So, I built a new PC and I am a little concerned about my benchmark scores. Intel i5-4690k and GTX 970

[Griffith]

Hey everyone,

 

                So, I just finished my second computer build and I am a little concerned that some of my benchmark scores are much lower than what I expected them to be. I am by no means an enthusiast, so I’m not too sure about what range of scores would be acceptable for the components in my build. However, plenty of Googling has revealed quite a few benchmarking scores that are significantly (?) greater than my own (granted, I could be overthinking this entire thing). I will try to make this post brief, so please let me know if you all need any more information!

 

*All benchmarks have been completed at stock speeds for both the CPU and GPU

 

3DMark FireStrike Score:

 

Also, what’s with the “Brawn” badge? I sure hope my CPU is willing to play, considering I spent $220 on it!

 

Cinebench R15 CPU & OpenGL Scores:

 

CPU-Z and Aida64 Information:

 

So, as stated above, I have been doing a lot of Googling in order to find someone’s benchmark with the same CPU/GPU combination. Unfortunately, I haven’t been able to find a great example, for most people post their overclocked scores instead of their stock ones. Here are the scores I am looking at that make me concerned:

 

This guy scored a 594 on Cinebench with his i5-4690k, a score that is 59 points greater than my own: http://www.bit-tech.net/hardware/2014/07/03/intel-core-i5-4690k-review/6

Lastly, a lot of the GTX 970 benchmarks for Fire Strike that I’ve seen are scored at 10,000 points or more. Mine, as you can see from the screenshot above, is at 9000.

 

On a slightly different note, I am having troubles playing DayZ Standalone at decent frame-rates. This guy (https://www.youtube.com/watch?v=UQAajBVdjCw) is running DayZ at 50-64 fps on max settings, but I am struggling to reach 40 fps on mid-high settings. Now, DayZ SA is way more CPU orientated, so maybe my CPU is at fault here? 

 

Now, I know enough about computers to build one and to fix my grandma’s internet, but I’m stumped here (unless I messed up during the build process or my CPU and GPU just aren’t as great as the rest). Anyways, any insight would be greatly appreciated and thanks for reading! 

 

P.S. Here is a photo of my computer... just in case any of you were wondering  :rolleyes:

 

[Fate]

Go test out a well optimized game, DayZ SA is a piece of poo and you shouldn't test your system on there. Your Cinebench CPU score seems about right but I'm not sure about GPU (not saying it's wrong, I just don't know).

[Tim Drake]

Can't help with your other problems but the best way to run any Bohemia Interactive game (Arma, DayZ) is to max out their graphics as it is the only way to even the load on your components.

 

I run Arma 3 15-20fps higher when I maxed it out rather than putting it on high. 

[Griffith]

Go test out a well optimized game, DayZ SA is a piece of poo and you shouldn't test your system on there. Your Cinebench CPU score seems about right but I'm not sure about GPU (not saying it's wrong, I just don't know).

 

Thank you for the reply! I am averaging 70-90 fps on Shadow of Mordor, which seems to be just as good as, if not better than, a lot of the videos I've seen online using the GTX 970. I can deal with a 59 point drop in the CPU score, especially when the scores are scaled in the hundreds. However, the 1000 point drop in the FireStrike score is a bit concerning, so I have no idea as well! Hopefully someone can offer some advice about the GPU score. 

 

Can't help with your other problems but the best way to run any Bohemia Interactive game (Arma, DayZ) is to max out their graphics as it is the only way to even the load on your components.

 

I run Arma 3 15-20fps higher when I maxed it out rather than putting it on high. 

 

Thank you for the reply as well. I will be testing that out as soon as possible. Do you think the developers will be optimizing DayZ SA anytime soon? 

[Tim Drake]

Thank you for the reply! I am averaging 70-90 fps on Shadow of Mordor, which seems to be just as good as, if not better than, a lot of the videos I've seen online using the GTX 970. I can deal with a 59 point drop in the CPU score, especially when the scores are scaled in the hundreds. However, the 1000 point drop in the FireStrike score is a bit concerning, so I have no idea as well! Hopefully someone can offer some advice about the GPU score. 

 

 

Thank you for the reply as well. I will be testing that out as soon as possible. Do you think the developers will be optimizing DayZ SA anytime soon? 

Ha! Rocket just left (main developer) and I am almost sure Bohemia have no bloody clue what they are doing. So expect updates to become pointless soon.

 

Also no. Not unless Bohemia invest in a much better engine, no amount of patches can fix the bad optimizations.

[TheMcSame]

The firestrike benchmark doesn't seem all that far off, I get roughly 11.5k with a 4690k (stock) and a (factory) overclocked 980 (1291mhz on the core, 1800mhz on the memory).

 

As for dayz, the thing is optimised like ass, the game runs on 2 core IIRC and a game like dayz really needs to run on more than 2 cores.

[CDHoward]

Hey everyone,

So, I just finished my second computer build and I am a little concerned that some of my benchmark scores are much lower than what I expected them to be. I am by no means an enthusiast, so I’m not too sure about what range of scores would be acceptable for the components in my build. However, plenty of Googling has revealed quite a few benchmarking scores that are significantly (?) greater than my own (granted, I could be overthinking this entire thing). I will try to make this post brief, so please let me know if you all need any more information!

*All benchmarks have been completed at stock speeds for both the CPU and GPU

3DMark FireStrike Score:

Also, what’s with the “Brawn” badge? I sure hope my CPU is willing to play, considering I spent $220 on it!

Cinebench R15 CPU & OpenGL Scores:

CPU-Z and Aida64 Information:

So, as stated above, I have been doing a lot of Googling in order to find someone’s benchmark with the same CPU/GPU combination. Unfortunately, I haven’t been able to find a great example, for most people post their overclocked scores instead of their stock ones. Here are the scores I am looking at that make me concerned:

This guy scored a 594 on Cinebench with his i5-4690k, a score that is 59 points greater than my own: http://www.bit-tech.net/hardware/2014/07/03/intel-core-i5-4690k-review/6

Lastly, a lot of the GTX 970 benchmarks for Fire Strike that I’ve seen are scored at 10,000 points or more. Mine, as you can see from the screenshot above, is at 9000.

On a slightly different note, I am having troubles playing DayZ Standalone at decent frame-rates. This guy (

) is running DayZ at 50-64 fps on max settings, but I am struggling to reach 40 fps on mid-high settings. Now, DayZ SA is way more CPU orientated, so maybe my CPU is at fault here?

Now, I know enough about computers to build one and to fix my grandma’s internet, but I’m stumped here (unless I messed up during the build process or my CPU and GPU just aren’t as great as the rest). Anyways, any insight would be greatly appreciated and thanks for reading!

P.S. Here is a photo of my computer... just in case any of you were wondering :rolleyes:

On your Cinebench score, what percentage does it say for how well your CPU score matches all other people who've done the test? Example: when I test my 8350, it says I have a 98% match with all others who've ran the test on the same chip.

Also, check to see if those scores amongst yours are from overclocked chips.

[sgloux3470]

Most 10,000+ scores in Firestrike are running overclocks.  

 

Here's my friends 970 run in Firestrike for example, and he actually fried the card doing this.

http://www.3dmark.com/fs/3056414

 

To be honest I wouldn't get caught up in synthetic benchmarks.  3Dmark in particular is very sensitive to overclocking, moreso than any actual games would be.

[ProKoN]

@Jumper118 has some great suggestion when running cinebench. have a look at tip #1.

 

Top tips to a high score:
1) Don't have any other programs running while you are doing the benchmark. This includes temperature monitoring software and cpuz. if you are concerned about your temps do one run with the monitor open to check they dont get too high, then another run without it open.
2) Overclock your cpu and gpu!
3) take your side panel off your case, open a window or go in a colder place. less heat+more overclock=higher scores.
4) Run the tests separately a few times this can get you that last extra point.
5) Run it on a fresh install if you can. fresh installs are always faster.
6) Don't use your 24/7 oc. top scores will be done with overclocks that might be too hot or too unstable for normal use.
7) more ghz on your cpu is always good, but make sure your ram is running at its full potential too. Normally 1600mhz cl8 will be faster than 1866mhz cl10. get that cl down as low as possible while trying to maintain a good memory frequency.

 

 

 

if your mostly doing gaming a better test would be to load up your favorite games and compare how others are doing at similar specs and settings.

[Griffith]

Ha! Rocket just left (main developer) and I am almost sure Bohemia have no bloody clue what they are doing. So expect updates to become pointless soon.

 

Also no. Not unless Bohemia invest in a much better engine, no amount of patches can fix the bad optimizations.

 

So, I carried out your suggestion to crank up all the settings. While the game does look beautiful, turning up the settings didn't really work. Either way, thanks for the response  :lol: I had no idea that Rocket left... seems a bit ridiculous considering how many people have bought the game and are waiting for the final version! I was excited to see the finished product; in spite of all of the optimization issues, I have had quite a bit of fun playing DayZ SA online. 

 

On your Cinebench score, what percentage does it say for how well your CPU score matches all other people who've done the test? Example: when I test my 8350, it says I have a 98% match with all others who've ran the test on the same chip.

Also, check to see if those scores amongst yours are from overclocked chips.

 

Thank you so much for this post. I didn't even realize I could check my score's percentile! Cinebench says "OpenGL Reference Image Match: 99.62%" I am going to assume that that means I am right up to par with other people running this same setup (without overclocking). 

 

Most 10,000+ scores in Firestrike are running overclocks.  

 

Here's my friends 970 run in Firestrike for example, and he actually fried the card doing this.

http://www.3dmark.com/fs/3056414

 

To be honest I wouldn't get caught up in synthetic benchmarks.  3Dmark in particular is very sensitive to overclocking, moreso than any actual games would be.

 

Thanks for the reply, sgloux3470. The fact that your friend fried his card getting a 10,308 does indeed seem to prove that my scores are pretty good after all. And I guess you are right, as long as the video card performs well in games then that's all that matters 

 

 

@Jumper118 has some great suggestion when running cinebench. have a look at tip #1.

 

Top tips to a high score:

1) Don't have any other programs running while you are doing the benchmark. This includes temperature monitoring software and cpuz. if you are concerned about your temps do one run with the monitor open to check they dont get too high, then another run without it open.

2) Overclock your cpu and gpu!

3) take your side panel off your case, open a window or go in a colder place. less heat+more overclock=higher scores.

4) Run the tests separately a few times this can get you that last extra point.

5) Run it on a fresh install if you can. fresh installs are always faster.

6) Don't use your 24/7 oc. top scores will be done with overclocks that might be too hot or too unstable for normal use.

7) more ghz on your cpu is always good, but make sure your ram is running at its full potential too. Normally 1600mhz cl8 will be faster than 1866mhz cl10. get that cl down as low as possible while trying to maintain a good memory frequency.

 

if your mostly doing gaming a better test would be to load up your favorite games and compare how others are doing at similar specs and settings.

 

Thanks, ProKoN. After comparing a well-optimized game like Shadow of Mordor to some YouTube videos, I can see that I am getting the same performance, or fps, as a lot of the other guys. As long as my GPU & CPU are performing as they should, I don't think I'll be too worried about my benchmarking scores anymore, as a lot of you all have convinced me that my scores seem to be fine.   

[deathjester]

 

Doesn't look like you set up the memory in the bios, unless you bought a 1600 CL10 kit.

 

Enable the XMP profile for memory in the bios. That is probably the discrepancy for your score and his. Memory speed can matter quite a bit in cinebench, even if it doesn't matter too much in a game. I saw some pretty impressive gains from 1600 cl9 to 2133 cl10, when i overclocked my memory in CINEBENCH (not games). Like I said in GAMES this doesn't mean much though. You still want to enable XMP though unless you bought a CL10 1600 kit.

 

Day-Z = CPU dependent on mainly one core like all MMO's. You are not going to hit a solid 60 on any MMO when there is a lot of crazy things going on, onscreen except maybe WoW with a OC Haswell (and even that might not get 60 at times). My 4770k at 4.5ghz can be at 30 in Guild Wars 2 at times.

 

The GTX 970? Not all of them are the same at stock speeds. Depends which one you got and add to that some models overclock like crazy. They have more power connectors and can push more voltage. All your higher end ones have this. Gigabyte G1 is notorious for hitting pretty high clocks, so it would beat a stock GTX 970 by quite a bit in a benchmark.

 

Overall? Relax . 30-40 FPS in a game like Day-Z with a ton of stuff on screen is very good. All depends where you are benchmarking a online game with a ton of players. You can benchmark a game like Guild Wars 2 over 100 FPS solo questing or questing with a couple players. In real game play in world vs world you, with a ton of players on screen, you are skyrocketing down on FPS. A low level API, which we are getting will help that a bit, but an I5 k is about as good as it gets currently. OC it will help. 

[Griffith]

 

 

Once again, thank you for the reply. I just found out that the RAM I purchased is in fact Cas Latency 10 (DDR3 1600 HyperX Fury 8gb). I don't think I knew that Cas Latency was a thing until reading your comment, so thank you for bringing that to my attention 

 

Your comment in regard to fps on MMO games is disheartening, but thank you for the information, haha. 

 

 

All in all, I would say this thread has been solved! Thank you to this awesome community for answering all of my questions with great detail and in a very, very short time-frame. Thanks for all of your help, guys!!

[Priller]

Once again, thank you for the reply. I just found out that the RAM I purchased is in fact Cas Latency 10 (DDR3 1600 HyperX Fury 8gb). I don't think I knew that Cas Latency was a thing until reading your comment, so thank you for bringing that to my attention 

 

Your comment in regard to fps on MMO games is disheartening, but thank you for the information, haha. 

 

 

All in all, I would say this thread has been solved! Thank you to this awesome community for answering all of my questions with great detail and in a very, very short time-frame. Thanks for all of your help, guys!!

 

If you play around alittle bit with the ram voltage you should be able to get atleast 1866 speed out of that ram. Although CL10 doesnt leave any room to play with.

[DildorTheDecent]

Nobody who owns unlocked overclockable hardware really keeps it at stock values. 

 

CPU OC guide: http://linustechtips.com/main/topic/41234-intel-haswell-4670k-4770k-overclocking-guide/

 

give it a go.