summaryrefslogtreecommitdiffstats
path: root/SampleClips
diff options
context:
space:
mode:
authorKonstantin <const@const.me>2023-01-20 15:22:25 +0100
committerKonstantin <const@const.me>2023-01-20 15:22:25 +0100
commita052894a31128ef62e53ae34efd3485c2bbd8569 (patch)
treef52931c14669c4abf62387616cd242f8eb5d7d6f /SampleClips
parent062d01a9701a11468b1a46bd11f9349bb98671a6 (diff)
Readme
Diffstat (limited to 'SampleClips')
-rw-r--r--SampleClips/Readme.txt6
1 files changed, 5 insertions, 1 deletions
diff --git a/SampleClips/Readme.txt b/SampleClips/Readme.txt
index 0945728..44d50db 100644
--- a/SampleClips/Readme.txt
+++ b/SampleClips/Readme.txt
@@ -9,6 +9,8 @@ The rest of the text files in this folder are the outputs of the in-app performa
The “1080ti” files are from my desktop, which has nVidia GeForce 1080Ti GPU.
+The “vega8” files are from the same desktop, when using the GPU integrated into AMD Ryzen 7 5700G processor.
+
The “vega7” files are from my laptop, the GPU is integrated into AMD Ryzen 5 5600U processor.
The laptop model is HP ProBook 445 G8. While running the tests, the laptop was on battery power.
@@ -18,6 +20,8 @@ The file names with “medium” in the middle were made with “ggml-medium.bin
In theory, 1080ti delivers 10.6 TFlops FP32 and 484.4 GB/second VRAM bandwidth.
+The AMD APU in that desktop delivers 2.0 TFlops FP32, and 53.3 GB/second memory bandwidth.
+
That variant of 1650 delivers 2.6 TFlops FP32, and 128.1 GB/second VRAM bandwidth.
-The AMD APU in that laptop delivers 1.6 TFlops FP32, and 51.2 GB/second memory bandwidth. \ No newline at end of file
+The AMD APU in the laptop delivers 1.6 TFlops FP32, and 51.2 GB/second memory bandwidth. \ No newline at end of file