Log in

View Full Version : madVR - high quality video renderer (GPU assisted)


Pages : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 [766] 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329

kasper93
7th June 2016, 19:04
* added workaround for weird crash with latest AMD beta drivers

They released new drivers with fixed DXVA processing (which was acting strange with 16.5.2 and 16.5.3 drivers). Not sure if this was the crash, but if the workaround is somehow hacky you might recheck if it's needed.

huhn
7th June 2016, 19:30
someone should first check if this issue is really fixed by using madVR 90.18 with amd crimson 16.6.1.

http://bugs.madshi.net/view.php?id=402

TobiMan
7th June 2016, 19:49
someone should first check if this issue is really fixed by using madVR 90.18 with amd crimson 16.6.1.

http://bugs.madshi.net/view.php?id=402

Good point... still crashes on my machine, so maybe we should keep the hack a little longer for now...

Sunspark
7th June 2016, 21:04
Here are the 3 GPU-Z screenshots.

Image 1: ReconSoft w/ Jinc AR

https://dl.dropboxusercontent.com/u/50794070/Screenshots/glitching.gif

Image 2: ReconSoft w/ Jinc (No AR)

https://dl.dropboxusercontent.com/u/50794070/Screenshots/not%20glitching.gif

Image 3: Left side is Jinc AR/Jinc AR and right side is DXVA2

https://dl.dropboxusercontent.com/u/50794070/Screenshots/jinc%20and%20dxva2.gif

Edit:

Added 1 more image, this one is VLC. Granted it doesn't have the stuff MadVR does, but the GPU load is even lower than that of DXVA scaling, which may or may not mean something.

https://dl.dropboxusercontent.com/u/50794070/Screenshots/vlc.gif

huhn
7th June 2016, 21:19
your GPU can't handle the load and the clock is very unstable.

jinc AR is already pushing it to it's limits.

Sunspark
7th June 2016, 21:49
your GPU can't handle the load and the clock is very unstable.

jinc AR is already pushing it to it's limits.

Thanks Huhn. So I am learning something here.

Is there a way I can stabilize the core clock? I see now that as gpu load goes up core clock seems to drop. I would have thought that it would have been the other way around, that as core clock goes up as load goes up.

It is too bad that this is not a bug after all.

huhn
7th June 2016, 21:58
better cooling should help. but i guess this is not a normal PC.
it looks like it is thermal throttling at ~72 °C.

using settings that are not going over 60% GPU usages should do the trick.

the shown 5 ms could be a bug. but even if it is a bug it doesn't mean it is a madVR bug.

Sunspark
7th June 2016, 22:15
better cooling should help. but i guess this is not a normal PC.
it looks like it is thermal throttling at ~72 °C.
using settings that are not going over 60% GPU usages should do the trick.
the shown 5 ms could be a bug. but even if it is a bug it doesn't mean it is a madVR bug.

It's a tiny little NUC5i5RYK. Basically a laptop computer on a desk. It's actually really neat, even comes with a plate to mount it to the back of a monitor's VESA mount compatible position. Of course that is usually occupied by the stand of the monitor itself. Due to being an ultra-small form factor PC, there is no way at all that I can add coolers to it. There is no room.

Thanks again.

chros
9th June 2016, 10:45
you should ignore HDR files that are just broken with incorrect infos. correct meta data is key to display them correctly!
That's correct.

it doesn't matter to what your display is calibrated as long as madVR knows that.
I think it's not true, it depends on what you mean. You can apply different primaries and matrixes to the same source and the end result will be different.

so the primaries info should be the primaries of the source no matter the calibration target.
I got this. But what if it's missing? e.g. this one from the demos: LG_2_DEMO_4K_L_N_08_Hoverboard 2160p_29fps_hevc.mkv . It's not HDR and still doesn't include any colorspace info.

huhn
9th June 2016, 11:13
I think it's not true, it depends on what you mean. You can apply different primaries and matrixes to the same source and the end result will be different.
a source has one matrix and one primaries and the creator should always add them.

so yes you can apply any conversation imaginable. but what is the point of this. there is a reason for these flags in files

and it doesn't matter to what your device is calibrated as long as madVR knows and supports that type of calibration.

if it is bt 709, bt 601, bt 2020 or stuff like YCoCg as long as the source is correctly flagged and supported by madVR (and madVR supports a lot more...) than madVR will handle the rest.

that's the whole point of flagging files to get correct colors.
I got this. But what if it's missing? e.g. this one from the demos: LG_2_DEMO_4K_L_N_08_Hoverboard 2160p_29fps_hevc.mkv . It's not HDR and still doesn't include any colorspace info.

sorry but i'm not going to download renamed MKV file.

if the encoder doesn't add them it is just guessing work.
try to use tags with bt 2020 this may works or may not work at all.

it is technically a bad file. only the creator can tell what the colorspace should be for sure. and to me it looks like he doesn't really care.

ajp_anton
9th June 2016, 12:39
It's a tiny little NUC5i5RYK. Basically a laptop computer on a desk. It's actually really neat, even comes with a plate to mount it to the back of a monitor's VESA mount compatible position. Of course that is usually occupied by the stand of the monitor itself. Due to being an ultra-small form factor PC, there is no way at all that I can add coolers to it. There is no room.

Thanks again.

The CPU+GPU combo is limited at 15W of power. When the GPU load is relatively light, it can keep its clocks up, but when the load gets too heavy it has to clock down to keep the power within 15W.
That's also why is fluctuates very much, and it's completely normal. It's trying its best to find the highest possible clock at every moment that doesn't use too much power.

strumf666
9th June 2016, 12:59
Maybe when combined with Kodi DSPlayer. Netflix addons exist. I think they have limitations when it comes the resolution and supported audio formats, but these addons will allow you to play Netflix content through Kodi.

DSPlayer is not available for Kodi v17. However, MPC-HC is and it can be used with the new interface, InputStream, which allows the playback of DRM-protected content. There is already an Amazon Instant Video addon that offers seamless playback similar to the Windows native app. Netflix will likely follow.

Kodi v16 allows playback of Netflix content -- but with restrictions, I'm sure. Kodi 17 is the way forward for streaming apps and Kodi. But no native madVR like DSPlayer.

MPC-HC + Kodi. See here for instructions: http://forum.kodi.tv/showthread.php?tid=209596.
:thanks: will take a look

andyvt
9th June 2016, 13:17
It's a tiny little NUC5i5RYK. Basically a laptop computer on a desk. It's actually really neat, even comes with a plate to mount it to the back of a monitor's VESA mount compatible position. Of course that is usually occupied by the stand of the monitor itself. Due to being an ultra-small form factor PC, there is no way at all that I can add coolers to it. There is no room.

Thanks again.

In my testing Lanczos 3 tap + AR is the max reliable scaling algorithm that this box can handle. I did not test the exact scenario you are trying to enable (scaling 1080p ->), used 720p60 -> 1080p60 as the "worst case" so if that doesn't work you will need to dial it back even more.

castaway
9th June 2016, 20:27
Hi everyone,

Is there a way to list madVR instances already running (ex: MPC HC playing a video) and hook to them (COM interfaces) ?

Thanks!

kanade
10th June 2016, 09:19
Hello madshi, despite using the latest madVR version (0.90.20) I'm still experiencing driver/madVR crash or even BSoD while using the latest AMD beta drivers (16.X).

Is there any fix for this? I've also tried using "madVR9019amdFix" before but I still experienced crashes and BSoD while using the latest beta drivers.

Downgrading the AMD driver to 15.12 will fix the problem though in return, I'm not able to play any newly released games with that older driver.

FYI, I'm currently using:
- R9 390
- Windows 7 64bit
- LAVfilter megamix: Potplayer slimmed down 64bit + LAVfilters + madVR

Thanks and I appreciate your help.

ryrynz
10th June 2016, 12:24
Hello madshi, despite using the latest madVR version (0.90.20) I'm still experiencing driver/madVR crash or even BSoD while using the latest AMD beta drivers (16.X).


You should pester AMD about it. It's not generally good practice to make workarounds for wonky drivers. I do hope they release a half decent set for the 470 & 480 (Polaris) wouldn't mind updating my 960.

burfadel
10th June 2016, 13:23
Hello madshi, despite using the latest madVR version (0.90.20) I'm still experiencing driver/madVR crash or even BSoD while using the latest AMD beta drivers (16.X).

Is there any fix for this? I've also tried using "madVR9019amdFix" before but I still experienced crashes and BSoD while using the latest beta drivers.

Downgrading the AMD driver to 15.12 will fix the problem though in return, I'm not able to play any newly released games with that older driver.

FYI, I'm currently using:
- R9 390
- Windows 7 64bit
- LAVfilter megamix: Potplayer slimmed down 64bit + LAVfilters + madVR

Thanks and I appreciate your help.

Have you overclocked your video card or computer at all? Newer drivers can trigger crashes even if older versions worked fine and the computer otherwise 'appears' stable, even in stress tests. If you have overclocked the card slightly try default settings. If you have overclocked your computer, take say, 0.2 GHz off it and see if that helps. If it does, it is indicative that your overclock isn't actually as stable as it appears. The new driver could be triggering the crash simply because it is now making use, or the right sequence of code relating to the instability.

Hope that makes sense :). Could also be just a Windows 7 issue, Windows 7 and 8.1 support is really only a passing thought for both Nvidia and AMD these days, almost all their focus is on Windows 10.

kanade
10th June 2016, 15:15
Have you overclocked your video card or computer at all? Newer drivers can trigger crashes even if older versions worked fine and the computer otherwise 'appears' stable, even in stress tests. If you have overclocked the card slightly try default settings. If you have overclocked your computer, take say, 0.2 GHz off it and see if that helps. If it does, it is indicative that your overclock isn't actually as stable as it appears. The new driver could be triggering the crash simply because it is now making use, or the right sequence of code relating to the instability.

Hope that makes sense :). Could also be just a Windows 7 issue, Windows 7 and 8.1 support is really only a passing thought for both Nvidia and AMD these days, almost all their focus is on Windows 10.

Thanks for the reply.

As of right now, I haven't overclocked anything as everything is running at default speed (auto). I'm currently using i7 4770k and although I overclocked it before, I revert it back to stock speed. Same with GPU and memory as well.

This is troubling.

LePousson
10th June 2016, 15:36
Hi ..

I just got my new laptop yesterday and I am getting problems with madVR versions from 0.89.7 until 0.90.20 ...

I am getting a lot of glitches when playing 1080p/24 files ...

Even if I know that my screen is 60Hz only, like nearly all the laptops, I can't understand why I am getting hundreds of glitches after a few seconds ...

The laptop config is :

Core i7-6700HQ / 8GB DDR4 / Hybrid Intel HD530 & GTX950M (2GB for the NVIDIA)

Thanks in advance for your help and your replies ...

huhn
10th June 2016, 16:09
most likely an optimus problem.

there is a thread about this on doom9.

you could post on OSD screenshoot so we have more information to work with.

there could other reasons for glitches like windows 7...

GREG1292
10th June 2016, 16:36
What's the best old version of madvr with basic performance and image with least amount of processing?

LePousson
10th June 2016, 16:40
most likely an optimus problem.



there is a thread about this on doom9.



you could post on OSD screenshoot so we have more information to work with.



there could other reasons for glitches like windows 7...

I doubt that Windows 7 is guilty as if I use the EVR or VMR rendering engine, it is working perfectly ...

Do you have the link to the thread for the Optimus problem ??

Thanks in advance ...

nevcairiel
10th June 2016, 16:48
What's the best old version of madvr with basic performance and image with least amount of processing?

All processing is optional, you can turn it all off and it would run on any kind of hardware. You don't need to find an old version for that.

huhn
10th June 2016, 17:12
I doubt that Windows 7 is guilty as if I use the EVR or VMR rendering engine, it is working perfectly ...

Do you have the link to the thread for the Optimus problem ??

Thanks in advance ...

http://forum.doom9.org/showthread.php?t=170937

windows 7 without aero is know for presentation glitches and the desktop composition (aero) is kind of buggy too. i could say more with a OSD screen.

VMR and EVR doesn't tell you if they got a presentation glitch they usually just have tearing without aero.

LePousson
10th June 2016, 18:05
http://forum.doom9.org/showthread.php?t=170937



windows 7 without aero is know for presentation glitches and the desktop composition (aero) is kind of buggy too. i could say more with a OSD screen.



VMR and EVR doesn't tell you if they got a presentation glitch they usually just have tearing without aero.

Thanks for the link ...

Here is a link to the YouTube video I uploaded to show you more precisely what problems I have : http://youtu.be/PAeovGIK1rE

If you can help me a bit, feel free to tell me ...

huhn
10th June 2016, 18:20
it's a private video so i can't watch it.
and i'm not an optimus user but the people in that thread are optimus user and if i remember right they figured something out about this.

LePousson
10th June 2016, 18:37
Yesssssssssssss.....

The fix indicated by Ceremony in the thread of which you gave me the link, is working perfectly ...

I just got the 2 or 3 dropped at the beginning of the video, quite logical under Kodi ...

madVR now plans a frame repeat every 3.5/4.0 hours, and still growing ...

Thanks a lot for your help ...

PS : Should be ok to play my video ...

tahaa7
10th June 2016, 23:57
Question regarding (de-)interlacing. I am using madVR in combination with MPC-HC (and built-in LAV filters). How do I set up madVR so that it outputs video to my HDTV in its original interlaced form, leaving the TV to deinterlace? I've played with the settings (display modes etc.) for hours, but I cannot get the TV to recognize that the signal it's getting is interlaced, and not progressive.

huhn
11th June 2016, 01:16
madVR is not able to leave an interlanced signal untouched.
i'm not sure if this is even possible with a PC.

just chroma upscaling to 4:4:4 will ruin an interlanced image if it is done before deinterlancing.

CarlosCaco
11th June 2016, 02:22
Thanks for the link ...

Here is a link to the YouTube video I uploaded to show you more precisely what problems I have : http://youtu.be/PAeovGIK1rE

If you can help me a bit, feel free to tell me ...

on my optimus system hd5500 / 930m i had this exact behave showed in the video only if set mpc-hc to run with intel gpu, to me this problem solves setting nvidia as preferred gpu on nvidia cp... and the if set the ceremony fix on nvidia inspector i don´t see any diference on perfomance...

Jtacdf
11th June 2016, 03:38
No problems here with madvr v0.90.20 and amd driver 16.6.1.
Those using newer amd gpu might try downloading radeon mod and disable crimson power saving.

robl45
11th June 2016, 04:45
Tried a test file of 3d mkv mvc tonight. Using latest lav and .9 madvr mpc-hc promptly crashed and I got a box onscreen saying the program terminated. Any ideas?

Olivier C.
11th June 2016, 14:11
@madshi :

Command line seems not to work.

I found this quite old open ticket :
http://bugs.madshi.net/view.php?id=210

Do you plan to fix it in a near future ?

Thx

robl45
11th June 2016, 15:37
Tried a test file of 3d mkv mvc tonight. Using latest lav and .9 madvr mpc-hc promptly crashed and I got a box onscreen saying the program terminated. Any ideas?

So I tested again and this time it did start and Lav showed it using the mvc decoder. Of course my 9 year old computer was maxed out on cpu and it was choppy, but it was working. However it displayed it side by side on mpc which I thought was odd. When I set my projector to side by side mode. The 3d did work just fine, is this the way its supposed to work? Also the first time I tried it, I opened the file in KODI and passed to mpc-hc. That is when mpc crashed, the second time I opened in mpc directly. Is there some reason it would crash like that?

Below is the sample file I used.

https://drive.google.com/file/d/0BwxFVkl63-lETWMzM1dRZF9XMDA/view?usp=sharing

shafat777
11th June 2016, 17:28
Long time fan and follower of this thread. Recently signed up.

I just purchased a 1440p monitor and now all of my 1080p blurays needs to be upscaled to fit my screen. Since madvr does a fine job in both up and downscaling, would it be wise to double luma and chroma of the 1080p movie using nneid3 and then using ssim to downscale? OR should i just let jinc take care of luma and the usual nneid3 for chroma?

I am having a hard time figuring out which one will technically give me a better result. Any help would be appreciated.

EncodedMango
11th June 2016, 18:28
I usually watch videos in a smaller window(720) instead of Fullscreen(1080) for multi-tasking purposes so I never noticed this before but I tried a few things fullscreen recently and all of them have a weird minute but noticeable lag when using FSE mode. As soon as I disable FSE the playback takes the 'normal' speed. I tried messing with all other settings and FSE turned out to be the culprit, nothing else helped. There were no dropped frames, delayed frames or any other issues.

Unrelated to above, for rendering performance should I care more about avg rendering time or max? Sometimes the max seems to shoot up high but avg remains below v-sync.

madshi
11th June 2016, 18:54
Thanks for the changes, is there now a grain detection implemented for deringing?
Yes, something like that.

Result in this cartoon looks very clear and also otherwise well to me. Look at how enormous the difference with super-xbr sharpness 150 is

Looks like pure magic is going on. :eek:
Glad you like it!

1) could you please add tags for borders cropping? some TV logos only need a few lines of cropping to fully disappear and I'd like to nail this once and for all.
Maybe some time in the future.

3) hardly worth a heart attack but why does the latter jump from 5 to 7? Eventually I got a video that would ideally need 6 ^^
What is "the latter"?

4) TV broadcasters like to mess with levels and even more AR, there's no tag for the latter according to this (http://forum.doom9.org/showpost.php?p=1709764&postcount=2) as it would appear? I've seen many 2.35 music videos airing as 1.78 :o
Managing AR is usually the duty of the media player. The media player usually has options to overwrite the AR, which is then passed to madVR, and madVR does what the media player says.

Sorry for the double post and I do realize that I'm extremely late to the party but I think a few more lines of cropping woulda been in good order?

Please advise whether that'd make a sample useful to anything as many music videos on that channel end up with that white bar in the top right corner :o
Euw, that's ugly. Atm I don't really have time for this. Maybe you can ask again a couple months later?

if I have Bluray 3D which I use . which is not SidebySide or Top&Buttom , it is Active 3D TV with Nvidia 3DTV software. I can't make it to show actual 3D when using 3D on AUTO in madvr with 3DTV Enabled. it says D3D11 Windowed or Fullscreen (3D) but there is nothing being visible as 3D. my TV goes to 3D mode when it detect supported 3D software or Nvidia 3DTV enabled, and Potplayer goes to 3D mode, but again nothing is visible as 3D , just plain 2D. can I fix that ? I can't watch the movie in PowerDVD because I need subtitles and the movie doesn't have subtitles and I can't add to powerdvd srt subtitles, is there other way for that too ? in PowerDVD it works properly, 3D is visible and pretty much perfect.
Is it an official unmodified (not remuxed or changed in any way) 3D Blu-Ray? Does 3D work at all for you with madVR with any 3D Blu-Rays? Which desktop resolution do you have? And is 3D enabled in your OS settings?

Can the anti halo algorithm address the halo generated by SVP?
I don't know, but I would guess the SVP halos are probably too wide. Not sure, though.

About that cleanup feature: I was pleasantly surprised to realize that madVR is apparently analyzing the image to decide if cleaning is required or not - you're saying that's what it's meant to be doing, right?

I definitely don't want any unnecessary cropping with high quality sources whereas with some DVD rips (Monty Python's Flying Circus being a prime example) I need to set the cropping to at least 3 or more for all borders.

Initially I kind of assumed all content would get brutally cropped, whether needed or not. But I have to say I really like the versatility of the zoom functions and that's saying a lot because previously I was biased against "any kind of zooming or cropping" to preserve the original aspect ratio.
The whole "zoom control" section is based on careful analyzation of the actual content to make sure cropping is only done if/where needed. Of course such automatic intelligent algorithms can sometimes fail, but for my own needs (mainly detecting different movie aspect ratios and scaling them accordingly for my CIH front projection setup) it seems to work very well, though.

What are the practical differences between the tags levels, blacklevel and brightness in correcting washed out blacks in an encode that otherwise has flawless PQ? Mediainfo reports "limited" but the black level is definitely off. Which tag would be the preferred option to fix this?
Usually the double and triple expanded tags should be the best choice, unless it's only the black level and not the white level which is washed out. In that case setting blacklevel would be preferable. "brightness" doesn't change either black nor white level, it changes the gamma curve to make the whole image look more or less bright, while keeping black and white level the same.

After cropping black bars, it seems possible to place txt/image subtitles over/under video area with potplayer's built-in subtitle renderer. I only enabled two options in madVR settings: "automatically detect hard coded black bars", "crop black bars" and let potplayer control the position of subtitle. It works well but there is one thing cannot be resolved within potplayer. The conflicting option of potplayer is to reserve two lines for bottom margin for subtitles. Could you look into the issue?
So what happens? How is it not working? Maybe you can make a screenshot of how the resulting image looks like, and draw into the image to show how you WANT it to look like, so that I can see what you mean. Right now I'm not sure.

In any case, I highly doubt there's anything I can do about it. The potplayer internal subtitle renderer is responsible for all the drawing, and it can get all information it needs from me (it just needs to ask). So this will probably have to be fixed by the potplayer developer.

I am happy using the deringing option, which I have globally enabled for all content. The good thing is it doesn’t do any harm when the content doesn’t have any obvious ringing.
:)

1) How does blacklevel work exactly - it compresses/expands the range of blacks below/above 16?
Looking at a grayscale, it makes the whole grayscale darker, but white stays untouched, and gray tones near white are changed only a little, while gray tones near black are changed a lot.

2) When correcting a washed-out picture using blacklevel, should whitelevel or some other parameter be adjusted too to compensate (and if yes, how/why exactly in relation to blacklevel?)
I can't say, it depends on the source. Usually, if the black level is washed out, so is the white level. But it doesn't always have to be that way.

I'm on the latest madVR and Kodi DSplayer but no luck.

I have no idea if it should matter but anyway I even tried setting the whole chain to full RGB (I usually have GPU on full while madVR, Kodi & TV are on limited). OSD tells me levels are changed but picture looks the same. Which, like I said, seems very odd as the blacklevel tag does work. Could the "culprit" be Kodi, which has its own range setting?
Sounds strange. Don't know why it doesn't show any affect for you - it should! It could depend on a lot of things, e.g. whether you use DXVA decoding, scaling or deinterlacing, or on some madVR settings. It's unlikely to be Kodi related, but I can't say for sure. Maybe you can figure out which setting is reponsible for it working or not working. If you do, please let me know. For all other users (and me) it seems to work, so it's hard for me to figure out what's going on on your side.

I have an odd but very specific bug to report. Using the latest AMD drivers (5/23 I believe), madVR v0.90.20, and HD7750.

When playing back 1080i content (either regular or film deinterlacing, doesn't matter), I get 1 line pixel flickering around all 4 borders, unless "use a separate device for presentation" is checked. This only affects D3D9 exclusive mode. Windowed and D3D11 are fine.

Edit: It's not just 1080i29.97 videos, it's also doing it on 720p29.97 videos.
Sounds like a driver or hardware issue.

Also, NNEDI3 isn't working at all.
Try reinstalling the GPU drivers. Or try a different GPU driver version.

1) "display" shows the refresh rate adjusted for the video clock deviation measured against the system clock
There's no adjustment for video clock deviation. "display" simply shows the number of VSyncs per second, measured by the system clock.

2) "clock deviation" shows the deviation for the audio clock, again measured against the system clock.
Yes.

I'll make an example of what we found for a 59.940 video content played back with bitstreaming audio at 59.940 display refresh rate.

Display: 59.94174
Clock Deviation: +0.00276%
1 frame drop every 3.17 hours
movie 59.940 fps

audio clock deviation 27.6ppm
video clock deviation: (59.94174 - 59.94) / 59.94 = 29.03ppm
total deviation: video clock dev - audio clock dev = 28.86ppm - 27.6ppm = 1.43ppm

frames in 3.17 hours = (3600 * 59.940) * 3.17 = 684,035.28

1 frame dropped every 684,035.28 frames. Accuracy of 1 / 684,035 = 0.0000014619, that is 1.46ppm (I guess this is well within the margin of error in madVR approximations once it reaches the x.xx hours stage).

The above tells me that you are using the approximation of 59.940 in the internal calculations, and not the 60/1.001 value. Is this correct?
If using the 60/1.001 value we would have:

audio clock deviation 27.6ppm
video clock deviation: (59.94174 - (60/1.001)) / (60/1.001) = 28.03ppm
total deviation: video clock dev - audio clock dev = 28.03ppm - 27.6ppm = 0.43ppm

frames in 3.17 hours = (3600 * (60/1.001)) * 3.17 = 684,035.96

1 frame dropped every 684,036 frames. Accuracy of 1 / 684,036 = 0.0000014619, that is again 1.46ppm

0.43ppm clock difference audio/video seems too low if compared to what drop/repeats at every 3.17 hours indicate, while 1.43 is very similar.
So, are you using 59.940 for internal calculations? Does this hold true for 23.976 as well?

Sorry, I know we are reaching the human limits of geekiness here... but we would really love to fully understand the underlying math :)

I also can't understand why my clock deviation sits constantly at about 0.00276% at 59.940 refresh rate while going down to 0.00242% at 23.976 refresh rate. It's the audio clock measured against the system clock. Why should display refresh rate influence the deviation? I have been checking and checking but the difference between the two refresh rates stays there... It's audio and video through HDMI on a GTX660, if it matters.
Ehm, I'm getting headaches thinking about this. From what I've seen, you guys have all this deeply discussed in some other forum, I don't think I can add much information to that. And I really don't want to dive into all this right now. I've dozens of posts to reply to, and not much time, sorry.

madshi
11th June 2016, 18:55
Thanks for the new version, the tweaked deringing algo is definitely better. Much less artifacts and deringing is still very powerful. Ok sometimes it removes some details but a lot less that others deringing algo I've seen so far so it doesn't bother me, I know it won't never be perfect.

The only real artifact that remains is the one I showed you... Do you need a sample of this video?

But I admit I find that the pros of deringing outperform the cons that remain this time :)
A sample would be useful, and maybe an easy way to reproduce the exact artifact you mean. E.g. first frame, with screenshots of how it looks with/without deringing on your PC. When I tried with your BMP/JPG, I couldn't really see much difference with deringing on/off, IIRC. The difference I saw here was much smaller than what your screenshots showed.

This CrossBilateral mpv shader works very well.
https://gist.github.com/igv/a015fc885d5c22e6891820ad89555637
I hope you will port it inside madVR soon!
It's not mpv's work. They've copied this one from Shiandow/MPDN. It's on my to do list to add it to madVR.

Could you bring the one-pass AdaptiveSharpen back into madVR?
https://gist.github.com/igv/8a77e4eb8276753b54bb94c1c50c317e
It uses much less gpu and looks pretty good.
I'm confused. I thought the 1pass algo was outdated and replaced with the 2pass algo? Is that not the case? Is the 1pass algo different (lower quality, but faster, maybe)? Is it still maintained?

They could have at least mentioned my name somewhere...
Agreed. But they don't do things like that.

How about graying out the checkbox when a 3DLUT that contains calibration information is loaded?
It's not that easy. There are multiple 3dlut slots, some of them might have a 3dlut file loaded with included GPU gamma ramps, some might not. What do I do then?

Anyway, this is not the time to do minor cosmetical things like this. At some point I'll invest time into making all the settings easier and nicer. Working on cosmetical things now would just cost me hours and hours of work, which I don't have.

Hi,My RAM is getting full after playing video say for 1hr.
Got 8GB RAM & at starting of player , it's 25%.
But after 1hr it becomes 95%.
Forwarding ,going into fullscreen or out of full screen takes time like forever.
it's like something is loading into RAM but not releasing.
Nobody else is reporting this problem. Could you please try to find out where it's coming from? It could be madVR. It could be the media player. Or the decoder. Or some other software running on your PC. E.g. you could try different madVR builds, different media players, different decoders/splitters, to figure out which exact piece of software is causing the issue. It could even be the GPU driver, or some 3rd party GPU monitor program or something.

the Dolby Vision clips works but madvr doesnt detect them as HDR and so they look washed out.
I've downloaded 2 clips. One was detected as HDR and played fine. One was not detected as HDR, but still played fine and was *not* washed out. It's too bad you didn't tell me exactly which file displayed washed out. I don't have the time to download 30 files and try them all one by one.

Is my CPU potentially bottlenecking the GTX 960?
No. Your first and only priority should be to make GPU accelerated decoding work. Until you succeeded with that, you can ignore all other problems.

Wrestling with my old laptop that I use as a HTPC (AMD Phenom II N930 2Ghz and AMD 5650m+4250 gpu), and after reinstalling windows 7 64-bit and updating mpc-hc and madVR (64-bit) to their latest nightly, I can barely get bilinear to work. Is there any change about performance in older rigs as mine from 89.x to latest? Or am I missing some requirement I am supposed to install (vcredist etc)? GPU-z shows that the clocks go up for 3D (550/800) and that the utilization is low (0%-20%), yet still I get "dips" in the queues occasionally and starting+switching to fullscreen results in around 20-30 dropped frames, no matter if exclusive or not. I am "forced" to use gpu-drivers from 2012 (Catalyst 12.4 to be exact) as I have a legacy GPU on my motherboard and AMD doesn't support them.
So the problem is not generally bad rendering times, but occasional dips? Do you have any 3rd party software installed which might be using the GPU? E.g. try closing the browser. Disable all GPU monitors and all other 3rd party software which might have anything remotely to do with the GPU.

New settings are great, low noise, sharp image.
New deringing artifacts filter + reduce dark halos is too heavy for my graphic cards (8750M), but without it, everything still great :D

If i enable deringing artifacts filter, rendering times increase from 18ms to > 40ms with high bitrate 720p video (>10000kbps), and framerate drops quickly.
Unfortunately there's not much I can do about deringing performance. It does eat some performance, but that's just the way it is.

When doing 720p -> 1080p, madVR OSD has started showing Jinc chroma upscaling (on the row after luma) in addition to the Super-XBR chroma upscaling that I've chosen for the profile.
It's hard to say something without seeing a screenshot of the OSD.

Some of them doesn't have any colorspace declaration (e.g. LG_2_DEMO_4K_O_H_2_Dolby Trailer.mkv) and MadVR uses BT.709 (best guess) for primaries and for matrix.
"LG_2_DEMO_4K_O_H_2_Dolby Trailer.mkv" looks ok to my eyes, so it doesn't really seem to be HDR at all.

If I hit source decoding matrix toggle, I can only select from these: BT.709 , BT.601 .
If I hit source primaries toggle, I can only select from these: BT.709 , SMPTE C, EBU/PAL .

Shouldn't I able to select the other ones (DCI-P3, BT.2020) as well with both of these options?
If I assign keyboard shortcuts to DCI-P3 and BT.2020 then I can select them as primaries. Though there's no option for the matrix.
I'm only toggling through the most common candidates. HDR content is still very rare. And *usually* it's tagged correctly. I'm not sure if I should extend the toggle for these extremely rare cases. It would make things worse for 99% of users who don't need it.

There's no matrix for DCI-P3. Content is not encoded in DCI-P3. Or if it is, it's RGB. So a matrix for DCI-P3 is not needed, so none exists.

One more question: If I'm certain that a source uses BT.2020 (let's say) what source decoding matrix and source primaries should I use on my BT.709 calibrated FHD display? (I guess BT.2020 for source primaries and BT.709 for decoding matrix, but I'm not sure).
It's unlikely the source uses BT.2020 primaries but a BT.709 matrix. If you're sure that madVR has guessed wrong, then you should correct both the primaries and matrix to what the video really uses. However, the file is seriously broken if it uses BT.2020 without saying so in the bitstream!!!

They released new drivers with fixed DXVA processing (which was acting strange with 16.5.2 and 16.5.3 drivers). Not sure if this was the crash, but if the workaround is somehow hacky you might recheck if it's needed.
Thanks. The workaround should not have any negative side effects, and it was not related to DXVA in any way. It had to do with my low level API hooks to make Direct3D use proper refresh rates.

Is there a way to list madVR instances already running (ex: MPC HC playing a video) and hook to them (COM interfaces) ?
Yes, after all madHcCtrl.exe (the madVR tray icon application) does it, too. It's not using a COM interface, though. Instead it loads "madHcNet32.dll" and uses APIs of that dll.

What purpose do you need this for?

Hello madshi, despite using the latest madVR version (0.90.20) I'm still experiencing driver/madVR crash or even BSoD while using the latest AMD beta drivers (16.X).
If you get a madVR crash, please send the crash report to me. If the GPU driver crashes or even creates a BSoD, then this is probably out of my control, and unlikely to be my fault. The driver should really not crash or BSoD.

Command line seems not to work.

I found this quite old open ticket :
http://bugs.madshi.net/view.php?id=210

Do you plan to fix it in a near future ?
At some point. Not sure if it will be the near future. What do you need this for?

So I tested again and this time it did start and Lav showed it using the mvc decoder. Of course my 9 year old computer was maxed out on cpu and it was choppy, but it was working. However it displayed it side by side on mpc which I thought was odd. When I set my projector to side by side mode. The 3d did work just fine, is this the way its supposed to work? Also the first time I tried it, I opened the file in KODI and passed to mpc-hc. That is when mpc crashed, the second time I opened in mpc directly. Is there some reason it would crash like that?
Which OS are you using? You need at least Windows 8.1 for frame packed 3D output to work. Windows 7 doesn't have APIs for that.

I just purchased a 1440p monitor and now all of my 1080p blurays needs to be upscaled to fit my screen. Since madvr does a fine job in both up and downscaling, would it be wise to double luma and chroma of the 1080p movie using nneid3 and then using ssim to downscale? OR should i just let jinc take care of luma and the usual nneid3 for chroma?

I am having a hard time figuring out which one will technically give me a better result. Any help would be appreciated.
Usually better quality luma processing is much more visible than better quality chroma processing. So I'd recommend choosing a sane but not too expensive chroma upscaling algorithm, and instead spend your GPU power on high quality luma upscaling and refinement.

But in the end, why don't you experiment and let your eyes be the judge?

I usually watch videos in a smaller window(720) instead of Fullscreen(1080) for multi-tasking purposes so I never noticed this before but I tried a few things fullscreen recently and all of them have a weird minute but noticeable lag when using FSE mode. As soon as I disable FSE the playback takes the 'normal' speed. I tried messing with all other settings and FSE turned out to be the culprit, nothing else helped. There were no dropped frames, delayed frames or any other issues.
I'm not sure what you mean. What exactly lags? The playing video? Or some OSD controls? Or what?

Unrelated to above, for rendering performance should I care more about avg rendering time or max? Sometimes the max seems to shoot up high but avg remains below v-sync.
As long as all queues remain nicely filled at all times, there's no need to worry. The queues have exactly the purpose to smooth out unexpected "max" spikes.

EncodedMango
11th June 2016, 19:04
I'm not sure what you mean. What exactly lags? The playing video? Or some OSD controls? Or what?
Hello, yes the video lags a little.

I'm having issue replicating it properly though, so will do more testing and report back. I just did a full settings and configuring madVR options again and the problem doesn't seem to be reappearing so far.

As long as all queues remain nicely filled at all times, there's no need to worry. The queues have exactly the purpose to smooth out unexpected "max" spikes.
I see, thank you.

castaway
11th June 2016, 21:26
Yes, after all madHcCtrl.exe (the madVR tray icon application) does it, too. It's not using a COM interface, though. Instead it loads "madHcNet32.dll" and uses APIs of that dll.

What purpose do you need this for?

To change madVR common enhancement settings instantly on a video already playing (on the fly).
I could successfully use the interface you provided to modify settings (ex: switch on lumaSharpen) , but they are not applied directly, only after restarting the player.

Is there a way to "apply" the changes made, through the interface ; or else would "madHcNet32.dll" have APi's to achieve that?
Thanks :)

Uoppi
11th June 2016, 21:27
When 720p -> 1080p upscaling chroma using Super-XBR (just upscaling, not doubling), how is chroma upscaled exactly? Is it doubled to 1080p (like luma) or is quarter of the resolution doubled to half of the 1080p resolution and then the rest of the upscale is done with Jinc?

Just trying to figure out if Jinc chroma should also be shown in the madVR OSD at the same time with Super-XBR chroma. I'm seeing both Super-XBR and Jinc chroma in the OSD and I can't remember seeing Jinc chroma there before.

robl45
11th June 2016, 22:11
I am using windows 10. So if its showing side by side like that and I put the projector into side by side mode, I'm not getting the full resolution? I have a feeling the drivers might not be correct in windows 10 on my system or its not detecting my projector correctly. But if i'm getting full resolution like this, then I don't really need to mess with it.

Quote:
Originally Posted by robl45 View Post
So I tested again and this time it did start and Lav showed it using the mvc decoder. Of course my 9 year old computer was maxed out on cpu and it was choppy, but it was working. However it displayed it side by side on mpc which I thought was odd. When I set my projector to side by side mode. The 3d did work just fine, is this the way its supposed to work? Also the first time I tried it, I opened the file in KODI and passed to mpc-hc. That is when mpc crashed, the second time I opened in mpc directly. Is there some reason it would crash like that?
Which OS are you using? You need at least Windows 8.1 for frame packed 3D output to work. Windows 7 doesn't have APIs for that.

Sunset1982
12th June 2016, 10:31
just ordered a msi gtx1070. Anyone using 1070/1080 already?

What settings are you using to max it out?

mueslibrown
12th June 2016, 13:46
Is there currently any reason not to use Windows 10 (nvidia plus i3) and madvr? Thx.

ashlar42
12th June 2016, 15:16
There's no adjustment for video clock deviation. "display" simply shows the number of VSyncs per second, measured by the system clock.Yes, but being measured against system clock (same reference as audio clock as you confirm), and knowing the target refresh rate of timings selected, it pretty much is a measure of video clock deviation. Tests seem to confirm this to several decimals of precision.

Ehm, I'm getting headaches thinking about this. From what I've seen, you guys have all this deeply discussed in some other forum, I don't think I can add much information to that. And I really don't want to dive into all this right now. I've dozens of posts to reply to, and not much time, sorry.We tried refining the process and came up with the guide you can read here: http://forum.doom9.org/showthread.php?t=173571

I'd be happy if you, whenever you'll have the time, could have a look and see if any glaring mistake jumps to your eye. We've repeated the process and it seems to work but... just in case. :)

Thanks! :)

madshi
12th June 2016, 17:41
Madshi, as you asked, here my settings.bin and my debug log from the supersampling situation where image doubling activates but upscaling refinement does not

http://www.filedropper.com/madvrdebugsettings

movie resolution : 1280x720
display resolution: 1366x768

thanks for your atention and patience
Can you please retry with the next build (v0.90.21). If the problem still occurs, can you please create another debug log with the next build? Thanks.

Houston, we have a problem(that's with .19 BTW):

http://thumbnails116.imagebam.com/48648/3c0d65486479708.jpg (http://www.imagebam.com/image/3c0d65486479708) http://thumbnails115.imagebam.com/48648/423fd3486479709.jpg (http://www.imagebam.com/image/423fd3486479709)

Seems boiling down to FS(whatever E or W) + SSIM 2D + black bars detection as either going 1D or disabling the latter works fine, here's a MKV sample (https://mega.nz/#!Xxx0gaYQ!fShBxbO0z3Xi2xGgWSwfmxIgpeOAtXWEFGDlix9h4n0).

So I captured a big .TS file off DVB-T2 and only this music video in the sample gave that first problem but once I enable doubling they all go bananas like this in FS: http://thumbnails116.imagebam.com/48649/440f09486480480.jpg (http://www.imagebam.com/image/440f09486480480)
Should be fixed in the next build.

Hello, yes the video lags a little.

I'm having issue replicating it properly though, so will do more testing and report back.
If it occurs again, please describe in more detail how the video lags. I'm not fully sure what you mean. But if the problem is already gone, as is well, of course.

When 720p -> 1080p upscaling chroma using Super-XBR (just upscaling, not doubling), how is chroma upscaled exactly? Is it doubled to 1080p (like luma) or is quarter of the resolution doubled to half of the 1080p resolution and then the rest of the upscale is done with Jinc?
It's doubled to video resolution (720p), then the whole image is upscaled to 1080p.

I am using windows 10. So if its showing side by side like that and I put the projector into side by side mode, I'm not getting the full resolution? I have a feeling the drivers might not be correct in windows 10 on my system or its not detecting my projector correctly. But if i'm getting full resolution like this, then I don't really need to mess with it.
Side-by-side loses half of the resolution. You need 3D to be enabled for your projector in the OS settings. For that to work your GPU must detect the display as being 3D capable. If you route your projector through some kind of AVR, make sure that AVR doesn't swallow the "3D capable" information (mine does!). You could try connecting the projector directly to the GPU to double check that.

Is there currently any reason not to use Windows 10 (nvidia plus i3) and madvr? Thx.
Yes, Windows 8.1 is currently more stable for video playback.

We tried refining the process and came up with the guide you can read here: http://forum.doom9.org/showthread.php?t=173571

I'd be happy if you, whenever you'll have the time, could have a look and see if any glaring mistake jumps to your eye. We've repeated the process and it seems to work but... just in case. :)
I'm sure you got it right.

To change madVR common enhancement settings instantly on a video already playing (on the fly).
I could successfully use the interface you provided to modify settings (ex: switch on lumaSharpen) , but they are not applied directly, only after restarting the player.

Is there a way to "apply" the changes made, through the interface ; or else would "madHcNet32.dll" have APi's to achieve that?
Thanks :)
Oh well, you can use this header:

http://madshi.net/madHcNet.h

Sending settings works by using "AutoNet_Send" with the "SetSettings" command. The whole operation is somewhat complicated, though. First you'd need to call "AutoNet_AddConnectionCallback" and "AutoNet_Listen" to be notified about existing madVR instances. Your connection callback will then be called with information about found instances and in the callback you could then call "AutoNet_Send".

All this stuff wasn't really meant to be used by anyone else, other than me, so the headers are not that well documented, and the process isn't well explained.

madshi
12th June 2016, 19:04
madVR v0.90.21 released

http://madshi.net/madVR.zip

* HDR: pixels between 0-100 nits are now left untouched (if possible)
* HDR: improved desaturation algorithms to reduce hue shift even further
* HDR: added option to choose between clipping and tone mapping
* HDR: added option for hue preservation quality
* HDR: added option to restore detail in compressed regions
* HDR: added option to measure each video frame's peak luminance
* HDR: added option to completely disable HDR processing
* HDR: added workaround for files with invalid metadata
* fixed: supersampling + 2D SSIM sometimes produced "green" video
I had planned to add Shiandow's latest Bilateral chroma upscaler this week, but didn't manage to squeeze it in, unfortunately. Hopefully next weekend.

FWIW, I'm probably going to have less time for madVR in the next couple of weeks, compared to the previous months. So don't expect big new features in the next couple of weeks. Just bugfixes and small tweaks, maybe.

kasper93
12th June 2016, 20:31
Something is wrong here. When display is set to 120 nits no matter if HDR processing is enabled or not or even if the file have HDR data. Video brightness is significantly higher than it should be.

madshi
12th June 2016, 20:40
Something is wrong here. When display is set to 120 nits no matter if HDR processing is enabled or not or even if the file have HDR data. Video brightness is significantly higher than it should be.
Ooops, that appears to be a bug, thanks for letting me know. I wouldn't recommend the 120 nits setting (or 180 nits), anyway. But of course this needs to be fixed. I'd recommend using 265 nits or higher, with which the issue should not occur.

The reason for this bug is that for 120 and 180 nits, I'm using the same tone mapping curves as for 265 nits, but I'm "secretly" increasing the brightness setting (for 120 nits more than for 180 nits, obviously). It seems that this secret brightness tweak isn't limited to HDR videos, although that was the intention.

tahaa7
12th June 2016, 21:16
madVR is not able to leave an interlanced signal untouched.
i'm not sure if this is even possible with a PC.

just chroma upscaling to 4:4:4 will ruin an interlanced image if it is done before deinterlancing.

But if I turn off deinterlacing in madVR, I do get an interlaced signal out (I see combing), but my TV thinks it's progressive. I've tried setting the "display mode" in madVR to 1080i50, but still my TV reads it as "1080p". Which also begs the question when might adding interlaced options to "display modes" of a TV/display be useful...