Author: Marshall Schott
In beer evaluation, one of the most oft cited off-flavors is dimethyl sulfide (DMS), a sulfur compound commonly said to impart a creamed corn or cooked vegetable character in beer. Most brewers are well aware of methods purported to reduce the risk of DMS making it into their beer, for example extended boil lengths and making sure to keep the lid off the kettle. But what exactly is DMS and where does it come from?
The most prominent source of DMS originates from the decomposition of another compound, S-methylmethionine (SMM), which is created during the germination stage of the malting process. SMM being heat sensitive, it easily converts to DMS during the mash. Since roasting and 
kilning reduce the presence of SMM, brewers tend to be most concerned about DMS development when using larger amounts of very light malts such as Pilsner. For a more in-depth look at DMS, I highly recommend Scott Janish’s excellent article, How To Prevent DMS In Beer.
Undeniably, DMS is real and can impact beer character. Equally as true is the fact humans have some capacity to perceive it in relatively low amounts. I’m not here to question this. Rather, what I find curious is the sheer frequency people note detecting DMS when evaluating beers. Whether in sanctioned competitions or less formal homebrew shares, those three letters are almost certain to be uttered or written by someone, and while I’ve no solid evidence to back this up, my hunch is this collective knowledge of DMS stems primarily from descriptions provided by popular authors and podcast hosts. I know I’ve done it, many times, and I couldn’t help but wonder– how can we be so sure?
For this xBmt series, Brülosophy partnered with FlavorActiV, producers of various off-flavor capsules that brewers can use to help them become better overall sensory analysts.
| PURPOSE |
To evaluate the ability of blind tasters to distinguish between a pale lager dosed with DMS from an unadulterated sample of the same beer.
| METHODS |
I initially planned to use a batch of my own beer for this xBmt and actually brewed 10 gallons of what ended up being what I think is my best Vienna Lager to date. However, as I thought about how I wanted to approach this variable, I realized it’d be prudent not to use a homebrewed beer that could possess a certain level of DMS and instead go with a commercial example know for being clean and crisp. Lauded by my entire circle of brewing mates, Bitburger German Pilsner fit the bill perfectly, so I headed to my local Trader Joe’s to pick some up.
Without lab data, which I can’t find online, it’s hard to know what the actual level of DMS in Bitburger is, but in the years we’ve been enjoying it, that descriptor has not once been used.
FlavorActiV provided me with 6 DMS capsules, which based on conversations with a representative from the company, was enough to dose 32 fl. oz. of beer to above standard threshold levels.
I wanted to ensure the DMS crystals were well dissolved and equally distributed in the beer, so I decided to use the different colored Nalgene bottles I often use when traveling for data collecting, emptying the contents of the 6 capsules into one of them then filling each with 2 cans of Bitburger. I noticed the DMS dosed beer was slightly more foamy than the other, but that dissipated after the 1-2 minutes it took for the crystals to fully dissolve.
The filled and sealed bottles were placed in a small cooler with enough ice packs to keep them adequately chilled until data collection commenced about an hour later.
| RESULTS |
With 32 fl. oz. of beer in each bottle and a serving size of approximately 3 oz per sample per participant, my participant pool would be restricted compared to other xBmts. Even though experience may not effect perceptual abilities, given the variable being tested, it was important to me that most of the participants have at least some familiarity with DMS. In total, 11 people consisting of 4 BJCP Certified beer judges, 3 Cicerone Certified Beer Servers, 1 experienced homebrewer who now brews professionally, 2 craft beer fanatics, and my Miller Lite drinking neighbor participated in this xBmt. All participants were blind to the variable being tested as well as the fact the beer was a respected commercial German Pilsner. Each taster received 1 sample of the DMS dosed beer and 2 samples of the unadulterated beer in different colored opaque cups then were asked to identify the unique sample. In order to reach statistical significance at this sample size, 7 (P<0.05) tasters would have had to accurately identify the DMS dosed beer as being different. Ultimately, only 3 tasters (p=0.77) made the correct selection, meaning a beer dosed with DMS was not reliably distinguishable from a clean sample of the same beer among this panel of participants.
Once all the data was collected, I revealed to tasters that this was an xBmt focused on off-flavors and that at least one of the samples they were tasting had been dosed with a commonly cited off-flavor. After going back through for sniffs and sips, I asked each taster what off-flavor they believed was used for the xBmt, only one person identified DMS, and he happened to be one of the few who was accurate on the triangle test.
My Impressions: I love Bitburger… and I’m pretty sure I hate DMS. From pouring the beers into the Nalgene bottles to cleaning them out the next morning, I was easily able to detect the dosed sample. To me, it absolutely did not have creamed corn character, which I quite enjoy, but rather smelled of overly boiled cabbage slathered in cheap ketchup. Curious if what I was detecting was a function of knowing which sample had the DMS or if I was truly smelling the compound, I participated in multiple “blind” triangles served to me by others. On my first attempt, my buddy asked me to close my eyes then handed me separate cups one by one, lo and behold, I couldn’t pick out the sample with DMS, which I interpreted as confirmation my brain had been playing tricks on me. But then, before opening my eyes, my buddy handed me one more cup and, sure enough, I detected that horrible smell… he’d initially tried to fool me by serving me 3 of the same non-dosed samples.
| DISCUSSION |
Humans are interesting. I don’t fancy myself an expert on our species and cringe a bit when others claim to be. If I’ve learned anything in my 15 years in this profession (psychology), it’s that human beings do weird things for weird reasons, often outside of our awareness. When it comes to the evaluation of beer, I can’t help but believe there’s far more going on than we’re aware of, that our perceptual idiosyncrasies, personal preferences, and past experiences all play a role in what we often think to be objective.
As I observed taster after trusted taster making the wrong selection in the triangle test, my mind began to wonder what could be going on here. After all, I was immediately able to identify the DMS dosed beer in multiple triangle tests, regardless of whether the beer was cold or warm. This was a very unique experience for me, as I almost always perform as well as if not worse than the participants. I was quite befuddled and naturally considered some some potential explanations.
I suppose it’s possible the concentration of DMS in the beer wasn’t high enough, but that seems like such an easy excuse, and I’m personally compelled to trust the manufacturer’s recommendation is sound. The odds seem awfully slim the intentionally dosed beer would contain less DMS than a non-dosed homebrewed beer, and the fact those who did detect the DMS beer seemed to do so rather easily suggests to me it was at least beyond standard threshold levels. Maybe certain people have a sensitivity to DMS, as was posited by a couple of my friends, both BJCP judges who I rely on for honest feedback, and both who were wrong on the triangle test. This being the case, those with the sensitivity should have been able to consistently identify the dosed beer over a series of multiple attempts, which is exactly what happened for 2 of the 3 correct tasters and myself. In the end, there’s really no telling why the results were the way they were, my guess is as good anyone else’s.
Personally, I’m far more interested in the implications that can be drawn from the fact a group of arguably qualified beer tasters were not only unable to identify the presence of DMS in a beer, but couldn’t tell it apart from clean samples. I’ve always assumed this compound was rather easily identifiable, and now I’m left shamefully aware that past DMS comments I’ve made have probably been wrong, that the “creamed corn” character I noticed after intentionally digging so hard for it may have been an acceptable component of the style imparted by ingredients… that my use of regurgitated jargon was likely motivated by my own desire to feel adequate.
I’ve found the extent of my rambling tends to correspond with my level of bewilderment, and while I could pontificate much more about this stuff, I’ll put the brakes on. For the sake of brevity, I won’t address the various contentions I’m presuming some will make against these results. But for those inclined to rely on the shitty palates argument to reconcile the dissonance experienced as a result of these findings, particularly if your only experience with DMS is what you believe you’ve tasted in beer, please consider getting some flavor standards and trying this out for yourself. You might be surprised.
If you’ve ever experienced a DMS dosed beer or even if you think you’ve had a beer with noticeable amounts of DMS, please share your thoughts in the comments section below!
Support for this xBmt comes from FlavorActiV who provide GMP Quality Flavour Standards, sensory panel taster training, and the world’s biggest beverage sensory proficiency schemes.
Support Brülosophy In Style!
All designs are available in various colors and sizes on Amazon!
Follow Brülosophy on:
FACEBOOK | TWITTER | INSTAGRAM
If you enjoy this stuff and feel compelled to support Brulosophy.com, please check out the Support Us page for details on how you can very easily do so. Thanks!
57 thoughts on “exBEERiment | FlavorActiV Off-Flavor Series: Dimethyl Sulfide (DMS) In A Pale Lager”
Awesome experiment. I’ve only had one beer that ever had a DMS problem and it was horrible, but I don’t know what I’d done to cause it. Every other beer I make with pilsner or other light grains has been fine, or maybe there was DMS and I couldn’t perceive it. Good work!
Thanks!
> I’ve only had one beer that ever had a DMS problem…
See, I’m wont to wonder how you can be certain it was a DMS issue?
Great start to a series. Guess you can no longer argue that you have a shitty palate, at least with respect to DMS.
I hope you will spike in some esters (other than isoamyl acetate). I’ve always been leery of claims like “ferment (insert ale yeast here) at 64 or expect fruit salad in a glass.” Never really experienced that myself, but maybe I’m using the “wrong” yeasts, or just have a shitty palate.
maybe DMS changes the flavor of beer differently at different concentrations.
“… methods purported to reduce the risk of DMS making it into their beer, for example … making sure to keep the lid off the kettle” is another old wives tale.
The idea of not using a lid is ostensibly to prevent SMS-laden condensate from dripping into the wort; however … this can be avoided easily by tipping the lid periodically to drain the liquid into a bowl.
Another thing to question; conventional wisdom tells us to chill the wort rapidly; otherwise … DMS may form. This flies in the face of the success we have whirl pooling / hop-stand.
—
And why is a 90 minute boil recommended routinely for pilsner malt (SRM 1.8) but not 2-row malt of the same color?
Lid condensation doesn’t stay put until you lift it off. It drops back into the liquid. Also it’s more complicated than that. Pot lids increase pressure which affects evaporation rate and boiling temperature. Even altitude can affect the ability the get rid of DMS.
Pilsner and 2-Row have different DMS requirements because they’re obviously different malts. Their colour is irrelevant. They are aren’t kilned the same way and so have different SMM levels which is the precursor to DMS.
Those who say something can’t be done should get out of the way of those who are doing it.
For what it is worth, I did an entire series of Flavor Active kits with my homebrew club and we concluded that the kits were extremely mild. Many of the flavors were very hard to detect even in Miller Lite. I promise the null hypothesis will loom large if you keep doing this series, but it isn’t because people can’t taste these things. We discovered that a 150 increase in concentration (2 cans of Lite instead of one) made a difference.
And yet the DMS dosed beer I made wave very easy to detect for those who could easily detect it. I’ll bring this up with them, but as I said in the article, I’m hard pressed to believe they’d recommend amounts that don’t push the compound beyond standard threshold, as all they’d have to do is recommend a more concentrated blend.
The more I have learned about beer and brewing in general, and DMS in particular, the more baffled I have become over the amount of claims about finding DMS in almost every pale lager, and many other beers for that matter. I believe it boils down to (a) some people being super sensitive to DMS (b) pure bullshit and/or self deception or (c) attributing some generic pilsner malt character to DMS. Probably mostly (b) and (c).
This result is consistent with some theories regarding the quality of the average BJCP certified judge or cicerone. Well done. ;D
I dumped a DMS bomb a couple months ago. Even tried to dry-hop the crap out of it to cover up the corn flavor to no avail. I was inspired to try a short boil after looking at some XBMNTS here and I generally brew experimentally (2.5 gallon batches). I attempted a 30 minute boil but actually had issues keeping the flame up because my propane was running low and there was a little bit of wind in play. The grain bill was 95%+ two row. I didn’t force any friends to drink it but both my wife and I could clearly pick out an unsuitable corn flavor.
I think it would be fair to double dose an XBMNT like this. It’s all about thresholds for people’s senses. I can’t say I’ve tasted diacetyl but some people call it out on a ton of beers that are perfectly fine, imo. Asparagus pee can only be smelled by a certain % of the population, right?
I’m wont to wonder if it was actually DMS you were detecting.
well if I’d have known you were going to disprove DMS flavor, maybe I would have saved it for everyone! 😀
My HBC used one of the flavor kits and honestly, we’d have been better off just doing samples of our homebrews the off-flavors were so mild. With DMS in particular, I think it’s a mix of some people have a sensitivity to it (ditto to other off-flavors), and it is so prevalent in commercial beers that it goes unnoticed by others. Of course that’s purely conjecture… and I didn’t score well on the DMS test.
Now please do an lid on/off boil exbeeriment! 🙂 I usually simmer with the lid on for most of the boil, lifting the lid and bringing it up to a rolling boil at the end. I am very curious if you would detect a difference between this (or only simmer and lid on) and a full rolling boil for a pilsner style beer.
Thanks for the great work
It’s interesting that tasters can’t detect a beer that is unequivocally different. What does this say about the many tests that have concluded that technique X makes no difference to a beer’s taste, flavour or aroma? I think taste tests must have poor sensitivity.
How else do you measure perceived differences besides perceptual analysis?
There’s no alternative, but my point was that more caution is needed in conclusions drawn from tests in which the different beer isn’t detected by a significant number of people.
That’s the whole point of the experiments, isn’t it?
The interesting question is how sensitive a tasting panel is to real differences. For various reasons, different beers may be perceived as the same when tasted in quick succession. I suspect many tasters wouldn’t be able to discriminate between different commercial beers of the same style under test conditions. That doesn’t mean those beers aren’t different. It just means the test is not very sensitive and may sometimes overlook real effects. Addition of DMS is a case in point but the principle applies to other tests.
Or maybe it’s not the test that’s not sensitive, but tasters in general.
Sorry, posted twice – my browser not updating.
It could be tasters in general or could be other factors, such as effect of multiple drinks on palate dulling olfactory sense or odours from beer mixing in the air. One interesting idea might be to repeat the DMS test at successively higher concentrations and see when significance is reached. Maybe guidelines for threshold of detection in an average person are wrong too.
Isn’t that the point? The beers could be chemically different but the general population cannot tell. If the population cannot detect a difference then the difference does not matter (unless you are one of the few that are sensitive to whatever chemical is different, DMS in this case).
That’s one possibility, but another possibility is that you might not notice a low level of DMS in one of three small beakers of beer swallowed in succession, but you might notice it when drinking a whole pint of contaminated beer. And you might notice it all the more if you’re looking for it. Perception is a fickle thing.
I asked every participant to repeat the triangle after they’d already done, and I told them what was up, those who got it wrong originally still couldn’t detect it. Even when I handed each the dosed sample and told them it was the dosed sample, they all said it smelled and tasted the same as the clean sample.
Re: Ben’s point on it maybe taking a pint.
I suppose that’s possible as flavour perception changes with temperature, but if it took me an entire pint to notice a “flaw”, I don’t think I’d be worrying too much about it.
I’d love to do these tests myself. I don’t notice many “flawed” beers. If it doesn’t jump out and slap me across the face I don’t care. Maybe I’ve got a shitty palate? Or just a shitty attitude?
@Marshall – that does suggest the dose is near the threshold of detection, but another confounding issue is that all participants now have a mixture of both beers in their stomachs. The flavours will inevitably be in their respiratory systems and nasal cavities, contaminating further tastes, so subsequent tests will be less sensitive.
Another issue with triangle taste tests is that detecting the absence of a flavour is harder than detecting it’s presence. If a taster happens to drink a non-DMS beer first, they may notice the presence of a new flavour on the next taste. However, if they drink DMS first and so have the DMS compound already in their body and potentially triggering olfactory receptors at some level, detecting the absence of DMS in the next mouthful of beer will be more difficult.
There may be no truly empirical alternative to triangle taste tests for flavour compounds, but the technique is a blunt instrument and it’s easy to over interpret non-significant results.
All of the tasters had a glass of water with them during the tasting, so they could rinse between samples. If the residual carryover from the blend of samples in their stomach is the culprit for their inability to distinguish the beers, I’d say that still supports the notion people may not be very sensitive to DMS.
A beer is never judged in isolation, but in a flight of beers that are often nowhere near as similar as the samples I served to the participants.
As easy as it is to over interpret non-significant results, I think it’s just as easy to come up with rather esoteric explanations for why the results failed to achieve significance. Blunt or not, the triangle test is the best we’ve got at this point, and I’m more willing to accept the information gleaned from it than I am anecdotal reports from folks whose only experience with DMS (or other variables) amounts to something they read in a book or heard on a podcast.
But that’s just me 🙂
Well, there is a difference between a difference that can be analyzed, or measured – ie in a lab with equipment, and a difference that may or may not be able to be perceived.
We encounter this often in our trials. We can see the difference of say gravity or pH via instruments, and yet no difference is recorded (statistically) by our tasters.
That’s what makes this particular test interesting. There’s an unequivocal, measurable difference between the beers that Marshall can taste but the panel can’t detect. If the two beers had been produced by a difference in technique (say, keeping lid on or off during the boil), the tempting conclusion would be that keeping the lid on doesn’t matter. That conclusion would be wrong though – at least for Marshall. The same principle applies to all tests for subtle flavours that produce non sig results.
On the other hand, tests that give significant positive results give more grounds for confidence in the conclusion, eg the water chemistry exbeeriments.
I’ve often observed commercial boil kettles with relatively small central vents (chimney?) for steam release and felt a large amount of condensate must drip back into the kettle. It is interesting that the homebrew norm is to keep the lid completely off during the boil to ‘avoid DMS’. Maybe as suggested a lid-on boil would be an interesting exbeeriment.
There was a podcast, Beersmith I think, with Charlie Bamforth about making sure you have a vigourous boil. He went on to describe a brewery that he consulted for which had a DMS issue. How you described the boil kettle setup is exactly what Charlie said they had and after removing the steam hood their DMS issue was solved.
The mildness of the off-flavor dose is worrisome for those of us who will be sitting for the Certified Cicerone test within a few weeks. I noticed the same thing months ago during a group off-flavor tasting and was hoping it was a fluke!
I apologize for long and somewhat rambling post (but I feel it’s important):
there is a bigger issue lurking in the background here that may undermine the entire brulosophy thesis (a site which I love to read, so it’s very disconcerting to me). And please keep in mind I probably read ALL of your experiments, ordered your merchandise, listened to you on various podcasts etc. You guys rock!
your p-test are based on assumption that all tasters are identical. Which is a reasonable first-order approximation.
But this experiment (and many others) seems to indicate, that’s not quite true. Some people are more sensitive to certain chemical compounds, and others can’t taste them at all. There are genetic differences among us, like smelling asparagus pee compounds – some people can smell it, some can’t. You can get genotyped by companies like 23andme and they can tell you whether you have this ability or not. Same thing may be happening with DMS and many other off-flavors, perhaps not so binomially (yes or no), but still.
This variation will undermine most (almost all?) your experiments so far that utilized a triangle test. For example, using triangle test of asparagus pee, if your panel has, say 30 members with 50%/50% split of those who can distinguish it and those who can’t – it will be proven inconclusive every time, regardless of the concentration of the asparagus compounds. You will never reach 95% threshold. 15 members who cannot smell the compound will throw it off for you, even if 5 of them randomly guessed correctly trying to please you for some reason instead of saying that all 3 samples smell the same to them.
The tempting result of this imaginary experiment – asparagus CLEARLY has no effect on smell of your urine! But the super-tasters, I mean super-smellers (50% of them) will strongly disagree!
***
You may want to consider modifying your triangle test a bit (run this by some statisticians to be sure) – currently you test every person with a triangle test only once. And you compare it to probability of selecting one by random (33.3%). But if you test the SAME person again, and effect is real – they should get it right, but probability of random guess goes down to about 10%. Test them again, and it’s down to 3%, and another try gets them to 1%. It’s almost impossible for someone to guess 3 times in a row, never mind 4, even in a small size sample.
Let’s say you do 4 triangle tasting tests on each person, in a group of 30 people. But 10 people can pick out the off-beer 4 times out of 4. and maybe 10 people are between 4 an 2, and 10 others are between 1 and 0.
What this will tell you is that about 1/3 or maybe a bit more people (10-15 or so) can definitely see the effect of the tested variable and it’s very much real. It cannot happen by pure chance. Maybe another 10-15 or so people CANNOT taste the variable being tested.
Another simplification of the protocol – and again, you may need to run this by statisticians, I am just a physicist – ask for multiple triangle test only from people who “got it right’ the first time. You are more likely to zero in on the super-tasters. But mathematically it may be easier to ask everyone.
There is additional value in being able to identify the correct sample every single time, as compared to random guessing. Just like you were able to do with DMS, and many other beerxperiments.
I realize there are complications. By 4th triangle taste everyone is a bit “drunk”. Ok, maybe do 3 triangle tests or even just 2. Maybe try lower-alcohol beers for some controversial tests. But you will get a better idea from those cross-correlations and auto-correlations (within the same taster) compared to “random chance” scenario, simply because you are no longer assuming all tasters are the same. Now you can assign a new variable – what fraction of population can reproducibly tell the difference with 95% certainty?
It would be very illuminating if, say, 20% of tasters could tell DMS within 95% certainty, and 40% of tasters could tell the difference between different fermentation temperatures within 95% certainty – as opposed to your current statement of “95% certainty has not been reached for the entire group, assuming all tasters are equally qualified to test the difference.
You are wrong, you do not need that assumption in order for the triangle test and the p-test to make sense. The null hypothesis here is “no one can detect any difference, and everyone is merely guessing”.
I appreciate your thoughtfulness. Considering the fact an entire sensory analysis industry uses the triangle test, I’m comfortable relying on it for the experimentation I’m doing.
That said, if there are indeed folks with sensitivities to certain compounds, than I’m wont to believe those who spout conventional wisdom ought to reconsider what it is they’re promoting.
it would be interesting to perform the test again after scrutinizing the panel further. I think this is would be a great/needed followup…then see where the results fall.
Dan
My club did a few technical meetings with these flavor kits, with Bud Light as a base (we of course first had to sample all of the american light adjunct lagers to determine which was the most flavorless).
I seem to recall that the following somewhat improved the overall perception of the Bud Light for me: DMS, isoamyl acetate, Diacetyl, Phenolic — and arguably geraniol, “grainy”, and “catty”. I’m sure that significantly higher levels, though, could definitely be considered bad.
Interesting! If I had to guess what might explain why the triangle test wasn’t significant, but you were still able to reliably detect the DMS samples is that DMS is extremely volatile. Is it possible that the cups or serving pitchers with the DMS sat longer when the triangle tests were performed on others versus when you tried the samples?
Interesting thought, Dan. The first triangle I did was maybe 2 minutes after the first 2 participants finished, a seemingly negligible amount of time. Every repeat test I did was within the following hour, probably less, pretty much after each participant finished, and I consistently selected the DMS sample.
I would say that that blows that theory away then! I believe that you’ve seen that not knowing the variable being tested versus knowing what the variable being tested is does not make a significant effect, however would there be exceptions to that? I can only think that there would be two reasons that you could detect the DMS while the triangle test was insignificant: 1) you know the variable, and/or are educated as to what the off-flavor tastes like (bad palate argument?), or 2) there is a significant portion of the population that genetically cannot detect DMS, or is not very sensitive to it. I know that the idea that some tasters can detect DMS more than others has been cited in literature, but I don’t know if there has been a lot of work to establish what the percentage is of the population that cannot detect it, or has a hard time detecting it.
Perception is Reality
My only question would be: How familiar with Bitburger was your panel?
To me, if you are familiar with the beer you are testing, you can pretty much pick up on any deviation from what you know to be its “true” taste profile more readily than someone drinking it for the first time or first few times.
I specifically chose Bitburger because the group of friends I had participate all drink it very often, it’s collectively one of our favorite beers.
Another aspect to consider is whether or not they could detect the DMS spike AFTER they were keyed in as to the variable. If they could not then it may have been a genetic ability issue on some, and unfamiliarity in other cases. Some of the “off flavors” were actually considered to have improved some of the beers.
I have helped and hosted a a good number of classes for judges and it wasn’t uncommon for many folks to not be able to detect a spike at first. Once they were told what it was and what to look for they were better able to perceive it later on. I have brought similar adulterants as well – toffee, actual artificial butter, runt candies, creamed corn, cabbage, raw pumpkin, raw apples…. etc to help them key into the spike.
Myself included, but many people improve later on even with a significant time lapse between the tests. I may have went 30-50% the first time I did one of the classes. I might be 75-85% now, but that is due to physical limitations (pure genetic threshold on some) and on the others it’s my ability to go through a mental check list – is it this? No. Okay maybe this….No no? And so on. I know WHAT to look for.
Ah, nuts! Shoots my idea down in flames. Just passing through…..
BTW, my favorite of theirs is Köstritzer.
Yeah, what did I win? Only one huh, I guess I like my corn, at least in popped form
I only had one beer which I thought had a massive DME problem. Corsendonk Summum Blonde. Now Corsendonk beers are generally OK ish but I think this is brewed specifically for budget supermarkets.
Generally though I am sure that individuals have certain abilities to detect certain things that others cannot. Tea or wine tasters for example. At an everyday level though why do some people describe certain tastes as bad or off. I think I can recognise a well made beer and not like it, but clesrly all tastebuds, olfactory perceptions are just not the same.
The recognised off flavours have a chemical formula, but beer is surely never as simple as a single isolated off flavour.
I love to follow your experiments, thanks.
I think Malcolm makes a good point about genetic ability to perceive certain flavors. I attended an “off favors” class at Raleigh Brewing early this spring and could not perceive DMS in the tainted sample of Pilsner. I also couldn’t perceive it even with the vial of taint directly under my nose. Sadly, the same was also the case for diacetyl. I could perceive the smell of oxidation from a mile away (and often have no difficulty perceiving corked wine when others cannot). I’m also in the percentage that smells asparagus in urine after a delicious dinner of lamb & grilled asparagus.
Funnily enough, DMS is one of the culprits in asparagus urine.
I’ve only really noticed it once, in a commercial beer–a microbrewery’s session IPA. It was not horrible, and maybe just a bad batch, but I thought it was very noticeable.
Please beer with me, as I may meander through my thoughts a bit…
First, I’m also bewildered by people who can, or cannot, pick off flavors. I don’t doubt that people can detect different levels, and others may be either immune to, or hyper-sensitive to certain off flavors. But I’m a bit astonished, when talking to some people about an off-flavor and they simply don’t detect a thing…be it any of the off-flavors. Is it me? Am I being hyper-sensitive? (well, probably…but I’m still surprised they detect nothing!)
Could this be similar to listening to a decent audio system? Where, for instance, someone who has never experienced pinpoint imaging or well constructed soundstage in music reproduction cannot distinguish these things from just casual listening. But then, once they hear it, you cannot un-here the imaging and soundstage elements…and they can recognize it easier going forward. I don’t know?
I’ve tasted DMS in beers before. Sometimes at local breweries, sometimes in commercial draft beer, and sometimes in commercial packaged beer. It doesn’t seem to be a super common off-flavor and seems to just creep up once in a while. It can vary from cooked vegetable/cabbage to oysters. The times I’ve really picked it apart it seems to be very prominent in aroma, but I wonder if I taste it at all (taking my nose out of the equation). Going back to the samples I’ve scrutinized I would easily double down on the aroma comment and double question the flavor.
We were recently drinking a real nice solid canned APA that has great hop flavors (drinks like a fantastic IPA). It had the aroma of oysters…no doubt. The flavor wasn’t there so much…it was easy to drink past the off-flavor and enjoy the beer. I found many off-flavors your able to isolate and drink past…detection, I believe, doesn’t mean undrinkable. After I mentioned the oyster, my buddy agreed and went further with his description. I wonder now if he would have noted or commented if I didn’t say anything.
You can also go into other off-flavors. I’m beginning to wonder if I should ever say anything anymore. I brewery opened up a year ago in the area…real nice place. They’ve got some consistent off-flavors that have been there since the beginning. I happened to mention it, politely, and they were polite but pretty offended. I didn’t mean it that way…but they’ve got plans for wide distribution, and that’s great…hope it works…but I mentioned they need to clean up the taste of their beer. They went on and on about how nobody has detected any off-flavors, they had some beers accepted at certain events and how BJCP judges have never said anything.
So I don’t know
interesting though…thanks,
Dan
Perhaps there are differences in people’s ability to smell/taste dimethyl sulphide levels in the first place. Such observations been raised in studies with broccoli and asparagus urine (some people catch the rotten smell, some don’t). http://www.huffingtonpost.com/2013/07/11/brocoli-taste-obama-bush_n_3575686.html
http://www.npr.org/sections/thesalt/2016/12/14/505420193/we-unravel-the-science-mysteries-of-asparagus-pee
https://udel.edu/~mcdonald/mythasparagusurine.html
Could your testers identify the properly dosed DMS water in the first place? If they couldn’t detect it well in plain water, I suspect they’d never detect it in a beer. Is DMS something you could sniff and identify, or only taste? In other words, I’m wondering if this is not about palates but innate ability levels to taste/smell DMS in the first place. This might explain your strong ability and others inability. Just a thought, as I’m no scientist, and I don’t know if I’ve ever smelled/tasted DMS myself.
I’m wondering if we can sense and taste SMM. hope that there’s SMM sensory kit.
I do believe SMM is either flavor neutral or is so high it is not widely discussed.
Please don’t think I’m being curt by responding with a links, it’s just these blogs are some of my favs for the subject matter and it’s explained so well here –
https://beersensoryscience.wordpress.com/tag/dms/
http://scottjanish.com/how-to-prevent-dms-in-beer/
Thanks for the information. I have made some beers which to me taste a bit off, and the only taste I can link it to use DMS, mainly because I could not detect any other off taste. Maybe it tastes a bit different to me than creamed corn or vegetables but I know I don’t like it. One common thread in these beers where I taste this I know I have cooled slowly while attempting to whirlpool. Maybe I should stop doing this.