djsubtronic
Silver Level Poster
Nice. I ran 3DMark Time Spy - 9700 graphics and 8700 CPU score. GPU on full stock, CPU with -80 mV undervolt.
Laptop configuration
- Thread starter djsubtronic
- Start date
barlew
Godlike
Was that in office mode or something. What was your combined score?
djsubtronic
Silver Level Poster
I didn't use the Clevo Control Centre, so it wasn't on any "mode" as such I just use the Balanced power profile in Windows which I have customized a little but that wouldn't have much impact on a benchmark while plugged in.
Don't remember the combined score exactly, it was somewhere around 9200.
Macco26
Expert
Those laptops can stretch their legs exclusively by invoking settings in their Control Center, deeply linked with the EC BIOS for fan curves, CPU power limits etc.
You should test again it with CC set in its Turbo mode and post results in the dedicated thread for 3D Mark in this forum, pretty please. I don't have on hand the thread link but surely someone can help.
You should test again it with CC set in its Turbo mode and post results in the dedicated thread for 3D Mark in this forum, pretty please. I don't have on hand the thread link but surely someone can help.
djsubtronic
Silver Level Poster
I tested both using CCC (Entertainment and Performance modes) and using my own custom config with CCC installed, the results were as follow:
entertainment: graphics 9151 cpu 8667
performance: graphics 9635 cpu 8616
without ccc: graphics 9700 cpu 8613
I've been using Obsidian PC Tools for some years now, including their Fan Control app which is actually way better than CCC and allows you to set much better curves on your own.
But I tried the benchmark using both CCC and without it just to get a comparison. Differences are small, but it does seem that CCC is not necessary to get the max out of the system. Note I had a web browser running in background (although minimized) so the fluctuation in CPU score is probably just because of that.
entertainment: graphics 9151 cpu 8667
performance: graphics 9635 cpu 8616
without ccc: graphics 9700 cpu 8613
I've been using Obsidian PC Tools for some years now, including their Fan Control app which is actually way better than CCC and allows you to set much better curves on your own.
But I tried the benchmark using both CCC and without it just to get a comparison. Differences are small, but it does seem that CCC is not necessary to get the max out of the system. Note I had a web browser running in background (although minimized) so the fluctuation in CPU score is probably just because of that.
barlew
Godlike
What are the temps like on your CPU? That score to me is short about 1000 points.
SpyderTracks
We love you Ukraine
Absolutely, that system isn't configured correctly, should be MUCH higher scores than that!
djsubtronic
Silver Level Poster
Temp had peaked 74C in that test, but I guess that CPU score was a bit lower because of the tweaks I made in ThrottleStop. Here it is at full stock:
This time the CPU temp hit 85 C in the Physics test.
barlew
Godlike
@djsubtronic I'm replying to your latest post here to try and keep the other thread on topic. The latest scores are more like what I would expect but still short. What temps where you getting before you put the undervolt on?
djsubtronic
Silver Level Poster
It was around low to mid 90s, and even with the undervolt it does hit 90+ but only stays there for brief periods of less than 1 second.
Are you sure the CPU is still short? I've had a look at a number of reviews on Notebookcheck for laptops with the 10870, and all of them have less than my score of 9270.
barlew
Godlike
Maybe not the mate perhaps I'm expecting too much of that CPU. Does the CPU stay low 90C constantly under load?
djsubtronic
Silver Level Poster
Not with the undervolt. I'm currently testing -110 mV, been stable for over 24 hours with a 5-hour gaming session. It does hit brief spikes over 90 C as I mentioned, but the average over the gaming session was around 79 C iirc.
barlew
Godlike
Bloody hell. 79C is a great average temp but what was it averaging before the undervolt?
djsubtronic
Silver Level Poster
To be honest, undervolt is the first thing I do so I didn't really do any prolonged heavy workloads before that. And I don't want to misguide you so I'll do a fresh measure of the average temp after a gaming session tonight with -110 mV, and let you know the exact average.
barlew
Godlike
The reason I ask is undervolts should be applied after temps are brought under control just to further optimise the system. It sounds to me if you are averaging in the 90C that you possibly have a cooling issue. Especially if your undervolt is reducing the performance of your CPU.
djsubtronic
Silver Level Poster
How would undervolt reduce performance, if it is able to run at max speeds and remain stable?
djsubtronic
Silver Level Poster
No, I had fiddled around with other stuff in Throttlestop that I probably shouldn't have, that gave me the lower score of around 8600 on CPU. The 9270 score is only with the undervolt, and Speed Shift EPP set to 100. It seems to be maintaining max clocks without a crash at -110 mV for about 48 hours now, with long periods of heavy load during that time frame.
Last edited:
barlew
Godlike
OK I'm with you. That's a decent undervolt.