Silverbench Benchmark

[KnightInd2000]

Would anyone be prepared to benchmark their CPU with Silverbench?
No install required, just need Internet Explorer with Silverlight installed and go to http://silver.urih.com/ (Chrome doesn't work).
My colleague wants to build a new machine for video encoding and is trying to compare processors.
He says his current machine, with an i5-3340 at stock speeds (3.1-3.3 GHz), scores P11140.
I said I'd check my old Gulftown when I get home tonight but he's looking for a more up-to-date processor.
I know there are benchmark websites we could look at but he's after real life experiences with sensible overclocks so please post your speed if it's not at stock.
Thanks friends.

[kalniel]

Why that benchmark? Surely if he's looking for encoding speed then he should check an encoding benchmark, ideally with the encoder he's going to use?

[KnightInd2000]

Because raytracing is processor heavy it pushes all the threads your cpu uses so it is a fair guide to any processor heavy task.
Yes there may be more 'accurate' tests out there but because there's nothing to install and it only takes a couple of minutes I thought more people might be willing to give it a go.

[g8ina]

I would, but Im not prepared to install yet more MS rubbish software.

[kalniel]

Because raytracing is processor heavy it pushes all the threads your cpu uses so it is a fair guide to any processor heavy task.

But not encoding, which is often accelerated through dedicated hardware functions, so you should test using an encoder.

Yes there may be more 'accurate' tests out there but because there's nothing to install and it only takes a couple of minutes I thought more people might be willing to give it a go.

No problem asking. Seems like quite a few ads on the site though so I'll pass.

[EndlessWaves]

The benchmarks in a hardware review are likely to be more accurate than random users with all sorts of different configurations of other components. Video encoding benchmarks should be quite easy to find and as long as they test a i5-3000 he can calculate performance next to that by the clock speed difference alone.

I would only expect modest improvements. Best case scenario for an overclocked i7-6700k or i7-5820k is probably 45% of the current encoding time.

http://www.hardware.fr/articles/940-...x264-x265.html

[KnightInd2000]

Thanks for your replies. A bit irrelevant now as his mate is giving him an i7-2600K for next to nothing so he's happy with that (for now!)
Just wondering though if anyone would still be prepared to benchmark and post results just for fun??
And if so, can anyone suggest a benchmark that we could use and people would be happy with?

[kalniel]

Thanks for your replies. A bit irrelevant now as his mate is giving him an i7-2600K for next to nothing so he's happy with that (for now!)
Just wondering though if anyone would still be prepared to benchmark and post results just for fun??
And if so, can anyone suggest a benchmark that we could use and people would be happy with?

Handbrake, as here:

http://forums.hexus.net/pc-hardware-...ded-power.html

[scaryjim]

The handbrake charts are a bit of a laugh; good fun, but it takes a lot of work to maintain the tables/charts. OTOH if you're volunteering to take that on...

p.s. I'd forgotten about that stonking overclock on my Pentium. Great chip, that one. Think I've still got it lying around somewhere...

[watercooled]

If you want encoding performance, you need encoding benchmarks, not some random synthetic one. Processors can perform extremely differently in different tasks - eg just because CPU a is faster than CPU b at gaming, doesn't mean that CPU b won't be much faster at video encoding. They're very different tasks and for example, video encoding tends to not benefit much from an i7's Hyperthreading.