Frame rate and jello

[ydgmdlu]

Mmmm... so the sensor scan rate is dependant on frame rate...

It's the other way around, in my opinion. I'm hypothesizing that the frame rate is created from the sensor scan rate.

Is there any reason why it couldn't select a scan rate of 60Hz for 24fps which would give it the same amount of skew/jello as for 60fps... because (for the GH1 abd 7D at least) the sensor is capable of the faster scanning?

If I'm correct and frame rate is derived from the scan rate, then the scan rate would need to be a multiple of the frame rate, which is why 60 Hz wouldn't work. The scan rate for 60 fps probably isn't 60 Hz anyway; that was just a number that I used for the sake of argument. It's probably 120 Hz (for 60 fps). Also, we don't know that the sensor is capable of scanning that fast. Why would you assume that? The 7D and the GH1 both have very different sensor designs from the 5D and from each other.

If you're wondering why the sensor couldn't just scan at the same maximum rate for all frame rates, then you've missed the whole point. I'm saying that the frame rate is tied to the read-refresh clock. In order to change the frame rate, you have to re-time the clock. That's why you can't use, for example, 120 Hz for all frame rates. Each frame rate must have its own clock speed.

Nobody here, that we know, is a Canon engineer who worked on the camera. So nobody can really answer your questions to your satisfaction. I started this thread merely to offer an explanation for the differing amounts of jello seen at different frame rates. There's no point in speculating beyond that. By that, I mean that it won't get us anywhere.

 

[ydgmdlu]

Canon already has a warning in their manual saying that extended recording can cause a drop in video quality (due to overheating). Running the chip faster than the minimum necessary speed will result in more heat generated.

Yeah, that's definitely another reason why Canon would choose such low scanning rates for the DSLRs. It's all about minimum necessary speed.

 

[mattsand]

Sensor scan rate is a fixed aspect of the hardware

max rate is, what they are actually clocking it at is variable, just like any chip.

It can't be changed.

in this case maybe, maybe not. if they are clocking it slower than possible to save battery, minimize heat, or whatever arbitrary reason like ease of design and minimizing buffer use, it could definitely be made faster.

 

[Luis Caffesse]

I think ydgmdlu's point in saying 'it can't be changed' was in the context of someone asking if the sensor scan rate was a selectable option in the camera (via menus, etc).
So sure, in theory, it could potentially be changed by the manufacturer - but not by the average user.

 

[ydgmdlu]

I think ydgmdlu's point in saying 'it can't be changed' was in the context of someone asking if the sensor scan rate was a selectable option in the camera (via menus, etc).
So sure, in theory, it could potentially be changed by the manufacturer - but not by the average user.

Yes, that's exactly what I meant. Apologies to everyone for sounding confusing and self-contradictory.

 

[mattsand]

i see, there are actually people who think they can change it themselves? the varying level of intelligence on this board amazes me every day.

 

[Ian-T]

...well I sure hope you didn't expect us all to be as smart as you..

 

[SimonPatrick]

i see, there are actually people who think they can change it themselves? the varying level of intelligence on this board amazes me every day.

Well, if you don't ask the questions, you might not learn. Don't confuse knowledge with intelligence. I'm pretty sure we don't come out of the birth canal with an inbuilt knowledge of sensor technology...

 

[SimonPatrick]

It's the other way around, in my opinion. I'm hypothesizing that the frame rate is created from the sensor scan rate.


If I'm correct and frame rate is derived from the scan rate, then the scan rate would need to be a multiple of the frame rate, which is why 60 Hz wouldn't work. The scan rate for 60 fps probably isn't 60 Hz anyway; that was just a number that I used for the sake of argument. It's probably 120 Hz (for 60 fps). Also, we don't know that the sensor is capable of scanning that fast. Why would you assume that? The 7D and the GH1 both have very different sensor designs from the 5D and from each other.

If you're wondering why the sensor couldn't just scan at the same maximum rate for all frame rates, then you've missed the whole point. I'm saying that the frame rate is tied to the read-refresh clock. In order to change the frame rate, you have to re-time the clock. That's why you can't use, for example, 120 Hz for all frame rates. Each frame rate must have its own clock speed.

Nobody here, that we know, is a Canon engineer who worked on the camera. So nobody can really answer your questions to your satisfaction. I started this thread merely to offer an explanation for the differing amounts of jello seen at different frame rates. There's no point in speculating beyond that. By that, I mean that it won't get us anywhere.

Hey thanks for being patient with my questions. I've learnt something, I think (especially after re-reading your original post - and finally realising its intent as an hypotheses)

 

[mattsand]

Well, if you don't ask the questions, you might not learn. Don't confuse knowledge with intelligence. I'm pretty sure we don't come out of the birth canal with an inbuilt knowledge of sensor technology...

what i meant was i *don't* think people are stupid. irony doesn't translate very well online i know. everyone who's brought up controlling sensor speed this way in this thread have done so to illustrate that it's quite obviously not possible.

 

[SimonPatrick]

what i meant was i *don't* think people are stupid. irony doesn't translate very well online i know. everyone who's brought up controlling sensor speed this way in this thread have done so to illustrate that it's quite obviously not possible.

Irony, on the other hand, can't be taught ;-)