I got my G100s early last week. Now that I've had some time with it, here are my thoughts.
Sensor: Avago AM010 N1301T, Optical* see Sensor Performance
Main Switches: Omron D2FC-F-7N (20m)
Wheel Encoder and Center Button: Kaihua Electronics Co. Kailh MI1267 (3m cycles rated)
*Native CPI steps: 250, 500, 750, 1000, 1250, 1500, 1750, 2000, 2250, 2500
Weight : 82.5g (-13g without weight)
Polling Rate: 125Hz or 500Hz (500 default)
Dimensions: 117x63x37mm (lwh)
Mouse Feet: Teflon
Ergonomics and Build Quality
Before I talk about the G100s, it needs to be said that I am coming from a mouse that comparatively speaking with the more current offerings in 'gaming mice', is decidedly simplistic. The iktec brand of mice are a pretty niche market as they are, or more precisely were, a Korean manufacturer that is now defunct. They made cheap but functional devices and were used by many in the pro Brood War scene. Compared with mice today, they lack the carefully nuanced ergonomics, great operational or tactile feel, or even a sensor that performs optimally. Even with all of these things seemingly against it, I fell in love with this mouse ergonomically. The reverse tapered design of the overall body (thin to wide from back to front) and the non-tapered sides while simple, has seriously made me question the incredibly unique and developed look and feel of certain mice, and I wonder if these shell designs aren't more of a subtle design choice in branding, rather than for actual ergonomic considerations, but that's a slight tangent.
With that in mind I was tentatively excited to get my hands on a mouse with good performance and in what many gamers consider to be a 'proven' shape. It's been around ever since the MX series was introduced in the early 2000's, and as you probably know, people still fight on ebay to get authentic G1's even today. Why? Well after a week I kind of know why.
The size and shape of the mouse is good. It feels natural in my hands, and for my sort of hybrid fingertip/claw style (more fingertip) it is just about the perfect dimensions. I should mention that I have been missing the ergonomics of my 9500+ and find the front and inward tapered design (top-down and back-front) to be the only thing that I am not ecstatic about, but generally speaking this mouse is almost the perfect shape for me. In the time that I've used it my grip has slightly modified itself to cope with these slight differences (right ring finger migrated upwards a little to help with stability and control that I felt wasn't as solid). Only time will tell if it feels more natural in my hand, but I can say that generally speaking that there's a good reason why many gamers love this shape as it's felt pretty good so far.
The main switches are omron switches, while the wheel encoder and cpi button are manufactured by Kaihua Electronics. In terms of their difference the omrons and kailh feel somewhat similar in their operation, though I have a little more of a tactile responsiveness with the main switches that I lack somewhat from sw3+4. I should make two notes about the switches here. First, the middle button for the wheel has a different feel than the cpi button behind it, and does provide a feedback more akin to the omrons. The second is something that was my first of only really two important gripes with the mouse.
The right click had a distinctively soggy or unresponsiveness about it, and in checking this out I was able to deduce that the travel distance from neutral to actuation for the main switches was longer for the right than the left (the button would move, but not activate until a certain point). This ended up with an unevenness in how each button felt. In SC2 this resulted in missing move commands occasionally which was a bother. There are however, two solutions to this problem. The first is something that Skylit let us know about. Having purchased the G100, he swapped out the top plate of the shells of each and this solved this issue for him entirely. Additionally I added ~4 layers of painters tape onto the shell to close the travel distance (just enough to cover the contact point), and this solved my initial gripe with this mouse. Now whether this is a design flaw that will eventually be fixed (quietly) or has been already I don't know, but as for now it may be an issue for some, though easily correctable (you will have to void your warranty to fix this yourself).
I should also note that my main switches are a little squeaky from time to time (moreso the right sw), but I can live with this if it doesn't impact the operation of the mouse
The cord is a tad thinner than others that I've had and initially was kind of stiff, but after some use it's relaxed enough that I don't even notice it anymore (I do not use a bungie as I cannot due to my mousing style in fps games).
Finally, there's a 13g weight inside the shell that can be removed (also voids your warranty). I immediately did this and am glad I did so as I tend to like very light mice.
Something that I need to quickly mention about the native CPI steps in the specs is that this information comes directly from CPate who works for Logitech, so we can only take him at his word that they are in fact all native. Personally I'd rather see what the native steps are for myself, but again, as this is a custom architecture it is not likely that we will ever see the datasheet (it's not publicly available).
Anyway, to start off the LOD of the mouse is between 2.4-3.6mm, though this fluctuated based on the surface. My Qck was 1-2CDs, Hayate was 2-3, and my Propus 380 still tracked at 3CD's, though barely.
Tracking wise there is no angle snapping, but I did want to mention one odd behavior that I noticed from my tests in paint. In moving my cursor quickly you'll note that there is an odd angularity in some of the lines below that makes me think that something else is 'helping' with the tracking (perhaps necessary? or another unlisted feature? idk). In practice, partly because most of my twitch movements are horizontal lines or very shallow curves I haven't really noticed it, and without the datasheet or other information I'm kind of blind in this area, but it was somewhat unexpected. And no, my hand does not naturally do this naturally. I would generally say that this is a minor thing that hasn't really affected me, but I do need to make note of it.
+ Show Spoiler [odd mspaint behavior] +
+ Show Spoiler [mspaint tracking tests, qck examples] +
I also noticed that there seemed to be a small amount of ripple or jitter in the mspaint tests. This happened at only at low speeds, and was additionally more pronounced at 500Hz than at 125. Between my three mousepads none seemed to really negate this behavior entirely, but it was slightly more pronounced on all of them beyond the 1750cpi range. Overall this is another relatively minor quip as once you past the 0.06m/s speed mark this behavior is far less noticeable and tracking is much smoother. Aside from these two things the tracking is for the most part predictable. I should touch on this 'for the most part' thing.
When it was announced, we were told the G100s could be expected to have an ips of about 120 or at around 3m/s. This made many of us excited as a good number of gamers have been waiting for a smaller form factor mouse that could perform reliably and at higher tracking speeds. In testing this out I found that surprisingly that enotus was recording a pretty similar range fluctuating just around the 2.7-2.9m/s range or 106-114ips. The most interesting part of this is that higher registers traditionally tend to drop off in their ips as you push the sensor to its maximum capabilities, but in this sensor it made very little difference whether the cpi step was 500, 1000, 1500 or 2000cpi. I should note that one consistent thing should be said in that the Propus 380 and Hayate tended to register just slightly higher ips than the qck at all ranges (though slightly more so at higher cpi steps). This seems to paint a relatively positive outlook for this mouse and its capabilities, but not satisfied with enotus being an authority (because it's a somewhat unreliable program) I had to test this out in an actual environment. This is where I ran into my other major quip about this mouse.
Being an low or ultra-low sensitivity gamer in fps games I need a mouse that can track reliably past the 3m/s range because I often will make sharp and very quick movements in these games (flick shots, fast rotational turns, etc...). Being that my rotation per 360° is close to 90cm (35in) I often need to get from point A to point B very quickly, but when I would do this instead I would arrive at point C, with a continual shortening and dropping of my view until I was looking at my feet if continued (see fig below).
+ Show Spoiler [tracking problem] +
Generally this started to happen at the 2.6m/s range or around 102ips and would increasingly get worse the further you went beyond this range. This paints a rather different picture, in that what enotus shows us is not exactly indicative of what the user may actually experience during play. So while the mouse can technically track at around 118ips, the perfect control speed is more likely around 100ips (2.55m/s).
In comparing this performance with its previous counterparts this is an improvement over the G100, and at least in terms of the perfect control speeds, the MX300 and G1. However, both the G1 and MX300 are still able to track at a higher ips without completely malfunctioning than the 100s, so certain individuals may be more hesitant to make the switch if they are still holding onto their older models (though this group is admittedly a smaller and more niche sector than most).
The new drivers are pretty easy to navigate and have some very interesting options about them. One thing to note is that the default CPI steps are 1000,1750, and 2500, and the default polling rate is 500Hz if you don't want to use the drivers. Inside tem, you can choose to have up to 5 different CPI levels (dammit Logitech, why do you still use dpi...), and even have the option to have separate values for the x and y axis of the sensor. Additionally there are some other rather interesting options available, such as assigning macros to buttons, assigning text blocks, shortcuts, ventrilo options, and a hilariously unnecessary but kind of neat scripting option where you can do some really cool things.
+ Show Spoiler [driver panel] +
It has been far too long since Logitech returned to this form factor, and at least in a general sense they have done a pretty decent job with their international rebooting of the G100. So after looking at it and its operation, who would this mouse be good for? As I am both a FPS and RTS gamer, I feel that I should try to tackle this most pertinent question on peoples minds regarding each environment specifically.
The average RTS player will likely find this an appealing and overall solid option to them as long as you don't have the r_click issue that I (and others) had or are willing to void your warranty to fix this problem (again could be an initial batch problem that has already been fixed or is being fixed, but idk, only time will tell). In terms of the tracking items I spoke of, most RTS players will never encounter these issues because of how they control their mice so they are really non issues for this group of people.
For FPS players this mouse is somewhat more dependent on your style. If you are a mid sensitivity player, you might find that the mouse will perform well enough for you depending on your overall rotation/360°. I found that with around 750cpi and 1 sens (approx 55cm/360° with a 0.022 yaw, or ~22in) you would only really encounter the malfunctioning problems near to, or at your top speeds. High sens players will of course not have any issues, though for low to very-low sens players the relatively consistent wandering of the sensor past 100ips or so (in varying degrees of severity after this point) will likely be an annoyance that you may not be able to live with depending on how you control your mouse and what your distance/360° is. Again though, kind of a niche group of people so keep that in mind.
Mouse comparisons with KTEC 9500+ and Azurues mini, documentation of internals, driver panels, and mspaint tests.