3Dmark Time Spy - Post your scores!

[Vellinious]

Yes, the CPU is 100% of the physics score, and supposed to be about 50% of the combined score. I've never sat down and figured out the math on how the calculate the overall score but, I know running a 5820k I can hit 18.5 to 19k physics scores and get a much lower overall score than someone with a 5960X, even with a higher graphics score. The 3D Mark benches are heavily weighted that way.

 

[EarthDog]

Maybe also another table/list with only the GPU score to take the CPU out of the equation?

The point here is to get the highest score at TPU. The futuremark website has all the tools to do that (sort by cpu) already I believe.

Ever since 3dmark 06, the cpu tests have played a fairly significant role in the overall score of these benchmarks.

 

[R-T-B]

The point here is to get the highest score at TPU.

Well shit, then I'm done.

Oh well.

 

[silkstone]

The point here is to get the highest score at TPU. The futuremark website has all the tools to do that (sort by cpu) already I believe.

Ever since 3dmark 06, the cpu tests have played a fairly significant role in the overall score of these benchmarks.

Ahw shucks. I was going to do a run on my HTPC just for the hell of it, but there's no way I could beat my gaming rig.

Good point about the 3DMark database, I notice I am top with a 2500k and 1060. I wonder how long that will last

 

[Vellinious]

Ahw shucks. I was going to do a run on my HTPC just for the hell of it, but there's no way I could beat my gaming rig.

Good point about the 3DMark database, I notice I am top with a 2500k and 1060. I wonder how long that will last

Probably quite a while....there aren't any full coverage blocks for the 1060, and the 2500k is getting a bit long in the tooth, so....I wouldn't be surprised to see it there for a good long while.

 

[silkstone]

Probably quite a while....there aren't any full coverage blocks for the 1060, and the 2500k is getting a bit long in the tooth, so....I wouldn't be surprised to see it there for a good long while.

I'm going to have to overclock to the point it's just stable enough for a single run, just to make sure.

The 2500k is pretty old, I am looking for a well priced 2600k to replace it unfortunately I blew this year's pc budget on a new gpu and psu.

 

[Recon-UK]

Honestly upgrading your CPU for a synthetic benchmark is stupid.
My CPU can handle much better than a 670 that i currently use, it will feed a 1070 just fine, however synthetics do seem to go a bit far beyond real world usage.

I think the closest i have seen a synthetic benchmark to real world is probably Unigine benchmarks, the higher the FPS you scale, the more CPU you need.

 

[Vellinious]

Heaven, yeah. Valley not so much. It's too CPU dependent.

 

[TheHunter]

Ok, re my original post

980ti Factory oc 1418mhz boost;cpu 4.7ghz

Score:5726
Gpu:5817
Cpu:5263

http://www.3dmark.com/spy/17651

"Max" gpu oc 1480mhz, gpu score only.. crashed later with tdr by cpu test lol

Gpu score:6109
http://www.3dmark.com/spy/18240

This is factory stock OC 1418mhz @ 1.187v boost


5726 - TheHunter / 4770K @ 4.7 GHz / ZotaC GTX 980Ti Amp! Omega @ 1418 MHz / 1805 MHz.

EDIT: cpu only part, had balanced power plan there..

Will try max OC again, maybe I can reach that magic 6K
gpu score does 6110, maybe its enough if it doesn't crash again by cpu test with tdr lol

 

[Vellinious]

Playing with the new toy....long way to go yet.

6939 - Vellinious / i7 6950X 4.0 / 980ti @ 1547 / 2153

 

[fullinfusion]

This FX CPU is killing me! Come on Zen or Intel will be taking my money!

6580 - JATownes / FX9370 @ 4.9 GHz / 2 x Powercolor 290x @ 1100/1400

View attachment 76960

@EarthDog and @JATownes

What's up with this score?

I mean something stinks here, either its my system that's f%cked up, or someone's playing games here, orrr just a glitched out bench mark

Let me say this, if I'm doing something wrong then PLEASE let me know... but JATownes score for what he's running is pretty darn high compared to what I'm getting.

The only difference is he runs twin 290X cards and I run a 290X and a 290 together and clocked even higher then buddie is, plus my CPU score is way higher

Ok now please don't take this the wrong way and any info would help heaps for me ok, but what setting are you using?

I ended up buying the bench to unlock the BS demo mode and have more control over the bench and the only way I can come close is to set my monitor to default 1920:1080

I run the same GPU driver, cpu is in my favor clock vs clock as the link shows and memory is on par so what gives... or what am I missing?

Here is what I get and its the actual online score.

http://www.3dmark.com/3dm/13668604?

Please take this post with a grain of salt and I mean no disrespect to any party but really what's up?

 

[silkstone]

@EarthDog and @JATownes

What's up with this score?

I mean something stinks here, either its my system that's f%cked up, or someone's playing games here, orrr just a glitched out bench mark

Let me say this, if I'm doing something wrong then PLEASE let me know... but JATownes score for what he's running is pretty darn high compared to what I'm getting.

The only difference is he runs twin 290X cards and I run a 290X and a 290 together and clocked even higher then buddie is, plus my CPU score is way higher

Ok now please don't take this the wrong way and any info would help heaps for me ok, but what setting are you using?

I ended up buying the bench to unlock the BS demo mode and have more control over the bench and the only way I can come close is to set my monitor to default 1920:1080

I run the same GPU driver, cpu is in my favor clock vs clock as the link shows and memory is on par so what gives... or what am I missing?

Here is what I get and its the actual online score.

http://www.3dmark.com/3dm/13668604?

Please take this post with a grain of salt and I mean no disrespect to any party but really what's up?

Heat issues?

I know i'd get wildly different results using my 280x crossfire on firestrike. One of the cards was downclocking when it got too hot and this would affect my overall score.
I also got different results between driver versions.

It could also be that time-spy doesn't support non-matched video cards correctly.

 

[fullinfusion]

Heat issues?

I know i'd get wildly different results using my 280x crossfire on firestrike. One of the cards was downclocking when it got too hot and this would affect my overall score.
I also got different results between driver versions.

It could also be that time-spy doesn't support non-matched video cards correctly.

Yeah very possible on the 2 different cards.

but with heat I don't have any heat issues, even clocked up I only run high 70's to low 80's

the 290 in the bottom slot only runs 300MHz on the core in the bench while the 290X is running full steam ahead.

Id like other mixed R9 series to post there findings here just to see whats going on.

I'm currently re-installing the bench and also went back to the newest AMD driver...

 

[silkstone]

Yeah very possible on the 2 different cards.

but with heat I don't have any heat issues, even clocked up I only run high 70's to low 80's

the 290 in the bottom slot only runs 300MHz on the core in the bench while the 290X is running full steam ahead.

Id like other mixed R9 series to post there findings here just to see whats going on.

I'm currently re-installing the bench and also went back to the newest AMD driver...

It just won't work with 2 different model cards. You'll need to wait for a 3DMark update.

It was frustrating the hell out of me too, until I did a bit of digging and found that anyone with mixed cards that still work in CF, do not work with Timespy.

Here's a discussion on the issue: https://community.futuremark.com/fo...-Time-Spy-and-Non-identical-GPUs-in-Crossfire

 

[fullinfusion]

It just won't work with 2 different model cards. You'll need to wait for a 3DMark update.

It was frustrating the hell out of me too, until I did a bit of digging and found that anyone with mixed cards that still work in CF, do not work with Timespy.

Here's a discussion on the issue: https://community.futuremark.com/fo...-Time-Spy-and-Non-identical-GPUs-in-Crossfire

Thank you very much on the info, I was litterly ripping my rig apart looking for any issues lol

Dam now I can go bench something else till this gets coded to work for us non matching card owners.

I wondered why there wasn't hardly any AMD R9 crossfire scores yet but yes it totally makes sense, but heck I'd wish there be an update already.

I can clock these cards into the 1200/1600+ range with ease

 

[silkstone]

Thank you very much on the info, I was litterly ripping my rig apart looking for any issues lol

Dam now I can go bench something else till this gets coded to work for us non matching card owners.

I wondered why there wasn't hardly any AMD R9 crossfire scores yet but yes it totally makes sense, but heck I'd wish there be an update already.

I can clock these cards into the 1200/1600+ range with ease

Me too. I was reseating the card, checking the bridges, trying different slots . . . nothing worked.

Not too many people use different cards in CF, so googling the issue wasn't that straightforwards.

 

[JATownes]

Whew! Glad nothing was fishy.

The explicit multi-adapter version they use seems to scale better than Crossfire, but definitely has drawbacks.

JAT

 

[EarthDog]

Updated!

 

[Vellinious]

Update

12611 - Vellinious / i7 6950X 4.6 / 980ti x2 @ 1531 / 2143

http://www.3dmark.com/spy/177380

 

[Laughing_Beast]

1861 - Laughing_Beast / G3220 @ 3Ghz / MSI GTX 960 @ 1190 MHz / 1753 MHz

 

[Lt_JWS]

4249 - Lt_JWS / 8320 @ 4.5Ghz/ MSI GTX1060 @ 1724/2242/1939Boost

http://www.3dmark.com/spy/201149

 

[TheoneandonlyMrK]

SCORE - THEONEANDONLYMRK / CPU 4.7 GHz / GPU @ ACTUAL 1266 MHz (actual boost clock) / Memory clock 2000 MHz

STOCK GPU CLOCKS XFIRE ON, CPU IN SUMMER FOLDING MILD MODE

 

[NinkobEi]

Not the greatest I guess but its pretty much stock.


4447 - Ninkobei / i5 3570k 3.9 GHz / R9 Nitro Fury @ 1060MHz / 500 MHz

 

[_larry]

Stock clocked XFX R9 290
Intel i5 3450 @ 3.9ghz

Score 3447

My CPU cannot be overclocked much more because it is a non - "K" series... yeah, I messed up when I bought it.
Going to see how far I can push my card though. I have copper heatsinks on the VRM chips which helps overclocking a TON on this card. I will do further testing until my card is maxed out and post another score.
This is simply the baseline score.

 

[EarthDog]

Will get caught up tomorrow..

_larry, if you would like your score posted on the front page, please follow the proper format.