Some people believe that image scaling using a monitors scaler is faster/better than GPU scaling. Depending on the GPU and/or monitor, this could be true. The opposite could be true too. Maybe neither is noticeable. If you have the choice, try both and see which one you prefer, if any.

Mine had a squeak but it went away after a while. It didn't really bother me personally, especially since I wear headphones at the computer most of the time.

I will buy that PCB

especially if it has a 3310 :-)

50ms delay between presses on a single switch is fine. The problem is that two totally different switches cannot actuate within 50ms of each other.

Thanks for your time in acknowledging this issue

This is great. If Valve didn't want people to remove the scope they would have not made this possible.

Thanks for your results Forsak3n
Can you tell me more about your beta version, like approx date you received it, anything else notable? I assume its still a 3090 sensor with the Holtek HT82A525R MCU?

Good luck finding anything as small as the asian salmosa with a good sensor. They are still available on ebay. I don't know if people who are recommending g100/fk really know how small this mouse is.

http://i.imgur.com/zCJV9KK.jpg

You could also put the spare set of feet on the existing set, which I did after I noticed the excessive drag.

Hey while I'm here, can I get some keen folks to test something for me?

I've been investigating this Aurora button issue, and why it is so much harder to sync the buttons for rocket jumping. I tested the button latency against the G100S, and it is only about 3ms slower. The interesting part is when you test the mice on their own in the Bloody Mouse Shooting Speed Test.

When I test the G100S on its own, the difference between left and right click is anywhere from 0-10ms generally. I accept that I am human and cannot press both buttons at the same time all the time in my natural grip.

However, when I test the Aurora on its own, I either get a value of 0ms or >53.9ms. I find it strange that I cannot get a click timing offset less than this value. It appears as though there is some kind of firmware issue causing delayed clicks when they are not synced within a very tight margin.

http://drok-radnik.com/junk/aurora_buttondelay.png

Would be nice if somebody else could verify these results, as it might just be an issue on some units (or just mine)

As promised, here's a gallery of the Aurora PCB in WMO shell mod.

http://drok-radnik.com/junk/wmora_mod/

Started by hacking away with a blade and bending the plastic off the bottom shell. Refined bottom shell with rotary tool and built wheel cradle out of old DVD case spine (pro tip from bst). Had to re-open and adjust the front PCB height a few times to get the best click response. The middle click is still a bit stiff, but I'm giving up for now until my potentially counterfeit ones arrive, which will probably solve that issue.

The top shell was done just with a blade and looks like shit. Next time I will have a bit more of a clue as to how much to remove.

The oldest demo I have handy is from December 2011, and it worked fine.

Thanks lol

I'll try to finish this normal WMO one first. I've got a bunch of pictures already. It will probably be 2-3 weeks before these potentially fake WMOs get to my door.

Quick update: I managed to lessen the disparity between mouse1 and mouse2 actuation force (mouse2 was a lot lighter than mouse1) by gently and incrementally bending the mouse2 button backward until the clicks were audibly synchronised. Bending it backward is easy, but if you go too far, you will have to open up the mouse and unhook the button assembly further back to bend it forward again. If you go too far forward, the shell will constantly push on the switches. There is a sweet spot and it takes time and patience. It is probably possible to snap the shell if you try this. If I didn't have a 2nd one, I would probably not have done this. All of these things will void your warranty, I accept no responsibility, etc..

Edit: I also attemped the WMO shell swap with only a stanley knife. It's mostly done but I will need to pick up a rotary tool to remove more debris on the bottom shell, as the distance from sensor to pad is too far, and loses tracking quite often. You will need to build your own wheel cradle, I used an old credit card. You will also need to remove a lot of plastic from the top shell buttons to make the larger wheel fit. I have also ordered what I believe to be counterfeit WMOs, which are apparently way easier to mod, and can use the original scroll wheel. When they arrive I'll attempt using one of those.

I just got mine from the original pre-order today. Thought I'd be waiting well into 2015. Here's my impressions after 30mins.

I've only opened one of them so far, seems like a decent sample. I am having trouble rocket jumping with mouse2 jump, as other people have noticed. Jump triggers late half the time, and it happens less frequently when i turn right to RJ, so I assume it can be solved by adjusting grip and/or getting used to the force required.

Flicks feel good, tracking feels good. I am accustomed to small/lightweight mice, so there isn't much adjustment period for me. I have been using a Kinzuadder primarily over the last week or two, and this is significantly more comfortable due to the narrower rear putting less pressure on my palm.

Default CPI steps are 800, 1600 and 4000, cycled with the button below scroll wheel. Polling rate is 500Hz out of the box. I wanted to turn off the LEDs, set 800 CPI only and set 1000Hz polling rate, so I had to install the software. After making my adjustments, it crashed when I hit Apply, and the mouse froze. Settings were still saved when I re-plugged mouse, and I have since uninstalled the software. I hear that I am not alone in this.

Scroll wheel is nice and smooth, yet has that satisfying mechanical whirring. If you need a scroll wheel with defined single steps, this is going to give you nightmares. Middle button is decent to click, slightly stiffer than m1/m2, which is about as good as you can get on most mice. Since the wheel is quite slippery, I have caused the mouse to scroll accidentally while middle clicking. May be a problem in TF2 where I use scroll wheel up/down to select slot2/slot1.

Lift-off distance is low. Not the lowest I've seen (eg: G502, Zowie), but low enough for it to not be a problem. I am a big fan of low LOD and this is acceptable.

Glide is fine, despite feet being a little thinner than I generally prefer. I don't think I would personally want a faster glide, as more friction tends to help stabilise light mice for me.

Cable is okay. For me it's similar to G302 and WMO level of thick/heavy. The texture of it is largely irrelevant for me, because it doesn't touch my pad at all. The cable is stiff enough that I can actually get it to suspend itself for about 25cm distance from the mouse to my desk in front of the pad.

Polling rate at 1000Hz is very consistent (largely within 0.5%), BUT every 15th (why??) update is 2ms/500Hz. I doubt this has any serious impact in general gaming usage.

Regarding side buttons, I don't really use them. The front one (mouse5?) is great, the rear one (mouse4?) is a little more difficult to get to with my grip. The only game I currently use a side button for is TF2 (slot3). If you need both buttons often, you may find you need to adjust grip to suit. The thing about squeezing the shell to activate side buttons never comes up for me in actual games. You REALLY have to squeeze it quite deliberately and hard. HARD.

Also the lights have that PWM thing going on, where you see stroboscopic artifacts while moving the mouse. A lot of mice do this. I turned all the lights off.

Anyway, main points TL;DR

+ Everything I want in a small, light mouse.
- Software buggy for me
- mouse2 feels delayed sometimes for some reason, likely grip/force related

Next step is checking my other sample to see if its better/worse, and deciding which one to put inside a WMO shell.

Now available from Amazon US
http://www.amazon.com/Ninox-Aurora-Professional-Gaming-Mouse/dp/B00QLJ57K2/