Panasonic AG-HPG20: an AVC-Intra P2 Gear

Jan_Crittenden said:
We are going to be testing the codecs. I don't really care about whether Sony's camera works well or not. I do know that the 3000 puts out a picture that will show the differences between the two codecs and the noise, and imperfections of the camera's output won't get in the way of the codecs' performance.
Click to expand...

Jan,

I agree that the HPX3000 outputs a great picture, and would be a great subject for testing the codecs. But since the HPX3000 can already record to AVC-I, chances are those owners will not be buying a data recorder. I believe that both the HPG20 and Flash XDR are targeted to smaller cameras costing less than $10,000. So why not test one of these cameras, too?

Out of all the smaller cameras, I believe that Sony's EX1 and EX3 are the only ones that output a true 10-bit HD-SDI signal. Since you want to show that a 10-bit codec (AVC-I) is better than an 8-bit codec (Sony EX), these cameras are really the only logical choices.
 
Hi,

The point of the shoot out is to test the codecs. The only camera that I know of that is better than the AVC-Intra Codec is the 3000, as it also has a 4:4:4 output. So by using the 3000 we can see the true nature of the 10 bit codec vs an 8 bit codec and not have a camera to blame. The 3000 is quieter than either of those two cameras you mention, has a 10 bit output and is based on 2 million pixel CCDs not CMOS imagers. So from that you can analyze the codecs and deduce where the strenths are. I happen to know that the AVC-Intra Codec is great codec and 10 bits will get you farther than 8 bits ever will but we will do the test.

Once the AVC-Intra codec shows its true nature vs the competition, I am sure that there will be many dealers that will be happy to run the test with any of the less expensive cameras for you.

Best,

Jan
 
Hi Jan-
OK, a post NAB test sounds great! Our Flash XDR and nanoFlash should be more than up to the task.

In the meantime, please don't tell our customers who are currently using the Flash XDR in F16 fighter jets, helicopters, race cars and acrobatic planes that the CF cards are not a reliable choice! Or that the Flash XDR does not produce gorgeous video.

You might have a very difficult time convincing them.

Mike Schell
Convergent Design
 
Mike Schell said:
OK, a post NAB test sounds great! Our Flash XDR and nanoFlash should be more than up to the task.
In the meantime, please don't tell our customers who are currently using the Flash XDR in F16 fighter jets, helicopters, race cars and acrobatic planes that the CF cards are not a reliable choice! Or that the Flash XDR does not produce gorgeous video.
You might have a very difficult time convincing them.
Click to expand...

Mike,

I don't say that it is not up to your task, I have said it isn't up to the task of what is happening in P2. This I will continue to say. People think that I am saying what you think I said, and perhaps if you take some of my words out of context you might ifer that is what I mean. But just like I told a young man that came into our booth at GV Expo yesterday, I do not want to have Compact Flash in the cameras, I like what P2 does for the reliability of the recording. I don't get the same warm fuzzies on CF.

But this request you ask of me would be similar to my asking of you to please stop implying that your 100mbs 8 bit codec looks as good or is equal to AVC-Intra, it's not and I have a whole bunch of customers that would say its not. Sort of the same kind of thing Mike, you would have a pretty tough time convincing them, because there is no way that an 8 bit codec can come close to a 10 bit codec. Different league entirely.

See you after NAB.

All the best,

Jan
 
Thanks for the very informative exchange and I look forward to future testing results.
I met Jan yesterday at GV Expo and was very impressed by the new P2 Gear. Curious to see HDX900 & HPX170 hd-sdi out recorded to 10bit AVC Intra on the HPG20.
I wish Mike had a booth there - that would have made my day.
CF card performance/reliability is pretty well entrenched in the stills world, but there must be a reason RED makes their own proprietary CF cards and drives.

Jan, as for the "young man" reference - I only look young in my avatar.


chris
 
monkeyking said:
Jan, as for the "young man" reference - I only look young in my avatar.
Click to expand...

Who says it was you? Unless you identify which person you were, I cannot say it was you or someone else. I had more that one, and I referred both of them to the HMC150 camera for inexpensive recording media.

Best,

jan
 
Jan_Crittenden said:
Hi,

The point of the shoot out is to test the codecs. The only camera that I know of that is better than the AVC-Intra Codec is the 3000, as it also has a 4:4:4 output. So by using the 3000 we can see the true nature of the 10 bit codec vs an 8 bit codec and not have a camera to blame. The 3000 is quieter than either of those two cameras you mention, has a 10 bit output and is based on 2 million pixel CCDs not CMOS imagers. So from that you can analyze the codecs and deduce where the strenths are. I happen to know that the AVC-Intra Codec is great codec and 10 bits will get you farther than 8 bits ever will but we will do the test.

Once the AVC-Intra codec shows its true nature vs the competition, I am sure that there will be many dealers that will be happy to run the test with any of the less expensive cameras for you.

Best,

Jan
Click to expand...

Jan,
I, as well as many/most people in these types of forums, can't afford the 3000. But, we come to these forums to discover/find/research affordable ways of achieving 3000 type standards (respectfully keeping the Panny reference, although the Viper, Red, F900, etc. are all on the standards list).

So, a portable recording device that not only records the out spigot from an "affordable" camera via its HD-SDI, but also records it to a codec that trumps the in-camera codec, comes visually close to uncompressed, that is NLE friendly, that is media affordable, is what would appeal to most readers here - I am betting. So, please don't be so dismissive to Mike Schell's request to test the respective recorders with the EX1, or if you can't see past your own product line, the HPX170. We know the electronics on the front end of the 3000 are clean. But we want to see how these recorders improve the image quality from our less expensive camera's front ends via the HD-SDI.

And you mentioned "blame on the camera" - there would be no blame on the camera - you both would be shooting the same scene via the same camera from a split signal. And then we, the consumer, can analyze the footage/stills to see whether the HPG20 or XDR performed a better job at the recreation, and whether one is truly superior in only image quality.

It seems you are forgetting/ignoring your audience base on these forums. We are not broadcast giants with deep pockets. And if you're going to just pan these "affordable" tests off on other dealers and not take us readers seriously - why visit these forums.

Respectfully,
Lonnie
 
Hi,

The point of contention is whether the 8 bit MPEG2 codec is a better codec than the 10 bit AVC-Intra. If indeed the AVC-Intra prevails like I believe it will, then one can look at the output of any HD-SDI camera and know that what you see on the monitor will be very close to what you get. It does seem that if you can see the difference between the two recordings with the cleanest signal, then one less pristine will have the same footprint and you should be able to see it, notably in the chroma gradations and overall noise footprint. Keep in mind the purpose of the test. With a lesser camera you may not see these as easily as those cameras do cost less for a reason.

And I am not failing to understand the audience. I want the audience to know that with the highest quality camera out there how the codecs will stand up. I am sure that there will be many that with do this sort of testing with less expensive cameras as it gets released but the purpose of the Codec Battle is the Codec Battle. That is what I am planning on doing as soon as I have a serial numbered unit in my hands.

Best regards,

Jan
 
I agree with Jan on this one. It's about the codecs and the best way to test those is with the highest quality optics. An essential concept of science research... limit your variables! If we add in all sorts of other cameras, that complicates the test. So I think settling on a single camera like Jan and Mike agreed to do is the best way to do it.

Perhaps you won't be able to see a huge difference between the two with a camera like the 170 or the EX1. But in the end, people that will buy these... high end professionals and rental houses... want to get the best bang for their buck. That's why the codec test will be exciting and great to see.
 
Jan,
I understand the methodology - I guess my concern is this... These portable recorders will have a great market sector that will become customers because they can't afford a $40000 camera. This same sector will already own $7000 camera(s) and will be looking to purchase a $5000+ external recording unit that will make their $7000 camera compete with the image quality of a $40000 one.

Now we all know the front end of the HDX3000 with great glass can produce stunning images. And testing from the front end of the HDX3000 will only prove that both yours and Mike's recorders, with their respective great codecs and high Mbps will both be awesome when used with an expensive camera such as the HDX3000. We all know this in advance - nothing learned.

But for the market sector I'm referring to, that are looking to improve upon the limitations of our existing cameras by bypassing the in-camera codecs and recording to a superior codec - our questions will still be unanswered: How much can they improve our less expensive cameras' images? Which recorder does it better? Or are both remarkably clean compressions and now it's more of a price, frills, and branding that matters?

And... I realize I am now beating a dead horse and yes, I too will still want to see the results of this test as well. But my prediction - they both will be gorgeous, thus Mike's less expensive module with less expensive media will most likely win this round.

Regards,
Lonnie
 
Last edited:
Lonnie said:
Jan,
I understand the methodology - I guess my concern is this... These portable recorders will have a great market sector that will become customers because they can't afford a $40000 camera. This same sector will already own $7000 camera(s) and will be looking to purchase a $5000+ external recording unit that will make their $7000 camera compete with the image quality of a $40000 one.

Now we all know the front end of the HDX3000 with great glass can produce stunning images. And testing from the front end of the HDX3000 will only prove that both yours and Mike's recorders, with their respective great codecs and high Mbps will both be awesome when used with an expensive camera such as the HDX3000. We all know this in advance - nothing learned.

But for the market sector I'm referring to, that are looking to improve upon the limitations of our existing cameras by bypassing the in-camera codecs and recording to a superior codec - our questions will still be unanswered: How much can they improve our less expensive cameras' images? Which recorder does it better? Or are both remarkably clean compressions and now it's more of a price, frills, and branding that matters?

And... I realize I am now beating a dead horse and yes, I too will still want to see the results of this test as well. But my prediction - they both will be gorgeous, thus Mike's less expensive module with less expensive media will most likely win this round.

Regards,
Lonnie
Click to expand...

How many people with $7000 cameras will buy a $5000 device just to get higher quality images in the field? I see people renting these for their shoot if they need the extra quality for CG or projection or anything like that. I mean for $5000 you can buy a new camera. Or sell your $7000 camera and get a better $12,000 camera. SO... I really think the validity of this test should be aimed at the high end spectrum of cameras. If you want to see what these can do for your lower end camera... look on Convergent Design's website for some demo footage comparisons. I'm sure there's some stuff out there from the version 1 of the P2 Gear as well.
 
Lonnie said:
Jan,
I understand the methodology - I guess my concern is this... These portable recorders will have a great market sector that will become customers because they can't afford a $40000 camera. This same sector will already own $7000 camera(s) and will be looking to purchase a $5000+ external recording unit that will make their $7000 camera compete with the image quality of a $40000 one.
Click to expand...


And the point here is to show what the respective codecs can do. If you take a lesser camera, then it becomes more about the camera and not the codec. Trust me, someone will do a follow-up with the less expensive cameras.

Now we all know the front end of the HDX3000 with great glass can produce stunning images.
Click to expand...

Actually even with lousy glass it takes great pictures and BTW, that is an HPX3000.

And testing from the front end of the HDX3000 will only prove that both yours and Mike's recorders, with their respective great codecs and high Mbps will both be awesome when used with an expensive camera such as the HDX3000. We all know this in advance - nothing learned.
Click to expand...

I would not be so sure about this. I have lived long and hard with 8 bit codecs and nothing is going to make an 8 bit codec act like a 10 bit codec. Keep in mind that DVCPRO HD is an 8 Bit 100mbs codec. Can I make it fail, yeah, a little effort and the right image/colors. If the physics and math get in your way, just look at the difference, 256 vs 1024. Something should guide youto think of a different look and feel, noise, color gradations. There will be a difference.


But for the market sector I'm referring to, that are looking to improve upon the limitations of our existing cameras by bypassing the in-camera codecs and recording to a superior codec - our questions will still be unanswered: How much can they improve our less expensive cameras' images? Which recorder does it better? Or are both remarkably clean compressions and now it's more of a price, frills, and branding that matters?
Click to expand...


You cannot tell how remarkably clean they are unless you send it something much cleaner and a know quantity. Branding is not a factor here, performance is. So with the test that we have agreed to, we should be able to tell the difference between the codecs, and from there other tests will be made, I do know what I have seen with the HPX170, and it does look quieter, and a little more pristine, but that was merely a subjective while we were working some tests for frame rate extraction.

And... I realize I am now beating a dead horse and yes, I too will still want to see the results of this test as well. But my prediction - they both will be gorgeous, thus Mike's less expensive module with less expensive media will most likely win this round.
Click to expand...

The less expensive media will win this round if the less expensive media is your priority. If picture quality is your thing then I wouldn't be so sure.

Best,

Jan
 
Jan,
thanks for taking the time - I will wait for the test results. Anything else would just be speculation...

Joseph,
a purchase of a $5000 recorder of this kind - that is camera independent, and just needs HD-SDI to play well with others - this type of toy will be part of your arsenal for a long time. A kind of future proofing... And a $12000 camera doesn't move you into any in-camera recording at 100Mbps to Mpeg2 Long-GOP or AVCIntra that I'm aware of...

Thanks for your time all,
Lonnie
 
Ok well how about you do the test on a hpx500? The target audience for that cam would love to get the most out of that cam and would be more willing to pay 5k for a recorder. HPX3000 doesn't seem like the best test unit as the camera would be best tuned to record to avc100.
 
Hi,

For all of the reasons I have said before about why we should use the 3000 for this test, we will be using the 3000.

All of the other camera tests will happen but not on this occasion.

Please understand we are looking at 8 bit codecs vs 10 bit codecs in this test. I do not want to be distracted by too many agendas on the plate.

Thanks,

Jan
 
Hi Jan-
Let's just proceed with the tests when possible. I know some users who are already planning independent tests using the EX1/EX3 and HPX170 cameras, so I think we get some unbiased results in the coming months.

I studied the AVC-I footage at the InterBee show in Tokyo last November. It does look gorgeous. On the same hand, several of our users, through extensive tests, have concluded that our 100Mbps CODEC is somewhere between HDCAM and HDCAM SR quality. The footage even looks spectacular displayed on a 25' screen, via a high-end Christie projector.

Undoutedly, each of our recorders has strengths and weaknesses depending on the application. So let's do the image comparisons and then customers can evaluate other differences, such as size, weight, media costs, workflow, metadata support, etc.

We know your recorder will outperform the Flash XDR / nanoFlash in certain areas, but we are confident that user will (have) find Flash XDR and nanoFlash to be very compelling solutions.

Mike Schell
Convergent Design
 
Jan_Crittenden said:
Yes the little guys is pretty featgure packed and on top of the HD-SDI input and the AVC-Itra Recording, it also has the ability to plyback in conficence mode from the HDD.

As far as recording to CF card, no way. I wouldn't want to put careers at stake on a $50 card that has not parity checking, no error checking and frankly isn't up to even my demos. ;-) If you have seen the abuse I had out to P2 cards, then no comment is necessary. But I only do what sometimes accidently happens to the cards under normal accidental use. And please this is not the place to have a very worn out discussion about whter CF would handle what we do, because if CF could have, then that is what we would have done 6 years ago, but it doesn't. It certainly would have been easier.

Additionally the Convergent Design is using long GOP from an MPEG2 codec, this is a very old codec that doesn't come close to the quality possilbe in an I-Frame codec like AVC-Intra.

It is identical in size to the HPG10, and save for a couple of markings and a extra BNC it could be mistaken.

Barry, If you would like some pictures I can send them on for hosting if you wish. Just let me know.

All the best,

Jan
Click to expand...

Jan,
first, I am fairly new to P2, but have a HVX200 and a 171 and am very happy with both of them. I am planning to get a 500, too. But can I use the AG-HPG20 to convert media shot in DVCProHD on these cameras to directly store them as AVCIntra on a hard drive?
Sorry if this seems to be a stupid question, but I couldn't find any answer to this on the board...

Thanks,

Michael
 
Hey Jan

We already did these tests with a HPX3700 and Canon HJ11x4.7 Lens. We compared the uncompressed footage from the camera to the compressed AVC-Intra 100. We went one step further and recorded the uncompressed footage in various decks and recording devices (including the Flash XDR). Of course we didn't use the HPG20 to record but the AVC-Intra from the camera should be the same.

We then brought the footage into Shake and viewed the difference between the compressed and uncompressed footage. After enhancing the difference the results were pretty interesting. I'd be happy to share them with you or anyone that would like to come in and see them.

The HPG20 is going to be a great product to enhance many tape based ENG cameras - including the F900R.

Andy
 
mikeandcamera said:
But can I use the AG-HPG20 to convert media shot in DVCProHD on these cameras to directly store them as AVCIntra on a hard drive?
Click to expand...


Hi Mike,

You would need to play out from the HPX170 over HD-SDI and go into the HPG20 via HD-SDI and have it set up for AVC-Intra, in the same format, i.e. 720P or 1080i.

Hope that helps,

Jan
 
You must log in or register to reply here.
Back
Top