Hvx200 vs XH-A1 lowlight

You know Barry, I think this subject is a dead horse too. You and I know that either cam has an obvious noise signature--it's just in the rendition of the noise pattern where people diverge in their impression of it.

It might be that the Canon look seems less noisy because the "chunks" of noise appear smaller...like you said, "grittier", but it certainly isn't noiseless by any means.
In marginal light it will show its dance of grit--the question is if it's tolerable in a pinch. Personally, I think it is and of course there are ways to kill noise with either cam, but they aren't "no light" cameras. They work well in most real world situations, but they are small sensor HD cameras.

Sorry for the rant.

OK, I think the fair thing to do would be to not crush the blacks on either cam and shoot with a preferred Cinegamma, since most will shoot at least with a Cinegamma engaged. Also, no detail coring engaged because that will soften the picture even if it kills some noise.

On the A1:

Everything neutral but these:

Cinegamma 2
Sharpness: -3
Color Gain: +15
Knee: Low
NR1: Low

Shoot with 1/48 shutter on either cam because no one likes the slurry look of 1/24.
Btw, since the HVX has no specific noise reduction modes I suppose NR1 Low should be optional to be fair.

Shoot a low light (candlelight?) scene with 0, +6 and +12dB on the HVX and -3 (since this is the lowest mode on the cam), +6 and +12dB from the A1.
 
Last edited:
btw, I know bandwidth is an issue and that's why pics will be the output, but I think it's a bad idea because analyzing stills with JPEG compression is just another layer of complexity. Besides, we're talking about video cameras here, right?

The only way to get to the bottom of the noise thing is to see a few clips from both cameras in their unaltered, native formats, which of course would be too bandwidth intensive, especially for DVCPRO HD @ 100 mbs.

We need to see the noise as it is rendered by each camera in its native recording format in first generation.
 
Okay, just played around with it some more. I took Ash's suggestions from DVInfo, then went further, trying to yank as much as I can from it. And there's a fair bit more that can be pulled out of the XHA1, I must say!

First, Gamma Normal is way way brighter than either of the Cine gammas. Same with the HVX, you can get a much brighter picture from Gamma HIGH than you can from Gamma CINELIKE-D. My prior comparisons were CINELIKE-D to CINEGAMMA-2, but since we're talking about finding maximum low-light performance, I figure I should optimize 'em both for the brightest I can get, regardless of whether it's the "look" we want or not, right?

Second, the noise reduction does help; using Ash's suggestion I set it to NR 1: LOW. With that in place the XHA1 does look less noisy than the HVX. I'm not sure what deletirious effects it may have on the image; I know there's some mention of ghosting tied to some NR function, and it's gotta be doing *something* to the image, or else they'd enable it all the time by default. There's a reason it's not set by default, but I haven't figured out what it is yet. Because on a static shot of some trophies on the wall, it seemed to just drop noise with no other obvious side effect.

I shot using one fluorescent 55w bulb, pointed away from the shot and bounced into a wall; couldn't find my light meter but I'd wager that it was really, really low illumination. Both cameras deliver approximately the same brightness, with the XHA1 having a very small edge, maybe (maybe?) 1/4 stop. When fully gained up to +12, they both were able to blow out highlights in the scene, which was impressive, but they were also both so noisy that I wouldn't use either under those circumstances. But I will say that the A1 was less noisy than the HVX, even with both at +12dB, by a noticeable amount (which is like saying a 400-pound woman is skinnier than a 450-pound woman; I mean, yes it's true, but does it really matter that much?) :) I think they were both too noisy at +6dB, but I know that people's opinions will vary. But at +12dB there was a difference not just in the character of the noise (HVX = smeary/splotchy with blotches of color, XHA1 = grainy/bitty and b&w) but also in the sheer amount, with the nod going to the XHA1.

(note: I was viewing the noise at something like 4x magnification on the DV Rack monitors; at fullscreen they both looked comparable with maybe a small advantage going to the Canon, but you'd have to stare at the screen pretty hard to notice and/or care.)

In short: yes the XHA1 can deliver a slightly brighter, less noisy picture under extreme low light conditions. With normal image settings I found them very comparable, but with adequate tweaking you can clean up the Canon's image to the point where it does make a difference. Not enough of a difference to say "hey, this one's usable and that one's not", more like judging a swimming suit competition between a 400-lb guy with a hairy back, and a 350-lb guy with a hairy back, both in Speedos. In other words, you don't really want to look at either of 'em... :)
 
Last edited:
Elton said:
btw, I know bandwidth is an issue and that's why pics will be the output, but I think it's a bad idea because analyzing stills with JPEG compression is just another layer of complexity. Besides, we're talking about video cameras here, right?
Click to expand...
Yep, and looking at stills is nearly pointless because you're looking for noise, and noise doesn't move in stills!

The only way to get to the bottom of the noise thing is to see a few clips from both cameras in their unaltered, native formats, which of course would be too bandwidth intensive, especially for DVCPRO HD @ 100 mbs.
Click to expand...
Well, and processor intensive for the HDV clips. My lame-o desktop can't play either of 'em back at full frame rate. The only way to really analyze and judge it is to use DV Rack's monitor on magnify, because it can show full frame rate and super-magnified so you really get to see what's actual noise in the signal, vs. what's amplified or introduced by whatever display system you're playing it back on.

We need to see the noise as it is rendered by each camera in its native recording format in first generation.
Click to expand...
Again, that's what makes the DV Rack solution practical -- you can even see it live, because it's monitoring the firewire output.

But it's all nearly pointless anyway, because you wouldn't use the image from these cameras under these scenarios anyway.

And in bright light condition, they both look fantastic, and in properly-lit conditions they both look fantastic.
 
Thanks for the thorough description Barry. I think it's a fair and realistic assessment, and I hope the subject can be put to rest now.

It's just great to see that there's a nice HD alternative at a very affordable pricepoint now. It'll be interesting to see if there'll be a few A1 entries in the dramafest.

What do you think about the idea of a future short competition in HD from various cams, with the top 10 ending up on an HD DVD?
 
So far, I think the A1 is the best HDV camcorder of the bunch. I don't know how the fest will go, I'm not one of the bosses on the fest, that's all in the hands of Jarred and Barry_S. I wouldn't expect a high-def DVD but maybe a computer-playable WMV9 HD DVD would be possible... I mean, at least that way everyone could play it! :)

Depending on how nicely everyone can play along, I'd like to see DVXUser be more open to the other cameras; use in festivals, included in discussions, etc. What I do not want to see is another repeat of last year's hatefest discussions. It's a different landscape now. We in the DVX world had, largely, the world to ourselves for a long time -- Sony never answered the DVX, and Canon's XL2 was very late and wasn't really price or feature comparable, it was a totally different product, so the DVX reigned supreme among indie filmmakers for three to four years. But now... now the little Canon and the promising new Sony are providing some serious alternatives, depending on what feature set people want. And AVC-HD is right around the corner.

Getting back to these two, the feature sets are still very different, even though the imaging is getting downright indistinguishable, at least for anyone other than the pixel-obsessed. I played about 20 minutes of footage on my XBR960 TV (top-of-the-line Sony CRT at the time I got it) and made my wife watch it as I swapped the cable between the two cams. She couldn't tell the difference, and usually when I was watching it (from two feet away, mind you) I found that sometimes I couldn't really tell the difference either -- as in, I'd forget which cam I had plugged in at the time, and try to fast-forward on the A1 and see no reaction on the TV so I had to say "doh! I'm plugged into the HVX..." Yes if you split-screened it you could tell they were different, but -- why bother? They both look frickin' fantastic. It's getting to the point where arguing about image is wasted breath. What matters now is things like zoom reach, frame rates, audio control, shoulder-mount vs. handheld, tape vs. tapeless, price point, 35mm adapter performance and whether MPEG affects it, all that stuff. I think there are still some very clear-cut reasons why someone should choose one of these over the other, and there are very good reasons why they're in different price brackets, but those are features differences, not necessarily image differences (at least, under normal circumstances).

I think the battle isn't really going to be A1 vs. HVX at all, because I see them appealing to two very different customers. Where the battle looms is going to be Sony V1 vs. Canon A1. The HVX buyer is going to buy an HVX, but the HDV buyer... well, there's the question. There the feature sets are matched across the board (20x vs. 20x, HDV vs. HDV, 24P/30P/60i vs. 24F/30F/60i, $3999 vs. steet $4199, etc). Neither really does anything the other one doesn't, excepting the laughable "smooth slow rec" function. And in the V1 vs. A1 comparison, image may very well be an issue, since according to Adam Wilt the V1 is about 1.5 stops slower than the HVX (which should mean it's about 1.5 stops slower than the XHA1 too) and that could make a difference. And, then again, the V1's CMOS chips might have a big latitude difference over the A1's CCD. And then there'll be vertical smear vs. no-smear, but also "rolling shutter" vs. CCD electronic shutter... that's where the battle royale is going to take place.

I just hope everyone can agree to play nicely with everyone else. Last time we had to call a time out and kick people out of the pool, let's try just behaving this time! :thumbsup:
 
Last edited:
Thanks, Barry. For a second, I really thought I was crazy when I kept seeing a "slightly" cleaner A1 image.

Now, the question I really wanna know is: The chroma smear/HVX noise. Is this why it's so difficult to correct an image in an NLE? If the Canon's noise is girtty/grain and black and white, is that why it seems to be more pleasing to nudge colors or contrast here without the blacks turning spotty-green?
 
Thanks Barry, :dankk2:
I know that 'bring down the gain en set in the NR1' was working.

Both cams are wonderful in terms of possibilities aimed to different users.
 
Barry again thanks for all your time and effort. I am ecstatic that everyone can now get over this and move on to more important things.

And also I promise to play nicely as the future brings change.
 
awesome work barry. i'm just happy that i picked up one of the top performing hd camera at an awesome price point. canon or panny, pixels or smears, it's all about the script.
 
Barry... your statement should be a sticky or a mantra.. or tatoo'd into people forheads.

Explained perfectly. I mean, I've seen some footage from these two new cams and they look great. It will come down to feature set and what people need and want to pay.. I'd still have bought my HVX, but I'm looking forward to getting another camera and with that new blackmagic card.. some of these look really good.
 
Kholi said:
Now, the question I really wanna know is: The chroma smear/HVX noise. Is this why it's so difficult to correct an image in an NLE? If the Canon's noise is girtty/grain and black and white, is that why it seems to be more pleasing to nudge colors or contrast here without the blacks turning spotty-green?
Click to expand...
I'm not a post-master, so I can't answer this; I do all my color correction in-camera whenever possible so I don't know. Most people that I've heard discuss it say that DVCPRO-HD can be pushed a lot further than HDV, so I don't know why you'd be having such difficulty with it, but that would be a topic for another thread. Now that I have similar imagery from both, once the article is posted you could take the same shot from both cams and tweak it (I'll supply an uncompressed version in TIF or TGA or BMP or whatever for those that want to do so).
 
Barry_Green said:
I'm not a post-master, so I can't answer this; I do all my color correction in-camera whenever possible so I don't know. Most people that I've heard discuss it say that DVCPRO-HD can be pushed a lot further than HDV, so I don't know why you'd be having such difficulty with it, but that would be a topic for another thread. Now that I have similar imagery from both, once the article is posted you could take the same shot from both cams and tweak it (I'll supply an uncompressed version in TIF or TGA or BMP or whatever for those that want to do so).
Click to expand...

Thanks, Barry! Sorry to ask for so much. LoL. But, I'll do it. TGA would be best!

I'll keep an eye out for the thread.
 
I've just spent a few hours going through and exporting them; they're PNGs right now because that's what Vegas exports as an uncompressed file. Still a couple more tests I want to run, but it's getting put together...
 
I visited Bel-Air and got a chance to play with the HVX200 and XH-A1 side by side. Connecting both to a JVC pro-HD monitor. It was great to be able to touch and hold this damned thing and really know more about it. A1 is a bad ass camera, all the way. I'm very much considering selling the HVX200 now and getting an A1 with Firestore.

But I had a thought about the LOWLIGHT thing and why it's important. For me, blasting a scene with light isn't out of the question because of where I work. But, I won't be able to carry those lights around all the time, you know? That's where Lowlight comes in handy.

Say you only have access to a few lights with an HVX200. Your only option is to get what you can and hope for the best. With the A1, you get an image that's probably going to be a bit better, regardless of the light that you can or cannot use. That's why the lowlight possibilities are such a great thing.

We can't all afford to blast a room with light so that we can avoid a noisey image. Even with the great settings that other have discovered. That's just a thought, and I think it's rather valid for indie filmmakers and users alike.

I also don't think that 24f looks anymore "video" than the HVX's 24p. Or, that Canon's rendition of life looks anymore video by any means. I certainly do think that the A1 looks more HD than the HVX200.

I am definitely going to shoot a short on the HVX200 really soon. That'll give me abetter idea on what I want to do as far as selling/trading.

Looking forward to seeing the PNGs, Barry. And again, thanks for the testing and all the info you've given me.
 
Barry_Green said:
In short: yes the XHA1 can deliver a slightly brighter, less noisy picture under extreme low light conditions. With normal image settings I found them very comparable, but with adequate tweaking you can clean up the Canon's image to the point where it does make a difference.
Click to expand...
bravo! that one me likes it! after all, coming from where's coming from..
 
You must log in or register to reply here.
Back
Top