Supercycler opened to the public on 28 April 2026. This first report runs to the end of July and focuses on what we are actually researching: which light cycles people pick when they can pick any, and what happens in those grows.
One comes out every month.
Each section is calculated over the zones that filled in that particular field, and they are different subsets: 275 have a photoperiod configured, 62 described their lighting, 53 have their climate evaluated, 49 described their substrate. Against 918 active zones, it is always a part.
That part is not a random sample. Whoever connects sensors, describes their room and keeps a journal is, almost by definition, whoever grows more methodically. It is reasonable to expect the network as a whole to look less like these numbers than the size of each cut would suggest. Every chart states how many zones it covers, and it is worth looking.
June brought 427 sign-ups, almost six times May; July closed with 230. During July, 306 of the 740 members logged in at least once — 41%.
Opening a zone and getting it measured are different things: of the 918 active ones, 282 have a device connected, 180 keep a journal and 101 have a sensor assigned. The rest of this report is about that instrumented subset, which is the one that leaves data behind.
The map is recalculated every day, so it does not match the report window: it shows where the network is today. We group by city and a dot only appears once there are at least five members; the dot sits at the centre of the city, never on an installation. The panel beside it is the opposite — frozen data as of 31 July: what light cycle each region runs, which is what the map cannot tell you.
The map could not load. Per-country totals are still available in the list.
Regions that do not reach the minimum number of cases on their own are grouped together. Base: 272 of the 275 zones with a photoperiod; three could not be placed in a region.
Data is shown grouped and may update with some delay. Cities or regions with fewer than 5 members are grouped or hidden. We do not publish anyone's location or activity.
This is the heart of the report. Supercycler lets you define a light cycle of any length, not just 24 hours, and the question that organises everything else is what people do with that freedom.
Of the 275 active photoperiods, 152 run a day that is not 24 hours long.
275 photoperiods configured
207 zones
65 zones
The proportion holds across both stages: 57% of zones in vegetative and 54% of those in flowering run a long day. This is not a handful of people experimenting, nor something only tried in one stage.
It is not a local phenomenon either. The panel next to the map, further up, carries the same cut by region: in the Southern Cone — where 84% of the zones are — supercycles reach 53%, but in Southern Europe and the rest of the world they climb to 67% and 71%. With 15 and 28 zones those two figures are fragile and may move a lot next month, so read them as a hint rather than an established difference.
A photoperiod is two numbers: how many hours of light and how many of dark. Added up they give the length of the day. There are 38 distinct combinations in use; these seven cover 226 of the 275 zones and the rest are grouped at the end:
13 hours of light and 13 of dark, on a 26-hour day, is the most used setup on the network: 93 zones. Then comes the classic 12/12 with 49, and vegetative's classic 18/6 with 44.
13/13 keeps the half-and-half ratio of flowering's 12/12, but stretches the day: each cycle delivers one more hour of light and one more hour of dark. The plant gets more of both without the relationship between them changing.
The cost is that the cycle drifts away from the solar day: after twelve 26-hour cycles, the lights come on at night. That is why a supercycle needs automation and cannot be kept up by hand.
207 zones · the classic is 18/6
65 zones · the classic is 12/12
Here is the most striking part of the cut. In flowering the classic 12/12 still comes first, barely ahead of 13/13 (22 against 19). But in vegetative, 13/13 beats 18/6 by almost double: 73 zones against 42.
Put another way: whoever adopts a supercycle in vegetative does not choose to lengthen the light period, which is what you would expect at that stage, but to split it half and half and stretch the whole cycle. The pattern is fairly uniform, though this data does not say why it gets picked — it only shows what ended up configured. That is what the next reports will be able to start cross-checking against results.
After 24 and 26 hours the count drops off fast. The tail runs to 56 hours, with one or two zones at each value: those are individual trials, not a trend.
A protocol is the grow's task calendar: what to do each week depending on the stage. Of the 172 journals opened, 154 have one assigned. These are the most picked:
The first four take most of it, and they follow the shape of the grow rather than any technical preference: one for vegetative, two for flowering and one for autoflowers. Behind them come the service rooms — mothers, clones, germination, drying — which are what people with more than one zone use.
Technique protocols and modifiers are still barely used: five journals in total across SCROG, organic and the 13/13 supercycle modifier. The cycle gets configured on the zone, not on the protocol.
Whoever lays out their zone in My Lab describes the space and the fixtures they have. 62 zones did, declaring 405 fixtures in total. It is a small sample against 918 active zones, but it is the best first-hand description the network has.
405 fixtures declared across 62 zones
LED took practically all of it: 371 of the 405 fixtures. HPS, metal halide and LEC appear in so few units that they do not even reach the minimum number of cases we publish separately, so they are grouped together with the ones left unspecified.
Adding up every fixture in each zone:
The bulk sits between 300 and 600 W, which fits a tent with 2 to 4 plants. The second group stands out: 18 zones go past 1000 W, almost a third of those that filled the field in. Weighted by number of units, the average fixture is 170 W, and the total declared power adds up to 68,677 W.
Adding up every zone that declared its lighting and also has a photoperiod running, the network drives 61.5 kW of lighting under automatic control, spread across 52 zones. That is 90% of the declared power: almost all the light the community described is in fact automated.
Light is the first device people connect and often the only one: 240 main lights against 40 irrigation pumps. That makes sense next to section 2 — a 26-hour cycle cannot be kept up by hand — but it shows the rest of the climate is still handled manually.
The scale is home-grow scale: 872 of the 918 zones declare between 2 and 4 plants, and only 22 declare ten or more.
Connecting a device and automating it are different things. Across the 354 active devices, this is what decides when each one turns on:
130 devices follow their zone's photoperiod — almost all of them lights — and another 46 respond to a sensor or a timer. But 177 devices, half the total, are still switched by hand: they are connected to the platform and visible in the app, but nobody gave them a rule.
There is a lot of room there. They are mostly the climate devices — ventilation, humidifiers, heaters — and it lines up with what shows up in section 8: humidity is precisely what drifts out of range the most.
133 sensors in total. A single unit usually measures more than one thing:
Temperature and air humidity come together in almost every sensor and cover practically the whole fleet. What barely gets measured is everything else: 13 substrate moisture sensors and 10 CO₂ sensors across the entire network. Without a probe in the substrate there is no way to know how much a plant actually took up, and that is why this report's irrigation cut is so small.
49 zones described their substrate in detail: the base medium, how biologically active it is and which amendments it carries. Not all of them filled in every field, so each chart states how many zones it covers. It is the smallest cut in the report and the most specific.
44 of the 49 zones filled this in
47 of the 49 zones filled this in
There is no dominant medium: peat, commercial mix, compost and coco split into similar shares. What does mark a trend is the other half of the picture: 47% describe their substrate as living or biologically active, against 17% working in inert media.
Across those same zones, how many declare each amendment:
The first four are all biological: mycorrhizae, worm castings, compost and trichoderma show up in close to half the zones that described their substrate. The corrective minerals — dolomite, gypsum, azomite — sit well behind.
This is a community that works the soil before the nutrient solution, and that talks to the next section: when the scanner finds a problem, it is almost always a deficiency.
When something looks off, the grower takes a photo and the system analyses it. Over the period there were 846 health scans by 167 growers.
What follows are suspicions, not diagnoses. The analysis is done by an AI model on a single photo, with no soil or tissue sample: accurate enough to point you somewhere, not to confirm anything. The percentages describe what the system suspected, not what the plants had.
846 scans by 167 growers
Almost a third of the scans find nothing: the plant is fine and the check served to rule things out. Of the rest, the system suspected a nutrient deficiency in 49% of everything scanned, far ahead of pests, which together come to 8%.
Across the 411 scans with a deficiency, counting every time a nutrient is named in the finding:
Magnesium and nitrogen take almost everything. It needs reading carefully: of those 411 scans, 266 name more than one nutrient — typically "nitrogen or magnesium" — because in a photo the two look alike; 131 point at a single one and 14 mention a deficiency without saying which. That is why the totals above add up to more than the number of scans.
Of the 846 scans, 70 ended in a pest: 8% of the total.
Spider mites are the dominant pest by a distance: 46 scans, more than double fungi and leaf spots. Thrips, whitefly and aphids show up so rarely that they are grouped; on their own none reaches the minimum number of cases we publish.
Traps are photographed separately and the system counts the insects stuck to them. There were 52 traps over the period, across 36 zones. Here each number is an insect counted, not a scan:
The order flips: on the traps whitefly is the most numerous with 355 individuals, even though it shows up in very few plant scans. They are two different things and worth not mixing — the trap measures what flies through the air of the tent and the scan measures what is already on the leaf. A flying pest saturates a trap long before it becomes visible on a plant; spider mites, which do not fly, barely land on traps and yet are what gets diagnosed the most.
A third of the trap counts came back unidentified: insects the system counts but cannot classify confidently from the photo.
We do not break pests down by region. Outside Argentina no country reaches the minimum number of cases our anonymity rule requires, and publishing "three zones with spider mites" in a small country points too closely.
226 growers opened 939 conversations with the assistant and wrote 2,532 messages. To know what they ask about we do not read what they wrote: we look at which tool the assistant ended up executing to answer, which is what the question was really asking for. That came to 1,538 executions.
1,538 tools executed · 226 growers
Almost half the queries are about status: what is happening in a zone right now, which devices are there, when the next light change is due. The assistant is used mostly as a dashboard you can talk to, rather than as an agronomy consultant.
The second block is different and worth separating out: searching the documentation. It is the query that reaches the most distinct growers — 119 — even if it is not the most repeated. Plenty of people ask the assistant how the platform works before asking about their grow.
Further down the list are the queries that interest us most looking ahead: 33 about fertilisation, 32 about pests and treatment, 26 asking for a hardware recommendation and another 26 changing the light schedule straight from the chat. Still few, but they are the ones that show the AI being used to decide and not only to look.
We neither read nor classify the content of the conversations. This whole cut comes from the name of the tool the assistant executed, which is a system record and not anyone's text.
Every zone has two ranges set by its grower: an ideal one and a critical one. Across 6,902 complete days from 53 zones, this is how the time split.
By time in the ideal range all three look alike, around 40%. The difference is at the extreme: temperature reaches risky values 8% of the time and humidity, 25%. It connects with section 4: almost nobody has their climate automated, and humidity is exactly what is hardest to hold by hand.
It stands as the baseline. If that figure drops in the next reports, the network improved, and it can be verified month by month without waiting for any harvest.
Being outside the ideal range is not being in trouble: the ideal is the tighter of the two, and stepping out of it means being uncomfortable. That is why the chart separates the yellow band from the pink one.
The 0.7% of critical VPD does not say VPD is fine: almost nobody configures a critical VPD range, so there is barely anything to cross. That number describes the configuration, not the grow.
Everything above is a snapshot closed on 31 July and it does not change. What follows is the opposite: it is recalculated every day, so it does not carry the same date as the rest of the report and it is not usable for citing the period.
Every month we gather the activity logged across the network: cycles started, new members and grow records. The series are kept apart so each number answers a single question and can be read cleanly.
Once a zone is connected, the system can read its environment, decide on a response and send it to the right device. These figures show how much gets measured, how many actions run and how many decisions are verified across the network. They are not promises: they are activity records.
Average of the last 30 days. The page updates automatically from the real system logs.
Connecting a device turns a routine into a process that can be measured and repeated. From then on the system can log the environment, run actions and leave evidence of what happened.
The underlying question — whether a 26-hour day yields differently from a 24-hour one — needs harvests closed and filled in. As of 31 July there are nine, and a single one with the yield recorded.
| Data | What there is | What would be needed |
|---|---|---|
| Grows with a harvest date | 9 | Only 2 with consistent dates |
| Harvests with the yield filled in | 1 | Around 30 per group to compare |
| Potency measurements | 0 | The integration exists and has not been used yet |
| Complete grow closures | 3 | It is the format that enables the comparison |
The main reason is the calendar: the network opened on 28 April and a full cycle takes longer than that. The first harvest with consistent dates ran 86 days. At that pace, the first comparable results arrive towards the end of the year.
We are not publishing an average cycle length either. Reviewing those nine records turned up harvest dates earlier than the start date and grows marked as harvested that are still active.
In the meantime there is already something to work with: two groups of similar size, 152 zones against 120, growing in parallel. It is not a controlled trial — nobody assigned anyone to a group, and the zones also differ in light, substrate and climate — but it is an observational base large enough to start looking for signals once the harvests arrive.
The data is taken from the network's grows and aggregated before being published. These are the rules each report is built with.
No equipment manufacturer, fertiliser, breeder or commercial protocol is named. What gear the network uses is not public information.
Everything published is totals and averages, and no group with fewer than five cases is broken out. No number corresponds to one member or one zone in particular.
The data is as of 31 July 2026 and is not modified afterwards. Corrections are published as a dated erratum.
Every number comes from a query saved alongside the report. The definitions and denominators are in the methodology annex.
The findings in section 6 are automated readings of a photo, not a lab analysis: they describe what the system suspects, with the margin of error that implies.
The environment cut covers 75% of the days with a configured range; in the rest, the field that separates "outside ideal" from "critical" is missing, because we started storing it later. That coverage grows with every report.
Data from before 28 April is excluded — it belongs to our own development zones — as are the automated accounts the app stores create when reviewing each release.
The methodology annex has the definition of every number, how many cases it is calculated over, and the log of corrections.
For the live state of the network, rather than this July snapshot, there is the live block, recalculated every day.
The next report comes out at the beginning of September. If you would like us to let you know, write to us using the form below.
This is the first snapshot and it works as a starting point. Every zone that begins measuring and every cycle that closes makes next month's number say a little more.