A55 review now up

[Greg Beetham]

Yes a clock visible in the image and a slave triggered flash, something of that nature perhaps.
Greg

[Neonsquare]

The delay of the A55 and the A580 (only in liveview) seems to be ~0.3-0.4s. My A550 has a delay somewhere between 0.0-0.1s (also in liveview).

There are some customers on different German boards, who seem to have been told by Sony Support that this is a hardware issue that cannot get fixed in a firmware release. A Sony employee in club-sonus.de promised to "escalate" the issue again when first notices about the "will not fix" appeared. It would not be the first time, that Sony support employees put false reports out, but it is certainly one where one can really hope for them being wrong .

[Dr. Harout]

Just to change the subject a bit
Shot with a55 + soviet made 500mm f/8 mirror lens, hand held

[pakodominguez]

Doc!
where is the donut-bokeh!

[OneGuyKs]

Not Found
Sorry, but you are looking for something that isn't here.

[David Kilpatrick]

That should now be changed to a registration message.

David

[Neonsquare]

I've created a small figure with common shutter delays using flash based on measurements collected from "Imaging resources" reviews:

[DavieR]

Thanks for the table Neonsquare. I have to admit that this flash/shutter lag issue is preventing me pressing the "Buy" button. Imaging Resource don't have any shutter delay with flash info for the A900, so I'm unable to compare.

Do you think this is a Firmware issue or a Hardware one, as I've read here - http://www.dslr-forum.de/showthread.php?t=776832


EDIT: Thanks now correctly attributed to Neonsquare for the table. My apologies.

[David Kilpatrick]

Anyone get any idea what kind of large clock might have a very fast sweep hand? I have a watch with this, but the second-by-second movement of any clock I can find isn't helping with my tests.

The delay is significant, but I think the table from Neonsquare allows for pre-flash operations (as the NEX for example has no option otherwise).

I'm getting, on preliminary tests, about 0.3 seconds delay for the A580 and 0.6 seconds for the A55 with manually set, MF, dedicated flash units not a sync adaptor.

David

[Philip]

David, can you not download freeware clock software to run on your Mac which has a millisecond counter, and use the computer screen as the clockface? Just an idea as there are a number of freebe programs out there.

Philip

[Greg Beetham]

I don't see the reason of pre-operation measurements, as far as I'm concerned the time should start from shutter button press to the image formed by the flash pulse, so I'm in the dark (ha ha) with the minus measurements on that chart.
As far as timing it goes, I think a dark room and the recording camera would have it's shutter open for the duration, but I can't figure out how to synchronize the shutter press and the time recorded start, at the same time regardless of what clock is used.
Greg

[David Kilpatrick]

The Mac screen would not be much used photographed using flash

It's simple enough to do what I am doing. I am using a single wireless remote control which triggers two or three cameras at once, maybe even four once I get the NEX set up as well. What results is a sequence of differently timed flash exposures.

This method does not allow for the latency if any present in the infra-red trigger command, but my experience is that this is not something which delays any more than a regular shutter release.

I need a clockface with a regular second hand plus a one second per revolution sweep hand, or something between that and a one minute per revolution second hand.

The intended camera will probably end up being A700 (about as fast as anything), A580, A55, NEX-5 but if I can find enough lenses and cards and batteries (not to mention tripods...) I may add the 900 as well.

David

[Greg Beetham]

The important bit is knowing the exact time that the shutter button is pressed, or the remote is pressed, that's when the elapsed time starts, how to begin timing exactly when the camera begins simultaneously without error...that's the leetle problemo. The remote would have to also start the timer at the same instant to stand a chance of being accurate. I know the 5600HS(D) has many available timed pulses at various Hz spaces and quantity but I just can't figure out a long exposure with those that would capture the entire sequence, there is still the question of exactly 'when' it was that the button/remote was pressed.
Greg

[Neonsquare]

The green bars show the difference to the timing of the A55 as reference value. So the A550 is 0,26s faster than the A55. The timings are done by Imaging Resource in their reviews and even show three digits after the comma. The timing would typically contain AF, metering and TTL pre flash. But their reviews also show other timings. Unfortunately there is no flash timing with shutter half pressed but only half-shutter timing without flash. I think the prefocusing and saving of exposure should give you somewhere around 1/10s (compare the times between the different alphas - thats quite constant).

Following this the A55 seems to be between 2/10s - 3/10s slower than the very good A550. It is 1/10s - 2/10s slower than more mediocre models. The A560 seems to be nearly 2/10s worse than the very good A550 and perhaps 1/10s slower than less stellar models.

To me three questions arise:

1) Why is the A560 nearly 2/10s slower than its successor A550? (presumably both in OVF mode!)
2) Why is there such a delay for dedicated system flashes even when the flash is set to "manual"?
(Remember: non-system flashes and radio triggers show no such a big delay)
3) Even if this is not fixable in software - why should that mean "not fixable" when Apple e.g. fixed peeping MacBook Pros by changing motherboards.

so far

[Greg Beetham]

I'm getting the impression that it's the processing time for the liveview side of things that's causing the delay; but the A550 has liveview anyway if I'm remembering correctly, so that might not be the whole story either.
Greg