Monday, August 13, 2007
Building the Platform and MIP stand for the last couple of weeks
Here are a few pictures of the base platform..
Platform with the top board installed is a 8'x8' square constructed out of the standard 2"x10"x 8' wood. I wanted top base board to overhangs 1.5" all around over the base structure. Hence the outside dimensions of the base itself as you see in the top picture is 93"x93". Further, I have horizontally divided it into 4 sections with alternating widths of 30" and 18" respectively. So, looking at the picture above, the first section (on which the MIP rests) is 30" wide, the next one is 18" (yokes bay), next is 30" and the last is 18" again. These dimensions are carefully chosen since I simply use two 4'x8' baseboards ripped length wise into two pieces of 30" and 18" wide and 8' long which is also easy to transport in a van. Notice the overhang all around the platform in this picture. That allows easy folding of the carpet around it.
I have also built the MIP stand out of 1/2" thick MDF/plywood. Here are a few pictures..
Notice, a different approach to installing the monitors. I have seen many builders building the base to rest the monitors. Instead, I have cut holes in the MIP base to drop the monitors down. This way, MIP panel simply rests on the mip base and montor drops down to correct height. I found this technique simple and effective for LCD monitors. I have simply installed few wooden brackets to hold the monitors. Also, I have cut slots into the CDU bay stand walls to support the monitors on either side in addition to the installed brackets.
Here is the first ever test pictures showing the monitors with PM software running..Now I understand how gratifying it is to see these pics and what every other builder must have gone through..
Those are the Engravity's excellent MIP panels installed on a temporary backplate that I made for test fitting. I got them custom made in Boeing 777 brown color. I have also painted the entire MIP stand with a matching brown paint. My local HomeDepot did an excellent job of matching the paint with sample from the Engravity panel.
If you are wondering, that is a 737 MIP panel, you are correct. Given the difficulty in finding hardware for 777, I have changed my plans to build my aircraft with a mix of 737 and 777 parts. Even though the MIP is a 737, the critical MCP, EFIS panels are going to be 777 ones. When finished, I will be in a position to fly either a 777 or a 737 at will. I think this is a lot better situation than simply getting stuck with a single model pit. Also, a 737 MIP works better for me personally since it occupies lesser space than a 777 MIP.
Wednesday, June 13, 2007
Starting to build now..
After a frustrating year long ordeal, I called the basement finished and finally I am ready for building the sim. I have picked up a lot of momentum once again and recently purchased a lot of parts for the sim. I will post the pics real soon..
Oh! btw, I have built a really cool theater in my basement and check all the scoop on it here: http://srivijdreamtheater.blogspot.com/
Tuesday, August 15, 2006
I wrote a program to speech control the FS9 default ATC from a remote machine
However, it still seems like a distant dream specially now the MicroSoft team has confirmed that there are no major enhancements to the ATC system in the FSX release. That leaves me still waiting eagerly for the technology to catch up and allow the simmers to realistically 'converse' with the ATC programs.
I know, serious simmers use VATSIM for realistic ATC and not depend on the software simulated ATC. If you are flying 'all the time' with VATSIM then good for you. But being a novice, I still want to use either the default ATC or programs like RC. And the major problem with these programs is - they all use keyboard as their user interface.
I am tired of using the keyboard to select the ATC menu choices. Specially in a homebuilt hardware based cockpit environment, having to use keyboard to interact with the ATC is so unreal and annoyingly distracting.
So, I went back looking at options to use speech to control the FS9 ATC. I am aware some people use speech programs like 'voice buddy', 'Flight Assistant' or 'shout' to map their voice commands to ATC menu keyboard keys. While this works ok, I have an additional important requirement that made me not use any of these known techniques. As far as I know, all the above programs require you to run them on the FS9 computer hence potentially impacting it's framerate.
So, my key requirement is - I want my speech recognition process to run on a remote computer and not on the FS9 computer.
As we all are well aware of to what extents a simmer would go for the ever elusive framerate nirvana :-), I ended up writing a program that uses the WideFS/FSUIPC to communicate and control the FS9 default ATC. I now can make ATC menu selections completely hands-free through voice. The freedom I gained is awesome and the command recognition is 100%. I can't be more happy.
Currently, I would say "Traffic Control" to either 'bring up' or 'Close' (useful to quickly get the atc transparent menu out of your way) the ATC window. Once the ATC window shows up, I would say "select one" (or two,threee,four..) to select the appropriate menu item.
Ofcourse, I know it doesn't sound realistic ATC communication. I thought of creating somekind of smart menu where instead of saying "select one", you would be able to say generic command like "contact ground control". But unfortunately I couldnt implement it due the fact that currently FS9 doesn't expose any of their ATC menu text details via offsets (confirmed by peter dowson) and there is no ATC api available either to hookup. Also, I wish I could display the ATC menu window on the remote machine rather than on the FS9 main display as it does now by default. But here also unfortunately I couldn't do much due to the same FS9 constraints.
But in the end, I am fully satisfied with having some kind of speech interface that I could run on a remote machine. My ATC experience is way better now than earlier when I had to deal with the keyboard. And I am thrilled with the reliability of the command recognition, which is 100%.
My next wish is..to add similar capability to the Radar Contact (a third part ATC add-on) if I end up owning a copy of it. I know RC currently suffers from the same draw back of keyboard as it's only user interface.
If you are interested in my program, drop me a line at: "projsri777 AT yahoo dot com".
Thursday, July 20, 2006
What is going on?
Answer is - A LOT!!
A lot going on with my life and I am busy on all fronts. Did not gather enough energy to come back and update this blog. A quick disclaimer - No, I haven't lost interest in pit building and simming (!!) On the contrary, it grew and pushing me to start building soon.
As a first step, my basement finish work has started and ongoing for the last month and half at a slower than anticipated pace. Part of the reason is - I have added another ambitious project to build a proper Home Theater also in the basement. When finished, it will be a room to watch movies on large screen. That is the deal I made with my sweet heart for building the sim in the basement and potentially spend all my free time on the sim. She hatched this plan to build the Home Theater, so that I come back to watching movies together with her :-) But anyway, now the HT theater project grew on me and I am all over it researching and building a properly sound treated theater room will all the bells and whistles I could afford.
Meanwhile, on the simulator front, I have built two more computers and bought PMDG 737 aircraft (detailed reasons later..), and FSBuild software for flight planning. Also, during the July 4th weekend, I went to Denver and attended the International Flight Sim conference. I will probably write about it in some other post.
So, I have lot more to write about including the cool xbox stuff I am doing right now and make few announcements :-) stay tuned (in a hurry to go to work right now..)
Sunday, October 09, 2005
Tips & Tricks - I
During this process, to my frustration, I realized that I forgot many simple things that I originally took trouble to figure out the very first time. So, this time, i decided to document these here as Tips & Tricks for simmers. Once again, this is to help me as much as it might others. So, here is the volume I of my tips and tricks. Let's see how this list evolves over a period of time..
1) To make the FS9 Flight Plans available in CDU..
a) Share the directory into which FS9 saves your flight plan files as "FSPLANS" with full read/write access enabled.
Typically this directory is:
C:\Documents and Settings\
b) Share the main Flight simulator program directory as "FS9" with full read/write access
1) Where is my FS9.cfg file?
C:\Documents and Settings\
2) Where is "no CD FS9" hack?
Not sure i can mention it here. But i know where it is :-).
If you don't know what is "no CD Fs9" hack means.. It allows to launch and play FS9 without having to insert the CD #4 into the cdrom!!
3) If you have added FSSound.dll to FS9 either knowingly or otherwise while installing your favourite aircraft/gauges and suddenly started getting the invalid dll message during FS9 startup..
then add the following lines to the bottom of your FS9.cfg file:
This dll is perfectly ok to user for FS9. The above lines will suppress the annoying startup warning.
Meljet Aircraft model
In the 777 VC mode, pilot side view is blocked and cannot see outside..
modify the eyepoing value in the aircraft.cfg file to the following:
eyepoint = 92.0, -2.65, 2.25
If the above didn't fix the problem, try changing the first value from 92 to 96 in the increments of 1 until it works..
Thursday, October 06, 2005
My CDU is ready
I also purchased the PSOne display holder alongwith the CDU so that I can secure the display in place easily. Notice the red insulation tape strips on either side to prevent the metal frame on the display to short the soldered pin connections on the back of the CDU.
The cool part is the default backlighting for the keys. Night time it is looking very impressive.
And btw, last time when i visited James, he generously gave me his old throttle and said - just play around until you get your own. It is quite unexpected and i was shocked at his kindness. Here it is.. I hooked it up to my computer and configured it using FSUIPC. And the throttles are working just fine. I used the console to temporarily mount my new CDU. Here is how it looks on James's console..
Friday, September 30, 2005
My first encounter with a home built 737 Sim
On Augutst 27th, I drove approximately 3.5 hrs each way to visit a fellow new england builder's home to see his 737 sim.
First, James..a big thank you for letting me have the time of my life and for your kind hospitality. You are amazingly kind to spend a lot of time with a total stranger and at the same time making me feel at home all the time.
After imagining how it looks for so long, finally it was exciting to see it in real. In terms of reality, it is more than what i imagined (see the pic for yourself). I jumped into the pilot seat and we tried a brief flight. But since i am not apt at flying, we didn't do much. James has built throttles all by himself and they were really cool and worked great. Overall, the cockpit looks so real, that it is surreal.
I stayed until late into the night and the long drive was worth every minute in my book. After visiting James's, i am eager to start building mine soon.
Saturday, April 02, 2005
The first year in review
Did I complete my sim in this one year? No, I haven't even started building yet.
Even though i knew a year back, that this is going to take time, i now have a clear idea of what lies ahead and what to expect realistically.
Besided hours and hours of research and reading posts on various newsgroups, here are my major accomplishments in the last one year:
- built the computer(s) and setup the network
- getting better at flying
- selected IOCards as the hardware interface system and built the cards
- bought Project Magenta, set it up and started climbing the steep learning curve
- started programming the iocards for various panels
- met some great, helpful builders online and made some good friends across the globe.
The best reward in this hobby, surprisingly came in the form of the last one in the above list, totally unexpectedly. I have never met so many helpful, generous and contagiously enthusiastic souls in my life before, than I did in this last one year. It almost feels like membership in an exclusive global club, where almost everyone knows the other. Major part of my learning came from these fellow builders, who happily shared their treasured collections, years of their research information, all just for simply asking (in somecases even without asking!!). A big thanks to you all.
In particular, i cannot miss this opportunity to thank the following folks individually.
- Chris W Balmer - for kindly sharing your docs
- Daniel, Tegelen - for the incredible information on the shell construction
- Hans Schuetz - for all your help and importantly introducing Klaus :-)
- Johan Nauwelaertz de Agé - for patiently explaining and helping me with the pedals design
- Klaus Troemer - thanks my friend for being the kind soul you are
- Thomas Konditi - still can't believe you took the trouble to setup a ftp site, just so that i can download information..
- Matthias Spahrmann - the very first builder friend who taught me a lot patiently
- Manuel Vélez - amazing example of how much difference one person can make in this world. Thanks for your iocards project and your relentless, day-in day-out support to builders all across the globe.
All of you made me a better person. You made me wanting to share more and help other builders in anyways i can. Thank you.
Last but not the least, this hobby demanded whatever the little free time, if at all any, that i have in my busy work life. That, need for time, is the hard part in this hobby. And it was harder on my sweet wife Vijaya than it was for me. Thanks Chinna, for all your understanding and support. You even learned how to solder!! and helped me with building the interface cards, just to spend more time together!! Thank you Chinna.
Looking into the future year, hopefully, i get to finish some area in my basement so that i can start the construction. One good improvement in this one year as i noticed is - panels market is further maturing and improved panels from FDS and others are coming into the market. So, currently i am not minding the wait.
In all, it was an intese research/learning phase of the project. So the new builders out there - be prepared for a looooooooong time before you see an AirBus or Boeing parked in your basement!!
See you all here with a similar report in a short year!!
Sunday, March 27, 2005
GenEx - My opensource contribution to IOCards Project
I named this program "GenEx" as it 'generates' SIOC programs from simple 'Excel' spread sheets. Users enter their iocards setup information(their input output connection numbers) in a spread sheet against a pre-identified function names. For example, this spread sheet contains a list of all functions such as IN and HPA selector or ND range 10,20,40 selector etc found on EFIS panel, one functin per row. For each function, users are expected to enter thier iocards connection number. For example, if you have connected ND range 10 selection switch to iocard input #32, then simply enter 32 in the cell against the function named 'ND range 10' in the spread sheet. That's it. And run the 'GenEx' against this spread sheet now, and GenEx generates a complete working SIOC program that works out of box with Project Magenta. No more need for knowing the details of IOCards programming or FSUIPC offsets.
A couple of following things motivated me to write this program:
First, when i started to write sioc program for my 777 efis panel, i realized that i was combing through a lot of example files from other users and also pouring over the project magenta offsets documentation. And then it struck me that many other iocard users before and following me will have to do the same. But in reality, there is no reason for that, since only difference between one users program and another's is really the connections information.
Second, I noticed the major trend in commercial interface cards vendor space with general curiosity. Most of their claim for charging premium is their software's user friendly claim of - "just connect, assign position numbers and go running without worrying about fsuipc offsets". It seemed, this kind of easy up and running nature is desirable for many builders and they are paying comparatively big bucks for these commercial solutions. While projects like IOCards are without doubt - better value and functionally competitive, if not better than any of these commercial solutions, but they still require some level of programming competency and not as user friendly as these other commercial solutions calaim themselves to be. So, future focus for these free projects such as IOCards should be of making them much more user friendly than they are today.
Third, i watched Project Magenta's PMSystems marketing video and one sentence Jonathan said caught my attention. While refering to Phidgets, he said - "programming with offsets is fast becoming a thing of the past". I thought - "yes, it can be done with IOCards too".
And finally, right when i thought no one would actively wish to distance away themselves from the programming fun (i obviously assumed that everyone would be dying to program their interface cards, and they'll laugh at me if i tell them to just forget about all the offsets and enter their connection numbers into a spread sheet), one of my builder friends told me that he is considering a more expensive commercial solution for precisely the same reason. Which made me believe that GenEx has a need and so then i decided to code it.
So, as a first step, in this regard, i thought with GenEx I can address all the above issues. It makes the IOCards simple by a) eliminating the need to learn and worry about SIOC programming b) no need to deal with Project Magenta/Fsuipc offsets and c) brings IOCards one step closer to commercial vendors claims.
Also, one more bonus advantage of using GenEx to generate your SIOC programs is - the excel file (that is used for generating the sioc program file) is also the automatic documentation of your cockpit project, since it neatly captures the information about every input/output connection on your iocards and it's assigned function.
So, by now if you are wondering what stage the program is? It's coding is complete in all aspects, except that it needs sioc program templates for various aircrafts and panels. Currently, it has template for generating sioc program for B-777 EFIS panel only. Adding these templates to cover more types and panels is external to the program code and will be taken up in the next couple of weeks.
I have sent the beta copy to IOCards team to evaluate and comment. Hopefully they'll accept the program and make it officially a part of IOCards project. If they like it, the program can be further improved. Currently it generates SIOC program file as ASCII text file. This generated file still needs to be compiled with the regular Config_sioc.exe compiler. This extra step can be eliminated if the sioc program compiler is also integrated into GenEx. Then it can directly generate the '.ssi' file directly from the input excel file. Sweet! I am only hoping IOCards project folks bless this.
So, now imagine - all you need to do is: connect your panels hardware to input/output ports on the iocards cards. Then open the included excel file and just type these connection numbers in the well identified cells and save. Run GenEx against the excel file. And you are ready to be up and running with Project Magenta. Sounds exactly like one of the commercial card vendor advertisments correct? But this time it's free, and it's IOCards.
Finally, I am glad that i am able to contribute something back to the IOCards project team whom I respect for what they gave us all for free.
Saturday, December 04, 2004
Modding PsOne 5" TFT display screen for CDU
There are detailed guides online explaining how to do the mod and make it work with normal computer VGA output.
I bought a Matrox Mystique graphics card (again on ebay) and tried to make it work with the VGA output. Spent long time messing with the Powerstrip settings but without any luck.
Frustrated, i finally decided to go with the alternative - connecting the psone unit directly to the video card's video out (s-video) port. So i used my other nVidia video card that has TvOut option, and connected it directly to the psone unit and voila...it works beautifully. Surprisingly the resolution is adequate and the display looks good even at an angle. (see the picture below).
However, one disadvantage with TvOut is - some video cards (including mine) don't fill the entire screen, instead they leave about half an inch border all around the image. This won't be good for CDU as you can imagine. But fortunately there is a solution (as always). TvTool software can 'overscan' and fill the image entire screen. Read this nvidia explanation if you want to know more about this.
I used Ctrl+cursor command to adjust the Project Magenta's image location and then Alt+cursor to adjust the text position. (Note: cursor mean up/down arrows keys on your keyboard).
And at the end, i have great looking 5" Project Magenta's CDU image on a sony PsOne screen. I am ready for rest of the hardware around it now...
Here is the image at an angle...
CDU on PsOne screen via TvOut of Video card
Below, you can also see the back of a stripped PsOne unit...
Sunday, October 24, 2004
IOCards - SIOC and IOCP
- Making steady progress with flying in the sim. Learning navigation and using FMC (CDU).
- Bought 2 PSOne lcd screens (more on this later) on ebay as an advance preparation for building CDUs. I will be hacking them later to accept VGA signal. Expecting the delivery tomorrow.
Now to the main topic.
This week, came a long way in my understanding about IOCards SIOC and IOCP software. Information about these is scattered and not concisely described in any single place. Unfortunately, the english SIOC user documentation doesn't give a good overview of the landscape. Hopefully, this summary note gets you off a good start.
Though IOCP and SIOC can also talk to XPlane and other dedicated IOCP servers, this discussion focuses on FS9 only.
First, IOCP is a communication protocol (not an application) for efficient network communication between IOCP clients and IOCP Server components. if you know IOCP protocol details, you can write your own IOCP client and connect to IOCP servers such as IOCP.exe or IOCPServer.dll).
On the other hand, SIOC is an application suite built using IOCP protocol. It contains both an IOCP Client module (besides other type of client modules) and an IOCP Server component. You can use it for:
a) controlling IOCards hardware using powerful scripting environment
b) communicating with simulator software such as FS9 (via IOCPServer.dll), X-plane and OnBoeing-737 using IOCP protocol
c) communicating the IOCards hardware status with FSUIPC dll using non-IOCP protocol.
Before we go into the details of sioc, let me clarify the confusion around sioc and IOCPServer.dll. IOCPServer.dll is a special purpose FS9 add-on module that gets installed into FS9 similar to the way FSUIPC is installed. It does the job similar to FSUIPC. However, it is also a IOCP Server. So IOCP client (such as SIOC's IOCP Client module) can connect to it while it is loaded into FS9 and communicate with it. For simple understanding, IOCPServer.dll is a combination of FSUIPC and "WideFS server" combined together in one.
The main sioc program "sioc.exe" can be thought of as both a IOCP server and sometimes as client as well. It contains the following 3 major client modules:
1) An IOCP client module that connects to another IOCP server available anywhere on the network. Two such servers that this module can connect to are a) another instance of SIOC.exe on the network or b) a special purpose IOCP server such as flight simulator's add-on module called IOCPserver.dll which does the same job as FSUIPC)
2) other special client module (doesn't use IOCP) that connects to the IOCards hardware installed on the same host and help control the hardware
3) a special FSUIPC client module that connects to FSUIPC.dll (installed inside FS9 and running anywhere on the network).
to be completed....
Tuesday, September 14, 2004
Testing IOCards - Master card /USB Card
As you might notice and as I mentioned earlier (while comparing various controller hardware), the major disadvantage with IoCards is the lack of strong user community support. I felt it hard while resolving the problems i ran into in the last couple of days. Undoubtedly, it is an excellent project, I only wish there is much more strong community around it.
There seem to be no definative way to test the USB card alone. After you hook up master card, test the input connectors (J3 and J4) by shorting the pins (read the annex-I doc) and make sure all inputs are showing up in the controller software.
Monday, September 13, 2004
Package from IOCards folks in Spain arrived in about 8 days (btw, i am in US). The PCBs are of excellent industrial quality, each component location well identified with clean legends making it really easy to solder the components. I had to rarely refer to the scheme diagram. Refer to the IOCards website for further information such as components lists and scheme diagrams etc.,
First of all thanks to IOCards folks for their excellent work and providing a great service by supplying these kits at a very reasonable price.
Few things before you start building the cards:
1) Needless to say - get yourself a good multimeter. I bought mine from a local Radioshack store ( http://www.radioshack.com/product.asp?catalog%5Fname=CTLG&category%5Fname=CTLG%5F011%5F008%5F002%5F000&product%5Fid=22%2D813 ).
2) IOCards kits does not include IC sockets. Get yourself a couple of these sockets as listed to plugin the ICs inlcuded in the kit:
For USB expansion card: 16 pin (2 nos), 20 pin (8 nos).
For Master card: 16 pin (8 nos), 24 pin (1 no)
For Displays card: 16 pin (4 nos), 20 pin (1 no), 24 pin (1 no)
One of the sockets (24 pin) i have ordered online turned out to be wider than the distance between the two rows (width of the ic)on the pcb. I solved the problem by breaking the socket in the middle into two seperate columns and soldered each seperately.
3) PCB mounting screws - one each on the four corners of USB and master card
I haven't had all these ready at the time i sat down to build these cards and local radioshack stores did not have all of them available. I had to go to 3 different radioshack stores and still did not find all of them. Finally ended up waiting for one more week and ordered them online at Mouser ( http://www.mouser.com ).
Things you need after you built these 3 cards:
1) a 25 pin Male to Female cable (DB25 M to F) to connect master card with the USB card
2) a 40 conductor cable (a flat ribbon cable is preferable) to connect 40 male IDC socket on the master card to the similar one on the displays card.
3) A steady 5V dc power supply (preferably from a PC powersupply).
As mentioned, soldering the components is quite straight forward. All component locations on the pcb are clearly marked.
Only issue that one will run into is with the USB Kit. USB kit as shipped has missing part: 220 nF Ceramic Capacitor (C1). Instead kit includes an extra 0.1 mF capacitor. This is ok and is due to a later design change. Go ahead and use 0.1 mF capacitor inplace of the listed 220 nF for C1.
POWER SUPPLY FOR MASTER CARD
Though USB expansion card is the one connected to the computer, it doesnot need external powersupply. However each master card connected to it requires a steady 5V DC power input. The best steady source is a PC power supply unit. I solvaged one from an old PC given to me kindly by my friend Prasad (Thanks Prasad). You can find a tutorial on how to do this on IOCards website or here: http://web2.murraystate.edu/andy.batts/ps/PowerSupply.htm
tips for building the cards:
1) Have a tool to cut the long component leads on the back of the pcb. You will have to cut a lots of them specially on master card. Having a proper tool is critical.
2) Do not use a high wattage solder. Mine is 15 watt type.
3) Do not weld ICs directly to the board. Use sockets.
The lessons i learned from this very first milestone activity is - this is going to be a long term activity and i need to learn to be patient. If a part or two is missing, i need to take it easy and learn to wait until the next weekend to get it from online stores. Impulsively buying from a local store may not be possible and/or it is going to cost you a lot more. So plan much in advance or take it easy and wait. For example, price of a 6ft 25db cable in local compUSA is around $20.00 and a 10ft cable of same type online is $5.00 approximately.
Oh btw, i cannot forget to thank my wife vijju for helping me with building these cards. She sat with me, learned how to solder and built the displays card all by herself. Thanks chinna.
Next week, i will be hooking up the boards with the IOCards software and start testing them....
Back to business
Meanwhile, i have downloaded the Project Magenta demo and played with it a little bit. Doing a lot of research on PC hardware as i immediately need computers to run the FS9 (Flightsimulator 2004) and PM (Project Magenta).
Also, there is competition began in the panels offerings and two new players showed up in the last couple of months. They are: Robert Prather's Simwerx (http://www.simwerx.com/) and soon to be opened (may be by next week) DakenSkys (http://www.dakenskys.com). FlightDeck Solutions (http://www.flightdecksolutions.com) still the leader but definately worth watching the competition - specially Dakenskys. Dakenskys panels seem to be made better quality material at competitive prices. So, as i am not in a big hurry - i am going to wait until i see the full line of offering from Dakenskys.
So, my short term plan is to make the iocards working and hook it up and experiment in a test mode. Next build the computers (at least two) and then move onto the panels.
Saturday, April 17, 2004
I made my first purchase towards the sim last week. I bought two second hand, viewsonic professional series (P95f), 19" CRT monitors from local supplier for $175.00 each. These monitors are of excellent quality and should serve well as the PFD/ND display screens for pilot and FO. Currentlyy I am doing a lot of research on computer hardware configuration. At somepoint, i will post my configuration information here. My goal is to make all my research information available online to the last nut and bolt. I have few computers (5 in total) and i have to plan the new ones taking what i have already into consideration. One of the things i am considering doing is to convert/build all of them into rackmount systems and mount entire system into a single rack.
more on computer hardware and networking later....
Wednesday, March 31, 2004
Even before the construction of our new home two years ago, i always dreamed about an integrated, networked multimedia system in our house and thought a lot about it. As a first step, during construction, i have installed a whole house audio system which will allow us to access and control music from every where in the house. But I always wanted and on the look out for a system that will allow me to access all forms of digital media (mainly audio (MP3, WAV, Real, Windows media), video, digital photos and internet audio steream) directly from the family room. As part of the plan, i turned my age old dell 64 MB PII computer into a Linux based file server in the basement and started ripping my CD collection onto the 120 gig hard drive that i installed on it. But in my search for a networked media server, i found that the market is not yet mature and the first generation consumer systems that are coming out are either expensive and still limited in features (for example they will handle only a limited media formats). Some mature systems like Audiotron are audio only and doesn't handle video and photos etc., At one point i was about to buy audiotron, but then enter Xbox....
Xbox hardware packed under the hood is really a powerful computer with excellent video and audio handling capabilities and can be used for more than just playing games. But naturally, microsoft doesn't want you to do that (instead, they wish you buy XP-Media version) and made sure that they locked the xbox hardware such that unofficial apps will not run on the box. But then some curious mind at MIT hacked the hardware and found a way to bypass the default microsoft bios by installing a mod chip. Once you install a mod chip and flash it with a hacked bios, your xbox turns into a powerful low cost computer and you can install a variety of apps or even an xbox version of Linux on it!!
The hardest part of the xbox mod is soldering the mod chip, in particular the dreaded d0 point. The moment i looked at the motherboard and the tiny d0 point that i needed to solder, my heart sank with fear. I remained staring at the soldering points through a magnifying glass for a looong time. To make it worst, this is my first real soldering job, though as a kid i might have soldered one or two wires. Installing rest of the modchip hardware such as flash programmer and jumper switches etc., turned out to be trickier than i thought. It took me couple of hours to finish the job carefully. And i was not sure whether it will work or not until i flashed the newly installed chip and switched the power on. But anyway, finally it was all successful and i can't wait to hook it up to my network.
Next day morning, when i first connected to the xbox over the network from my office room via ftp, it was exciting. Next couple of hours i spent configuring samba on my linux box as well as my other xp computer to connect to the xbox. Once that is done, I hooked up xbox to the family room AC3 receiver and i am ready to stream all my digital media (audio, video, photos) directly into the family room from my basement file server and another computer in my office room.
Trust me, having all photographs in the digital form and accessible at the press of a button is going to make a huge difference in the way and frequency at which you watch your long forgotten photos. I watched many old photos that were lying on my computer for a long time in a slide show mode while music being played in the background. Ofcourse, the quality of pictures on a regular tv tube is no where near watching them on a computer monitor (you can guess, my next wish is a large LCD or Plasma TV :-)
Another really cool feature with this setup is, i now am able to connect to internet broadcast stream (at present Shoutcast streams only) and listen to internet radio. Once again, quality of your audio will ofcourse depend on the bit rate of the broadcast stream. A 128 kbps station will sound way better than a 24 kbps stream.
Ofcourse, one other thing i have in my mind for this project is to write a wireless remote controller application to run on a 802.11 enabled pocketPC to control the media player. It will be a cool little .net app. Currently it is not high on my priority though.
Besides all the fun, this project gave me a swell in my confidence to handle other flightsim PCB soldering jobs.
back to flightsim...
Friday, March 26, 2004
It is going to be ..IOCARDS. Not FSBUS for control system...
Last week i thought it is definately going to be FSBUS. But later on the weekend and during this week, i researched a lot about IOCARDS project , read all the manuals, played with the software and decided to go with it instead of FSBUS.
Both FSBUS and IOCARDS are more similar than they are different. Both are:
a) non-commercial projects, making the technology available for free
b) modular and extendable
c) support various kinds of inputs/outputs
d) doesn't require programming but provide api if anyone wants to code
For my detailed comparison spread sheet, you can contact me at: email@example.com
and the major difference is in the technology they use. While FSBUS uses the more latest and popular 'microcontrollers' based solution, IOCARDS main board is based on slightly complex ICs solution. However, IOCARDS uses USB interface for communication while FSBUS uses serial port. Finally, each one has it's one strengths and weaknesses. So you wouldn't go wrong going with either one.
My reason for choosing IOCARDS is not necessarily technical merits as there is not much to differentiate strongly there. IOCARDS while making the pcb designs available for anyone to make them on their own, also ships the PCB hardware at a very nominal, reasonable price. I failed to locate a PCB manufacturer who would make them for me in US. So with IOCARDS, i don't have that problem. Also, in my opinion, IOCARDS project is better documented (compared to poor to none documentation of FSBUS). I was also impressed with their software suite which consists of 4 different programs allowing easy simulation, control, identification of connected devices. I thought the software is very user friendly and make the life easier for the builder.
On the other hand, IOCARDS is a spanish project, that is not as popular as FSBUS which seem to have a wider user base. I posted a lot of questions in the forum on their website and always got the response back (though from the administrator and not much user interaction). Neverthless, i got my questions answered. It may be a risk going with a relatively less popular project. But the fact is i liked it better than FSBUS and actually noticed couple of FSBUS users switching to IOCARDS at a later date. So i am going to go with IOCARDS.
I received my xbox mod chip in the mail yesterday. I am all set for working on my xbox mod this weekend. I am crossing my fingures as this is going to be my first soldering job and i am starting out with a complex assignment.
Friday, March 19, 2004
This week...I reconsidered doing this
All this week long, i was reading a lot and doing my research. I sense that it is going to take atleast a couple of months of planning before actually start building the pit. This week i focused mainly on deciding the control system i want to go with and did considerable amount of research on this.
The choices came down to:
1) EPIC -- the mother of all
With EPIC, issue is the price. It seems to be the most versatile, flexible and powerful system but expensive. Requires a fair amount of custom programming in C like language, which i am not worried about. However, i decided not to invest a lot of money upfront in a system like this.
Looks like a winner. All the information (including the circuit designs) available online for free. However, you need to build these boards yourself which requires fair amount of electronics knowledge. This could be the cheapest solution amontg all. Major drawback is for help, you are on your own or dependent on online communities.
Relatively new to the market. With the available support, it seems to be a fair balance between price and ease of use. Probably very good for someone looking for quickly up and running without much hassels.
Easy, quick keyboard emulator based solution. Limited for future expanding needs and one may not learn much. Good for someone with limited programming and electronics knowledge and don't care to learn either. All you want is to have a cockpit and quickly get to flying, this is the one.
This is only a quick and dirty summary. I have actually compiled a detailed feature by feature comparison of all these solutions. You may contact me if you want this excel file and i'll email it.
Anyway, in my final analysis, I am inclining to go with FSBUS for two reasons:
First it will be less expensive but more importantly it allows me to learn electronics and data communication. For making this decision, at the end, i asked myself - what is my goal for this project, what do i want to really accomplish. The came out - learn as much as possible besides having a sim to enjoy realistic flying experience. I was almost decided to go with Photon but ruled out against it in the final analysis, since both Photon and Hagstrom doesn't make me learn much. They will be quick point solutions.
Also, with less money invested in FSBUS(compared to others) and armed with all the knowledge i gain from struggling to make it work, i can easily venture out and move to another system in future, fairly easily, without much trouble. This is the most important reason why i chose to pursue FSBUS. Inorder for making me to switch the system in future, i plan to build the control interfaces in a modular and easy to disassemble fashion. This mean, no hardwiring of components to circuit boards. Instead i plan to use connectors for all connections to FSBUS boards.
It took me a week of research to come to this decision.
I considered making the PCBs myself but ruled out since having a professional good quality boards for starters seemed like a good idea. So the next step is to locate a PCB manufacturer who will make the FSBUS boards at a reasonable price.
meanwhile, i have a Xbox now and i need to work on the long awaited xbox mod project next week. I am planning to mod xbox and make it a networked media player streaming all my audio and video files over the network from my basement computer to the family room entertainment center. I tried to play a game on the xbox and it didn't interest me at all..i was bored in the first fifteen minutes.
Thursday, March 11, 2004
Why am I writing this?
Ofcourse my motivation is selfish - i don't trust my memory well enogh to feel confident that i can relive the pains and joys that i needed to go through this project at a later date.
Also, my inspiration came from visiting other's websites and watching what they have done. So hopefully, someday i too might be able to inspire someone to jump onto this wonderful hobby.
Another important reason is not just motivate but provide them with a detailed trial map of each step that a novice like me has to take for a project like this. All the websites i visited so far have wonderful pictures and some tips and limited guidence on how to build a sim. But not a single one i could locate that has the detailed account of what one has to go through. How and what decisions have to be made along the way? What their pains and triumphs along the way? what worked? what failed? and why? No one site has this information available which i am badly looking for at this stage.
I have so many questions starting out when i look at the pictures of someone's wonderful sim on the web. How did they build the projection system? But that is only the front view, what about the side views? what did they do - are they using monitors for the sides and a projector only for the front view? - You get the idea.
So i thought, recording a novices journey will help others who are not already somehow hooked into the industry....
P.S. I am not a pro who is trying to build a sim, instead i am starter and a student.. a good one. I am here to learn and hopefully someday teach! So if you read these logs and find me struggling with a problem, a question, please jump in and post a reply if you can. Even if you have no answers, a simple word of encouragement is also welcome.
You may also write to me at firstname.lastname@example.org
CH Pro Yoke (USB)
btw, buy.com was excellent in shipping my order. I ordered it on monday morning and tuesday by 9.45 AM. it was at my doorsteps (not bad for free shipping!!)
As soon as i hooked it in, XP recognized it instantaneously. However, note this, it is not caliberated correctly. Also you cannot use default XP's general calibration interface to caliberate this yoke properly.
Instead download the latest Calibration Manager software (it is v3.0 for me) from CH Products website. Do not use the software from the included CD (toss out the CD).
Anyway, Caliberation manager installs bunch of drivers and wait patiently meanwhile. Once installed, use this software to caliberate the yoke properly. One of the dilemma for me is how to caliberate the throttle lever (as a result of not having flown a real aircraft)? I wondered, when is the max power setting - when throttle lever is pushed away from you or is it when the lever is pulled towards you? Finally, i figured that it is full when you push the lever away from you. Another small detail i have no answer yet is - there is a small turn wheel on the lower left side of the housing. I wondered what it is, there is no calibration available for it. Probably i can map it for pitch trimming in the MSFS. Currently i did not map it to any. I left the default mappings in MSFS as is which i found are convenient enough.
Until now i was using MS Sidewinder 3D Pro. What a difference. Using Yoke, it felt much more real. It always centered correctly (only after caliberating ofcourse!)
I played enough last night to appreciate the yoke.
Even if you are not planning to build a simulator, but if you play MSFS, i highly recommend you treat yourself with this yoke. Once again, thanks to all those who influenced my decision by posting their comments/reviews.
Wednesday, March 10, 2004
I am not sure yet abot the exact type of the aircraft. I still need to do more research. Main criteria would be the simplicity of the control panel and ease of duplication. Also, more importantly availability of the software and hardware supporting the choice aircraft. Currently, i am assuming it coule be 747-400. Once again, i am prepared for a change from the begining.
So with my initial excitement and my nervousness about my construction skills(structures, housing etc,), i thought i need to start with building a platform and seriously started thinking about AutoCAD, materials like epoxy coatings etc., reference plans etc.,
Then i took a deep breath, stepped back and thought over about my general approach and these are the steps i feel comfortable with:
1) Master the critical software pieces and work on inter-connecting them
2) Experiment with the control hardware(both analog, digital and mechanical ones)
3) Interface control hardware and the software
4) Build each system and interface with the simulator software
5) Plan and build the minimum necessary computers and related hardware (no emphasis on the visual part yet. Just a simple monitor would do)
Ones all electronics, software and control hardware is connected and available in modular fashion, then start housing them and put them into the final structure and add big bang visual elements.
So, once again taking cues from my professional life, i am going to prototype first at each phase and only then move towards a final version even that means some throw-away work (nothing is throw away, you learn a great deal from your failures).
I think i already know what my critical software components are:
2) Project Magenta (critical for the glass-cockpit look)/FSUIPC
3) EPIC board (for interfacing all the hardware with the sim software) programming (also some tool like EpicMapper (http://www.epicmapper.com) might help.
One issue in selecting MSFS'04 is - it seems not all other SW/HW seem to support it yet. Many of them are still on 02 version. But i am counting by the time i am ready, they all will support 04.
So, for the initial phase of my project, my goal is to learn these pieces as thoroughly as possible. Which calls for some prototyping. So during this phase i am not going to bother about hardware and systems, instead i am going to run MSFS and Project Magenta at the most on two seperate systems and learn as much i can.
Also, i might go ahead and get an EPIC board and do small experiments with it such as:
a) connect my old sidewinder joystick through epic board and control MSFS and later switch to connecting CH Pro Yoke.
b) connect a switch through the board and be able to perform an action in MSFS such as turning aircraft engines on/off or switch on the cabin lights.
for some of my friends walking into my home office, this might look like a small step, but hey i know and you know that it's a giant step for the "Project Sri 777"
Oh btw, yoke that i ordered from buy.com has already arrived yesterday and i missed the delivery. They delivered it today and i am going to open the box tonight. Yeah!
Questions...Questions...More Questions and some decisions...
- how much it is going to cost?
- how long is it going to take?
- where to build?
- what skills i need?
- do i need any help?
- what is the right scope of the project?
I am full of questions that i need to answer to myself.
Project Cost & Duration:
About the cost and time, i realized that this is going to be an expesive and time taking hobby. But hey, what's the use of money if it can't help you follow your heart? So the best way i am going to handle it is by planning it carefully and building it over years (that is what it took for many people on the web who have done this before me). After careful consideration, the best approach for me seem to be pre-allocating X dollars per annum and spend it without worrying every time i have to order something.
Ofcourse basement is going to be the new home for the project. However, one big problem..i live in NewEngland and for 7 months a year the basement is a freezer. I have to do something about it unless i want to freeze myself.
About the skills...I know i have a lot to learn, some very fundamentals like how to solder and some as complex as control systems. Of all the things, one that most concerns me is the design of connected yoke and rudder pedals, hydraulics and mechanics involved in connecting and interfacing them with the system. One crucial skill i am more comfortable with is computers and programming. Ironically, i have a masters in mechanical engineering (with specialization in Machine Design..huh!) and the only mechanical engineering i practiced is - 'none'. Ever since i came out from the school, i am a software engineer by profession. I know i studied designing complex mechanisms to trace specific motion profiles while in school and learned to solve some complex problems there. But still i am a lot nervous about the very thought of it. Other day, i am reading about the forces acting on a flight and i am ashamed at my in ability to draw out the force field, write the equations, and resolve the forces on my own. Gosh, i can't even remember the equation for resolving vertical and horizontal component of a force vector!! Is it a "F cost(Theta) or F Sin(Theta)?" Shame on me.
On the other hand, programming, though a different kind this time (i know writing business applications is so different from writing control systems logic) i am up for the challenge. In general, i have some engineering background that i can count on for this project.
When it comes to help...i realized many pioneers who built wonderful, impressive sim systems at home have a friend or a partner working with and helping them out in most cases. Also, more importantly they have strong background in Aero industry and some how connected to it through friends or some like. In my case, this is going to be a disadvantage since i don't have a partner. I am on a solo flight here...
Scale and the scope
This is the most difficult part for me. I am ambitious and i want it all - the best, biggest and the craziest stuff. But the reality wants me to define some boundaries and set limits. So here is what i decided... i donot want to limit myself in my ambition, i am going to do this as long as it takes and as long as it is fun. I will aim for the bigger and the better but in a modular fashion. Each step builds on the previous one and adds to and enhances it. I can't escape but use my years of software project management experience here. Why not?
So, I recall the golden rules of software PM:
1) Requirements are bound to change during the course of the project
2) rather than fighting the change, be prepared to deal with it.
So how do we deal with the change? The answer in my profession(atleast the one i believe in) is - "Architecture centric, use-case driven, iterative and incremental" process.
That is exactly how i want to approach this project. I will have a system architecture that allows me to build the system in small incremental iterations, each iteration delivering some value so that i won't loose interest in the project waiting for a big kahoona to fly overhead some 7 years later....
Next thing is ..I ordered a CH Product's Yoke (excellent in many user's opinion) and the .audio 90 headphones from buy.com.
Well, i went for this yoke based on my earlier assumption of building a Cessna. Later I realized that for truely modeling a Boeing, it may not be useful (Boeing has two inter connected yokes (pilot side and First Officer(FO) side) mounted on columns . Oh! well, my plan is to adopt it to boeing style on the pilot side for sometime and when the time comes, then i can go for something else.
Also, even though it is less than a week, i did a lot of research and my mind is full of plans on how I want go about this project..
Germination of an ambition..
Initially I wanted to build a cessna sim. Next two days of intense research and what I saw on the web has inspired me to drop my modest goals and shoot for higher altitudes..i.e., building a real big bad Boeing 7xx class simulator.
Next came the crucial and critical step: My wife's approval.
I showed my sweet wife a couple of home built simulator photographs and posed the question! And i am not surprised at all by her response. As usual, she said - do it by all means if you really want to and if that makes you happy. That's my wife. Always behind me supporting on everything i do without even for once making it difficult in all these years. Only thing she wanted is a promise that i will complete it and not leave it in the middle by loosing interest (well, that says a lot about me i guess and a testimony to many other earlier passions that were never pursued to the end :-)
But something tells me that this time it is going to be different. I am going to do this.
This project has all the challenges i would love to face and i am sure i wouldn't get bored. The steep learning curve that i need to climb starting from learning about the fundamentals of electricity, electronics (not to mention, I need to learn how to solder), control systems, computer hardware, software (lesser of my concerns), video graphics, carpenting, and all other craftsman to be able to build space in the basement etc., etc.,
A life time of learning ahead waiting....and I am excited about my first kindergarden class...
Birth of "Project Sri 777"
By the end of the week, I know i am going to build a modest simulator of my own - probably a Cessna. Well there began my journey....code named "Project Sri 777".