Let’s look at 4iiii’s latest power meter from last year. This is a review of the dual-sided Precision 3+ Pro on the 165mm Dura-Ace R9200 12-speed crankset from my Cervelo S5 – my main bike, so it had better be accurate!
Two central claims stand out – 1% error and 550 hours of use. Whilst that accuracy claim is par for the course for a top-end power meter, over 500 hours of use is market-leading.
Two compelling bonus features are ‘Auto Terrain Selection’ and Apple FindMy integration. All the other features are comparable to the competition, so at least on paper, this stands out as a product I could use.
Heads up: It’s a $500 freebie from 4iiii on my crankset. Make of that what you will!
4iiii Precision 3+ Pro Power Meter - 1-minute Review
89%
Provisional Summary
Accurate overall, but the left/right balance seems skewed. Actual battery life is great but doesn’t match the claim of 550 hours.
The Apple FindMy is a great bonus feature for me, and there’s a great way to keep the overall cost down with a factory installation option.
You can quickly check out this comparison chart to see that the headline improvements are battery life, Terrain Selector, and FindMy. Although nuanced cadence limit support and weight differences exist, other features are similar.
Image: 4iiii (2025)
4iiii Precision 3+ Pro – Visuals and a quick once-over
Much of the drive-side power meter and the entire left side are hidden, though the branded drive-side pod remains visible inside the chainrings.
The left-side CR2032 battery is accessed via a twist mechanism, while the right-side requires three small screws to be undone. Each CR2032 battery is inserted with the ‘+’ facing outwards.
The design of both sides’ caps could be improved. The left-side cap is easy to replace incorrectly, risking the loss of both the cap and battery. The right-side pod is more time-consuming to open, needs a screwdriver, and overtightening the screws will break the cap. Both sides have integrated waterproofing rather than easily lost circular washers used in other brands – great!
The left-side pod offers plenty of clearance from my frame (chainstays), though there is a tiny risk of debris getting caught in the gap and pulling off the power meter. The right-side pod has 5mm of movement, which surprised me. However, it seems unlikely that it would flex repeatedly or that anything would get caught around it.
Supported Cranksets
Power meters are hard enough to choose and buy at the best times. 4iiii offers a wide range of options that will probably fulfil your need to work with a modern crankset. Still, there are too many options for me to quickly summarise them here, especially if you consider what is available as a factory installation vs new. You’ll need to think which of these options you want or need
Crankset brand – Shimano is the most common, and FSA, SRAM, and Cannondale are also offered.
Then, which ‘quality’ do you want – Dura-Ace is the pro-grade I’m using, but Ultegra and Shimano 105 are perfectly fine for competitive amateurs. GRX and XTR should be considered for non-roadies.
Crankset model – do you want 12-speed or 11-speed, and which crank length? 172.5 is most commonly sold, but the trend is to go shorter; 165 mm will probably work for anyone under six feet. Further complications are that a 12-speed Shimano crankset will work on an 11-speed setup (I do it)
Do you want single-sided (left only) or dual-sided or an upgrade from single-sided to dual-sided?
Single-sided cranks are a great entry-level option – they’re cheap, and the data is actionable but best suited if you only have one bike. Get 4iiii on a Shimano 105 L.
What chainrings do you want? 50:34T is probably best if you don’t know what this means – dealers usually offer bigger chainrings.
Do you want to buy a whole crankset or add the power meter to your current crankset at the factory?
4iiii’s older models offer wider compatibility with older model cranksets.
In the Box – What you get
Typically, with a 4iii dual-sided power meter, you would get the entire crankset, including chain rings and batteries. You don’t get pedals. If you go for the factory-install options, you get back what you sent.
Installation
Installation of a dual-sided crankset should take no more than 30 minutes and as little as 5 minutes. The caveat is that things can go wrong, and then you need special tools. You probably should use a torque wrench.
It’s a good bet that a single-sided installation will be hassle-free, and you can get by with a single Allen wrench, a hammer and a bit of wood to stop the hammer from damaging the axle. That installation should take 5-10 minutes.
You probably want a unified channel, even if you’ve no idea what that means 🙂
With a Unified Channel setup, the right-side power is sent to the left side, which transmits both sides of the data to your head unit. You only pair your bike computer with the left side. In rare circumstances, you might need to have each side as a separate power meter, and then your sports device will combine them.
Each side of the power meter has a different ANT+ID; the easy way to remember which one to use is to think of people who only have a single-sided PM. The process needs to be the same for everyone, so we all pair the left side.
The first issue is that my crankset came without the two sides linked. I had to download the 4iiii app, associate the two IDs, and calibrate with the app for good measure. Apple FindMy also needs pairing to your iPhone.
The 4iiii App
4iiii’s app is straightforward to use.
Check your two power meters are linked (paired) together. If not, pair them in the app. Place the cranks vertically and check it calibrates. I get a result of 1010 – a ’10’ from each side is the correct result.
There’s an optional tweak you can make in the app. If you have many sports sensors, giving the 4iiii power meter a recognisable name makes sense – I chose 4iiii3. That will be the name that appears when I want to pair it, and it will also be the name that pops up on your Garmin at the start of a ride to say it’s connected and prompt you to do a pre-ride calibration. I also renamed the right side to remind me not to use it for pairing.
Scaling Factor
A scaling factor option exists, but you typically want to avoid it. That can be used to fine-tune the 4iiii to align with another power meter or trainer you might have.
What is the Terrain Selector, and how do I set it?
Terrain Selector is a fully automated feature. It’s good to know it exists!
When you are pedalling on smooth surfaces, the details of the circular motion are predictable, meaning that less frequent readings are needed, and battery life can be better preserved. Put another way, the power meter uses more battery juice on bumpy roads.
If you ride on bumpier roads like I tend to, this might explain why the battery life is significantly less than the 500+ hours you were expecting. Anecdotally, I may have noticed this in my smooth rides around Richmond Park, where the battery life hardly changed compared to Surrey rides, where the needle shifted notably after even a 2-hour ride.
What is the Temperature compensation, and how do I use it?
Again, this automated feature normalizes the effect of temperature changes on the strain gauge’s resistance, ensuring accurate power readings. It’s a fairly standard feature handy over longer rides with significant temperature variations.
Setting Up Apple FindMy
Remove and replace the battery on your left-side PRECISION 3+ Powermeter three or more times to enable Apple Find My pairing mode. Use the Apple Find My app to add your PRECISION 3+ Powermeter as an ‘Other Supported Item’.
You must use the 4iii app to reset the FindMy feature, for example, if you sell the crankset.
How Apple FindMy works with 4iiii Precision
If you share your FindMy with friends or family, it’s an excellent tool for specific scenarios. However, it doesn’t offer live tracking like Garmin’s LiveTrack or Strava Beacon, which require an open smartphone app and web access. FindMy, on the other hand, doesn’t need that.
Theft: A thief is unlikely to realize the power meter has FindMy capability and would probably look for an AirTag hidden elsewhere on the bike. As far as I know, FindMy cannot be reset to a new owner, making your crankset unsellable by a thief.
Peloton Live Tracking Without an iPhone: If, like me, you don’t like carrying an expensive phone, you don’t even have to ask your cycling buddies if you can share their data plan while you ride together. FindMy will send your location to your friends and family back home at regular intervals (every few minutes) via their iPhone.
Location with a Dead Phone Battery: If your phone’s battery dies while an app is running, any nearby stranger with an iPhone will update your location as long as they stay near you for about a minute.
Battery Issues
Several reports suggest battery life is significantly lower than the claimed duration.
I observed the same issue when checking battery levels frequently in the 4iiii app, with a loss of several percentage points over a few hours of riding, suggesting a battery life of around 50 hours. However, the battery loss wasn’t consistent, so it’s unclear whether this was due to incorrect calibration and reporting by the app (meaning it’s not a problem), terrain compensation during an especially bumpy ride, or something else.
My guess, at the moment, is that I will get over 120 hours. That’ll do nicely.
4iiii’s solution to battery drain issues is to reboot the pod, which can be done by incorrectly replacing the battery a few times on each side.
Daily Usage
Initially, I tested the 4iiii Precision 3+ on a Wahoo Bolt/Roam/Ace, Forerunner 965, Edge 540, and Zwift as a backup power meter on MyWhoosh. It seemed fine.
Once satisfied, I made it my main power meter, primarily using it with the Wahoo Bolt. I relied on its data, recording backup data elsewhere with my ageing Assioma pedals as my “gold standard” reference.
I calibrate before every ride, though that might not be strictly necessary. Once the Bolt is started, it takes about 15-20 seconds to pair and recognize the 4iiii, adding up to roughly a minute in total, which is annoying. Assuming I’ve angled the cranks correctly, I consistently get a 1010 calibration result on the first try.
10 – Zero offset successful, ready to ride
20 – Unstable data (bike not stationary)
30 – Low battery (less than 10%)
50 – Calibration error – contact support
99 – Communication error – try reconnecting with the 4iiii App
With the 3-second power display on my Wahoo, the numbers react quickly to changes in effort when riding. The displayed power may take slightly more than 3 seconds to match your actual output due to the moving average calculation and slight delays in transmission and processing. This delay is generally not noticeable, though.
One unique thing about Wahoo devices is what happens after an extended coffee break. Sometimes, but not always, the power meter won’t reconnect when you start pedalling. I believe this is a longstanding bug with Wahoo and its failure to wake up the connection to the power meter (not just 4iiii). The trick is to press the left-side button to go to the ‘settings’ menu and leave it there for about a minute as you continue riding. Press it again, and the connection should re-establish. This is quite annoying, but it’s especially troublesome in a triathlon—I typically set my Bolt/Edge running in transition before the race starts, then press lap when I begin riding. If the connection doesn’t wake up, I get no data. Grrrr.
Daily Stats
You get cadence and dual-sided data for power, pedal smoothness (PS), and torque effectiveness (TE). That data is transmitted over both two Bluetooth channels and over ANT+, although PS and TE are ANT+ only.
Q: Bluetooth or ANT+?
A: It doesn’t usually matter
ANT+ is significantly easier if you have multiple recording devices (I use this)
ANT+ might use slightly less power
BLE is more secure
BLE might pair quicker
ANT+ is generally more stable
ANT+ gives the extra data metrics
Both might conflict with 2.4GHz WiFi indoors, causing dropouts
BLE might be the only option, e.g. for smartphone apps or Apple Watch.
Q: Are the extra cycling dynamics stats useful
A: Probably not.
Why power is useful.
Appropriate power averages for your riding style will enable you to pace your efforts more efficiently. After your workout, power durations are used to determine complex performance and recovery metrics – all of which use the basic power data and offer some great insights.
Cadence can be helpful. Generally, you want a higher cadence to avoid muscular fatigue at any given power level. You have to train this ability over several months.
PS and TE are perhaps ‘occasionally interesting’ and unactionable for most people. If you pedal with a more correct technique using more muscle groups, it will be more efficient if you are trained. Looking at how you apply the power throughout the stroke might lead you to think that shorter cranks or oval chainrings will be better for you (4iiii data is incompatible with oval chainrings, only ROTOR and maybe ASSIOMA will work from the more common brands)
Dual-sided cycling dynamics data is even less actionable. Correcting imbalances is very hard, but it might prompt you to do a good thing and go for a bike fit or go to the gym to train asymmetrically.
Data Accuracy Tests
A significant advantage of 4iiii’s data is that it collects data from each side, so it has a better chance of being correct than data taken from one side and doubled. A pair of power meter pedals shares the same theoretical advantage and uses the exact data collection mechanism on each side. The issue for 4iiii is that each side has a different construction and installation – even though the actual strain gauge may or may not be identical (IDK).
Tl;DR – superficially, this appears to be an accurate power meter, but the left-right balance seems wrong. I suspect (IDK) it was factory-tweaked during installation to get the system-level data correct.
I’ll caveat that by saying I’ve probably tested about 20 different power meters, whereas someone like Dcrainmaker will have tested significantly more. I’ve had issues with widely trusted brands, which I’ve not written about due to lack of time and inclination. It must be an issue that some testers get ‘special’ models – not necessarily better components but just ones validated before shipment (more than simple factory calibration). I’m sure I hardly ever get such equipment (hence the issues I’ve had over the years). However, this one from 4iiii was, I think, their first UK Factory Install, so it might have had more factory validation than was routinely undertaken by other brands. I know dcrainmaker sometimes buys PMs from retail stores to avoid precisely this issue. I specifically noted to the factory installer that the PM would be used for a published test – primarily because I planned to use it afterwards if it was OK!
power Accuracy Testing
I’m happy with this first stage of indoor and outdoor testing against Kickr Move and Assioma Duo – two devices I generally trust but which are not without the occasional feuble. I have some new power meter pedals lined up that I’ll use later in 2025 to re-test the balance info (ping me in June in the comments if I haven’t done that). I’m reasonably confident about the data I get from Assioma, but the pedals are old, so I have a lingering doubt about their current accuracy.
Testing and Accuracy – what’s important to me.
I want the numbers I see to be meaningful and actionable. Essentially, that means accurate. But I suspect many happy cyclists are improving their fitness with one single-sided power meter that is 3-5% inaccurate.
I mostly ride outside on bumpy roads, but I find it annoying when cycling indoors if my Kickr data doesn’t match. Have I moved the trainer? Is the drive chain misaligned, giving more drivetrain loss than expected, and so on – the annoying bit is that I like things to be correct, but it often takes SO much time to achieve that, so I compromise in some way. I have a few sports data platforms where I consider my data complete and nearly correct. I periodically swap different power meters as the source for that data, but only when I trust the source.
Cadence data accuracy.
This is boring but important data. Cadence is an input to the wattage calculation, so the power might be wrong if the cadence is wrong.
This is an indoor trainer ride with varying cadences. Averaging is set to zero, which makes the dropouts more clear.
This appears to be a problem. However, the dropouts are at very low power or simply represent not pedalling. We can’t compare the accuracy of KICKR cadence as it is estimated. Maybe the Assioma is reading cadence too highly here? Otherwise, it doesn’t seem to coincide with errors in power, which I’ll discuss later.
The following chart is the cadence from an outdoor ride on the bumpiest part of the day.
There seems to be a similar issue, again coinciding with low power. Again, this made me think this was simply a bit of coasting and that Assioma might be wrong.
Let’s see what happens with the power data
Overall Power Accuracy
Looking at these four rides, there is a decent match between the two sources, but Precision 3+ (red) often seems to overestimate higher power levels slightly.
Ride: 12 Feb – A few efforts around 250 watts, then higher. Outdoor, bumpy, traffic.
This ride data looks about right in total, but the left side is lower than Assioma, and the right side is higher. The two curves mostly track each other except for one section toward the end when Assioma peaked 50w higher.
12 Feb LR – b
12 Feb LR – a
12 Feb Total – b
12 Feb Total – a
Ride: 1 Mar – Smooth, outdoor. 200w for an hour ish
This ride data looks about right in total, but the left side is lower than Assioma, and the right side is higher. The two curves mostly track each other.
1 Mar LR
1 Mar
Ride: 2 Mar – Selection from long bumpy group ride
This ride data looks about right in total, but the left side is lower than Assioma, and the right side is higher. The two curves mostly track each other.
2 Mar LR
2 Mar
Ride: 4 Mar – Smooth, outdoor. Short surges from 300w-550w
This ride data looks about right in total, but the left side is lower than Assioma, and the right side is higher. The two curves mostly track each other, with exceptions at some peaks.
Something very odd was going on here. I got dropouts from 4iiii. It may have been due to the WiFi-induced dropouts (I disabled the Wifi on the subsequent workout on 18Mar), and 4iii suggested I try a Garmin to receive the data and also to reset the 4iiii PM by inserting the battery the wrong way around. But there were also issues with the KICKR data connected via Direct Connect (LAN), so that couldn’t have been WiFi-related dropouts.
At times KICKR recorded higher watts, which shouldn’t be the case. 4iiii still was high on the left side.
Ride 18 Mar – indoor.
This was another odd test using someone’s Minute Monster workout. I got bored and recorded one part using MyWhoosh and the second using Zwift. I also dual-recorded the 4iiii data with a Wahoo Bolt 2 and an Edge 540. You can look at the source data here on dcr’s analyzer. Note I’ve coloured two different sets of 4iiii data in the same colour covering the Zwift/MyWhoosh change. Confused? I was! Also, note that you can’t compare entire ride averages for all the data, so highlight either the first or second section.
Indoor LR
Indoor Total
The data looks pretty good. Assioma is a bit higher than 4iiii, which is, in turn, higher than KICKR – you’d expect the latter due to drive train loss. However, 4iiii still records highly on the right side.
Price and availability
4iiii are widely available; even the factory install option is now available in Europe. These are list prices, but click through to the latest price.
I’m generally happy with the Precision 3+ Power meter and can use it to train and race. The overall power level is plausible and actionable on bumpy roads and indoors.
It’s staying on the bike!
My concerns are these.
Battery life appears nowhere near as good as the claims, but at over 100 hours, I don’t mind.
I still haven’t gotten to the bottom of why the left side over-reads. Oddly, the overall level appears factory-calibrated to be correct. It could well be that Assioma underreads, even tho it uses identical tech on each pedal – unlike 4iiii, which measures differently. Some other pedals are arriving soon, so I’ll try those for a new comparison.
From a market perspective, 4iiii’s products have experienced a positive bump-up in sales following the closure of Stages – the two companies had similarly positioned offerings. 4iiii gains big time for non-drive side sales but is currently (March 2025) impacted by US-Canada tariffs as 4iiii is Canadian. Much of 4iiii’s other competition is in an odd place; some new challengers like Magene and Sigeyi can’t offer a spider solution for Shimano cranksets – 4iiii can offer the factory install.
