@kas1e New test - Mesa 1.1 QEMU Pegasos2 PCI passthrough Radeon HD Interesting result: GLMark2 Score: 954 Is this result better than the Sam460 with a Radeon RX? The Pegasos2 QEMu machine is the same one used in this test: https://www.amigans.net/modules/newbb/ ... id=161731#forumpost161731
Here’s the full test: Quote:
======================================================= glmark2 2023.01 ======================================================= OpenGL Information GL_VENDOR: AMD GL_RENDERER: VERDE (radeonsi, verde, ACO, RM 1.0, 54.57 (29.7.2023) Pegasos 2 release) GL_VERSION: 4.3 (Compatibility Profile) Mesa 26.1.5 (git-2e7650ee6e) System Info: mesa.library 1.1 (1.9.2026), RadeonHD 3.7 Surface Config: buf=32 r=8 g=8 b=8 a=8 depth=24 stencil=0 samples=0 Surface Size: 800x600 windowed ======================================================= [build] use-vbo=false: FPS: 54 FrameTime: 18.633 ms [build] use-vbo=true: FPS: 1276 FrameTime: 0.784 ms [texture] texture-filter=nearest: FPS: 1156 FrameTime: 0.865 ms [texture] texture-filter=linear: FPS: 1161 FrameTime: 0.862 ms [texture] texture-filter=mipmap: FPS: 1169 FrameTime: 0.856 ms [shading] shading=gouraud: FPS: 1296 FrameTime: 0.772 ms [shading] shading=blinn-phong-inf: FPS: 1312 FrameTime: 0.762 ms [shading] shading=phong: FPS: 1297 FrameTime: 0.771 ms [shading] shading=cel: FPS: 1315 FrameTime: 0.761 ms [bump] bump-render=high-poly: FPS: 1284 FrameTime: 0.779 ms [bump] bump-render=normals: FPS: 1251 FrameTime: 0.800 ms [bump] bump-render=height: FPS: 1245 FrameTime: 0.804 ms [effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 1590 FrameTime: 0.629 ms [effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 1576 FrameTime: 0.635 ms [pulsar] light=false:quads=5:texture=false: FPS: 791 FrameTime: 1.265 ms [desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 307 FrameTime: 3.266 ms [desktop] effect=shadow:windows=4: FPS: 261 FrameTime: 3.841 ms [buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 39 FrameTime: 25.780 ms [buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 26 FrameTime: 38.500 ms [buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 43 FrameTime: 23.454 ms [ideas] speed=duration: FPS: 119 FrameTime: 8.404 ms [jellyfish] : FPS: 848 FrameTime: 1.180 ms [terrain] : FPS: 157 FrameTime: 6.409 ms [shadow] : FPS: 608 FrameTime: 1.646 ms [refract] : FPS: 400 FrameTime: 2.504 ms [conditionals] fragment-steps=0:vertex-steps=0: FPS: 1379 FrameTime: 0.726 ms [conditionals] fragment-steps=5:vertex-steps=0: FPS: 1375 FrameTime: 0.727 ms [conditionals] fragment-steps=0:vertex-steps=5: FPS: 1387 FrameTime: 0.721 ms [function] fragment-complexity=low:fragment-steps=5: FPS: 1385 FrameTime: 0.723 ms [function] fragment-complexity=medium:fragment-steps=5: FPS: 1363 FrameTime: 0.734 ms [loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 1342 FrameTime: 0.746 ms [loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 1339 FrameTime: 0.747 ms [loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 1366 FrameTime: 0.732 ms ======================================================= glmark2 Score: 954 =======================================================
What have you been doing with GART optimization? Does this have anything to do with that result?
------------------------------------------------------------ FixedGL - Fixed-Function OpenGL Benchmark ------------------------------------------------------------ GL Vendor : AMD GL Renderer : POLARIS11 (radeonsi, polaris11, ACO, RM 3.0, 54.57 (29.7.2023) Sam460 release) GL Version : 4.6 (Compatibility Profile) Mesa 26.1.5 (git-2e7650ee6e) System Info : mesa.library 1.1 (1.9.2026), RadeonRX 2.12 Surface : 640x480 windowed Workloads : 29 (6 seconds each) ------------------------------------------------------------ [batched-triangles] vertices=3072:calls=6146: FPS: 359.24 FrameTime: 2.784 ms VertexRate: 1.10 M/s CallRate: 2.21 M/s [split-triangles] vertices=3072:calls=8192: FPS: 112.54 FrameTime: 8.885 ms VertexRate: 0.35 M/s CallRate: 0.92 M/s [legacy-quads] vertices=2048:calls=4098: FPS: 448.13 FrameTime: 2.231 ms VertexRate: 0.92 M/s CallRate: 1.84 M/s [attribute-rich-quads] vertices=2048:calls=8194: FPS: 303.85 FrameTime: 3.291 ms VertexRate: 0.62 M/s CallRate: 2.49 M/s [matrix-object-quads] vertices=2048:calls=6144: FPS: 23.78 FrameTime: 42.044 ms VertexRate: 0.05 M/s CallRate: 0.15 M/s [medium-batch-triangles] vertices=3072:calls=6208: FPS: 336.45 FrameTime: 2.972 ms VertexRate: 1.03 M/s CallRate: 2.09 M/s [points] vertices=2048:calls=4098: FPS: 451.50 FrameTime: 2.215 ms VertexRate: 0.92 M/s CallRate: 1.85 M/s [lines] vertices=2048:calls=4098: FPS: 451.32 FrameTime: 2.216 ms VertexRate: 0.92 M/s CallRate: 1.85 M/s [line-strip] vertices=528:calls=1088: FPS: 723.36 FrameTime: 1.382 ms VertexRate: 0.38 M/s CallRate: 0.79 M/s [line-loop] vertices=2048:calls=5120: FPS: 176.10 FrameTime: 5.679 ms VertexRate: 0.36 M/s CallRate: 0.90 M/s [triangle-strip] vertices=1056:calls=2144: FPS: 573.29 FrameTime: 1.744 ms VertexRate: 0.61 M/s CallRate: 1.23 M/s [triangle-fan] vertices=3072:calls=6144: FPS: 163.70 FrameTime: 6.109 ms VertexRate: 0.50 M/s CallRate: 1.01 M/s [quad-strip] vertices=1056:calls=2144: FPS: 573.46 FrameTime: 1.744 ms VertexRate: 0.61 M/s CallRate: 1.23 M/s [polygons] vertices=2048:calls=5120: FPS: 179.30 FrameTime: 5.577 ms VertexRate: 0.37 M/s CallRate: 0.92 M/s [texture-only-quads] vertices=2048:calls=6146: FPS: 367.87 FrameTime: 2.718 ms VertexRate: 0.75 M/s CallRate: 2.26 M/s [lighting-only-quads] vertices=2048:calls=6146: FPS: 360.75 FrameTime: 2.772 ms VertexRate: 0.74 M/s CallRate: 2.22 M/s [vector-entrypoint-rich-quads] vertices=2048:calls=8194: FPS: 317.79 FrameTime: 3.147 ms VertexRate: 0.65 M/s CallRate: 2.60 M/s [multitexture-lightmap-quads] vertices=2048:calls=10242: FPS: 259.51 FrameTime: 3.853 ms VertexRate: 0.53 M/s CallRate: 2.66 M/s [fog-secondary-texture-quads] vertices=2048:calls=10242: FPS: 72.81 FrameTime: 13.734 ms VertexRate: 0.15 M/s CallRate: 0.75 M/s [per-vertex-color-material-quads] vertices=2048:calls=6146: FPS: 394.98 FrameTime: 2.532 ms VertexRate: 0.81 M/s CallRate: 2.43 M/s [array-element-rich-quads] vertices=2048:calls=2050: FPS: 68.81 FrameTime: 14.533 ms VertexRate: 0.14 M/s CallRate: 0.14 M/s [non-begin-end-draw-arrays-rich-triangles] vertices=3072:calls=1: FPS: 364.84 FrameTime: 2.741 ms VertexRate: 1.12 M/s CallRate: 0.00 M/s [non-begin-end-draw-elements-rich-triangles] vertices=3072:calls=1: FPS: 314.54 FrameTime: 3.179 ms VertexRate: 0.97 M/s CallRate: 0.00 M/s [large-display-list-rich-quads] vertices=2048:calls=1: FPS: 1073.28 FrameTime: 0.932 ms VertexRate: 2.20 M/s CallRate: 0.00 M/s [small-display-lists-rich-quads] vertices=2048:calls=512: FPS: 30.04 FrameTime: 33.290 ms VertexRate: 0.06 M/s CallRate: 0.02 M/s [mixed-primitive-surfaces] vertices=2254:calls=5532: FPS: 165.61 FrameTime: 6.038 ms VertexRate: 0.37 M/s CallRate: 0.92 M/s [static-vbo-draw-arrays-rich-triangles] vertices=3072:calls=1: FPS: 1164.41 FrameTime: 0.859 ms VertexRate: 3.58 M/s CallRate: 0.00 M/s [static-vbo-ebo-draw-elements-rich-triangles] vertices=3072:calls=1: FPS: 1142.42 FrameTime: 0.875 ms VertexRate: 3.51 M/s CallRate: 0.00 M/s [dynamic-vbo-subdata-ebo-rich-triangles] vertices=3072:calls=2: FPS: 545.22 FrameTime: 1.834 ms VertexRate: 1.67 M/s CallRate: 0.00 M/s ------------------------------------------------------------ FixedGL Score: 397 ------------------------------------------------------------
======================================================= glmark2 2023.01 ======================================================= OpenGL Information GL_VENDOR: AMD GL_RENDERER: POLARIS11 (radeonsi, polaris11, ACO, RM 3.0, 54.57 (29.7.2023) Sam460 release) GL_VERSION: 4.6 (Compatibility Profile) Mesa 26.1.5 (git-2e7650ee6e) System Info: mesa.library 1.1 (1.9.2026), RadeonRX 2.12 Surface Config: buf=32 r=8 g=8 b=8 a=8 depth=24 stencil=0 samples=0 Surface Size: 800x600 windowed ======================================================= [build] use-vbo=false: FPS: 151 FrameTime: 6.664 ms [build] use-vbo=true: FPS: 872 FrameTime: 1.147 ms [texture] texture-filter=nearest: FPS: 839 FrameTime: 1.193 ms [texture] texture-filter=linear: FPS: 840 FrameTime: 1.192 ms [texture] texture-filter=mipmap: FPS: 828 FrameTime: 1.208 ms [shading] shading=gouraud: FPS: 870 FrameTime: 1.150 ms [shading] shading=blinn-phong-inf: FPS: 871 FrameTime: 1.149 ms [shading] shading=phong: FPS: 865 FrameTime: 1.156 ms [shading] shading=cel: FPS: 867 FrameTime: 1.154 ms [bump] bump-render=high-poly: FPS: 867 FrameTime: 1.154 ms [bump] bump-render=normals: FPS: 836 FrameTime: 1.197 ms [bump] bump-render=height: FPS: 826 FrameTime: 1.211 ms [effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 987 FrameTime: 1.014 ms [effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 986 FrameTime: 1.015 ms [pulsar] light=false:quads=5:texture=false: FPS: 665 FrameTime: 1.504 ms [desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 201 FrameTime: 4.980 ms [desktop] effect=shadow:windows=4: FPS: 204 FrameTime: 4.914 ms [buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 35 FrameTime: 28.850 ms [buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 23 FrameTime: 43.970 ms [buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 54 FrameTime: 18.800 ms [ideas] speed=duration: FPS: 100 FrameTime: 10.085 ms [jellyfish] <default>: FPS: 563 FrameTime: 1.777 ms [terrain] <default>: FPS: 100 FrameTime: 10.046 ms [shadow] <default>: FPS: 447 FrameTime: 2.238 ms [refract] <default>: FPS: 299 FrameTime: 3.349 ms [conditionals] fragment-steps=0:vertex-steps=0: FPS: 912 FrameTime: 1.098 ms [conditionals] fragment-steps=5:vertex-steps=0: FPS: 904 FrameTime: 1.107 ms [conditionals] fragment-steps=0:vertex-steps=5: FPS: 912 FrameTime: 1.097 ms [function] fragment-complexity=low:fragment-steps=5: FPS: 903 FrameTime: 1.108 ms [function] fragment-complexity=medium:fragment-steps=5: FPS: 911 FrameTime: 1.098 ms [loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 903 FrameTime: 1.109 ms [loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 906 FrameTime: 1.105 ms [loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 896 FrameTime: 1.117 ms ======================================================= glmark2 Score: 648 =======================================================
@Kyle That not lower performance, that noise : you can try to power off, and repeat, results may differs a bit always. Benchmarks are mostly to see general things if it start to me MUCH worse, or MUCH better, like was 3100, and now 1500, or was 3100, and not 5100, then those things to consider, when it 3100 vs 2900, or vs 3300, its all the same mostly, don't worry about small differences.