Global Shutter Coming to Canon DSLRs? [CR1]

[DudeInTheSky]

CFast is based on Serial ATA which is not being developed past the current 6Gbit/s. Read for yourself at http://serialata.org http://www.compactflash.org

16Gbps SATA:
http://en.wikipedia.org/wiki/Serial_ATA#SATA_revision_3.2_.2816_Gbit.2Fs.2C_1969_MB.2Fs.29
http://en.wikipedia.org/wiki/SATA_Express

It's not 16Gbps SATA, it's 16Gbps SATA Express. Which is different. SATA Express is essentially SATA software stack on top of electrical PCI-Express. (So different connector, etc, etc)

There is of course something to be said about what works here and now! Even if it isn't upgradeable in the same way we have become accustomed to with CF cards.

Yes, SATA Express is the follow-on to SATA, and XQD is the follow-on to CFAST.

http://en.wikipedia.org/wiki/XQD_card

Successors aren't always backward compatible.

I don't see XQD 2.0 as a 'follow-on' standard to CFast 2.0. Both cards are available today at similar speeds. The question is why bother putting CFast 2.0 in a DSLR today if you need to upgrade to a PCI-Express based memory card standard anyways in the near future?

 

[Lawliet]

No, I'd still say it's fantastic - the 1/500 sync that is.

Well, while the D8x0 officially only goes to 1/320 a fast trigger also allows for 1/400 - almost there. Or up to 1/640 in crop mode

 

[Joe J]

I'd like to hear why people want a very fast sync speed. I know quite a lot about this topic and can see direct applications for sync speeds up to about 1/1000th, but at that point applications get a lot more limited.

1/1000- to at least 1/2000 sync works for one primary situation- freezing action in mid-day ambient light, without an impractical car-battery sized power source. Any good portable strobe (i.e. Einsteins) will put out enough output at a fast enough t.1 duration to properly light a subject at lower apertures. Jester237 nailed it with the 1D Mark I mention. Electronic shutters and higher sync speeds aren't new; just Canon and other manufacturers choose to ignore the possibility of advancing them and making better tools for photographers, like the holy grail of true fast sync speed built into cameras. All the HSS nonsense is still flawed (Profoto B1's latest update is the closest to perfection to date) and shouldn't be necessary anyway!

 

[Lee Jay]

I'd like to hear why people want a very fast sync speed. I know quite a lot about this topic and can see direct applications for sync speeds up to about 1/1000th, but at that point applications get a lot more limited.

1/1000- to at least 1/2000 sync works for one primary situation- freezing action in mid-day ambient light, without an impractical car-battery sized power source.

The issue there is that you need to be very, very close to your subject for the flash to make any real contribution to the ambient light. So this tends only to work for macros.

 

[Lawliet]

The issue there is that you need to be very, very close to your subject for the flash to make any real contribution to the ambient light. So this tends only to work for macros.

Lets see, a Move1200L/Para222 nets me f/16@Iso100@10m, about 3 stops over ambient at high noon when used with the LS lenses. With recharge times about in line with the frame rate of the back.
Not my definition of very close or macro.

 

[privatebydesign]

The issue there is that you need to be very, very close to your subject for the flash to make any real contribution to the ambient light. So this tends only to work for macros.

Lets see, a Move1200L/Para222 nets me f/16@Iso100@10m, about 3 stops over ambient at high noon when used with the LS lenses. With recharge times about in line with the frame rate of the back.
Not my definition of very close or macro.

I agree, I can get f16 at 100iso at 10' with a single $500 Einstein. What we don't get is short flash duration at that power, the Einstein is around 1/666 sec t1 duration so much faster than that and you are losing flash power anyway, the Broncolor is an even slower 1/375 sec t1 duration at full power.

In these situations, where the flash exposure becomes the shutter speed, we are still not gaining much because nothing puts out huge amounts of light fast enough. This will change over time, until we get true global shutters there is no inherent need for full power flashes to be short duration most of the time, when there is then the gear will be made, but it is going to cost! Or we need to use more smaller heads for faster t1 times.

What came first, the chicken or the egg? The fish........

Below, midday Florida sun, f16, 100iso, 10'. 640Ws.

 

[Lawliet]

The issue there is that you need to be very, very close to your subject for the flash to make any real contribution to the ambient light. So this tends only to work for macros.

Lets see, a Move1200L/Para222 nets me f/16@Iso100@10m, about 3 stops over ambient at high noon when used with the LS lenses. With recharge times about in line with the frame rate of the back.
Not my definition of very close or macro.

I agree, I can get f16 at 100iso at 10' with a single $500 Einstein. What we don't get is short flash duration at that power, the Einstein is around 1/666 sec t1 duration so much faster than that and you are losing flash power anyway, the Broncolor is an even slower 1/375 sec t1 duration at full power.

That's why I used numbers one gets in speed mode (at 400Ws, not full power), resulting in a t0.1 better than the sync speed of the lenses and the triggers. Helpswith the recycle time as well.

 

[3kramd5]

I'd like to hear why people want a very fast sync speed. I know quite a lot about this topic and can see direct applications for sync speeds up to about 1/1000th, but at that point applications get a lot more limited.

1/1000- to at least 1/2000 sync works for one primary situation- freezing action in mid-day ambient light, without an impractical car-battery sized power source.

The issue there is that you need to be very, very close to your subject for the flash to make any real contribution to the ambient light. So this tends only to work for macros.

Lets see, a Move1200L/Para222 nets me f/16@Iso100@10m, about 3 stops over ambient at high noon when used with the LS lenses. With recharge times about in line with the frame rate of the back.
Not my definition of very close or macro.

But note it doesn't meet one of the original requirements (bolded).
(Granted you could get a lot closer than 10m and use a substantially smaller pack)

 

[Lawliet]

But note it doesn't meet one of the original requirements (bolded).

So much for spec reading vs. actual experience - the latter says it meets the requirement. Compared to speedlights it's actually a reduction in bulk and weight. thats if your intent involves taking pictures.

 

[3kramd5]

But note it doesn't meet one of the original requirements (bolded).

So much for spec reading vs. actual experience - the latter says it meets the requirement. Compared to speedlights it's actually a reduction in bulk and weight. thats if your intent involves taking pictures.

I don't doubt it. It's just silly to respond to a desire to avoid a car battery-sized power source with a rig including a car battery-sized power source.

 

[Lawliet]

I don't doubt it. It's just silly to respond to a desire to avoid a car battery-sized power source with a rig including a car battery-sized power source.

Do you really believe the Move is the size of a car battery?! So much for preconceptions...

 

[3kramd5]

I don't doubt it. It's just silly to respond to a desire to avoid a car battery-sized power source with a rig including a car battery-sized power source.

Do you really believe the Move is the size of a car battery?! So much for preconceptions...

Heh, climb down from your horse, realize that not everything is literal nor serious, but that a 500in3 battery may not power a Tesla P85, it's not small by any stretch of the imagination.

 

[Tinky]

Posted by LuckyDude in another thread, makes sense to post here too as there is a very good demonstration of global vs rolling shutter, and why it's important, towards the end...

http://youtu.be/CmjeCchGRQo