How Is Panasonic Keeping Up With Sony

Wow $8,000 for the HPX 500 nice... This client wants cameras to think I was going to get the EX1 I might just go ahead and get the 500 and a 200A.

Any links barry?
 
After playing around with EX1, the EX3 and the Hpx170, out side all day at cinegear, I must say the 170 is very nice, but I do believe that the EX cameras still have a FEW more stops of latitude than the HPX, other than that the HPX is an awesome cam, I am going to get to borrow the ex1 in a few weeks for some tests, after that is done, I will see if i am stay with panny or going sony. BANDPRO sure makes sony sound good, but at the same time Jan at Panasonic make the HPX sound like a bad ass.


Jon
 
Yes I thought it very weird that Bandpro would have a flyer that said, "Thinking about buying a Panasonic, call us." I think that is weird. Why not try to sell the product or perhaps theyre aren't enough customers interested in the Sony product so they have to intice Panasonic customers into their booth so they can try to switch them. ;-)

And unless they had both cameras in the booth for comparison, I do not think you will see the dramatic difference that you think is there. Barry and I have have them seen them side by side.

While you have the EX1 for the test, make sure you analyze the pan and the clarity of the background once you stop the motion. Look at the buildings and look for skew.

Try to input metadata before you record, and I mean something really meaningful like scene name, actors in the scene, location that sort of stuff.

Try to go from record to playback in less than 2 seconds and vice versa.

And don't forget to look at the color black in pure daylight or under tungsten lighting, or just any footage under HMI lighting and not just the flicker-free.

Also try to set up the flip in the LCD, or any pre-record, or loop function. Compare the Waveform Monitor and Vectorscope function, along with the Chroma Aberrations during the zoom. The decision should always be about the pixels in the imager, as the camera is a system of features and functions and how they respond to a myriad of challanges.

There are a lot of issues with CMOS and the EX1 shows all of them.

Best,

Jan
 
dantewaters said:
Wow $8,000 for the HPX 500 nice... This client wants cameras to think I was going to get the EX1 I might just go ahead and get the 500 and a 200A.

Any links barry?
Click to expand...
B&H sells it for $9,995.00, and after the $2,000 mail-in rebate your cost is $7,995.

That doesn't include a lens or power though; 2/3" cameras are usually (always?) sold as body-only.
 
Don't know what other territories may be offering. The $2,000 rebate offered in the USA is from Panasonic US. Every territory sets their own marketing policies, so you'd have to ask Panasonic Europe what their offers are.
 
While you have the EX1 for the test, make sure you analyze the pan and the clarity of the background once you stop the motion. Look at the buildings and look for skew.
Click to expand...

Just a quick note... maybe I'm doing something wrong with my EX1... I'm not seeing skew in my pans? I've tried all kinds of settings and I'm whippin' that sucker faster than I would ever use for footage, and I don't see skew. The only time I saw a small amount was when I was fully zoomed, wide open at 1080/24, and I was zooming quite fast (not a useable pan speed, in my opinion, for any kind of good footage... not even of Sasquatch). Even then I only saw it when I paused the video.

I've also walked around shaking the camera and wobbling it quickly (not like I'd even do that while shooting 'real' footage) and I was never able to create the jello effect, not even a wobble effect (apart from my own wobbles from tilting the camera rapidly back and forth). So seriously, I either got an amazing production model or I'm doing something wrong.

Now I must admit I do see some jittering motion effects on pans (some not even so fast), especially at 24p (remember, this video stuff is new to me, I'm a still photographer) so I did some checking on these forums and I see many threads about it, even in the HVX threads, and I'm reading that is common to all cameras shooting 24p, and even common with fast video pans at most settings. So that's not a CMOS effect. I'm not doubting CMOS issues can exist, but with my EX1 I'm just not seeing it (not yet anyway) and I'm panning and wobbling it faster/more than I'd ever do when shooting for "realz".

I'm not endorsing CMOS cameras, I'm just trying to reconcile what I read here and what I'm seeing in reality.

My monitors are a 30" Apple Cinema Display and a JVC DT-V24L1DU. I'm giddy with this footage... my wife, who is not a media-freak like me (ha!), comes in to watch my footage and she's blown away too! Especially with the slow-mo stuff of my 2 year old boy... this is so cool! I can't wait to slap this thing on the deck of my kayak and paddle through the tufa towers at Mono Lake.
 
Last edited:
Thats funny. BandPro guys say they are strictly Sony, weird, I didnt see that, I will know after I get an ex1 tester and sit down with it an figure it out, we did an ad with one, but I was only on set with it.



Jon
 
Skew due to rolling shutter is easy to calculate

Assume a vertical line moves from one side right or left to the other side in 3 seconds. The Sony EX-1's rolling shutter has been measured to take 1/60 of a second. Thus, in 3 seconds or vertical line travels across 1920 pixels and the shutter takes 1/60 of a second.

1920 pixels / 3 seconds for pan / 60 (the fraction of a second) = 10.7 pixels offset

Thus our vertical line as viewed from top to bottom is offset horizontally by 10.7 pixels.

ARCSIN (10.7 pixels / 1080 pixels) = about 0.6 degrees
So, possible to see, but slight...

Now let's assume the pan took 1 second to travel from one side to another:
1920 pixels / 1 second for pan / 60 (the fraction of a second) = 32 pixels offset
ARCSIN ( 32 pixels / 1080 pixels) = about 1.7 degrees

A whip pan of 0.5 seconds to travel from one side to another:
1920 pixels / 0.5 seconds for pan / 60 (the fraction of a second) = 64 pixels offset
ARCSIN ( 64 pixels / 1080 pixels) = about 3.4 degrees
Now we are getting to a point, where it's easier to spot.

The skew really stands out when we have some sort of reference. If there's a vertical line that does not move that is next to the line that does, it's easier to spot the skew. Also a quick right to left pan followed by a quick left to right pan will make the skew appear to double and it will stand out even more.


Vertical pans present an interesting distortion. Panning upward stretches the item and panning downward compresses the item. The 1080 vertical pixels are shifted as follows:

1080 pixels / 0.5 seconds to go from top-bottom / 60 (the fraction of a second) = 36 pixels

This compression/expansion made for a weird effect when someone put a CMOS camera on their motorcycle and drove it around. The bumps made the world expand and compress in the vertical direction.


IF any pan on a CMOS camera is 3 seconds or longer from side to side, the skew is hard to see.


Bob Diaz
 
Bob, that's an interesting formula. Thanks for the science behind it. I'll have to try counting when I'm panning to see how long it takes for pans that I consider 'faster than I find useable'. It might still fall within the acceptable confines of your formula, I don't know. It gives me another test to perform! I have been using fence lines and houses for my vertical lines.

The only 'fast pans' I would realistically encounter with what I'll be shooting would be running deer/bear, etc., and birds in flight (which probably would be less noticeable against a sky or a very blurred background). And my personal holy grail of footage will be of Bigfoot. If I can get a crisp shot of Bigfoot on an EX1 as he lumbers slowly along a river bank, I'll be really really rich and famous. Maybe I'll still hang around here, maybe I won't. I'll try and stay 'me', but no guarantees.
 
Buck Forester said:
I've also walked around shaking the camera and wobbling it quickly (not like I'd even do that while shooting 'real' footage) and I was never able to create the jello effect, not even a wobble effect (apart from my own wobbles from tilting the camera rapidly back and forth). So seriously, I either got an amazing production model or I'm doing something wrong.
Click to expand...

Put the EX1 on a tripod. Put the tripod on a hard surface like a sidewalk. Turn steady shot OFF. Lock everyting down. Rap on the handle (lightly) with your knuckles. You can easily see the jello-wiggle on the LCD.
 
BobDiaz said:
Skew due to rolling shutter is easy to calculate
Click to expand...

Shouldn't the speed of the pan be in that equation? A 180 degree pan (rotation of tripod head) could take 3 seconds and so could a 15 degree pan. Wouldn't this create a greater off set? Or would the degree still be the same but the faster speed would create a missaligned blocking look?

Example:
||||||||||||||
||||||||||||||
||||||||||||||
Normal CCD Image

//////////////
//////////////
//////////////
CMOS Degree Skew

//////////////
//////////////
//////////////
CMOS Degree Skew with speed offest
 
Put the EX1 on a tripod. Put the tripod on a hard surface like a sidewalk. Turn steady shot OFF. Lock everyting down. Rap on the handle (lightly) with your knuckles. You can easily see the jello-wiggle on the LCD.
Click to expand...

Thanks, Tom, I'll have to try that. I'll be sure to put wife in front of the camera so I have something pretty wiggling back at me. :)
 
Zander said:
Shouldn't the speed of the pan be in that equation? A 180 degree pan (rotation of tripod head) could take 3 seconds and so could a 15 degree pan. Wouldn't this create a greater off set? Or would the degree still be the same but the faster speed would create a missaligned blocking look?
...
Click to expand...

The speed of the pan is in the equation. The speed is measured for a line to go from an edge, like the left edge of the image sensor to the right edge of the image sensor.

This could be a telephoto shot, which requires only a slight turn of the tripod to make a line move from one edge of the image sensor to the other edge OR it could be a wide angle shot which takes a bit more of a move to go from one edge to the other. Without question, when in wide angle, we must turn the tripod faster, but the end result is the same; the line moves from one edge to another in xxx seconds.

By calculating it my way, the setting of the zoom does not matter, as long as the line takes xxx seconds to go from one edge of the image sensor to the other edge.


I hope this helps,


Bob Diaz
 
I'm not sure the two companies really have much in the way of competing products, you get what you pay for really. An EX1 costs more than an HVX200/170 and turns in (IMO) superior image quality in most (though not all) situation. The HPX500 with a decent lens will cost more than an EX3 and turn in superior image quality than the Sony camera.

The more you can afford to spend, the better a camera you're going to get.
 
You must log in or register to reply here.
Back
Top