HMC40 - First Look

The HMC40 was, in our view, sharper than the 170. That surprised us.
Duh. The 170 is pixel-shifted. This is why the 150/170/200 have lower than average resolution among HD cams including consumer cams, this is why they deliver 3 stops over the HMC40. The result was to be expected.

I also said that in CCInfo tests the point-and-shoot Canon HF20/200 palmcorder resolved about 900 lines in good light. I cannot see numbers past 800 on this chart, don't know how they measured 900 lines. Still, not bad for 600 bucks, huh?

Canon_HF20_3000_lux_auto_noise_crop.jpg
 
Not sure what is meant by "Duh" but David did say that his tests were at 720p24 and at that resolution I would also have also been surprised to see the HMC 40 sharper. Still what do I know as I have only been using the HMC151 since November!

I am also sure that David is fully aware of pixel shift technology.

Ken
 
Last edited:
Could it possibly be because pixel shift technology, as applied to this camera, is easily able to resolve 1280 X 720?

If I am talking rubbish I apologize.

Ken
 
Cranky said:
Duh. The 170 is pixel-shifted.

In this case, why he was surprised?

I am aware of the pixel shifting in the 170, 200, and 200a. Been aware for nearly four years.

However there are other factors we were testing for the first time in circumstances where we usually shoot using these two cams.

Such as, among others:

1/4 sensors versus 1/3 sensors.

AVCHD versus DVCProHD.

CMOS versus CCD.

Light sensitivity.

Transcode to pro res HQ versus native DVCProHD.

These discussion threads work best when comments are neither condescending nor pretentious.

Try to remember that if possible.

best
 
Last edited:
These discussion threads work best when comments are neither condescending nor pretentious.

Try to remember that if possible.
Well, people should not only "try to remember that" but start complying with it, or find themselves disinvited from the discussions...
 
David, I am sorry if my reply came out as condescending. In fact, I felt that you were being patronizing a little bit, sort of saying "Lookie here... A camera that is 3 times cheaper than my REAL PROFESSIONAL camera that delivers a comparable or even better image." So the HMC40 has a smaller sensor, but apparently it has a matching lens, so the whole combination together resolves more lines. I am also not suprised that AVCHD holds well against DVCPROHD, after all the difference in bitrate is only 4x, but AVCHD is long-GOP, which means it is more efficient. And it is more modern. And DvCPROHD simply records less pixels in the frame. CMOS vs CCD? If any, CMOS sensors are usually higher-res than CCD because there is less wiring on the sensor. Light sensitivity should not affect resolution in good light.

It seems like point-and-shooters, pros, still photogs and others... they only know what happens in their own neighbourhoods. The Sony HC1 resolved 550+ lines in 2005, the Canon HV20 resolved about 700 lines in 2007, the current pocket-sized HF20 resolves almost 900 lines, this is Sony EX territory. People are locked in their environments and don't care looking outside. At the same time, the 200/170/150 use pretty much the same 960×540-pixel sensors block as four years ago. The HMC40 -- along with the HPX300 -- breaks the mold, but pays with sensitivity.

Modern consumer cams can shoot video on par or better than pro cams that cost 5 or 10 times more (the HF10/11 can be found for $500-700, down from $1200 MSRP last year). The usual issues with consumer cams is usability, manual controls and other "pro" features. But the sheer image is good enough. Here is another link for you: http://provideocoalition.com/index.php/awilt/story/review_canon_vixia_hf11_avchd_camcorder/ If Canon did not use interlaced chroma with 24p and 30p video, the competition -- with the EX no less -- would have been even tighter. (Sorry if I did not use the tenses correctly). The comparison is to MPEG-2, not DVCPROHD.

The chart shown was shot using the camera’s highest-quality mode, 24 Mbit/sec 1920x1080 PF30, but I saw little variation across frame rates or recording formats. Both interlaced and progressive shooting modes gave me similar results, as did different recording formats: on a still image, as I’ll discuss, even 5Mbit/sec LP mode gave a comparable picture, despite its 1440x1080 recording (which tops out at 787 TVl/ph resolution in the horizontal direction).
...
In 60i mode, I found that the HF11’s 17 Mbit/sec images held up almost as well as the EX1’s 35 Mbit/sec images, while its 24 Mbit/sec pix were visibly less affected by compression artifacts. Likewise, the HF11’s 12 Mbit pix almost equaled the EX1’s 25 Mbit pix, while its 17 Mbit pix were better. AVCHD really is roughly twice as good as MPEG-2 in terms of bit-rate for a given quality of image.
 
I'm just reporting what we saw -- that's all. That's what the boards are for.

I said: "OTOH, we shot some colorful, outdoors, sunny footage using a HMC40 and a HPX170 at 720/24p in both cams.

The HMC40 was, in our view, sharper than the 170. That surprised us."

I DIDN'T say: "Lookie here... A camera that is 3 times cheaper than my REAL PROFESSIONAL camera that delivers a comparable or even better image."

Those are two materially different statements.

What I and some others are doing is going out and shooting stuff at real gigs with real people in real environments and reporting here.

We are fortunate enough to have these two cams shooting side by side in a paying shoot, and then report here to assist others in making decisions.

I also did this with a 170 and a 150 we previously owned. And reported our results for others to learn or ask questions about.

I don't have any axe to grind about either cam or the 200 we owned and the 300 we tested.

If a $50.00 Kodak HD cam shot better than a 170 or 300, we would use it.

The point being is that I'm not interested in how the cameras resolve or anything similar. I'm interested in HOW the video LOOKS. Others may be interested in the other stuff. That's great.

If I had my choice, I'd take a full raster 1/3 inch CMOS p2 based AVCINTRA camera in the form factor of a 150 or 170 at the price of the latter.
 
OTOH, we shot some colorful, outdoors, sunny footage using a HMC40 and a HPX170 at 720/24p in both cams.

The HMC40 was, in our view, sharper than the 170.

That surprised us.
David, thanks. Can you provide any opinions on a comparison between the HMC40 and the 170 beyond resolution (e.g., dynamic range, color reproduction, Panasonic "mojo", etc.).

Have you compared the implementation of gamma, color matrix, etc?
 
We shot with near identical scene settings.

My spouse handled the 40, so I had that set to automatic because she still isn't conversant enough to go manual. I had the 170 set to manual, and worked that way.

The 40 was sharper, brighter, but with less of the "mojo" than what I saw in the 200/170 series. But it looked so darn good in full sunlight. Really nice stuff.

The color reproduction was also somewhat more saturated than the 170, but not overly so. All in all my spouse liked what she shot more than what I shot insofar as color reproduction.

A low light presentation with flo bulbs and some natural sunlight produced a different story for B shots with the 40.

We knew that going in, so were careful in what we shot and where we shot.

These were locked down shots, so we found the best light for the 40. It measured well, but I could see some of the gain kicking in. We had it limited to 12.

I like the 40 a lot, and I think it's worth every penny, but any user has to be aware of its capabilities going in.

And a good support system is needed if you don't or can't use a tripod.. The cam is so light that hand held with OIS still produces some bounce.
 
It seems like point-and-shooters, pros, still photogs and others... they only know what happens in their own neighbourhoods. The Sony HC1 resolved 550+ lines in 2005, the Canon HV20 resolved about 700 lines in 2007, the current pocket-sized HF20 resolves almost 900 lines, this is Sony EX territory. People are locked in their environments and don't care looking outside. At the same time, the 200/170/150 use pretty much the same 960×540-pixel sensors block as four years ago. The HMC40 -- along with the HPX300 -- breaks the mold, but pays with sensitivity.

Modern consumer cams can shoot video on par or better than pro cams that cost 5 or 10 times more (the HF10/11 can be found for $500-700, down from $1200 MSRP last year). The usual issues with consumer cams is usability, manual controls and other "pro" features. But the sheer image is good enough. Here is another link for you: http://provideocoalition.com/index.php/awilt/story/review_canon_vixia_hf11_avchd_camcorder/ If Canon did not use interlaced chroma with 24p and 30p video, the competition -- with the EX no less -- would have been even tighter. (Sorry if I did not use the tenses correctly). The comparison is to MPEG-2, not DVCPROHD.

With respect, I think the reason why people that use HMC150s and above don't compare them to cheap consumer cams is because consumer cams will be totally unsuitable for them. Image quality isn't just resolution you know? Have you ever had a camera like the HMC150 next to a consumer cam in low light, this is where having 3 x 1/3" sensors shines compared to 1 x 1/3" or 3 x 1/4". It is literally like night and day.

Then you've got other factors such as XLR audio, manual controls, handling etc.

There really is no point in comparing a HV20 to a HMC150 as they are totally different and designed for different purposes.

One last thing, I believe you are wrong about the HMC sensor being four years old. The sensor in the 150/170/200a is much more sensitive than the previous sensor in the 200, just because it has the same resolution does not make it the same sensor. Personally I would much rather have a sensitive CCD sensor than a higher res one or CMOS.
 
I got to thinking a bit more about the equivalent ISOs for the different gain settings on the HMC40. As I said in a past posting:

HMC-40
+0dB = ISO 64
+6dB = ISO 128
+12dB = ISO 256
+18dB = ISO 512
+24dB = ISO 1024
+34dB = ISO 3200 (VERY bad image sensor noise at this setting)

If we look at the HMC-150, here are the numbers:
+0dB = ISO 500
+6dB = ISO 1000
+12dB = ISO 2000
+18dB = ISO 4000

In addition the HMC-150 has an F1.6 lens and the HMC-40 has an F1.8 lens.

(1.8 x 1.8) / (1.6 x 1.6) = about 1.27 times more light on the image sensor at maximum opening.

Ignoring the issue of image sensor noise, if one pushes both cameras to the maximum limits, the HMC-150 (+18db gain) is still 1 and 1/3 stops faster than the HMC-40 (+34dB gain). Even at +12dB gain the HMC-150 is 1/3 a stop faster than the HMC-40 at +34dB gain.


Bob Diaz
 
Image quality isn't just resolution you know? Have you ever had a camera like the HMC150 next to a consumer cam in low light, this is where having 3 x 1/3" sensors shines compared to 1 x 1/3" or 3 x 1/4". It is literally like night and day. Then you've got other factors such as XLR audio, manual controls, handling etc.
What about 3 x 1/6? ;-) Yes, I know that there is more to a pro camera than sheer resolution. If you re-read my message you will see that I said the same thing. My original point was that there is nothing unusual in a fact that a consumer camera can produce image comparable to a professional camera costing 10 times more. Consumer products improve faster because they have shorter development cycles, afforded by larger consumer base.

In this regard, the HMC40 is an obvious cost-cutting solution, exploiting a consumer product that is already good enough, and adding to it professional bells and whistles. Panasonic took the lens/sensor/LCD from a popular consumer camera that has its cost distributed to a large consumer base, and sticked it into an updated DVC30 body; reused the handle; reused the XLR block. Save, save, save. Now you have a $2K camera with video quality of a modern consumer camcorder -- read: with video quality comparable to a 4-year old pro cam -- and even with consumer-oriented features like spot focus or face recognition.

I will not be surprised if in the next year Panasonic takes the lens/sensor from the next consumer model and sticks it into the existing HMC40 body.

What I am surprised of, is why Panasonic did use this trick earlier, with the 3CCD 1/4-inch system off the SD1? I think this would have been a better complement to the 200/170/150 family. Maybe the sensors were too expensive, the original price of the SD1 was $1,400 and Panasonic did not sell many of these. It did use these sensors in the HMC70 though.

I am also very surprised why Canon did not do the same thing with the HV20 lens/sensor block and the GL-2.
 
To add to what Cranky just said, these are cameras for different uses, making the comparison fun but perhaps not critical. The 150/170 are clearly for lower light situations, more control (i.e. manual zoom), a more "professional" appearance due to larger size, etc. The 40 is for those who don't need those extra characteristics OR want a small footprint (such as those would be attracted to the HM100), or as a target price for the consumer looking to step up to prosumer.

As for the 40 being higher res, I attribute that to a simple fact: Panasonic hasn't upgraded the 150/170/200 yet with higher resolution sensors, not because there is any other good reason their $3500 and $5000 cameras have lower res than their $2000 camera. I don't know the inner workings of a camera company, but I would guess it's an expensive, time consuming process to develop new chips. I would bet if they had higher res 1/3" ccd's for those cameras they would have put them in by now - no reason to hold back while other companies have them. With all of the other great qualities Panasonic cameras have, there would be no other reason to buy a Sony if they had 1920x1080, for example.
 
Cranky,

Rather than get all hung up on what Panasonic did, didn't do, or labels like "Consumer Camera", "Professional Camera"; the real issue is, does the camera fit one's needs or not? Will the camera produce results that the user will be happy with?

I have the Samsung SC-HMX20C video camera. For a single chip camera, it does OK considering its price. As far as I can tell, it produces 800+ lines of resolution. (I'm not sure how many exactly, because my chart maximum is 800 lines.) The active area for the image sensor is 0.45 inches and the sensor is 6MP.

While the camera has a lot of very nice features and can take reasonably good video, the design of the camera can drive me nuts. I have no control over the gain setting of the camera, I can set the shutter speed, but when I power off the camera, it goes back to auto shutter speed, I have no real control over the iris, and it does not shoot any of the 720p modes. At the time I got the camera my other camera was dieing and I couldn't afford to get an HMC-150.

In as much as I'd love to get an HMC-150, I would be hard pressed to spend the money considering the fact of the poor economy and California State budget problems, and as a contract worker with the college, I'm only working at about 67% of normal load.

The HMC-40 has all the controls over the camera that I would want and it shoots great looking video. So for me, it would work very nicely. However, in as much as I'd love to get it today, I have to wait until February 2010 to see how the new contracts work out.


As I've said before:

(1) The prefect video camera does not exist, there's always some sort of trade-offs in selecting a video camera.

(2) Much of the selection process is subjective and what seems OK for one, will not be OK for another. Different users have different values.



Bob Diaz
 
Bob, I am not bashing the camera nor Panasonic. I am just saying that the lens/sensor block in modern consumer cams is good enough for use in pro cams, which Panasonic did to everyone's delight. If/when I decide to upgrage, the HMC40 will be among the first two cameras I will consider (another one being the HMC150).
 
As for the 40 being higher res, I attribute that to a simple fact: Panasonic hasn't upgraded the 150/170/200 yet with higher resolution sensors, not because there is any other good reason their $3500 and $5000 cameras have lower res than their $2000 camera. I don't know the inner workings of a camera company, but I would guess it's an expensive, time consuming process to develop new chips. I would bet if they had higher res 1/3" ccd's for those cameras they would have put them in by now - no reason to hold back while other companies have them. With all of the other great qualities Panasonic cameras have, there would be no other reason to buy a Sony if they had 1920x1080, for example.

The 150 has only been out about a year, it's hardly out of date. There is no CCD 1/3" camera with the low light capability of the 150 and greater resolution (the Sonys you mention have CMOS , not CCD). Engineers have to make trade offs with cost, size, low light, resolution, you can't have it all I'm afraid. I suspect the 150's CCDs are pretty close to state of the art. If you want higher resolution Canon offer this with poor low light performance, or Sony offer it with CMOS and all the problems that brings. For many people, CCDs and low light are more important than resolution, if you want all 3 you need a 2/3" camera, but that comes at a much higher cost.
 
Well said. Don't you think, though, if the 150 had higher res that it would end the need to look at any other cameras in this price range? Why do people go with the Canons or Sony's if the 150 is so solid? Because people still want that HD bang that a higher def image brings? I am both stating that and asking a question. Anyone out there with a 150 that can get eyepopping hd from it? I'm still considering buying one, but the purported softness of the image has held me back, and leaning towards the HMC40 now, even though I'd like better lowlight performance once in a while.

Like you said, tradeoffs, tradeoffs, but the day is near that one camera for $3k will have it all.
 
Back
Top