Jump to content
XPEnology Community
  • 0

For DSM 6.2.3 upgrade CPU to i7-5775C (65W) OR E3-1265L v4 (35W)


robert36

Question

I im having Gigabyte GA-H97N-WIFI mini-itx board and i5-4460 (LGA1150) that is 84W of TDP so i would like to upgrade to 4c/8threads CPU and get lower TDP and temperature inside my case. On website supported CPU list is here:

 

https://www.gigabyte.com/Motherboard/GA-H97N-WIFI-rev-10/support#support-cpu

 

And i get two very interesting CPU and comparation table:

   

CPU                Core i7-5775C   |   Xeon E3-1265L V4   

------------------------------------------------------

TDP                        65W            |              35W

CLOCKSPEED       3.3GHz        |              2.3GHz

SINGLE TREAD     2312            |               1977

CPU MARK           7843            |               6493

MULTIPLIER           33              |                  23

UNLOCK MULT.     YES            |                 NO

IGPU                    IRIS P6200   |                 IRIS P6300

------------------------------------------------------

 

I need CPU to:

- serve 10-15 file readings from HDD (single file size around 2,0-2,5GB)

- have to use HW Transcoder in DS VideoStation or Plex

- have to be LOW power hungry (running 24/7 so to make noise and bill at minimum)

- have to be faster that i5 that i have

 

So what is you opinion what CPU to buy? On ebay i see that both cpu are same price so i im thinking to buy i7-5775C that is 65W and lower multiplier from 33 to 23 to get lower TDP and temperatures...but what i don't know how many performance i will lose? Or is better and faster to go to XEON and at base speed od 2.3GHz will be faster and cooler that i7 running at same speed of 2.3GHz (lowering 33x to 23x)?

 

In my opinion that i im not sure calculating from benchmark result:

 

i7-5775C        takes 7843 rating / 3300MHz => 2.376 rating per MHz

E3-1265L V4  takes 6493 rating / 2300MHz => 2.823 rating per MHz

 

On i7-5775C running @ 2300MHz(23x multiplier) x 2.376 rating we will have 5466 points that is slower that 6493 rating for E3-1265L V4...so in theory XEON is faster and cooler. Is this true or not?

 

And second question is Intel Iris Pro 6200 or 6300 supported in VideoStation or Plex to support HW Transcoding and how much x264 transcoding on the fly i can use on first or second CPU? I read that for one x264 SW transcoding you need 2000 benchmark points so both CPU when using SW x264 transcoding on the fly will run 3 streams? (3streams x 2000points is 6000points minimum to use)...so this why i can't decide what CPU to buy as upgrade...i chose these two because MBO supports it and it is 14nm (lower temperatures and current), and have powerfull IGPU that i need to know if it will work in plex and diskstation as HW transcoder...main target is only x264 files.

 

Thanks.

 

 

Link to comment
Share on other sites

12 answers to this question

Recommended Posts

  • 0

does not look like there is much difference between these cpu's, even when it comes to the gpu its the same  (both GT3e)

https://en.wikipedia.org/wiki/Intel_Graphics_Technology#Broadwell

 

should be supported with the i915 of 918+ dsm 6.2.3

i listed all the supported gpu pci device id's here

https://xpenology.com/forum/topic/28321-driver-extension-jun-103b104b-for-dsm623-for-918-3615xs-3617xs/

click on "iGPU device ID's supported by synology's i915 driver" to open the list

the i7 device 0x1622 is in that list and as the xeaon is also GT3e i would expect it to be supported too

 

you might get the same tdp when you underclock the i7, the i7's official tdp-dowm mode @2.2GHz is 37W, the thin you would need to find out is if that lower core clock has influence on the gpu

https://ark.intel.com/content/www/us/en/ark/products/88040/intel-core-i7-5775c-processor-6m-cache-up-to-3-70-ghz.html

 

 

when it comes to features of video decoding the 4th and 5th gen cpu's are not that different

https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video#Hardware_decoding_and_encoding

 

 

Link to comment
Share on other sites

  • 0

OP, if you are going to move to hardware transcoding, why does it need to be faster than your current i5?

 

And your current Haswell chip is not using 84W TDP unless it is maxing out its workload. For all workloads other than transcoding, monitor your current chip (htop is useful for this, or just use Resource Manager).  If you are like most people, you cannot max out 4 cores with any normal system usage.  So performance is not much of a differentiator.

 

So it sounds like you want a new chip so that you can hardware transcode - that's fine.  I otherwise agree with IG-88, there isn't much of a difference when you can tune the clock to your power taste.  Xeon is only useful if you can use ECC RAM which your mobo does not support.  Just buy the least expensive option available.

Link to comment
Share on other sites

  • 0

Thanks..that info i need to know...so using XEON is waste of money because motherboard does not support ECC RAM...my goal using XEON was to fix restart issues on DSM6.2.3 and to get lower TDP (35W)...but eather MBO does not support ECC (so restart won't work), nor iGPU in XEON is same as in i7 (it is different name but performance is same).

 

So you suggest me to buy i7-5775C, lower CPU to 2.2GHz and get nice 37W TDP (like xeon), it supports non ECC RAM that my MBO have (2x8GB DDR3 @ 1600MHz CL9), and iGPU device id: 8086:1622 is supported in DSM918+ model so HW transcoding should work.

 

I have watching this CPU to buy (and using picture i see that this is not engenering sample, it is real stuf, pulled from some server), and price is a bit high but it is the lastest i7 cpu that supports LGA1150 with most powerfull Intel Iris Pro 6200 that have 128MB of L4 cahce to speed transcoding a liitle more.

 

https://www.ebay.com/itm/Intel-Core-i7-5775C-3-3GHz-LGA1150-Graphics-6200-CPU-Processors/174444989228

 

Using benchmark on what i have difference will be on high load 84W vs 37W (idle temp will be the same, but load temp i expect to be around 52-55C on i7, now i have on full load around 68-70C what is a high to me, but i have 2pcs of WD RED 10TB and push pull vents in case so i would like to lower the power draw and lower a CPU temp).

 

I see when i download DS3617xs DSM 6.2.3.pat that in description it says Broadwell so can this be used or i must use DS918+? (i want to use plex HW transcoding and my mobo have 6xSATAIII connectors so i want to use all 6 SATA connectors when time arrive for to buy more HDD).

 

Thanks and this i all i need to know.

Link to comment
Share on other sites

  • 0
1 hour ago, robert36 said:

lastest i7 cpu that supports LGA1150 with most powerfull Intel Iris Pro 6200 that have 128MB of L4 cahce to speed transcoding a liitle more.

just to make sure, you know your i5-4460 can be used with 918+ and hardware transcoding, just for this you dont need a newer cpu, 4th gen haswell is the minimum and the i5-4460 is 4th gen (the last thin in my post was about that, your 4th gen can do the job and a 5th gen might not be that much better)

maybe just make a test install on a single disk, copy e few movies to it and test transcoding (before buying a new cpu you might not need for what you want)

Link to comment
Share on other sites

  • 0

I have put x265 to x264 High Profile..and i see in Synology 39% CPU usage (this seems to high to me)...so i run htop in SSH and see that ffmpeg is using /dev/dri/renderD128 to transcode video...im im not sure if this is OK CPU usage (as i expect GPU to transcode and CPU to stay around 2-3% while it is doing only transcode in DS VideoStation)...


This is commands that i put in terminal using SSH and return...

 

ls /dev/dri
card0  renderD128

cat /usr/syno/etc/codec/activation.conf
{"success":true,"activated_codec":["hevc_dec","h264_dec","h264_enc"],"token":"5695bd565998486974465f1d690ad5a1"}

iGPU? cat /sys/class/hwmon/hwmon0/temp1_input
42000
Core0 cat /sys/class/hwmon/hwmon0/temp2_input
41000
Core1 cat /sys/class/hwmon/hwmon0/temp3_input
40000
Core2 cat /sys/class/hwmon/hwmon0/temp4_input
42000
Core3 cat /sys/class/hwmon/hwmon0/temp5_input
38000

 

 

DS918+ HD4600 transcoding.png

Link to comment
Share on other sites

  • 0

And another from SSH terminal: so last 11 items have Unknown header type so kernel driver is not loaded...how to fix that? I set only UEFI in MBO UEFI settings sto i see that i im missing some drivers..any help where to find drivers if is compiled or any tutorial how to compile it so that i can try to compile.


I im attaching dmesg where i power on xpenolgoy so that you can see if there is some errors about drivers....thanks.

 

root@Synology:~# lspci -v
0000:00:00.0 Class 0600: Device 8086:0c00 (rev 06)
        Subsystem: Device 1458:5000
        Flags: bus master, fast devsel, latency 0
        Capabilities: [e0] Vendor Specific Information: Len=0c <?>
        Kernel driver in use: hsw_uncore

0000:00:02.0 Class 0300: Device 8086:0412 (rev 06)
        Subsystem: Device 1458:d000
        Flags: bus master, fast devsel, latency 0, IRQ 29
        Memory at f7800000 (64-bit, non-prefetchable) [size=4M]
        Memory at e0000000 (64-bit, prefetchable) [size=256M]
        I/O ports at f000 [size=64]
        Expansion ROM at <unassigned> [disabled]
        Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
        Capabilities: [d0] Power Management version 2
        Capabilities: [a4] PCI Advanced Features
        Kernel driver in use: i915

0000:00:03.0 Class 0403: Device 8086:0c0c (rev 06)
        Subsystem: Device 8086:2010
        Flags: fast devsel, IRQ 11
        Memory at f7e30000 (64-bit, non-prefetchable) [disabled] [size=16K]
        Capabilities: [50] Power Management version 2
        Capabilities: [60] MSI: Enable- Count=1/1 Maskable- 64bit-
        Capabilities: [70] Express Root Complex Integrated Endpoint, MSI 00

0000:00:14.0 Class 0c03: Device 8086:8cb1 (prog-if 30)
        Subsystem: Device 1458:5007
        Flags: bus master, medium devsel, latency 0, IRQ 27
        Memory at f7e20000 (64-bit, non-prefetchable) [size=64K]
        Capabilities: [70] Power Management version 2
        Capabilities: [80] MSI: Enable+ Count=1/8 Maskable- 64bit+
        Kernel driver in use: xhci_hcd

0000:00:16.0 Class 0780: Device 8086:8cba
        Subsystem: Device 1458:1c3a
        Flags: bus master, fast devsel, latency 0, IRQ 11
        Memory at f7e39000 (64-bit, non-prefetchable) [size=16]
        Capabilities: [50] Power Management version 3
        Capabilities: [8c] MSI: Enable- Count=1/1 Maskable- 64bit+

0000:00:19.0 Class 0200: Device 8086:153b
        Subsystem: Device 1458:e000
        Flags: bus master, fast devsel, latency 0, IRQ 28
        Memory at f7e00000 (32-bit, non-prefetchable) [size=128K]
        Memory at f7e38000 (32-bit, non-prefetchable) [size=4K]
        I/O ports at f080 [size=32]
        Capabilities: [c8] Power Management version 2
        Capabilities: [d0] MSI: Enable+ Count=1/1 Maskable- 64bit+
        Capabilities: [e0] PCI Advanced Features
        Kernel driver in use: e1000e

0000:00:1a.0 Class 0c03: Device 8086:8cad (prog-if 20)
        Subsystem: Device 1458:5006
        Flags: bus master, medium devsel, latency 0, IRQ 16
        Memory at f7e37000 (32-bit, non-prefetchable) [size=1K]
        Capabilities: [50] Power Management version 2
        Capabilities: [58] Debug port: BAR=1 offset=00a0
        Capabilities: [98] PCI Advanced Features
        Kernel driver in use: ehci-pci

0000:00:1c.0 Class 0604: Device 8086:8c90 (rev d0)
        Flags: fast devsel
        Bus: primary=00, secondary=01, subordinate=01, sec-latency=0
        Capabilities: [40] Express Root Port (Slot-), MSI 00
        Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit-
        Capabilities: [90] Subsystem: Device 1458:5001
        Capabilities: [a0] Power Management version 3
        Kernel driver in use: pcieport

0000:00:1c.3 Class 0604: Device 8086:8c96 (rev d0)
        Flags: bus master, fast devsel, latency 0
        Bus: primary=00, secondary=02, subordinate=02, sec-latency=0
        I/O behind bridge: 0000e000-0000efff
        Memory behind bridge: f7d00000-f7dfffff
        Capabilities: [40] Express Root Port (Slot+), MSI 00
        Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit-
        Capabilities: [90] Subsystem: Device 1458:5001
        Capabilities: [a0] Power Management version 3
        Kernel driver in use: pcieport

0000:00:1c.4 Class 0604: Device 8086:8c98 (rev d0)
        Flags: fast devsel
        Bus: primary=00, secondary=03, subordinate=03, sec-latency=0
        Memory behind bridge: f7c00000-f7cfffff
        Capabilities: [40] Express Root Port (Slot+), MSI 00
        Capabilities: [80] MSI: Enable- Count=1/1 Maskable- 64bit-
        Capabilities: [90] Subsystem: Device 1458:5001
        Capabilities: [a0] Power Management version 3
        Kernel driver in use: pcieport

0000:00:1d.0 Class 0c03: Device 8086:8ca6 (prog-if 20)
        Subsystem: Device 1458:5006
        Flags: bus master, medium devsel, latency 0, IRQ 23
        Memory at f7e36000 (32-bit, non-prefetchable) [size=1K]
        Capabilities: [50] Power Management version 2
        Capabilities: [58] Debug port: BAR=1 offset=00a0
        Capabilities: [98] PCI Advanced Features
        Kernel driver in use: ehci-pci

0000:00:1f.0 Class 0601: Device 8086:8cc6
        Subsystem: Device 1458:5001
        Flags: bus master, medium devsel, latency 0
        Capabilities: [e0] Vendor Specific Information: Len=0c <?>
        Kernel driver in use: lpc_ich

0000:00:1f.2 Class 0106: Device 8086:8c82 (prog-if 01)
        Subsystem: Device 1458:b005
        Flags: bus master, 66MHz, medium devsel, latency 0, IRQ 26
        I/O ports at f0d0 [size=8]
        I/O ports at f0c0 [size=4]
        I/O ports at f0b0 [size=8]
        I/O ports at f0a0 [size=4]
        I/O ports at f060 [size=32]
        Memory at f7e35000 (32-bit, non-prefetchable) [size=2K]
        Capabilities: [80] MSI: Enable+ Count=1/1 Maskable- 64bit-
        Capabilities: [70] Power Management version 3
        Capabilities: [a8] SATA HBA v1.0
        Kernel driver in use: ahci

0000:00:1f.3 Class 0c05: Device 8086:8ca2
        Subsystem: Device 1458:5001
        Flags: medium devsel, IRQ 18
        Memory at f7e34000 (64-bit, non-prefetchable) [size=256]
        I/O ports at f040 [size=32]
        Kernel driver in use: i801_smbus

0000:02:00.0 Class 0200: Device 1969:1091 (rev 10)
        Subsystem: Device 1458:e000
        Flags: bus master, fast devsel, latency 0, IRQ 30
        Memory at f7d00000 (64-bit, non-prefetchable) [size=256K]
        I/O ports at e000 [size=128]
        Capabilities: [40] Power Management version 3
        Capabilities: [58] Express Endpoint, MSI 00
        Capabilities: [c0] MSI: Enable+ Count=1/16 Maskable+ 64bit+
        Capabilities: [d8] MSI-X: Enable- Count=16 Masked-
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [180] Device Serial Number ff-25-2c-cc-fc-aa-14-ff
        Kernel driver in use: alx

0000:03:00.0 Class 0280: Device 8086:08b1 (rev 73)
        Subsystem: Device 8086:4070
        Flags: fast devsel, IRQ 11
        Memory at f7c00000 (64-bit, non-prefetchable) [disabled] [size=8K]
        Capabilities: [c8] Power Management version 3
        Capabilities: [d0] MSI: Enable- Count=1/1 Maskable- 64bit+
        Capabilities: [40] Express Endpoint, MSI 00
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [140] Device Serial Number a0-a8-cd-ff-ff-d5-8c-6b
        Capabilities: [14c] Latency Tolerance Reporting
        Capabilities: [154] Vendor Specific Information: ID=cafe Rev=1 Len=014 <?>

0001:00:12.0 Class 0000: Device 8086:5ae3 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:00:13.0 Class 0000: Device 8086:5ad8 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:00:14.0 Class 0000: Device 8086:5ad6 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:00:15.0 Class 0000: Device 8086:5aa8 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:00:16.0 Class 0000: Device 8086:5aac (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:00:18.0 Class 0000: Device 8086:5abc (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:00:19.2 Class 0000: Device 8086:5ac6 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:00:1f.1 Class 0000: Device 8086:5ad4 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:01:00.0 Class 0000: Device 1b4b:9215 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:02:00.0 Class 0000: Device 8086:1539 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

0001:03:00.0 Class 0000: Device 8086:1539 (rev ff) (prog-if ff)
        !!! Unknown header type 7f

 

dmesg

Link to comment
Share on other sites

  • 0
root@Synology:~# cat /sys/kernel/debug/dri/0/i915_frequency_info
Video Turbo Mode: yes
HW control enabled: yes
SW control enabled: no
PM IER=0x00000470 IMR=0xffffff8f ISR=0x00000000 IIR=0x00000000, MASK=0x0000002a
pm_intrmsk_mbz: 0x00000004
GT_PERF_STATUS: 0x00001677
Render p-state ratio: 22
Render p-state VID: 119
Render p-state limit: 255
RPSTAT1: 0x00048b00
RPMODECTL: 0x00000d92
RPINCLIMIT: 0x000019fa
RPDECLIMIT: 0x00003a98
RPNSWREQ: 1100MHz
CAGF: 1100MHz
RP CUR UP EI: 6747 (8636us)
RP CUR UP: 4607 (5896us)
RP PREV UP: 0 (0us)
Up threshold: 85%
RP CUR DOWN EI: 13307 (17032us)
RP CUR DOWN: 4940 (6323us)
RP PREV DOWN: 0 (0us)
Down threshold: 60%
Lowest (RPN) frequency: 200MHz
Nominal (RP1) frequency: 350MHz
Max non-overclocked (RP0) frequency: 1100MHz
Max overclocked frequency: 1100MHz
Current freq: 1100 MHz
Actual freq: 1100 MHz
Idle freq: 200 MHz
Min freq: 200 MHz
Boost freq: 1100 MHz
Max freq: 1100 MHz
efficient (RPe) frequency: 350 MHz
Current CD clock frequency: 540000 kHz
Max CD clock frequency: 540000 kHz
Max pixel clock frequency: 540000 kHz

 

Link to comment
Share on other sites

  • 0

I figure out what is the problem....i try x264 1080p to play and it is using HW Transcoding (i see ffmpeg uses /dev/dri/renderD128) and CPU usage is 0% so it works. Then i try x265 1080p and CPU usage is between 34-39% and because of little research of CPU i have i5-4460 i im using Intel HD 4600 iGPU (Haswell CPU Code Name) and from Hardware Decoding And Decoding page:

 

https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video#Hardware_decoding_and_encoding

 

I see that GPU built in CPU have HW Support for MPEG-2 and AVC (x.264) so this is reason why when i play x265 i get high CPU usage because HD 4600 iGPU does not have HW decoder for x265.

 

So now i im thinking to add on mini ITX board one GPU from nVidia with low TDP and to transcode around 10 files of x265 simultaneously, i found that webpage:

 

https://www.elpamsoft.com/?p=Plex-Hardware-Transcoding

 

And would like to buy used Asus GeForce GTX 1650 4GB GDDR5 to handle 10 x265 streams simultaneously (320MB/per h265 1080p stream x 10 wanted streams gives me 3200MB used GPU memory of 4096MB memory available, and that is 78% GDDR usage so it will be stable and 4GB of GDDR5 is enough because we use only 78% of that memory). Another great thing is TDP of GPU and under full load is 75W (GTX 1060 uses 165W when connected all monitors to card, but without monitors it will consume around 80W of TDP):

 

So my calculation is what GPU synology is supporting (that we have original synology compiled nVidia drivers that card, i im using DS918+ and iGPU works, but i will disable it when nVidia GPU will work as expected so as HD4600 consumes 20W TDP at full load, and GTX 1650 or 1060 uses 80W TDP at full load, disabling HD4600 we lower system power usage by 20W, and 80W TDP - 20W we get only 60W increase of power use in out pc case, and that is OK.

 

So what gpu should i buy? I know that i need recompiled ffmpeg to support nvidia gpu encoding but maybe we can extract from synology pat nvidia drivers (kernel modules) and install nvdidia runtime, copy ffmpeg from synology that uses nvidia card (that must have compiled nvidia support) and try if it works.

Link to comment
Share on other sites

  • 0
22 hours ago, robert36 said:

o last 11 items have Unknown header type so kernel driver is not loaded...how to fix that? I

afaik these are are dummy devices made present to fool dsm about being on the original hardware, dsm checks the presence of some pci device id's to make sure its running on a synology hardware, thats part of the "magic" in jun's loader (and one of the reasons a loader is specific for a certain dsm type like 3615, 3617 or 918+)

 

5 hours ago, robert36 said:

I figure out what is the problem....i try x264 1080p to play and it is using HW Transcoding (i see ffmpeg uses /dev/dri/renderD128) and CPU usage is 0% so it works. Then i try x265 1080p and CPU usage is between 34-39% and because of little research of CPU i have i5-4460 i im using Intel HD 4600 iGPU (Haswell CPU Code Name) and from Hardware Decoding And Decoding page:

 

https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video#Hardware_decoding_and_encoding

 

 

that was the reason for the link and my comment about 5th gen cpu not having much gain over the 4th gen cpu you already use

you did not specify h265 as needed for you project or i would have marked that explicitly a  problem with your plan to use a 5th gen cpu

 

5 hours ago, robert36 said:

So now i im thinking to add on mini ITX board one GPU from nVidia with low TDP and to transcode around 10 files of x265 simultaneously

you file system (with its underlying raid)  might not be up to the task to deliver a constant 4k file stream to transcode, just having the cpu/gpu does not make it work, the system as a whole needs to be seen and i'm not sure the low iops of a raid from conventional hdd's can do this, you should read about that or ask people with more experience before assuming it will work (imho)

 

GPU Nvidia is not supported and the last experiment with that and a plex owner did not work

https://xpenology.com/forum/topic/22272-nvidia-runtime-library/

just making a kernel driver is possible but the whole driver construct around it with its libs and dependencies in a appliance with all its limitations ... its no like on a normal (maintained) linux where you can use apt and you are done

5 hours ago, robert36 said:

I know that i need recompiled ffmpeg to support nvidia gpu encoding but maybe we can extract from synology pat nvidia drivers (kernel modules) and install nvdidia runtime, copy ffmpeg from synology that uses nvidia card (that must have compiled nvidia support) and try if it works.

you can join the effort and find out whats possible (like above), i already lined out what can be tested (driver and ffmpeg) without compiling anything new

 

if you thinking about a new systemboard anyway, you could just use a gemini lake to have h265 support (or skylake or newer cpu), no need for nvidia at all with a intel qsv h265 capable cpu

 

 

Link to comment
Share on other sites

  • 0

I im very happy with my gigabyte board and i5-4460, because when i work in Diskstation CPU usage is 0-6% so CPU is more powerfull that i need.

 

My concentration is on GPU, as my MBO have 6xSATA and 2xGbE Ethernel LAN Cards integrated, i don't need to go to new version of CPU because of IGPU.

 

Question is i have pat file from DVA3221 and DVA3219 that use NVIDIA GeForce GTX 1650 and NVIDIA GeForce GTX 1050 Ti, so where can i find in pat file what device id is supported? Because i need to know before i buy new NVIDIA card...i have GTX 1050, 1060, 1070 and 1080 cards to buy (searching for better prices), so i would like to buy but what is supported in compiled synology kernel module driver if my custom compilation fails..to try original synology driver....i see that GTX 1650 is slower in transcoding x.265 that GTX 1060...this web page states:

 

https://www.elpamsoft.com/?p=Plex-Hardware-Transcoding

 

GTX 1650 can handle 14 h.265 streams and is 128-bit memory inteface (this really slows as memory conumpation is more important here around 320MB/s per x.265 transcoding....so 4GB is minimum)

 

Eather GTX 1060, 1070 or 1080 i would like to buy , it is non TI version...4 or 6GB of GDDR needed as i can hadnle around maximum of 20streams...i need only 10....for now...so to be available for feauture use. Will buy one card just to check pat file for ID that is supported by original Synology driver so that i can search that model for better price and try to find the solution about driver comiling and ffmpeg...i think when we fix that that DS Station, Jellyfin and plex will work in containter but will need to find way out how to add to containter gpu devices (like we did on intel IGPU), and patch nvdidia driver to allow mora that 2 transcodes (for that exist patch for linux so that is not a problem).

 

And i have x.265 1080p because 4k will use more internet bandwidth, so x.265 1080p...10 transcoding at same time is what i need currently to be happy many many years.

Edited by robert36
Link to comment
Share on other sites

  • 0

I have one more question about UPS...as i would like to buy UPS that when power if gone send to synology signal that does synology power off (like you do in synology browser clicking on upper right icon of your username and choose Shut Down). This way is elegant because HDD will save all activity and power off all server.

 

So what interface need to UPS have? USB is good to comunicate to synology and is there synology kernel driver or module that communicate with ups so that i can do programatically set when power is off (we are running on UPS battery), shut down synology. When power goes on after 1minute power on synology.

 

What interface i need to buy to be compatibile with synolog and what software or kernel driver i need to use...my power draw is around 45-55W

 

I have found CYBERPOWER BU650E for about 45€ (and for 60W load it lasts 30min which is very good, and is small factory...only i don't see how to replace battery when time goes on).

 

https://www.cyberpower.com/eu/en/product/sku/bu650e#specification

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Answer this question...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...