test1
April 25, 2017, 05:23:53 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
News: You may have to login twice the first time,  but we don't know why - Erm I mean it's a security thing yeah that's it - security.
 
   Home   Help Search Login Register  
Pages: 1 ... 9 10 [11]
  Print  
Author Topic: jittering pointer  (Read 67646 times)
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #150 on: March 14, 2013, 11:16:44 AM »

You're close! Wacom tablets operate at a frequency of around 666KHz (depending on model), see http://forum.bongofish.co.uk/index.php?topic=2198.150 reply #155.

It is interesting that Hanvon tablets should use a frequency not a million miles away from Wacom, probably to do with physical constraints ie proximity of pen from the surface during normal use, maximum area of the antenna while maintaining an aestetic and practical overall tablet size, maximum power consumption requirements via USB etc. I suppose once you've planned the design to be a specific size and fine tuned the circuitry to resonate at some frequency you end up with something between 500KHz and 800KHz.

I also tried to observe the operation frequency with a Hall probe but didn't really pick much up, think this is because the operation frequency is modulated beyond the scope of the ability of such probes. The closest I got to measuring it was using a search coil and oscilloscope, but unfortunately even the scope couldn't differentiate the modulated signals effectively.
Logged
pntbll248
Jr. Member
**
Posts: 34


View Profile
« Reply #151 on: March 15, 2013, 07:05:11 PM »

Haha, guess I shoulda searched better Tongue But yeah, interesting that they both fall in relatively the same range, maybe that means that Hanvon tablets may be useful for these builds afterall?



Pro tip: Cut your pen, cut your jitter. Dunno if anyone's covered this yet. The wacom Intuos3/4/5 grip pens have approximately 5mm of plastic at their tip that can be cut off without interfering with any electronic functions of the pen. The location/pressure sensor is located far behind this. Cut off 4mm of this plastic (I whittled it down with a sharp knife), but be sure to leave at least 1mm of this plastic tip to guide the pen nib. If you don't, it's no big deal, but the separate plastic housing that actually grips the nib inside the plastic outer cover has about 0.5mm of clearance on the sides, so it will move around ever so slightly if you don't leave some of the tip intact. Cut the pen nib to size, making it approximately 4mm shorter than standard. The end result is that the location sensor is 4mm closer to the digitizer's antenna -- stronger signal, significantly less jitter, at least from my limited experimentation. I was able to take a B154SW01 from approximately 3 pixels of jitter after shielding to virtually no jitter whatsoever just by bringing the pen sensor closer to the screen. Of course, I'm not telling anyone to go hack up their pen, try it at your own risk, could set ya back $70 for a new one if it doesn't work for you. No idea about the art pen or the classic (slim) pen.

Edit: For clarification, it's best to slowly sand or whittle down the pen tip so as to avoid cutting too far. You do want to try to keep some outer plastic to stabilize the pen nib.

Also, 64.2hz seems to be the sweet spot for several AOU 15.4" screens, followed by 59hz.
« Last Edit: March 15, 2013, 09:00:56 PM by pntbll248 » Logged
bernard
Administrator
Hero Member
*****
Posts: 2584


pato mania


View Profile
« Reply #152 on: March 16, 2013, 01:55:00 AM »

Shortening the nib is an awesome tip!!    Cool

But if you tilt the pen at, say 45 degrees, does the tip still "align" correctly?

Logged
pntbll248
Jr. Member
**
Posts: 34


View Profile
« Reply #153 on: March 17, 2013, 04:52:08 AM »

Tilting the pen results in movement away from the nib at approximately the same distance as the pen nib has been cut short. So if you cut the pen nib 3mm shorter, it seems that 60-degree tilt results in a 3mm movement away from original calibration. Personally, that's not a huge big deal for me, I normally keep the cursor about 5mm to the left of the pen nib, lets me see exactly where I'll be drawing while still preserving the hand-eye coordination of using a LCD graphics tablet. I guess I got into that habit after using the Wacom 12WX, which I could never get reliably calibrated (realizing that switching positions changes effective calibration).

So it's a trade-off, a few mm in calibration accuracy when tilting the pen, or a decrease in jitter for jitter-prone screens (the B154SW01 is a bit more prone to jitter than the B154PW01 from my experience).

Edit: Oh, also, the tilt angle still seems accurate.
« Last Edit: March 17, 2013, 03:26:05 PM by pntbll248 » Logged
lukepjsmith
New Poster
*
Posts: 5


View Profile
« Reply #154 on: May 11, 2013, 02:35:49 PM »

Hi,
Fairly noob to all this but do have vague knowledge in some of the related areas. I started my build and *was* experiencing terrible jitter. After a few hours of playing around with various settings (changing resolution, backlight on/off, etc) I somehow managed to remove almost all of the jitter.
The two factors that I believe randomly affected it were the updating of the drivers for both the wacom (intuos 2) and the video controller in my laptop.

The strangest thing I noticed (and hope is not a common thing that everyone knows, although through my browsing I've not seen anyone say it) was that when I had outrageous jitter, using the eraser on the pen was almost unjittered. Don't know how much of a game changer this it but thought it might be useful to report to the big dawgs on the forum.





Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #155 on: May 11, 2013, 04:52:41 PM »

I've not heard that one before. Which tablet do you have?
Logged
lukepjsmith
New Poster
*
Posts: 5


View Profile
« Reply #156 on: May 13, 2013, 03:17:58 PM »

Its a wacom intuos 2 and the screen is an LG LP154WX5(TL)(A1)
Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #157 on: May 13, 2013, 03:23:32 PM »

That's interesting, my build is based around an Intuos 2 but I can say I've noticed this before. The next time I come to playing around with shielding I'll give this a test, but I can't imagine what would be the reason for it......Hmm interesting.
« Last Edit: May 15, 2013, 06:43:20 AM by Aerendraca » Logged
bernard
Administrator
Hero Member
*****
Posts: 2584


pato mania


View Profile
« Reply #158 on: May 13, 2013, 04:44:31 PM »

If the eraser works but not the main pen, then your pen is not tuned correctly.  You need to re-tune your pen.  There is a (couple of) potentiometers (pots) inside.

The pots are under the side switches plastic cap. These are typically removable *but* extremely fragile and you have to know where to apply force else it will break.

Logged
lukepjsmith
New Poster
*
Posts: 5


View Profile
« Reply #159 on: May 15, 2013, 12:24:05 AM »

Well the odd thing was that when I updated drivers and changed to duplicate screen mode instead of extended, the jitter almost completely went, which is why I suspect the pen is fine (that and the fact that without the screen it was fine too). I'm assuming it was the updating drivers that had most effect because now its not as bad with extended screen mode. I've now opened the tablet and am preparing to build an enclosure. With a 1.5mm sheet of PET the jitter is mostly gone, at times is comes back but it almost negligible when it does.

The point is that I was assuming it was that the signal from the tablet was being distorted by the screen but it seemed that in my case the pen was being confused by it. I guess I just got Lucky   Cheesy  
Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #160 on: May 15, 2013, 07:00:06 AM »

Sounds like lukepjsmith's pen works fine, no need for a retune, just noticed that the communication with the eraser wasn't as jitter affected as the pen tip.

I think that the reason you see less jitter in clone mode vs extended is that the graphics card isn't working as hard and is able to output two perfectly synced signals at each (dvi?) connector. Alhough there is some regulation for the clarity of the signal output by a graphics card ut is not all that strict which allows some cards (even expensive relatively high end) to get away with noisy digital signals. It would also seem that the more you have going on inside the card the more you introduce noise. There's plenty of sites on the web which show the output signal graphically for various cards and alot of info about this too, think you just need to search for something like 'lvds noise graphics card', 'graphics card eye diagram', or 'DVI compliance eye diagram'.

Also, I notice you want to use polyester (pet) sheet but because of it's low coefficient of friction it is not easy to glue without specialist adhesive. Don't want to put you off just giving you something to think about.
« Last Edit: May 15, 2013, 11:14:58 AM by Aerendraca » Logged
bernard
Administrator
Hero Member
*****
Posts: 2584


pato mania


View Profile
« Reply #161 on: May 15, 2013, 07:12:20 AM »

Ah ok, sorry. You are right, a detuned pen never works obviously.

Logged
lukepjsmith
New Poster
*
Posts: 5


View Profile
« Reply #162 on: May 15, 2013, 03:00:13 PM »

Ah this explains a lot as I'm using my (fairly average) laptop. Also explains why when photoshop slows down, jitter seems to come back, and why it was bad when I was using linux on my laptop because the drivers are a little messed up, I think.

I'm thinking the PET should be okay because its a solid sheet, not the fibre/textile version (polyester), plus my main aim is to design an enclose that utilises the holes in the tablet board and all components are clamped downwards using just 3 or 4 screws. Thanks for the heads up though.

Logged
Aerendraca
Administrator
Hero Member
*****
Posts: 1057


View Profile
« Reply #163 on: May 17, 2013, 06:56:07 AM »

Polyester is just a generic name given to a group of long chain polymers whose main group is comprised of interlocking esters. Apologies, i confused the situation by saying polyester but i did mean the solid sheet variety when i gave the caution about adhesive.
Logged
Sewje
New Poster
*
Posts: 5


View Profile
« Reply #164 on: June 18, 2014, 09:39:09 AM »

Not sure if this was mentioned anywhere but I've been testing Driver revisions for reducing jitter, and I've noticed version Wacom 6.3.5-3 seem to give the least jitter on my intuos4 xl. Still jitters but not as much as other drivers including the latest 6.3.8-3.

Some of the newer driver features are lost but jitter removal is more important.
Logged
Pages: 1 ... 9 10 [11]
  Print  
 
Jump to:  


Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!