Eric Cosky

In the field of USB missile launchers, there is a bit of competition. This year I was able to conduct side by side tests of two of the more well known missile systems: The Chinese V2 Foam 3-G Missile Defender and the Dream Cheeky USB Missile Launcher. In a strange twist of fate, both systems arrived within days of each other and I could finally perform a comparison between the systems without either manufacturer giving my some line about how it was an old model. These are fresh, right off the line, state of the art USB missile launchers and neither brand will have any leeway from me during the analysis. At this time, I am not aware of any other entries to this sector so this may very well be the first and only product comparison ever made on the subject. This article will determine which of the two is the winner in the contest for USB Missile Launcher of the Year, 2006.

First, the Chinese V2 Foam 3-G Missile Defender.

The name is apparently classified, because it came in a box marked only as "Made in China", so I coined the name based on what information I had: the place of manufacture, the general shape of the missile, the construction of the missile, how many missiles there are, and finally "defender" because it is a stationary unit as opposed to a mobile unit which would have suggested a name more suited for aggression. The "-G" notation is because of the generic packaging and the absence of any name at all to discuss this unit with.

The CV2F 3-G MD is itself is a 3 missile system is perhaps best described as a stationary, miniature form version of the Spyder Surface to Air system with one less piece of ordinance. The exterior is a rigid monochrome gray plastic with a power switch and an access plate to the power chamber on the bottom of the unit. This system requires 3 AA LR6 1.5v power cells and will not function without them, despite the fact that another power source is available via USB. This redundancy would have been a bonus had there been a wireless control system, but unfortunately the USB must be connected in order to operate the control system at all, rendering the redundant power system pointless.

While the launching system bears resemblance to a Surface to Air system, the ordinance itself is most closely related to the German V2 rocket systems of World War 2, which would make it suited for the typical installation environment where it's most likely intruders would be ground based (possibly even stationary).

Something to note in the above picture is the plastic rod sticking out the back of the missile. This is what is inserted into the launching system with a significant amount of force to compress a spring system that is then locked in place, much like an old fashioned dart gun as shown here:

Speaking of installation, this process is fairly typical of most USB devices with no surprises, with one potential catch: The printed instructions are not as clear as they could be about imparting the importance of not connecting the unit to the PC until the software is installed. While the instructions, if followed in order, would result in a clean installation, it is plainly possible that a person excited about setting up this system may overlook the fine print and connect the device right away expecting Windows XP Plug & Play to know what to do. I didn't want to take the risk of this, so I was very careful to install the software first. If the manufacturer should ever read this, I suggest putting a label on the USB connector that must be removed before it can be used which explains the installation order requirements.

The installation itself is done through a SETUP.EXE which installs an icon on the desktop for all users. Unfortunately, this installer does not support Add or Remove Programs features of Windows XP, as required by section 2.7 of the Microsoft User Interface Guidelines. The absence of a start menu icon was also mildly annoying, a user should not be required to see the icon for an application on the desktop at all times, no matter how important the software company may feel their product is - even a missile control system such as this. Since I run as a limited user, this non compliance is made even more of a problem because I cannot remove the desktop icon at all; I had to manually clean up the start menu and desktop so that the application can be reached by all users instead of just the user who installed it. 

This probably explains why the disk says "Compatible with Windows XP" and not "Designed for Windows XP". I can only imagine the compatibility problems this device will have with Vista if such a simple guideline wasn't followed right out the gate like this. Making matters worse, the software did not install in an obvious folder such as "C:\program files\missile launcher", it instead put the software in a folder called "c:\program files\USB Missile Launcher". While I suppose it's unlikely we will ever see a FireWire missile launcher, incorporating the name of the cable used to connect something to the computer into the name of the folder used to store the software is an odd thing to do as far as I'm concerned. I don't recall ever owning a peripheral that installed the software in a folder where the name was based on the cable used to connect it.

The icon itself was clearly not subjected to usability testing. Here it is:

As you can see, it's a shortcut. Shortcuts can have any name they want, including things other than "Missile_Lau...". Had they placed a space in the name, it would have appeared as such:

Unfortunately, since the icon was installed on "All Users" desktop, I had to switch to administrator mode just to rename this eyesore before I eventually moved it to the start menu to clean up the desktop.

Once the software was installed, I connected the USB cable. The cable itself is 1.2192 meters of standard construction with no EM shielding, protective sheath, gold contacts, or anything beyond the bare minimum USB cabling. But it is, thankfully, a full 1.2192 meters which I will spoil a bit by saying it took the prize the cable length category this year.

The operating system detected the USB device correctly. However, it detected it as a generic HID device as opposed to a specific "Missile Launcher" device. Surprised by this, I reviewed the filesystem behavior during the attachment of the device but I didn't see it access anything that appeared device specific. After scanning & filtering the filesystem logs for the installer, which oddly included hundreds (if not thousands) of accesses to a file called "C:\STOPTHIS.NOW", I didn't see anything so I have to conclude this launcher does not have a custom USB driver.

When I ran the Missile_Launcher.exe application, I was greeted with the following window:

This was due to running in limited user mode. However, the mouse cursor was hung up and I had to control-alt-delete at which time the cursor freed up, I was able to close the window by hitting the close button (despite it being hung before showing the task manager). I have no idea why it was creating this file. Finally, I was able to see the master control window:

Sometime later I noticed a new control down in my toolbar:

The toolbar icon is a nice touch; when the primary window is minimized the toolbar icon can be used to bring up the master control system or to shut it down entirely.

Unfortunately, while I was able to work through the problems getting the application to start, the "Cannot create" dialog came up every time I tried to actually control the unit. In order to solve this problem without dropping into administrator mode, I enabled folder writing on the "c:\program files\USB Missile Launcher" folder so it could create it's temporary files. This is another violation of the XP guidelines (section 3.1 and 3.4) with respect to where shared data can be stored, and general lack of support for limited user mode. Its unfortunate when this happens, often seen with products in this category, but fortunately this is often solved by allowing all users write access to the folder that contains the application. Take care not to enable write access to all files in the folder, or you will risk enabling malicious software to manipulate executable files.

Now that we're done with the grunt work of installing a new missile defense system and figuring out how to make it not crash, we can now test it's performance. For testing, the unit was physically installed upon a desk approximately 0.7874 meters from the ground. The instructions are quite clear about turning off the power switch when not in use, so I had to pick up the unit and turn it on because the switch is on the bottom. This has the unfortunate effect of losing any element of surprise unless you plan on buying batteries just to keep the unit on standby mode for the likely to be increasingly rare event that you feel the need to activate it.

The turning radius is 180 degrees horizontally, and the vertical range was approximately -10 degrees to 40 degrees from the horizontal plane. I would have preferred a zero to 50 degrees range, but presumably the downward angle would facilitate installation on higher location while retaining the ability to directly target below the plane. 10 degrees downward doesn't seem like it would buy you much elevation though. The gear action of the control system was reasonably tight for horizontal movement, but there was a fair amount of play in the vertical control system, suggesting these components may not have had the best testing at the factory, or possibly even a design flaw. When turning limits are reached, there is no feedback in the UI to indicate this has occurred. However, the unit itself does not push against the gears when this happens, suggesting a firmware update could add this feedback feature in the future. It's not clear if this unit supports firmware updates however.

Firing the unit is fairly straight forward (or at any point within 180 degrees/-10 to 40 degrees, anyways). Just press the red Fire button on the UI. After one second of internal calculations and associated rocket launching sounds, the unit launched one missile. However, spamming the fire button multiple times did not launch another missile - it apparently needs an internal cool down period before multiple missiles could be fired. Spamming the fire button repeatedly, I was only able to get the three missiles to launch in just under 15 seconds. The missile flight itself was not as smooth as I had hoped, either. They were not aerodynamic and tumbled more than flew. They individually flew 15, 14, and 13 feet from the fixed launch position 0.7874 meters above the ground.

So it appears at this point to be a functional device that does what it is supposed to, albeit with significant shortcomings. The battery requirement is unfortunate, the firing speed is sub par, and the lack of control feedback even though there is already bidirectional communication clearly available is unfortunate. Suspecting a flaw in the software, I fired up the system analysis tools and discovered the control system was actually spamming disk writes during the operation of the system! Observe:

Constantly creating and deleting the tttttt.jpg file, its apparent the system programmer was unaware of the feature of computers called "memory" with which temporary data, such as visualization systems, can utilize to store, combine and process information related to real time visualization such as the blips on the UI that move around despite the absence of any sensor device on the launcher itself. If there is a sensor, it did not appear to track my motion at all and I have to conclude radar system is completely non functional.

But now for some bad news. The missile units have a severe design flaw; the missiles are part foam and part hard plastic. If someone were to pull on the missile so they could, say, measure it, the foam can (and will) separate quite easily from the rigid plastic projectile that remains in the launch device.

I exerted what I feel was less than 2 newtons of force attempting to remove the missile for measuring and the foam detached from the plastic rod which was mechanically locked in place to the launcher. The only way to apparently remove what is now a plastic bullet from the device is to actually launch it, or take a hammer to it and extract the pieces from there. I haven't decided which technique to use just yet.

Moving on to the other serious contender, we have the Dream Cheeky USB Missile Launcher.

Like the Chinese V2, the Dream Cheeky also has 3 foam missiles, connects via USB and has a computer control system. At first description the Dream Cheeky has nothing special over the CV2F 3-G MD, but one look at the box itself after the last unit's bland presentation was enough to put me in a good mood about this system. This manufacturer clearly had the customer in mind not only when it came to actually choosing a name for the product, but also the exterior packaging of the box. It is quite shiny, and has pictures, including those little market friendly squares that say it does this or that including: "Win XP", "Software Included", "USB Connect", "VMS: Virtual Motion Sound Technology", "Plug & Play", and "Online Support". Quite a raft of features for any product, and the VMS was something new that might have potential to bring something fresh to the foam rocket defense industry.

Despite all the advertised features, I have to take some issue with "Dream Cheeky" as a brand name. My guess is the marketing folks responsible weren't aware that the (I expect is the) target audience of U.S. market missile purchasers tend to prefer more masculine names, such as "Fiery Armageddon" or "Sweet Justice".

Fortunately, the missiles themselves don't say "Dream Cheeky" on them, they are a down to business construction with a yellow foam body, harder foam tip for weight, and orange fins on the back. The launch unit itself is a two tone black and forest green plastic which reeks of quality after the experience with the Chinese V2 Foam 3-G Missile Defender. The package included 0.4191 meter target printed in 3 colors, the installation CD with a full color image upon it, the launcher assembly packaged in double sided protective plastic, and 3 pressure driven projectiles packaged with shipping rods in place to keep the form in good shape for a pressurized launch system. There were no printed instructions. Everything was taped up nicely and clearly a professionally assembled package although they did take the tape a bit far by taping up the CD case, forcing me to tear it open.

First impressions of the hardware and accessories were good. So I inserted the disk and took a look. The base SETUP.exe application was dated 1999, which was a bad sign, but the actual data files used by the installer were dated 2/12/2006 which put me at ease. I ran the installer as a regular user with limited permissions, and it failed to install but it did at least roll back the partial install. Again, not a surprise for this kind of product to have Made for Windows XP compliance problems, it's fairly typical even of more widely used software. So I installed it again as an administrator via "Run As..." menu item and it gave me the option to not put an icon on the desktop (+1) and put the application in the start menu visible to all users. Bonus points for Dream Cheeky. The installer could have been smoother by providing an option to choose which sub-menu on the Start menu I would prefer to use. Again, not uncommon for this range of software so I can't be too hard on it. Besides, it made up for it by providing an uninstaller. That's right, it has an uninstaller, and it actually worked.

Unfortunately, the desktop icon it installed was installed to the desktop of the account that installed the software (that Administrator), rendering it invisible to everyone else. It was, however, at least using an appropriate link name that enabled the text to be visible. Rather surprised it didn't look like a missile, or a launcher, or some combination of the two.

At this point I'm winging it because there is no documentation to speak of; none in print or on disk. I plug in the device, and I observe the Windows XP device discovery message telling me the Dream Cheeky USB Missile launcher was detected as a standard USB Human Interface Device. Excellent news, no fear of driver problems deep in the OS.

Astute readers will note the conspicuous absence of a discussion about power supplies, batteries or power switches. That is because the DC USB ML has better technology than the competition, at least in this regard, and draws all power from the USB cable. The USB cable, while slightly heavier than the Chinese V2's cable, is however shorter at 0.9144 meters in length which may be a slight problem but in reality either device is likely to need a USB extension for "creative" placement of the device. Just be sure not to exceed 3 meters or you may find the signal degradation affecting the reliability of communications.

Starting the application in user mode was successful, in that it did not lock up or completely crash. I was however presented with the following control display:

And yes, the borders were extended into blank space. Hovering the mouse over, I found that it would highlight various pieces of text, such as this:

It's pretty clear there is a simple scaling problem going on. Thinking it might be related to the limited user mode I was running the application under, I restarted as an administrator with no improvement so I returned to restricted user mode despite the fact I had already ran the application with full privileges meant any potential damage was already done. This control system also had a toolbar icon, which was the same as the desktop icon so I won't show it here. It has the same functionality of "show" and "exit" that the Chinese V2 did, with the same "minimize to toolbox" behavior.

I thought perhaps the image scaling could be fixed with a data file change, so I reviewed the installation folder which was in the logical "C:\Program Files\Missile Launcher" folder, although given the steep competition in this arena I think it should have been "C:\Program Files\Dream Cheeky\Missile Launcher" to accommodate users with a broad collection of missile launchers from unrelated manufacturers. It's almost like people forget there is competition out there or something.

At any rate, I noticed the kind folks who wrote the software provided us with everything necessary to link the application together again after making any useful changes we might have to the .obj files. That would of course be a bit difficult without the .C/.CPP/.WTF files they actually wrote this software with, but not entirely out of the question for a determined programmer.

At this point, the device is connected, the UI is running albeit with scaling issues, and the missiles have been mounted upon what will apparently be pressurized launch rods. So I seek the hotspots where the arrow buttons flash, and press the mouse to initiate a spin. It moves one small amount, perhaps a few degrees which is a nice level of precision, but it requires a click per micro-movement. It has sounds of movement, which is a nice touch. Unfortunately, after just a few clicks I am presented with this display:

Being a programmer, I decide to look at the details because I can in fact sometimes solve these problems using that information. I see this:

Well, the szModName is "hungapp" which is itself surprise, most apps don't know they are hung until it's too late. The offset of 00000000 tells me it's a null pointer exception of some sort, and it's not likely I will be able to diagnose the problem without some substantial effort which I will not take the time to do. Perhaps Microsoft will notify Dream Cheeky about the crash report I submitted so they can fix it and notify me of an update via the online support.

Or, perhaps the crash has been fixed? I began looking for the online support mentioned on the box. There is no icon for a web site, and the software doesn't appear to have any mysterious hotspots that take me to a site before the application crashes, so I visit Google to see if perhaps the online support is well known enough for Google to know about it. In other words, if it exists now or sometime in the distant future, I should see it on Google. Happily, I discover that DreamCheeky.com does in fact exist and they have an update available! I find the download page, and it wasn't obvious at first my pop up blocker was preventing it from downloading but once I did I saw their download page, which looked exactly like this, blurs and all:

Thankfully, I already tested the installer and knew it would work. So I uninstalled, grabbed the files, and proceeded to insatll (sic) the new one. Excited, I fired it up and had some success in that the UI appeared at least scaled correctly so my hopes began to rise.

However, upon using the app it exhibited the same micro-movement and nearly immediate crashing behavior. Thinking it might be sound related, I disabled that to no avail.It's unfortunate the aiming mechanism was failing because that red dot in the UI actually indicated where the launcher was aiming and that kind of device feedback was exactly what I was hoping to see from the Chinese V2.

Despite the failure of the device to aim via the telemetry system, I had hoped I could aim it manually and at least utilize the launch button itself. When triggered, the sound effects of a launch and "the explosion" sound we all know from every game that ever used an explosion sound would soon follow. Unfortunately, no missiles would fire and the application would then hang. Some time after I gave up, I nearly jumped out of my seat when the device began moving around all on its own nearly 20 minutes after I last manipulated the UI. The program was still running, so perhaps it was in some kind of first time initialization where it finally finished whatever it has to initialize. Or so I hoped. Attempting to use the launcher after it stopped moving on its own proved fruitless, as the application was still hung.

In the end, despite a fantastic first impression, the great promises made by Dream Cheeky did not translate into an even salvageable foam missile defense system. I am disappointed, the pneumatic launch system appeared to be a safer solution and the projectiles looked like they would fly nicely with that extra weight on the tip.

So, without further ado, here are the final results for the...

USB Missile Launcher of the Year, 2006!

SECOND PRIZE:

Chinese V2 Foam 3-G Missile Defender

anonymous Chinese Manufacturer, circa 2002

This entry's most notable quality is that it actually worked. Despite the fact it could be transformed into a nail launcher, the software did not crash and if used as directed by documentation that did in fact exist the missiles would actually fire.

FIRST PRIZE:

Foam Sponge Tied to a Back Scratcher with a USB Cable and a Laser that has a Half Dead Battery

Eric Cosky, 2006

The simplicity of this design is pure and to the point. It has USB, it has foam, and the leverage of the stick makes it an effective missile with far more range than the Chinese V2. While it may not have the outward appearance of a missile launcher, the foam will in fact detach from the stick and become a missile when launched by the user. And it has a laser. In the end, this device will provide the most entertainment of all the foam missile contenders in the market today and so is awarded USB Missile Launcher of the Year, 2006.

Well, that takes us to the end of the first annual USB Foam Missile Launcher of the Year award. With luck, the next year will bring fantastic and interesting changes to this young industry which will propel an actual product to first place!

EDIT: It has come to my attention there are is in fact another competitor: The Striker II USB Laser Guided Missile Launcher. Had this "been on my radar" I would have included this in the contest and from the looks of it, it may have won so it is truly unfortunate that I heard about this one only too late.