lowlight performance on XL2 and old XL1: no difference ?!

[Jan De Wever]

Well, as said in the title of this thread ...

I am willing to spend the money, and have been using my old XL1 for about years now. Today I had the chance to try them next to each other, in a quiet pro-shop (ESVideo.be). They are Sony Broadcast dealers BTW.

I put both cameras on a tripod, 0dB gain, no ND, 25i (PAL cameras), no shutter, full manual. I set the zebras at 100% (the ones in the XL1 are at 100%). I had both cameras in full wide, no zoom.

When both aimed at the same scenes, and trying to get the same aparture (using the zebras in the picture and a waveform to be sure), both cameras ended up at 3.2 ... I was stunned ... I tried different scenes (less light ...) but the net result remained the same: same aparture.

I was totally disappointed, I must say. I expected to see a major improvement in the low light capacity of the XL2. Take into consideration that all cameras since years have seen a boost in low light performance. Look at the DVX100: in the past I put it next to my old XL1 and it was 2 stops better: same scene, 4.8 on the DVX, 2.8 on the XL1.

I'm really willing to hear about you guys' experience with the XL2, compared (!!!!) the the XL1. I'm willing to buy, but I need better low light.

Please shed some light on the subject ...

[Don Palomaki]

FWIW, the NTSC XL1 zebra is set for 95 IRE per the service manual. Not sure about the PAL version.

[Daniel von Euw]

Look at this file shoot by Johnnie Behiri:

http://www.24puser.com/nightxl2.wmv

All night shots 3200k preset
1/50 o db
From 2:11 1/25 +3 db
Last shot Just for fun,18 db


Brillant. No way to do this with XL-1 in same quality.


regards
Daniel

[Robin Davies-Rollinson]

.such beautiful "clean" images...

Thanks for posting the clip. I'm very impressed once again!

Robin

[Greg Boston]

That night footage from the XL2 has me drooling all over myself. Mine is supposed to get here next week. I will post some footage if Chris has room to host it.

-gb-

[Chris Hurd]

I've got plenty of space, so send me an email when it's ready -- just make it good!

;-)

[Barry Goyette]

Jan

Something isn't right in your math....I haven't put the xl1 next to the xl2 yet, but I have put the dvx100 next to the xl2, and the two cameras are within 1/3 of a stop of each other...not 2 stops.

Barry

[Matt McDermitt]

Wow, that is absolutely amaizing. Mine arrives in a week. This footage is looks awesome, THANK YOU for posting!!

[Jan De Wever]

Thanks for the replies guys!

But, I didn't get a real answer to my real question: is the XL2 really more sensitive and better in low light than the XL1?

I've seen the site with the beautifull night footage of Vienna, I knew that allready. Based on that footgae, I had high anticipations.

But, when placing both cameras side by side, I could not find any proof of better low light on the XL2! I expected more from a camera that should be based on 7 years of development (the age of the XL1). Any brand has made tremendous progress: look at the PD170: it's by far more sensitive (allthough more grain in the picture, due to electronics and a bad lens). And I stand by my point: I tested the DVX100 compared to my XL1, side by side, in the same way: in this case in my livingroom, full wide, same subject, using a scope and zebras to make sure I get the same level out of it. The DVX100 beat the XL1 by 2 full stops.

Now under the same circumstances, I can't find a difference between the XL1 and XL2.

I'm one of the biggest fans of the XL1. I've owned it since it came out, have defended it against many techheads and Beta-lovers. Taken it all over the world. But it is lacking in low light by today's standards. The XL2 should be significantly more sensitive than the XL1. By my test, it is not. And that after 7 years?

Has anyone actually put the XL1 and XL2 next to each other? Can we compare tests? The Vienna footage does indeed look amazing. No discussion about it. But there is no reference. If the same footage had been shot with the XL1, and put in A/B (splitscreen), that would really show something.

I'm holding off buying untill I've seen more of these comparisons.

[Kevin Lepp]

im confused. if you saw that nighttime footage, then how can you say that there has been no improvement? No camera out there in this category matches up to this outstanding lowlight performance. Even if it is only that the pictures are super clean with almost no noise, that in itself is what makes lowlight capabilities acceptable. Other cameras may be able to go to the same area, but none are as clean and efficient in low light.

And still thats only whats been proven so far... much more may come into view as the days go on.

[Rob Lohman]

Jan: I think the difference is not in how much lights the camera
needs in comparison to the XL1S. I think the difference is in the
quality of (very) low light footage. In other words you can
get by with much less light since it will introduce much less noise
and other things like artifacts. So the signal is more clean then
was the case in with the previous camera.

So I think the right comparison would be in a dark environment
to see how the images compare with both camera's at identical
configurations.

In essense I think it is about quality / performance and not about
how much light you will need (to expose properly).

[Jan De Wever]

Hey Rob: to a certain level, I can agree with you.

I'm from a broadcast background, and I've seen the evolution of modern cameras there. But modern BetaSX and BetaDigi cameras can shoot scenes with only candlelight, at 0dB gain, with a poperly lit face! That's pretty amazing. That's now. 7 years ago, you couldn't do that at all! You needed lights to add light ...

Yesterday, I was shooting at a party-event. A party tent, scarcely lit, but with a lot of atmosphere. A 7 year old Beta couldn't shoot there without extra lights. So couldn't my XL1. I needed to use lights, thus ruining the atmosphere, and giving me a lot of angry looks. A modern broadcast camera could shoot that scene whitout any additional lights!

Don't get me wrong, I don't ask Canon to produce a camera that can compete with such cameras in the low light department. But I expected to see some progress. And despite the better image quality, you should be able to shoot the same scene, correctly exposed with less stops open than the XL1 on a modern camera. The DVX can do it, the PD150/170 can do it. Why not the XL2?

Who has done comparisons? Chris? Ken? Don? ... ?

[Don Berube]

Jan,

I'm not questioning your skill or experience here, so please do not take this the wrong way.

Your statement that the XL2 is not any better in low light sensitivity than the old and long-discontinued XL1 is not entirely true. It is no secret that even the XL1s was better in low light than the XL1, by a measured value of +4db which pretty much equated to 2 stops.

While I have not yet done any testing to determine the exact value, it is at least well known that the XL2 is even better than the XL1s. That much has been noted by many people. The Xl2 is not only better in low light sensitivity than the XL1s that it replaced, it is significantly cleaner. As we all know, it is also *significantly* more resolute.

Here's an example: The XL1s was known to be rated at 320 ISO. The XL2, in 60i Mode at 0db and 1/60th sec shutter speed, has been rated at 400 ISO.

I can also add that with the XL1, shooters would typically set the XL1 to -3db *only* when you were in bright lighting or outdoors in bright sunlight. With the XL1s, because it was more sensitive, we found that we could stay in -3db much more often than the XL1. Now, with the XL2, you can often times simply leave it in -3db, as it is even more sensitive than the XL1s. At the trade shows, there were times I would set my demo XL1s in -3db but would often have to put it back to 0db. Now, I have found that I can pretty much always leave the XL2 in -3db.

Another somewhat 'real-world' example, as Barry noted and many else have seen, is that the low-light sensitivity of the new XL2 is similiar to the low-light sensitivity of the DVX, deviating by just a small amount. Again, as Rob points out, the XL2 is one camera that is super clean and remarkably noise-free to boot.

So Jan, I am not entirely sure just exactly what procedure you are using, but I will confidently and respectfully say that your statement that the XL2 is no better in low light than the XL1 is incorrect.

I will make it a point to do some more exact and 'scientific' testing in the following weeks, schedule permitting. I invite anyone who has the time to offer their input to please do so.

Wishing you the very best,

- don

[Jeff Donald]

The difference between 320 ISO and 400 ISO is only about 1/3 of a stop.

[Don Berube]

Yes Jeff, that is correct.

Jan wrote: The DVX100 beat the XL1 by 2 full stops.

If you read my post, it reminds that the XL1s was better than the XL1 by a factor much greater than 1/3 stop. The point is that the new XL2 is even better than that.

Thanks for sharing,

- don