Wacom EMR vs. N-Trig vs. Wacom AES

So I gather from your post that AES does not have N-trig's precision problem in the video Marty linked, or did you mean it has better precision only on the edges?

The reason I ask is EMR is more precise than the resolution of the screen, so it is hard for me to understand when you say that AES is more precise. How could you tell?

 

It has better precision AND accuracy than EMR everywhere IMO. I don't have an Ntrig pen to compare. I have Excel on the 8" Encore and its trivial to resize columns - it never fumbles. But its possible Excel has a more forgiving divider detection scheme than Libre (or whatever was used in the video). I just set my Excel window to 25% zoom to make it more challenging. At that zoom level the standard columns are about 10 pixels wide, the divider bars ~1 pixel. I can still resize without a fumble. I can even adjust row height which are about 5 pixels high with a target less than 10 pixels from the left edge of the screen.

All the usual caveats apply: personal preferences, device setup, subjective biases, etc. But I think you know from my other posts Meso that I attempt to be objective about these tests and try to find some standard of measure that can be used between devices/users. If someone has a testing protocol I'd be glad to run it.

I may try to start posting some videos but I didn't really want to create a whole Youtube account for these. I don't see any way to upload video directly. But I'm making a number of discoveries about this device that really have to be seen. The only complaints I have about the TruPen are its slick finish, overly integrated buttons,and lack of backup pen supply. Some Sugru mods will take care of that but I wanted to buy an extra pen before I started modding the only pen I have.

 

It looks like your reply beat in my edited in final paragraph =P

But as far as I can tell in my precision tests for EMR, it is essentially "pixel perfect". Also the specs of 500 LPI should result in a precision that is better than pixel perfect (in theory). What are you seeing that is leading you to say AES is more precise? How do you improve on "perfection".

I'm certain you are seeing something that is leading to you saying that. I do know you make an attempt to be objective (and I know how hard/impossible that can be with this sort of thing). I'm just trying to get a handle on what properties you are seeing that leads you to that conclusion... as I live vicariously through you =)

 

I'm curious about this too.

The only occasion where I've found the UD digitizer's precision sometimes wanting, is when taking pixel-perfect screenshots with an assisted magnifier (eg. FSCapture). About 10-20% of the time I try to perform a 1px shift with the pen tip, the cursor may shift 2px instead of 1px.

Yo Meso, speaking of which, I think the above scenario would make an excellent "standardized precision test" . We could call it the "1px shift test", where users assess the difficulty of perfectly screen capturing a 1px border box (of course, adjusting for PPI, user dexterity etc.)

@Precurve: how is the hover latency (i.e. responsiveness) compared with the SP3 and Wacom UD? Can you get the cursor to trail behind like ron2k mentioned?

 

I don't know how to respond to this without being subjective. We need some other kind of diagnostic. May be I'll look at FSCapture and see if that can become a workable test.

@Marty, I don't have an SP3. I have a Fujitsu t904 (penabled), Cintiq 21UX, and Toshiba Encore 8" (AES). The "hover latency" subjectively appears to me to be identical to the Wacom on my t904.

Meso, I guess my conclusion is based on overall responsiveness, lack of parallax and near perfect registration across the screen rather than sub pixel resolution. It doesn't just seem easier to write and draw with I can do things in tests; eg, matching line endpoints, smoothly drawing 1-1000 pixel line, manipulating near-edge UI elements (toolbars, close boxes, hidden task bars) more easily on the AES than EMR. That last item in my list is by far the most obvious. The EMR pens by comparison have really sloppy registration across the screen. So, I think its a "perceptual" sum of factors. I'm still looking for someone to give me an EMR test that the AES pen fails. I can make toolbar buttons that the EMT pen will struggle with because it loses accuracy near the edges.

Another small thing, but with EMR on my t904 I have the registration set up to work in upright landscape. It is then slightly off in portrait - or any other odd angle, which can be important when sketching and rotating the device (it goes from being accurate and precise to being precise but not accurate by virtue of device rotation as well as nearness to the edge. With the AES pen I never recalibrate. It just works in all views and angles. This is a perhaps a different topic but it continues to add to the perception of a more accurate and precise experience. It has made a huge difference in my ability to doodle and rotate the device to get better arm and wrist angles without the cognitive overhead of compensating for the out-of-calibration pen tip.

 

From what I am understanding from the examples you give above. I would classify all of them as accuracy problems rather than precision (except the edge issues). That in no way diminishes them as important factors of course. I would just say that the problems that lead you to the subjective impression of worse performance with EMR is the sum of a bunch of different accuracy problems.

Also, as you say, paying attention to the hover cursor rather than the pen tip is an extra "cognitive step". I can't say it is a step that causes me a great deal of effort, but that may just be because of having over a decade of practice training myself not to even pay attention to the pen tip. If a wacom device wont calibrate well for me, I tend to bias my calibration so that the cursor sits just to the left of my pen tip (so that I can always see it).

When I say EMR seems to have "pixel perfect" precision, I'm saying the position of the pen cursor can be controlled to that level. This has nothing to do with systematic differences in the position of the pen tip relative to the cursor (which I would classify as an accuracy problem).

Really all of this is just a minor issue for me. For most of my writing it is just muscle memory so I'm not paying attention to either the pen tip or the cursor. But for fine corrections, like changing a minus sign to a plus sign, I pay attention to the cursor and am able to control where the ink is laid down to the precision my hand steadiness allows, and it sounds like from what you are saying (particularly from ron2k and marty's tests) that this is equally easy to do with AES. Whether or not EMR really has a precision disadvantage, I think it is saying a lot for AES that it can even equal EMR in this regard (particularly considering its edge and accuracy advantages).

 

That's fair and seems correct.

Lets just put it this way: If you made me play a target game for prize money using a stylus, and the target was a one pixel spot that bounced around the screen off the edges I would today choose to play that with Wacom AES not EMR. No question about it.

 

From your description of the pen performance it definitely sounds like it would win in that contest. Particularly considering the fact that the tip/cursor biases vary across the screen and with pen angle.

AES is sounding very appealing to me the more I learn about it. I could definitely see myself going full AES at some point in the future. A big detractor that probably doesn't apply to most users is that I use 3 different penabled devices and carry an EMR pen in my pocket. Having to carry two pens (and I'm sure grab the wrong pen for the wrong device 50% of the time) is a huge detractor for me... Plus, as I said, EMRs downsides don't really effect me much. In a vacuum though, I could see myself potentially preferring it to EMR based on what you and others (Kuma) have said about it.

One linger fear I have is the long term performance. Things like contact corrosion, unexpected battery drain and generally intermittent flakiness seem to be a common problem with active capacitive pens. My feeling is that part of the problem is a fundamentally more complicated pen which leads to more potential things to go wrong. A wacom pen is basically just a passive LC circuit, but I am not enough of an expert at these techs to really call it more than a feeling =) Also, I haven't heard anything like that about AES yet, so that may be a problem Wacom managed to solve.

 

Ok, this is huge.

If my ears are not mistaken, this means an AES Atom tablet can match the hover latency of a full-blown EMR Core i7 tablet! And have equal or better tracking precision and accuracy. This is OMFG$%^ news!

But I still hold some reservation: doesn't this mean Wacom essentially engineered its main products lines into obsolescence?!

Not that I doubt you Precurve, but could you do some other tests: fast circles, straight line jitter, initial activation force, palm rejection reliability, and vectoring on small/quickly-printed letters?

If the AES performs as well as your T904 on all these, it's hallelujah for all tablet PC owners.

 

You should check the other thread for some of that if you haven't already (hes already done extensive testing there spread across multiple posts). You may find some of your answers there.