The problem with gaming mice

It’s been nearly 4 years since the original Razer Deathadder was released. After some initial issues, a few firmware updates made it a nearly flawless mouse. Since then, Razer, Logitech and other manufacturers have released many new mice, none of which I’d recommend over the trusty old Deathadder.
Here’s my list of the mice I’ve tried and what’s wrong with them:
  • Logitech MX518, 125Hz polling rate by default, prediction, otherwise flawless afaik. Both the old 1600DPI, and the new 1800DPI.
  • Razer Boomslang CE07, rubber buttons wear very quickly.
  • Razer Lachesis, z-axis tracking
  • Razer Mamba, z-axis tracking, jitter on old firmware, dynamic dpi scaling on new firmware (to fix jitter :/ )
  • Razer Abyssus, jitter on cloth pads
  • Razer Deathadder 3500, jitter on cloth pads
  • Razer Imperator, z-axis tracking, dynamic dpi scaling (hardware acceleration :/ )
  • Logitech G500, hardware acceleration due to Avago ADNS-9500 sensor
  • Logitech G9x, hardware acceleration due to Avago ADNS-9500 sensor
  • Steelseries Xai, hardware acceleration due to Avago ADNS-9500 sensor
Apparently, releasing a new mouse with OVER 9000 DPI AND BUTTONS is more important than flawless tracking to all these mice companies.

86 thoughts on “The problem with gaming mice

  1. Hi. Thank you for this little summary. Now it looks to me like there’s no way I can get a good mouse like my old Diamondback 1600, which unfortunately died last month after almost five years of extensive use(r.i.p.). So I decided to buy the new Imperator, but i’ve sucked in every game I used to pwn with my old mouse since i got it. Today, thanks to your pages, i found out why is my aim so messed up. The “for gamers by gamers”(bah!) company thought that some mouse acceleration would be useful. Which is f***ing not! I’ve never encountered mouse acceleration before with mice like Intellimouse Explorer and Diamondback 1600, so it kinda surprised me.
    So i threw my Imperator out of the window and am looking for something to play with. So is there a !currently available! mouse without all these mouse acceleration, Z-tracking and jittering problems? I use the Destructor mousepad.
    Thank you for your advice, fox.

  2. The 1800DPI Deathadder should still be around. The MX518 is great if you can live with drift-control/prediction/angle-snapping after boosting the USB rate to 500Hz.

  3. I’ve been looking for this OLD Deathadder in my country (Denmark) and it has been replaced by the NEW Deathadder by all the online retailers – it’s not really obtainable any more.

    The MX518 may be the only viable option left – which is really strange, since it’s such an old mouse (and I’ve personally worn out 2). This is why I’m kind of interested to know how this firmware update to the Razer Imperator worked out. People were commenting on this:,76

    So, is it possible to get a comment or an update from some one with an updated Imperator?

    As the above poster Fox probably helps to illustrate, it’s not easy to actually get a hold of a mouse that works.

  4. I have the Imperator with upgraded firmware, it still sucks.

    Main problems:
    It still has dynamic DPI scaling to fix jitter. While this fixes the jitter on some pads, it also introduces acceleration.
    Pick up/put down, still causes to cursor to shift down-right.

    I wouldn’t recommend the Imperator to anyone.

  5. Thanks for replying to my comment… and so fast!

    Just to get this straight, with your knowledge and ruling out the old Deathadder, then there is no superb gaming mouse available at the moment?

  6. I’m afraid so.

    I’m ‘stuck’ with my 1800DPI Deathadder because I like to play on cloth pads. If you play on a plastic pad, the 3500DPI DA shouldn’t have any issues.

  7. Okay! Well, be sure to make a big headline when you find a successor to this beast of a mouse that is the old Deathadder 🙂

    I’m stuck with the Abyssus for the time being. Not too unhappy with it beside the cheap feel and the lack of thumb-buttons. It’s currently living on top of a steelseries cloth pad and is not really jittering noticeably. Am I missing something here about jittering?

    Thanks for entertaining me with your interesting blog and great responses 🙂

  8. It doesn’t have jitter on all cloth pads, but it does on my Razer Goliathus and a few no-name pads I’ve tried.

    To test for jitter, simply open MS Paint, take a small drawing tool, press and hold mouse button 1 while you slowly paint from top to bottom and left right.
    It’s most notable on 1800DPI and higher.

  9. Well said! All these DPI gimmicks are really more for layman than “real” gamers… I am still looking for a new gaming mouse to replace my trusty old WMO. (which i was luckily enough to stock up 4 of them) But after 5-6 years of searching, buying, nothing same close to replacing it.

    Interested to try the new Zowie EC2 though (
    ) But having problems getting it in my country. It looks promising and i prefer WMO size mouse (IE3.0, Deathadder, MX5xx are slightly too big for me).

    If you happen to come across a EC2, please share your thoughts. Your reviews are probably the only mouse reviews that is worth reading on the Internet today.

    Thanks again for sharing.

  10. (Don’t wanna smart-ass but to not confuse anybody: “Lycosa” = “Lachesis” in Arie’s original post.)

    I am impressed with the number of mice you have tried already. I am surprised to see only Razer and Logitech mice in there (apart from the Xai). Of course, nobody has the money to try all, but nowadays there seem to be awesome mice from other companies like Roccat, Cooler Master, Mionix, Ozone!!

    Have a look at:
    Though their reviews don’t always seem (technically) detailed enough for me, they have a very broad selection of manufacturers to get an overview what’s available (I am in no way personally affiliated with them, it is simply what I think of them).

    I haven’t tried any of these myself yet and am somewhat hesitant to order four mice and send three (worst case four) of them back again…

    The thing is also that the shape and physical feel of a mouse have utmost importance for me. So I am unsure if I will ever find replacement for my current mouse.
    It is an embarrassing “Trust MI-6950R”. It might be a cheap one and not even intended for gamers but technically it performs solid in all areas AND to me it is just a perfect match shape-wise – like it had just been designed for my very hand 😀 (38mm high, a swift 84g, on-the-spot thumb buttons, mamba-style index finger buttons, running OCed @500Hz)

  11. Thanks for pointing out the Lycosa/Lachesis mistake, I’ve corrected it.

    The problem with sites like xsreviews is that they don’t spot the obvious flaws these mice have for gamers. Their review of the Abyssus doesn’t point out the vertical tracking issues on cloth pads. The review of the Imperator doesn’t mention the very obvious z-axis tracking issues.

    Besides the Logitech, Razer and Steelseries mice, I’ve also tested the Saitek R.A.T. 7. It uses the same horrible sensor as the Razer Imperator, only with worse firmware.

    If you don’t have any issues with your current Trust mouse, why are you looking to replace it?

    The reason I try so many mice is because I somehow like it. But it’s a bit disappointed to end up with the same mouse after 3 years of trying different ones (DA 1800DPI). I also manage to test quite a few mice at LANs every now and then. It’s a great opportunity to test drive lots of different mice and to ask questions to actual gamers that use them every day. Much better IMO than depending on some reviewer who tries to please a company or doesn’t know what to look for.

  12. for z-axis tracking, is this a problem if you don’t lift your mouse because you can reach everywhere on the screen without doing so?

  13. I’ve finally had to replace my trusty Copperhead due to the left button wearing out internally

    As a left hander the choice of similar replacements was very limited.

    The Lachesis I ended up with suffers from z axis tracking and not just when lifted from the surface it is on. Even the weight of my palm on it tracks the cursor left/up and it returns when I lift my hand.

    When I press down on the top of the mouse the tracking is even more pronounced. Soft mats are unusable, including the Goliathus that was bundled with it. The problem is reduced on a hard surface but its still noticeable.

    So far I’ve found the best way to minimise z tracking and maintain reasonable accuracy is to turn down the sensitivity to 5 and use the DPI scale to increase movement rate.

  14. I just tried the imperitor before i read this review. I too was haveing the crazy Z-axis issues. I was using the MX518 prior. Is there any other mouse out there that we can use with the number of buttons of the MX518? I already gave up one going to the Imperitor but justified it… now seeing all these terrible short commings of the new mice.. i feel its better to just stock up on MX518s.

  15. I want a 9,000 dpi mouse, and would probably be more satisfied with a 12,000 dpi mouse. 15,000 dpi might be a little much.

    Using my current 5,600 dpi mouse, playing minecraft, when I set the turning sensitivity all the way down for maximum, per pixel aiming precision, it doesn’t turn as fast as I would like. My hand is still capable of much more precise aiming than a 5,600 dpi mouse allows.
    I figure doubling the dpi from what we have now should allow per pixel aiming with reasonable speed.
    Of course, once I have a 5,760×3,600 monitor (which should finally match my ability to see), then a 20,000 dpi mouse might be necessary.

    I appreciate the info on hardware acceleration. Anything that meddles with my input is very irritating.

  16. Have you any updates on this? Tested any new mice, any new mice on the way? I just bought the imperator and I’m really feeling the dynamic scaling bullcrap. The z-axis mapping isn’t a huge problem for me, but I do see it.

  17. Nope. Razer still uses the crappy Philips sensor of the Imperator in other mice, Logitech is still on the Agilent with acceleration issues.

  18. Hello Arie,

    I’m stuck with a similar conclusion as you’ve stated on that post. Right now I have those mice with those problems :
    · MSIE 1.1 : No problems apart that it’s impossible for me to get it to 1000Hz on Windows 7 x64 – which means negative accel, so unusuable.
    · WMO 3.0 : See above + wheel randomly going nuts.
    · Mx518 : 1600dpi version. I should test if I can get it to 500Hz via drivers, else same problem as above + Angle Snapping being a pain.
    · Lachesis : z-axis (not a big problem) + some axis lockups + when mouse is slightly tilted it locks too.
    · Mamba : z-axis + what I call ‘lazy aiming’. Sometimes I do the same movement twice, and it ends up with slightly different outcomes. I might try to install an older firmware update since jittering ain’t a problem (using SteelSeries SX mainly). Build quality, though, is best from all the other mice.
    · G500 : Some acceleration – not what I expect when I’m playing competitively + too stiff side buttons + weak wheel (tilting mechanism was broken in a couple of days, still wondering how).
    · DeathAdder : 1800dpi version. First batches. Wheel is driving me nuts sometimes (not acceptable for competitive play), it started to get hard locks sometimes – could it be the mouse cord getting detached from the mouse as old DiamondBacks had?

    So I’m still for the quest of a good, solid mouse to play with.

    I’m considering the Xai (even with the acceleration) since it sounds like it has only one problem that might affect me.

    However I have two questions :
    – Using hard mousemats, is the 3500 DA as accurate / great tracking as the 1800 DA? Does it have the ‘fixed’ mousewheel?

    – Have you heard of any case where a 1800 DA would lock for a second during gameplay? Could it be the cord detaching itself from the mouse’s circuits?

    Thanks in advance.

  19. -With a black non-cloth mousepad the 3500 dpi will be fine. I never had problems with the scrollwheel on either of them.
    -No, but if you hear the usb unplug/plug sound at the same time it’s very likely the usb cord. The 1800 dpi has a thin, vulnerable cord, much like the diamondback had.

    Have you seen the zowie mice? Angle snapping is their only fault afaik, haven’t used them myself yet though.

  20. – I might pick one of those 3500 DA then, after checking with the shop if I can return it. Good thing that the mouse wheel works good. I hope it tracks as good or close to the old DA.

    My 1800 DA is very old, it was one of the first batches and was one of the first batches to come into my country. Still, it keeps working like a champ, but playing competitive at pretty high level, I can’t tolerate the ‘new’ behavior that the mousewheel has taken lately. More often than not I find myself heading towards a pack of enemies with my pistol :/

    – It’s weird, because it’s as if the whole computer would lock for a second, but game would still running. Can’t do anything with either mouse or keyboard. I’ll scan for spyware again, but it happens with DA only. Could be one of the DATA USB cables would be detached from the mouse internals and the CPU/IRQ would go nuts until it recieves info. I’ll try opening it once the 3500 DA gets here, see if I can fix it.

    I saw the ZOWIE EC mice, but angle snapping is a big turn off for me right now. Even if I play with low-ish DPI (400~800) and low sensitivity (~25cm / 180º), I tend to make smooth upwards and downwards movements, and that would be probably killed by angle snapping.

    Sometimes the difference between winning a round for your team and losing it is just that couple of pixels that separate head from torso. I’ve been there with the Mx518 and kind of left it on the shelf for that reason. Now I’m playing on a higher level, so it would be even more of a problem, even if I absolutely loved the Mx518 back then.

    However, I might pick the Mx518 again and see if I can notice it as greatly as I remember. If I can’t, they’re a good option. If I still notice it as intrusive as I think, I’m heading for a 3500 DA (my current mousemats are : Destructor, SS SX & Exactmat, and it should work on all of them).

    Thank you!

  21. And one more thing, saw some rumors that the Avago ADNS-3090 was the 3888 (DA 3.5g) for the open market? Doesn’t there exist a ADNS-3668 version for the open market as well?

  22. DeMS: Be sure to keep us informed with the behavior of DA3500 + Destructor combo. I already have the mousepad, and I am considering buying the DA. 😉

    I’ve had much work in the RL lately, having played only some indie games every few days, but now I can start playing again. And i’m still stuck with that f…..g Imperator. 🙂

    Thank you, fox.

  23. Does the dynamic dpi persist when used on pads that would ideally not give jittering? maybe it just downscales dpi when jitter is detected?

  24. What good plastic pads would allow one to use the da 3500 and still feel like a good pad?

  25. Hey again,

    about my Trust mouse:
    First thing is, I figured out it only runs at 800CPI – the rest is interpolated (A7550 sensor). However, I don’t mind – 800 is enough imo.
    The other thing is that I used to play high-sens. Moving more to the low-sens end I realized the malfunctioning speed is considerably low.
    Other than that, it is probably pure curiosity and the thought that there could be something better that gets me looking for an alternative 😉

    Right now I am on a Trust GXT14(s). It’s optical but I don’t know the exact sensor. It is 800CPI (+interpolation) as well but has good performance otherwise (relatively low lift-off, no angle snapping, no jitter or similar).

    Promising might be the CM Storm Spawn. They finally ditched the TwinEye and put something in there that looks very much like the new Avago ADNS-3090. Screenshots of the driver UI imply that the dreaded angle snapping can be disabled! Other than that, I like the claw-grip style which DA lovers probably won’t. Then again with 142g the weight is fairly DAy but might be too much for me :-/

  26. Maybe we should make a page on wikipedia called “Gaming Mouse Comparisons” or something, so everyone could add their mouse and their experiences. Kinda objectively state, Brand, Product, Sensor tech, Features, Problems and such. That would be great, could quickly be updated with new mice as well.

  27. There is also a pretty informative thread on named “OCN Mouse Sensor Reference and Performance sheet”

  28. Looks like wikipedia refused that article for a number of reasons… however a register like that would still be kinda nice to have, and to have on a wiki somewhere so more people could contribute their experiences. Also, a pure sensor guide is great, but sometimes certain products add/remove features which make just go sensorhuntig problematic. Like prediction and stuff. That’s why I like per product.

  29. that thread seems very nice, it should include include a column for jitter issues, hardware acceleration issues and other problems though. according to that thread all mice has perfect tracking up to some speed, and some have prediction…

  30. I won’t pretend to be an expert on gaming mice, having only been introduced to them a year ago, but doesn’t the Xai have ways of shutting off acceleration? I’m using an oversized Goliathus from Razer with the Xai and switch across the whole range of sensitivities from 450 to 5,000, depending on what task I’m doing ingame. Oh, and the ingame sens is 1.0, always. I was using an Abyssus before this one. And also, if the Xai’s tracking issue is a real serious problem, can I correct it by purchasing a hard mousemat like the 9HD/4HD? I know that hard mousemats have severe longevity issues. I don’t know how the old DA runs since I’ve only used the Abyssus and Xai. Kind of a noob here.

    “Some games and applications
    have you working on a huge
    2D mousing environment with
    resolution of 1900×1200 (or
    even higher), not to mention
    multi-screen setups. In these
    specific cases you can
    either configure the
    SteelSeries Xai to operate
    with very high CPI values or
    you can utilize hardware
    acceleration. With
    acceleration you can bridge
    large distances quickly,
    while still being able to
    work flawlessly in small
    areas of the screen, without
    having to switch your CPI
    with the CPI toggle all the
    time. The main difference
    between this setting and
    Windows acceleration is that
    this setting is running in
    the hardware. Not only are
    you able to turn it on and
    off, you are also able to
    set the degree of dynamic
    acceleration that is applied.

    Recommended use: Most
    gamers will probably find
    that a setting in the lower
    region should be more than
    enough, but if you’re a real
    speed freak you can turn it
    up high. For FPS gamers and
    games with a 3D mousing
    environment we recommend
    not using acceleration.” -Excerpt from the usage guide on the Xai’s software configurator.

    ExactAccel can be turned on or off and then set to 0-100. So is it fully disabled when set to 0 or shut off, or do we still have the positive acceleration issue? I hope not, since this feature is one of the main reasons I got the mouse to begin with, although FreeMove is pretty important, too.

  31. The Xai has positive acceleration caused by the sensor used. The Logitech G9x/G500 has the same sensor/problem.

  32. Sorry for the double post, but I wanted to add that it would seem to me that if the ADNS-9500 was designed to use hardware acceleration, then wouldn’t the right control software be able to shut it off? That’s why I thought that maybe ExactAccel does have the ability to shut it off. Not unlike, say, disabling a core in a CPU.

  33. Never mind, I’ve more or less confirmed that there is positive acceleration. Well, it does seem cut it down, but that’s still not zero percent. The tester was using the latest firmware, too. This article has the info: He’s not entirely sure, but seems to think that it’s the case. That’s really too bad. I’ve got no other complaints about this thing. Also, WindspEEd? I consider myself a “real” gamer and I do make use of high sensitivities, especially in close-quarters FPS situations. High DPI isn’t just a gimmick. Just ask the high-sens snipers/living aimbots on FEAR 2.

  34. If you wanna know to what extend high cpi (“dpi”) is a gimmick, have a look at

    How much cpi you need, depends on your screen resolution and what real sensitivity you want in FPS-games (X cm/360º).

    For example:
    With 800cpi @ 1920×1080 (->fov:106) you can get as high sense as >14.4cm/360º (->ingame:28.9cm/360º (that’s kinda “low sens”)
    1200cpi: real sense >9.6cm/360º (pretty “high sens”)

    With less screen resolution, you need less cpi from your mouse.
    Of course, if you want insanely high mouse speeds on your dektop, higher cpi might be usefull but not really for FPS-games.

  35. Something got messed up… reposting:

    If you wanna know to what extend high cpi (“dpi”) is a gimmick, have a look at

    How much cpi you need, depends on your screen resolution and what real sensitivity you want in FPS-games (X cm/360º).

    For example:
    With 800cpi @ 1920×1080 (->fov:106) you can get as high sense as >14.4cm/360º (->ingame:28.9cm/360º (that’s kinda “low sens”)

    1200cpi: real sense >9.6cm/360º (pretty “high sens”)

    With less screen resolution, you need less cpi from your mouse.
    Of course, if you want insanely high mouse speeds on your dektop, higher cpi might be usefull but not really for FPS-games.

  36. Is it reading the “>””fov: 106) you can get as high sense as >14.4cm/360º (->ingame: 28.9cm/360º (that’s kinda “low sens”)

    1200cpi: real sense >9.6cm/360º (pretty “high sens”)

    With less screen resolution, you need less cpi from your mouse.
    Of course, if you want insanely high mouse speeds on your dektop, higher cpi might be usefull but not really for FPS-games.

  37. If you wanna know to what extend high cpi (“dpi”) is a gimmick, have a look at

    How much cpi you need, depends on your screen resolution and what real sensitivity you want in FPS-games (X cm/360º).

    For example:
    With 800cpi @ 1920×1080 (->fov: 106) you can get as high sense as >14.4cm/360º (->ingame: 28.9cm/360º (that’s kinda “low sens”)

    1200cpi: real sense >9.6cm/360º (pretty “high sens”)

    With less screen resolution, you need less cpi from your mouse.
    Of course, if you want insanely high mouse speeds on your dektop, higher cpi might be usefull but not really for FPS-games.

  38. This is ridiculous, it just won’t post the way I write it…
    Someone go ahead and clean it up, I didn’t do anything wrong…

  39. High DPI is nice for fast movement on a large desktop without having to use acceleration or skipping pixels.
    In-game it has the advantage of having the same cm/360 but with smaller discrete screen movements. In-game resolution and mouse sensitivity shouldn’t be related to eachother. It should take the same amount of centimeters to make a 360 degree turn on 640×480 as it would at 1920×1080.

    So the highest sensible DPI is the smallest amount you can move your hand 🙂

  40. Hi Arie,

    May I ask what your impressions are on the Tt eSports Theron and Saphira?


    P.S. Is there a way to get e-mail notification for your reply?