AVS › AVS Forum › Display Devices › Display Calibration › AVS HD 709 - Blu-ray & MP4 Calibration
New Posts  All Forums:Forum Nav:

AVS HD 709 - Blu-ray & MP4 Calibration - Page 4

post #91 of 3625
12/10/07 at 9:13am
Quote:
Originally Posted by jvincent View Post

If you ever add deinterlacing or audio/video sync patterns then keeping it at 29.97 would be better. If it's easy to change, changing it to 23.976 might be even better.

What is the advantage of 29.97 for deinterlacing/sync patterns? What is the advantage of 23.976 over 29.97 or 30?

30 is just such a nice, round number
Reply
Reply

AVS Top Picks

post #92 of 3625
12/10/07 at 9:22am
Quote:
Originally Posted by hwjohn View Post

What is the advantage of 29.97 for deinterlacing/sync patterns? What is the advantage of 23.976 over 29.97 or 30?

30 is just such a nice, round number

23.976 matches film-based telecine speeds so you can check if your VP correctly deinterlaces this material, but you'll also have to add cadence flags which might not be too fun. If you add these tests it might be better to find a pre-made clip somewhere.
Reply
Reply
post #93 of 3625
12/10/07 at 9:27am
Quote:
Originally Posted by zoyd View Post

23.976 matches film-based telecine speeds so you can check if your VP correctly deinterlaces this material, but you'll also have to add cadence flags which might not be too fun. If you add these tests it might be better to find a pre-made clip somewhere.


Adding the flags probably wouldn't be a big deal, but we are talking strictly film mode deinterlacing for that type of pattern. I can see where there might be an advantage for that one pattern (I'd have to think how such a pattern would be made), but I don't see where it really matters for anything else. Everything on the disc doesn't have to be the same frame rate, so if we wanted to include a special pattern like that then we could use a different frame rate for it than the rest of the disc.

Any other advantages of 29.97 or 23.976 over 30?
Reply
Reply
post #94 of 3625
12/10/07 at 9:31am
Since you can do individual patterns at different frame rates then it probably doesn't matter what rate you use for the static patterns.

The only real reason for using the film friendly rates is to make audio/video sync simpler.
Reply
Reply
post #95 of 3625
12/10/07 at 10:58am
Thread Starter 
Quote:
Originally Posted by jvincent View Post

If you ever add deinterlacing or audio/video sync patterns then keeping it at 29.97 would be better. If it's easy to change, changing it to 23.976 might be even better.

24p is out of the question if we have to transcode the AVCHD, because the Ulead software can only do 60i.

Ron posted a film judder mpeg2 in the w6rz thread that I think was supposed to be 24p. I put it on an HD DVD test disk and my XA2 player on my A3000 TV didn't seem to play it as smoothly as my computer on 24p with the A3000. Based on that one test, I'm not so sure MF6+ deals with 24p correctly.

I'm not doing the encodes, but I think that we should just stick to the tested HCenc clip. If the time misses because of the difference, then miss with a little longer than 2 minute time.
Reply
Reply
post #96 of 3625
12/10/07 at 11:02am
Quote:
Originally Posted by alluringreality View Post

24p is out of the question if we have to transcode the AVCHD, because the Ulead software can only do 60i.

That's ok, you want to simulate what you get on commerical DVD's anyway which is a 24p source embedded in a 60i signal. In other words you don't need 24p clips, you need to encode a 24p set of frames with the proper field cadence into a 60i clip.
Reply
Reply
post #97 of 3625
12/10/07 at 11:12am
Thread Starter 
I thought Blu-ray was 24p on disk.
Reply
Reply
post #98 of 3625
12/10/07 at 11:16am
Quote:
Originally Posted by alluringreality View Post

I thought Blu-ray was 24p on disk.

I think both Blu-ray and HD-DVD allow for 24p framerates as well as 50i/60i but 24p won't test your deinterlacer.
Reply
Reply
post #99 of 3625
12/10/07 at 11:27am
Thread Starter 
My experience is basically using Avia, DVE, Getgray, w6rz, and HCFR. I've looked at a number of the computer patterns also, but a lot of them don't apply to film sources. Out of all of that, I'm not familiar with testing the deinterlacer idea.

First of all, which deinterlacer would you be wanting to test - the player or the TV? (I'll guess TV)

Second, which patterns would be used for testing the deinterlacer anyway besides maybe the single pixel patterns?

What would the generator patterns for testing deinterlacing be like? For instance Cnet said the A3000 failed their deinterlacing test from a generator, but I think UMR said that it passed his test.

Maybe it's just because I use 24p, but I don't get how we would go about testing 1080i anything more than Ron's rotating bar pattern. If we were to use that, then it would probably have to be joined in with VideoReDo and the Misc Patterns would have to match that same encoding. Then we would have to test it again when it was transcoded from mpeg2 to H264.

Sure I get the idea about having an audio/video sync pattern, but I would have to call that beyond our current technical level. Based on what my XA2 was doing with the Getgray pattern and the judder test, and the issues we've encountered with video encoding, I see that as beyond the current realistic scope with our current software and knowledge.
Reply
Reply
post #100 of 3625
12/10/07 at 11:38am
Quote:
Originally Posted by alluringreality View Post

My experience is basically using Avia, DVE, Getgray, w6rz, and HCFR. I've looked at a number of the computer patterns also, but a lot of them don't apply to video. Out of all of that, I'm not familiar with testing the deinterlacer idea.

First of all, which deinterlacer would you be wanting to test - the player or the TV?

Second, which patterns would be used for testing the deinterlacer anyway besides maybe the single pixel patterns?

Sure I get the idea about having an audio/video sync pattern, but I would have to call that beyond our current technical level. Based on what my XA2 was doing with the Getgray pattern and the judder test, and the issues we've encountered with video encoding, I see that as beyond the current realistic scope with our current software and knowledge.

I think we could add some of those patterns in the future, but they will definetely take more time to create and verify. I'd say the audio/video sync pattern would be the most difficult, mainly because it isn't trivial to get the audio and video perfectly synchronized when you mux it.

I personally would like to have a pattern to check deinterlacing if we could come up with one that makes it easy to indentify deinterlacing problems. If anyone has an idea for a pattern I think it would be worthwhile to post it, even if it didn't make it to the disc for a while (or ever). I think either of these two patterns would be down the road a bit as they would most likely be significantly harder to create and we have bigger fish to fry at the moment.

As long as Ron is around to prod us in the right direction, I'd say that our current software will be more of an issue than knowledge. In the end it is alluringreality's decision as the disc was originally his idea, so he steers the ship.
Reply
Reply
post #101 of 3625
12/10/07 at 11:43am
It would test the player if it outputs a progressive signal or the display if the player just passes 1080i. I don't know what pattern is best to use for this test, probably just single spaced horizontal lines.

I'm not saying it should be included on your disk, I was just responding to what a 24p frame rate could be useful for. Here is an article that goes into a little more detail concerning HD resolution and interlace test.
Reply
Reply
post #102 of 3625
12/10/07 at 12:00pm
Thread Starter 
That link reads "The Onkyo DV-HD805 and Toshiba HD-XA2 HD DVD players will deinterlace 1080i and properly handle the 3:2 cadence if you set them for 1080/60p output because of their built-in Silicon Optix HQV processor."

When I tried the rotating bar pattern that I think is meant to show deinterlacing on the HD DVD that Ron hosts, the rotating bar seemed messed up at 1080p/60 from the XA2 as far as I could tell.

I'm willing to accept input on deinterlacing or audio sync, but they're currently outside our scope.
Reply
Reply
post #103 of 3625
12/10/07 at 12:33pm
Quote:
Originally Posted by alluringreality View Post


When I tried the rotating bar pattern that I think is meant to show deinterlacing on the HD DVD that Ron hosts, the rotating bar seemed messed up at 1080p/60 from the XA2 as far as I could tell.

Don't know why the XA2 fails. Note that Ron's site has tests for deinterlace artifacts both with and without 3:2 cadence flags.
Reply
Reply
post #104 of 3625
12/10/07 at 2:25pm
Quote:
Originally Posted by hwjohn View Post

Well, that explains it. I should have thought to calculate it myself.

I think I'm going to change the framerate for the second beta of the HD DVD and the first release of the AVCHD from 29.97 fps to 30 fps, just for simplicity in timing. There is no advantage to using either one is there?

Also, HD DVD and AVCHD is a world wide standard, correct? Would we need to change anything to allow playback in PAL countries?

30.0 fps isn't allowed on HD-DVD or Blu-ray. You must use 29.97 fps (or 59.94 fps for 720p).

For progressive film content, things get more complex. On Blu-ray, you can use either 23.976 or 24.0 fps (although almost all commercial Blu-ray discs are 23.976). On HD-DVD, you must use 29.97 fps with pulldown flags.

I don't believe any of the players support 25 fps at this time, although it is allowed in both HD-DVD and Blu-ray (along with 50 fps 720p).

Ron
Reply
Reply
post #105 of 3625
12/10/07 at 2:54pm
Quote:
Originally Posted by dr1394 View Post

30.0 fps isn't allowed on HD-DVD or Blu-ray. You must use 29.97 fps (or 59.94 fps for 720p).

For progressive film content, things get more complex. On Blu-ray, you can use either 23.976 or 24.0 fps (although almost all commercial Blu-ray discs are 23.976). On HD-DVD, you must use 29.97 fps with pulldown flags.

I don't believe any of the players support 25 fps at this time, although it is allowed in both HD-DVD and Blu-ray (along with 50 fps 720p).

Ron

Well, that settles it then. Figures they would be more restrictive than ATSC.

I thought that the pulldown flag would tell the player to repeat a field to go from 23.976 (as stored on disc) to 29.97. HD DVD stores the repeated field and the pulldown flag simply tells the player that the repeated field is present and to throw it away for 24p playback?
Reply
Reply
post #106 of 3625
12/10/07 at 3:17pm
Quote:
Originally Posted by hwjohn View Post

Well, that settles it then. Figures they would be more restrictive than ATSC.

I thought that the pulldown flag would tell the player to repeat a field to go from 23.976 (as stored on disc) to 29.97. HD DVD stores the repeated field and the pulldown flag simply tells the player that the repeated field is present and to throw it away for 24p playback?

The repeated field isn't stored. There are 23.976 coded frames per second, and the pulldown flags tell the decoder when to repeat a field from a coded frame. The final rate is 29.97 fps (and that's what is present in the bitstream syntax). For 24p output, the decoder just ignores the pulldown flags. Ignoring the flags is a little kludgey, and that's why 24p output on HD DVD took some time to arrive.

It's possible to convert HD DVD film content from 29.97 fps to 23.976 fps by removing the pulldown flags. I have a tool that I wrote for VC-1, but it doesn't do the complete job required for Blu-ray. It just removes the pulldown flags and changes the frame rate. It doesn't change the stream from interlaced (you can only have field repeats on an interlaced stream even though the actual frames are progressive) to progressive, which is required for Blu-ray.

http://www.w6rz.net/vc1conv.zip

For H.264, it's much more complex. We've been fooling around with it in another thread (with mixed results).

http://www.avsforum.com/avs-vb/showt...815296&page=27

Ron
Reply
Reply
post #107 of 3625
12/10/07 at 7:10pm
Quote:
Originally Posted by dr1394 View Post

The AVCHD format uses specific PID's. Here's the demux of the Ulead file.
Code:
C:\\xfer>xport -h Export20071209-006.mpg 1 1 1
xport Transport Stream Demuxer 1.00
program = 1, video channel = 1, audio channel = 1
Program Number = 0 (0x0000), Program Map PID = 31 (0x001f)
Program Number = 1 (0x0001), Program Map PID = 256 (0x0100)
program descriptor = 0x05, 0x04, 0x48, 0x44, 0x4d, 0x56
program descriptor = 0x88, 0x04, 0x0f, 0xff, 0xfc, 0xfc
Video PID = 4113 <0x1011>, type = 0x1b
ES descriptor for stream type 0x1b = 0x05, 0x08, 0x48, 0x44, 0x4d, 0x56, 0xff, 0x1b, 0x44, 0x3f
Audio PID = 4352 <0x1100>, type = 0x81
ES descriptor for stream type 0x81 = 0x05, 0x04, 0x41, 0x43, 0x2d, 0x33
ES descriptor for stream type 0x81 = 0x81, 0x06, 0x04, 0x20, 0x05, 0x00, 0xff, 0x01
0 frames before first I-frame
High Profile
Level = 4.0
First Video PTS = 0x00004b03
Audio Bitrate = 128000, Audio Sampling Rate = 48000
Audio Mode = 2/0, bsid = 4, bsmod = 0
First Audio PTS = 0x00004b03, 0
ts rate = unspecified
packets for pid    0 <0x0000> = 1366, first = 1, last = 36840
packets for pid   31 <0x001f> = 1366, first = 4, last = 36843
packets for pid  256 <0x0100> = 1366, first = 2, last = 36841
packets for pid 4097 <0x1001> = 1366, first = 3, last = 36842
packets for pid 4113 <0x1011> = 19868, first = 5, last = 36735
packets for pid 4352 <0x1100> = 11517, first = 131, last = 36849
packets for pid 8191 <0x1fff> = 15, first = 36850, last = 36864
coded pictures = 7364, video fields = 0

C:\\xfer>
The PMT is on PID 256, the PCR is on PID 4097, the video is on PID 4113 and the audio is on PID 4352. There's also a NIT on PID 31. If you look at a demux of a Blu-ray file, it's the same (except there are more audio PID's, sub-title PID's and the DTCP descriptor indicates copy never).
Code:
C:\\xfer>xport -h 00009.m2ts 1 1 1
xport Transport Stream Demuxer 1.00
program = 1, video channel = 1, audio channel = 1
Program Number = 0 (0x0000), Program Map PID = 31 (0x001f)
Program Number = 1 (0x0001), Program Map PID = 256 (0x0100)
program descriptor = 0x05, 0x04, 0x48, 0x44, 0x4d, 0x56
program descriptor = 0x88, 0x04, 0x0f, 0xff, 0xff, 0xfc
Video PID = 4113 <0x1011>, type = 0x1b
ES descriptor for stream type 0x1b = 0x05, 0x08, 0x48, 0x44, 0x4d, 0x56, 0xff, 0x1b, 0x61, 0x3f
Audio PID = 4352 <0x1100>, type = 0x80
ES descriptor for stream type 0x80 = 0x05, 0x08, 0x48, 0x44, 0x4d, 0x56, 0xff, 0x80, 0x61, 0xff
ES descriptor for stream type 0x81 = 0x05, 0x04, 0x41, 0x43, 0x2d, 0x33
ES descriptor for stream type 0x81 = 0x81, 0x04, 0x06, 0x48, 0x0e, 0x00
ES descriptor for stream type 0x81 = 0x05, 0x04, 0x41, 0x43, 0x2d, 0x33
ES descriptor for stream type 0x81 = 0x81, 0x04, 0x06, 0x48, 0x0e, 0x00
ES descriptor for stream type 0x81 = 0x05, 0x04, 0x41, 0x43, 0x2d, 0x33
ES descriptor for stream type 0x81 = 0x81, 0x04, 0x06, 0x48, 0x0e, 0x00
ES descriptor for stream type 0x81 = 0x05, 0x04, 0x41, 0x43, 0x2d, 0x33
ES descriptor for stream type 0x81 = 0x81, 0x04, 0x06, 0x29, 0x04, 0x00
0 frames before first I-frame
High Profile
Level = 4.1
LPCM Audio Mode = 3/2+lfe
LPCM Audio Bits/sample = 24
LPCM Audio Sample Rate = 48000
First Video PTS = 0x000ffff0
ts rate = unspecified
packets for pid    0 <0x0000> = 21, first = 1, last = 50437
packets for pid   31 <0x001f> = 3, first = 3, last = 48197
packets for pid  256 <0x0100> = 21, first = 2, last = 50438
packets for pid 4097 <0x1001> = 22, first = 4, last = 50757
packets for pid 4113 <0x1011> = 36755, first = 5, last = 52083
packets for pid 4352 <0x1100> = 11720, first = 1542, last = 52084
packets for pid 4353 <0x1101> = 963, first = 1704, last = 51719
packets for pid 4354 <0x1102> = 963, first = 1724, last = 51721
packets for pid 4355 <0x1103> = 963, first = 1751, last = 51723
packets for pid 4356 <0x1104> = 403, first = 1825, last = 51649
packets for pid 4608 <0x1200> = 243, first = 6, last = 36726
packets for pid 4611 <0x1203> = 7, first = 80, last = 3124
coded pictures = 67, video fields = 0

C:\\xfer>
So it's very likely that Ulead is looking for this structure. It may also be looking for the descriptors (the string 0x48, 0x44, 0x4d, 0x56 is "HDMV" in ascii).

However, the re-encode looks okay to me. You can decide for yourself with the JM reference decoder. It's located at http://iphome.hhi.de/suehring/tml/ and I have a compiled version of 13.0 here:

http://www.w6rz.net/ldecod_patch.zip

Since ldecod outputs to a single 4:2:0 file, I have a couple of tools to make things easier to work with.

http://www.w6rz.net/clipyuvhd.zip

http://www.w6rz.net/yuvtorgb.zip

clipyuvhd converts the single 4:2:0 file to a series of 4:2:2 files in UYVY format. yuvtorgb then converts those UYVY files to .tga format. The command lines are:

ldecod -i bits0001.mpv

clipyuvhd 1920 1080 test_dec.yuv

yuvtorgb 1920 1080 test0000.yuv test.tga

For video RGB levels:

yuvtorgb -v 1920 1080 test0000.yuv test.tga

Ron

EDIT: Use yuvtorgb utility instead of yuvtobmp.


Ron,
I think that the test clips have the right levels, although Photoshop keeps reporting 235,17,16 for 100% red primary. I have been looking at the math in the yuvtorgb source so that I can check it in UYVY space, but its getting late here any I'm going to have to head off to bed. Here is the UYVY for the 100% red primary:

F0 3F 66 3F

I think that is in the right order. This is the first time I have tried to look at a UYVY file directly. I can upload the whole thing if you would like. Can you tell us if that us 235,16,16 in RGB, or is it 235,17,16 like Photoshop is reporting?
Reply
Reply
post #108 of 3625
12/11/07 at 3:43am
Thanks for a great resource, guys !

hwjohn, I haven't used Ulead and I'm curious how you added the pop-up menus to this, because as I understood Ulead only creates Standard Content material, which isn't suitable for use with HDi code. Is that wrong, or did you generate the TMAP files needed yourself ? Or...?

Thanks !
Reply
Reply
post #109 of 3625
12/11/07 at 5:17am
Quote:
Originally Posted by mistertones View Post

Thanks for a great resource, guys !

hwjohn, I haven't used Ulead and I'm curious how you added the pop-up menus to this, because as I understood Ulead only creates Standard Content material, which isn't suitable for use with HDi code. Is that wrong, or did you generate the TMAP files needed yourself ? Or...?

Thanks !

alluringreality will have to answer that one. I do most of the Avisynth/encoding work, and he does the authoring. I do know that it is possible within the Ulead software.
Reply
Reply
post #110 of 3625
12/11/07 at 6:41am
Thread Starter 
Quote:
Originally Posted by mistertones View Post

I haven't used Ulead and I'm curious how you added the pop-up menus to this

There are a couple HDi menu programs I've seen listed on doom9, but the menus were created with http://www.ulead.com/dmf/plugin.htm There's only three other options for popup menus with the plug-in, but it's also what allows us to go to BDMV and AVCHD for Blu-ray.
Reply
Reply
post #111 of 3625
12/11/07 at 8:52am
OK.. Please Help: I'm in a technical field and have lots of computer experience but very limited in the A/V realm.

I really wanted to give the HCFR software a shot. I already own the Monaco Optix XR, and updated the dlls for HCFR BUT I don't have any targets to use with the HCFR software. I thought this was gonna do it for me but I think I might be out of luck since I have a PS3.

I am unclear by the language in the original post but it seemed that while these targets had a problem with PS3, it wasn't unsumountable. I thought I might have been able to read the target files individually instead of from the menu. ( If this is not true please say so). I burned these targets to Disc and PS3 could read the folders but did not see the files.

The HCFR provided targets didn't work for me because they're PAL. With Nero8, I tried the 'decode' business with the HCFR targets and while it seemed to recognize the screens (and I could see them) when the burn process was over PS3 still rejected it as PAL and even Windows thought the Disc was empty.

Honestly the reason I'm doing all this is because I already owned the Optix XR and thought this would be an inexpensive way to accurately calibrate the new TV. My picture looks pretty decent but I do wonder about the picture and brightness settings. Also being a serious amature photographer I know what it is to have accurate color - and want it on the new TV ( 40xbr4)

Now with the fact that I'm an A/V noob should I just bale out on this HCFR software for now and wait for HCFR 2.0 ? Is HFCR going to be over my head anyways having never used the AVIA on DVE products? I've already dropped close to 3k between 40XBR4, a new amplifier STR-910DG and the PS3. Should I just bite the bullet and get the AVIA or DVE Disc?

I've really been trying to help myself but I could use some direction here please. What I would really like are some targets to use with HCFR. When I set out doen this road 4-5 days ago I thought that would be the easy part. I guess that alone shows my noobness.
Reply
Reply
post #112 of 3625
12/11/07 at 8:59am
Slightly off-topic.

Reefbone, you can also use the GetGray disc to start with your PS3.

I'd suggest reading the calibration stickies at the top of this forum and then the GetGray thread.
Reply
Reply
post #113 of 3625
12/11/07 at 9:00am
Quote:
Originally Posted by reefbone View Post

OK.. Please Help: I'm in a technical field and have lots of computer experience but very limited in the A/V realm.

I really wanted to give the HCFR software a shot. I already own the Monaco Optix XR, and updated the dlls for HCFR BUT I don't have any targets to use with the HCFR software. I thought this was gonna do it for me but I think I might be out of luck since I have a PS3.

I am unclear by the language in the original post but it seemed that while these targets had a problem with PS3, it wasn't unsumountable. I thought I might have been able to read the target files individually instead of from the menu. ( If this is not true please say so). I burned these targets to Disc and PS3 could read the folders but did not see the files.

The HCFR provided targets didn't work for me because they're PAL. With Nero8, I tried the 'decode' business with the HCFR targets and while it seemed to recognize the screens (and I could see them) when the burn process was over PS3 still rejected it as PAL and even Windows thought the Disc was empty.

Honestly the reason I'm doing all this is because I already owned the Optix XR and thought this would be an inexpensive way to accurately calibrate the new TV. My picture looks pretty decent but I do wonder about the picture and brightness settings. Also being a serious amature photographer I know what it is to have accurate color - and want it on the new TV ( 40xbr4)

Now with the fact that I'm an A/V noob should I just bale out on this HCFR software for now and wait for HCFR 2.0 ? Is HFCR going to be over my head anyways having never used the AVIA on DVE products? I've already dropped close to 3k between 40XBR4, a new amplifier STR-910DG and the PS3. Should I just bite the bullet and get the AVIA or DVE Disc?

I've really been trying to help myself but I could use some direction here please. What I would really like are some targets to use with HCFR. When I set out doen this road 4-5 days ago I thought that would be the easy part. I guess that alone shows my noobness.


You would probably have better success posting this in your own, separate thread, but I'll give a short answer.

As for AVSHD, the HD DVD version probably can't be read at all by the PS3. The version we were referring to that could be read in file format was a prototype version that alluringreality and myself created for testing. We are currently working on an AVCHD version that would work on the PS3 as our top priority.

Outside of our disk, there are other options that may serve you better. Avia and DVE both have good explanations, where our disk only contains patterns and is geared toward users that already know what to do. I would suggest that you read the stickies in this forum to learn a little more about the process, and then decide which disk you need. By that time we may have an AVCHD version available.

EDIT: I would avoid trying to convert the PAL HCFR disk, or any other disk for that matter. It isn't trivial to ensure that the patterns are correct. That is what makes a good calibration disk a challenge to make.
Reply
Reply
post #114 of 3625
12/11/07 at 9:45am
Thread Starter 
Quote:
Originally Posted by reefbone View Post

I really wanted to give the HCFR software a shot. I already own the Monaco Optix XR, and updated the dlls for HCFR BUT I don't have any targets to use with the HCFR software. I thought this was gonna do it for me but I think I might be out of luck since I have a PS3.

As hwjohn stated, this is an HD DVD disk at this time. We are working on an AVCHD version that should work on the PS3. If you just want to do grayscale then GetGray will work. GetGray will have 601 color rather than 709. You can get GetGray in about a day and we certainly cannot match that.

I would prefer to release the next disk(s) to have some way to set sharpness, look at overscan, and adjust centering. That issue along with encoding will probably be what determines the time for the next release. My own tenative timeline might be around two weeks, but that ends up in the middle of the holidays. Realistically I'm not going to say another release will come this year. The only chance of another disk in 2007 might be some sort of a beta test disk.

If anyone is familiar with open-source patterns for setting sharpness, like ISO 12233, IEEE, or EIA, I would be interested to hear about it. I have seen http://www.graphics.cornell.edu/~wes...res-chart.html but I'm not impressed by the PDF at 100%. I also questioned HCFR about their EIA pattern but I have not received a response. At this time we have a tenative agreement to be able to use http://www.imatest.com/docs/testcharts.html but their timeline is more in line with a release in 2008.
Reply
Reply
post #115 of 3625
12/11/07 at 10:27am
Quote:
Originally Posted by jvincent View Post

Reefbone, you can also use the GetGray disc to start with your PS3.

I'd suggest reading the calibration stickies at the top of this forum and then the GetGray thread.

Quote:
Originally Posted by hwjohn View Post


As for AVSHD, the HD DVD version probably can't be read at all by the PS3.

other options that may serve you better. Avia and DVE both have good explanations, where our disk only contains patterns and is geared toward users that already know what to do. I would avoid trying to convert the PAL HCFR disk, or any other disk for that matter.

Quote:
Originally Posted by alluringreality View Post

. If you just want to do grayscale then GetGray will work. GetGray will have 601 color rather than 709. You can get GetGray in about a day and we certainly cannot match that.

Thanks for the replies all. Sounds like your all suggesting I bale on HCFR for now. I'll get on the stickies but would anyone care to offer a quick answer on "GetGray will have 601 color rather than 709" ? I guess this means I want 709 color huh? : /
Reply
Reply
post #116 of 3625
12/11/07 at 10:41am
Thread Starter 
I really didn't want to touch that question because I can't seem to get a clear answer myself in reading these forums. This disk basically uses 709 because DVE HD uses 709 color. On second thought, I would recommend the free disk at http://www.avsforum.com/avs-vb/showthread.php?t=852536 In my opinion the only thing it's really missing is a pattern for black-level. It will also be 601, but it won't cost anything. If you live in a big enough city the library might have Avia or Digital video essentials for a black-level pattern, and if not then netflix has them. w6rz.net is another option, but I don't know how the PS3 deals with levels. At this time there are really not many options for PS3 calibration with 709 color.
Reply
Reply
post #117 of 3625
12/11/07 at 10:50am
Quote:
Originally Posted by reefbone View Post

Thanks for the replies all. Sounds like your all suggesting I bale on HCFR for now. I'll get on the stickies but would anyone care to offer a quick answer on "GetGray will have 601 color rather than 709" ? I guess this means I want 709 color huh? : /

Typically, "601" color is used for SD content while 709 is used for HD. Its a bit more complicated than that, but that is what you basically need to know.

HDTVs typically use 709 primaries and can use either a 601 or 709 color decoding matrix depending on the input.

It is possible for playback devices to take material that originated from 601 primaries and was encoded with the 601 matrix and convert that material to 709 (provided the target display has 709 primaries). These are two distinct and different steps. This is the dilemma. If we knew all playback devices followed the correct rules and did the correct conversions, then the difference between 601 and 709 would be moot. The problem is that we don't know what each playback device does, and if it does it correctly.

AVSHD just bypasses all these problems by using an HD resolution with 709 encoding. All HD DVD and Blu-ray players will (should) use 709 to decode the HD material, and you will get the correct colors for calibration.

None of this is to say that AVSHD is better than any other disk. Each disk is very useful in the correct context.
Reply
Reply
post #118 of 3625
12/11/07 at 11:50am
Quote:
Originally Posted by hwjohn View Post


It is possible for playback devices to take material that originated from 601 primaries and was encoded with the 601 matrix and convert that material to 709 (provided the target display has 709 primaries).

Just a small clarification, there are no sources using "601 primaries". The rec601 encoding matrix was formulated for NTSC primaries which are no longer used, instead SMPTE-C primaries are most commonly (if not exclusively) used to master film and broadcast material regardless of the encoding matrix. This may be true of most HD content as well as I haven't seen anyone that can definitively point to any source material mastered with rec709 primaries, maybe Ron can chime in on that. If that's the case then even if you have rec709 encoded material such as your disk, you should calibrate your display to SMPTE-C primaries.
Reply
Reply
post #119 of 3625
12/11/07 at 1:27pm
Quote:
Originally Posted by zoyd View Post

Just a small clarification, there are no sources using "601 primaries". The rec601 encoding matrix was formulated for NTSC primaries which are no longer used, instead SMPTE-C primaries are most commonly (if not exclusively) used to master film and broadcast material regardless of the encoding matrix. This may be true of most HD content as well as I haven't seen anyone that can definitively point to any source material mastered with rec709 primaries, maybe Ron can chime in on that. If that's the case then even if you have rec709 encoded material such as your disk, you should calibrate your display to SMPTE-C primaries.

Thanks for the clarification Zoyd.

What I was trying to get at is that there is a set of primaries associated with each Rec. If the guys mastering the material want to mix and match, then I guess that is their choice, but it makes a bad situation for the consumer in that we are left to guess what the film was mastered to, which leads to all this confusion.

Our disk was made to be consistent with Rec. 709 decoding matrix. Most people don't have a choice in changing their primaries, but for those who do, I guess it is a crapshoot as to what you want to calibrate to.
Reply
Reply
post #120 of 3625
12/11/07 at 1:50pm
Quote:
Originally Posted by hwjohn View Post

Ron,
I think that the test clips have the right levels, although Photoshop keeps reporting 235,17,16 for 100% red primary. I have been looking at the math in the yuvtorgb source so that I can check it in UYVY space, but its getting late here any I'm going to have to head off to bed. Here is the UYVY for the 100% red primary:

F0 3F 66 3F

I think that is in the right order. This is the first time I have tried to look at a UYVY file directly. I can upload the whole thing if you would like. Can you tell us if that us 235,16,16 in RGB, or is it 235,17,16 like Photoshop is reporting?

It's just a rounding issue. The 601 and 709 specifications want you to round up for integers. If you do the floating point math, you'll see that Y = 63 (0x3f), Cb = 102 (0x66) and Cr = 240 (0xf0) converts to R = 235.4406, G = 16.502108 and B = 15.831311. The 16.502108 gets rounded up to 17. Here's the floating point code (for video RGB levels):

Code:
/*
Converts YUV to RGB floating point
*/

#include 
#include 

#define TRUE            1
#define FALSE           0

/* color space conversion coefficients
 * for YCbCr -> RGB mapping
 *
 * entries are {crv,cbu,cgu,cgv}
 *
 * crv=(219/224)*(1-cr)/0.5
 * cbu=(219/224)*(1-cb)/0.5
 * cgu=(219/224)*(cb/cg)*(1-cb)/0.5
 * cgv=(219/224)*(cr/cg)*(1-cr)/0.5
 *
 * where Y=cr*R+cg*G+cb*B (cr+cg+cb=1)
 */

/* ITU-R BT709-5 (2002) coefficients */
/* cg = 0.7152 */
/* cb = 0.0722 */
/* cr = 0.2126 */

int main(int argc, char **argv)
{
        int             y, cr, cb;
        double  r, g, b;

        if (argc != 4) {
                fprintf(stderr, "usage: yuvtorgbcalc   \
");
                exit(-1);
        }

        y = atoi(argv[1]);
        cb = atoi(argv[2]);
        cr = atoi(argv[3]);

        r = (double)y + 1.5396482142857 * (double)(cr - 128);
        g = (double)y - 0.4576750704098 * (double)(cr - 128) - 0.18314292755273 * (double)(cb - 128);
        b = (double)y + 1.814180357142857 * (double)(cb - 128);

        printf("R = %f, G = %f, B = %f\
", r, g, b);

        return 0;
}
http://www.w6rz.net/yuvtorgbcalc.zip

Ron
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration

AVS Top Picks

AVS › AVS Forum › Display Devices › Display Calibration › AVS HD 709 - Blu-ray & MP4 Calibration