Do companies differentiate camera ranges by firmware switches?

Canon have their lines, Panasonic have their lines. Usually we think of hardware as being the main differentiating factor between product ranges. But recently, we have had some unusual situations occur.

The consumer hardware in the GH1 has trickled up to the AF100. This is really unusual. The Pro AV Division’s AF100 is based on the GH1’s sensor. And on the Canon side you have a semi-professional camera, the 7D sharing an almost identical sensor to the entry level consumer T2i / 550D.

It is cheaper for a manufacturer to make one sensor and put it into two cameras and use firmware to differentiate the feature set, with the justification being that if you want more features you’ll have to pay more.

Back in the days of Intel versus AMD, the computer chip manufacturers used to clock their silicon based on the quality of the wafer. They used internal settings to set the clock speed. All silicon was not born equal and so the high end chips that didn’t perform quite as expected were under-clocked and sold as ‘lesser’ models. This of course led to the huge overclocking scene on internet communities, with inexpensive CPUs being over-clocked to near identical performance of more expensive models, with the odd crash on the way.

That’s similar to the GH1’s overclocked bitrates. But the difference with camera ranges is that the silicon is identical, and only the firmware is different. Panasonic could have shipped the GH1 with 44Mbit AVCHD which would have made it a formidable opponent to Canon, but they did not anticipate the market demand from videographers and consumers alike – helped by the excitement and publicity such cameras garnered. Things like HOUSE being shot on a DSLR were not anticipated by anyone, least of all even us!

Canon and Panasonic thought consumers would prefer low bit rate codecs and easy operation. Canon thought that 30p and zero manual controls would be fine for their target audience of photographers, because customers said they liked ‘smooth’ footage of 30p and that 24p ‘looked jerky’ on a PC screen and that video should be automatic and easy to shoot like a camcorder. Such customers are not filmmakers and aren’t experienced in the aesthetics of video. But now video modes are selling cameras by the bucket load and the DSLRs are becoming true hybrids will we videographers have a say in what happens in stills camera land? I sure hope the focus group co-ordinators recognise this trend too and give Canon’s product planners a properly representative picture of the market.

I covered a lot of that in my article about there being no conspiracy at Panasonic, but in light of recent developments I am forced to change my view slightly. Is there a conspiracy to limit camera performance via firmware? On some levels, yes of course there is.

Take these examples:

The Panasonic GF1 has an older sensor than the GH1, the same in fact as the original G1. But the G1 lacked video altogether because Panasonic wanted to add all the bells and whistles and the proper HD video lens, and so they chose to do video properly 2nd time out with the GH1. Nothing wrong with that.

Now the the GF1 only had 720p AVCHD Lite, and we assumed the hardware wasn’t good enough to do better. Wrong. As Tester13’s custom firmware shows by changing just 4 parameters in the firmware the GF1 can do 1080p at high bit rate AVCHD just fine.

The second (less extreme) example comes from Canon, with the 550D. It shares the same core hardware as the 7D but whilst the 7D has a full range of ISO increments, the 550D only has the basic 100-200-400 increments. Clearly a firmware limited feature. Would it have cost Canon more money in terms of testing and development to implement a feature already supported on the 7D and nearly all their other semi-pro cameras on the 550D? I doubt it, since as we have seen with the GH1, firmware features already exist on the most part and can simply be reconfigured or switched on and off.

To compensate, Canon sells the 550D at much cheaper price. But a lot of the cost saving comes from materials – no water proofing, a plastic body, a cheaper kit lens, etc.

So is it right for companies to differentiate between products by purposefully turning off existing capabilities in firmware?

The engineer in me says “no”, of course not. The marketeer in me says “yes, show me the money George”. I can see why it makes business sense, and without business sense Canon would be making croissants not cameras.

But still it isn’t clear cut. The debate that I have in mind is that maybe it doesn’t really make business sense at all. Why turn off a desirable feature like full HD in a camera? It seems they don’t fully understand the demands of the market. I can think of many whose purchase would have been influenced big-time by the presence of 1080p in the GF1.

The GH1 costs more because it is a better camera than the GF1 – better sensor, rotating screen, built in EVF. People don’t need ‘lower’ cameras artificially crippling for them to see this. The differentiating factors in camera line-ups should always be hardware based. I’d love to get an answer from Panasonic on why they left 1080p out of the GF1 because there doesn’t seem to be a technical or cost or business justification for doing so.

Does a lack of 1080p push people to pay more for the GH1 if they want full HD? No. It just pushes people away from buying a camera in the GF1’s small form factor and harms the GF1’s sales. The GH1 does well regardless because people understand the hardware differences and it’s aimed at a different kind of user. But denying the other kind of user 1080p for no reason at all seems a bit silly.

The fear is that Panasonic are going to a lot of effort to develop a stunning sensor for the GH2, only to cripple the firmware and have the AF100 aimed at cinematographers for a higher price.

Again we have a similar situation but with an unusual twist. The AF100 will have hardware features not available on cheaper cameras but the GH2 will have a better sensor. Are Panasonic going to risk having a GH2 with 70Mbit 1080/60p from a global shutter CMOS whilst the AF100 is basically a GH1 in a shoebox? Of course not.

The GH2 will have to have a shit codec.

The issue for Panasonic is, people like me would much rather buy the GH2. I prefer the form factor now I’m used to the DSLR stealth and portability. A camera should be as small as is possible without severely compromising handling.

Our demand for new DSLR models during the DSLR revolution is a growing one. If Panasonic ignore the demands of this growing market – the enthusiast cinematographer – and cripple the GH2’s firmware to separate it from the professional Micro 4/3rds video line, I think it would be bad for business. The GH2 would loose more sales than it would gain, or the AF100 at double the price would gain.

With the boom in internet video sites like YouTube and Vimeo, excitement via blogs and forums, online distribution, and eager actors connected to eager directors via the internet the DSLR market is shifting toward video. Customers are lapping up cheaper and more capable cameras, creating a new niche for Panasonic to cater for with a high bit rate GH2 with some enthusiast bells and whistles.

Panasonic and Canon must stop the differentiation of lines via firmware, and knock the conspiracy theories on the head. If consumers begin to believe that whenever they buy a camera they are getting a product which isn’t optimised to perform at it’s best, the malaise will spread throughout the industry and hand a growing market sector to RED on a plate if they ever made a sub $2000 DSLR style video camera.

Would you rebel from the big corporations and buy a RED DSLR with lens adapters for Nikon, Canon and Contax Yashica for under $2000 if you felt the next generation of DSLRS were crippled? That’s a more attractive product than Scarlet surely and I for one would buy a video orientated camera like that over a 17Mbit AVCHD GH2.