🎙 [Review] FunSR PRO Sound Test & Review

[Falafel]

FUNSR PRO: The Most Powerful Stroke Device, Featuring Joint Motors, Is Launched

FUNSR PRO This is currently the most powerful stroke device on the market, and will remain so for years to come. FUNSR PRO – FUNOSR [微信图片_20251017165903] OSR2 and SR6 use servos as their power units; FUNSR1, FUNSR1 2.0, and the HANDY2 series use brushless motors as their power units; This time, however, the power unit used in FUNSR PRO is the next-generation motor—the joint motor currently used in the joints of humanoid robots! [微信图片_20251017153711] Reasons to Choose FUNSR PRO 1. Adeq…

What is this?

The FunSR Pro is a custom-built OSR2+ made and sold by FunOSR. It uses joint motors designed for serious robotics instead of hobby servos. It is claimed to be extremely silent all whilst delivering superseding performance.

I have been using this thing for over a month now and finally got some time to do a proper review. In this post you’ll see videos of it running in action as well as noise measurements. Hope this will be informative and addresses some concerns.

Disclaimer: I got my unit for free but I’m not paid to write this. I do have a discount code - you may use the code [FALAFEL] at checkout if this post helps your research.

---

Overview

Weight: 3.25kg (7.2 lb) with mount
Dimensions: 410 x 180 x 150 mm (16 x 7 x 6 in) with mount

The device follows the classic formula of the OSR2+, though heavily redesigned to fit the GIM6010-8 motors onto it. The casing is in PETG though the motor housing is metal. The whole unit is almost twice heavy compared to the FunSR6 thanks to all the metal parts combined.

---

The Motor

The GIM6010-8 is a motor engineered for precision robotic arms and robot dogs. Under 24V it has a rated torque of 5.0 N·m and a stall torque of 11.0 N·m. How much is that you ask? A 30kg·cm servo has a stall torque of roughly 3 N·m. The GIM6010-8 is 3.5× stronger than a 30 kg servo at stall. Safe to say it can carry more than a standard Onahole.

Raw power aside, they are very silent, making only a humming sound at the fastest section of a script. They also produce a very smooth movement that feels controlled and isn’t sharp around the edges.

---

See It in Action

Below is the FunSR Pro moving to the Conqueror by かやま script. Feel free to compare it with the SR6 video. Turn up the volume - the video uses actual recording of the device’s sound 30cm away.

Click if video doesn’t load

The manufacturer is confident that it can lift up to 3KG without problem. I unfortunately don’t have a 3KG hip for testing, so here goes a ride with my 1KG Puni Virgin 1000 Fuwatoro, running the “stress test” script I did earlier:
Stroker_Stress_Test.funscript.

Alright… I guess load really isn’t an issue here.

---

Overclocking It

Its potential doesn’t end here - tests above are done under the default settings, whereas the motor can be set to go faster and further. The maximum speed and stroke range can be adjusted following this tutorial:

How to adjust the speed/Maximum stroke of the FUNSR PRO - Google Docs.

• Kp value affects speed. It can be set from 0.5 to 5.0 (default = 1.2). Higher value makes the device “snappier” but also louder.

• Angle Gain affects stroke range. It can be set from 2.0 to 1.0 (default = 2.0), corresponding to a stroke range of 12.43cm (default) to 23.21cm (maximum).

This way you can adjust the speed and stroke range according to your own need, so there’s no potential wasted.

Start-up Behavior

Unlike the SR6 / OSR2+ which “snaps” into its default value (50%) upon boot, the FunSR PRO sets the current position of the receiver as the default position when it is turned on. This default position can be configured by sending the following command to the device:

• DSP1 - Start at bottom (0% L0)
• DSP2 - Start at middle (50% L0)
• DSP3 - Start at top (100% L0)

By setting the device to start at the bottom (0% L0), and its range adjusted to match your penis, you can switch on the device after a full insertion and be confident that it should not move beyond the tip of your penis, causing a slip-out.

---

Noise Test

So here’s some data on the noise. The recording was taken the exact same way I did with the FunSR6 in this post: 🔬 [Review] “Silent” SR6 From AliExpress: How Quiet Can It Be?. It is taken 30cm away from the device which is pressed against the floor running the following script:
speedtest.funscript (7.5 KB)

Device	Avg dB	Max dB
FunSR PRO Idle	20.6	24.8
FunSR PRO Stroke (400g load)	33.3	42.5
FunSR PRO Stroke (400g load, Overclocked)	42.2	53.2
FunSR PRO Twist	53.3	63.4
The Handy	50.8	60.7
FunSR6 Stroke	57.9	67.1
Coffee Grinder	78
Hairdryer	84

Summary

• The pitched “hum” you hear is the GIM6010-8 motor working. It is much more quiet and subtle compared to hobby servos of the SR6 and step motor of the Handy.

• By increasing the Kp value, the motors also becomes louder. You can also hear collisions of the receiver parts as it swings with greater momentum. I’ve had a much harder time pressing the unit against the floor.

• The Twist module of the SR Pro is still the loudest bit, with very pronounced “click” as it reverses direction. It is slightly louder than the Handy at full speed.

notes on recording

• I used the app DecibelX on a Samsung S20 FE. It’s not professionally calibrated, but should be enough for comparison.
• Some of the data comes from last year. Recording devices and location are the same.
• To make the sound more audible in the video, they have been turned up +12dB in editing software.
• Full length recording Here (3:53)

---

Hardware

Desk Mount

The desk mount is made with 2 metal C-clamps bolted onto aluminum extrusions, which are then bolted onto a connector plate. The device slides onto the connector plate and is secured with 2 bolts at the bottom and an anti-collision wedge on top. It’s an assemblage that you don’t want to undo often.

I added some rubber pads underneath the top of the clamp to protect my desktop, also adjusted the mounting height by sliding the clamps along the extrusion. The process is a bit cumbersome, but solid once all bolts are fastened. 4 clamp screws ensures a very stable grip onto the desk.

If you don’t want to use the clamps, it’s also possible to DIY your own mount - it’s a standard 100*100mm VESA at the back.

Receivers

You get to choose between an open receiver (that holds larger toys), and a Twist receiver (that spins). The Twist receiver is directly compatible with Fleshlights, but can also hold Onaholes with the included adapter.

Just like the FunSR1 2.0, this Twist module uses a step motor, which makes it rotates up to 360° compared to the typical 270° servos.

If you buy the Twist option you’ll also get the open receiver.

The Fan

A 40mm fan is attached to the front of the unit to help with ventilation. The fan will spin as long as the device is turned on. The fan noise is almost negligible.

According to the dev, the fan is only ever necessary if the device is operating under heavy load for extended period, such as swinging a 1kg+ hip for several hours. I initially had issue with this design choice since my fan was making some very audible noises (32dB), which I thought had broken the device’s main feature as being silent. I later solved it by replacing the dud fan with a new one that makes almost no sound, which I’m happy with it constantly blowing.

I notified FunOSR about this and they promised improvements on the fan. If you are buying this after me, you should get a silent fan mounted at the bottom of the device instead of front.

---

“Unboxing”

Opening it up was surprisingly easy. Simply undo the 4 screws holding the front cover and disconnect the left motor cable, then you can remove the cover.

And here are the internals:

I did this whilst replacing my fan and it was much easier compared to doing maintenance on an SR6.

---

Concluding Thoughts

It’s a jacked up OSR2+ that doesn’t scream like the SR6.

I was originally unimpressed by the form factor of the unit, but the fact that it makes almost no sharp noise during operation has changed my mind. This is now my go-to stroker for ASMR and VR porn for the best immersion. With some noise-cancelling earbuds I hear nothing coming from the device, and what a bliss to be able to ignore it.

What I liked:

• Silent (without quotation mark).

• Smooth. I found the movement to be a lot smoother than a servo-driven toy, but it can still be made “snappy” through configuration.

• Powerful. I’m not an avid user of heavy hips, but can definitely see the potential. Temperature control is also well-considered in the use of metal housing and a fan.

• Customizable. Can set stroke speed, length and startup position according to one’s need.

• Flexible default position. I switch between different onaholes and they all mounts a bit differently. This device provides a solution to that problem.

• User-repairable. Just like the OSRs, it’s build with standardized parts which I could source and replace if they ever breaks after the warranty.

What can be improved:

• Heavy and bulky - It’s not very portable. I don’t want to keep carrying it between places. This is a device that’s best stored in one place and used in one place.

• Desk mount is difficult to adjust. It’s an unpleasant process that involves a lot of screwing and unscrewing. Thankfully you shouldn’t need to do this very often.

Overall, this is a bold initiative to fit industrial-grade motor onto our beloved sex robots, and the outcome isn’t bad at all. It’s a delight to witness such progress and I look forward to the dev’s next steps

---

Links & Coupon

If you enjoyed this review and would like to order one, feel free to use my promo code for a $10 discount at purchase!

• -$10 with FALAFEL

Product Link:

FUNSR PRO

The product is currently only available through FunOSR’s site, but will also be on AliExpress soon.

---

[roa]

Yeah needs to get fanless and way smaller for me to jump and no it does not need to be able to lift my grandma.

[Falafel]

@roa

it does not need to be able to lift my grandma.

No it probably cannot… unless your grandma is less than 3KG

[randomus]

Thanks for this review. It’s good to see some transparent data concerning the noise and it was generally illuminating. I really like the simplicity of this mechanical design and think that - aside from the size/weight - the small weaknesses it has can be corrected quite easily. So looking forward to their progression and/or measures people take on their own using their printer.

For scripting purposes: Would it be possible to confirm something about the geometry? To be precise: The length of the arm (center joint - center joint) and the distance on the receiver between the main arm joint to pitch arm joint, as well as the max roll/pitch angles? From what I was able to gather this far, it’s a bit different to the OSR2+ and some of it might be prone to change together with the parameters.

@Falafel

Kp value affects speed. It can be set from 0.5 to 5.0 (default = 1.2). Higher value makes the device “snappier” but also louder.

The way this reads and how the video sounds, it seems to me, we are either directly changing the position/velocity controller proportional gain or some filter. Which is cool that they would allow for that. Is it also more jittery at rest with high Kp values? Do you see changes in timing?

@roa

Yeah needs to get fanless and way smaller for me to jump and no it does not need to be able to lift my grandma.

If I had to guess, it’s because the smaller the motor gets in that region, the more expensive it gets. I mean, if I interpret the pictures correctly, even the pitch motor is one of these big ones. Probably not to be able to really snap it but to save on the cost and use the same motor controller and accessories (and software and configs) as the other ones.

Concerning the fan… From their other thread, most would probably be fine without one. But I still think it’s a good idea to have at least a small bit of forced convection for the power electronics. They could probably change the fan and its speed to both save on space and make it more silent pretty easily. So I’ll be observing their updates ^^

[GarbageMan]

@Falafel

Unlike the SR6 / OSR2+ which “snaps” into its default value (50%) upon boot, the FunSR PRO sets the current position of the receiver as the default position when it is turned on. This default position can be configured by sending the following command to the device:

• DSP1 - Start at bottom (0% L0)
• DSP2 - Start at middle (50% L0)
• DSP3 - Start at top (100% L0)

Could you give me a small guide on how to set it to dsp1. I tried to do it like how the guide says for speed, but it stays at default when i connect it to MFP.

[Falafel]

@GarbageMan

it stays at default when i connect it to MFP

You’ll need the latest firmware to be able to use this feature. Just checked their discord and it should be released this week.

If you meant it snaps up by 50% after connecting to MFP - This is normal, as it is now controlled by MFP. You can change this by setting a default value in MFP’s “device settings”.

[GarbageMan]

Oh i didn’t know they had firmware updates, I got to wait till that’s out then!

[Kenny]

That was an awesome review. Needless to say, my decision to purchase one is now made

@renwoxing let us know when its shipping again. Thank you

[Falafel]

@randomus

For scripting purposes: Would it be possible to confirm something about the geometry? To be precise: The length of the arm (center joint - center joint) and the distance on the receiver between the main arm joint to pitch arm joint, as well as the max roll/pitch angles?

Now you mentioned I will look into it. Here’s some rough measurements

• Arm length: 175.0 mm
• Distance between the bearings on the receiver: 39.5 mm
• Distance between the bearings on the motor: 26.0 mm

Angles (total range)

DAG2.0 (default)

• Roll: 40°
• Pitch: 23.8°

DAG1.0 (max)

• Roll: 63.0°
• Pitch: 46.8°

We’ve took measurements on the SR6 and OSR2+ angles earlier here: OFS Simulator3D Mod: Surge & Sway Fix - v1.6 - #22 by Falafel. They were all a bit inconsistent between devices (each with different code) and the SR6’s range also differs to that of the OSR2+.

In a nutshell, the FunSR Pro’s pitch on default AG value is a bit limited compared to the SR6 and OSR2+ (with Khrull’s firmware at least) but can be increased by dialing up AG, then limiting L0 inside MFP. There’s also some leeway around the pitch link arm so I bet it’s possible to stretch out the pitch range even further via firmware updates. It may not be necessary though, as extreme angles on the OSR2+ mostly leads to collision with the lid, and your penis can only bend that far.

I do hope this could be made a bit more consistent with other devices in the lineup, or perhaps allowing the user to set the pitch and roll gain independently without affecting the L0. So consider that a feedback for future OTAs @renwoxing.

Will this mess up the standardization of scripts? Well my take is that it’s already quite mess up. Yoooi please save us.

[randomus]

Thank you for those quick measurements. Are those angles total range or deviations from the center position?

In either case that’s exactly what I had feared. The scripting gets even less standardized, especially when you don’t want to rely on Smart Limits in MFP.

@Falafel

Will this mess up the standardization of scripts? No, because it’s already quite mess up. Yoooi please save us.

Yoooi and I had a little discussion about this issue a while back. (From this comment on and after). The more devices with different geometries come out, I have to agree with them, that it should be the firmware doing all the compensations and safety features. Which means that the axis setup should also happen there and not in MFP. Or of course, we keep the devices dumb and do everything in the player. Fun discussions to be had in the future.

[Zel]

Thanks for the review Falafel. I’m glad to see more competition coming in as that’s always good for us consumers. The movement here looks super smooth compared to the sr6 even with decent servos, and quieter is always better even if it isn’t as quiet as I’d initially hoped (The twist still sounds like someone running on a treadmill, though that’s nothing a blanket tent can’t help with).

It’s also a shame that we’re missing an axis (or two? I’m not entirely sure) compared to the sr6.

A very tempting purchase but I’ll have to hold off for now and see if a DIY version comes along at a lower or similar price point or that we can build up the knowledge ourselves with and potentially modify rather than having to rely on an international seller to help us with it. I didn’t have a great experience with the funsr1 v2 support but I believe that was largely due to the language barrier at the time.

One question I have is if there’s a noticeable difference on the vibration/shuddering of the desk that it’s mounted to compared to the sr6? My tempered glass desk doesn’t handle it well as is so I might be on the lookout for a new desk at this rate too

[Falafel]

@Zel

It’s also a shame that we’re missing an axis (or two? I’m not entirely sure) compared to the sr6.

Yes, this is based off a OSR2+, not an SR6. I’m not sure how an SR6 would look like with 6 of these chonky bois fitted.

@Zel

if there’s a noticeable difference on the vibration/shuddering of the desk that it’s mounted to compared to the sr6?

I felt it’s similar to the SR6 experience back then (with default settings) , I unfortunately can’t compare right now because I killed my SR6 in an accident.

What I can say is that it causes noticeable vibration if you dial up the Kp value to make it go faster. It also visibly shakes my desktop under high speed + heavy load as seen in the 1KG stress test video. My desk is a cheap IKEA desk, not very sturdy.

I wouldn’t use this clamped on a glass desk, or any of these devices…

[Zel]

Yeah I ended up using an unused cat tree to mount my sr6 instead of my desk, but I think the weight of this would be enough that it’d tip over without a counterweight added to the other side

[Falafel]

@randomus

Is it also more jittery at rest with high Kp values? Do you see changes in timing?

Just tested the device with KP5.0. There’s no visible jitters. Timing doesn’t go wrong. The device just got more responsive and snappy. It is now able to reproduce those tiny vibrations people add to their scripts, and perhaps a bit too faithfully. And while it is doing that, there’s an earthquake on my desktop.

But one can also set it to KP0.5. It’ll be much slower and won’t be able to attend some positions in time (fast strokes become shortened, timing is still correct). So when I said “actions feel smoother around the edges” it’s probably just the motors being less responsive with lower kp. Under KP5.0 everything feels super linear and sharp (which is unless I turn on pchip interpolation in MFP).

Found my personal sweet spot to be KP1.8, still get the nuances but don’t have to look at a shaky monitor.

[randomus]

@Falafel

The device just got more responsive and snappy. It is now able to reproduce those tiny vibrations people add to their scripts, and perhaps a bit too faithfully. And while it is doing that, there’s an earthquake on my desktop.

Ngl. Doesn’t come unsuspected.

@Falafel

which is unless I turn on pchip interpolation in MFP

Just from afar, I would assume a combination of high Kp (which most certainly is the proportional gain of the controller) and the use of PCHIP offers the best combination of high responsiveness for quick and short movements, as well as be less linear and mechanical for a lack of a better term and generally deliver less mechanical shocks (better durability and sound emissions) to the whole setup and table. Either way… a sturdier desk is needed. Just imagine it with a 3 kg sleeve ^^

@Falafel

• Distance between the bearings on the receiver: 39.5 mm
• Distance between the bearings on the motor: 26.0 mm

Hold on… I just clocked that they’re different. Does that mean that a simple L0 0-100 movement changes the pitch angle of the receiver as well? Or do they compensate for that in the firmware?

[Falafel]

@randomus

Hold on… I just clocked that they’re different. Does that mean that a simple L0 0-100 movement changes the pitch angle of the receiver as well? Or do they compensate for that in the firmware?

Irrc during a simple L0 stroke, the pitch motor also moves to keep the receiver upright, if that helps.

[randomus]

Thank you. Yes that helps. And is a good thing

[Falafel]

Double checked and I was wrong… Pitch doesn’t move during L0 movement, mine moves because I left “link” on in MFP.

So yeah L0 stroke does change the receiver’s pitch by a little.

[nannyskexis]

@Falafel
I have the version with the fan mounted at the bottom and used the same app as you for the recording. The fan is still audible (36db, way louder than my PC case fans in idle). Could you send me the link where you bought the replacement fan, please?

[Falafel]

I used one of these red “Usongshine” fans bought off Amazon. Stock varies according to region. If you can’t find it, get whatever 5010 brushless fan marketed as silent - doesn’t matter if they have a slower RPM.

As you replace it, could you take a picture of the fan installed on your device? I thought they have switched to better ones.