Skip to content
This repository was archived by the owner on May 17, 2023. It is now read-only.

Commit 98c0c36

Browse files
daleksanOleg Nabiullin
authored and
Oleg Nabiullin
committed
Update release notes for 20.5.1 release
1 parent b92339e commit 98c0c36

File tree

1 file changed

+26
-25
lines changed

1 file changed

+26
-25
lines changed

doc/mediasdk_release_notes.md

+26-25
Original file line numberDiff line numberDiff line change
@@ -35,54 +35,51 @@ In this document we will refer to processor families by their former codenames f
3535
- Intel® Celeron® and Pentuim® processors codenamed "Apollo Lake I" will be referred to as “Apollo Lake”
3636

3737

38-
## What’s New in Media SDK 20.3.0
38+
## What’s New in Media SDK 20.5.1
3939
**New features**:
40-
* **API 1.34**
41-
- Added Keem Bay platform support
42-
- Added AV1 decode support
40+
* **AV1 decode**
41+
- Refactored handling errors in SubmitFrame
4342

4443
* **AVC encode**
45-
- Added implicit Motion-Compensated Temporal Filter (MCTF) to improve coding quality on certain scenarios
46-
- Improved CodecLevel initialization from rate control parameters
47-
- Added support of 4k streams with look-ahead bitrate control mode
48-
- Enabled QPOffset by default
44+
- Improved handling support status for "Sliding window" feature
45+
- Added EncTools support
46+
- Added vaSyncBuffer support
47+
- Added external BRC support for Rocket Lake
48+
- Added TCBRC support
49+
- Added MCTF support
4950

5051
* **HEVC encode**
51-
- Improved efficiency of FPQ
52-
- Enabled QP modulation (Icelake+)
52+
- Extended B frames support across all target usage with LowPower on
53+
- Added TCBRC support
5354

54-
* **VP9 decode**
55-
- Added 16k support
56-
57-
* **VPP**
58-
- Added A2RGB10 input in case of passthru copy
55+
* **MPEG2 encode**
56+
- Added vaSyncBuffer support
5957

6058
* **Samples**
61-
- Added -dec::sys flag for setting of the output memory to system type in sample_multi_transcode
62-
- Added p016/y216 support in sample_encode
63-
- Added i420/nv12 support in sample_multi_transcode
64-
- Fixed chroma size for Y210/Y216 formats in sample_deocde
65-
- Added handle for iterpolation methods for resize in sample_vpp
59+
- Added AdaptiveI and AdaptiveB frames support in sample_multi_transcode (#1401)
60+
- Added NV16 output support
61+
- Added 12 bit support in sample_multi_transcode, sample_decode and sample_vpp (#2055)
62+
- Enabled MBQP for all encoders in sample_multi_transcode
63+
- Added -disable_film_grain flag to disable film grain application in sample_decode
6664

6765
* **Misc**
68-
- Added support for discrete Intel Xe graphics codenamed DG1 and SG1
69-
- Added support for Rocketlake platform
70-
- Actualized Elkhartlake device id support list
66+
- Disabled plugins loading (Plugin implementations were moved to library runtime, except LA and FEI)
67+
- Added support for dynamic GPU session priority
7168

7269
* **Software requirements**
7370
- Libdrm 2.4.84 or later
7471
- Kernel 4.14 or later (5.4 recommended, consult [kernel support matrix](https://github.com/Intel-Media-SDK/MediaSDK/wiki/Intel-Graphics-Support-in-Linux-Kernels) wiki page for details)
7572

7673
* **Known issues**
7774
- Kernel 5.0 have known issue with endurance on Skylake see https://bugs.freedesktop.org/show_bug.cgi?id=110285 for details.
78-
- Media Stack 20.3 with Linux kernel 5.0 may cause higher CPU usage and instability in endurance testing. Issue root caused to media driver see https://github.com/intel/media-driver/issues/671 for details.
75+
- Media Stack 20.5.1 with Linux kernel 5.0 may cause higher CPU usage and instability in endurance testing. Issue root caused to media driver see https://github.com/intel/media-driver/issues/671 for details.
7976
- AV1 Decoder doesn't properly support asynchronous decoding (AsyncDepth > 1) and sometimes may return *MFX_ERR_UNDEFINED_BEHAVIOR* and *MFX_ERR_DEVICE_FAILED* from DecodeFrameAsync calls. See #2244 for details.
8077

8178
## <a id='System_Requirements'>System Requirements</a>
8279

8380
**Hardware**
8481

85-
Intel® Media SDK supports the following platforms with the integrated graphics:
82+
Intel® Media SDK supports the following platforms with the Intel graphics:
8683
- Intel® Celeron® and Pentuim® processors codenamed "Apollo Lake I"
8784
- Intel® Xeon® E3-1200 v4 Family with C226 chipset
8885
- Intel® Xeon® E3-1200 and E3-1500 v5 Family with C236 chipset
@@ -97,6 +94,7 @@ Intel® Media SDK supports the following platforms with the integrated graphics:
9794
- Note: chipset must have processor graphics enabled; make sure to check the datasheet.
9895
- Having a C226/C236 chipset is necessary but not sufficient. Make sure to consult with specific platform or board vendor regarding processor graphics being supported. Check Media Server Studio website for the list of “Known OEM/ODM Functional Platforms”:
9996
https://software.intel.com/en-us/intel-media-server-studio/details
97+
- Intel® Iris® Xe MAX Graphics require special setup decribed in the following online guide: https://dgpu-docs.intel.com/devices/iris-xe-max-graphics/index.html
10098

10199
**Software**
102100
- **Linux kernel 4.14 or newer** of 64 - bit architecture. Kernel 4.19 or newer recommended, as it contains stability and performance improvements over 4.14. Release was validated against 4.19.5 for gen8 and gen9 graphics, 5.4 for gen10 (Icelake) graphics, and 5.5 for gen11 (Tigerlake) graphics.
@@ -106,6 +104,7 @@ Intel® Media SDK included in this package implements SDK API 1.34 and contains
106104

107105
Component | Supported features | Limitations
108106
--- | --- | ---
107+
AV1 decoder | Supported Profiles:<ul><li>0, 8bit and 10bit</li></ul> | Maximum supported resolution:<br/>16384x8704
109108
H.265 decoder | Supported Profiles:<ul><li>Main</li></ul> | Maximum supported resolution:<br/>8192x8192
110109
H.265 encoder | Supported Profiles:<ul><li>Main</li></ul>Supported BRC methods:<ul><li>Constant QP (CQP)</li><li>Constant Bit Rate (CBR)</li><li>Variable Bit Rate (VBR)</li><li>Variable Bit Rate with Constant Quality (QVBR)</li><li>Software BRC</li></ul> | Maximum supported resolution:<br/>4096x2176
111110
H.264 decoder | Supported Profiles:<ul><li>Baseline</li><li>Main</li><li>High</li></ul> | Maximum supported resolution:<br/>4096x2304
@@ -166,6 +165,8 @@ NOTE: Please use *Query* functions to check feature availability on any given ma
166165

167166
Please see the Intel® Media SDK Reference Manual for details *&lt;sdk-install-folder&gt;/doc/mediasdk-man.pdf”*.
168167

168+
Please also check the collateral materials: https://dgpu-docs.intel.com/devices/iris-xe-max-graphics/guides/media.html
169+
169170
## <a id='GPU_Hang_Reporting_And_Recovery'>GPU Hang Reporting And Recovery</a>
170171

171172
Intel® Media SDK supports reporting of GPU hangs occurred during SDK operations. This feature allows application to establish proper GPU hang recovery procedure without the need for additional monitoring of the system (e.g. without checking dmesg logs).

0 commit comments

Comments
 (0)