3DMark Firestrike 2013 Benchmark Thread - Post your scores!

[prf007]

Intel 4770k @ Stock speed(need liquid cooling still) & MSI GTX 780 LIGHTNING [1254/1630] Score - 10018

http://www.3dmark.com/3dm/2270415

 

[ikanos]

Consider most runs are over clocked. My run I believe was 4GHz CPU/1150MHz core | 7012MHz memory GPU.

Correct me if I'm wrong, but another possibility could be the 2TB green drive and the fact you're running in single channel being only one RAM module is installed.

You're right man...single channel RAM might also be the issue, but I have been saving money for a decent CPU cooler. Next time I'm doing an upgrade, filling the 2nd ram slot will be a priority..!

And woaaah! so many replies. My bad, have been caught up with college work. 0_0

Totally forgot to mention this:

Download EVGA's Precision X. I have found it to be the easiest tool to use while overclocking. You can find it here:
http://www.evga.com/precision/

If you need help using the tool, post here. I have used it countless times now to get 10K runs on SLI 760's. I am sure others would help you as well!

NOTE: For starters on you 780, push Power Target to 105%, push GPU Clock Offset to +100, push Mem. Clock Offset to +125- +150. This should be a decent AND safe starting point for your 780.

I have EVGA PrecisionX, but I've never over clocked till date, I'll try it out and get back hopefully with a better score

and thanks a lot guys 😀

 

[Wydomkr]

http://www.3dmark.com/fs/1262591

I love my 780 classy acx

 

[gr3gario]

i7-4820k and nvidia quadro 6000.

http://www.3dmark.com/3dm/2461913

Surprisingly low... I must be doing something stupid.
It says I'm using non approved drivers, I'm on 332.50.

I don't really know why, any tips from people who are smarter than I?

 

[king smp]

Hmm usually workstation graphic cards will use a different set of drivers from the desktop cards.
The WS drivers are more stable and generally more focused on OpenGL than DirectX.
A part of the high cost of WS cards is the more work that goes into their drivers.
Generally a WS card will not do so well in gaming benchmarks will beat desktop cards in OpenGL and productivity apps like Maya and AutoCad
chances are the benchmark program doesnt recognize non Desktop drivers

 

[stickg1]

New card, new scores! 780 @ 1267/1778

http://www.3dmark.com/3dm/2467060

 

[Noshiz]

I just got my self a brand new GTX 780 from Asus and did a 3DMARK.

Score is 9252 http://www.3dmark.com/3dm/2476933

How did you manage to pull off 10126? I messed a bit with the overclocking tool from Asus and i managed to get the 9252 score (previously it was 8216), but it seems i can't "suck" more power from it, if i tune up a bit more the memory clock the system freezes and crashes during the benchmark.

 

[stickg1]

What speed is your 780 running at?

 

[Noshiz]

Core clock
1,149 MHz
Memory bus clock
1,540 MHz

You can see it at the benchmark results.

 

[stickg1]

Well mine was at 1267/1778 so that would explain the difference.

 

[Noshiz]

Yes I saw that, but didn't you overclock it yourself? Or was it overclocked from EVGA? I am not that much into overclocking but seems a bit wired that i can't get the numbers you have, probably i am missing some basic knowledge here.

 

[stickg1]

Naw I overclocked myself. I have a water block that cools the core, memory, and VRMs. And I unlocked the voltage and disabled GPU Boost through BIOS modification. All stuff I wouldn't recommend unless you've been overclocking a while and comfortable with it.

 

[RobCrezz]

Naw I overclocked myself. I have a water block that cools the core, memory, and VRMs. And I unlocked the voltage and disabled GPU Boost through BIOS modification. All stuff I wouldn't recommend unless you've been overclocking a while and comfortable with it.

Yeah you can get a lot more out of them using KGB and KBT if you have the right cooling in place. Ill do a firemark on my oem gtx 660 that ive been heavily overclocking!

 

[4Ryan6]

Well mine was at 1267/1778 so that would explain the difference.

The CPU he is running would also make a difference and if Nick had a hyper threading capable CPU it would be even higher than it is now.

 

[RussK1]

Ry is right... the thing to look at is GPU score and turn a blind eye to physics and CPU scores. I wish 3Dmark would remove the latter two because they affect the scores highly and confuses people.

Look, clocks and cpu mean all the difference.

http://www.3dmark.com/compare/fs/1731101/fs/1724109

 

[stickg1]

I want to grab a i7. But this 3570K is by far the best clocking Ivy specimen I've ever had. This is my sixth sample and the only one that can do 5GHz. This one does it at only 1.4v! I typically keep it at 4.7GHz with only 1.3v.

 

[Noshiz]

I am pretty sure than Hyper threading does not matter in gaming and probably in 3dmark benchmark. So i believe Ryan is wrong about this. I remember reading benchmarks between the 3570k and the 3770k, even tho the 3770k has 8 hyper threading and 8 mb cache, it had the same exact scores with the 3570k because in games and generally in 1 application the hyper threading does not matter.

 

[Noshiz]

I just got and installed another pair of 8gb rams and now i have a total of 16gb, i did a benchmark again and results went higher
http://www.3dmark.com/3dm/2492339

Is it possible? I didn't really expect to see any better result with 16gb ram vs the 8gb.

 

[stickg1]

Half of the 3DMark benchmark is exclusively for the CPU. So for the overall score, hyperthreading makes a significant difference. Your first post in this thread was referring to the overall score, which compares the entire system. In which case CPU, RAM, SSD v HDD all make a difference.

This thread is mostly to compare the graphics score part of 3DMark.

 

[RussK1]

I am pretty sure than Hyper threading does not matter in gaming and probably in 3dmark benchmark. So i believe Ryan is wrong about this. I remember reading benchmarks between the 3570k and the 3770k, even tho the 3770k has 8 hyper threading and 8 mb cache, it had the same exact scores with the 3570k because in games and generally in 1 application the hyper threading does not matter.

i7-3770K @ 4GHz vs i5-3570K @ 4.7 GHz.

http://www.3dmark.com/compare/fs/634985/fs/1724109

Ry is right, it does matter.

This should further clear the confusion:

http://www.futuremark.com/hardware/cpu/Intel+Core+i7-3770K+Processor/compare

 

[RobCrezz]

RobCrezz - 660 oem at 1254/1820, 2500k @ 4.6Ghz 5932 Graphics score

Pretty happy with that from the humble 660 oem!

 

[4Ryan6]

I am pretty sure than Hyper threading does not matter in gaming and probably in 3dmark benchmark. So i believe Ryan is wrong about this. I remember reading benchmarks between the 3570k and the 3770k, even tho the 3770k has 8 hyper threading and 8 mb cache, it had the same exact scores with the 3570k because in games and generally in 1 application the hyper threading does not matter.

I did not say anything about gaming, I was referencing the 3DMark Benchmark scoring and hyper threading does and always has made a difference in every Futuremark Benchmark there is, including Firestrike!

Below is a 3770K at 5ghz with hyper threading disabled:



Below is the same clock with hyper threading enabled:



I think that pretty much covers it!

 

[RobCrezz]

It did sound like he was basing them opinions on theory rather than experience 😉

 

[smeezekitty]

5 GHZ :O

 

[RussK1]

... there are plenty of games out there that utilize hyper-threading (or multi-core). Most notably current games that use HAVOK physics...