Maybe I don't have to reinvent the wheel after all -- just found this!
Maybe I don't have to reinvent the wheel after all -- just found this!
Wow, that's nice, too bad its so freakin expensive. I definatly like the idea.
Heh heh, well I hope this could start a revolution! I admit to not knowing a whole lot about Windows hardware so I've been digging around and doing lots of research. It seems that there is just no way to plug in a less-than-VGA LCD screen to a PC and have it work. You must build some sort of proprietary hardware/software solution. That means building an embedded system -- just to output your work on a small screen. Yikes!
If this setup can allow you to make a screen-sized window and divide that window into "panes" and output each pane to a different LCD, that could have far-reaching implications IMO. Designers could deploy apps to small screens without the need to do anything outside the Windows environment. Right now, building an embedded system with a color GUI is a pretty complex and daunting project (not to mention expensive!). But, in the not-too-distant future, if Windows could run reliably on a single board computer with no moving parts and boot instantly, you could build a dedicated system using PC architecture instead. That could open up the world of design for embedded-like systems to enthusiasts and hobbyists instead of just hardware engineers.
Also imagine having a dedicated PC for your in-car entertainment with a main screen in the dash. Now add a second PC for rear passengers and each one can have their own touchscreen control. Or, use one system for all and allow the rear passengers to have their own mini control panel. I can think of lots of applications for this technology.
Yeah, cost is an issue. But as this becomes more widely available it might get cheaper. What I am hoping will happen is for them to produce some sort of "module" that will allow you to plug several different small LCD screens into a VGA connector. That gives some flexibility in choosing the right screen for the right application.
Also notice their screens are fully contained and hook up via an ethernet cable! That is perfect for a car environment. You could put your PC in the trunk and run ethernet cabling under the carpet to the dash. The LCDs are powered by the ethernet cables so all you have to do is figure out how to actually mount them in the dash. No other wiring is needed!
Without something like this, I simply cannot believe how hard it is to add a small color LCD to a system. It seems no one has produced any hardware or software to make it easier. You have to pick a screen and a controller and then become a microelectronics engineer to figure out how to make the controller work to put something on the screen. Unbelievable!
Well, I spoke with the president of the company and he answered all my questions. This was the missing link I needed for my project! I will start working on the design side and put up a page on the project when I have time (it will be between now and the end of the year before I can devote any serious time to it though). The design plan so far is as follows:
1. Microcontroller to interface the PC with all the inputs needed for the instrumentation. This will be modifiable using code.
2. "Generic" VB program to accept data from the microcontroller and display the gauges as well as process user inputs for controls.
3. Flash graphics and routines to actually draw the gauges.
4. A dedicated PC to run the system.
The instruments will have three screens. The main screen will display speed, tach, turn signals and certain indicators as well as messages to the driver including trip computer info. The left screen will be a gauges module and display fuel economy. The right screen will be a Vehicle Information Center with a diagram of the car, warnings and indicators, tire monitor for all four tires and a digital compass. The system will interface with a second CarPC running entertainment/GPS, etc. to display vehicle maintenance reminders and data as well as redundant trip information when the passengers need to see it.
The main carputer can have a backup speedo and other critical indicators in case the cluster PC crashes while driving. A touch of the screen and your info will be there until the main unit comes back online. This will cover the bases -- unless both units crash at the same time!
The design work for all this has been done for a while now, but building it will take a lot of work and time! :D
Originally Posted by star-art
Yes and no, Most smaller screens are for embedded stuff which don't really have a lot of horsepower to begin with so most of the stuff ends up being oriented for slow updates.
I don't see why you couldn't just get a quad vga (like a Matrox) and hook up 4 monitors to that. That would probably be ideal (and cheaper). That just leaves the small screens.
If you could get S-Video/composite output, it would be fine. No, it's not as sharp, but when a screen is only 2.5" to begin with, it doesn't matter much. For gauges, you need to be able to see it at a glance so the graphics relatively speaking, will be big.
Having said that, 640x480 via S-Video was suprisingly usable on a 2.5" screen.
Though I still think it's possible to get VGA going on the nitemax lcds.
Here's why I think this is ideal:
- They will have 1/4-VGA which is what I plan to use (1/8-VGA is pretty small).
- Their current screens are transflective!
- They will have high-bright screens soon.
- All I have to do is create an application window which takes up the entire desktop at 640X480. Then divide this into "panes" and each pane will appear on a different LCD automatically! You will be able to mix and match 1/4-VGA with 1/8-VGA.
- I can use the PC program to overlay VIDEO on any part of the window when I want to. I could, for example, have the center screen show the backup camera view when the car is put into reverse and each side screen show an outside "blindspot" view when the turn signal is activated. All I need to do is get the video to show up on the desktop -- no other wiring or programming is needed!
How much simpler can you get? :D
BTW, I've seen small displays (5 inch or so) that have a composite video input. But 1/4-VGA with a SuperVHS connection I have not yet seen. Still, good alternatives. :)
I agree with this approach. With a Windows extended desktop to 2 or more screens, the software would still control what is displayed in each monitor. With 1/4 and 1/8 VGA, not only are you more limited in your hardware selection, there is also less graphics resolution (splitting 640x480 four or eight ways). If I pursue this project for my own vehicle, I will probably integrate two 7" or 8" widescreen (800x600) LCD VGA monitors side by side in the space where my current instrument cluster is located. I would use a readily available VGA card (matrox or other) to drive two monitors, and write the software in VB. I will monitor the Class II datastream for all the existing gauge data, and I will have to integrate some individual DC inputs for the handful of discrete signals that are currently wired directly to my cluster. I think that larger monitors would allow for more programming options than small 2.5"-5" monitors would. The goal would be to leave the PC in hibernation or standby mode for minimal startup delay on ignition.Quote:
Originally Posted by shotgunefx
As a side note, I did find a source for very small LCDs: http://lcdtft.com/search_result.asp?...LCD%20Monitors if it is something that anyone would like to use. They are mostly composite video and are not the superbright type. Also, some of them are fairly pricey for their size.
I like the idea for the mini lcd's as well, it is a good alternative and I do hope the price comes down, but I don't see a point of having two Carputers, one for doing each entertainment and one for gauges. I could see using one as a backup unit incase the other crashes, but you should be able to do multiple monitor outputs, using the PCI card they offer as the primary display, and use the second monitor as a secondary, this gives you an extended desktop as well, and you can do your gps/entertainment on the second monitor without the need for a secondary cpu.
I'm building mine with three 7's just because I can and have the parts readily available. Keep in contact with your program though, it will be very easy to write something if you need help with let me know. The only thing I dont like is 512 colors..you'd have to stick to 256 colors on the desktop, which is very possible and doable, but, ick.
Again, I'll keep everyone updated on my progress, I've been painting parts all weekend, so hopefully soon I can start piecing the car back together.
For a progress report: http://mobilecpu.net:8080/installs/invadergt/
If compostie, SVIDEO is ok, pm Nitemax_Mark he had a ton of 2.5 LCDs with controller, I believe 2 for $25 :)
The only problem you might have is where to put the controller boards. Do a search on "nitemax" and you should find ton's of threads.
More good points. But I can't use large screens, though, so that is just not an option for me. I drive a big car, but my dash is not that big! ;)
Keep in mind I prefer to maintain a somewhat OEM appearance so I don't want a "radical" restyling to the interior. The central instrument stack going to LCD is pretty radical already (I'd like a two screen stack which replaces all the controls and buttons). I'd like my instrument cluster to resemble a modern digital dash which is more subdued -- classy and elegant but not "hip" or "trendy." I'm not a hotrodder! ;) For me, it should look slick but still resemble what the factory could have done if they had the technology (and the guts) to do it. The small screens will be perfect for this effect.
Personally, I don't like the idea of two screens which splits the display right down the middle, but that certainly works and is practical. If I had room for large screens, I'd be tempted to use three (one in the middle flanked by one on either side). But that is WAY too large for most modern car dashboards (trucks, maybe).
As far as a second PC, well the instrumentation is a mission-critical function. I am counting on the fact that hardware costs will continue to go down. Ideally I would like to see my instrumentation system deployed as a "dedicated" setup (like a present day embedded system would be) but one that runs on PC hardware -- perhaps a single-board-computer with "instant on" capability. We seem to be headed in that direction in terms of technology. Already you can build in a Windows visual environment and deploy to an SBC running WinCE that boots in mere seconds. That sort of system should get even easier to build and deploy as technology improves.
OK, that's my 32 cents for what it's worth! ;)