Quest Pro has finally been revealed, and while it’s a much more expensive headset, inevitably people are interested in knowing how it compares to Meta’s next best headset, Quest 2. So here we’re going to look at Quest Pro vs Quest 2 specs side-by-side and break down what they really mean in detail.

There’s so much about XR headsets that is difficult to communicate from paper specs alone. So similar to my PSVR vs. PSVR 2 spec comparison here we’ll look at the numbers and I’ll use my recent experiences with the headset to explain more about what they do or don’t mean.

Quest Pro vs. Quest 2 Specs

Quest Pro Quest 2
Resolution 1800 × 1920 (3.5MP) per-eye, LCD (2x) 1,832 × 1,920 (3.5MP) per-eye, LCD (1x)
Refresh Rate 72Hz, 90Hz 60Hz, 72Hz, 80Hz, 90Hz, 120Hz
Optics Pancake non-Fresnel Single element Fresnel
Field-of-view (claimed) 106ºH × 96ºV 96ºH × 96ºV
Optical Adjustments Continuous IPD, continuous eye-relief Stepped IPD, stepped eye-relief (via included spacer)
IPD Adjustment Range 55–75mm 58mm, 63mm, 68mm
Processor Snapdragon XR2+ Snapdragon XR2
RAM 12GB 6GB
Storage 256GB 64GB, 128GB, 256GB
Connectors USB-C USB-C
Weight 722g 503g
Battery Life 1–2 hours 2–3 hours
Headset Tracking Inside-out (no external beacons) Inside-out (no external beacons)
Controller Tracking Inside-out (headset line-of-sight not needed) Headset-tracked (headset line-of-sight needed)
Expression Tracking Yes (eyes, face) none
On-board cameras 5x external, 5x internal 4x external
Input Touch Pro controllers (rechargeable), hand-tracking, voice Touch v3 (AA battery 1x), hand-tracking, voice
Audio In-headstrap speakers, dual 3.5mm aux output In-headstrap speakers, 3.5mm aux output
Microphone Yes Yes
Pass-through view Yes (color) Yes (B&W)
MSRP $1,500 $400 (128GB), $500 (256GB)

So then, let’s talk about all of this.

Display & Optics

This is one part of the spec sheet that’s actually pretty straightforward. With nearly the same resolution between Quest Pro and Quest 2, there’s just not going to be a real jump in resolving power between them. That said, Meta claims Quest Pro should look at least a bit sharper thanks to the optics.

The big deal with the new pancake optics in Quest Pro is that they’ve allowed the headset to shrink so it doesn’t feel quite as much like a box on your face. But Meta also says they are clearer, offering a 25% increase in sharpness at the center, and a 50% increase in the peripheral region.

That should mean the Quest Pro has a much better ‘sweet spot’ (the variance in clarity from one edge of the lens to other). However, it’s not apparent to me yet if these supposed increases in sharpness are specific to just the optics or if they include Quest Pro’s display as well—after all, if your resolving power is fundamentally limited by the number of pixels on the display, increasing the clarity of the lens won’t make any difference (I’ve been attempting to clarify this with Meta).

Granted, we know that the periphery of Quest 2’s lens is definitely limited by the lens rather than the display, so any increase there is sure to be an improvement in the overall sweet spot of the lens, especially if it’s really by 50%!

Regarding the refresh rate of Quest Pro and Quest 2… while the latter originally shipped with a 72Hz refresh rate, it can now run up to 120Hz. Quest Pro on the other hand tops out at 90Hz, and sources tell me it’s unlikely that it will increase after launch.

90Hz is largely considered the industry standard for a ‘good’ refresh rate on a VR headset, whereas anything higher has proven to be a nice-to-have but not quite essential feature. On Quest 2 only a small number of apps actually run at 120Hz. I’d bet the majority actually target 72Hz.

Optical Adjustments

Photo by Road to VR

Here Quest Pro gets some upgrades that I’m happy to see. While both headsets technically have a physical IPD adjustment and eye-relief adjustment, Quest 2 is somewhat hampered in this regard because the lenses can only be placed in three discrete IPD positions (58mm, 63mm, 68mm) and the eye-relief only has two positions. Neither of these adjustments can be easily made while wearing the headset itself.

Quest Pro on the other hand gains a continuous adjustment for both IPD (55–75mm) and eye-relief. And both adjustments are made with dials on the headset, making it easier for anyone to find the best lens position with less fiddling.

And one small but meaningful bonus on Quest Pro comes from its eye-tracking sensors… the headset will automatically measure the distance between the user’s eyes and tell them to position the lenses in the correct place. This is great because most people don’t actually know their IPD value, despite it being pretty important to XR devices, nor are most people particularly good at finding the right IPD position by visual inference.

Processor & RAM

Qualcomm’s Snapdragon XR2 is the industry-standard chip for immersive XR headsets, and it has faithfully powered Quest 2 since the headset’s launch two years ago. But for Quest Pro, and all of its extra sensors, Meta needed more power. Instead of waiting for the next-gen XR2 chip, the company worked with Qualcomm on an interim solution which it calls Snapdragon XR2+.

My understanding is that XR2+ is—transistor for transistor—literally the same chip as XR2 but backed up by twice the RAM (12GB) and a better cooling solution which allows it to run at higher speeds without overheating. Anyone who has ever overclocked their PC knows just how much heat is a limiting factor in processor performance; Meta’s claim that Quest Pro is getting some 50% more power out of the same chip is not unreasonable.

SEE ALSO
Expect 1-2 Hours of Quest Pro Battery Life, Says Meta; Included Dock May Ease the Pain

The big question on my mind is how much of that additional power translates to a headset that can run existing content at higher fidelity as opposed to how much of the extra power is reserved for the headset’s system-level functions. I would guess the majority of the extra power is going to the latter—after all, Quest Pro not only has higher resolution sensors, it has six additional sensors compared to Quest 2. All of that extra incoming data needs to be processed with as little latency as possible.

Assuming that’s the case, existing Quest 2 games probably won’t run any better on Quest Pro, while the benefits in processing power will come in the form of new apps that take advantage of system-level capabilities like improved passthrough and face-tracking.

I’ve asked Meta and third-party developers for more info on this front. Specifically I’m interested to know if an app that doesn’t need face-tracking can tap into the cycles that would have otherwise been reserved for that feature.

Field-of-view

Photo by Road to VR

This is another place where the specs as we see them are fairly clear: there isn’t a particularly obvious difference to me in the field-of-view between the two headsets.

However, a nice little bonus from the new continuous eye-relief adjustment on Quest Pro is that everyone will be able to move the lenses as close to their eyes as they are comfortable with. With varying facial topography, people can experience fairly different fields-of-view when the lens position is static. The ability to move the lenses forward and back means more people will be able to experience Quest Pro’s maximum field of view than if the lenses couldn’t move.

Another note on field-of-view is that Quest Pro is designed with an open peripheral view (compared to Quest 2 which blocks as much of the outside world as possible). This is intentional to make the headset’s passthrough mode feel more natural (since the real world in your periphery matches up with the passthrough view in the center of your vision.

Thoughtfully, Quest Pro also ships with ‘light blockers’ which magnetically attach to hide much of the outside peripheral view. An optional ‘full light blocker’ accessory goes even further and provides a more complete face-gasket for blocking out as much of the outside world as possible.

In the future it would be nice to see some kind of dynamic solution to opening or closing the peripheral field-of-view to the real world, but in any case, Quest Pro gets good marks for a high level of flexibility in this regard.

Continue on Page 2: Weight, Battery Life, Controller Tracking, & More »

1
2

This article may contain affiliate links. If you click an affiliate link and buy a product we may receive a small commission which helps support the publication. See here for more information.


  • mellott124

    The lack of increased resolution is a bit disappointing. The battery life is shocking. Quest 2 already feels a bit short. And 1-2 hours tells me it will be closer to the 1 hour end, otherwise they’d just say 2 hours. Everything else I’m excited to try out.

  • Duane Aakre

    Any word on the battery life of the controllers? I can see doing the extra battery in a pocket to extend the time of the headset, but if the controllers are only good for the same 1-2 hours, then an extra battery for the headset may not gain you any additional time for the overall system.

    • Zombie

      The rumor is as you fear. A 1-2 hour battery life for the controllers as well.
      So if you get the controllers, stock up on rechargeable batteries.

      • ViRGiN

        Both Carmack and Boz said they never run out of controller power during their few hour long sessions.

      • Jistuce

        I read that the new controllers have integrated rechargable lithium cells. Which, while I dislike the long-term implications of cell degradation, is a usability win over replacing a pair of AA cells constantly.

        • JanO

          Use rechageable AA cells and have ALL the options… !!?

          So many of my tech toys have become unusable overtime, just because of battery wear, that I now dislike NOT having the possibility to easiliy change batteries on the fly…

          • Jistuce

            Rechargable AA cells is a nice retrofit, and it is what I use on my current Quest controllers.

            But I like 18650s. I wanna use ’em in more things.

      • Mr.Philgood

        According to Meta it’s up to 8 hours of battery life for the controllers

    • jarrodbraun

      UploadVR reports 8 hours for the controllers

    • Boz said up to 8 hours

  • Mike

    no thanks, ill pass.

    • Yencito

      Are you business?

      • I’m business, and literally their demographic for this device and I’m gonna pass.

      • Andross

        as I large explained in the quest pro presentation article, this device seems unable to cover the most needs of our and our customer businesses, and I asked someone to tell me an example of business coverage beyond flexing face expressions in a short VR presentation with some geek customers.
        AR assistance is for Hololens, other than that most of business needs require much time, and most of the time wireless is more an handicap than a feature.

  • Michael Barkdoll

    Did your sources say why 120 Hz wouldn’t be supported? Without 120 Hz I don’t think I’d enjoy gaming on the device.

    • MosBen

      From the article it sounds like what they’re saying is that higher refresh rates are nice to have but require more computing power to push, while 90 Hz is good enough for most people. And maybe it’s the case that the the XR2+ doesn’t provide enough of a boost to process all of the new input data while maintaining the higher refresh rate.

    • Ben Lang

      Display limitation they said.

      • Mico

        Since the Pico 4 uses the similar pancake lenses will the Pico 4’s display also be limited to 90Hz or do you think they’ll eventually support 120Hz via Virtual Desktop?

        • Christian Schildwaechter

          The impact of the lenses on the refresh rate should be only indirect. The only part that could technically restrict the refresh rate is the LC filter part of the display, as the lenses don’t require any refresh and any LED used as a backlight should support 1000Hz or more.

          HMDs don’t run the backlight constantly, but instead flash the screen for a few milliseconds to reduce the time in which the “wrong” image is displayed during head movements. This means that the actual brightness of the display is much lower than what it could technically achieve. This is made worse by the inefficiency of pancake lenses that waste even more of the light.

          The Quest Pro uses a quantum dot LCD with mini LED backlight, which allows do adjust brightness for several regions instead of only the whole backlight at once, while the Pico 4 uses a more traditional LCD with a single brightness LED backlight. It is quite likely that both are actually capable of running at 120Hz, but that would cause the image to either be darker or, with a longer per-frame display time, could increase nausea for some users. Alternatively a brighter backlight might have been used, further increasing power consumption and heat issues.

          Obviously running at 120Hz also requires a much higher rendering performance, so it is of limited use on any XR2 device, and only a few apps really benefit from it. So the limit to 90Hz may be less due to the display and more a design decision based on mix of brightness, comfort, power management and a too limited use case for the higher refresh.

          • Mico

            How much do you think Wi-Fi 6E will help with the render resolution capabilities? Meta said Quest Pro will have twice the wireless speed releasing later next year as a software update compared to Quest 2. Also, Pico 4 connects at 150MBps on Wi-Fi 6 maybe 50% improvement over Q2. If cost was not a factor and a person were willing to play games at 90Hz then which headset do you think will have the better display for games?

          • Christian Schildwaechter

            Not a lot. AFAIK all compressed VR streaming is less limited by bandwidth and more by compression time. Technically a USB 2 connection with 480Mbit/sec has more than enough bandwidth to carry the 150Mbps typically used by Link/VD. A Wi-Fi 6E router will give you more bandwidth, but a Wi-Fi 5 router is already fast enough. I forgot the exact reasoning why the full bandwidth isn’t used, but it basically boils down to higher/better compression on the GPU taking more time and thus adding latency, so the current compression rate is mostly limited by timing, not by available bandwidth.

            The main reason why people suggest better routers are the overcrowded frequencies. If you are streaming to your device in a hut in the woods, you may even be able to use a Wi-Fi 4 router, but if you are not the only human around, there is quite a chance that esp. the 2.4Ghz band is completely saturated by dozens of Wi-Fi and bluetooth devices, portable phones, smart switches and running microwaves. Moving to 5GHz removes anything but other Wi-Fi devices, and cheap ones like smart light bulbs usually all stick to the 2.4GHz, so your chance of a connection/stream without tons of collision increases. Whats really new for Wi-Fi 6E over Wi-Fi 6 is that it now also allows using the 6GHz band, so if you live in a very crowded area and not even a dedicated Wi-Fi 5/6 5GHz router is able to give you glitchfree streaming, a Wi-Fi 6E router will probably help, at least for some time.

            The increased speed may even shave of a few milliseconds of latency. But with most of the bandwidth currently unused, the stream is actually send in bursts, with longer times of inactivity during which the PC is encoding or the HMD decoding the next frame. The theoretical top speed of Wi-Fi 5 is 6933Mbit/sec, meaning that in an ideal case the connection can transfer all the frames needed for 1sec in just 0.03sec total, the rest of the time between the single frames it idles. So if you increase the max bandwidth by 50% with Wi-Fi 6E, you save 0.01sec for each second of streaming, meaning 0.12ms per frame @90Hz or about 1% of the 11.1ms frame time.

          • Mico

            A typical airlink at 1200 MBits/ps on Wifi 6 is around 105 MBytes/ps which is actually a bit slower than this theorical maximum due to environmental factors that you’ve noted. Your theorical maximum of these routers is based on standards that most clients don’t support. E.g., the Q2 XR2 supports a 2400 Mbps connection in a 2×2 client to wifi 6 routers, but they choose to only use one of those 160Mhz channels. Also, a compressed data connection (it is on Q2) over USB from pcvr to quest 2 headset can have better quality if your argument is the GPU is the limiting factor. Clearly a wired compressed connection with more bandwidth results in better quality without the GPU being the limitation. I’m starting to think how much a factor this is situational. E.g., the GPU needing to render frames not in the FOV, which I’ve now found super sampling reduces this latency over wifi since the frames are what I believe is a pre-rendered state. Anyway, this was actually not the question I was most interested in since you seem to understand the display technology best. How noticeable better do you think the quest Pro display is over the pico 4?

          • Mico

            You do realize USB from PCVR to Q2 is compressed and not lossless right?

          • Christian Schildwaechter

            I do.

            uploadvr_com_playstation-vr2-hands-on-preview/#comment-5981891332

  • Kevin Brook

    The headset has some interesting features, and is hand’s down the sexiest looking device so far relesead. The charging dock is also a great premium feature to have. They have made some very strange decisions as regards hardware specs though. Firstly, this XR2 plus chip is simply the old XR2 with better cooling and seems insufficient for the device, in a similar way to the Quest 1 was underpowered at launch. There was no race to get this out, and I can’t for the life of me understand why they delayed it from last year, only to release now with two year old tech running it. The XR2 gen 2 launches next month so why not release this early next year with that chipset so the device has longevity?

    The same could be said for the screen too. 1800 x 1920 per eye frankly sounds embarrassing for a $1500 headset releasing as we approach 2023. I’m looking for something sharper than my 2160 x 2160 per eye Reverb G2, I certainly don’t want to pay three times as much for a device that goes backwards.

    Finally, for existing VR users, who want a high end experience, no full light blocking facial interface in the box and no display port connectivtity largely kills this device dead as a viable high-end PCVR headset, which is what most of Meta’s existing enthusiast customers wanted.

    Disappointing, on to the next one I guess.

    • kraeuterbutter

      have the Reverb G2 too – lot of that 2160×2160 resolution is vasted because of the bad E2EC
      so even when the resolution is lower on the quest pro,it can mean that you can read text better, specially outside the very center – as on the Reverb G2

    • kraeuterbutter

      for the release date:
      maybe Pico has something to do with it ?
      they have released the pico4, which is a Quest2 competitor, but better specs (pancake lenses, lighter, more comfort, …) for a cheaper price
      and soon pico will release the pico 4 business, which also supports eyetracking and facetracking – but supposed to have a lower price then the meta quest pro

    • MosBen

      I feel like the Pro Quest models will be the test bed for tech that gets improved and brought down in cost for future non-Pro Quest models, while maintaining compatibility with the current non-Pro generation of hardware. Which is a complicated way of saying that I bet that the Quest 3 has the XR2 Gen 2 while incorporating the face tracking, lenses, tracking, etc. features from the Quest Pro. The Quest Pro will have worked these features out and given developers some time to get comfortable using them so that when the Quest 3 launches as a heavily subsidized device developers will be ready to release compelling content for it, where Meta hopes to recoup its hardware losses. Then the Pro 2 will come out a year or two later and will similarly use a beefed up SoC from the Quest 3 while adding new features that Meta wants to incorporate into the Quest 4.

      This is a very long way of saying that the Pro models are the non/less subsidized product where the professionals hand hardcore audience pay Meta to bring features out of research and into a commercial device so that they can find ways to bring the cost down when its incorporated into the heavily subsidized consumer device.

      • Christian Schildwaechter

        The problem with the test bed for developer theory is that the most important changes of the Quest 3 will probably be a vastly faster processor and an increased resolution, both of which will allow for a qualitative jump in games, and none of which the Quest Pro provides. Eye tracking could be a more interesting aspect, not necessarily for DFR, but for improving interactions in VR. I seriously doubt that outside of social VR many of the existing Quest apps will find much use for the face tracking.

        I assume that the Quest Pro was intended pretty much exactly as what Meta says it is: a headset aimed at VR conferencing to be used with their own Horizon Workrooms and now announced cloud based Microsoft products. Which automatically means a different target group less sensitive to a higher price IF the headset does indeed provide the productivity benefits Meta claims. In a market that doesn’t necessarily care if the chip design is cutting edge, as long as the promised performance is met. Businesses often prefer somewhat older, but stable products for which someone else already figured out most of the glitches.

        The Pro is based on the Quest 2 because it makes no sense to develop a second, incompatible platform that will sell only in small numbers. They probably expect some enthusiasts to buy it too, but it was never truly intended for gamers, not even for game developers, and John Carmack just warned against developing on/for the Pro and then trying to get it to run on the Quest 2 afterwards.

        The “I am disappointed” reactions and wild guesses why the Quest Pro turned out this way only make sense if one expected it to be an improved Quest 2 for everyone (ignoring price), while the specs should be looked at asking if the product in its current form makes sense for the actual target group that Meta has left little doubt about since they first presented Cambria. The “will it work for business applications” barely depends on the hardware specs and much more on Meta providing significant or any improvements over the Zoom conferences everybody is currently relying on, a least enough to justify the extra hassle required with any headset. This will make or break the success of the Quest Pro, and so far the results were unimpressive.

        • silvaring

          I’m starting to wonder if the reason Facebook didn’t launch a consumer friendly Quest Pro style device is because they couldn’t secure enough supply for the pancake optics in the headset. Would make sense given a lot of Chinese investment seems to being put into pancake optics over the last few years (Huynew).

    • ViRGiN

      Quest users are enthusiasts, PCVR users are fanatics.

  • kraeuterbutter

    a comparisson between Quest Pro (1500US$) and Pico4 (429 EUro) would have been (more) interesting…
    Pico4: 2160×2160, also Pancake lenses, also continouse IPD (but motorized), more RAM than Quest2, more Power than Quest2 (better cooling than Quest2), very light, battery also on the back, also very good edge2edge clarity

    but: Display-colors more like quest2 level, not the supposed much better quest Pro Level

    price: lower than Quest2 now

    • eadVrim

      With Pico 4 you need to wait at least one year more until it reaches the level of the current software and games library of the Quest 2, and it will never reach the level of XR that the Quest Pro can provide.

      • silvaring

        Pico 4’s main drawback seems to be the PC wired compatibility issue, lack of DP. But given this is the first affordable pancake lens headset on the market and Pico have been in the biz for so many years with multiple SKU’s, I think its a good launching point for the brand in spite of the PC drawbacks.

  • Christian Schildwaechter

    Regarding pancake lenses and extra power:

    The added clarity towards the edges compared to only about 20° to 30° of full clarity with Fresnel lenses on Quest 2 is of course welcome, but now unbalances the device. On the Quest 2 you get around 25% performance increase by using fixed foveated rendering, which reduces the resolution in the peripheral vision, the same areas where the image is somewhat blurry due to the Fresnel lenses too, so the one limitation hides the other at least partly. With the better clarity of pancake lenses, developers will have to be less aggressive with fixed foveated rendering, as it will be more noticeable.

    And AFAIK John Carmack confirmed during his keynote that motion estimation for eye tracking currently adds about 50ms, similar to what developers have measured with Tobii eye tracking on the HTC Vive Pro Eye, making it basically too slow/unusable for dynamic foveated rendering due to the eye permanently jumping towards areas still rendered at low resolution. As Sony didn’t seem to be having these problems during their PSVR 2 hand-ons, maybe they really don’t rely on reducing the render resolution, but instead reduce geometry complexity based on eye position as mentioned in a patent earlier this year, potentially making artifacts less visible.

    Without good dynamic and reduced fixed foveated rendering, the Quest Pro would have to rely on a higher GPU performance to render at the same level as the Quest 2. Which the XR2+ can provide, though I’m still not sure how much. I’ve read about the “extra 50%” in two versions. Most often it is given without context as a performance increase, but others sources mentioned that this is actually the extra power that the XR2+ itself has available, meaning that thanks to the RAM no longer being placed on top of the SoC and other measures the chip can now consume more power without overheating. Qualcomm themselve mention that the XR2+ has up to 30% higher GPU performance or a 30% lower energy consumption, so the 50% is most likely a mix of no longer running the CPU part underclocked like in the Quest 2 plus gains from an improved process. And not a 50% gain in frame rates for a comparable render resolution.

    Clocking the SoC higher with performance gains is as welcome as the higher clarity of pancake lenses, but both are less efficient than the solution in the Quest 2. Higher core voltages provide extra performance, but power requirements rise exponentially for linear performance gains, which is why mobile gaming handhelds are usually underclocked to improve battery life compared to overclocked gaming PCs where battery isn’t a limit. The same is true for the pancake lenses, which lengthen the optical path with multiple internal reflection of polarized light, but without having a true polarized light source this means that a lot of the power/light generated by the display is actually lost in the display. So both improvements are bought at a high energy cost and most likely responsible for the bad battery life. Meaning that while the tech is impressive, it leaves the whole Quest Pro as a somewhat unbalanced headset.

    • Newlot

      I really admire your knowledge

    • silvaring

      Very interesting, thank you!

  • Rogue Transfer

    Indeed! Even more so, if you’re using it for social interaction. Not to mention that constantly leaving it on the charging dock is bound to degrade the battery longer term, as it keeps topping it up over the time you’re not using it. Though, this device isn’t likely to have a long shelf-life before they replace it with an upgraded Quest Pro 2 with a newer SoC.

  • G’Zilla

    This seems like a major fail for biz use…unless plugged in. And, i got to have all the new shinny toys, but the price. So its wait for the quest 3 to play my games.

  • Well whaddaya know … the most shrill Meta Shill is in tears. Did your buddy Mark Zuckerdroid let you down? Sure seems like he’s letting the whole VR industry down. Meta is TANKING.

  • Jorge

    There’s no way in hell, I’m going to fork over $1500 for that slightly upgraded Quest 2. Now I have a great reason to get the Valve Index, unless I wait for the New Valve release. In any case the only reason there was a Quest on my desk is because it is a great machine for $300. I wanted to get into VR, I have and now I know enough to make a very informed choice and it won’t be the Quest Pro…

    • Cless

      Look, I like it as much as you but… its not for you, unless you are thinking about using it as an enterprise. Getting a Valve Index is a bad idea. It was pricy for what it offered in 2019, now its just flatout overpriced and starting to show its age, so please wait for any other new headset.

      • Snekertechie

        Seems like valve never drop the price of their headset. I agree the price does not make sense anymore. I hope valve index 2 will have a good wireless option.

  • Trish Rolt

    It would be worth mentioning that the Quest Pro supports eye and face tracking as an input method.

  • Matthew

    Damn. I just wanted to buy them ( i’m still using rift S bought about 3 years ago) but if pro isnt as much progress and improvement vs quest 2 i need to wait untill quest 3 will be relased. What a shame. And this price and weight …

  • duked

    Just buy the Pico 4. Much better hardware than Quest 2 and some features similar or better than the Quest Pro. Already before launch, it has lots of games and the price is lower than that of the Quest 2.

  • Jasmin Frenette

    The only real problems with the Pro is the price. make it 699 $ and it’s a good evolution.

  • ViRGiN

    Just bought 3