Monthly Archives: April 2013

What should the Combimouse sell for?

(Please let us know in the comments section in the Combimouse Indiegogo campaign what you think the Combimouse should sell for.)

There is a large market for keyboards and mice, with companies like Apple and Logitech offering numerous different designs and combinations.  Logitech offers a wireless mice and keyboard bundle for $110 while Apple sells their Apple Wireless Keyboard for $69 and their Apple Magic mouse for $70 – a total of $139.

What then should the Combimouse sell for (including free shipping in the US)?

Based on early Indiegogo campaign feedback a sale price of $80 and $100 has been suggested as significantly more attractive.

Can we achieve this and still be profitable?

The Combimouse uses similar components as other keyboards and mice so material costs should be similar.

However, until we obtain accurate manufacturing quotes it is not possible to confirm this. To obtain accurate quotes we need to provide final design documents to our suppliers. To date we have only asked for very conservative estimates based on the current prototype.

The keyboards are the most expensive items.

We will be using the same keyboard technology that is used on the Acer Aspire S3 laptop keyboard as designed and supplied to Acer by Darfon.

Darfon will only respond to our enquiries when they can see that the project is a real commercial project.

For now an estimate for the keyboards was supplied by a keyboard company that doesn’t have experience in the specialist field of notebook keyboard technology – they provided the quote in order for us to at least have some conservative estimate, with the understanding that we would be approaching Darfon for actual quotes.  This company will be manufacturing the plastics and will do the assembly of Combimouse, and so will get the bulk of the business.

Initial estimates from our supplier were $18 for the left keyboard and $19 for the right keyboard. These are the estimates that we have used up until now.

However, we think that Darfon can supply us with much cheaper keyboards. We will be using the same keyboard technology that is used on the Acer Aspire S3 laptop keyboard (made by Darfon). It costs $13 to purchase one of these keyboards on eBay, including a profit element.

It is difficult to determine whether this is a realistic and representative value for the cost of keyboards. The keyboard we purchased appeared to be a genuine Acer/Darfon keyboard and worked well. We will assume for now that $13 is what we can expect to pay as it also ties in with the low price of complete keyboards available for sale.

As you can see we have estimated conservatively and this has hurt our Indiegogo campaign.

To better reflect customer price expectations and what we think is achievable, we have decided to introduce new Perks to our Indiegogo campaign.

As of the date of this blog post we have added more affordable Perks with a limited number available.

Should we make a Left Handed version Combimouse?

Here is a concept of what a Left Handed version could look like.

This is just a quick sketch and isn’t a definitive solution. It is presented for discussion purposes only. For example a left handed version wouldn’t have the right part being mobile with the big palm/grip rest.

Should we make a Left Handed version - Big

What do you think?

Consider the following:

15% of people are left handed.

50% of left handed people use a mouse in a right handed way.

If we did a special left handed version, the Function keys could be located at the top of the right unit which would be stationary.

It is very expensive to make a special shape. Ideally one design can satisfy the majority of users.

We would like to build an optimized left handed version but at his stage it is not economically viable for us.

Scroll solutions for the Combimouse

A scroll wheel has been added to the prototype. Previously we considered putting it to the left of the comma key as shown in the sketch below. However we have put it on the right side of the comma key. As of October 2014 it appears to work well but hasn’t been fully tested. To save battery life it is only active during mouse mode. This isn’t ideal as one has to be gripping the unit to activate mouse mode and often during scrolling the unit is lightly gripped. This will be fixed in the future.

Combi right 730 x

The following paragraphs talk about the various scroll solutions considerations prior to June 2103. I have left them in the blog as a reminder.

There is no ideal scrolling solution for the right hand Combimouse part.

On the earlier prototype a Scroll wheel was located on the left part but that is not ideal and will be removed.

Currently the thinking is to implement both the Gesture Scrolling and Click Scrolling methods as described in the first two solutions below. The third solution is likely to also be present.

Gesture scrolling by moving finger along surface of the Comma key

This is the same that is done on the Apple Magic Mouse and some Microsoft mice.

By moving your finger along the surface scrolling will happen.

This hasn’t been implemented yet on the Combimouse, but it is expected to work and is practical to implement even though it is fairly complicated. I did say it is practical to implement – that needs to be confirmed.

There is a disadvantage of this system as it is very sensitive and doesn’t give as much control as a scroll wheel.

The other disadvantage is that it is only available for Mouse Grip 1 and not Mouse Grip 2

Mouse grip 1



Mouse grip 2


Click scrolling

This has been implemented and works well. The disadvantage is that like the Gesture Scrolling method it doesn’t provide fine control.

One advantage is that works for both mouse grip modes.

During mouse mode, if the P or Semi-Colon is pressed and released, the unit goes into scrolling mode. If you move the mouse up or down from this zero position scrolling goes up or down. The further you move from the zero position the faster the scrolling. If you keep the mouse at a position away from the zero position scrolling accelerates quickly. This makes it easy to jump to the top and bottom of documents.

To get out of the scrolling mode, any key is pressed or you got out of mouse mode by stop touching the touch sensor on the right side of the unit.

Alternate Click Scrolling

This method was prototyped but is not practical.

If you were to press a designated key during mouse mode, scrolling would occur by moving the mouse. The problem is you need to keep moving the mouse for scrolling to continue. By the time you are finished scrolling the mouse has moved very far.

Pointing stick

The pointing stick as shown below has been considered. It wouldn’t move the cursor as it normal does – rather it would produce scrolling in the direction it  pushed. I discarded this option – I must confess that I can’t remember why. I think it was due the high power consumption.

One advantage would be that the same pointing stick  could be used with both mouse grip modes. It also doesn’t add significant mass far from the centre of gravity

Scroll - Pointing stick


Thumb Wheel Scrolling

This method has been suggested and is being considered, although it would seem that it probably is not a good solution.

The big advantage is that it is accurate and has good control.

Two scroll wheels located on the mouse grip as shown in the diagram below would be needed. Two scroll wheels would be required – one for each mouse grip mode.

thumb scroll

One problem is that there there would not be enough space for the thumb between the Space key and the side of the side when using the top scroll wheel. In addition the height below the keyboard to house the scroll wheels is not sufficient. There is only 7mm available.

Another problem is that the thumb may not be suitable for continuous scrolling and may cause RSI – this is just a gut feeling and may not actually occur.


What’s it like using the Combimouse?

Updated – 1 July 2013

So, what’s it like using the Combimouse?

I am working on the latest prototype. I have just completed the new right part which has a number of new technologies including a new keyboard technology. I have also got the right part keyboard brake to work. I haven’t upgraded the left part yet and it is still a black part using older keyboard technology but it is 100% functional and doesn’t hinder use.

There are still a number of problems and so I can’t use the prototype full-time but I am working on it and expect to resolve the major issues soon.

I have been vague as to what the keyboard brake is. I explain the reason for this in this blog.

We are currently looking for an Australian tech blog to review the Combimouse. The aim is to validate the Combimouse concept. We could wait a bit longer but we need to show the potential of the Combimouse. Ideally we should wait until the left part is upgraded, wait until the external circuitry for the Keyboard brake is put into the unit, wait until the scroll wheel is implemented etc. But we have decided that it is better to let someone evaluate it as is, with the understanding that it is close to the finished pre-production status.

(JULY 2ND UPDATE – THIS FOLLOWING PROBLEM HAS BEEN MOSTLY RESOLVED. The key is a lot more sensitive and is now okay. It still needs to be firmly pressed but not as much as before.)I am using the new keyboard technology and I am having a problem with some of the keys as discussed below. The problem is that the flexible membrane that I had made is not optimum for this keyboard technology. In production this won’t be a problem as an optimised membrane will be used. AS I MENTIONED ABOVE THIS HAS BEEN RESOLVED.

Any small annoyance is easy to ignore for a few minutes or maybe an hour, but with extended use it becomes a problem. Up until June 2013, the right part moved slightly during typing and this was a problem. The keyboard brake was devised and this overcomes the problem. I can now use the Combimouse for much longer periods. There are still some annoying problems. Some of which are because the prototype is handmade. I am working on these and will resolve them soon.

Others are more intrinsic to the design. All these problems are listed as Disadvantages on the Combimouse site on this page.

Having a mouse at one’s fingertips is of course is the main advantage and it is great.

As a mouse.

When it is used with Grip 1(shown below),  it works as well as a regular mouse. This is how a mouse is normally gripped.

Grip 1 250 x 250


Scrolling is currently a problem. I have implemented Click scrolling (see the “Scroll solutions for Combimouse” blog) and this doesn’t work well and is frustrating. As discussed in the Scroll blog, in June 2013 I worked out how to add a Scroll wheel. I hope to add it soon. I will do that as a background task.

Here is a photo of Mouse grip 2

Grip 2 250 x 250


Using this mode is really great. It is especially useful when you need to regularly move between mouse and keyboard, for example during CAD work. The problem is that it is not as accurate as mouse grip 1 because you are not resting your palm on the table, but it is nevertheless very accurate and very usable. It is also not as comfortable to grip as grip 1 but still it is okay. Mouse grip 1 is identical to a normal mouse. I have become lazy and want to always use grip 2. Just like it is frustrating to move from a keyboard to a normal mouse I have become even lazier and moving from the keyboard mode to mouse grip 1 has become frustrating even though it doesn’t take much effort to move my hand to grip 1. It’s amazing how lazy people can be – or should I say it’s amazing how lazy I can be!

So, for occasional mouse usage I would say that Mouse grip 2 will be used, but for extensive mouse usage and for accurate mouse usage I would say that Mouse grip 1 will be used.

(THIS FOLLOWING PROBLEM HAS BEEN RESOLVED)During Mouse grip 2, the J key is used as the left click button. Because this is a normal keyboard key it needs to be held down firmly during drag operations, else a number of key releases and presses may be detected. I have added software filters and modified the mechanism slightly and this has made some improvements. I can’t be too aggressive with the software filter else it won’t register double clicks.

This should solve another problem. At the moment I have to press the J key in the middle when I use it as a left click button during Grip 1. THIS PREVIOUS PROBLEM HAS BEEN RESOLVED.

During mouse mode with Grip 1, the left click button is the comma key – the comma key doesn’t use the flexible membrane – a switch from a mouse is used and so false key releases do not occur during dragging.

As a keyboard it works as well as a normal keyboard. Some keys have been moved. Some keys like the backspace and delete keys have been moved to very convenient positions. I am used to the new positions for these keys and find them very convenient.

Here are some other keys that have been moved:

Back Slash, Forward Slash, Home, End, Page Up and Page Down.

You have to lift your palm off the right part when typing else you will bump your wrist into the part and it will move. I am used to it and don’t find it an issue. Trained typists are trained to keep their hand parallel and above the keyboard during typing.

The palm rest/palm grip is low and flat and so it is easy to lift your palm. It is very similar to the Apple Magic Mouse which Apple have sold with their computers for the past few years. Apple have done this because the surface of the mouse is a gesture surface.

So the palm rest on the Combimouse is very comfortable as a palm rest during pauses between typing as it is flat. As a palm grip during mouse mode  it does not fill the bottom of the palm. Apple obviously don’t see this as an issue and it feels okay on the Combimouse. So if you want to get a feel as to how the Combimouse feels with grip 1 try using an Apple Magic Mouse. It is similar but with some differences. With the Apple Magic Mouse occasionally the hand grips the sides near the bottom near the bottom but on the Combimouse the hand occasionally touches the top surface – I hope that made sense and you could follow what I said.

The Return key is possibly too small. I will look into extending it a bit to the right.

AS MENTIONED ABOVE THE FOLLOWING PROBLEM HAS BEEN RESOLVED. Some of the keys on the prototype are currently a little difficult to press and this is annoying. I am currently fixing this.

The Space key on the right part is still a little uncomfortable. I have some ideas why that may be and I am working on it.

There is a side effect from the keyboard brake which is distracting and I am working on resolving this. Once again I apologise for not providing details on the keyboard brake.

Combimouse Design Evolution

In this blog I will give you a brief summary of the Combimouse design history. I have made many prototypes over the years. At the time it was invented, technology wasn’t available to make it a reality. There were also difficult design obstacles to overcome.


I will refer to the Old prototype shown below as “OLD”. OLD has a mass of 194 grams. The centre of gravity is located somewhere along the red line. This is what the Combimouse looked like in 2012.


The latest prototype (as of April 2013) is shown in the following sketch and has a mass of 74 grams (this includes 26 grams of lead weights to optimise the centre of gravity and weight distribution). I will refer to it as “LATEST”. The centre of gravity is located somewhere on the red line. It is 120 grams lighter than OLD and lighter than most regular mice.


As a comparison, below is the Apple Magic which has a mass of 108 grams. The centre of gravity is located somewhere along the red line. It is 34 grams heavier than the Combimouse.


Originally with OLD, I thought that the palm grip had to be bulging. This was necessary so that it would fill the palm of the hand while using as a mouse. This was especially true up until a few years ago, when the right part was heavy and a good grip was necessary.

The problem is that this bulge made it uncomfortable as a palm rest during typing. As a palm rest it has to be flat and low. Also when I used to move my hand when typing I would bump into the large bulge and the whole unit would move.

On OLD, I was using AA batteries that have a diameter of 14mm. Also on OLD I wasn’t aware that height is an issue – I have subsequently discovered that the top of the keys must be as low as possible. The height of the top of the keys above the desk on OLD is 27mm. On Latest it is 14mm. So LATEST is much lower and that makes typing much easier.

In addition the mouse click keys are now much lower and better.

Apple started selling the Apple Magic Mouse a few years ago. I was surprised that it didn’t have a bulge to fill the bottom of the hand. They made it very flat and low. This is probably a compromise so that they can make the surface a flat touch sensitive surface for gestures. Apple supplies this mouse with all their Macs.

Last year (2012) I realized that if Apple can do this then so can I. I can now have a low flat surface for the palm rest/palm grip, which is ideal as a palm rest during typing. It is also good enough as a palm grip during mouse mode especially now that LATEST is light and doesn’t need a firm grip. This was a major breakthrough.

In addition electronics have reached a point where power consumption is very low. Today a common Logitech wireless mouse has over 1.5 years battery life. So the Combimouse, which has a relatively high current consumption, can use the lighter and smaller AAA battery. This makes LATEST lower and lighter.

Mass of Keyboard Section

It is critical to make the mass of the Keyboard section as low as possible so that it isn’t top heavy during mouse mode. Even 1 gram is a lot. Notebook Keyboard technology has slowly improved to a point where keyboards are very light.

In addition thin plastic walled devices are now possible allowing for a light Combimouse. One mm thick walls are now easy. This is mainly due to advances in mold flow analysis software, which occurred in the last few years.

This progress in technology makes a light design possible.

There are two main conflicting design obstacles that had to be overcome with the Combimouse:

  • Make it mobile as a mouse
  • Make it immobile as a keyboard.

Mobile Mouse

To make it mobile as a mouse it was necessary to reduce the mass and achieve a favorable center of gravity and low moment of inertia. What this comes down to, is making the mass of the keyboard section as low as possible. I have got it to a point where the area bounded by the Equal, Seven, Space and Shift keys is about 15 grams. This took a long time. I have been liaising with the manufacturer to check that this solution is manufacturable. They think it is manufacturable, but would need to use software flow analysis to confirm this and this costs money.

Immobile Keyboard

The problem of making the unit immobile as a keyboard was a difficult problem. Earlier this year (2013)I decided that I needed a radical change to solve this problem, as I wasn’t getting anywhere. I have now devised a solution. Unfortunately I can’t talk about the solution, as I will be patenting it.

After all these years it is a relief that this problem has been finally resolved.

As a result the Combimouse works well as described in the blog “What’s it like using the Combimouse?“.

Even with touch devices there is a huge market for Combimouse

The news these days is that “PC sales plunge as tablets bite“.

Does that mean that the Combimouse has missed it’s opportunity?

Or as the this article suggests is it that:

“The reason people aren’t buying new PCs isn’t that they don’t need a PC. It’s that, for the most part, they’re getting along just fine with the one they already have.”

and as this article says:

“Tablets and smartphones aren’t replacing PCs, says Sacconaghi, but they are likely “lengthening the replacement cycle” as they cut into some of the PC’s workload and make their replacement less urgent.”

This slide from Business Insider summarises it best.Computers in Use

Why not use Bluetooth – Question from Indiegogo

There was a question about using BLUETOOTH, but there is a problem – see below.
Bluetooth will have to come later. I need to check but certainly a few years ago it was expensive and difficult to implement.
The problem is that there needs to be interaction between the left and right units. For example when you hold the 2nd Fn key down on the left unit it enables alternative functions on the right side. These alternative functions include modifying cursor up to page up for example. Another alternative function is temproarily modifying space to back space. There are many others.

The left and right units transmit data to the dongle and the dongle can combine the functions described above.