Justin Dunham

's journal about making things

The Pepsi Challenge

Filed under: Cooking Journal — Tags: , — Justin Dunham on January 25, 2011

The results.

How consumers perceive tastes when they don't know what they're drinking.

How consumers perceive tastes when they're influenced by brands.

Remember the Pepsi Challenge? (Well, perhaps they still do it, but do you remember it?)

The deal was that people would take a blind taste test of Pepsi and Coke, and say which one they preferred. The reason this is called the Pepsi Challenge (and not the Coke Challenge) is that Pepsi promoted the results – because they often won.

The Challenge raises several interesting problems. First is whether this is a fair test. Pepsi has the advantage of being somewhat sweeter; many people prefer a sweeter flavor if you only take one sip, which is how the Challenge was set up.

Secondly, there’s a branding issue. Apparently, if the taste test is blind, more people prefer Pepsi. But if you know which brands you’re drinking, more people prefer Coke. In that sense, the Challenge is kind of a confession of weakness on the part of Pepsi’s marketing department, no? Knowing which brand you’re drinking actually undermines your taste experience.

Brands in general have a huge impact on taste perception. Check out the branded vs. blind taste test “maps” to the left. This is my approximate recollection of an actual study we were shown in marketing class. People were asked to taste beer samples without knowing what they were, and then intuitively arrange them on a map, which produces a vague grouping by flavor experience. Subjects were then asked to do the same thing with the same beers, but this time they knew what they were tasting. Incredible demonstration of the impact of branding.

Anyway, as part of discussing the dynamics of the soft drink industry, we actually took the Pepsi Challenge in Strategy class a few months ago. Each person stated their preference (Pepsi vs. Coke). Then, with the participants out of the room, three cups were filled, two with Pepsi and one with Coke. Participants listed which cup was the odd one out, and which they preferred (cup 2, or cups 1 and 3).

As you can see from Xs and Os in the far-right column, most people got this right. Pepsi and Coke are actually pretty different, though some people don’t seem to think so. About a third actually got this wrong, of which 2 thought they preferred Pepsi when they really preferred Coke (and vice-versa for the remaining person).

One thing that’s illuminating about this experiment is that it shows you how the soft drink industry remains profitable. Coke and Pepsi only go up against each other in the Pepsi challenge – there’s no RC Cola, and no other colas either. They compete, but only against each other, with the result that they have a cozy duopoly in the industry which maximizes profits for both of them over the long term. If you follow the pattern of new product introductions, you also notice that Pepsi and Coke tend to introduce similar products very close in time to one another.

Coconut cake with hibiscus sauce, salad with pomegranate seeds, salmon and mashed potatoes, duxelles puff pastry

Blanched and peeled pistachios. They look like little avocadoes...

Cake + sauce.

Dried hibiscus flowers. Image by Wikimedia user Thelmadatter.

Salad with pomegranate seeds and red onions.

Salmon with a cream sauce, over mashed potatoes and peas

Wow, this entry is late. Anyway, just before the holidays I had the chance to cook for some people from my class at school, which was really fun.

Let me start at the end and work backwards. Dessert was a coconut cake, with a hibiscus sauce and some toasted pistachios.

Coconut cake’s been on my list for a while now, and I’m glad I finally got a chance to make it. I am also frustrated that I burned the bottom of the cake slightly. There are several ways to test the doneness of a cake, including inserting a toothpick and seeing if it comes out clean (the toothpick test), seeing whether the cake springs back when you depress the top, and also waiting for it to start pulling away from the sides of the pan. I generally use the toothpick test, and I’m worried that I’m being too stringent since things I bake often come out a little overdone. In writing this, I realize the other possible problem is that I am using very long toothpicks that we happen to have. I wonder whether the point of the toothpick test is that it is only supposed to check the middle of the cake?

But what was really fun was making the hibiscus sauce. First of all, hibiscus isn’t that easy to track down, especially in Philadelphia. What saved me were some totally awesome Mexican grocery stores in the Italian Market neighborhood, specifically the one right next to Taqueria Verecruzana at 930 Washington St. They stock an incredible array of stuff in a tiny space! I didn’t really know what hibiscus flowers even looked like but fortunately was able to recognize their Spanish name, jamaica. Oh yeah, and the sauce was really delicious, and a beautiful, deep purple color.

For the main course I baked some salmon and served with a really tasty mashed potato, pearl onion, pea and dill mixture. At the last minute, I figured it could use a sauce, so I whipped something up with lemon, white wine, dill and cream.

Before that, I served an appetizer of mushroom duxelles in puff pastry, with a little thyme and a smoked tomato sauce. Duxelles is basically a mushroom paste with cream and white wine, named after the duke of Uxelles, who employed the chef who created it. (I think it would be a little bit of a stretch to call what I made duxelles, but I’m working on it!) I got the general outline of that recipe here.

And finally, for the salad, I served this pomegranate and spinach salad. I think I have written about this before. Note that it only works with fresh pomegranate seeds, and also almonds really do work better in this recipe than walnuts, which is what I used.

Beet and Carrot Salad, Green Tea Cheesecake, and Roasted Chicken

Filed under: Cooking Journal — Tags: , , , , , , , , , , , , , , , , — Justin Dunham on January 8, 2011

Cheesecake with green tea leaves and raspberry / mint garnish

Beet and carrot salad, with plenty of parsley

Before... you can see the chicken in the center, coated with a honey-mustard mixture and surrounded by tomato, red onion, and rosemary and thyme

...and after.

Between exams, other projects, and travel I hadn’t really been in a kitchen for a little while. So I decided to cook something, even though I’m away from home at the moment (in Portland).

Other people’s kitchens make me a little nervous – I can’t count on things that I need being available when I need them. So I decided to make some dishes that I could make without too many special implements. This meant a beet and carrot salad, roast chicken with tomatoes and onions, and a cheesecake.

I decided to go with this green tea cheesecake since I was also looking for an excuse to go to a nearby location of the amazing Asian supermarket Uwajimaya. I freaking love Uwajimaya. If I lived in Portland, it’s where I would shop even though it’s about an hour of extra travel. They have incredible fresh seafood of all types, as well as lots of really interesting fruit – fresh rambutan, lotus root, and buddha’s hand, anyone? I was pretty close to making a last-minute decision to incorporate some of these ingredients in what I cooked, but given that Epicurious had only one recipe including dragonfruit, I decided to wait.

Anyway, so, cheesecake, chicken, and beet and carrot salad. The beet and carrot salad recipe was fairly intimidating, even though I really like roasting beets from scratch. I decided to dispense with most of the complexity of the recipe, particularly the idea of buying golden and red beets and making a separate dressing for each.

Instead, I bought only red beets and made one dressing for the whole thing, combining the ingredients of the dressings from the recipe. I used cumin, shallots, lemon juice, balsamic vinegar, and olive oil. I also decided to use a ton of parsley, since it was getting overshadowed by everything else, but lends a really nice fresh flavor to the salad. The recipe calls for leaving the tops of the carrots on while you roast them. This also looks pretty nice, but I’m not sure it was worth the trouble.

The chicken was fun – I’ve never roasted a whole one before. I started with this Easy Provencal Lamb recipe from the Barefoot Contessa, who’s usually pretty reliable. But when I got to the store, I found that bone-in lamb now costs $10 per pound. So for 7 pounds… yeah. A chicken that could feed everyone cost me 10% of that price. I removed the giblets – had never done that before, and they weren’t even bagged. This was less disturbing than I thought it would be. I then smothered the chicken in the honey-mustard sauce and put it in the roasting pan. The chicken gets surrounded by diced tomatoes and onions in oil and honey.

Interestingly, it turns out I put the chicken in the roasting pan upside-down. Apparently, people do this on purpose, because it helps keep the meat moist, especially in this case since the chicken was basically poaching in all the collected liquid from the tomatoes, onions, and sauce. So the chicken came out really well.

Uh, the second time, that is. As soon as we cut into it the first time, it turned out was still raw! How embarrassing. I had used the “juices run clear” test, which it turns out is not accurate. Instead, insert a meat thermometer in the thigh, and wait until the internal temperature reads 165 degrees. I promise your chicken will not dry out (well, it won’t dry out because you cooked it fully – I can’t guarantee it won’t dry out for other reasons). Once I cooked it through again, it was fine. Overall, I recommend 425 or 450 degrees for an hour or 90 minutes.

The cheesecake may have been the best part of the meal, and it was certainly the easiest. Cheesecakes are made from a simple batter which includes cream cheese, sugar, and eggs. In this case, we also added crushed-up green tea leaves, and some greek yogurt. These were both substitutes for the far more expensive ingredients that the recipe calls for: powdered green tea and fromage blanc. The green tea complements the savoriness of the cheesecake really nicely, as does the shortbread crust. All three of shortbread, green tea, and cheesecake straddle this savory / sweet line pretty well.

The cheesecake is garnished with raspberries and mint, and you serve it with a tea, also made from raspberries and mint. I was pretty skeptical about this, but it actually tasted pretty good along with the cake.

Building a Desktop Fabricator (3D Printer), Day 5

Filed under: Everything Else — Tags: , — Justin Dunham on January 5, 2011

Soldering.

A properly soldered joint.

Everything's completed... technically. Extruder is on the right.

Alright, today I thought I was going to be able to finish this whole project off, but I didn’t quite get there. Got close, though.

The first thing I had to do was finish building the extruder, which took quite a while. I had a fair amount more soldering to do, including hooking up the thermistor, which measures temperature and allows the information to be transmitted electronically.

I ran into lots of problems today, including wires breaking (and therefore needing to be re-stripped and re-soldered), tight fits, and various other random issues. At one point, I needed to use a small steel rod to determine the appropriate spacing between two components. But everything seemed to come together in the end. Until I started hooking the Makerbot up to my computer to start doing some calibration and testing of the extruder.

Makerbot has great software called ReplicatorG that helps you interact with your Makerbot. Specifically, you can send it things to print, and you can also manually move the build platform around and do other things as well. Unfortunately, when I went to use ReplicatorG, I was unable to get it to communicate with my extruder. Therefore, no printing. I took a pretty careful look around some help forums, and while I found some good ideas that I will try tomorrow, I’m not optimistic about getting this to work soon.

The good news, however, is that everything else seems to work well – the platforms move just as they should, which proves that that’s all built correctly. More as it develops on the extruder issue.

Building a Desktop Fabricator (3D Printer), Day 4

Filed under: Everything Else — Tags: , — Justin Dunham on January 4, 2011

The setup for building the extruder.

Learning to use a multimeter...

The multimeter. The dial on the top measures current flow, the knob allows adjusting the range that the dial shows. The red and black leads connect to an electrical source.


When I left off yesterday, I was about to start building the extruder for my Makerbot. The extruder is the part that heats up, and then lays down, the plastic used as printing material – it’s the 3D printer equivalent of a printer head.

Building the extruder, for an electronics novice like me, is intimidating. This would be first time I’ve ever really soldered, used a multimeter, etc. (A multimeter, by the way, is a device that allows you to measure how much electricity is flowing across a circuit.)

The first step was actually to go shopping! I needed a bunch of stuff that wasn’t included in the kit, such as the multimeter, wire strippers and cutters, and some other odds and ends. I also thought I would need kapton tape, which is a special type of insulating tape that turned out to be included in the kit. Thank goodness for that, since it wasn’t available at any of several electronics stores I tried in Portland. Neither was a certain type of wire that’s called for, which also turned out to be included in the kit… I think.

These slight oversights bring me to one thing that bothered me generally about this part of the build process: it’s a level up in difficulty from the rest of the build, and the instructions provide lower quality, and often less, guidance than before. I think this is one thing that will definitely need to be addressed if the Makerbot is to have wider appeal (or perhaps, as I said yesterday, the extruder should just be pre-built for you). As an example, to quote from the instructions:

The nichrome wire should have a resistance of approximately 6 Ohm. Usually this amounts to about 300 mm of wire, but use a multimeter to be on the safe side. Cut the nichrome wire to a length which amounts to about 6 Ohm of resistance (+/- 1 Ohm).
NOTE: It’s difficult to measure the resistance of the nichrome until you strip the ends, so cut it long, then trim it down until you reach 6 Ohms.
NOTE: Make the wire too long, and your heater takes longer to heat up. Make it too short, and you burn out the MOSFET on the extruder board.

So, what does this mean? You have to cut a certain type of wire to a certain length. In order to get the right length, you have to measure the resistance with a multimeter. If you haven’t used a multimeter before, you have to learn how to do so. And then if you get it wrong, you may end up destroying a critical component of the machine when you start to use it.

Fortunately, learning to use a multimeter isn’t that hard, and I was able to figure it out. So what am I complaining about? I guess I don’t really mean to complain – this kit is clearly targeted at hobbyists, and as I’ve said many times, it’s an ingenious machine and up to now extremely thoughtfully documented. But I think it would be a good idea to simplify build process steps such as this one, and that will be necessary to broaden the appeal of these fabs beyond hobbyists.

There are some other really touchy steps, too – for example, when building the idler wheel, you have to put a circular bearing within a larger plastic ring. But the bearing must be glued to the outer wheel at a very precise depth – i.e. it shouldn’t be flush with either the top or the bottom of the wheel. Moreover, it has to be superglued in place, so, no second chances. Lastly, there is some kind of problem with the instructions here, where someone has inserted a note to update the instructions, without actually doing this. I had to tread carefully, and I’m not 100% sure that I did it right.

A soldered (and then taped) wire connection.

Things like this aside, construction actually went pretty well. I did my first real soldering – hooray! – and you can see the results of that in the pictures accompanying this post. Tomorrow, I’ll try to get a picture of me actually performing the soldering operation, so anyone unfamiliar can see what that actually looks like. This was a pretty forgiving situation, but I was happy that that went well, and I was able to confirm that everything was OK, using the multimeter (which is a very cool tool).

I also built the supports for the extruder, which are called “Weird Dinosaur” and “Big Dinosaur” on account of their shapes. I’m not sure if this is helpful nomenclature or not…

Tomorrow I am hoping to complete the build of the Makerbot, though there’s a lot to do between now and then.

Creative Commons License
.