Search the Community

iGPU hw transcoding only works on plex for DS918+ (arc loader 24.1.17 + addon i915) for other models the /dev/dri folder is empty ((

Please try using SA6400 with the loader you are comfortable with, whether it is rr or my mshell. This is the topic of @MoetaYuko, who developed the transcoding i915 module for SA6400 last month. He was the first to test and stabilize SA6400 transcoding on Alder Lake N100. https://xpenology.com/forum/topic/69865-i915-driver-for-sa6400/ However, if the i226-v chipset is used in N100, the NIC chipset may become an issue.

Hello community , I'm having some issues with HW transcoding and need some clarifications System: DSM918+ OS: Synology DSM 7.2.1-69057 update 3 Loader : ARC 24.1.6 Hardware: unknown vendor, bought from Aliexpress. CPU N5095 with integrated GPU, 12SATA ports (used 5), 2 NICs. BIOS from AMI, project W-N95-R build 20.09.2022 From what I see, N5095 is JasperLake generation 11. Is this currently supported by DSM 7.2 ? I can see the GPU in Plex Transcoder settings (JasperLake). I have a HVEC 10 bit file used for testing. If I select transcode from the plex menu (web) it keeps loading, system CPU spikes and gets a bit unresponsive..and nothing, no play. root@b443vpn:~# cat /sys/kernel/debug/dri/0/i915_frequency_info Video Turbo Mode: yes HW control enabled: yes SW control enabled: no PM IER=0x00000000 IMR=0x00000000 ISR=0x00000000 IIR=0x00000000, MASK=0x00003ffe pm_intrmsk_mbz: 0x80000000 GT_PERF_STATUS: 0x00c0610f Render p-state ratio: 97 Render p-state VID: 15 Render p-state limit: 255 RPSTAT1: 0x06000006 RPMODECTL: 0x00000d92 RPINCLIMIT: 0x00002c88 RPDECLIMIT: 0x00004fb0 RPNSWREQ: 200MHz CAGF: 200MHz RP CUR UP EI: 4692 (6256us) RP CUR UP: 0 (0us) RP PREV UP: 0 (0us) Up threshold: 95% RP CUR DOWN EI: 4698 (6264us) RP CUR DOWN: 0 (0us) RP PREV DOWN: 0 (0us) Down threshold: 85% Lowest (RPN) frequency: 200MHz Nominal (RP1) frequency: 450MHz Max non-overclocked (RP0) frequency: 750MHz Max overclocked frequency: 750MHz Current freq: 200 MHz Actual freq: 200 MHz Idle freq: 200 MHz Min freq: 200 MHz Boost freq: 750 MHz Max freq: 750 MHz efficient (RPe) frequency: 300 MHz Current CD clock frequency: 24000 kHz Max CD clock frequency: 675000 kHz Max pixel clock frequency: 675000 kHz root@b443vpn:~# lspci -tvnnq -+-[0001:03]---00.0 Intel Corporation I211 Gigabit Network Connection [8086:1539] +-[0001:02]---00.0 Intel Corporation I211 Gigabit Network Connection [8086:1539] +-[0001:01]---00.0 Marvell Technology Group Ltd. 88SE9215 PCIe 2.0 x1 4-port SATA 6 Gb/s Controller [1b4b:9215] +-[0001:00]-+-12.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series SATA AHCI Controller [8086:5ae3] | +-13.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series PCI Express Port A #1 [8086:5ad8] | +-14.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series PCI Express Port B #1 [8086:5ad6] | +-15.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series USB xHCI [8086:5aa8] | +-16.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series I2C Controller #1 [8086:5aac] | +-18.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series HSUART Controller #1 [8086:5abc] | +-19.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series SPI Controller #3 [8086:5ac6] | +-19.2 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series SPI Controller #3 [8086:5ac6] | +-1f.0 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series SMBus Controller [8086:5ad4] | \-1f.1 Intel Corporation Celeron N3350/Pentium N4200/Atom E3900 Series SMBus Controller [8086:5ad4] \-[0000:00]-+-00.0 Intel Corporation Device [8086:4e24] +-02.0 Intel Corporation JasperLake [UHD Graphics] [8086:4e55] +-04.0 Intel Corporation Dynamic Tuning service [8086:4e03] +-08.0 Intel Corporation Device [8086:4e11] +-14.0 Intel Corporation Device [8086:4ded] +-14.2 Intel Corporation Device [8086:4def] +-15.0 Intel Corporation Serial IO I2C Host Controller [8086:4de8] +-15.2 Intel Corporation Device [8086:4dea] +-16.0 Intel Corporation Management Engine Interface [8086:4de0] +-17.0 Intel Corporation Device [8086:4dd3] +-19.0 Intel Corporation Device [8086:4dc5] +-19.1 Intel Corporation Device [8086:4dc6] +-1c.0-[01]----00.0 JMicron Technology Corp. JMB58x AHCI SATA controller [197b:0585] +-1c.1-[02]----00.0 JMicron Technology Corp. JMB58x AHCI SATA controller [197b:0585] +-1c.6-[03]----00.0 Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] +-1c.7-[04]----00.0 Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] +-1e.0 Intel Corporation Device [8086:4da8] +-1e.3 Intel Corporation Device [8086:4dab] +-1f.0 Intel Corporation Device [8086:4d87] +-1f.3 Intel Corporation Jasper Lake HD Audio [8086:4dc8] +-1f.4 Intel Corporation Jasper Lake SMBus [8086:4da3] \-1f.5 Intel Corporation Jasper Lake SPI Controller [8086:4da4] root@b443vpn:~# lsmod | grep i915 i915 1314764 0 drm_kms_helper 123302 1 i915 drm 313498 3 i915,drm_kms_helper iosf_mbi 4298 1 i915 fb 35077 2 i915,drm_kms_helper video 27424 1 i915 backlight 6170 2 i915,video button 5161 1 i915 i2c_algo_bit 5876 1 i915 [ 268.970218] audit: type=1105 audit(1705421941.685:200): pid=20965 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:session_open grantors=pam_env,pam_env,pam_permit,pam_umask,pam_unix,pam_limits acct="webuser" exe="/usr/sbin/cron" hostname=? addr=? terminal=cron res=success' [ 269.477092] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 270.083998] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 270.687320] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 271.290573] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 271.894298] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 272.499581] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 273.103407] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 273.132983] audit: type=1104 audit(1705421961.880:201): pid=20965 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:setcred grantors=pam_permit acct="webuser" exe="/usr/sbin/cron" hostname=? addr=? terminal=cron res=success' [ 273.138513] audit: type=1106 audit(1705421961.906:202): pid=20965 uid=0 auid=4294967295 ses=4294967295 msg='op=PAM:session_close grantors=pam_env,pam_env,pam_permit,pam_umask,pam_unix,pam_limits acct="webuser" exe="/usr/sbin/cron" hostname=? addr=? terminal=cron res=success' [ 273.706607] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 274.311265] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 274.915158] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 275.518378] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 276.121588] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 276.725612] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 277.330218] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 277.934320] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 278.538331] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 279.142898] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 279.749588] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 280.354620] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. [ 280.958268] [drm:fw_domains_get_with_fallback [i915]] *ERROR* media: timed out waiting for forcewake ack request. What else should I check? Thank you

In case someone else comes across this question - yes I was able to get this working with quicksync hardware transcoding with ESXi (8) (6700k skylake) Restart without any monitor connected (to be safe) Under Host - Manage - Hardware - Intel 530 - Toggle Passthrough (activate) On the VM - add PCI device (Intel 530) Start VM For my VM I rebuilt the loader (DS920+ dsm7.2.1 Update 1) and startup (install/migrate) once in UI, ssh in as su (or use script) and "modprobe i915" I was able to see /dev/dri then.

The suggestion that a more powerful GTX GPU might also work on DVA3221 for hardware transcoding, with the possibility of NvidiaRuntimeLibrary handling various GTX GPUs, is an interesting thought. It demonstrates a forward-thinking approach and an openness to exploring potential alternatives.

And, although it may be a waste of your effort, @MoetaYuko has already completed a stable i915 module to enable transcoding on SA6400, as shown below. This i915 module has already been applied to rr and tcrp-mshell. However, the Intel CPU generations for which transcoding has been confirmed to operate are These are the 4th, 11th, 12th, and 13th generations. For other generations, the /dev/dri directory is checked and whether the i915 module is loaded is checked, but it does not work. I am still finding out what more detailed adjustments are needed.

HW transcoding works fine when you use native compatible loader and hardware. Like a 4th to 9th gen intel iGPU with DVA1622 / DS918+ / DS920+. If it is your main goal, SA6400 and 10+ gen iGPU may not be the best choice for a loader. There is currently no official Synology running 10th+ gen intel iGPU ... so don't expect it to be a smooth work to be done.

I gave up.. with the retirement of wjz304 I decided to move away from a dedicated plex box. Synology and Plex (hw transcoding has always been hit or miss except with the Nvidia NVR route.)

"The device management of the company acquired by Intel is really a mess." Very much so, yes! Thanks for your attention to detail and application of scripts as it relates to this part/driver. Good to know that the SA6400 already has OEM support for the E3100G. I'll definitely try a SA6400 build, and testing the HW transcoding as well.

It seems that sudo is being used redundantly when the user is already a root account. It looks like you have a plex pass since you can select the hw transcoding device. Processing using i915.enable_guc to enable hw transcoding device acceleration has not yet had successful feedback for generations 5 through 10. The situation seems to be the same for the 8th generation. @MoetaYuko , the i915 developer for sa6400, couldn't tell me any more about how to enable it.

Thank you so much for sharing Peter Suh, this is awesome I think I might be doing something wrong, I cant seem to change the /sys/module/i915/parameters/enable_guc with vi or by echoing an update? Is there another way to correctly update it with a new value? using vi to update: "/sys/module/i915/parameters/enable_guc" "/sys/module/i915/parameters/enable_guc" E514: Write error (file system full?) WARNING: Original file may be lost or damaged don't quit the editor until the file is successfully written! trying to update with echo: root@slave:/# sudo echo 2 > /sys/module/i915/parameters/enable_guc -name: /sys/module/i915/parameters/enable_guc: Permission denied I can see the device in transcoder: As well as DRI in the system folders: Just everything is software transcoding:

I recently applied S.M.A.R.T activation of HBA (mpt3sas) on DS918+. It looks like there are still some problems left. I don't know if you can use a translator to translate this post from the Korean server forum where I work. https://svrforum.com/nas/1068754 We conducted transcoding tests on Kernel 5 SA6400 for all generations from Intel's 1st generation. This content also includes feedback from users. I thought there was already feedback on the 8th generation, but I was mistaken. No users have given any feedback yet. And if transcoding acceleration doesn't work, the only thing you can adjust is /sys/module/i915/parameters/enable_guc. I use a setting of 2 or 3, but I'm not sure what each number means. -> I found a setup guide for i915.enable_guc . https://wiki.archlinux.org/title/intel_graphics#Enable_GuC_/_HuC_firmware_loading

Thanks Peter Suh, it is great that these do support transcoding. They still are having issues reading information from the disks. This means I can't see disk temperatures or smart information at all using this version. Mpt3sas does work correctly with ds3617xs (without these below errors), so is likely a driver issue with DS918+, DS1019+ etc? These messages are flooded in the /var/log/messages file. Do you know if support for 8th Gen or others is still being worked on for Kernel 5? Would be great to be able to move to the latest versions 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2611]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sda/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2611]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sda/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2614]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdb/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2614]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdb/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2617]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdc/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2617]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdc/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2620]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdd/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2620]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdd/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2623]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sde/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2623]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sde/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2626]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdf/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2626]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdf/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2629]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdg/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2629]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdg/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2632]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdh/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2632]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdh/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2629]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdg/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2629]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdg/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2620]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdd/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2620]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdd/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2611]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sda/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2611]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sda/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2623]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sde/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2623]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sde/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2617]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdc/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2617]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdc/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2626]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdf/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2626]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdf/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2632]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdh/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2632]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdh/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2614]: disk/disk_is_mv_soc_driver.c:71 Can't get sata chip name from pattern /sys/block/sdb/device/../../scsi_host/host*/proc_name 2023-12-15T13:00:25+11:00 Beast synostgd-disk[2614]: disk/disk_is_mv1475_driver.c:66 Can't get sata chip name from pattern /sys/block/sdb/device/../../scsi_host/host*/proc_name

Activation has been verified only for Intel 4th generation, 11th to 13th generation IGPU, which is the transcoding of SA6400. I think the 8th generation will be difficult. There is a DS918+ (Apollo Lake) that supports both transcoding and mpt3sas. All Apollo Lake supported models support the above in the same way. DS1019+, DS620slim etc...

Hi Peter Suh, I'm running a 3617xs with mpt3sas drivers to support my raid card. Underneath its an 8700k CPU with 64GB RAM. Would the SA6400 be an appropriate upgrade pathway? Is there anything special required to enable fast transcoding for the CPU? I'm keen to upgrade as I'm still on an older version of your loader "0.9.5.0".

Grazie delle risposte. Io prima avevo un NAS "originale" Synology DS213J con 1 GB di ram e sta ora iniziando a perdere un pò i colpi con DSM7, pensavo con 4 GB di avere un miglioramento, visto che solitamente i NAS non hanno grosse ram installate. Il Terramaster F4-423 ha queste caratteristiche: Processore Intel® Celeron® N5105/5095 Architettura X.86 64-bit Frequenza Quad Core 2.0 GHz (Max burst up to 2.9) Hardware Transcoding Engine H.264, H.265, MPEG-4, VC-1; Massima risoluzione: 4K (4096 x 2160); FPS: 60 Memoria preinstallata 4GB (1 x 4GB) Disponibili 2 slot DDR2 SODIMM con massima memoria 32GB (16GB+16GB) Slot 4 dischi Allora Arc è una loader custom Redpill per DSM 7.x da quel che ho capito, non so se sia la soluzione migliore o meno, ma vedo che è in continuo aggiornamento già dalla versione scritta sono passate 2 minor release, ora siamo alla arc-23.12.11 (non so se questo sia un bene o un male). In teoria io utilizzo il NAS come contenitore foto e video, server multimediale e volevo provare ad utilizzare docker, che non era supportato dal mio vecchio NAS.

you're right. Synology's original specifications for SA6400 are based on AMD Epyc CPU. iGPU transcoding is only available on Intel CPUs. You can see it as hacked to enable the use of Intel CPU in SA6400 and then transcode to Intel iGPU. SA6400, the only model in XPE that supports kernel 5, is considered advantageous and suitable for new hardware of the 11th generation or higher. There are continued reports of cases where 12th and 13th generation MOBOs, which did not work on XPE kernel 4 models, work on SA6400.

Well it's definitely been mentioned here by several people, that the DSM Kernel is hardcoded for the numbers of supported cores/threads, based on model. Of course, maybe that is just. visual thing, in the Info Center, maybe in reality, it uses all there is available, this, I can't remember for sure. I'm on a 920+ which is also build on a 6 Cores / 12 Threads CPU. The 920+ officially is stated to support 4 Cores / 4 Threads, per the link. So if I am remembering the hardcoded support issue right, then we'd like both benefit to the SA6400. Interestingly, my Info Center is showing 6 Cores / 8 Threads. 🤔 I too also already have perfectly running HW Transcoding. Assuming there are no draw-backs from this model. 🤠

I don't think the model limits you to the actual number of cpus that be used by the system. Sure dsm might display all the CPU/cores but does that mean the rest are not being used? My 918 is built on a 6core, 12 core CPU. TOP only shows me number of cpus to be 8. I already have HW transcoding. So wondering if the move to SA6400 would be beneficial

I think I can see, aside from the great news @Peter Suh regarding the "Fast Transcoding" which at the moment, I assume to be HW, even though the system is built with an AMD by default, is that it's kernel should support 12 Cores and 24 Threads. So depending on your chosen hardware platform to build your NAS, your 918+ might support less than what processor is actually capable of. 🤞🏼 It also supports up to 32GB of ECC (don't know if required) RAM. https://kb.synology.com/en-nz/DSM/tutorial/What_kind_of_CPU_does_my_NAS_have

Sorry i was incorrect maybe. Is it possible to start P400 for transcoding in Emby/Plex? AI in SS is not inteded (it requires 4-9 gen Intel CPU). Thanks.

Do you use a translator? I have a hard time understanding your German-English translation. I don't even know the meaning of the German in the capture. The recently added SA6400 has many better features than DS3622xs+. It supports fast transcoding in the 12th generation, is device-tree based with the advantage of easy device mapping, and even supports HBA. The only issue is that I226 is not yet stable.

Please try my mshell. We recommend the new model SA6400, which uses kernel 5. Fast transcoding is also possible in Haswell. https://github.com/PeterSuh-Q3/tinycore-redpill/releases/tag/v1.0.0.0

Yes, that's right. Upgrading from kernel 4 to kernel 5 affects performance. It is also related to improving transcoding performance. -------- Oui c'est vrai. La mise à niveau du noyau 4 vers le noyau 5 affecte les performances. Cela est également lié à l’amélioration des performances de transcodage.

Great, at least now you can use hardware transcoding for Plex / Emby / Jellyfin If you are going to use h/w transcoding, I'd advise using the patch for nVidia drivers which allows more than 3 simultaneous NVENC video encoding sessions , which is a driver limitation imposed by nVidia on their consumer GPUs. https://github.com/keylase/nvidia-patch Clone the repo, run the patch, then restart the custom nVidia driver and you're good to go.