Created attachment 100261 [details] lspci -vv output After upgrading my AMD E-450-based notebook to a newer one (HP 4545s A4-3300-based) i noticed that in spite of noticeable higher clock rate the video performance is about 15-20% worse than on E-450. It also got much hotter initially, which I could cure by passing pcie_aspm=force to the kernel. In spite of this I still have a feeling that it probably eats more power than it should (which I can not prove with numbers, of course) comparing to E-450 (and fglrx) experience. (Yes, I know the proprietary driver performs better in terms of power management, too, but the question is to what extent?) Anyways, it shouldn't underperform the older and weaker model... It also shows the message at boot time: [drm:radeon_acpi_init] *ERROR* Cannot find backlight controller but it seems to work fine after that. (Shall I post another bug on this?)
Created attachment 100271 [details] Kernel config
Please attach your dmesg output.
(In reply to comment #0) > After upgrading my AMD E-450-based notebook to a newer one (HP 4545s > A4-3300-based) i noticed that in spite of noticeable higher clock rate the > video performance is about 15-20% worse than on E-450. How did you measure that?
Created attachment 100301 [details] dmesg output
Created attachment 100311 [details] Xorg.0.log
(In reply to comment #3) > (In reply to comment #0) > > After upgrading my AMD E-450-based notebook to a newer one (HP 4545s > > A4-3300-based) i noticed that in spite of noticeable higher clock rate the > > video performance is about 15-20% worse than on E-450. > > How did you measure that? - gtkperf (which is the only "test" that shows an improvement about 30%) - compiz benchmark (shows a regression) - glxgears in both "dafault window" and full-screen modes - shows a regression, too. (Yes, I'm aware it's NOT a benchmark, for instance I noticed it's very dependent on CPU part: simple change of a CPU governor may change figures dramatically, and so on...) But in my case (I use exactly the same environment except the hardware) its numbers apparently correspond well to what I "look and feel" while using the computer: how smooth the cube rotates, firefox scrolls, etc. Unfortunately, it *is* slower, hotter and gives a shorter battery life... And while I found open source driver on E-450 quite usable, I've to switch to fglrx on my new notebook so far -- having all it's disadvantages and headaches... BTW, if you can suggest a reliable benchmark, I could do the measurements...
Does not enabling Option "BackingStore" in xorg.conf help? The high power usage might be due to the discrete GPU being powered on, but the X server isn't using it, so it doesn't help performance.
According to powertop it saves about 0.5W or a little bit more. Not a break-thru but still useful. Thanks!
(In reply to comment #8) > According to powertop it saves about 0.5W or a little bit more. That's just from not enabling backing store? Not bad, considering backing store is basically useless with a composited desktop and probably hurts performance. :) I suspect powering off the discrete GPU could save a lot more than that though.
It's already off... The first thing to do if you don't want you laptop to die in 1.5 hr :-)
For me this problem happens as following: My HD5770 card could clock up to 850MHZ - highest state, 600MHz in medium power profile and 150MHz is low power profile. However automatic management (dynpm) would only cycles 600 and 850MHz clocks and never goes to 150MHz unless I explicitly force power profile to low. Should I tell such power management is dumb? It is! You can be pretty sure that down-clocking 850MHz card to 600MHz does not makes big difference. However down-clocking to 150MHz makes a really drastic difference in power consumption and cooler noize. Should I admit current Radeon dynpm implementation suxx?