Trying to find a smart trainer/bike in-stock? No problem! Check out the new consolidated trainer stock & alerting tool, to get alerted the second your hard-to-find smart trainer or smart bike is in-stock and ready to buy & ship. (For US & European users)
- Weights & Sizes
- General Use Overview
- Compatibility with head units
- Power Meter Accuracy Test Results
- Power Meter Durability
- Power Meter Recommendations
- Found this review useful?
- Discussion (534)
The Favero bePRO Power Meter In-Depth Review
In many ways, the bePRO power meter is the little unknown underdog that snuck onto the scene earlier this summer with a very simple declaration: We’re here, we’re new, we’re cheap, and we’ve already started shipping. How ‘bout them apples?
The unit retails for €749EUR globally, with the local price simply based on the current exchange rate. So for US folks, that’s approximately $835USD, for those in Australia it’s $1,196AUD, and for those in Uzbekistan it’s 2,185UZS. Note that this is for the dual-leg system (two pedals). Whereas they also make the bePRO S version, which is a single pedal for 499EUR. For this review, I’m reviewing the dual-system. The single leg system simply doubles the left-leg power, but otherwise works identically.
These prices are notable because this is the cheapest dual-sided system on the market today actually shipping. Sure, there are others that have promised shipping soon or planned to ship in the recent past – but nobody else is actually shipping a product in this price range today.
Favero, the manufacturer of the bePRO, sent me a unit back in July to try out, and I’ve had it on my bike now for about 45 days through a wide assortment of weather and conditions. As is always the case, once I wrap-up with the product review here shortly, I’ll box it back up and send it back to them. Just the way I roll.
With that – let’s get into the unit.
To start, we’ve got a fairly large box, sorta like a skinny shoe box, but longer.
Inside facing upwards towards you, you’ll find the two pedals. As you can see, the pods are permanently affixed to the units.
Next, we’ll remove the upper layer of the box, where you’ll find the second elevated tier. This includes tools that’ll be used to attach the pedals to your bike.
If we work our way down to the bottom of the box, we’ve got a slew of parts. So I’ve taken them all out and laid them on the table. Here’s what we’ve got inside (roughly from top to bottom):
A) Two pedals
B) One alignment tool (long skinny/flat black tool)
C) One pedal wrench (black, looks like a wrench)
D) Long hex wrench
E) Two cleats and shoe installation screws
F) Two pedal alignment tools (silver, round and long)
G) USB charger with two micro-USB cables, crapton of adapters for different countries
H) Manual and other paper stuff
I) Table not included
J) Small sticker/card with ANT+ ID numbers, serial numbers (sitting atop left manual page)
K) Four alignment stickers (on photo just below cleats)
Phew – lots of stuff!
Grouping things together for a better look, here’s the installation pieces in one pile:
Then we’ve got the power adapter (universal), plugs, and extra long micro-USB cords:
Then the cleats:
Finally, the pedals themselves. Each pedal is identical. Note that there are two different bePRO product options though. One that has a single pedal, and one that has two pedals (sorta like Garmin Vector & Vector S). I’m only testing the dual-pedal system.
While the pedals can screw into virtually any crank arm out there, it does take a special tool to do so correctly.
With everything unboxed, let’s briefly look at how it compares.
Weights & Sizes:
Perhaps the most common question is whether or not the bePRO pedals change your ‘q-factor’. No (in case you were wondering). The above is an example of the bePRO sitting atop a standard Look KEO pedal; they’re identical (the photo skews things slightly).
Never mind the fact that your q-factor actually changes if you go from road to mountain bike, or that somehow some imply that as humans we’re all destined to have the same q-factor despite varying sizes (or the fact that we tweak every other aspect of the bike down to the last millimeter, but almost never q-factor). Yes, I’m highly cynical of q-factor concerns since there are actually no real published scientific studies on the topic (that cover longer timeframes than just a single point in time snapshot).
But of course, none of that matters (and my tangent can be saved for another day) – since the q-factor here is identical.
As for weight, here’s a few common competitors. In a nutshell, the 156g of the bePRO pedal is the lightest of the ones here. I didn’t weigh the Polar/Look KEO pedal, but officially it comes in at 199g per pedal. Obviously, I’m only showing one pedal – simply use those advanced calculus skills to double the number seen for total weight.
Note that all of the images above include all mounting hardware required for a base installation; it’s possible some bikes may require an extra washer or two per crank arm (less than a gram). If that concerns you, don’t eat that next potato chip. It’ll work out.
Now, I’ve gotta be a bit more blunt than usual here. When I first opened up the box and looked at all the installation parts, I basically muttered a string of expletives. I wanted something that’d be quick and simple to get going – and in looking at the box or parts, it looked like a complete cluster-duck.
But here’s the thing I found out – in reality, it’s actually not that bad. The first time you’ll probably be a bit confused, but once you do it once it makes total sense. And now I’ve moved it between numerous bikes without issue and can do it easily in less than 5 minutes. Here, let’s start.
In my case I was using an updated PDF copy of the manual, hence the laptop:
First, you’re going to grab one of your long silver tubes. These are used as braces for applying the alignment stickers to the cranks. You can see I’ve screwed it into the crank arm. However, I forgot to put in the thin (flexible) black tool you’ll see in the next photo.
So, in my case I un-screwed it and put in that piece first, and then simply twisted in the other side. Captain Obvious notes that if you do this correctly, you shouldn’t be able to turn your crank arm very far.
Next you’ll grab your four-pack of stickers. These are simply used to allow you to align the pedals later on. After you finish installation, these stickers serve no purpose on your bike.
Now, here’s something I’ve learned in re-installing the stickers multiple times: You’ve only got four of them. Thus if you have two bikes and want to install it somewhere else (a weekend getaway onto a rental bike), you’re hosed. Instead, you can simply not push the sticker on very hard, and once installation is done, just peel it gently off and put it back on the sticky paper. Forever stickers!
See, you just stick it in the little rectangular window:
Now, do the whole thing on the other side again:
With your stickers set, remove all of the tools you just had and throw them back in the box. You won’t need them again. Next, grab your pedals, the lock nut (usually on pedals already, but just in case), and a spare washer.
Then, somewhat gently screw it into your crank arm. The manual implies doing this by hand, but realistically that’s a pain in the ass. Instead just use the hex wrench to do it quickly, but don’t tighten it.
At this point in the manual, Favero would like to remind you that if you screw this up – it’s totally your fault:
Next, you’ll take your pedal and loosen it (yes, loosen) until the line on the pod lines up with the line on the sticker. Roughly like this (minus the wrench, didn’t have the right photo on the plane):
Finally, you’re going to do a two-handed maneuver. This involves both the pedal wrench and the hex wrench. You’ll simply pull them in the directions that the arrows on the pedal wrench tell you to.
There is no need for a torque wrench here, just screw it down until it feels tight. Pedals will automatically self-tighten and settle out over the next few rides.
Repeat for the other side:
You’ll simply go ahead and pair your head unit to your bePRO using either the search function (for a power meter), or by entering in the ANT+ ID manually if you want:
With everything physically installed, we’ll move onto using the system.
General Use Overview:
By and large the bePRO system is pretty easy to use and pretty straightforward. They recommend that prior to each ride you do a simple ‘calibration’ procedure (aka Zero Offset), from your head unit. This is simply just waking up the cranks and then pressing the calibrate button on your head unit.
The process only takes a few seconds to complete, in which case you’ll get a calibration value (0 hopefully) returned.
In addition to this they also offer a ‘Dynamic calibration’ option. This is sorta similar to what Garmin does with Garmin Vector, but just with a little less finesse than Garmin does (in large part because Garmin owns the head unit and can thus make this process neat and tidy for their own power meters). This process actually just changed slightly in the latest firmware update. Previously you used the calibrate menu to trigger it, but now you backwards pedal 10 times. Beyond that the steps are the same.
With the dynamic calibration option you’re aiming to correct for any tiny little shifts to how well the pods fit against the crank arm. As some of you know, certain power meters require a bit of settling in, and bePRO is one of those. That means that you want to get in a few rides, with some varied efforts (i.e. short hard sprints) to try and get things all settled.
What the dynamic calibration does is to recalibrate the pedals based on that. To do a dynamic calibration after backwards pedaling 10 times you’ll start forwards pedaling at an 80RPM cadence.
You’ll need to continue pedaling at 80RPM (+/- 5RPM) for another 30 seconds, where the bePRO units will continue transmitting your cadence values. Previously you received a calibration value, however now you do not. That value was actually kinda useful in figuring out if things were still shifting a bit.
What I’ve found in changing the pedals to/from different bikes four times now, is that it takes on average 2-4 rides until things settle with the bePRO power meter. That’s roughly in the ball park of some other power meters, such as Vector and Quarq. As noted above, I’d highly recommend doing some short sprints (i.e. 15-20 seconds) at whatever max power level you can spit out. Ideally if you’ve just installed the pedals inside, just do these on a trainer.
(Tip: As a general rule anytime I make pedal/bottom bracket/crank changes on a bike, I do a few short sprints on a trainer to validate everything is good and sound. My reasoning being if I’ve totally screwed something up, at worst that part would break and I just semi-harmlessly fall to the ground inside going 0MPH, versus in a sprint going 30MPH on asphalt with traffic/etc.)
Note that you can do this process both inside and outside. I’ve done it both places. Do note however that if you do it outside, during the time period where you’re doing the dynamic calibration, it will not transmit ANY ANT+ power stream to your head unit, so you’ll get a little time gap like this:
No big deal of course (and kinda logical), but just to be aware of later when analyzing your files (it’s also been noted in the latest edition of the manual).
Next, we’ve got battery usage and consumption. The unit has rechargeable batteries within each of the small pods. Those pods are charged via micro-USB port, just like most cell phones (non-iOS). The bike comes with a dual USB port charging cable setup, which works well:
It has nice long cables for each one, making it super easy to get to the right spots on your bike.
Now in theory the system is rated for 30 hours per pedal. And at first I was on-target for that, no problem. In fact, I didn’t charge it once from the end of July till almost the end of August. Great! But then I found a snag: Movement.
See, during that one month period I didn’t travel with my bike, rather, it left the DCR Cave/Studio and went for a ride and returned to the same spot and then waited for me to ride it again. But then I went on a roadtrip, and with that there was the train ride down, lots of driving, etc… Before I knew it, I was re-charging my batteries again.
And then a mere 4-5 days later, I was doing it again. It appears that there might be some bug in the unit not properly falling asleep when moving. Though, even then it doesn’t really account for the massive gap in battery performance in doing it again just a few days later (i.e. that’d be way more travel than I did). Since I’ve stopped travelling with the bike, I haven’t seen any battery issues since. Perhaps it was just a one-off, I’ve noted it to Favero and they’re digging into it a bit, though it may be tougher to replicate.
Compatibility with head units:
The bePRO power meter is compatible with any ANT+ head unit that supports power meters. This would be most of the Garmin Edge units above $200, as well as the PowerTap (Joule), Sigma and O-Synce head units. Along with many other devices and apps out there.
It should be noted that it does not support Bluetooth Smart at this time, which is a bit odd in that many sensors and power meters are now shifting towards dual (concurrent) ANT+/Bluetooth Smart. So the bePRO would not work with the Suunto Ambit 3 or the Polar M450/V800/V650 units.
Still, for those devices that it does support, the pairing process is pretty simple. I’m using a Garmin Edge 810 here for simplicity’s sake, but you can use any device you’d like. First up, you’ll search for the unit:
Then, once you find it you’ll want to set the crank length on that bike/sensor. To do so you’ll go into either sensor or bike settings (depends on the bike computer).
With that set up, you’ll want to do at least one calibration here per the earlier section. After that, the unit is ready to roll.
While riding you’ll get metrics such as power, cadence, torque efficiency, and pedal smoothness. Because this is a crank region based power meter, you will not get speed (that will instead come from GPS or a speed sensor). Within the settings, the unit will also transmit current status:
Note that the bePRO units do NOT support Garmin Cycling Dynamics. Garmin makes that fully private and is only available to Garmin Vector & Vector 2 devices, not 3rd party devices.
Below is a compatibility table. There are essentially a few key items that are required in order for the bePRO unit to be fully supported. They are:
A) Ability to set crank length
B) Ability to display left/right power
C) Ability to display Torque Effectiveness and Pedal Smoothness (optional)
This is actually the exact same listing of requirements for the recent PowerTap P1’s. And in general I find that the P1 and bePRO units act exactly the same with respect to behaviors on different head units. Below, the compatibility table for the bePRO:
|Display Unit||Compatibility Status|
|Garmin Edge 500||Fully functional, can set crank length via bike profile|
|Garmin Edge 510||Fully functional, can set crank length via bike profile|
|Garmin Edge 520||Fully functional||can set crank length via bike profile in FW 3.00 and higher"|
|Garmin Edge 800||Fully functional, can set crank length via bike profile|
|Garmin Edge 810||Fully functional, can set crank length via bike profile|
|Garmin Edge 1000||Fully functional, can set crank length via bike profile in FW 5.10 and higher|
|Garmin FR910XT||Fully functional, can set crank length via bike profile|
|Garmin FR920XT||Fully functional, can set crank length via bike profile (Beta Firmware 5.28 now available)|
|Garmin Fenix3||Fully functional, can set crank length via sensor settings|
|Garmin Epix||Fully functional, can set crank length via sensor settings|
|Garmin VIRB XE||Fully functional, can set crank length via sensor settings|
|Polar V800||Not supported, BLE-only device - bePRO is ANT+ only|
|Polar M450||Not supported, BLE-only device - bePRO is ANT+ only|
|Polar V650||Not supported, BLE-only device - bePRO is ANT+ only|
|PowerTap Joule 2||Not compatible.|
|PowerTap Joule 3||Not compatible|
|PowerTap Joule GPS||Fully functional, can set crank length via sensor settings|
|PowerTap Joule GPS+||Fully functional, can set crank length via sensor settings|
|Suunto Ambit3||Not supported, BLE-only device - bePRO is ANT+ only|
Again, I’ll update this from time to time, but it’s also worthwhile searching the comments section, as oftentimes readers will update compatibility items from time to time there.
Power Meter Accuracy Test Results:
The bePRO pedals actually found their way back and forth on two different bikes. First was my triathlon bike – a Cervelo P3C, and the second was my primary road bike – a Giant Defy. Both of these bikes are setup fairly similar in terms of power meter counts. Each has a rear PowerTap G3 wheel on it. And each has a different crank based power meter in test on it (the PowerTap C1 on the Cervelo, and the Verve InfoCrank on the Giant). For the purposes of these tests, I’m mostly going to use the road bike data, since the PowerTap C1 unit was an earlier prototype and I try not to mix prototype units within production reviews. Don’t worry, a PowerTap C1 production hardware/software unit should hit posts soon.
From a data collection standpoint, virtually all of the data used in the analysis was collected using product Garmin Edge 520/810/1000 units. In all cases the bePRO was using the Edge 810’s, since the Edge 520/1000’s don’t yet support crank length for non-Vector power meters (firmware update coming shortly). Meanwhile the PowerTap G3 and PowerTap C1, as well as the Verve InfoCrank were captured using the Edge 520/1000 (since crank length was irrelevant there).
With that, let’s dive into things. I’m going to pick a few random rides that talk to specific conditions (i.e. temperature changes), sprints, stable power, cobbles, etc… The goal being to ferret out any oddities that I might see. You can download all the ride data at the end of this section.
I’m going to first start with my 2nd ride, to show a little bit of the settling period. Favero recommends that you re-do a dynamic calibration after a high intensity ride. This is because of slight shifts in the attachment of the pedal to the crank. These aren’t usually visible to the human eye, but they are there (just like Vector and others). What this means is that you may see a slight offset, such as below:
You can see that all three power meters agree very well, it’s just that the bePRO is running a few percent higher. Now some could argue that’d be normal – and, a portion of that would be true. After all, the bePRO is closest to your foot and thus would have the least amount of ‘loss’. Below, I’ve zoomed in on a section (which changes the scale), note that we’re only talking about 13w on 260w here.
If we look at the Mean/Max data, which helps to visualize max power outputs over a time graph, you’ll see they align spot-on perfectly. Again, a tiny pre-settling offset – but really, very impressive.
Next, we’ve got a ride I did with the Verve InfoCrank in the mix. So now we’re on a different bike, and I left things settle a couple rides on that bike. Nonetheless, you can see things are basically exactly on top of each other – there’s virtually no appreciable disagreement there.
If we zoom in, we see the same – especially true on some of the sprints that I did. For example, here’s zoomed in on a sprint. This is with a 10s smoothing factor to make it a bit easier to see. You’ll notice that all three units do almost exactly the same thing. You see itty-bitty variations, which is normal because of the 1-second recording rates and transmission differences.
Here, I’ll pick another sprint. In this case, I’ve reduced again the smoothing rate down to just 3-seconds, and all three still look stunningly close. I specifically choose one point at the top where they were wider apart (standard delays), yet the next second they’re all within 18w (at 650w – or 2.7%). That’s totally normal variance for three power meters measuring in dramatically different places on the bike.
And the cadence data here? Just as good. You’ll always see slight difference in cadence data on stop and go activities like this, because of the way power meters differ in updating cadence when you go from 0RPM to 90RPM. But you can see that once I’m pedaling all three of them completely agree. The only exception being that sometimes coming down from sprints you’ll see the PowerTap G3 be a bit wobbly, but that’s because it has more of an estimated cadence than truly measured.
And again, here’s the Mean/Max graph for that ride too:
Lastly, we’re going to look at a mountain ride I did. I want to include this because the temperature drops as I climb. For this ride I ascend about 3,000ft over the course of a fairly short time period (35mins):
And thus the temperature drops about 20 degrees:
Yet still, everything aligns right on top of each other:
Let’s zoom in on the climb as well (I never stopped during it):
The key item above is that you don’t see any drifting on the bePRO. You’ll see that in general, the bePRO records the highest power track, but that’s mostly consistent throughout the climb. You see a few bumps of more variance in the last 2-3 minutes of the climb, but I’d be guessing that’s probably to do with me being on steep gravelly and potholed roads and I believe I was only clipped in one pedal just in case I had to bail (so my contact with the power sensor in the pedal wasn’t clean). Thus I’m not really concerned there as I haven’t seen any issues with cobbles or the like in and around Paris.
And finally, if we look at cadence on that climb, it looks pretty darn good. Though there are two brief seconds for a few seconds where there’s some variance. It’s not clear to me if I did something different in those moments that might have triggered that (i.e. being unclipped briefly for one-legged or what). So it might be a bug that’s since been fixed in the latest firmware.
Overall, I’m not seeing anything on the accuracy front that’s of any concern. It’s consistently been spot-on with other power meters that I’ve thrown at it.
Now one interesting item I saw on my very last ride before publishing was a condition where one-legged pedaling on the right side caused the power for both units to drop. However, Favero confirms they also saw this bug and have already fixed it in a firmware update that was just released.
Here’s a Dropbox link with all raw ride data.
(Note: All of the charts in these accuracy sections were created using the DCR Analyzer tool. It allows you to compare power meters/trainers, heart rate, cadence, speed/pace, GPS tracks and plenty more. You can use it as well, more details here.)
Power Meter Durability:
I’ve noted numerous times in responses to comments since receiving the units that my primary concern has been with the durability of the unit, in terms of how well it’d stand up to wear and tear. Not so much the pedals themselves, but rather the pods that sit in between the pedals and the crank arm.
To begin, we’ll start with the micro-USB port covers. These are kinda crappy.
They easily fall off when poked, and Favero seems to even implicitly acknowledge that. After all, they included a spare set of the doors/caps within the box. I’ve never had them pop off out on the road, but did have one pop off/out while on a trainer when I clipped in just the wrong way. They also readily fall out when trying to open/close them for charging.
When I asked Favero about this, they noted that the battery covers aren’t needed to protect the port itself. Rather, the port is internally waterproofed to IPX7 standards (so 30 minutes at 1-meter deep). Thus the port cover mostly serves to keep dirt/sand/etc out of there to keep a good connection. Still, it’s not ideal.
My next concern was around the plastic siding on the pods. It doesn’t seem crisp and I found it easily got damaged over the course of my time using them due to clipping in/out. My concern here being that this could lead to water leakage inside the unit should cracks develop (a seemingly likely scenario after months of usage based on the trend I see).
I also went back to Favero on this as well. They noted that in actuality the plastic is pretty much just cosmetic. The pods themselves are fully sealed in a hardened resin. They provided a photograph (below) of what one looked like prior to the pod casing being added (below shows it sanded down a bit as part of a test they were doing). They noted that this fully waterproofs the entire pod itself, and also makes it completely enclosed. In fact, once the resin is added the unit’s electronics are actually unreachable. Then entire pod is then laminated as well.
So I’d say that mostly solves my concern about water getting into the pod (though it doesn’t solve the aspect around how pretty it might look after many months of usage).
Lastly, for lack of another spot to stick it – a few readers had purchased bePRO units earlier this summer and expressed concerns around the lack of support in August (Favero essentially went on vacation for two weeks, leaving customers with no support mechanism). I enquired about this as well, and the company straight-up admitted it was a screw-up and that they’ve learned from it. Here’s what they had to say:
“It’s true, we had not sufficient time to organize a proper support during these 2 weeks of August. Actually, [employee name] respond to many emails and phone call even [on] vacation. When we went back to work, we saw so many requests (bePRO reservations for next production batches, information requests, some support on installation, …), for sure we will keep support active even during holiday.”
So it sounds like they’re putting in place the required staffing to ensure continuous support year-round.
Power Meter Recommendations:
As a general rule of thumb, when purchasing a power meter you should focus on your specific requirements. Meaning that if you ask a bike shop “Which power meter is best?” and they give you XYZ brand, then you must fire them as your power meter consultant. Rather, they should be asking you about things such as do you swap bikes? Do you swap wheels? What type of pedals/cleats do you use? What type of chainrings do you use? And so on. That allows them to identify a specific power meter type that might be best, and then from there a specific product.
Now, within the context of pedal based power meters, the bePRO ranks fairly highly on my list. The pedal based power meter world as of today consists of the following players:
– Garmin Vector/Vector 2
– PowerTap P1
– Polar/Look KEO Power
– Favero bePRO
There are no doubt products announced by other companies for products in the pedal or pedal connectivity region (i.e. Brim Brothers, Watteam, Limits, Look-specific offering), but only the above four companies have actually shipped something to consumers today as of the date of this publishing. That’s an incredibly important distinction in a segment of the industry where delays are often years in the making.
So given those four players, I think it’s fair to say the best value for the money in a pedal based power meter at present is the bePRO. Whereas my favorite in terms of portability, durability and general ease of use is the PowerTap P1. And then the most metrics comes from the Vector/Vector2. But I still think at this juncture the Vector/Vector2 and the Polar offerings are overpriced. If perhaps Garmin or Polar did some form of rebate or price drop down to $999, then things might change.
Note I don’t believe it’s worthwhile to look at single-pedal offerings (i.e. Polar Power Essential or Garmin Vector S), given the prices those companies are charging for single-pedal. There are far better deals from Power2Max, Quarq, and PowerTap for crank/chainring based systems.
Finally, as usual I’ll be doing my Annual Power Meter Buyers Guide the week after Interbike (so that’s two weeks from now). This will encompass a bit of a ‘state of the industry’, futures, and thoughts on power meter purchasing.
Overall I think the bePRO power meter is a great value today. It might not be the prettiest power meter out there, but it’s the best priced pedal based power meter on the market. It’s reasonably easy to install after you do it once and understand what’s going on. Further, from an accuracy standpoint, metrics look very strong. I haven’t had any issues in shifting temperatures (mountain climbs), nor rain (or shine). The pedal body itself hasn’t posed any problems and performed well, and I’ve been able to use it successfully without any incidents.
While I do have concerns about long term exterior durability, bePRO seems to have architected the internals of the product to stand up, even if the outside shows more wear than I’d like. Further, their 2-year warranty (required in the EU) should help ease any concerns there. However, the unit today only supports ANT+, which could be a deterrent for some that are using Bluetooth Smart head units. But I suspect that won’t be a blocker for most.
Of course, the power meter market is moving very quickly with new entrants announcing products frequently. While Interbike is next week, I don’t expect to see any earth shattering news there at this point – most of the major power meter players have made their announcements. But of course, ya never know.
With that – thanks for reading!
Found this review useful? Or just wanna save 10%? Read on!
I’ve partnered with Clever Training to offer all DC Rainmaker readers an exclusive 10% discount across the board on all products (except clearance items). You can pickup the bePRO pedals from Clever Training. Then receive 10% off of everything in your cart by adding code DCR10BTF at checkout. By doing so, you not only support the site (and all the work I do here) – but you also get a sweet discount. And, since this item is more than $75, you get free US shipping as well.
Thanks for reading! And as always, feel free to post comments or questions in the comments section below, I’ll be happy to try and answer them as quickly as possible. And lastly, if you felt this review was useful – I always appreciate feedback in the comments below. Thanks!