XTU tweak (big boys already know time to share)

Score is calculated within the first 1.5s of the bench. Sooooo set your clocks, open bench run 1.5s then flick slow mode switch. I prefer slow mode switch because it will down clock your CPU speed and your Cache speed to x8 multi. Will take 2-3 minutes to finish the bench, I usually go sweep or do chores till I hear the fans going full speed again (they will spin down a little under load). Now you know what most of the pros know about XTU. 6.1ghz garbage 1m chip can run 5.82ghz 4c8t using this method. Your chips should not get as much strain as well using this method, there is very little heat at 1ghz I dont even pour while its running, just when I need to turn slow mode off and get screenshot. Enjoy the level playing field!

good guy splave^^^^

Score is calculated within the first 1.5s of the bench. Sooooo set your clocks, open bench run 1.5s then flick slow mode switch. I prefer slow mode switch because it will down clock your CPU speed and your Cache speed to x8 multi. Will take 2-3 minutes to finish the bench, I usually go sweep or do chores till I hear the fans going full speed again (they will spin down a little under load). Now you know what most of the pros know about XTU. 6.1ghz garbage 1m chip can run 5.82ghz 4c8t using this method. Your chips should not get as much strain as well using this method, there is very little heat at 1ghz I dont even pour while its running, just when I need to turn slow mode off and get screenshot. Enjoy the level playing field!

ty Splave

Is this legit? For me this sounds like an bug of this bench:D

101% bug or bad design. GG Intel.

So giving big fat globals for an bugged Benchmark lol^^

So giving big fat globals for an bugged Benchmark lol^^

Sure. Same for a couple other benches too.

101% bug or bad design. GG Intel.

Seems like lazy code or something. Not a bug, bench works right just a workaround lol intel should realize we are a crafty bunch.

thanks very much splave for level play field.

Andre posted this tweak for it after first XTU contest

Didn't know it

I thought that after Andre told this bug someone fixed it on the newer version of XTU. Hmm too bad it's Sunday here and my LN2 supplier is closed. Maybe tomorrow I will give it a try.

Thanks 4 sharing!

If only this Gene board had a Slow switch thanks for sharing Master !!

Gigabyte Z87X-OC has slow switch too. Called "OC Trigger Switch"

Andre posted this tweak for it after first XTU contest

This is more direct procedure and much easier

no slow mode on maximus gene lol! will have to manually clock down

don't you have an oc panel it works on gene too

Thanks for sharing but WOW my opinion of XTU just took a nosedive.

Thanks for sharing, but I consider it a seriously flawed benchmark.

OK guys there are two ways to calculate a score in a benchmark:

1) Run through sequence. Calculations per second overall is score. Like overall frames per second.

2) Run through sequence bits, calculate per second in each bit, score is the peak in each bit over all bits. Like peak frames per second.

XTU clearly does number two. So technically you can still get a higher score further down the benchmark, it has to beat the peak bits (peak FPS) at the beginning to register.

OK guys there are two ways to calculate a score in a benchmark:

 

1) Run through sequence. Calculations per second overall is score. Like overall frames per second.

2) Run through sequence bits, calculate per second in each bit, score is the peak in each bit over all bits. Like peak frames per second.

 

XTU clearly does number two. So technically you can still get a higher score further down the benchmark, it has to beat the peak bits (peak FPS) at the beginning to register.

But would it be worth it?

Law of diminishing returns would most definitely apply.

Well it depends if performance decreases over time. Law of diminishing returns doesn't apply: the time to do a calculation should stay the same, as you don't need to throw more power to keep a game running at 30fps indefinitely. (3D analogy for 2D benchmark, but yeah same principle applies). But as we know with older 3D stuff like Aq, fresh boot does it best.

Also each bit of the benchmark could be calculating something different. This is where L1/L2/L3 caches come in, and the benchmark keeps ramping up the P95 size until it overflows L3 cache. It's going to affect different CPUs at different times. But in XTU's case it seems to be doing the same calculation over time and finding the peak score.

Law of diminishing returns does apply. You can flip to slow mode after a second and a half and save LN2 and beating on your hardware.

Question: any idea of percentage of time your score would increase if you finish the bench full tilt?

Law of diminishing returns does apply.

You can flip to slow mode after a second and a half and save LN2 and beating on your hardware.

Law of diminishing returns in the context of code applies to performance.

You mean to apply it to hardware stress, which is valid but confusing without an additional qualifier in your phrasing.

Law of diminishing returns in the context of code applies to performance.

You mean to apply it to hardware stress, which is valid but confusing without an additional qualifier in your phrasing.

But you knew what I was getting at.