Overclock.net banner

1 - 20 of 27 Posts

·
Registered
Joined
·
779 Posts
Discussion Starter #1
AVIOR 7000 Firmware 3.08
-Interpolation fix from 5000 to 7000dpi
-Disable profile fix
-New default colors on profiles

http://support.mionix.net/entries/66256899-AVIOR-7000-Firmware-Version-3-08

NAOS 7000 Firmware 3.09
-Interpolation fix from 5000 to 7000dpi
-Disable profile fix
-New default colors on profiles

http://support.mionix.net/entries/66256599-NAOS-7000-Firmware-Version-3-09

Both released November 19th. I've updated my NAOS 7000 and I can't really notice an obvious difference in smoothing at 800 DPI, but I'm not that well-trained on pinpointing this. If I remember correctly, the rumor was there was a new firmware Mionix was testing that was supposed to reduce smoothing at non-native DPIs. I've just been drawing some lines in Paint but I'm about to test it out in some CS:GO and TF2. The mouse does seem a little snappier on the desktop, but that may just be placebo effect.

Anyone noticing any differences or improvements with this new firmware?
 

·
Registered
Joined
·
2,479 Posts
I'm currently using the Avior 7000 right now and going to keep on v3.38 firmware unless the changelog says the lag reduction was included since this might be an older firmware.
 

·
Registered
Joined
·
4,892 Posts
This is probably the finalized version of 3.08. Keep in mind that 3.38 is test only. A fully working version of 3.38 would be their next firmware.
 

·
Registered
Joined
·
779 Posts
Discussion Starter #4
Where are people getting the 3.38 Firmware? On the Mionix site, they just list the new 3.08 Firmware and 2.34 BETA Firmware for the Avoir 7000. Is there 3.3x Firmware floating around for the Naos?
 

·
Registered
Joined
·
779 Posts
Discussion Starter #6
Ah right, I remember reading that thread now. Any insight into the techinical differences between the 2.x, 3.0x, and 3.38 Firmware? Thanks c:
 

·
Registered
Joined
·
4,892 Posts
3.08 had a bug always selecting profile 3 upon plugging in the mouse, which might or might not be fixed with the final release. 3.38 a lot more is broken like the LOD customization and lighting customization IIRC. It has no smoothing though.
 

·
Registered
Joined
·
779 Posts
Discussion Starter #8
Here's some interesting data I collected while testing the new 3.09 Firmware. I was trying to simulate the mouse input of checking corners in CS:GO with different length swipes and varying the pause in between direction change. Show is 1000Hz @ 400, 800, and 1600 CPI.

At the lower CPI, you can see what I think is the "rubberbanding" at the black arrows that some have noted in the 3310 sensor, while the green arrows show some questionable peaks and jitters at higher movement speed.

At the native 1600 CPI however, it seems both of those phenomenon are not present, and you can see especially from the much more subtle y-axis movement in red how much smoother the response is. I'm really not sure if that's just because of less processing at the native CPI or something like the smoothing algorithm behaves much better at the native CPI. But it would seem counter-intuitive that this would be just from more software smoothing than the lower CPIs because why would the native CPI have MORE software smoothing.

Anyway, this is making me consider switching from 800 CPI to 1600 CPI for gaming and halving my in-game sensitivities. Anyone else have any alternative/additional analysis of these charts?

http://i.imgur.com/Z1Wrkau.png
 

·
Registered
Joined
·
123 Posts
Quote:
Originally Posted by discoprince View Post

i have the 3.38 beta driver installed and i have no issues, should i even bother?

this thing feels really tight now btw.
Unless you really need LOD or lighting customization, I say keep the 3.38 beta driver until they have a finalized version of the 3.38 driver out.
 

·
Registered
Joined
·
318 Posts
I also don't know if I should update the firmware of my Naos 7000. I've only had it for a week and I'm not experiencing any problems. Its firmware is 1.63 and I'm actually using 1600 dpi and everything feels great.
 

·
Registered
Joined
·
779 Posts
Discussion Starter #13
I definitely still feel like my Naos 7000 is snappier with the new firmware. I've also changed to 1600 DPI and that feels really good on then desktop and in games. There doesn't seem to be any real downside to upgrading the firmware unless you just do it wrong, you could potentially brick your mouse.

Also, I want to put forth a strange problem I've always had with the Naos and how i fixed it. I frequently will use my left hand for the mouse when I'm just casually using the computer, which puts the mouse at about a 45 degree angle relative to my 'gaming' position on my 5mm Roccat Taito King-size pad. It would start skipping and registering very poorly at that angle (in fact, I thought I had spilled something on it to cause this and washed it with warm/cold water and microfiber shammy several times). Then I noticed simply reopening the Naos software would completely fix it. I think there had even been times in game when i rotated my hand too much and had this effect. So to fix it permanently, I just put the link to the Naos software in Windows Startup folder (or you could make a task in Task Scheduler). It's somewhat annoying that you can't start it minimized in the system tray, but it's worth never having to worry about that odd tracking bug.

I'm not sure if that's a problem somehow specific to my mousepad or hardware or software or who the f-- knows. It's just odd that simply opening the Naos software and hitting OK and/or Apply (just Cancel doesn't do it), which causes the mouse to blip off then back on for a second, causes it to track properly at the 45 degree angle and the normal gaming position. Just thought I'd share in case anyone else had experienced this problem or knew what caused it. I'm pretty sure I've seen at least one other person describe the same issue in a Naos 7000 thread at some point.
 

·
Registered
Joined
·
2,479 Posts
Quote:
Originally Posted by VolsAndJezuz View Post

At the lower CPI, you can see what I think is the "rubberbanding" at the black arrows that some have noted in the 3310 sensor
An Avior 7000 on v3.38 firmware feels like a laser beam glued to the screen to me without any rubberbanding whatsoever. I use 800dpi 1000hz.
 

·
Registered
Joined
·
779 Posts
Discussion Starter #15
Yeah looking back on that picture, I suspect I forgot to change the CPI from 400 to 800 for the second graph, because I did some more extensive tests and the rubberbanding you can see at 400 CPI clears up for the most part at 800, but 1600 CPI is still better yet and has less spikes and jittering during faster motion with the 3.09 Firmware. I hope us Naos 7000 users can get access to that 3.3x Firmware soon, because it seems to really sort out the few niggles you can find to complain about with these Mionix 3310 mice. I'm going to stick with 1600 CPI til then.

I have to say, with the new firmware, optimal CPI/polling rate settings, and using your input lag optimizations as a launching point, it really does feel like a completely new mouse compared to when I first got it. And it was already a tremendous step up over the 9800 laser sensor I was using previously. Cheers!
 

·
Registered
Joined
·
406 Posts
The lines in MouseTester are generated by the software as some kind of smoothed average or something, and shouldn't be taken as a reference for tracking quality. The points are the actual counts from the mouse. The reason the Y axis response is 'smoother' at 1600cpi is because you are using more counts to represent the same motion. The 400 and 800 y axis is literally a single count in magnitude, you can't have a smooth response when it is no larger than the smallest quantity possible.
 

·
Registered
Joined
·
779 Posts
Discussion Starter #17
Yeah I get that, but I was probably overestimating how much the averaged lines represent tracking quality. Still, I think there is a profound difference in going from 400->800** CPI as far as rubberbanding (which I can actually feel at lower CPI with my lower sensitivity I use in CS:GO) and fluidity of the count changes at higher speed movement (there are some pretty wild changes in counts that caused most of the erratic behavior of the averaged lines). Furthermore there's a smaller but noticeable difference, at least with 3.0x Firmware, in going from 800->1600 CPI. I suspect from others' comments that the 3.3x Firmware address these issues at the non-native CPIs.

**as mentioned in my previous post, I think my second graph was actually a screw up where I didn't change the CPI in the mouse software, so the first two are actually both 400 CPI. The test movements I did were all similar speed, so the 800 CPI counts should be about double.

Regardless, do you or other posters have suggestions for other ways to compare tracking quality, etc, with quantifiable data like with MouseTester, to have some hard data to look at and compare to qualitative assessments? I mean more than just observing the feedback from the MouseMovementRecorder or mouserate programs.
 

·
Registered
Joined
·
406 Posts
The problem with software is that it only knows about what the mouse is reporting, not what is happening physically to the mouse in real space. We can check for things like polling rate consistency, max count per interval and angle snapping (to some extent) pretty easily with software, but the more subtle subjective things that come up in tracking quality (smoothing, pixel walk) are difficult to create data on because they are being tested using human hands that can't do the same thing twice to the same mouse, let alone mice of different shape/weight/glide. Until we get our own turntables and robot arms, we are going to have to endure claims of swamp mice and clown cursor without any hard evidence, however true they may be.

Even with good software, people can still misinterpret what they see. I've seen mspaint tests where somebody has done the fastest circles they can do on a high cpi and called 'angle snapping' because the circle looks more like a decagon.
 

·
Registered
Joined
·
779 Posts
Discussion Starter #19
Well when I look at the graph and it looks like a smooth, differentiable function, it just makes me feel better, ok?

 

·
Registered
Joined
·
10 Posts
Quote:
Originally Posted by metal571 View Post

3.08 had a bug always selecting profile 3 upon plugging in the mouse, which might or might not be fixed with the final release. 3.38 a lot more is broken like the LOD customization and lighting customization IIRC. It has no smoothing though.
Did Mionix indicate when the final 3.38 firmware will be released?
 
1 - 20 of 27 Posts
Top