12bytes Mumble meet every Sat. night!
Coronavirus information & resources
Vaccines - What You Need To Know

VPN Provider Review: AzireVPN

AzireVPN, operated by Netbouncer in Sweden, was recommended to me by one of those geeky, super knowledgeable hacker types who detailed some really interesting differences between Azire and other VPN providers. And what are those differences you ask, mouth watering in anticipation?

Well, first let's get something straight regarding VPN providers: there isn't a damn one that can be fully trusted, at least none that i know of. They can tell you whatever they want about their security and privacy and no-log policies (many of them are flat out lying when they state this), but unless there's an information leak, or you discover a security or privacy issue yourself, or you personally know the people running the company, your confidence in their service will always be blind. Tor advocates like to use this ammo to suggest that Tor is far better in this regard because it's open source and uses multiple nodes and multiple layers of encryption, yada yada yada, but i find their claims of security to be less than concrete. For example a bad actor, such as your ISP, can apparently run an entire Tor network on a single machine using something like The Shadow Simulator and god knows what the intelligence community can do. Tor has other problems as well, some of them detailed in my article Tor versus a VPN - Which is right for you?.

Understand that i'm not suggesting that a VPN is necessarily superior to Tor in every case, but i think that what path is best chosen depends on what you're trying to achieve and i think that for the average user who's downloading ... things ... or wants to circumvent YouTube's idiotic geo-restrictions, a VPN may be the better option, though unlike Tor, VPNs are not free and any provider that claims this is a good one to run the hell away from at maximum velocity.

Back to Azire...

AzireVPN claims to do things very differently. Unlike every other (or mostly every other) VPN business where one can sit behind a keyboard and provision as many servers around the world as they please, Azire tells us they physically own, configure, secure, install and maintain each server they operate (if you search the images on their domain you can find some circumstantial evidence to support this). From a security/privacy viewpoint i see this as a huge advantage over other mega-VPNs like NordVPN, ExpressVPN, AirVPN, etc., who are potentially more open to hacking and government snooping.

Azire makes the following claims...

  • they own and maintain the hardware
  • nothing is stored physically on the servers (no hard drives) - the entire system runs in RAM (more here)
  • all USB, VGA and serial ports are sealed to prevent tampering (more here).
  • they support WireGuard which is apparently faster, better, easier and less bloated than the OpenVPN protocol
  • no logging
  • no port restrictions (torrenting, etc., is allowed)

AzireVPN was featured in TorrentFreak's article, Which VPN Providers Really Take Privacy Seriously in 2021?.

Sounds like the berries, right? There is one downside to managing your own hardware though in that they can't provision equipment as quickly as the fast-food VPNs and so Azire doesn't have a heck of a lot of servers, but the ones they do have are located in quite a few countries and they seem to be slowly expanding (see their blog for more). Azire does offer SOCKS5 proxies, however you must be connected to one of their VPN servers to use them and there is no encryption at the proxy level. Still, their SOCKS5 servers make it easy to change your location/IP in order to circumvent geo-restrictions. For those like myself who run their VPN client on their router this is a plus because, while it isn't as straight forward to swap locations, there are plenty off web browser extensions available that provide the ability to quickly switch between SOCKS proxies.

I started with AirVPN several years ago then moved to NordVPN, but being with a huge company like Nord, who seems to be less than transparent, has always bothered me and i'm glad to have found an alternative which i think is better all around. Although it wasn't an issue when i first signed up, Nord's servers have become blacklisted by quite a few sites and it started to get annoying, as did the lack of the connection stability.

Getting AzireVPN set up on my router was a bit of a pain in the ass. At first i was using the DD-WRT firmware and even after contacting Azire support i could not get OpenVPN or WireGuard working. Truth be told, their setup guides are out of date and, although they say they support OpenVPN, i'm not convinced they do, at least for some configurations which they claim to support. Azire seems to be moving away from OpenVPN in favor of WireGuard, but this is all pretty new stuff and so there can be hitches in setting up WireGuard as well. I finally got the tunnel working with WireGuard after switching to the OpenWRT firmware and a lot of fiddling around plus still more help from Azire support. Azire definitely loses points here though their support has been mostly OK (i'll get more into that in a bit). If you decide to use their app however, you can likely avoid the hassle i had and they have a healthy selection of apps for different platforms/devices.

Another big plus with AzireVPN is that you don't have to give them any personal information to open an account and you can pay with cryptocurrency, so acquiring their service can be totally anonymous if you want, especially if you use another VPN or Tor to sign up. There aren't a lot of other VPN providers that go this far to protect your privacy.

As far as the tunnel itself, all ports are open and bandwidth is unlimited. I've only been using their service a short time, but speed seems really good in tests, though this may have a lot to do with the WireGuard protocol since it has less overhead. Also i haven't yet had much trouble accessing sites which had blocked Nord's IPs. The stability of my connection has been very impressive whereas this was not the case at all with Nord.

Now, back to that support thing...

Because i couldn't get DD-WRT working with the OpenVPN protocol, a configuration which Azire claims to support, i was offered some free time without having to ask for it. I appreciated the offer and viewed it as the right thing to do, especially for a smaller company which is apparently interested in growing. Problem is, they didn't follow through and so i inquired again about their offer. Crickets. In the end i inquired four times before i got a response, and their response was to renege on the offer because i didn't help them figure out how to get DD-WRT/OpenVPN working on their tunnels, a condition which was absolutely never stated nor implied. Here's what they said, emphasis added:

We thought that our offer was pretty clear while saying the following statement:

"Whether you manage to find a solution to your issue, we will be glad to give you free time and eventually we will make a quick update to our guide."

In other words, if you were able to find a solution which we could integrate into our guide to update it, we would give you free time. I think our sentence was poorly written, but that is what we meant.

Their offer was unconditional. It did not hinge upon anything. Needless to say, their blatant twisting of their own words pissed me off and so i fired back a rather terse reply calling them out on it. Shortly after receiving my mail they did extend my service time for a month, so in the end they did what was proper and ethical, but what they should have done was not ignore three mails regarding the issue.

All in all i think AzireVPN offers some uniquely attractive and important features and they manage to do it at a very competitive price. If you decide to go with Azire please consider using my referral link which helps me out a bit.

UPDATE, 18-Apr-2022

I introduced AzireVPN in a Github repository, Lissy93 / personal-security-checklist, in issue #140, [CONTENT-CHANGE] Privacy-Respecting Software > Virtual Private Networks, and 'Lissy93' brought up several concerns, some of which i shared, and so i emailed Azire. Following is their reply to these concerns which i'm personally fairly pleased with:

Q: Client applications not open source. And their only GH repo is very stale
A: It is true that the source code of our current WireGuard applications is not released yet. It will be when we feel confident that the code is ready and mature enough so that everyone will be able to review, submit issues, and contribute with merge requests.

Our GitHub currently hosts the source code of our old OpenVPN client, which is now deprecated and not maintained anymore.

Q: Android App only available through Google Play, no F-Droid or APK
A: We are planning to, at a minimum, release our Android application on F-Droid, probably at the same time we release the source code.

Q: Unsure why the Android app needs external storage read/write permissions
A: The Android application needs external storage read/write permission to be able to write debug logs, which are available from the hamburger menu. Users can then send us the log for support inquiries.

Q: No kill switch option on client apps, and Linux app disconnected several times
A: It is planned to integrate a kill-switch in our clients on all platforms where it makes sense and can be properly implemented.

The Linux client is deprecated. Linux users can use WireGuard's wg-quick directly, or better, use systemd services, for now. They also can use NetworkManager's OpenVPN GUI applet to ease the establishment of an OpenVPN tunnel.

Q: Their only DNS servers are in Denmark, part of the 9-Eyes
A: Our static public DNS servers are located in Sweden. When connecting to our service, users will be assigned with the endpoint's local DNS servers, which should keep the DNS requests internal to the location's local network. It is therefore possible to avoid country deemed untrusted.

Our static public DNS servers are listed on this page, under the "DNS servers" section:

https://www.azirevpn.com/docs/servers

Q: No security audit. And no evidence to backup any of their claims
A: We are planning to make an audit of our back-end infrastructure when we feel ready to do so. For the moment, the back-end is reworked for the release of port forwarding, which should happen in the incoming months.

Q: My traffic was flowing through shared data centers, they cannot / do not physically maintain these themselves, like they made it sound like
A: We buy all our hardware (servers and switchs), seal it, and then send it to data centers around the world. It would not be feasible to own our data centers, although we have close business links with some of them, so we know they can be trusted.

More information on these pages:

- https://www.azirevpn.com/docs/environment
- https://www.azirevpn.com/docs/security

Q: Relatively few locations, and expansion seems to have slowed down
A: See answer number 7. It is less easy to find trusted and quality data centers to send our hardware to, than simply leasing a server which can be terminated at any time.

During 2022, we are striving to expand our locations on the West Coast of the United States.

Q: Surprisingly small throughput compared to other providers, possibly making identifying individuals easier
A: We are not sure if "small throughput" refers to "low traffic" on some locations from our Status page, or if the speed when testing was not great. It usually depends on a lot of factors, but our locations are, for some of them, using Tier 1 providers directly (Cogent, Telia) so the speed should be there. Also, our servers are for the most part using 10 Gbit/s full duplex links.

Q: When trying it out, I found performance was quite poor, and not all their advertised servers were connectable. But this could be due to my location
A: Unless indicated otherwise on our Status page, all our locations are available for use. We have automatic ways to detect down locations on our side, so there should be no issue connecting to them unless an Internet Service Provider banned some of our locations' IP addresses.

We are open to answer other questions or clarify some points if our answers were not complete enough. Alicia can contact us directly to our support email address.

Product Review: Schiit Hel 2 External Headphone AMP/DAC

Dealing with Schiit Audio was not a pleasant experience. Perhaps some of my frustration can be chalked up to the alleged COVID-19 "plandemic" (yes, the spelling is intentional), but i'm not sure that accounts for all the trouble.

Second of all, i'm not very technically knowledgeable regarding digital audio, but i do have a basic understanding and i just wanted to share a potentially unique and early review of the Schiit Hel 2, "early" meaning i've only had the unit in my hands for a week or so.

The Hel 2 Gaming AMP/DAC is primarily aimed at headphone and headset users who also want to run their microphone through the unit, thus it suits gamers well. The $199 USD Hel is an upgrade to the $109 Fulla 4 (up from $99) for those who require more power and options. I no longer do much gaming, nor is that why i bought it. My reasons for wanting the Hel were because the Fulla 4 i was sent was DOA and i wanted to see if could gain a better listening experience verses my on-board sound system. Also i needed to clean up the noise floor on my cheap microphone.

While the Hel looks and feels great on the outside with its weighty steel and aluminum exterior and smooth, precise knobs and switches, there seems to be serious shortcomings with the internals. While most of the reviews i've read and watched were enthusiastically positive, others were quite critical. While i was writing this review and listening to music, my Hel started producing the same 'popping' every few seconds that quite a few other customers have mentioned. For example there are several people on Reddit that very recently have had problems similar to what i experienced, both with Schiit and its products. Here's a couple partial quotes from the beginning of the thread...

'HuskySlim' opened the thread with...

All i’ve seen so far are a lot of people with a lot of issues on their new Fulla 4 / Hel 2 units. Is this something that Schiit is aware of and actively correcting with the outgoing units? Seems like a lot of units are effected.

'b34tn1k' replied with...

I got my Hel 2 yesterday and it's cut out a bunch of times. I've emailed their support detailing the steps I've taken to trouble shoot and their reply was to ask if if I'd taken any of the steps that I already detailed.... It was like they didn't read my email beyond the subject line.

Another user replied...

Audio randomly cuts out/pops for a second and I have to unplug the power cord from the hel everytime I turn on my computer or my mic wont work

The 'popping' issues people are reporting, which i would characterize as a roughly 100 millisecond skipping in sound every second or two, as though the unit were turned off and back on again, is not exclusive to the Hel. I noticed the exact same issue with an ARC MK2 which causes me to wonder if the issue is due to a Linux kernel setting or a problem with USB on Linux.

As i mentioned, the Fulla i received was DOA which is one reason i went with the Hel. There's no excuse for this. Schiit is shipping products which i believe they know are faulty. In Schiit's defense, 'HuskySlim' apparently wasn't using the required power supply and this may have been an issue with 'b34tn1k' also, i don't know. What i do know is that many other customers are complaining, and have been for years, and that i completely agree with the statement, "It was like they didn't read my email beyond the subject line". I'm certain this is what happened in my case, no less than three times when my concern regarding the quality of their 1/4" to 1/8" headphone adapter, which others also complained about, went unanswered.

Another Reddit thread, this one from 2 years old, reveals more complaints. Here 'deepskypics' opens the thread with...

Why the hate for Schiit these days?

'verifitting' replied with...

Schiit got called out for their quality control, safety concerns, poor objective performance, and engineering pitfalls by an experienced audio engineer with an impressive résumé. Rather than own up to it, provide their own measurements, or their own review units, they attacked his character and credentials, along with many others from SBAF and Head-Fi. He then, in turn, reviewed all of their gear and pointed out all of the flaws, albeit using pretty biased language in his reviews.

It was a big controversy on headphone/audio forums (here, ASR, Head-Fi, SBAF), with many people taking sides. When it was all set and done, Schiit ended up buying their own audio analyzer and started measuring their equipment, so the desired end-result was met (i.e., audio manufacturers being more transparent in their objective performance/engineering).

Had i found these comments earlier i would not have bought from Schiit. If you're considering purchasing a product from them, i'd highly recommend finding plenty of the most recent reviews before you decide to pull the trigger.

Another problem i had with the Hel was latency. Listening to music or doing voice chats was fine, however watching videos resulted in roughly a ~300 ms delay between someones lips moving and the sound hitting my ears. I don't know whether Schiit could have done anything different regarding design since this seems to be an OS issue. In my case i run a Linux box (Manjaro) with PulseAudio and once i screwed around with the configuration file for PulseAudio, the latency issue seems to be resolved, or at least nearly so. Here's what i did...

  1. Make a pulse folder in the .conf directory of my home directory ( ~/.config/pulse ) and copy /etc/pulse/deamon.conf to ~/.config/pulse/deamon.conf
  2. In the daemon.conf file, change default-sample-format = s16le (the 'le' could be 'be) to default-sample-format = s24le
  3. After saving the file, open a terminal and restart PulseAudio: pulseaudio -k

If you want to double check that the default-sample-format setting matches the spec for the Hel 2, or if you're tweaking your config for another sound card, open a terminal and run pacmd list-sinks and look for the sample spec: line for the card you want to configure. For the Hel 2 the spec is: s24le 2ch 44100Hz , thus why i used the s24le setting.

Here's the resources that helped me:

PROS:

  • Made in the U.S.
  • All metal, solid build quality with silky smooth volume controls and quality toggle switches
  • Flexible in terms of input/output capability
  • Can drive most any headphones thanks to the hi/lo gain switch
  • Microphone volume knob
  • Unlike the Fulla, the Hel has a mechanical switch to toggle between the USB and optical inputs
  • Power switch

NIGGLES:

  • Requires a separate (included) power supply - the data USB cable alone won't cut it, but this is the price of a rather powerful amp.
  • Microphone gain is weak, but OK, at least for me.
  • The microphone gain knob is positioned on the front of the unit which makes it a bit difficult to adjust. I think they should have stuck it on top near the output volume.
  • Price may be an issue for a dedicated gamer who is not an audiophile or doesn't need the feature-set of the Hel 2.

CONS:

  • Long standing quality control issues make buying Schiit's products a roll of the dice. In my case they shipped me a Fulla 4 which was DOA and a Hel 2 that started having problems shortly after i received it. Many others are complaining as well.

Here's the specs for the Schiit Hel 2...

Frequency Response: 20Hz-20Khz, +/-0.3db
Maximum Power, 16 Ohms: 1350mW RMS
Maximum Power, 32 ohms: 1200mW RMS
Maximum Power, 50 ohms: 800mW RMS
Maximum Power, 300 ohms: 200mW RMS
THD: <0.0008%, 20Hz-20KHz, at 1V RMS
IMD: <0.0008%, CCIR
SNR: >110db, A-weighted, referenced to 1V RMS
Crosstalk: >-80dB, 20Hz-20KHz
Output Impedance (headphones): 0.25 ohms
Output Impedance (line out): 75 ohms
Gain: Low and High, selectable via front panel. Low has maximum output of 1.3V RMS, high has maximum output of 8VRMS.
USB Receiver: C-Media CM6635
DAC: AKM AK4490 with TI OPA1656-based filter stage
Sample Rates and Bit Depths: 
USB Playback: 16/44.1 to 32/384 supported without drivers on Windows 10, Mac, Linux, Android (any UAC 2 device) with autoswitching to UAC1 for PS4, PS5, and Switch consoles.
USB Input: 48kHz
Optical Input: 16/44.1 to 24/192
Output Stage: TI OPA1656 (4 amp stages per channel) 
Power Supply: Via USB, with +/- 12V rails via high-current dual-polarity switching regulator, with inductor filtering and local regulation
Power Consumption: 2.5W typical
Size: 5 x 3.5 x 1.375” (including knob)
Weight: 13oz
APx Report for Hel 2

Dealing with Schiit Audio

I was originally going to review the Schiit Hel 2 external sound card, a headphone DAC/AMP combo, but i'm really not in any position to do so since a) it's the first and only decent external sound card i've owned and b) i only have one pair of headphones and they aren't great. What i am in a position to do however is review Schiit, the company, based on my interaction with them. Here's the story...

I originally ordered a Hel 2 from their website and waited for it to show up in my mailbox. And i waited, and waited, and nothing. I was never sent a tracking number or anything else regarding the status of my order and so i finally contacted them and asked what's up. The rep said that there was a supply problem and that the Hel wouldn't be available for another 6 to 8 weeks. That's great, i said. Were you planning on ever telling me this??? Again i waited and i eventually decided that i didn't really need the Hel and so i canceled the order and ordered a Fulla. During this period i had sent a few emails informing them that, because their 1/8 to 1/4 in. stereo adapter that shipped with their devices was considered junk by several reviewers, i wanted to make sure i was sent a good one since i had none. Although they (as in, more then one person) addressed part of my brief mails, not once did anyone acknowledge my concern regarding the adapter and i mentioned it no less then 3 times.

To my surprise i received the Fulla shortly after i order it and was quite excited to have it ... right up until the point where i plugged the f'n thing in to test it. It was DOA. The power LED never lit. I tested with a powered USB hub and, nothing. I tested with a dedicated power supply and, nothing. I tested with different cables and, noting. Most of the reviews i've read regarding Schiit equipment have been exceedingly positive, however a few reviewers mentioned quality control issues. You can add another one to the bunch. I sent the Fulla back and re-orderd the Hel 2, pleading with them to place me high upon their list since the Fulla was DOA. I didn't expect them to do so since they obviously don't fully read or respond to emails, but to my surprise i received the Hel in short order, and it worked!

Given my limitations i mentioned earlier, i'm reasonably happy with the Hel, however there is a latency issue and the microphone gain seems to be anemic. Regarding latency, this may be due to my operating system; Manjaro Linux with the KDE desktop and Pulseaudio sound system. Listening to music is totally fine, but watching a video results in roughly a 200-300 ms delay between a mouth moving and the sound hitting my ears. Hopefully i'll find a way to fix the problem. Regarding the mic gain issue, one of the major reasons i wanted an external DAC/AMP combo with a mic input was to cleanup the microphone noise floor which the Hel does wonderfully, however i have to crank up the gain very high (about 85% or better) to get a good sound level and i expected more from a $200 dollar AMP. The Hel also takes a rather long time to boot after powering it on. If you keep it powered on this obviously isn't a problem, but i tend to turn it off when i'm not using it.

In the end, i'm not trying to sway people away from Schiit, but i would suggest using caution. The problem i had with their Fulla DAC/AMP had nothing to do with my machine since i tried to power it with 3 different devices, yet it never powered up. When i returned the Fulla i enclosed in the package a note stating the procedures i used to troubleshoot it but, as i expected, i never received a reply. If you decide to deal with Schiit Audio, good luck. I suspect you won't experience the problems i did but, then again, you never know.

Product Review: Corsair M65 RGB Elite Gaming Mouse (on Linux)

Rather than critiquing the gaming performance and digging deep into the specs of the Corsair M65 RGB Pro, the focus here will be to inform others as to how well i personally like this mouse and its usability and compatibility with a Linux based OS.

the TL:DR version

Unless you're willing to break out the soldering iron, stay the hell away from this mouse! Corsair is using horrible, poor quality micro-Omron 50m switches for both the primary and secondary buttons and they fail very quickly according to many people, including myself. I would suggest avoiding any mouse using the "50 million click" buttons, or at least make sure they aren't the same brand.

intro

My Cooler Master Inferno finally gave up the ghost after 11 years. It started clicking the primary button by itself and although i might have been able to resolve the issue, i've had it apart several times to clean crap from the wheel sensor, chipping the case in the process, and its feet were just about worn down to the nub. Oh, and all that nice rubberized coating was worn off and it didn't take long for that to happen either. Time for a new rodent.

Corsair M65 RGB Elite Gaming MouseThe Corsair M65 RGB Elite is a wired, right-handed, 7 button (plus the scroll wheel and its button) gaming mouse with solid specs, like a very accurate 18,000 DPI optical sensor which provides a max resolution that's an order of magnitude higher than any human could ever manage, but bigger numbers are more impressive apparently. With the KDE desktop and the mouse sensitivity set to 50% and acceleration enabled, the M65 is set at around 600 DPI for a 17 in. 1080p laptop display.

looks and feels

I think the M65 is a very nice looking mouse. I'm not entirely sure if it looks more like a weapon or a mouse with its somewhat prominent red 'sniper' button and vented aluminum frame, but i do like its angular yet smooth looks. The one piece top seems to be made of silicone and it feels velvety smooth. The plastic, concave sides have a hard, sandpaper-like texture which, as i suspected it would, is already wearing smooth after only several weeks of daily use.

The M65 feels good in the hand, but it's not as ergonomic as my Inferno was and because of the way the sides are shaped it's a little harder to lift off the pad, even with the rough textured sides and even though all the rubbery coating was worn off of my Inferno. I removed all the weights to make the mouse lighter and although this surely made a difference, it wasn't one i could readily feel. It also feels narrower between my thumb and pinky fingers, yet it's actually a little wider in that area compared to the Inferno. The M65 is a shorter mouse designed to be 'claw' or 'fingertip' gripped and this is certainly my preferred style of holding a mouse.

As implied, the M65 has three removable weights, each located in a different spot, however the way this feature was implemented wasn't well thought out. The weights are held in place with screws and replacing a screws after removing a weight is a little tricky because the hole in the weight guides the screw into it's threaded socket, though one could just leave the screws out entirely for more of a weight reduction. Had the engineers had their thinking caps on, they might have used these screws to hold the mouse together as well instead of adding two more which, to their credit, at least aren't located under the damned mouse feet, thus disassembly is very easy requiring only a Phillips driver. Although on the smaller side this mouse is quite heavy, even with the weights removed. It also has a very solid feel thanks to the rigidity of an aluminum frame and there's no detectable rattles when shaking it or squeaks when squeezing it.

Like many wired gaming mice, the M65 uses a braided cord which i find to be a gimmick. These cords are often rather stiff and unlike the older rubber/plastic/silicone sheathed cords, the slight noise that the nylon braiding might make as it drags across a rough textured mouse pad could be annoying to some, though this can be managed with some creativity.

the buttons

The primary and secondary buttons, both rated at 50 million clicks, are significantly quieter and lighter feeling than those of most mice and while i personally don't like loud buttons, these are so quiet that there's hardly any audible feedback and so sensitive that i couldn't rest my fingers on the buttons without clicking them. Whereas with the Inferno i knew i had clicked something because of the audio and tactile feedback, i now find myself looking at the display for confirmation. If we were to compare the feel of these buttons with the trigger of a gun, the words 'hair trigger' would certainly be appropriate. The throw of the buttons -- how far they move before a click is registered -- is much less than with the Inferno and i'm fine with that part of it.

Overall the click of the right and left buttons just doesn't feel good with the right button feeling slightly crisper the the left, possibly because its switch is positioned slightly differently than the one for the left button. Several times i had the mouse apart to see if anything could be done to adjust the pressure required to click and though that was easy enough to do with some strategically placed foam tape, nothing solved the problem of the very mushy feel of the buttons.

Corsair is using the Omron 50m switches for the primary and secondary buttons and though they are rated at 50 million clicks, they are absolute garbage and apparently quite a few gaming mice use these switches. Mine lasted a few weeks before developing what i might call 'switch chatter' and clicks not registering. Besides the laughably short lifespan of these switches, they are so sensitive that i couldn't rest my fingers on them and so mushy that you actually end up pressing the buttons much harder in order to make sure the click registered. These attributes actually make the mouse more fatiguing to use. I ended up replacing both the primary and secondary buttons with a pair of Huano Blue Shell Pink Dot switches.

The down-click of the wheel button is significantly stiffer and crisper than that of the primary and secondary buttons, leaving me with no complaints here. One thing i very much disliked with the Inferno was the barely noticeable notches of the wheel when scrolling and the M65 is slightly better in this regard, though still not what i prefer. The wheel is nearly completely silent which is nice. As with many mice, one can scroll the wheel a half notch and it will stay there instead of rotating to the nearest detente in the wheel. Unfortunately there is no side scroll function, though pushing the wheel to the right results in a middle click because the switch is located on the right side of the wheel shaft.

The M65 has three buttons on its left side and, thankfully, none are of the 50m type. The 'sniper' button requires more effort to press than any of the others and this is fine given that the thumb rests on it. The back and forward buttons narrow in height but OK.

The last two buttons behind the scroll wheel control sensitivity by default and they require more pressure to operate than is necessary, though i'm not complaining. Their location is good, making them candidates for binding to other functions for game use.

Missing from the M65 are the 'wing' buttons of the Inferno which were located on the outside edges of the primary and secondary buttons. In my opinion these were great buttons to have and there's no reason to not have them on all gaming mice since they don't interfere with any other operation as long as they're placed smartly.

configuration

The M65 will of course work as a standard 5 button mouse on Linux without any additional software, however when i first plugged it in, it defaulted to a 'breathing' mode where the LEDs slowly changed colors. I find this to be really annoying and Corsair's ultra-bloated, ultra-buggy and Windows-only iCUE crapware is needed to configure the RGB lighting, DPI and button assignments since they don't do Linux. I grabbed a Windows 8 image and installed it in a VM in order to be able to use their shitty software and the process was about as frustrating as it could be. Many times it refused to write to the internal memory of the mouse or froze the OS. Changing something as simple as the USB report rate caused all of the settings in that section of the GUI to become disabled. Sometimes re-plugging the mouse worked, other times a soft reset did the trick. The latter is accomplished by unplugging the mouse, holding down both the primary and secondary buttons and plugging the mouse back in while continuing to hold the buttons down for about 15 seconds or so. After a while the mouse seems to reboot and all settings should be at their defaults.

You can also access the firmware file directly which might of been helpful were i able to download different versions to try but Corsair doesn't make them available. To boot the mouse in this mode, unplug it and unscrew the upper-right weight retaining screw. In the hole where the weight goes is another smaller hole with a button in it that you can press with something small like a toothpick. If you press that button while plugging the USB connector back in, the mouse boots in something like USB storage mode and you can open a file manager and see the new device/folder inside of which is a firmware.bin file. Apparently you can replace it with another version and, i presume, reboot the mouse again by re-plugging it.

While there is a Linux configuration utility for the M65, the ckb-next software means having to run a daemon in the background since, at the time of this writing, they haven't yet reverse engineered the communication protocol that Corsair is using to talk to the device and so you can't store any settings on the mouses internal memory. Also i was warned about a stability issue when i installed it on Manjaro. Using the the utility, any button can be assigned to any other button, macro, or keyboard key, or it can be used to invoke a program and, unlike Corsair's iCUE crapware, you can also adjust the lift-off distance with ckb-next. Honestly it would be a great utility were it able to write to the mouses internal storage.

If you don't want to use any 3rd party software to configure the mouse, there is yet another potential option for Linux by using Xorg configuration files, though there appears to be limitations.

conclusion

I'm quite sure some people will not like the way the primary and secondary buttons work since the throw is very short, they require very little pressure to operate and they are very quiet. If you're wanting a satisfying tactile feel when you actuate these buttons, look elsewhere. On the other hand, these characteristics may be a plus for some, especially if working or gaming in the vicinity of those trying to sleep, however the atrociously short lifespan of the Omron 50m micro-switches used for the primary and secondary buttons makes buying this mouse a fantastically poor decision in my opinion. Stay away from this mouse (and Corsair products in general as far as i'm concerned), as well as any mouse/device that is using the Omron 50 million click switches.

resources