• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

ASUS Matrix is a joke OC'er

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
3 cards and two threads later we have similar results from the first card for 24/7 ops. Nice!
 
Hmm no idea as to why I got that score then. I run dual monitors and it went to fullscreen mode on my main one and I could see it using the crossfire on the 2nd one.. Perhaps I will try it again!

You could try disabling one of the monitors, having the second monitor active might be using up a little bit of resources on one of your cards, or both, when my wife had 2x6870's she was in the 9300~ range, but she was overclocked to 1000mhz core/1250mhz memory on both of her cards, we ended up selling the second one about 2 months a go just because we needed the cash at the time., so i would forecast, non-OC, you should be around the 9300 range as well, simply because you have more shader engines rather than clock speed, seems like you're not overclocked so really anything in the 9000 range wouldn't surprise me on them, that's just where i feel you should be.

2x OC'ed 6870's should = 2x non OC'ed 6970's, or atleast you would think...
 
3 cards and two threads later we have similar results from the first card for 24/7 ops. Nice!

except the first card wouldn't get 1000mhz :bday:, there was a noticable ease difference overclocking this one, it didn't feel like i was walking on eggshells to stabilize the clocks, and it didn't actually crash the driver but once, once i hit 950mhz/960mhz at 1.1v i had a huge sigh of relief. I think i'm going to leave the memory alone, i don't see where it would make much of any difference being any higher than 200Gb/s for a single monitor.
 
You could try disabling one of the monitors, having the second monitor active might be using up a little bit of resources on one of your cards, or both, when my wife had 2x6870's she was in the 9300~ range, but she was overclocked to 1000mhz core/1250mhz memory on both of her cards, we ended up selling the second one about 2 months a go just because we needed the cash at the time., so i would forecast, non-OC, you should be around the 9300 range as well, simply because you have more shader engines rather than clock speed, seems like you're not overclocked so really anything in the 9000 range wouldn't surprise me on them, that's just where i feel you should be.

2x OC'ed 6870's should = 2x non OC'ed 6970's, or atleast you would think...

Yea I was thinking that also... Did you ever run 3dmark11 with the dual 6870s? I edited my last post with my info on that..
 
Hmm no idea as to why I got that score then. I run dual monitors and it went to fullscreen mode on my main one and I could see it using the crossfire on the 2nd one.. Perhaps I will try it again!

*Edit*

Is there a screenshot somewhere of dual 6870s for the score? I had my cpu @ 4.6 ghz, both cards @ 950 core mhz and 1425 memory mhz for that run and that is what I ended up with.. I also ran 3dmark11 just before that FFIV benchmark and came up with P10113 marks

That cant be right, I get 10k at full stock :confused:

http://3dmark.com/3dm11/1497479
 
Yea I was thinking that also... Did you ever run 3dmark11 with the dual 6870s? I edited my last post with my info on that..

She did, but we never recorded any screenshots or anything on it, it was in the neighborhood of 9500-9600 stock from what she remembers, i was thinking it was closer to the 9200 range but women :shrug: it's best not to say they're wrong. Overclocked was 10k ish, i don't remember much else than seeing 10,000, sorry i wish i had those screens now. I'm assuming a pair of stock 6970's should out-perform a pair 570's, let alone a pair of 6870's.
 
Last edited:
That cant be right, I get 10k at full stock :confused:

http://3dmark.com/3dm11/1497479


It has to be my 2nd screen then. I also have cpu-z coretemp etc up on it along with the afterburner fan and temp info window.

I will try turning them all off tonite, disabling the 2nd screen and running one of the benches.. See what happens!

Also your dual 570s are no slouch cards either! ;)
 
Congrats on getting a non-bum card, I'm glad that this one is working for you.
 
She did, but we never recorded any screenshots or anything on it, it was in the neighborhood of 9500-9600 stock from what she remembers, i was thinking it was closer to the 9200 range but women :shrug: it's best not to say they're wrong. Overclocked was 10k ish, i don't remember much else than seeing 10,000, sorry i wish i had those screens now. I'm assuming a pair of stock 6970's should out-perform a pair 570's, let alone a pair of 6870's.

Well I unhooked my 2nd monitor, stopped all apps that did not need to be running, removed afterburner, removed my old drivers, rebooted, installed 11.8 with CCC. Ran the FFIV bench and managed this:

ffiv-run2.jpg


cpu @ 4.6ghz

gpu @ 950 / 1425 (both)

ram @ 7-8-7-21-1T

PCI freq @ 103

This is a new system. New mobo, gpus, new install of windows 7 (as of July) updated. I don't know what else to try. Any ideas would be appreciated!
 
This might seem like a lot of work, but try this, shut down and take a card out, run the benchmark with one fo the cards, record the score, shutdown and try the other card solo, if one of them is giving you 1000-1200 points you might have a bad GPU, since the ATi/AMD cards are less likely to artifact and more likely to just slow down, it could be that.

http://www.ffxivorigin.com/content/benchmark-scores-72/

This link can provide you some food for thought on where you should be performance wise with one card, you should be just slightly above the 5870 with a single 6970, just because the 5870 has more shader engines, you have a slight advantage bbecause you have faster clocks at stock and a different architecture, i would guesstimate a single 6970 doing 5000-5200, and the 6000 series has a 90% average scale rate, so your dual GPU scrore should be closer to 10,000 than you are.

When i had my 5970 i was able to do 8700 overclocked, a pair of 6970 should be spanking it, and not just by a couple hundred points. And the 5870 chips did not scale as well as the 6000's, not nearly as well.

9300
9400
9500
-----You want to be somewhere around here
9600
9700
9900
10,000
10,500
------580's in SLi is around here
11,000

if you can, open up fraps and keep and eye on min, Max, and avg framerates, the more information we have the more we can help without telling you to try stuff :)
GPUz screenshots of both cards would be good too, to make sure you don't have one running at x1 PCie or something.
 
Last edited:
This might seem like a lot of work, but try this, shut down and take a card out, run the benchmark with one fo the cards, record the score, shutdown and try the other card solo, if one of them is giving you 1000-1200 points you might have a bad GPU, since the ATi/AMD cards are less likely to artifact and more likely to just slow down, it could be that.

http://www.ffxivorigin.com/content/benchmark-scores-72/

This link can provide you some food for thought on where you should be performance wise with one card, you should be just slightly above the 5870 with a single 6970, just because the 5870 has more shader engines, you have a slight advantage bbecause you have faster clocks at stock and a different architecture, i would guesstimate a single 6970 doing 5000-5200, and the 6000 series has a 90% average scale rate, so your dual GPU scrore should be closer to 10,000 than you are.

When i had my 5970 i was able to do 8700 overclocked, a pair of 6970 should be spanking it, and not just by a couple hundred points. And the 5870 chips did not scale as well as the 6000's, not nearly as well.

9300
9400
9500
-----You want to be somewhere around here
9600
9700
9900
10,000
10,500
------580's in SLi is around here
11,000

if you can, open up fraps and keep and eye on min, Max, and avg framerates, the more information we have the more we can help without telling you to try stuff :)
GPUz screenshots of both cards would be good too, to make sure you don't have one running at x1 PCie or something.

I will have to yank them and see whats up.. Yeah I don't want to deal with another crap card. If I find the time this weekend I will dig into this!
 
I will have to yank them and see whats up.. Yeah I don't want to deal with another crap card. If I find the time this weekend I will dig into this!

I wish you luck man!

But since its gone this far, I will let you in on my secret..

I have a magic 970.. Really.

You should trade me your 980 for it :)
 
I wish you luck man!

But since its gone this far, I will let you in on my secret..

I have a magic 970.. Really.

You should trade me your 980 for it :)

Hah! I KNEW it! You people and your magical cards. Its all making sense now. While the rest of us deal with the regular stock.

I have been blessed with a nice 980x I believe.. It will pass LinX and Prime95 @ 4.2 ghz with 1.28v.. I only run it at 4.6 for some benchmarks. Thats on air also!
 
Hah! I KNEW it! You people and your magical cards. Its all making sense now. While the rest of us deal with the regular stock.

I have been blessed with a nice 980x I believe.. It will pass LinX and Prime95 @ 4.2 ghz with 1.28v.. I only run it at 4.6 for some benchmarks. Thats on air also!

Mine needs 1.45 for 4.2 :(

I wouldnt say cursed.. because the 10 cpus I had before my last 2 were good to awsome clockers :D

:cry:
 
Ran each card separately:

6970-1.jpg


6970-2.jpg


cpu @ 4.2ghz for both.. each gpu was left at stock settings..

I ran driver sweeper before each install and reinstalled ccc 11.8 each time.

I then ran driver sweeper again and installed both in crossfire mode and ran the bench with the cpu @ 4.2ghz and both cards @ stock settings and I managed:

6970-bothnew.jpg


:shrug:

Maybe Windows 7 pro 64 bit does not like crossfire still..

:cry:
 
Back