double odd, since what you described is sleep; have you changed your power buttons default action? Even when smart shutdown or whatever the hell its call, back when i had it enabled it didnt appear to do anything. Post-creator update, it behaves as expected, too. My PC is built from older componenets, but i doubt thats the issue.

Your pc is now puzzling me; can you 1) post the forum where they talk about the power button being weird (and, lets be clear, not the physical one on your case, right? You and I are both talking of the start menu here)? And 2) perhaps we should hop on discord and screenshare some time, so we can crawl through your settings and drivers to try and get this solved?

I don't doubt the VObench score necessarily but the fact that I'm watching 2FPS on 4x4 and get a moderately high score at all is most puzzling don't you think ?

Not at all. That's how the benchmark works. As it mentions in the startup, it renders a 4x4 set of frames, showing them on-screen when they're completed. So, that isn't a single frame, that's 16 different frames. Thus, your "2 fps" is actually 2 * 16 = 32fps.

The benchmark mode is not intended to run visually smoothly or be appealing to look at, "demo" mode is available for someone seeking the aesthetics. The benchmark is really just a benchmark, the only reason it shows anything at all on-screen is to validate the content and quality of what is being rendered.

It's only the final benchmark result numbers that are of relevance here.

"fast start-up" is actually a form of hibernation? has been since i remember but it is only offered on bios's that offer full support for it. but meh, if you say so. mine still shuts off when i click shut off. im personally preferring windows 10 over any current linux distro, just too many problems on newer hardware.

@incarnate ok 2fps is ok then. I thought others were indicating in the forums that this was not normal. If that is the way it is suppose to work then ok. So 2FPS would actually look like time lapse then. It's not smooth it's very choppy as I posted in the youtube link. Is that normal for 2fps then ?

@Luxen

This seems to be a known issue but I never had this happen to me except for the last update
I've never had the "shut down" menu item not actually shut down the computer.

Also it's strange that the feature itself is located under the "choose what power buttons do".
Why would I select choose what teh power buttons do to effect what the "shut down" menu items does.

Anyhow, it appears to have been existent for longer then I thought but I just noticed it when windows updated.

It appears to have turned this item on by default when it updated. I have no idea why it would have done this if I have not selected it or even knew about this feature.

Normally you select from the start / shut down menu and shut down the computer without a fail of this feature no matter what settings you select. I've never experienced this alternate feature and never even know you could change the "shut down" menu item to actually do something other then shut down.

Here is a link talking about it too. Just google windows 10 shuts down to login screen it's out there which is how I found the solution after Windows updated and started doing this.

https://www.drivethelife.com/windows-10/windows-10-wont-shut-down-after-fall-creators-update.html

I never even had a reason to look in that section EVER except to figure out why I couldn't simply shut down the computer like most people do when they want to shut down the computer.

@yodaofborg

I've never used this feature nor checked off anything in that section. It's lame they put it under the section that says "choose what the power buttons do".

Anyhow search google for Windows 10 shuts down to login screen it seem to be a known problem for some, but I've never thought in a million years windows update would default to this sort of hibernation that takes you login screen.

Here are some 3Dmark tests.

https://www.3dmark.com/3dm/23460035

Time Spy: 9543
I'm not sure if this is good or not but according to them scores better then 94% of all results. I'm running it clock not overclocking

Also I have G-sync turned off because a different bench didn't allow it to be turned on for some reason.

I'll ad the other scores for the others tests in 3Dmarks once they complete

Firestrike: 21,284
It says better then 98% of all results. I'm assuming only among the free users and not the paid users, but I could be wrong it could mean all results as it suggests and doesn't really indicate that there is a different set of results broken out for paid vs free users but I don't really know anything about it.

I do expect to be in the top results when comparing to lessor spec hardware.
I mean shouldn't this be expected ?

https://www.3dmark.com/3dm/23460261

I'm running them both on native resolution of my monitor 2560x1440

Finally cracked 20,000 and amusingly got the best composite score at only 4.4ghz and no GPU/mem clock offsets (though 4.6ghz and +100/+700 offsets did produce scores above 20,000 more consistently)...Pascal is a strange beast.

Running arch linux 64bit
x99 i7 6core (12threaded) @3.30GHz
32GB DDR4
RX Vega 56 (w/ opensource driver built into kernel and git-mesa)
intel i530 ssd for linux install
m.2 ssd for the vo install

1080p: 10369
4k: 10256
Combined: 20625

------------------------------------------------------------------------------------------

cpu OC @3.80GHz

1080p: 10836
4k: 10586
Combined: 21422

is that what you've been doing in Jallik E-15 for the past week, Drazed?

6 core i7 at 3.3? Which one is that, drazed? [edit]ah, I assume it's a 5820k...very nice scores, particularly in light of the lower CPU clocks and midrange GPU[/edit]

Finally got around to swapping in a R5E10 + 32GB of 3333mhz ram, which aside from the obvious ram increase mainly fixed some annoying issues I'd been having with the m.2 boot drive on my R5E. With no more system changes on the horizon, I proceeded to dial things in.

Looks like 10836(1080p)/10711(4k)/21547(composite) is as good as I'm getting. VMark seems to make very little use of GPU resources (my 1080ti shows no temp increases during a run) and next to no use of extra cores, and my 6850 just won't hold 4.6Ghz no matter how crazy I get with the voltages.

A new benchmark!

Old Score (7700K, 1080 GTX)

1080p Score 11097
4K Score 10698
Composite Score 21795

--

New Score (Intel i9-9900K @ 5.0GHz all core, NVIDIA RTX 3090)

1080p Score - 18669
4K Score - 18464
Composite Score - 37133