Review & Teardown: Intempo HD Action Camera

In recent years, the action camera market has been hotting up, primarily due to the breakthrough success of the GoPro line. Other established names, as well as less familiar names, have tried to challenge the GoPro for success, but none have yet de-throned the king just yet.

The GoPro features high quality footage, and a wide array of mounting options and accessories. Recent products bring greater flexibility to shoot in stereoscopic modes, with Wi-Fi remote and display, interchangeable batteries for longer shooting and higher frame-rates and resolution than ever before. It has gotten so good that many TV shows have used them to provide broadcast-quality footage, for example, Mythbusters.

Unfortunately, this typically comes at a high price, upwards of $300 for the better models. Many people are not willing to shell out this price for something which they might not use very often.

That’s where lower cost alternatives might be the better option. In this article, I will be looking at the Intempo HD Action Camera, generously provided by MobileZap, currently listed for AU$60.99 plus shipping.

The Package

The Intempo name has not really been used anywhere else, and it was a bit perplexing when I received the box that it was instead branded Ocean. It is, however, the same product judging from the appearance.

DSC_4418

It is packaged in a colour cardboard box with viewing window, showing off the camera in its fully waterproof case. An array of mounting options, including a tripod mount, a waterproof tripod mount, handlebar mount and helmet mount are illustrated. The text is mostly in French.

DSC_4416 DSC_4414 DSC_4415

A photocopied and stapled English manual is provided with the product, outside of the box.

DSC_4419

The model number is EV0200, and it was made in China in March/April 2013.

DSC_4417

Inside the box, not visible from the outside, is the tripod mount, the helmet mount plate, a velcro nylon strap to secure the helmet mount plate to the helmet, and a handlebar mount with detachable tripod screw mount. There is also a manual in French, and a USB A to miniB lead (note: not a microUSB connector but the older miniUSB type).

DSC_4420

The waterproof case for the camera appears solid and has a special viewing window for the lens made of more optically transparent material. The main body appears to be some sort of perspex. Access to the top two buttons (power, and shutter) are provided through the case.

DSC_4421

The camera sits inside with the rear buttons for OK, up and down being inaccessible from the back. The LCD is clearly visible without obstruction. The camera itself is CE marked, and is marked with a model number of EV0200GFDIR.

DSC_4422

The waterproof case has a rubber seal all the way around and a latch system to keep a tight seal. The bottom offers a standard tripod screw mount.

DSC_4424

With the camera out of the case, the top shows a recessed power button to avoid accidental activation, and a shutter button. In the centre is a hole for the mono microphone.

DSC_4425

Along one edge is a slide switch which toggles the camera between photo, video and review modes.

DSC_4426

The bottom allows for the insertion of a microSD card, up to 32Gb, and the miniUSB-B port for charging and data transfer.

DSC_4427

There is a provision on the other edge to thread a lanyard strap into the camera.

DSC_4428

From the front, the lens is diminutive, which does not bode well for image quality.

DSC_4429

The rear is mostly covered by an LCD, with buttons for up and down and OK to make menu selections, review photos and control the digital zoom. There are also LEDs for charging (red) and activity (green).

DSC_4430

Teardown

Removing the small screws at each end allows the top cover to unclip from the bottom. It reveals a total of 440mAh of Li-Poly battery capacity attached in parallel to the top cover. Internally, the main PCB is black, featuring an iCatch/Sunplus SPCA1227A System-on-a-Chip which is commonly used in novelty spy pens and mini cameras. There is a Nansi N25S16 serial EEPROM with configuration data, and a PM Tech PMS307416ATR-6CN DRAM chip for the processor.

DSC_4435

The lens assembly can be seen firmly epoxy glued to the PCB, and the focus was not alterable as the threads were epoxied as well, likely from the bottom. I did give it some force, but it would not budge. It is very similar to those used by manual focus webcams.

DSC_4437

The rear of the PCB shows the connections for the LCD soldered down, as well as a thin flat buzzer. The electret mono microphone is just visible at the top.

DSC_4438

The LCD itself is marked BKT180C-A61, dated 13th March 2012.

DSC_4439

The PCB itself is identified as DV123SC-MAIN Ver 0.2 dated 9th August 2012.

DSC_4440

In Use

The camera itself is very small, and easily fits within the palm. Starting up the camera is as simple as pressing and holding on the power button.

DSC_5647

The firmware is VER.SV123SC 20130410. The camera defaults to booting up with a French menu. It took a little bit of wrangling to change the language, as you need to press OK to bring up the menu, then power (quickly) to toggle to the settings menu, hit down until you get to language, OK, then select English, then OK. Only then, does it become usable.

Selecting the shooting mode is as easy as sliding the switch on the side of the camera. The screen appears as follows in the three modes – photo, video and review.

DSC_5631DSC_5632DSC_5633

The LCD itself is somewhat grainy and pixellated, and there is no zoom on review mode. Pressing up and down in shooting mode controls the digital zoom, up to 4x. No internal memory is provided, although in review mode, quick pressing the power button toggles to a menu stating “No Files”. A properly formatted microSD card is absolutely required to use the camera, inserted into the slot at the bottom of the camera.

It’s important to navigate most of the options in the menu to choose the highest quality options available. The menu looks like the following images.

DSC_5635 DSC_5637 DSC_5638 DSC_5639 DSC_5640 DSC_5641 DSC_5642

The battery takes two hours to charge from flat to full. The display features a battery indicator with three segments, although, it is very inaccurate. Soon after charging, it falls to 2/3rds, and it seems to last a long time on 1/3rd. Testing the continuous shooting time resulted in 68 minutes of footage recorded, although the LCD went to sleep automatically a few minutes after shooting commenced. Despite the LCD sleeping, the recording continued. The battery is not easily replaceable in the field.

The unit features quiet key beeps which cannot be switched off. When the battery is about to be depleted, the unit flashes up a Low Battery warning and beeps three times before turning off.

The files recorded by the device are stored in DCIM\100MEDIA\SUNP****.JPG or .AVI. Images are stored as JPEGs of about 1.2-2Mb size, with a resolution of 2592 x 1944 when set to 5MP. The EXIF data states a camera name and model of Icatch Spca1628 V233-00-01. It claims an aperture of f/3.2, and all the images appear to be recorded with a sensitivity of ISO100 or ISO200.

Videos are recorded as 25Mbit/s MJPEG videos, up to 1280 x 720 at 29.97fps with a mono 11.025khz 16-bit audio track in an AVI container. Files are automatically split at 4Gb increments due to the FAT32 filesystem limitations.

The lens itself is fixed focus, so there is no focus lag, although there is a significant shutter lag, and the shutter speeds are quite slow due to the limited sensitivity of the sensor, resulting in many blurred images in indoor lighting.

Unfortunately, the sample I received seemed to have an anomaly which results in a dark speck appearing in the top right hand corner of videos and on the border of images, especially when on light coloured backgrounds. Despite numerous attempts to clean the front lens, it was determined that this was likely to be a piece of dirt resting on the sensor.

As the lens was epoxied onto the PCB and sensor, and not removable, the dirt could not be removed. However, I have been assured that this is a rare one-off defect, and that customers should not see this in their photos.

SUNP0003.AVI_snapshot_00.03_[2014.04.11_10.14.03]

Performance Testing

In order to find out how it holds up in real life circumstances, I had to take it out for some “action”. This review took a little longer than expected, due to the poor weather in Sydney over the past few weeks. I had been meaning to take it out using the handlebar attachment on my bike, riding the local cycleways, but it was raining week after week. Luckily, last Sunday provided a good break, so I decided to strap the camera on and give it a try.

The handlebar mount itself secures quite well, although the rubber strips do overlap somewhat and it can be tricky at first. Trimming the overlap makes it easier to fit. The plastic cap that keeps the nut from getting lost did pop out, but some super-glue fixed that.

The waterproof casing provided additional assurance should there be an unexpected downpour, and was easy to open and close for those “photography” moments.

Photos

As it would not be fair to compare this 5MP camera to something touting many more megapixels, I decided to use my “the new” iPad 3rd generation with a 5MP sensor as the benchmark. We already know that phone/tablet cameras are not the best due to optical limitations, but it seems fair to compare two sensors with the same pixels and similar optics size (i.e. small units, not DSLR or Point and Shoots).

I chose a scene near my cycleway of choice, which contained much fine detail as well as variation in brightnesses. This would allow me to illustrate the relative merits of the dynamic range and resolution of the sensors. It would also let me illustrate the focal length differences. Below are the uncropped originals – click for full size.

              Intempo Action Camera                                   Apple iPad

SUNP0019 IMG_0906

Immediately evident is the focal length difference – the action cam has a smaller field of view than even the iPad, which itself is not as wide as many phones of today. The dynamic range of the iPad surpasses the action cam dramatically, by preserving details in both the light and dark areas.

Cropping to 100% (320×240, one pixel captured is one pixel on your screen), it is evident that the low cost action cam falls greatly behind the iPad’s 5MP sensor. In fact, it’s clear that the details are smeared and non-existent. Note that because the focal length is different, it appears the images are zoomed in different amounts when they are in fact both at 100%.

               Intempo Action Camera                                         Apple iPad

SUNPonetooneiPad-OnetoOne

A further view of the EXIF data of the shot from the action cam gives a hint that the sensor itself appears to be only 1296 x 972 pixels, despite the file being 2592 x 1944 pixels.

EXIF Inconsistency

Unfortunately, this is not a commendable performance, and resembles more of a 1.3MP webcam quality.

Video

I was able to get a total of 55 minutes of footage and three pictures out of the battery before it shut itself off with a few quiet beeps and flashes of the screen.

Unfortunately, the footage itself suffered quite strongly from rolling shutter artifacts due to the way the camera utilizes its sensor. The dynamic range is also fairly poor with a slow automatic exposure causing highlights to blow out and a “washed out” look with little saturation. The detail is not exactly where you would expect it to be with 720p, and the rolling shutter artifact made even a leisurely cruise down a cycleway path into an almost uncomfortable piece of footage to watch.

The footage itself was recorded continuously, and a segment of two minutes was cut by using Virtualdub’s Direct-Stream-Copy feature which causes no loss in video quality. This was then uploaded to YouTube as an MJPEG AVI file, which was then compressed by YouTube into H.264. By avoiding multiple recompression runs, we can preserve as close a quality as possible to that provided by the camera – which is not particularly remarkable.

With the bright sun, I would have expected the shutter speed to shorten, reducing the shutter angle, and the artifacts to be less severe, but this was not the case. Maybe for lower action footage, mounted to a helmet, it may have been less nauseating, although swinging your head around is almost certainly going to cause the issue to reappear.

Conclusion

While the Intempo HD action camera is a small, “low cost” device, its performance is relatively poor. While it claims to be a 5MP device, the photos it produces fall far short of what an Apple iPad’s 5MP sensor can produce in terms of detail and dynamic range. This results in an image which lacks details in the highlights and shadows, and can appear smeared and washed out. This can be understood when its internals share more in common with novelty spy camera pens, webcams and watches, than a serious camera. Even smartphones of today easily outclass the 5MP figure that is touted, however, I have reason to believe that the sensor in this does not provide even 5MP natively.

The lack of fisheye lens, and a relatively “zoomed in” focal length of approximately 64mm (in 35mm terms) makes for very little action being captured except what is directly in front of the person, into the distance. The fixed focus means no problems with focus hunting, but however, trades off sharpness as a result.

The video is recorded as MJPEG in AVIs, split at 4Gb, rather than more modern H.264 which means it consumes a lot more space for the same quality, and the audio is mono, only sampled at 11025hz which makes it more “AM radio” speech quality.

Unfortunately, the video is riddled with rolling shutter artifacts (as are most mobile phones, but not dedicated video cameras) due to the way the sensor is used, and with only two sensitivity levels (ISO 100 and ISO 200 as alluded to by the EXIF data), even at high brightnesses, the artifacts are still very apparent and make handlebar footage look rather disturbing (the jello-vision effect). As a result, it is really only suitable for “low-motion” action, and even then, the video quality isn’t anything outstanding with 720p being “outclassed” by many smartphones.

The non-removable battery, with a typical shooting time of under an hour, may prove to be a limitation for some users as well. However, it is possible (although limited in practicality) to attach a large external USB power bank to power it or charge it in the field.

The sample video and images illustrate typical performance of my review unit. Where quality isn’t the primary concern, and instead, small size, waterproof casing and mounting are important, then this product may be suitable for your needs. Unfortunately, given the performance it displayed in use, I find this product hard to recommend to most users.

Thanks to MobileZap for providing this unit for review.

About lui_gough

I’m a bit of a nut for electronics, computing, photography, radio, satellite and other technical hobbies. Click for more about me!

This entry was posted in Computing, Electronics, Photography and tagged , , , , , . Bookmark the permalink.

2 Responses to Review & Teardown: Intempo HD Action Camera

  1. boloshon says:

    Hello I’m wondering if we can find somewhere a way to update the firmware? I expected this cam to be able to record automatically when you move your car (like a dash cam), and I know that on some shop the very same does, so I guess it’s just a problem of updating it, right?

    • lui_gough says:

      Not necessarily. There are many products with the same exterior but different interiors, which unfortunately would preclude this. Other low cost products may include one-time programmable ROMs or not bother installing flash memory to hold the firmware itself. Finally, some chips (e.g. cheap MP3 players) have internal fusible links which are blown to disable functionality not paid for by the manufacturer at time of ordering. So no, it’s not necessarily that easy.

      The other issue is that there’s no exposed firmware updater facility in the existing software itself, nor do you have a way of taking the firmware out of an existing device that does what you want. If I’m to guess, U3 is probably the serial EEPROM that contains configuration values for the main Sunplus chip (its capacity is too small for firmware), but the firmware itself might be stored inside the Sunplus chip itself either as Mask ROM or integrated Flash which doesn’t allow us a convenient way of extraction.

      That being said, for such a low performance, low cost product, I don’t see it being worth the time to hack it either.

      – Gough

Error: Comment is Missing!