HPX170 ProRes (HQ) and DVCproHD Comparison (footage and full res stills)

K

Kholi

Section Moderator
Alright... this one took a moment and it's nothing special. A few notes before you check the links:

--Letus Ultimate and Zeiss C/Y 135/2.8, 50/1.4
--HPX with adjusted CineD and CineV Files
--Captured through an AJA to ProRes (HQ) 4:2:2
--Shot through a window with "Tint" on it already, so these could have been even sharper. Food for thought


Stills --

www.kholi.net/CarAndGreenDVCProHD.tga
www.kholi.net/CarAndGreenProRes.tga
www.kholi.net/LeavesDVCproHD.tga
www.kholi.net/LeavesProRes.tga
www.kholi.net/OfficeDVCproHD.tga
www.kholi.net/OfficeProRes.tga
www.kholi.net/TreeDVCproHD.tga
www.kholi.net/TreeProRes.tga

Video --

www.kholi.net/Comparison1H264.mov -- Small File
www.kholi.net/ComparisonH264Large.mov -- Larger File

The LEFT side is DVCproHD and the RIGHT side is ProRes (HQ)

The differences are a bit more prevalent in the "raw" footage.

How is DVCproHD? DVCproHD is quite a fine codec in comparison to a Full Raster and efficient codec. Keep in mind that ProRes (HQ) is damned near indiscernible when dealing with Uncompressed Comparisons. It's quite nice.

Is the resolution jump needed? Needed? That's a strong word. Welcomed? I'll take every little bit I can get within reason. Which means I'll take this as well. Mostly because I use a 35mm Adapter ritually. Those who don't? I wouldn't bother.

Is the image fidelity far off? This is my personal favorite. Check out the difference in noise patterns and image clarity all around. ProRes is awesome. Even tough the HPX is kicking out some gained-up images (I will still stand by the HPX being noisier than the 200) the ProRes Codec brings it back to a place that I actually like as much as the 200A. It's smooth, less macro-blocky -- which isn't a term-- and you should go and throw the footage or stills through some grading and see how it holds up. DVCproHD starts showing banding and brings out the blocks VERY swiftly. ProRes? Just rocks. It also seems that DVCproHD washes colors a bit. ProRes retains a fair amount of color detail which is sweet. Throw all that in with the very slight boost in resolved detail and the overall package is a lot sweeter.

Is it practical? Hell no. And, if I had to be tethered to a deck just to get this image quality I better have the budget. AKA two people to sit at a computer, a really computing rig that's extremely mobile, lotsof SDI cable to stay out of my way. Or...

Convergent Design's XDR and NanoFlash units.

What's that mean? I have a strange feeling that Convergent Design's mobile CF reader could be a nice thing to have around. The Sony Codec being used is Full Raster 4:2:2 MPEG-2 and you've got the choice between i-Frame @ 160mbps or LongGOP at 100mbps. Records sound through the HD-SDI port as well. So, it's an alternative that I want to test out here really soon. Seems like the mileage comes in post when you want to grade a bit heavier and you want the extra resolution.

The real weak link? 1/3" CCD Technology... I hope this is the last CCD Based Camera that Panasonic releases.

Thoughts welcomed.
 
Hi Kholi,

If you are comparing, you need to butterfly the image. This way the image on the left is mirror on the right, anything else doesn't count, and you can actually compare the same image albeit reversed.

On the HPX170 being noisier than the 200A, please compare either the compnent analog to the component analog signal or the recorded signals and both over the same signal connection, component analog.

Best,

Jan
 
Jan_Crittenden said:
Hi Kholi,

If you are comparing, you need to butterfly the image. This way the image on the left is mirror on the right, anything else doesn't count, and you can actually compare the same image albeit reversed.

On the HPX170 being noisier than the 200A, please compare either the compnent analog to the component analog signal or the recorded signals and both over the same signal connection, component analog.

Best,

Jan
Click to expand...
Thanks for the advice on comparing.

The ProRes is definitely cleaner than the DVCproHD, not by a huge margin but it's definite.

And, I'm trying to locate an HVX200A right now to test. I know for a fact that they're different. While you're here, I have a question:

Do panasonic techs have the ability to adjust the gain amps in the HPX170? If that makes any sense. The HPX170 looks like the 200A @ +3dB gain.
 
I refer you to the second part of my earlier post. Please compare the recordings or the same analog outputs.

Let me know the difference.

And ofcourse ProRes should be better it is a 10 bit codec. Everything has a price.

Best,

jan
 
Jan,

I spent two days looking at my 170 A-B'd with a 200A through an Extron RGBHV switcher so I could use component out on both cameras. At first I was monitoring composite downconverted SD which was useless, but let me use my Tek scopes. Then I did the component monitoring with an LH80 and couldn't see much difference in noise. It was only when I got my 1700 back from rental that I became aware of the 170 being about 2-3 db noisier than a 200A, no matter which scenefile/gamma/detail setting if the two cameras had the same settings. I even took the HD SDI output of the 170 and compared it to the 200A via component by switching inputs in my 1700 and the 200A still looked cleaner, even though I was monitoring a digital vs. an analog signal.

Surprisingly, the 170 was faster than the 200A, with both at a fairly wide focal length. The extra noise I'm seeing is not just mid-tone, but in the blacks, in chroma and luma. I think the Panasonic engineers had a camera whose lens is 10mm smaller in diameter, which effected long focal length sensitivity, so they raised the gains in the preamps. Effectively, the HPX170 at 0db is like an HVX200A at +3db.

At first I thought the HPX170 wasn't white balancing at tightly as the HVX200A, but now I realize I was seeing more noise on the scopes. I would also like to know if the preamp gains can be reduced or if a firmware update allowing -3db can be looked into for the 170?
 
And it would be faster if the camera was "pre-amped" for production models, right?

I'm really glad I'm not the only one seeing the obvious difference.
 
Kholi said:
And it would be faster if the camera was "pre-amped" for production models, right?
.
Click to expand...

Right. Adam Wilt told me that the pre-production HPX170 was as slow as an HVX200, but quieter. Obviously, the production HPX170's have been altered for more sensitivity at the expense of noise.

We were led to believe that the HPX170 would exhibit an image that is at least as good as an HVX200A. This is not the case. I want to reduce the noise in my 170.

My camera was at Panasonic in NJ this week and I wrote a note asking the techs if they could reduce the noise by bringing down preamp gain. I don't expect that to happen, but I thought I would try.
 
If the case was a "pre-amp" and it could be brought back to where the 200A is/was this camera would be a 9 on a scale of 10. And that's only cause nothing can be perfect.
 
Not to be a positive pansy or anything... but with the drastic improvements made in the HPX170 all thanks to Jan's involvement with the user community especially here on DVXuser, I have no doubt that she'll help us figure out what's up!
 
For some reason, this reminds me of the way the audio levels of the DVX100A were "boosted" over the DVX100 because people were requested a stronger playback monitor level. When I bought the 100A upon its initial release, the static noise through the headphones was so bad in the A model as compared with the 100, I was convinced the camera was defective and returned it. When I received the second one, I realized that is just the way it is... So, I don't really understand taking extremely delicate calibrated technology (generalizing obviously) and slapping a "boost" on it as if that is going to really do something positive.

I have really mixed feelings about these various early overviews of the 170 camera that were not based on the final shipping version of the camera. I'm not faulting the authors, but there have obviously been some significant issues to crop up over the last 4 weeks with the release models that really would have influenced the likelihood of me purchasing the 170 over the 200A or an EX1 for that matter.
 
CW-35 said:
So, I don't really understand taking extremely delicate calibrated technology (generalizing obviously) and slapping a "boost" on it as if that is going to really do something positive.
Click to expand...

In the case of the HPX170, it looks like Panasonic wanted to make sure that it is as sensitive as the 200A, despite the smaller diameter lens. It is now close to a stop faster than a 200A at a wide focal length, but at the expense of added noise.

I shoot with my HDX900 at -3db as much as possible because I find noise to be a bit too high at 0db. I would be happy to shoot with the 170 at -3db as well if it offered the setting. Obviously "0db" is an arbitrary number that manufacturers decide what the signal to noise ratio compromise will be.
 
We'll have to produce some comparison video to even be heard. And, then, I wonder if they'll even do anything about it...
 
Have faith young Kholi have faith! You may want to produce a video of doing the comparison video... just to confirm methodology and technique to these Panasonic guys though.
 
I don't expect Panasonic to acknowledge it regardless. You see how much it took to get the Lens Wobble issue out there. It was being called a "feature" as the excuse and it took a hellstorm to even get someone to listen.

And that was a manufacturing issue. Something they did on purpose? It's almost pointless to even bother, we're going to be forced to eat it regardless.
 
i will be visited by two panasonic people from osake and one from panasonic germany.
i would like to speak with them about this issue.

thomas
 
Does anyone know if the suspected "gain-boost" would apply to the HMC150? I'm quickly being turned off the HPX170 but don't like the size of the 200A.
 
My only experience is with the DVX and the HVX felt unwieldy next to that. I'm doing preliminary research... just wondering if anyone's seen the same noise problems with the 150. At the end of the day, at 720/24p the HPX170 and HMC150 are very similar beasts on a purely visual level from everything I've read.
 
You must log in or register to reply here.
Back
Top