Radeon glmark2 scores

Below are to be found some Radeon scores from the glmark2 benchmark. Version 2014.03 of the benchmark was used, compiled on OpenSuSE 13.1 (gcc 4.8), and run on Ubuntu 16.04 LTS (kernel 4.4.0, Xorg 1.18.3, module version 7.7.0 of free radeon driver, Mesa 11.2.0, DRI2, LLVM 3.8). The default window size of 800x600 was used on a 1920x1080 display, and the window manager was fvwm2 (i.e. free of compositing effects). The machines were otherwise idle.

All of the cards were passively cooled, and were purchased for under £50 (some second-hand).

Scores are in frames per second (FPS).

Some of the scores are repeated, using the same binary, but running under Ubuntu 18.04LTS (kernel 4.14.0, Xorg 1.19.6, Mesa 18.0.5, DRI3, LLVM 6.0). For the Terascale cards, these show scores almost a factor of two faster, showing how much of a difference software can make even for a rather mature hardware product. The column headed 18.04 gives these scores.

So far tests with Ubuntu 20.04LTS (kernel 5.40, Xorg 1.20.8, DRI3, Mesa 20.08, LLVM 10.0.0) have shown little difference on Terascale cards, but the performance of the GCN 1 cards is back to 16.04LTS levels, the 15% regression of 18.04LTS having disappeared.

CPU	Bus	GPU	class	cores	MHz	VRAM	score	score	score
							16.04	18.04	20.04
Kaby Lake QC 3.7GHz	PCIe v3 x8	GT 1030	Pascal	384	1500	64 bit GDDR5/6000		7192³
Haswell QC 3.1GHz	PCIe v3 x8	RX550⁴	GCN 4	512	1206	128 bit GDDR5/6000		6682
Haswell QC 3.1GHz	PCIe v3 x8	R7 250E	GCN 1	512	800	128 bit GDDR5/4500	4703	4032	4689
Ivy Bridge QC 3.3GHz	PCIe v3 x8	HD 7750	GCN 1	512	800	128 bit GDDR5/4500	4610	3984
Haswell QC 3.1GHz	PCIe v3 x8	R7 240	GCN 1	320(?)	780	128 bit DDR3/1600		2050
Ivy Bridge QC 3.3GHz	PCIe v2 x8	HD 6670	T'scale 2	480	800	128 bit GDDR5/4000	1757
Ivy Bridge QC 3.3GHz	PCIe v2 x8	HD 5670	T'scale 2	400	775	128 bit GDDR5/4000	1719
Ryzen 5 QC 3.2GHz	PCIe v2 x16	HD 5670	T'scale 2	400	775	128 bit GDDR5/4000	1494	2975
Haswell QC 3.1GHz	PCIe v2 x8	HD 6570	T'scale 2	480	650	128 bit DDR3/1333	968	1729
Core2 QC 2.4GHz	PCIe v2 x16	HD 5570	T'scale 2	400	650	128 bit DDR3/1000	634	1371
Haswell QC 3.1GHz	PCIe v2 x8	R5 230 Flex	T'scale 2	160	625	64 bit DDR3/1600	542	940
Haswell QC 3.1GHz	PCIe v2 x8	HD 6450	T'scale 2	160	625	64 bit DDR3/1333	483
Core2 QC 2.4GHz	PCIe v2 x16	HD 5450	T'scale 2	80	650	64 bit DDR3/1333	388
Haswell QC 3.1GHz	PCIe v2 x8	HD 5450	T'scale 2	80	650	64 bit DDR3/1333	378
Sandy Bridge QC 3.2GHz	PCIe v2 x8	HD 5450	T'scale 2	80	650	64 bit DDR3/1333	378	802
Kaby Lake QC 3.5GHz	PCIe v2 x8	R5 230	T'scale 2	160	625	64 bit DDR3/1066	377
Core2 DC 2.4GHz	PCIe x16	X550	RV370	4	400?	128 bit DDR/500?	298^1,2
Athlon II DC 3.0GHz		RS880/HD 4250	T'scale 1	40	500	Hyper & sideport	275
Core2 QC 2.83GHz	PCIe x16	HD 3450	T'scale 1	40	600	64 bit DDR2/800	250
Core2 QC 2.4GHz	PCIe x16	HD 4350	T'scale 1	80	600	64 bit DDR2/800	246	488
Core2 DC 2.4GHz	PCIe x16	X1550	RV500	4	550?	128 bit DDR2/800?	141¹	141¹
Athlon II DC 3.0GHz		RS880/HD 4250	T'scale 1	40	500	Hyper & sideport	117
Core2 DC 2.4GHz	PCIe x16	X300SE	RV370	4	325?	64 bit DDR/400?	84^1,2

T'scale: Terascale
GCN: Graphics Core Next

Memory speeds given in MT/s. Memory clock speed is one half this (one quarter for GDDR5).

¹ does not support GL version 3.0 and unable to run the Terrain benchmark from this suite. Score is average of those it could run, which will flatter the result as the Terrain benchmark is generally the slowest.

² failed to complete four other benchmarks correctly, substituting a dummy shader, and failed to run the Jellyfish correctly. Again, this flatters the results as the dummy shader is less work than a proper shader.

³ using nvidia's binary drivers. With the open-source noveau drives, this scores under 300 on 18.04. (This passive card is both newer and more expensive than the Radeon ones.)

⁴ the card used was actively cooled, but it is quite low power and a passively-cooled version would be technically possible. This card used the AMDGPU driver, not the radeon driver.

The performance of the RS880 may depend on the exact memory configuration (no sideport, exclusively sideport, sideport and hypermemory interleaved in different ways). The precise configuration of the two machines above is not known.

One result for Ubuntu 20.04 (kernel 5.4.0, Xorg 1.20.8, Mesa 20.0.8, LLVM 10.0.0), HD6570 with DDR3/1333, Haswell QC 3.1 GHz, 1710 FPS, and 1680 FPS with a 2560x1600 monitor. Thus it seems that there has been little change for Terascale 2 between Ubuntu 18.04 and 20.04, and even with the benchmark running in the same sized window, the extra memory bandwidth consumed by the signal generator at the higher resolution makes a small, but measurable, difference.

Comments

The benchmark does seem to be testing things which are entirely offloaded to the graphics card, in that very significant changes in CPU power have no impact.

The benchmark seems to be heavily influenced by GPU memory speed, particularly when running Ubuntu 16.04. The HD 5670 would be expected to be about 20% faster than the HD 5570 if the constraint was GPU processing power. It is over 2.7 times faster, presumably reflecting a memory subsystem four times as fast. Similarly the only difference between the R5 230 and the HD 6350 is that the memory is 20% faster on the former, leading to a score over 12% higher. Conversely the HD6670 has much better GPU performance than the HD5670 (20% more cores and 3.2% higher clock speed), yet the same memory, and scores just 2.2% better. One should be careful when purchasing cards: I purchased the HD 5570 believing it to have DDR3/1333 memory, which would probably have increased its score by about a hundred FPS if it were true.

The GCN-based card seems to be able to make much better use of the available memory bandwidth.

The Terascale and GCN-based cards switch between different power states, changing GPU and memory speed, and GPU voltage. The Terascale-based cards seem incapable of transitioning from their high-power state whilst a 2560x1440 monitor is active, save for the XFX HD5670 (the other cards were mostly Sapphire).

Upgrading from Mesa 11.2.0 (Ubuntu 16.04) to 12.0.6 made no significant difference to the above scores. The Haswell with the HD 5450 achieved an identical score, for instance.

Comments on the different DVI, HDMI and DisplayPort standards.

Video Decoding

The above Radeon cards all include some version of AMD's "Unified Video Decoder" (UVD). This may sit on the same die as the GPU, but it is separately clocked, and is mostly independent in generation from the rest of the GPU. Firefox and Chromium are notorious for not using it, but mplayer and VLC can. UVD 2 can decode two H264 1080p streams simultaneously, and is found on the HD4000 series and later. Only very recent cards (and none of the above) can decode H264 at 4K, H265 or VP9.

The line in /sys/kernel/debug/dri/0/radeon_pm_info reading

uvd    vclk: 0 dclk: 0

refers to the current clock rates of the UVD. Here, quiescent. The program vdpauinfo will describe the capabilities of the UVD.