Consistency, consistency, consistency. That is the key. I have used uber flat temp profile, hump profile, inclining profile, declining profile, etc... machines and gotten good shots from all of them. The trick is to get repeatable temperatures. That is, in my opinion, what you are paying for when you make the jump from the prosumer to commercial machines. It is not a quantum jump in cup quality, most of the time it is a relatively small change but what you get for your dollar is consistency. That is not to say that every coffee behaves best on any of these machines. What works good on a very flat profile may be undrinkable on a machine with a different profile and vice versa. You just have to find the blend that you like when pulled on any particular machine. The same coffee ground on the same grinder will taste different between two different machines.

The quest for uber flat intra shot temperature profiles was driven by engineers. You can measure temperature, you cannot measure taste, it puts an objective quality on a subjective variable. In the end you find the coffee that works best for your machine on your grinder and your palate.

If one is brewing entirely to please one's own palate one needn't worry about any kind of objective testing criteria. I know what I like and I like what I taste if sufficient. If one wishes to recommend to others which methods, coffees, equipment produces better espresso, one ought to take the time and effort to do the required blind testing.

Humans are simply too subject to subjectivity. To paraphrase nobel-prize winning MD Sir Peter Medawar, if you doubt the ability of subjective thought to control human physiological response, take the simple case of blushing, wherein a simple mental stimulation can cause the well-known red flushing of the skin.

The point is, your mind can control your body, probably even to the point of knowing espresso A is better than espresso B, because it came from machine C rather than machine D. Taking the time to do a blind test eliminates this possibility.

bostonbuzz wrote:Or there is something being overlooked?

Dave nailed it:

cannonfodder wrote:The quest for uber flat intra shot temperature profiles was driven by engineers. You can measure temperature, you cannot measure taste, it puts an objective quality on a subjective variable.

I elaborated on a similar point in Espresso machines at $2000 budget. That is, when all you have is a hammer, everything looks like a nail.

And when you look like a nail, you spend a lot of time looking up and cringing at shadows.

HB wrote:I elaborated on a similar point in Espresso machines at $2000 budget. That is, when all you have is a hammer, everything looks like a nail.

I am so tired of statements like this, and plead guilty to jumping in with a knee-jerk defense of engineers when I see them.

It's just too black-and-white a statement, Dan. The truth is really somewhere between science and art. Shades of grey man, shades of grey.

These discussions remind me of similar debates between audiophiles. Years ago, the editor and writers of Stereophile magazine took the position that only measurable differences, such as total harmonic distortion, played a role in how the gear sounds. The editor and writers of of the Absolute Sound took the position that measurements are irrelevant and the the only thing that matters is whether the music sounds like a live performance.

IMHO, both camps were wrong. Sure, the sound is produced by physical phenomena, some of which we can measure, but there are so many interacting variables that no one has yet figured out how to reproduce sound perfectly with electronics. And sure, what you hear is really the bottom line, but sound perception, like taste, varies considerably from one individual to another, so without using objective measurement tools we can't begin to track down what's causing the differences.

Too much reliance on a purely sense-oriented diagnosis leads to all sorts of voodoo-like and subjective hand-waving like $10,000 1-foot interconnects and trying to get a perfectly symmetrical extraction cone when using a bottomless portafilter. No one has been able to prove that these sorts of things make any real difference at all because the effects are extremely subtle and they can't be measured with current technology. And since sense perception varies greatly from one person to another, opinions about sound and taste simply don't tell us a whole lot about what a given individual will perceive.

But too much reliance on measurements misses the point of the whole thing: is the listener or taster pleased with the results? Are all of the desirable sensory stimuli making it from the source to the brain?

In espresso production, we can measure some things, like time, temperature, pressure, dose weight, beverage weight, %TDS, etc. But as has been noted here ad nauseum, there are a lot of inter-dependencies, and we still don't have equipment capable of accurately measuring what's going on at the molecular level when the water hits the coffee, or what's going on at the cellular and molecular level when the coffee hits the palate.

Many of us are trying to slog through the mud using as scientific an approach as we can, and whatever measurement tools we can find, to deal with a characteristic that's highly subjective and varies greatly from one person to another: taste perception. It's not an exact science by any means, but I feel it's important to try to correlate these measurements with taste. Bear in mind that without engineers and their measuring tools, we wouldn't have Titan grinders, Speedsters and advances in roasting and brewing techniques. But by the same token, without the final taste-test, the measurements would be meaningless.

Since it's not likely to be proven either way any time soon I would be interested in seeing how analytical versus intuitive thinkers feel on the subject.

First, answer the question. If a ball and a bat cost $1.10 and the bat costs $1.00 more than the ball how much does the ball cost?

My theory:
If you answered .10 chances are you believe in different temp profiles producing different results.
If you answered .05 chances are you're skeptical.

Peppersass wrote:...Many of us are trying to slog through the mud using as scientific an approach as we can, and whatever measurement tools we can find, to deal with a characteristic that's highly subjective and varies greatly from one person to another: taste perception. It's not an exact science by any means, but I feel it's important to try to correlate these measurements with taste. ...

The lasting impression of the audio wars overall is that measurement engineering eventually caught up (somewhat anyway) with the ears and bore out in numbers just about everything that the Harry Pearsons of the era said they were hearing. Once the technology was improved to quantify the phenomena, just about everyone came around to agreeing after all that absolute phase did matter, that the CD bit rate was too coarse, that interconnects that correct for propagation delay did sound better, that power line hash did destroy wave-forms and so forth. (If all you have is a 44k, everything sounds like a hammer...)

When selling - or arguing - numbers are convenient replacements for facts, and big or constant ones become false proxies for quality. Since most buyers do not have the means, knowledge, patience or attention span to develop their own judgements from experience, even a most rudimentary measurement that sounds meaningful will assume a convincing power.

Brillat-Savarin's seminal 16th century works notwithstanding, quantification methodologies of taste sensations are probably well behind that of aural sensations; so instead of "goodness" of a cup of coffee we find ourselves thrashing over the goodness of a specification of a machine that makes that cup of coffee. That we can put a number to, and so, fresh from the Kelvinator, we convince ourselves that we know something about it.

It's funny: we all talk a good game, universally agreeing that "the taste is the final arbiter," but 99% of the pixels spilled in these forums focus at least one or two levels of abstraction above the actual extractions themselves. Myself included, it is - as here exampled - so much easier to discuss a machine's temperature profile than the real brass tacks of how such a profile might taste.

Coming full circle then, items like the coffee refractometer do start to enable measurement of the coffee itself. Total Dissolved Solids may only be our current analogue to the Total Harmonic Distortion of audio's analog, but anything that moves us to a better understanding of the espresso - and not just the machinery - has to be a step in the right direction.

another_jim wrote:The second problem is that it's inconceivable, at least to me, how anyone can even can come up with a halfway plausible model on how the taste could be systematically different on these different profiles. Therefore, this debate is not only senseless in this reality; it is senseless in all possible realities.

I think we have lost our way in this discussion. Correct me if I'm wrong, but I don't think anyone is saying that one machine's temperature characteristics are better than another's. I think the premise of this thread is that a given machine should have a consistent temperature profile that repeates itself shot to shot. Using Jim's three profiles as an example, I can't say which one would taste better but I wouldn't want a machine that threw out any one of those different profiles randomly. It's been said before and I'll say it again, consistency is the key ingredient.

another_jim wrote:... it's inconceivable, at least to me, how anyone can even can come up with a halfway plausible model on how the taste could be systematically different on these different profiles. Therefore, this debate is not only senseless in this reality; it is senseless in all possible realities.

Uh ... did anyone else glance downward and happen to notice a gauntlet at your feet?

Jim presents a fair challenge. While not likely to provide a comprehensive or fully definitive answer, I propose the following experiment. The idea is to execute shots with temperature variations as Jim charted them, but varying those temperatures to such an extreme that the differences will be exaggerated and obvious, thereby arriving at any tendencies we might reasonably expect in the taste of espresso when the brewing includes the smaller temperature shifts present in most machines:

• Choose a familiar coffee
• Pull one shot at your favorite temperature. (It would be nice if your machine is reasonably temperature-stable, but it shouldn't be strictly necessary for these purposes.) This shot approximates Jim's first (flat) curve above. Taste it carefully.
• Prepare two baskets to be as similar as possible. Pull one shot with your machine set 5F lower than your standard temperature, but cut the shot off half way through your typical duration and set the cup aside.
• Now reset your machine 5F higher than your standard temperature, and pull the second of this pair of shots, this time keeping only the second half of that shot.
• Combine it with the first half-shot you made at the lower temperature. You now have a shot that should approximate Jim's third temperature curve - not in any way perfectly, but at least in its general effect.
• So: how's it taste compared to the first "flat-line" shot? What's different that you notice using a too-hot start and a too-cool finish?
• Repeat, but this time go 10 degrees higher and lower. One imagines that any differences would be exaggerated, and we might start to see some general tendencies.
• What happens if you reverse the temperature direction? I've suggested going from cool to hot, but that's just because most machines can heat up faster than they can cool off, making the process easier. But temperature changes within a shot going the other way are just as instructive.

Of course there are plenty of technical problems and variations that, strictly taken, compromise these results. No two shots are identical, the first half-shot will have cooled a little, the temperatures within the variations will vary, the machine isn't linear, coffees are different (someone will try Sumatra) and so forth. But we still should be able to develop some kind of consensus on the way that temperature variation and direction influences the cup.

The routine can also be done with three shots, with one third of the shot at low temp, one third normal and one third hot, each for 1/3 the usual shot duration. This could give some approximation of the effect of Jim's sine-wave curve.

Any results here would be very gross, with mileage variances all over the place. Still, most people currently seem to buy into a few general rules about temperature and taste, such as too-cool brewing leads toward sour taste and too hot leads to bitterness. With enough machines and coffees and people trying this out, maybe some general tendencies about the effects of within-shot temperature variations can be unearthed.