How do you benchmark your Clevo?

Hi all.

I'm owner of a new P650RP6-G from Obsidian-PC and I want to do some benchmarks to see the overall performance of this bad boy. I only know 3DMark (paid) and the CPU-ID software (free) to do this.

Can you give me some tips/software recommendations?

Thank you

 

Minesweeper

But actually OCCT, and Unigine Valley are some other good benchmarks.

 

The real test for your machine is how well you think it does in the tasks you actually bought it to perform. If gaming is a priority, play them. If it's content creation, create content. IMO, the only really useful thing benchmarking programs like Heaven or 3DMark can do is test overclocking.

 

Unigine - Heaven
3DMark (paid or not)
Furmark (very "stress-happy," some users even reporting that it kills video cards)

This is the real benchmark. Synthetic tests can be helpful, but mean little to nothing if your primary applications aren't working well on the laptop.

 

Uh yeah I would avoid Furmark. That and Prime95 are fairly unsafe compared to most other options.

Edit: both are probably all right at stock, but I've heard horror stories when testing OCs.

 

Agree with first 2, avoid Furmark like a plague.. Prime95 or Throttlestop benchmarks are also something you can do

 

I don't really benchmark, I just run Prime95 26.6, Unigine Heaven, and some games to make sure everything is in working order and stable.

 

Some of the latest games is usually the best way, that way you get to enjoy the experience and check everything out

 

A good amount of games will have some benchmarks built in to them. Gears of War 4, Tomb Raider, Shadow of Mordor, Bioshock Infinite are just a few (though some dated) that come to mind.

The 3DMark benchmarks are nice because when they are done it spits out a statistic like "Better than 86% of all computers." Cinebench also has in stats like that for comparison when it is done worked in to it too.

 

And yes this, if games work fine, just keep it and enjoy..

 

While testing, I would recommend using a program like HWINFO64 for system statistics monitoring and recording to log file. You could then use RTSS and genericlogviewer to view and compare the log files between tests.

 

Check out post #1 in this thread http://forum.notebookreview.com/threads/clevo-overclockers-lounge.788975/ ...and, share your results with us in that thread. Almost any benchmark will do a decent job of testing the components it is designed to test. The important thing is to use default preset values so your results are relevant. If you use custom settings it makes it impossible to compare the results with others because you're comparing apples to oranges with different settings.

The exceptions are abusive long-term or looping stress tests like Furmark (which can kill GPUs) and Prime95. There is no point in torturing and potentially damaging your components. Doing so proves nothing but user stupidity and having dead parts takes all of the fun out of overclocking and benching.

My favorite CPU dependent benchmarks are wprime32M and 1024M, Cinebench R11.5 and R15, 3DMark 11 and 3DMark Vantage. I include the last two because you can have an amazing GPU and a wimpy CPU and that will give the sucky overall score that a machine deserves to have. In contrast, you can have an anemic CPU and still get a pretty nice and fluffy-looking overall score with Fire Strike as long as you have a powerful GPU.

 

I totally agree with you. Yesterday I was playing The Witcher 3 and it runs like a charm. I come from a 560Ti desktop

Ok, I will try to do that this weekend.

Thanks you all for the responses!

 

For me, when ocing (since thermals are issue on laptops) I run video conversion software, cinebench r15 and compress file on ram. rather than using prime 95 and occt as they stress every cpu component (fpu is basically a water heater coil inside cpu lol) unrealistically making you reduce oc due to insane temperatures, while you will never reach these temps in any real life scenario. e.g cpu running at 4.5ghz that failed in temps when doing occt performs better than 4.2 ghz that made it out in stress tests. (I also think mentioned tests are enough for stability when run all day)