Marathon: February CineBench

[OCMusicJunkie]

PS- Has anyone sorted out if this thing is impacted to even a slight degree by the GPU? I know that in every 2D bench I've ran before, overclocking the daylight out of my card netted me something. None of those were so focused on rendering as they were motion though.

 

[bmwbaxter]

Don't think an octo PhII would even be close to a intel hexa with HT.

 

[EarthDog]

Don't think an octo PhII would even be close to a intel quad with HT.

Fixed.

It PhII was still slower clock for clock than Nehalem IIRC, its not going to beat SB or IB? Not sure though, LOL!

 

[Bobnova]

Octo PhII? That'd beat nehelem, quite possibly SB/IB too.
Too bad it doesn't exist.

 

[OCMusicJunkie]

Of course we'll never know for certain, but I can't wait until I get this OS reinstall done and can run a bench with my PhII clocked 4500+. Assuming that a linear improvement would be the minimum if they were true individual cores and not the shared design of the BD and PD chips, I'd only have to hit half of what would match the intel hexa chips.

Again, all this is sort of silly to debate considering it's an imaginary chip that will never be produced, but it's interesting nonetheless.

 

[Mandrake4565]

PS- Has anyone sorted out if this thing is impacted to even a slight degree by the GPU? I know that in every 2D bench I've ran before, overclocking the daylight out of my card netted me something. None of those were so focused on rendering as they were motion though.

I'll give it a shot the next time I run my FX, I think if I test it on my Ph II with the 7600gt I have on it overclocked, it will not end well. I already have bulging caps on the gpu I don't think overclocking it will help

As far as the octo Phenom II I was just looking at my run at 4.2 on the PH II and doubling it, then comparing it to the FX @5.3. I know it doesn't work like that, nor would it be that easy, but fun to dream.

 

[EarthDog]

I cant imagine the GPU has anything to do with test (the opengl test does that), but... worth a try.

 

[OCMusicJunkie]

Yeah, I am asking before I try overclocking my 6870 or sticking my 6850 in for a temporary crossfire. I've seen supposed measures of what an OC'ed FX8350 can draw from the PSU, and I don't want to risk asking for much more from my 620w PSU unless it will matter.

 

[OCMusicJunkie]

I cant imagine the GPU has anything to do with test (the opengl test does that), but... worth a try.

You wouldn't probably believe just how much difference there was in my HyperPi times last month depending on whether I was running 32-bit 60hz or 16-bit 30hz display settings then.

 

[Mandrake4565]

You wouldn't probably believe just how much difference there was in my HyperPi times last month depending on whether I was running 32-bit 60hz or 16-bit 30hz display settings then.

And you didn't want to let me in on that little secret?

 

[EarthDog]

Well are we talking overclocking or adjusting settings as they are different. The settings mentioned are not overclocking? Im a bit confused, sorry.

Wait? Pifast? So lost... LOL!

 

[OCMusicJunkie]

And you didn't want to let me in on that little secret?

I would have, except you were close enough the last day that you could have actually caught me just by running my display and windows settings I think. I assume this is some of the secret sauce they share in the OC Team lounge I can't get into yet.

Well are we talking overclocking or adjusting settings as they are different. The settings mentioned are not overclocking? Im a bit confused, sorry.

Wait? Pifast? So lost... LOL!

They need to get more original with these names- hyperpi, superpi, pifast, raspberry pi.... stop with the pi, I can't keep them straight! I meant last month's Pifast event.

I was just pointing out that for anything that gets displayed on the screen, the GPU settings/speed has some role. Maybe too minor to even be worth a single point in a benchmark most of the time, but sometimes maybe more...

 

[Mandrake4565]

I just ran my [email protected] with my 5850 at stock 775/1000 I scored 7.76 and with it at 1000/1145 scored 7.75.

 

[OCMusicJunkie]

Just for kicks, what happens if you underclock it?

 

[Mandrake4565]

Ok just ran the cpu at the same speed and underclocked the GPU to 584/804 7.74

 

[OCMusicJunkie]

lol okay I'll quit using you as my guinea pig. I was wondering if maybe having to send fewer instructions to the gpu (since it would be running its own clock slower) would unload the CPU a bit. Doesn't look like that works though.

 

[ChanceCoats123]

That small of variance is just operating system loss and such. No tweaks so far.

 

[dejo]

just a place holder at this point, but this is what I can do on a quick oc on water. may have to drop the ss onto the chip and see what is what.
dejo/intel i7 3770k/ 4.9ghz/ 9.87

 

[ChanceCoats123]

Dang, that was 4.9 and I was hoping for a little better scaling on IB.

 

[bmwbaxter]

Updated!