Thursday, July 2, 2009

My New Hobby AKA I Kanna the ShutterBug

Yes, I mentioned that the 90 size crash and stopping of R/C flight was actually a blessing in disguise...You see, I have always wanted a better camera than the two compacts that I have because it gets a little hard to produce sharp pictures with them especially when I am fixing my heli in the middle of the night in my not so well lit kitchen. For those who are wondering whats the link between kitchen and heli, basically I am very particular(or some call OCD-obsessive compulsive disorder) about keeping not scratching my flooring in other parts of the house, so the kitchen which has ceramic tiles becomes the place where I build and rebuild pretty much ALL my R/C.

The reason I want a DSLR is soley for the low light capability, basically high iso and larger aperture. It is perfectly possible to produce good pictures with a compact camera, but the problem with them is that in a dimly lit room, you need to increase the exposure time to keep the ISO(refer to 1. at the end of article) low for nice and clean photos. In normal situations, I can do handheld exposure of up to 1/3 second because the built-in stablisation of my compact is really quite good. However, building a heli in the middle of the night is not normal and most of the time, I am tired and my hands are sore, which all add to shaky hands and blur pictures.

And so with a little bit of budget, I set out to buy my first DSLR a month ago when my school holiday really started. I had originally wanted to get a Nikon D90 with a Tamron 17-50mm f/2.8 lens. However, the Tamron was really slow in auto focus and the D90 did not come with micro-focus adjustment, which is used when your lens and camera body are not "in sync" with each other. What happens is that your camera body want to focus on the middle point, but the lens end up focusing on something a little in front or a little behind thereby causing a slight out-of-focus. It can be solved by calibration at the service center, but that would mean having to go down to the service center every time you get a new len. I didn't thought it was such a biggie until I tried a D90 at the shop and it was constantly back-focusing with both kit lens and the the Tamron.

That pushed me towards the canon 50D, which although had a nosier sensor, came with the AF-microadjustment function. I was certain I was going to buy it when I started playing with it and realise that it came with a jog-wheel and joystick EXACTLY LIKE HOW THE FUTABA 10C that I had. As for lens, I settled for the kit lens over the tamron because it focus faster and the image quality is not really that bad. Best of all, the price tag was smaller and it came with image stabilisation which the Tamron does not have. However, the Tamron will probably be great for low light because it has a larger aperture. Consider a larger aperture as a larger hole for light to go in, so obviously more light will enter in the same amount of time. So the difference between the Tamron and the canon 18-55 kit lens is that one came with stabilisation, one came with a bigger aperture(refer to 2. at end of article).So it was the 50D+18-55 kit lens.



Everything that came along with the kit, including memory card and bag.


Bulk of the cost, the Canon 50D body. It is suppose to be "semi-pro" camera body, but I feel that it is more of a hobbyist-consumer body.


Canon 18-55 f/3.5-5.6 IS kit lens attached. While it is really cheap and plasticky, the image quality really isn't that bad.


The jog-dial and joystick that is commonly found in futaba TX and on all the TX that I use/used.



The excitment and happiness of buying the camera was just like when I got my Trex 700N. Something that words cannot describe.
Here are some photos I took using the kit lens, nothing impressive, but good enough for my use. I thnk it is more of a issue with the user rather than the camera or lens.

Although the high iso is not as great as the nikon, some high ISO sample shots of my Ino-lab servos proved to be ok. Photographers might look a crop from the 100% photos, but how often will we really use that 15megapixels?

ISO200 Noise Reduction(NR) STD


ISO1600 NR STD


ISO3200 NR STD


ISO3200 NR HIGH


ISO6400 NR STD


ISO6400 NR HIGH


ISO12800 NR STD


ISO12800 NR HIGH


ISO3200 is perfectly usable for me and banding starts to appear at ISO6400. ISO12800 is really quite bad and not something that I will ever use.

In addition to the kit lens, I also got a canon 50mm f/1.8 prime lens which to me is really sharp and really bright. Again, it is really cheap too, at just $130. I settled for these cheap lenses because of my limited budget and I really don't see the need for expensive lenses yet. I mean these lens are way more than sharp enough for me, just that they feel a little cheap and plastic. More about it another day...

1.For those who don't know, ISO is basically the light sensitivity of the sensor. When you have a higher ISO, the sensor in your digital camera can capture more light in the same amount of time than at lower ISO, therefore, higher ISO means that you can reduce your exposure time required for photos of proper brightness. However, as ISO increases, noise starts coming in and it can be observe in the form of those grains and coarse looking texture, especially in the shadow region. Therefore, we tend to try to keep the ISO low for shooting. As sensor size increases, the high ISO noise reduces, which is why most DSLR have better high ISO photos than compacts because of their much larger sensor. However, not all DSLR have the same sensor size, some are larger(and more expensive), some are smaller and equally expensive. Mine happens to be somewhere in between. In heli term it is like comparing a 90size(Large sensor), 50size(medium size sensor) and micro electrics(small size compact camera sensor) and it doesn't require a Ph.d to know which is nicer to use, although all can work great in the right hands- just not mine.

2. Stabilisation allows you to have longer exposure when taking photos, so more light can enter into the sensor. What the stabilisation does is to compensate for movement of the camera due to handshake when the camera is taking a photo.

A larger aperture on the other hand allows more light to enter in the same amount of time, so your camera spend less time trying to capture the image.

The main difference is that stabilisation minimize the effect of motion-blur created by the photographer, while a larger aperture and shorter exposure time will minimise motion-blur created due to movement of the subject in the photo. I.E. You want a big aperture for moving stuff like a flying heli in a cloudy overcast day, you want stabilisation for taking photo of a static heli in the same cloudy overcast day.

So how powerful a computer do you need to run Realflight G4?

A while back, I started playing some games like call of duty-world at war, grid and farcry 2. However, the low end card I have in my computer proved to be a very severe bottleneck to the smoothness of the game. While those game were not as demanding as realflight G4, they were somehow still unplayable on my Geforce 7300. Coupled with the fact that I had always wanted a more powerful card for running realflight in the rendered fields, I started looking at a graphic card upgrade. It was obviously stupid for a person like me who so rarely play games to be spending a lot on a GFX, so my I wanted something that is cheap, good and supported by my other components in my current com.

And with my tiny budget... I got a Asus EAH4830


From 50D



It is the slightly overclocked version of the standard ATI 4830 and is cheap and good enough for the 1280x1024 resolution that I run my games and simulator at. This is a full size, dual slot card which is quite a bit longer than my old card and requires one PCI-E power connector, which just so forunately is supported by my P.S.U.


From 50D



From 50D


And this is how it looks like when installed in my rig. I was lucky enough that it fits my small casing after I shifted the harddisk down. Just a bit longer and I will probably have to change the casing with computer for videos in my bedroom which has a old and ugly full size tower.

From 50D

From 50D

This card is IMPRESSIVE. At my monitor's native resolution of 1280x1024, everything was much much quicker during gameplay and there was no lag even during large explosion. Impressive card.

Now the important part...How does Realflight G4.5 runs with this?
Good enough for me. For the rendered fields which my previous card was struggling to do, framerate never for once, dropped below 30fps with this new card. That is, even with all the bloom effect, shadows, soft shadow and what not turned on the max.

Night flying in the circus looks awesome and is perfectly playable now. On my old card, I was more like looking at slide show, rather than flying.

Water flying is the tricky bit. With all the effects turned on to the max, I am still not getting the 30fps that Iwant. It hovers somewhere around 25.5-27fps, which I guess is good enough considering that this is a budget gaming card. I suspect the bottleneck is my cpu which is the first generation core 2 duo.

All in all, it was well worth the money and it is something that I regret not doing earlier.

However, there is one tiny problem. Because of the amount of power that it consumes in gameplay, it produces a lot of heat that my mini-tower finds hard to dissipate. System temperature shot up by a whole 4 degree when I am playing game and my P.S.U. seems to be running a low warmer. Probably need one or two more fans to increase air circulation in the casing.

Tuesday, June 30, 2009

Ino-Lab 760mg, one year down the road...

THEY ARE TOTAL CRAP AND SHOULD BE AVOIDED AT ALL COST.

On my first crash of the 700N, all 3 of them got exceptionally tight to even rotate by hand. A trip the local distro suggested that I might have gotten the servo case screws too tight, which IMHO, is total absolute rubbish because the servos were free before the crash and so the crash just suddenly tightened all the screws? Wow that is some magic. And not the mention that the crash was a super minor one that only cost like less than $300 including blades, which is really cheap for a 90size.

Post repair flights revealed that the ino-lab 760mg that I have are drawing an insane amount of amp in flight. More than 3amp. Considering that I am running 5.1v output on my regulator, a 4amp load on my regulator will mean trying to dissipate (7.8v-5.1v)*4amp=10.8w of heat, which is definately impossible for any R/C regulator in the market now, not even the very expensive ones which have got even smaller heatsink. What that means is that....the insane amount of amp draw resulted in a second crash, from what I thought was servo failure in flight when in reality was just my regulator going into thermal protection.

So how did I know it was the ino-lab and not the rest of the electronics causing the insane amp draw? Two signs...

1. The ino-lab's got exceptionally hot after a flight. Still remember the comment I made about the lack of a heatsink on the ino-lab being the weakest link? It has been proven by yours sincerely that it does indeed need a heatsink to be cool because if even the plastic case can get hot to touch, then the internal motor temperature is obviously going to be molten hot.

2. I shifted them to my raptor 50 and from the average 1.5-1.8amp draw, with just TWO ino-lab 760 on the cyclic, it shot up to over 4amp, 4.5 to be exact. Only god knows where all the current is running to. Perhaps you may wonder, how come the regulator didn't go into thermal protection this time round? Simple. I increased the output voltage to almost 6v, which means a whole lot less heat for the heatsink to move.

As such, this has proven to be an expensive lesson. $246 worth of servo for just 80 or so flights, excluding the cost of the second crash. Painful.

On a side note, this crash has prove to be a blessing in disguise in some other ways, as will come later on....well...a partial change in hobby. In fact, I am considering quitting from all online R/C activities, including the postings and forums....

My flight log for the months of may and june: 1/3 of a gallon in may and june, combined. So few that I have lost count. Oh well, in any case I was already disgusted by the price hike of almost everything at the local hobby shops.

The wreakage...been sitting there for months. Probably engine and bearings are wasted, but who cares. Canopy is semi-repaired by a half arse job.

From 90size
From 90size
From 90size

WAIT A MIN.... YOU CRASHED THE HELI AND ANY SERVO CAN GET DAMAGE IN A CRASH...HOW COME YOU ARE BLAMING THE INO-LAB??

My point is simple. If one or two servo conk out in a serious crash, it is user error. If all three servos fail( and fail in a weird way) from just a light crash that didn't even destroy the canopy, obviously something was not quite right with the servos in the first place. I have seen worse crashes on my raptor and the hitec never had such problem. Hack, even the $20 hitec 325hb on the throttle in the 700N didn't give me any problem after the crash. The futaba BLS251 on the rudder have the tail rod control pulled out from it when the boom ejected upon crash impact and all it took was just a small plastic gear replacement. So how can all 3 servos fail from such light crash?