won the big one and lost the little one.

Here’s the question…

What’s the function of this circuit? Plot Vout vs. Vin. What’s the gain?

Feel free to put your answer in the comments.

UPDATE @ 11:13 PM EDT on 8/15/2010:

Some clarifications

1. Assume both PMOS devices are identically sized.
2. Assume the two resistors at the top are the same. Let’s call them R1.
3. Might as well call the resistor at the output R2.
4. The two current sources are the same. In the schematic they are ideal current sources, but really they are NMOS devices that are biased to pass a current I.

I will post the solution to the problem eventually, but first I want to see what people have to say about it.

I’ve also added a new category called “Interview Prep” and I am anticipating future posts like this.