![]() |
one of my all-time favorite teas; it was fresh, bright, sweet, fruity, and approachable |
I wrote about this theme long ago (in 2018, here). Before posting this I'll re-read that, but it's really on a slightly different topic. In the past people that's what people called sheng pu'er that was approachable when young (newish), or else "pulong," if it wasn't bitter and astringent. That could have related to some oxidation occurring as a processing input, but the tea character could have worked out that way for whatever varying reasons. Plant type material and processing varies, and climate and weather input does, and so on, so a tea could be approachable and lack challenging forms of aspect intensity for different reasons. One other input could be that it was over-heated, so more similar to green tea, which works out in some comparable ways, but that isn't the same thing.
The main idea, as I remember it, was that sheng pu'er that is sweet, mild, floral, and fruity (often including that range) early on won't usually age as positively, and it's more suitable for drinking within a couple of years of production. Or maybe that was just my take, which could've shifted some since then?
This is about pu'er with actual oxidation input, more specifically about a few related versions like that from Vietnam, so really "pu'er-style tea," given the Yunnan origin naming designation limitation. That naming convention is a bit odd, to me. Matcha isn't restricted, that name, or Oriental Beauty isn't (which is kind of politically incorrect now anyway), and limiting something like Darjeeling to that named origin area is completely different. But I get it, and will just move on.
![]() |
the 2023 (left) and 2024 versions; it may not only be extra age making one darker |
Quang Tom Lao Cai origin examples
This theme comes up a good bit related to South East Asian origin sheng pu'er (or the same tea under whatever name you apply), but not as explicitly as in this case, where the vendor is open about the input. It's written on the label (for a tea I just reviewed, from the Quang Tom producer out of Lao Cai, Vietnam):
Name: Freshly Fermented Green Tea Cake (Green Tea Series)
Ingredients: 100% ancient Shan Tuyet tea buds, sun-dried and fermented.
Storage: Store in a cool, dry place with humidity < 70%, away from direct sunlight.
Shelf life: Suitable for long-term storage under appropriate conditions.
Instructions for use: Take 5g of tea and brew with 250ml of boiling water (>95 degrees Celsius), steep for 30 seconds, then pour into a cup or mug to enjoy.
Food Safety Certificate: 04/2024/NNPTNT Phone: 0961.129.186 Address: Bac Ha Town, Bac Ha District, Lao Cai Province
Produced by: Quang Tom Cooperative, Bac Ha District, Lao Cai Province, Vietnam Production date: 06/2023 Net weight: 350g
Then again it's not really green tea (mentioned on that label), but what else would they call it? Some South East Asian producers call it pu'er anyway, but I can see why more don't. The most common work around is to call SE Asian "pu'er" dark tea, a reference to hei cha, and that doesn't work very well either, because pu'er may or may not be hei cha, depending on how one interprets the category breakdowns.
That reference to fermentation could just mean aging input, change related to fungus and bacteria activity since it was produced, but I'm taking this to relate to an addition of extra oxidation. Whatever words cover those two processes in Vietnamese may not be clearly distinguished through translation, or a decade ago it was pretty common for Chinese producers to refer to oxidation as fermentation (again probably related to a translation error).
I usually use "sheng" to describe the tea type, but if they translate that on to "raw," that actual meaning, then it makes no sense. Raw what? The old Vietnamese name related to the sun-dried form, per some hearsay input, calling it "dried tea," which doesn't really help. I've seen it marketed as snow tea or shan tea, but none of that is a real common type name (and per some recent input some of that is a reference to plant type range there, so I guess to some range within the Assamica variety, not processed tea type). So green sort of works, even though it's not really that.
Related to this tea, to 2022, 2023, and 2024 versions of it that I've reviewed here and here, base flavors of honey and dried apricot stand out, with citrus or other spice range varying by version. That makes it sound like there is some pleasant complexity, and that's true; to me the teas were quite enjoyable, complex, and well-balanced. They are personal favorites. They just lacked typical sheng bitterness and astringency, which for many would be giving up a lot. For people who have only tried factory tea versions it might be hard to relate to that being a problem, but those inputs balance a lot better in higher quality, more whole leaf versions.
In discussing this with Steve of Viet Sun he raised a good point about why some might not like teas like this:
I think they are interesting. And also very easy drinking compared to traditional or semi traditional young sheng.
One of the big things people are into with puerh is the terroir inputs so making reddened puerh really takes a lot of that away.
So you can have teas from different places that taste similar. That kills the experience for a segment of that community.
I agree completely, even though I hadn't been thinking of it that way. A version tasting like pear, honey, dried apricot, and citrus is distinctive, but somehow a lot of the complexity young sheng expresses does seem to get narrowed down in related versions. It can be hard to write out the list of a half dozen distinctive flavors, as people tend to, but even without generating such a list you generally get a sense that there is a lot going on with many forms of sheng, usually seemingly accurately attributed to a narrow origin terroir input.
Bitterness balances the rest in an interesting way, and that astringency adds depth. There might not be more flavors expressed, so complexity could be similar in one sense, but somehow it often seems there is more going on. It's not always completely positive, but in the better half of all cases it all is generally good. Transitions across infusion rounds can vary more, even though that also just depends.
Farmerleaf more oxidized sheng pu'er examples
This theme comes up elsewhere, in other cases than atypical South East Asian tea processing. Here is an online post from Farmerleaf on this subject; that I see as interesting background:
That's labeled as "Partially oxidized Puerh tea: let's try it again!," relating to a 2023 experiment.
William regularly explores and experiments with tea style variations; his channel includes lots of reference to that. To me it's especially interesting related to descriptions of sheng and black tea processing.
Of course I'm not saying that this processing is somewhat identical (in the Farmerleaf experiment and the Quang Tom producer version). That tea, in his video, was oxidized through a very long wither, prior to the frying (sha qing / kill green), and per discussion with Phuoc, the Quang Tom producer, at least some of the oxidation in the other Vietnamese versions occurred after that step instead, and maybe more of it. I'm listening to that video again now as I write this; I lose track of what I've seen.
The darkening of his leaves during the frying step does look similar to the darkened Quang Tom version. It's interesting that William mentions using a low heat to limit the charring of the leaves, in that extra dried conditioned (the withered leaves, left to rest for an extended time, weren't re-wetted). His description goes further; it's really worth watching. He says that roasting the tea can be included but it will simplify the tea character, so it has to be used sparingly, and that he isn't including that extra step for that batch.
I think I want to keep it that way [not pressed, as moacha] because it's an entertaining tea of course but maybe not achieve that a tea Enthusiast would would like to have every day.
He ends up saying more about how processing inputs, the frying approach, changes things than how the extra oxidation makes a difference, but that's especially interesting:
I don't Shuffle it too much because I want the steam to build up in the batch, so I will use minimum shuffling just to prevent any strong burning. And then if we want to make a roast I think we'll try to do that a bit towards the end. But I'm a bit scared of doing that roast because a light roast can really enhance the complexity of the tea, but on the other hand if you do it too much it also simplifies everything, so roasting is a bit like putting sugar in the tea. It's a bit like it somehow when you smell it's more complex but the mouth feel tends to be a bit more simple...
He never did include video content description from when it was finished, in that clip, beyond this comment:
I can smell both and you could say a new long-like fragrance so an oxidized fragrance mixed with some notes of roasting...
![]() |
photo of the leaves from the product listing, probably before sun-drying |
A bit of Google search turned up the product listing:
Spring 2023 Reddened Puerh tea
Processed on April 8th 2023
Ancient tea garden leaves from Nuo Gu Wan, Jingmai village
24 hour-long withering, check the video to know more
This is a continuation of an experiment we made last year. While conventional Pu-erh tea benefits from 2 to 6 hours of withering, we let these leaves on our bamboo baskets for a whole day. This leads to a reddened tea, which has to be processed carefully due to the lack of water in the leaves. After pondering the question over the length of a video, I decided against roasting the tea, and kept the temperature on the lower side during the session.
Still no aspects description there, but we can accept a customer comment as limited feedback:
Nice experiment. I was lucky to get a bag of 70g.
On then first infusion seems to be a white tea, fragrant, elegant smooth sweet thin no astringency and bitterness. With the second infusion it is showing slowly up its raw puer character . I like it very much.
Another comment mentioned the approachable nature, limited astringency and bitterness, and a hint of citrus.
If the Quang Tom producer had used an extra roasting step that could explain the darkened leaves in the 2023 version, since they look to be charred a bit in processing. Or maybe they also pan-fried (kill-greened) partly dried leaves, also using a long wither, and that led to some being singed a little during the frying step. William stressed that you need to be careful to avoid over-roasting, and the related trade-offs he is cited as mentioning here. In a limited form it might be fine, but extending those comments a little going too far it could add a flaw to the tea, not just a tradeoff, but also some actual char effect, as we more often see in roasted Wuyi Yancha oolongs.
You really should watch his video instead of accepting my summary, if those ideas are of interest. He goes further with background and outcome descriptions, which I could cite more of word for word to avoid mischaracterizing that or getting it wrong, but it seems better to just cross reference it here. It's not that long a video, and it's interesting.
That earlier post on oolong pu'er (in 2018)
This tied to a past favorite sheng version that I was talking about just then (in 2018; before I had the same degree of exposure to pu'er experience, only a few years into main exploration then):
A comment about a Moychay Nan Nuo sheng relates directly to that (here on Reddit, by Jay, if he's familiar, who sells tea here):
It'll be interesting to watch it age over several years but my gut instinct is that it will fade rather quickly in BKK. The fruity flavors often indicate oolong style pu that will fade out with age.
I had framed a similar idea in that original review more as a question.
I love the fruit in it, and that overall "bright" effect. Often when I'm drinking sheng made within the past year I'm saying this might be better in a year or two, a little less edgy, but in this case I'm not so sure. If that brightness were to decline, as it would to some extent, the balance might be just as positive, or more so, or the tea could've been best drank when very young like this.
All that is probably as good a subject theme intro as what I've written. That tea did die a couple of years later; it shifted to include warmer and deeper tones, and lost the brightness and fruit, most of what was positive about it. If my tea budget had been deeper I would've re-bought that, or the next year's version, to try aging what I didn't drink right away, but I didn't. Kind of unrelated background, I bought that tea in a shop in St. Petersburg at New Years time in 2017-18.
![]() |
so many amazing sights on that Russia trip, but this captures my favorite part |
![]() |
the more iconic scene version |
I retried that 2017 Moychay Nan Nuo in 2020, documenting where it was then, and it was fine as a 3 year old version, but as I remember from trying it since it was going quiet.
It seems like a good place for me to add the next source of evidence, something that really brings all this into clearer focus, related to a tea I've tried over a decade of aging. I've tried Thai versions that were somewhat approachable early on that I've experienced changing over a few years, but the starting point was never the same (really bright, sweet, and fruity, with really limited bitterness and astringency), and I haven't tried any aged longer than this last 4 year example (nearly that; it will be 4 in 4 more months). I've probably tried a good bit that was older that had already went dead; I mean examples of the entire sequence.
I've tried at least a couple versions of somewhat older Wawee Tea Thai sheng, but if anything their versions run a little closer to standard Yunnan style, so that really doesn't work. Of course the same applies to Hong Tai Chang versions, if that is familiar.
Other cited input in that Tea in the Ancient World "Oolong Pu'er" post
Returning to this post section theme, I re-read that 2018 post from this blog, and it's interesting (kind of new to me too, for being that far back). It includes input from William, again of Farmerleaf, citing him as the Bannacha owner. His 2023 thoughts on the subject represent what he now thinks, instead of from way back when, but I'll cite part of it to get a feel for the input range:
By controlling the Sha Qing time and the wok temperature, you can make redder or greener tea, and, as long as it's not extreme, I think there's a large span of well processed tea. Then, it depends on your personal taste, what kind of aroma you prefer, and which mountain you're dealing with. I like a greener processing on Mengku tea, because it brings out the high pitched fragrance of this area. Redder tea might be good to emphasize on the body, the mouth feeling.
The subject of tea community discussion history is interesting to me (which is something of a tangent here). It's no secret that William re-branded to Farmerleaf as a new business when moving to Yunnan; this is from the Farmerleaf "about us" page:
Farmerleaf was created in 2016; it is the continuation of our efforts to provide high-quality teas worldwide. We started selling tea online in 2011 with www.bannacha.com which is based in France and offers mostly Pu-erh tea. We are now based in Puer city, Yunnan, in which we have our office and storehouse. We have chosen this city because it is in the center of the Pu-erh tea production areas and the region produces a wide range of white, green and black teas, it is also a good place for storing Pu-erh tea.
So some of that blog post is about discussion of this theme on Tea Chat prior to 2016, written about in 2018. This post reviews a 2016 Jing Mai Miyun version from Farmerleaf in March of 2017; maybe I knew how it all connected in 2018, that his business had either renamed or reset, or maybe that had got by me.
It's all an interesting read, to me, that old oolong pu'er post, but it's lots of mixed input from lots of sources. People talk about primary flavors, body / feel, and aroma varying, probably using terminology in comparable but perhaps slightly different ways. Shah82 passes on input in an extended citation, that doesn't narrow down to a short phrase that I can pass on about this subject here. His accounts of his experiences were always interesting. And probably still are; after checking in on that shaving forum he had discussed tea in, Badger and Blade, the last post by Shah8 (probably the same Shah) was last Monday, on page 401 of a sheng of the day thread.
Here is what he said about "oolong pu'er" in that post citation (of Tea Chat discussion):
I will say that taste itself is highly deceptive. It depends on your water, whether you let the dry leave sit in the bowl a week or two, or any number of things. It can change with the years. The classic way people buy bad tea is by preferentially buying "approachable" flavor or aroma puerh. Which is why you see so much problematically tinkered puerh that gives a oolongly fruity or floral taste (or more red and mellow malt).
For me, the easy way to tell, again, is to see whether that aroma *endures*. Many of these teas also will betray badness by becoming hard to drink or pointless to drink by about brew 6-8. *Few* puerh, however, are done straight processing green. A little butteriness is sought after. Or a little hongcha, or some nice smokiness. That's alright, so long as most of the underlying qualities are still there to age.
To me the rest doesn't condense down to ideas that I can simplify and clarify here. Everyone isn't saying exactly the same things. Intuitively the tea would become less like normal sheng for including this processing variation, and less like green tea by extension of a sort of relationship between the two types, and more like black tea. That doesn't really work as a summary. Bitterness and astringency decrease, and mouthfeel and flavor profile changes, in the ways described here, and in varying ways mentioned in discussions.
One interesting recurring theme was that overheated sheng, which would be much closer to green tea than oolong, could get mixed in with that category. The typical flavor range wouldn't be the same, but the tea would be aromatic and fresh, losing bitterness and astringency (although it would retain a different kind of edge), and it wouldn't age in the same way. Put another way it would include a different balance of main compounds than conventional sheng. It still might seem like "oolong pu'er" to most people, but really it would be more of a hybrid matching green tea range.
I tried a version of Thai sheng from Aran tea that reminded me of this theme, reviewed here in 2020. They called that pu'er, related to the naming convention issue I covered earlier. It might sound like I'm saying that they don't know how to make sheng properly, which I don't intend, but even read that way they could have adjusted processing approach in the past 6 years. I just keep repeating that the character reminds me of green tea in that review, and guess that it might not age-transition well. I liked the tea; something in between normal sheng and green style would just be heated slightly more in pan-frying, and it could still be pleasant, novel, and fresh, as that was.
Conclusions:
So do these teas ever retain great aging potential, in any form? I'm not sure. The 2022 Quang Tom is still pretty good, just shifting to warmer flavor tones, which seems fine, except that I really liked the brighter, fruitier versions earlier.
How did William and Farmerleaf's versions work out? There's a little on that in that section, a website comment, but trying such a tea fills in a lot more.
To me it's an even more interesting question if moderate aging ever works really well. I've tried plenty of Yunnan sheng that worked well aged in the 5 to 7 year range under much cooler and drier conditions than in Bangkok. Here is an example: a 2021 Tea Mania Gua Feng Zhai version.
I don't keep track of any but the most common village names, and I suppose that could be one, with it presented as from an in-demand Yiwu local area. That tea is still bright, fresh, and sweet, after five years. That probably wouldn't be true if it had spent those years in Bangkok, where it's always 25 - 38 C (mid-70s to 100 F), and almost always 65 to 80% relative humidity. At the coolest and driest times maybe only 60% RH, but that comes and goes fast.
Of course I'm not claiming that the Yiwu / Yunnan tea version I just mentioned was "oolong pu'er," that it was extra oxidized, or overheated in a kill-green step (so more like green tea), or that it doesn't have great aging potential (although a good number of Yiwu origin versions that I've tried didn't). It seems likely that there is a direct opposition between versions being approachable early on, bright and sweet, highly floral or fruity, and lacking significant bitterness and astringency, or full and structured mouthfeel. That tea had good feel at 5 years along, as I remember. This is a mention of that, in that post:
...Spice range definitely picks up, but it's broader than the sassafras was last round... Then that's rounded out with root spice, more like ginseng, maybe just not quite as punchy and medicinal as ginseng (which is also subtle, in a different sense). It has changed a lot over the last three rounds [it had shifted from early pear and dried mango range on to floral scope, and then warmer spice tones].
Of course these flavors link well with the pronounced base mineral tone, which isn't dropping out. Feel and aftertaste are significant, but those can be more intense in some other versions. Overall balance is the nicest part, the way it all comes together each round...
So none of this condenses down to a simple generality, beyond that one (approachable new sheng probably won't age transition as positively, at some point). But that is only part of a broader story, that would vary depending on why the tea is like that, and that generality could be wrong in some cases.
I think the positive complexity that we expect from sheng, as it ages / fermentation transitions over 20 years or so, wouldn't be present, across many different cases. That last review excerpt only implies how that goes in lots of cases, mentioning warmer spice range entering in, but that related to a more standard form sheng. I haven't clearly established that "oolong pu'er" will always just go dead, even though to me that's implied quite a bit here, with sources openly stating that. William--again of Farmerleaf--explicitly stated that he was curious about this himself, in that video, wondering how the last experiment before that one would transition later on.
If a somewhat oxidized "oolong pu'er" example didn't balance well at 5 to 7 years along it still might not be clear if it's going dead or just in between character forms. I would expect that I could tell the difference, from the interim aspects form, but it might not be as clear as I'd expect.
None of this relates to that 2018 Xiaguan example, surely; the starting point must have been very different. I could re-try the Nan Nuo version I referenced at the beginning to add more related background, but for sure it was never oxidized to the degree these Vietnamese versions are. These are "oolong pu'er" in two different senses; that Nan Nuo one related to it being approachable, for whatever reasons, and then these Quang Tom teas are intentionally quite oxidized.
The "somewhat green" overheated during pan-frying pu'er theme is surely a third and different thing. One would expect that to age even more poorly than the other two types (oxidized sheng, and sheng that just happens to be approachable early on, due to mixed other inputs). But it would be interesting to try a couple of versions to know better.


.jpg)



No comments:
Post a Comment