Performance degradation after the new update

[spaghetti-]

This may be anecdotal, but after yesterdays 400MB update that added compute shaders, Dota 2 went from 60fps stable @ 1440p on my Radeon Pro 560 4096 MB to struggling at 30-40 fps. I had to lower the graphics quality one notch (in the automatic settings slider from high to medium) to make up for this.

Anyone else noticed the same?

System information:

Computer Information:
    Manufacturer:  Apple
    Model:  MacBookPro14,3
    Form Factor: Laptop
    No Touch Input Detected

Processor Information:
    CPU Vendor:  GenuineIntel
    CPU Brand:  Intel(R) Core(TM) i7-7920HQ CPU @ 3.10GHz
    CPU Family:  0x6
    CPU Model:  0x9e
    CPU Stepping:  0x9
    CPU Type:  0x0
    Speed:  3100 Mhz
    8 logical processors
    4 physical processors
    HyperThreading:  Supported
    FCMOV:  Supported
    SSE2:  Supported
    SSE3:  Supported
    SSSE3:  Supported
    SSE4a:  Unsupported
    SSE41:  Supported
    SSE42:  Supported
    AES:  Supported
    AVX:  Supported
    CMPXCHG16B:  Supported
    LAHF/SAHF:  Supported
    PrefetchW:  Unsupported

Operating System Version:
    MacOS 10.13.5 (64 bit)

Video Card:
    Driver:  Radeon Pro 560
    Driver Version:  1.2
    Driver Date: 0 0 0
    OpenGL Version: 2.1
    Desktop Color Depth: 32 bits per pixel
    Monitor Refresh Rate: 59 Hz
    VendorID:  0x1002
    DeviceID:  0x67ef
    Revision:  0xc0
    Number of Monitors:  2
    Number of Logical Video Cards:  2
    Primary Display Resolution:  2560 x 1440
    Desktop Resolution: 5120 x 2880
    Primary Display Size: 23.46" x 13.19" (26.89" diag)
                                            59.6cm x 33.5cm (68.3cm diag)
    Primary VRAM: 0 MB

Memory:
    RAM:  16384 Mb

Miscellaneous:
    UI Language:  English
    Media Type:  DVD
    Total Hard Disk Space Available:  476802 Mb
    Largest Free Hard Disk Block:  331040 Mb
    VR Headset: None detected

[danginsburg]

I measured a performance gain from compute shaders on my AMD FirePro D500. Does toggling Compute Shaders have any impact on perf (keeping the settings at high)? A performance drop would be unexpected.

[jeanlain]

As a comparison, I'm seeing a ~15% performance boost with compute shaders on my iMac radeon 580, macOS 10.3.5. With shaders disabled, I think performance is roughly the same as before the update.

[maximea1r]

Im actually experiencing the same thing. Playing a full match with compute shaders on would result in the fps tanking to about 20 to 30 fps on my case.

[olddonkey]

I experiencing the same thing.

[tsaber]

they work fine on manjaro linux intel iris crystallwell C220 Chipset until minute 27 of a match. Then stuttering with still max fps

[spaghetti-]

@danginsburg How can I get you quantitative analysis of this problem? Do you have a tool that can profile fps/tick/replay? I can play a game each with both on and get you the replay, but I don't think that information is tracked in the replay files.

I can also test it for you on my desktop which is running Gentoo (4.17.6, and I can emerge vulkan support to test) but I don't think it will be noticeable because its a 1080GTX.