Incremental Blogger

Lunch with FinePoint

I had a fascinating lunch with Steve Caldwell (CEO) and Rodney Standage (Major Account Sales) of FinePoint Innovations yesterday. You may know of FinePoint as the developers of the original digitizers for the HP TC1000 and the PaceBlade Tablet PCs. But this only hints at FinePoint's long history in the digitizer market. FinePoint is a third generation company with an engineering lineage that stretches back to 1979 and the founding of Kurta—also rooted here in Phoenix.

FinePoint has been involved with the Tablet PC from its beginnings. The original Tablet PC prototype used a FinePoint digitizer. Oh, and the original IBM ThinkPad—no, I mean the real pad-like ThinkPad, not the follow on notebooks—used a digitizer developed by some of the current FinePoint team. So Steve and Rodney know their Tablet stuff.

Rodney and Steve invited me to lunch because they wanted to hear a Tablet user's perspective. Their direct customers are the Tablet OEMs/ODMs and so they naturally focus on their needs. Lower cost, smaller, lighter, less power. But what is important to a Tablet customer? Steve and Rodney want to know. With their long engineering experience in pen computing they understand the issues well, but they want to make very sure they understand the end customer even better than before.

We met at Z-Tejas, a home grown restaurant chain with a reputation for unique Southwestern food. It's the hot spot for business lunches. However, in a matter of minutes, the crowded restaurant melted away as we got into a packed conversation about Tablets, weblogs, Microsoft's MVP program, digitizers and pens.

They wanted to know what I thought was important in a Tablet digitizer. I told them cursor accuracy and a good set of pens. I want the digitizer to record faithfully ink as close as possible as to what would appear on paper. I don't want to have to mentally adjust my handwriting as I get close to the edge of the screen or worry about the ink stuttering when I first use the pen.

We talked a lot about pens. What makes a good pen? I'd like something more substantial than the lightweight Wacom pens that come with most Tablets. To the OEMs a better pen is a lighter, smaller pen that fits within the ever-thinner Tablet shell. But to me, I'd like something with more bulk with good balance and feel in my hand. Something designed for comfort. If you open my desk drawers you'll see a whole slew of pens. None of them are this plain. Well some are. Some freebies I've picked up at tradeshows are. The majority of the pens I've bought though have a rubber grip for my fingers and an attractive durable case that fits nicely in my shirt or pant's pocket. Now the bigger the pen, the bigger the storage problem in terms of Tablets. But to me there's nothing like a good engineering problem like this to spawn some creative thinking.

FinePoint digitizers are different from the traditional Wacom hardware although both are similar in many respects. They both employ a very thin 2D array of antennas behind the display to pick up electromagnetic waves emanating from the pen. The difference is how the signals are generated. The original FinePoint digitizer produces a signal using a coil and small battery in the pen. Wacom flashes a signal within their antenna board which is picked up by a coil in the pen and then as the energy is released from the coil it is sensed back in the antennas on the digitizer.

What are some advantages of the FinePoint approach? First, by using a battery in the pen, the signal that the antennas need to pick up is stronger. Plus the signal is making only one trip—from the pen to the antenna rather than a two-way trip for the Wacom pen. The result is that it's possible for the pen to be detected slightly faster and more accurately than the Wacom pen. It also means that the Tablet's battery is not being wasted trying to induce a signal in the pen. The pen has its own battery and is generating its own signal. This means the FinePoint digitizer consumes less power. If I remember correctly it's something like 10mW for the FinePoint digitizer versus 300mW for Wacom's implementation. Thirty to one difference is huge when you're accounting for every drop of battery life. One other advantage is that the FinePoint technique scales quite nicely. It doesn't take anything special to place a digitizer behind a 21” display whereas the Wacom approach will demand greater power consumption to push the signal greater distances out to the pen.

Now there are some drawbacks to the FinePoint strategy. For instance, the original FinePoint pens have batteries and are slightly larger. FinePoint is working on this. To me it's an issue of refinement versus wholesale reworking since the approach has several benefits that are ideal for the Tablet.

Oh, we also talked about pen tilt. I've heard requests for pen tilt at every developer session I've been at. And I passed this along. They were curious as to what ways people would use tilt information. To them it's not hard supporting tilt, but it's a matter what it costs. From their perspective, tilt information can already be picked up by the antennas. This isn't a big deal. However, it does add to the packet size. For every packet of data coming from the digitizer you might imagine another couple bytes of pen tilt information. And bigger packets require faster microcontrollers processing the data. This is where the OEM catch-22 is. Because faster ASICs or whatever cost more money. And in the OEM world every penny (quite literally) counts. It's going to take continuing nudges from the user community as well as continuing price pressure at the parts level to make the OEMs/ODMs more comfortable about adding pen tilt as well as other features in Tablets.

I also mentioned to them, that from the first time I worked with Tablets I've wanted to measure the cursor/pen inaccuracies. I've wanted to create a 2D disparity map that displays let's say green where there's little error between the pen tip and Tablet cursor and maybe red where the error is large. Characterizing digitizer inaccuracies for each Tablet would be useful as a consumer and as a techy I want to understand the positioning inaccuracies better. We talked about how this might be done. Who knows, maybe with some help from their test equipment it'll be possible to do this. We'll see.

After a long lunch Rodney and Steve invited me over to their office and gave me a nickel tour. They showed me an antenna grid—something I hadn't seen before—and their x-y position table which they use for calibrating the digitizers.

We talked about a lot more, but this post is getting long and I need to vacuum before some friends come over this morning, so I'll leave it at this for now.

If anyone has some thoughts on what they'd like to see in a Tablet digitizer, post your thoughts here or better yet on your blog. The FinePoint team is reading them.

Posted by Loren at June 12, 2004 11:23 AM